Technical Assistance Consultant’s Report

Project Number: 7059-GEO August 2009

Georgia: Preparing the Ajara Bypass Road Development Project

Prepared by

Engconsult, Ltd.

21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada

For

Asian Development Bank 6 ADB Avenue, Mandaluyong City Philippines

This consultant’s report does not necessarily reflect the views of ADB or the Government concerned, and ADB and the Government cannot be held liable for its contents. (For project preparatory technical assistance: All the views expressed herein may not be incorporated into the proposed project’s design.

EXECUTIVE SUMMARY

Georgia is located south of the mountain range, with Russia to the north, Armenia and Turkey to the south, Azerbaijan to the east, and the to the west. It has a population of 4.5 million. The road network (2008) consists of 1,495 kilometers of international roads, with about 92% in good to fair condition; 5,446 kilometers of secondary (interstate) roads of which 28% are in poor condition and need rehabilitation; and 13,480 kilometers of local roads. The 81 km – – Sarpi (S2) road along the western coast of Georgia, mainly located in the Ajara Autonomous Republic, is a key international highway and international transit route in Georgia. It is connected to the major Black Sea ports of Georgia, viz. Batumi and Poti, and a number of holiday resorts, the two major ones being Batumi and . Due to heavy traffic on this road there has been significant increase in congestion and accidents particularly in Batumi and Kobuleti and especially during the tourist season. The Government plans to construct two bypass roads around Batumi and Kobuleti to address these problems The Government requested ADB’s help to prepare and finance the construction of these two bypass roads. The Technical Assistance (TA) “Preparing the Ajara Bypass Roads Development Project” was approved on 28 January 2008. After discussion between ADB and Government of Georgia it was agreed that the bypass roads would form two projects (tranches) of a Multi-tranche Financing Facility (MFF). The first tranche would comprise the Kobuleti Bypass (28km) and 6 km of existing road widening while it is envisaged that the second tranche would comprise the Batumi bypass (14km). The Project aims at improving transport reliability by providing a cost-effective investment program to improve traffic flow between Poti and Sarpi, with considerable savings in both travel time and accident reduction costs. To achieve the above objectives, the TA Project’s scope of work comprised the following components: • prepare a feasibility study for construction of bypass roads and to undertake a road sector and institutional analysis. • Prepare preliminary engineering design for the preferred alignment; • Prepare a road safety component for the new road; • Identify the needs for capacity building of Roads Department of the Ministry of Regional Development and Infrastructure; • Establish benchmark indicators and monitoring system for the project implementation;

The following conclusions and recommendations are made by the Technical Assistance (TA) study team based on their investigations and analysis. These are set out in more detail in the main text and appendices of the report. Road Alignment: The road alignment provides a bypass of the developed tourist area of Kobuleti for 28km, then joins the existing coast road for 6km and then provides a bypass of Batumi for 14km. The total road is 48km in length and stretches from the Choloki to the Chorokhi Rivers. The first 34 kilometers is proposed as tranche 1 of a Multi-tranche Financing Facility (MFF) with the Batumi bypass proposed as tranche 2. Road Standard and Design Speed: The new road should be built as a two lane high class road with limited access except for the 6 km section where the existing coast road is to be widened to four lanes. It is recommended that it be generally constructed to the new Georgian design standards for an international road; where these are not applicable other international standards should be used. Traffic volumes would indicate that a two lane road will be sufficient through the 20 year evaluation period. Road Access: It is important that traffic on the project road be allowed to travel quickly and safely and that there is no interaction between through traffic and local traffic and pedestrians. It is therefore recommended that access to the new road be only at interchanges, located near traffic generation centers, and that local traffic, including tractors, pedestrians, bicycles and animals should not be allowed on the highway except for the six km section of coastal road widening. Costs: Based on the preliminary costs the estimated construction cost for the total project is GEL 523 million ($ 314 million equivalent) at current prices, including, land, design, supervision and physical and price contingencies but not including any interest during construction PMU costs or additional TAs. For the first tranche road section the cost is GEL 258 million ($155 million equivalent). Economic and Financial Analysis: Based on the preliminary costs the economic rate of return (EIRR) of the total project (48km) is 14.8 percent; for the first tranche it is 16.2 percent. Organization: It is recommended that the Ministry of Regional Development and Infrastructure (MRDI) be the executing agency (EA) and Roads Department (RD) be the implementing agency (IA). It is proposed that a project management unit (PMU) be formed within RD. Capacity Building: The construction and operation of this high class road will require technical and operational skills that are not yet at present in Georgia. It is recommended that, as part of the project, an extensive training program be initiated to develop these skills during construction and before the project opens. It is recommended that, as part of this capacity building training be carried out in conducting an axle load survey to ascertain the extent of the problem of overloaded vehicles. Other technical assistance is recommended in management of private sector involvement in infrastructure projects and in preparing a roads transport strategy. Road Safety: In order to improve road safety on the project and other roads it is recommended that road safety audits, awareness campaigns and training in road safety issues be included in tranche I of the project. Environment: RD and its related divisions should review the suggested environmental clauses included in the EIA/SEIA and ensure that they are inserted into the contract specifications section of the bid Land Acquisition and Resettlement. The RD, with due assistance from the Ministry of Finance, should ensure that implementation of land acquisition and resettlement are carried out promptly and efficiently in a legal and transparent manner following the Land Acquisition and Resettlement Framework (LARF) and Land Acquisition and Resettlement Plan (LARP) agreed with ADB. The RD should ensure that implementation of the LARP is monitored, evaluated, and reported to ADB as required in the LARP and LARF. Social. RD should ensure compliance with ADB’s Policy on Gender and Development (1998) during implementation of the Project and should take necessary steps to encourage women living in the area of tranche I and further tranches of the MFF to participate in such project implementation, including causing the contractors to maximize employment of women. RD should ensure an independent monitoring of the social impacts throughout implementation of the Project,

FINAL REPORT

Asian Development Bank Roads Department Ministry of Rural Development and Infrastruucture of Georgia

Main Report

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 , Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Contents I STUDY CONCLUSIONS AND RECOMMENDATIONS ...... 1 II INTRODUCTION ...... 7 A. Background to the Study ...... 7 B. Description of Project ...... 8 1. Project Area ...... 8 2. Components ...... 8 C. Conduct of the Study ...... 9 1. Executing Agency and Counterparts ...... 9 2. Study Team ...... 9 3. Summary of Methodology ...... 9 4. Progress ...... 10 D. Contents of the Report ...... 11 III RATIONALE: SECTOR PERFORMANCE, PROBLEMS AND OPPORTUNITIES ...... 12 A. Performance Indicators and Analysis ...... 12 B. Analysis of Key Problems and Opportunities...... 13 1. Problems and Constraints...... 13 2. Revenues and Expenditure ...... 15 C. Stakeholder Participation and Consultations ...... 15 IV THE PROPOSED PROJECT ...... 17 A. Project description ...... 17 1. Project Area ...... 17 2. Components ...... 17 B. Project Rationale ...... 17 C. Impact and Outcome ...... 17 D. Outputs ...... 18 1. Output for Tranche 1 ...... 18 2. Outputs for Tranche 2 ...... 19 E. Cost Estimates ...... 19 F. Implementation Arrangements ...... 21 G. Implementation Period ...... 21 V PROJECT BENEFITS, IMPACTS AND RISKS ...... 22 A. Benefits ...... 22 1. Traffic Analysis ...... 22 2. Economic Analysis ...... 22 3. Financial Analysis and Tolling ...... 23 B. Safeguard Impacts ...... 23 1. Environmental Impacts ...... 23 2. Social and Poverty Impacts ...... 25 C. Resettlement Impacts ...... 26 D. Project Risks ...... 26

Appendices 1 Design and Monitoring Framework 2 Transport Sector Review 3 Technical and Design Summary 4 Road Safety 5 Stakeholder Participation and Consultations 6 Cost Estimates 7 Implementation Schedule

Engconsult Ltd. ii

8 Procurement Plan 9 Traffic Forecast 10. Economic Analysis 11 Institutional and Financial Assessments 12. ToR for Construction Supervision Consultants 13. ToR Road Safety Enhancement and Awareness Training 14 ToR Capacity Building 15 ToR Public-Private Partnership (PPP) Capacity Enhancement 16 Summary Poverty Reduction and Social Strategy 17 Summary Land Acquisition and Resettlement Framework 18 Summary Land Acquisition and Resettlement Plan 19 Environmental Assessment and Review Framework 20 Summary Environmental Impact Assessment 21 Cumulative Impact Assessment

Volume II Supplementary Appendices 1. Transport Trends and Economic Profile 2. Road Standards, Road, Structures and Tunnel Design 3. Project Preliminary Cost Estimates and Schedules 4. Road Safety 5. Detailed Design ToR 6. Traffic Forecasts and Economic Evaluation of the Project 7. Institutional and Financial Assessments and Financial Evaluation of the Project 8. Land Acquisition & Resettlement Framework (LARF)

Volume III Preliminary Design Drawings

Stand-alone documents Land Acquisition and Resettlement Plan (Tranche 1) Environmental Impact Assessment Report. Tables Table 1. Project Area Districts ...... 8 Table 2. Project Cost Estimate and Financing – Tranches 1 and 2 ...... 20 Table 3. Project Cost Estimate and Financing – Tranche 1 ...... 20 Table 4 Disbursement Schedule – Tranche 1 ...... 20

Maps Map 1: Area of Project Map 2: Project Area Map 3: Alignment 0-24km Map 4: Alignment 24-48km

Engconsult Ltd. iii

ABBREVIATIONS

ADB Asian Development Bank AH Affected Household AP Affected Person CIA Cumulative Impact Assessment EA Executing Agency EARF Environment Assessment Review Framework EIA Environmental Impact Assessment EIRR Economic Internal Rate of Return EMP Environment Management Plan EU European Union GDP Gross Domestic Product GNI Gross National Income GoG Government of Georgia Ha Hectare IA Implementing Agency JICA Japan International Cooperation Agency LARF Land Acquisition and Resettlement Framework LARP Land Acquisition and Resettlement Plan LSG Local Self Government MED Ministry of Economic Development MFF Multi-tranche Financing Facility MRDI Ministry of Regional Development and Infrastructure NGO Non-Government Organization NPV Net Present Value O/D Origin and Destination PBC Performance Based Contracts PMU Project Monitoring Unit PP Patrol Police PPP Public Private Participation RAMSAR Ramsar Convention on Wetlands RD Roads Department RoW Right of Way SEIA Summary Environmental Impact Assessment SPRSS Summary Poverty Reduction & Social Strategy

TA Technical Assistance ToR Terms of Reference UTA United Transport Administration WB World Bank VOC Vehicle Operating Costs

CURRENCY EQUIVALENTS (As of May 2009)

1 $US = 1.65 Georgian Lari (GEL)

Engconsult Ltd. 1

I STUDY CONCLUSIONS AND RECOMMENDATIONS

1. The following conclusions and recommendations are made by the Technical Assistance (TA) study team based on their investigations and analysis. These are set out in more detail in the main text and appendices. 2. Road Alignment: The road alignment is set out in Maps 3 and 4. It provides a bypass of the developed tourist area of Kobuleti for 28km, then joins the existing coast road for 6km and then provides a bypass of Batumi for 14km. The total road is 48km in length and stretches from the Choloki to the Chorokhi Rivers. The first 34 kilometers is proposed as tranche 1 of a Multi- tranche Financing Facility (MFF) with the Batumi bypass proposed as tranche 2. 3. Road Standard and Design Speed: The new road should be built as a two lane high class road with limited access except for the 6 km section where the existing coast road is to be widened to four lanes. It is recommended that it be generally constructed to the new Georgian design standards for an international road; where these are not applicable other international standards should be used. However it is recommended that a reduced standard for tunnel width be used consistent with international standards and that the road pavement lane width be 3.5m instead of 3.75m to reduce earthwork and bridge quantities. Traffic volumes would indicate that a two lane road will be sufficient through the 20 year evaluation period. 4. Road Access: It is important that traffic on the project road be allowed to travel quickly and safely and that there is no interaction between through traffic and local traffic and pedestrians. It is therefore recommended that access to the new road be only at interchanges, located near traffic generation centers, and that local traffic, including tractors, pedestrians, bicycles and animals should not be allowed on the highway except for the six km section of coastal road widening. 5. Costs: Based on the preliminary costs the estimated construction cost for the total project is GEL 523 million ($ 314 million equivalent) at current prices, including, land, design, supervision and physical and price contingencies but not including any interest during construction PMU costs or additional TAs. For the first tranche road section the cost is GEL 258 million ($155 million equivalent). 6. Economic and Financial Analysis: Based on the preliminary costs the economic rate of return (EIRR) of the total project (48km) is 14.8 percent; for the first tranche it is 16.2 percent. 7. Organization: It is recommended that the Ministry of Regional Development and Infrastructure (MRDI) be the executing agency (EA) and Roads Department (RD) be the implementing agency (IA). It is proposed that a project management unit (PMU) be formed within RD. 8. Capacity Building: The construction and operation of this high class road will require technical and operational skills that are not yet at present in Georgia. It is recommended that, as part of the project, an extensive training program be initiated to develop these skills during construction and before the project opens. These skills would include, tunnel design and construction, operation and maintenance, pavement design, environmental and social/resettlement monitoring. It is recommended that, as part of this capacity building training be carried out in conducting an axle load survey to ascertain the extent of the problem of overloaded vehicles. Other technical assistance is recommended in management of private sector involvement in infrastructure projects and in preparing a roads transport strategy. 9. Road Safety: In order to improve road safety on the project and other roads it is recommended that road safety audits, awareness campaigns and training in road safety issues be included in tranche I of the project.

Engconsult Ltd. 2

10. Environment: RD and its related divisions should review the suggested environmental clauses included in the EIA/SEIA and ensure that they are inserted into the contract specifications section of the bid documents and in particular implement the following actions: • Implement vegetation rehabilitation and soil conservancy actions during construction and operation; • Minimize woodland acquisition; • Protect local natural landscape, avoid high fill and deep cut section when possible. • Manage traffic congestion during construction so as to avoid long delays and maintain safe conditions, and, • Employ local farmers for environmental mitigation activities which they are skilled to undertake, e.g. the revegetation and replanting activities. 11. Land Acquisition and Resettlement. The RD, with due assistance from the Ministry of Finance, should ensure that implementation of land acquisition and resettlement are carried out promptly and efficiently in a legal and transparent manner following the Land Acquisition and Resettlement Framework (LARF) and Land Acquisition and Resettlement Plan (LARP) agreed with ADB and in accordance with Georgian laws and regulations and ADB’s Policy on Involuntary Resettlement. The RD should ensure that implementation of the LARP is monitored, evaluated, and reported to ADB as required in the LARP and LARF. 12. Labor, Social, Gender and Development, Health. RD should ensure compliance with ADB’s Policy on Gender and Development (1998) during implementation of the Project and should take necessary steps to encourage women living in the area of tranche I and further tranches of the MFF to participate in such project implementation, including causing the contractors to maximize employment of women. RD should ensure an independent monitoring of the social impacts throughout implementation of the Project, in consultation with local governments and beneficiaries. In this respect, RD should ensure that all civil works contractors (i) comply with all applicable labor laws; (ii) use its best efforts to employ women and local people, including disadvantaged people, living in the vicinity of the projects financed under the MFF; (iii) disseminate information at worksites on the risks of sexually transmitted infections for those employed during construction; (iv) provide equal pay to men and women for work of equal type; (v) provide safe working conditions for male and female workers; and (vi) abstain from child labor. Contracts for all projects financed under the MFF should include specific clauses on these undertakings, and compliance will be strictly monitored during implementation. The RD should ensure that compliance with these provisions is monitored.

Engconsult Ltd.

Engconsult Ltd.

Map 1: Area of Project

3

4

Map 2: Project Area

Engconsult Ltd.

Engconsult Ltd. Engconsult Ltd.

Map 3: Alignment 0-24km

5

Engconsult Ltd. Engconsult Ltd.

Map 4: Alignment 24-48km

6 7

II INTRODUCTION

A. Background to the Study 13. Georgia is located south of the Caucasus mountain range, with Russia to the north, Armenia and Turkey to the south, Azerbaijan to the east, and the Black Sea to the west. It has a population of 4.5 million. Gross Domestic Product (GDP) and Gross National Income (GNI) per capita have increased significantly in the past few years, although it is expected to drop in 2009 partly because of the global economic crisis. The transport sector is one of the fastest growing in the Georgian economy. Transport, storage and communications have substantially increased their contribution to the total output, from 4.6% of GDP in 1996 to 11.5% in 2006.The country's GNI per capita of US$1,560 in 2006; however, is one of the lowest among lower middle-income countries. 14. Official development assistance is derived from bilateral, multilateral, and non- government organization (NGO) resources. The principal bilateral sources are Germany, Japan, , and the . Among the multilaterals, World Bank (WB) and European Union (EU) are the largest loan providers. The United Nations system is also a significant multilateral grant provider. 15. The road network (2008) consists of 1,495 kilometers of international roads, with about 92% in good to fair condition; 5,446 kilometers of secondary (interstate) roads of which 28% are in poor condition and need rehabilitation; and 13,480 kilometers of local roads (2006) the responsibility for which was transferred to Local Self-Government (LSG) bodies in Jan 2007. 16. There are thirteen trunk highways in Georgia, totaling 1,525 km. These are: • S1-Tbilisi-- ( border) 435km • S-2 Senaki-Poti-Sarpi (Turkish border) 110km • S-3 --Larsi (Georgian Military Highway) 139km • S-4 Tbilisi-Red Bridge (Tsiteli-Khidi) (Azerbaijan border) 46km • S-5 Tbilisi-Bakurtsikhe- (Azerbaijan border) 140km • S-6 Ponichala--Guguti (Armenian border) 94km • S-7 Marneuli-Sadakhlo (Armenian border) 34km • S-8 -Vale (Turkish border) 97km • S-9 Tbilisi bypass 49km • S-10 Gori-Ergneti (S Ossetia border) 25km • S-11 - (Armenian border) 110km • S-12 -Grigoleti (beach area N of Kobuleti) 57km • Koda-Ninotsminda (no number) 189 km 17. The often rough topography, the low-capacity highways, the inadequate maintenance regime, the poor technical standards of vehicles, and other adverse conditions mean the traffic flow is often slow and the risk of accidents is high. 18. The 81 km Poti – Batumi – Sarpi (S2) road along the western coast of Georgia, mainly located in the Ajara Autonomous Republic, is a key international highway and international transit route in Georgia. It is connected to the major Black Sea ports of Georgia, viz. Batumi and Poti, and a number of holiday resorts, the two major ones being Batumi and Kobuleti. Due to heavy traffic on this road there has been significant increase in congestion and accidents particularly in Batumi and Kobuleti and especially during the tourist season. The Government plans to construct two bypass roads around Batumi and Kobuleti to address these problems. 19. The Government requested ADB’s help to prepare and finance the construction of these two bypass roads. The Technical Assistance (TA) “Preparing the Ajara Bypass Roads

Engconsult Ltd. 8

Development Project” was approved on 28 January 2008 and, subsequently, on 1 September 2008, the TA became effective and the TA letter was signed by the Government. In January 2009, ADB selected Engconsult Ltd., Canada to carry out the Project. 20. After discussion between ADB and Government it was agreed that the bypass roads would form two projects (tranches) of a Multi-tranche Financing Facility (MFF). The first tranche would comprise the Kobuleti bypass (28km) and 6 km of existing road widening while it is envisaged that the second tranche would comprise the Batumi bypass (14km). In this report the two tranches are called tranche 1 and tranche 2, while both bypasses are named “The Project”. 21. Engconsult Ltd. mobilized its team in Tbilisi on 10 February 2009 together with associated local consultants. The Inception Report was presented on March 6th, the 1st monthly report on April 5th and the 2nd monthly report on May 6th. The Draft Final Report was presented on June 28th and, after considering comments by ADB and Government, this Final Report is presented as required by the contract1. B. Description of Project 1. Project Area 22. The direct project area includes the three districts of Kobuleti, Batumi and in the autonomous region of Ajara2, through which the existing highway and the proposed bypasses traverse. The total population of the three Rayons, as set out in Table 1, is 204,800 (2008) of whom 122,000 were resident in Batumi municipality who will benefit from reduced congestion and safer passage on the existing road as well as faster travel for through traffic using the bypass. Over 50 percent of the rural population of this area lie below the poverty line. 23. The indirect project area covers most districts between Batumi and Tbilisi and, to a smaller extent, most of the rest of Georgia (Population 4.4 million) as they will ultimately benefit from the lower transport costs of agriculture and manufactured goods traveling over the new bypass roads, particularly to and from Turkey. Table 1: Project Area Districts Population Districts January 2007 City of Batumi, 122.2 89.4 Khelvachauri Municipality 92.2

2. Components 24. The Project aims at improving transport reliability by providing a cost-effective investment program to improve traffic flow between Poti and Sarpi, with considerable savings in both travel time and accident reduction costs. 25. To achieve the above objectives, the TA Project’s scope of work comprised the following components: • prepare a feasibility study for construction of bypass roads (about 48 km) for Batumi and Kobuleti in Ajara Autonomous Republic, including engineering, economic, social and

1 The LARP{ for tranche 2 is not included with this report due to unavoidable delay in completing the survey work. It will be presented later as a separate document. 2 The first four km of the Kobuleti bypass are in Rayon of Juria region

Engconsult Ltd. 9

environmental criteria for ADB financing and to undertake a road sector and institutional analysis. • Prepare preliminary engineering design for the preferred alignment; • Prepare a road safety component for the new road; • Identify the needs for capacity building of Roads Department (RD) of the Ministry of Regional Development and Infrastructure (MRDI); • Establish benchmark indicators and monitoring system for the project implementation;

C. Conduct of the Study 1. Executing Agency and Counterparts 26. The Executing Agency (EA) for the study was the RD and the Counterpart for the study was Mr. Tengiz Guelesiani, Head of Technical Department. 27. The Project team would like to express its thanks for the help, support and advice given to them by Mr. Guelesiani and other staff of RD and the support of staff from Georgia Resident Mission during the project. 2. Study Team 28. The study team for the Project consisted of: Team Leader/Road Engineer G. Connor Transport Economist T. Yates Tunnel Specialist S. Mitani Environment Specialist M. Karim Social Specialist S. Jena Road Engineer B. Kopaliani Structural Engineer D. Ukleba Transport Economist B. Kakhadze Transport Economist T. Todradze Financial Specialist G. Shihashvil Social Specialist J. Gogodze Social Specialist L. Janjgava Environment Specialist O. Turmanidze Senior Surveyor

3. Summary of Methodology 29. There had been a previous prefeasibility study, carried out in 2005/06 for the two bypass roads. The consultants reviewed this and other relevant past studies and an initial selection of possible alignments was made based on the recommendations of the pre-feasibility study and available mapping. Field inspections were made of each route and plotted on digital mapping. 30. Traffic studies were carried out including classified 12 and 24 hour counts and Origin and Destination (O&D) surveys. Forecasts were made for normal traffic, generated traffic, traffic diverted from other roads and modes and traffic from proposed or possible future development in the project area. 31. The existing road is at present the main transit route for trade from and to Georgia to Turkey and beyond and also as a transit route to and from Armenia and Azerbaijan to Turkey. A study was made of future development projects which would affect this traffic and it was

Engconsult Ltd. 10 determined that some of the transit through traffic could be diverted to a new border crossing between Turkey and Georgia and, if relations between the two counties improved, directly between Turkey and Armenia. It was decided to evaluate scenarios for both options. However, the chances of either or both of these options were considered to be small and they were only used as a special case (traffic case B) in the economic evaluation sensitivity testing. 32. Preliminary engineering and safeguard studies were carried out on the preliminary alignments and a recommended alignment presented to Government for approval and further study. 33. It was agreed with GoG and ADB that the road should be constructed as a two lane two way limited access road except for 6 km of existing road which would be widened to four lanes with no access control. Design, however, should allow for a four lane road to be constructed at a later date, depending on actual traffic diversion and growth. The section from Batumi to the border at Sarpi, originally in the scope of work, was dropped as it was considered economically infeasible. 34. A preliminary topographic and geotechnical survey of the preferred alignment was carried out. 35. Based on these surveys and further site inspections preliminary design was carried out for alignment, pavement, bridge and other drainage structures and tunnels. Following this, cost estimates were made of the preferred alignment. These were based on the preliminary design, availability of materials and the maximum use of local labor. 36. Preliminary design drawings including long sections, plans and standard cross sections were produced. 37. Terms of Reference (ToR) for detail design consultants were produced as were draft ToR for supervision consultants, and capacity building TAs to be included in tranche 1. 38. A road safety audit was carried out on the preliminary design of the project road and a road safety component program prepared to prepare drivers and local residents of the particular safety problems of a high speed road. 39. Using forecast traffic volumes and construction and operation costs, economic analyses were carried out for each of the two sub-projects and for the project as a whole. Environmental, land acquisition, resettlement and other social impacts were also assessed and included in the analyses. Sensitivity and risk analysis were also carried out. 40. Safeguard studies and surveys were carried out for the whole project, an Environmental Impact Assessment (EIA) and Summary Environmental Impact Assessment (SEIA) and Land Acquisition and Resettlement Plan (LARP) were prepared for tranche 1 and Environment Assessment Review Framework (EARF) and Cumulative Impact Assessment (CIA) for tranche 2 and subsequent tranches. 41. Further details of the work carried out are set out in later sections of the Report and, in more detail, in the Supplementary Appendices. 4. Progress 42. An Inception mission by ADB was made in late April 2009 to assess the progress of the project and to carry out discussions with GoG and a site visit to the preferred alignment was carried out with ADB safeguard specialists. 43. In April 2009 GoG accepted the recommendations on the alignment. The team completed the preliminary design and analyses in May 2009 and the results are presented in

Engconsult Ltd. 11 this Final Report. The preliminary design drawings and the LARP for tranche 1 are also included with the Final Report. 44. A fact finding mission was made by ADB during the first week in June and a further review mission in late July. D. Contents of the Report 45. The main body of this report summarizes the results of the Feasibility Studies, Preliminary Design and other work carried out during the TA project. The appendices to the report describe in more detail the methodology used in of the study, the design standards and methodology, cost estimates and implementation plans, and summary of the economic and financial analyses. The SEIA, summary LARP and LARF and a social assessment (SPRSS) are also included in the appendices. The remaining sections of this report are as described in the following paragraphs. 46. Section II gives the rationale for the project and gives the background to the transport and road sector in Georgia. 47. Section III gives the impact and outputs from the proposed project and sets out the cost estimates for the preferred alignment. It discusses the stakeholder participation and consultation carried out in the project and also discusses possible implementation arrangements for the project. 48. Section IV describes the technical benefits and risks for the preferred alignment based on traffic projections and construction and operation costs. It gives the results of the economic and financial analyses. It also summarizes the impact of the alignment on resettlement, minorities, poverty and the environment. It also describes the risks associated with the Project. 49. There are 21 appendices included in Volume I which describe the study methodology, technical analyses, and conclusions and recommendations in more detail. 50. The report is accompanied by 8 Supplementary Appendices describing in more detail the work carried out. The design drawings are presented in a separate Volume III. 51. The EIA and LARP for tranches I are presented as separate stand alone documents.

Engconsult Ltd. 12

III RATIONALE: SECTOR PERFORMANCE, PROBLEMS AND OPPORTUNITIES

A. Performance Indicators and Analysis 52. Georgia’s transport infrastructure consists mainly of roads and rail. Roads carry about 99 percent of all passengers and at least 55 percent of freight within the country. A key part of Georgia’s road system is the 383 km East West Highway (E-60) which runs from Senaki on the Black Sea coast to Tbilisi through the center of the country ending at the Red Bridge on the Azerbaijan border. In addition there are eight international and national highways linking major towns that serve much of the population. The major NS highway goes from Sarpi on the Turkish border to Poti, the major sea port in Georgia continuing to Sokhumi in Abkhazeti; this is connected to the E-W Highway at Poti and , the start of the project road. 53. Road Network: Responsibility for road construction and maintenance in Georgia is split between the Roads Department, districts and cities. The Roads Department is responsible for the main road network, comprising international roads (i.e. those used by international transit traffic) and interstate roads. In 2007 RD lost responsibility for local roads, however, 2,000km of what had hitherto been local roads were re-designated as interstate roads. Cities remain responsible for urban roads. The main road network was 6,941km at the end of 2008, of which 347km lay within Ajara. The local road network is approximately 13,388km (Ajara: 965km). In terms of length both the main and local networks have hardly changed since 2000, the investment emphasis having been on rehabilitation rather than network extension. All but 55km of the 1,495km of international roads are paved, but in the case of interstate roads the percentage falls to 61, while only 23 percent of local roads are paved. As with its total length, the proportion of the national network that is paved has hardly changed since 2000. National and Ajaran road densities are 290km and 160km per 1,000km2 respectively. Densities are low by regional standards: Azerbaijan has 690km/1,000km2 (both paved and unpaved roads) and Armenia 270km/1,000km2 (but paved roads only). 54. Vehicle Fleet: For administrative reasons vehicle fleet statistics are of poor quality and almost certainly overstate the total stock of vehicles that are in use. According to the Ministry of Internal Affairs the total fleet in 2008 comprised 613,000 vehicles, of which 543,000 were passenger vehicles, 58,000 goods vehicles and 12,000 categorized as “other”. Average annual growth from 2000 to 2008 was 12 percent. Vehicle ownership per 1,000 people has increased from 56 in 2000 to 140 in 2008. Georgia’s car ownership is broadly consistent with those of its neighbors, bearing in mind the differences in GDP/head. The increase in vehicle ownership is attributable to sustained economic growth (GDP per head grew by an average annual 7.9 percent from 2001 to 2008) and has been helped by tariff reductions following Georgia’s accession to the World Trade Organization in 1999. The fleet has been modernized in the process, as Soviet era vehicles are replaced by western and Japanese imports, improving vehicle safety and reducing individual vehicles’ fuel consumption and their emissions. 55. Organizational Structure: Institutional arrangements changed in February 2009 when a new ministry was created, the Ministry of Regional Development and Infrastructure (MRDI). In that month the Roads Department and the United Transport Administration (UTA) were moved from the Ministry of Economic Development (MED) to the new ministry. The Roads Department is solely responsible for construction and maintenance of the main road network, all of which is contracted out to private sector companies. Strategy and planning are handled by the Transport Department (whose line ministry is the MED). 56. From 2005/6 annual licensing of vehicles ceased and was replaced by a requirement to register only at a change of ownership. Vehicle registration and driver licensing is handled by the police. There are no longer any vehicle roadworthiness tests and owners are not obliged to

Engconsult Ltd. 13 carry even third party insurance. (The abolition of annual licensing accounts for the unreliability of vehicle ownership statistics: there is no longer any mechanism for determining that a vehicle is no longer used on the road). 57. Road Transport: Georgia’s road freight industry is made up of private companies and is completely unregulated. Entry to the market is easy. It is difficult to judge the extent of market efficiency, but freight rates of between GEL0.20 and GEL0.27 per tonne-km for hauls of 100- 500km are at least consistent with costs predicted using HDM-4, which suggests that excess profits are not being earned. 58. Intercity bus transport is completely deregulated and there is no official control of prices. Intercity bus services are extensive and frequent. Although most routes are plied by both mini- buses and larger buses, there is little to choose in terms of journey time as all buses will stop to pick up passengers en route if they have room. Fares between bus stations (e.g. Tbilisi and Batumi) are controlled by bus station operators, who take a cut from the fare. Bus station operators also appear to control capacity – a new entrant would find it difficult to operate a new service from an existing bus station. This control of both prices and capacity may be hindering the development of new services (e.g. express services). 59. Where a direct comparison can be made, bus and rail fares are comparable; for example, rail passengers between Tbilisi and Batumi pay GEL20 for a single trip, while bus passengers pay around GEL18. Trains are slower but are more comfortable. The introduction of limited stop, comfortable bus services at a premium fare would change the market considerably and be likely to increase bus ridership. 60. The development of the Kobuleti and Batumi bypasses should encourage the development of faster bus services between the main centers along the Ajaran coast. B. Analysis of Key Problems and Opportunities. 1. Problems and Constraints. 61. Road Construction and Maintenance: Since 2004, the Government has invested some GEL990 million in road construction, rehabilitation and maintenance, with financial assistance from the World Bank, the Kuwait Fund, Germany (KfW) , Japan International Cooperation Agency (JICA) and the United States (Millennium Challenge Georgia Corporation). Foreign assistance for the road subsector is increasing. Despite the surge in road sector investment since 2004, the maintenance backlog for the road network is estimated at GEL970 million. Sustained financing for highways and roads will continue to be needed to bring the road system (particularly secondary and local roads) to a maintainable standard, while adding essential new capacity to meet the needs of trade (transit), rural development , agriculture, industry and tourism. Increased mobility and access to secondary urban centers is expected to help with job creation. 62. The RD maintenance program 2009-2013 forecasts an increase in funding from $118 million in 2009 to $158 million in 2013. In 2007, responsibility for local roads was transferred from RD to 69 local authorities, as part of broader public sector reform aimed at decentralizing government functions. An estimated GEL 1,450 million is needed to rehabilitate the local road system (90 percent in poor condition) and bring it up to a maintainable condition. Rehabilitated local roads would require GEL1,500-4,000 per kilometer annually for routine and winter maintenance. 63. RD has no direct labor force: it contracts out all its maintenance and construction activities. The country is divided into 34 routine maintenance zones. Contractors bid against bills of quantities drawn up by the department. The last round of bidding took place in 2007 and

Engconsult Ltd. 14 was for a period of 27 months. These contracts will expire in 2009 and will be re-bid for a period of 12 months. Supervision and verification of quality and quantities were also contracted out to a private company but has now been brought into the department. 64. The World Bank is supporting the RD in its efforts to improve the efficiency of its road maintenance activities. It believes that a shift from contracted out maintenance to performance based road maintenance contracts (PBCs) could produce further savings (up to 50 percent). A proposed project component of the Third East West Highway Project will assist in developing a comprehensive framework for the introduction of PBC in Georgia. The international consultant, in collaboration with a small RD team, will apply the PBC framework to a preliminary assessment of a concession for the maintenance of the East-West Highway and an area-wide contract for rehabilitation and maintenance for 250-300km of main and secondary roads centered on a Regional RD Office. 65. The national contracting capacity in the road subsector includes 13 large contractors with an annual turnover of more than GEL10 million per year, about 30 medium size contractors with an annual turnover between GEL5-10 million and more than 60 smaller contractors. 66. Road Safety is a growing problem in Georgia. In 2008 there were 867 fatalities from road-traffic accidents, a rise of over 18% from 2007. The fatality rate relative to the vehicle fleet at 16 per 100,000 vehicles in 2007 is significantly higher than other Eastern European countries and much higher than other developed countries in EU3. Both the fatalities and serious injuries resulting from traffic accidents in Georgia over the past few years show an upward trend and road-safety is thus becoming a serious social issue. 67. World Bank review of statistics for 19964 showed that about 66% of deaths from road accidents occur in "urban" areas and 34% occur on "rural" roads. Over 30% of those killed nationally in 2006 were pedestrians. This is almost double the percentage in many West European countries5. The fact that seatbelts are not worn and vehicle speeds are often excessive in "urban" areas and that there are very few facilities to ensure pedestrian safety or to guide drivers through hazardous locations on the road network are major causes of these accidents and fatalities. The project will provide a bypass to two congested areas, Kobuleti and Batumi which, at present, incur traffic and pedestrian congestion, especially in the summer holiday season. 68. Recognizing the growing problems, the GoG has established a more effective traffic policing capability, will be developing a multi sector road safety strategy and develop and implement a multi sector road safety action plan based on the final report of consultants who have been undertaking a road safety study of Georgia during 2007 and early 20086. A road safety component has been included within the World Bank Second East West Highway Improvement Project, to strengthen the RD road safety unit, to implement a hazardous locations program along the East-West Highway corridor and to develop traffic and safety engineering standards that can be used on urban and rural roads of Georgia. A further component is proposed in the Third East-West Highway Improvement Project to develop road safety audit manuals. 69. A road safety TA is proposed for the project. Its aim is to educate the domestic road safety professionals, road users and pedestrians in safety procedures related to high speed access controlled roads. The Terms of reference (Appendix 13) set out the objectives,

3 Unpublished World Bank working paper 2004. 4 World Bank Appraisal Report 2nd E-W Highway Improvement Project 2008 5 e.g. Germany 14%, France 11%, Sweden 14%, Netherlands 8% 6 Sweroad consultants for PHRD February 2008

Engconsult Ltd. 15 scope, deliverables and timing for a project to provide the required technical assistance to achieve the aim above. It is envisaged that the work shall be done by a team of domestic and international consultants with counterparts from RD and the Patrol police (PP) and other Georgian professionals so as to ensure local capacity to provided ongoing support to the RD and PP. It is important that the consultants for the TA coordinate with the World Bank projects. 70. Road Development: The road program is expanding, with expenditure on construction, rehabilitation and maintenance planned to increase from GEL272 million in 2008 to GEL521 million in 2009, with an increase to GEL1 billion planned for 2010. 71. The Government began a systematic upgrading of subregional corridors in 2006. Progress to date and planned is detailed in Appendix 2. External support to the road subsector, principally in support of these projects, is being provided by World Bank, JICA and MCG, in addition to ADB. Upgrading of corridors north to the border with Russia is currently in abeyance, due to the political situation. Topographic and geological constraints limit the number of practical all-year round corridors. Shorter alternatives to existing routes all cross passes of over 2000m. In the long term, other corridors may be developed to serve subregional traffic, but at high cost. 72. Institutional Capacity. Despite some progress on policy and institutional reforms, institutional capacity in the road subsector remains limited. This is reflected by (i) lack of a legal and regulatory framework for managing the road network; (ii) weak enforcement of some existing regulations such as those relating to vehicle axle-load controls and road safety; (iii) lack of experience and knowledge about private-public partnership in road construction and maintenance; and (iv) limited staff experience in management of road projects, particularly in the area of tunnel and bridge engineering and environmental and resettlement safeguards. Although the WB has continuously supported necessary institutional strengthening, further external assistance is needed to help RD in implementing the planned large-scale investment program (Appendix 14). 2. Revenues and Expenditure 73. 74. Up to 2004 there was a road fund, financed from the proceeds of the fuel tax, the international transit tax and a road usage tax levied on individuals and enterprises. In 2004 the government abolished special funds, including the road fund, and since then the sub-sector has been funded directly from the state budget and from donor funds. The proceeds of the remaining direct taxes on transport (fuel tax, import duties and VAT on vehicle sales, vehicle registration fees and the Rikoti tunnel tolls) are part of state revenue and are not earmarked. 75. The RD budget for 2009 is GEL521m. This is financed as follows: GEL284m from the state budget (of which GEL102m are counterpart funds) and GEL237m from donor grants and credits. Routine maintenance accounts for GEL26m (GEL 3,800 per km). GEL119m from state funds is budgeted for periodic maintenance and rehabilitation. If donor funded rehabilitation projects are included, total funding for rehabilitation and periodic maintenance amounts to GEL150million, or GEL 22,000 per km. C. Stakeholder Participation and Consultations 76. Public participation is essential for every developmental activity and is a statutory requirement in Georgia. To ensure peoples’ participation in the planning phase and to treat public consultation and participation as a continuous two way process, public consultations were held during the initial screening and preliminary design stages at different places along the tranche 1 and 2 alignments in order to gather views regarding the need of the proposed roads in their vicinity.

Engconsult Ltd. 16

77. Extensive consultations were also held with government agencies involved in environmental protection, water and forest resources management during the preparation of the Project. Government officers in relevant ministries and departments, transport and commerce related organizations, and regional leaders were consulted during the feasibility and design studies and were apprised of the Project objectives and asked for comments. The results of the consultations were taken into consideration in the design of the tranche 1 and 2 roads. 78. During the social, environmental and resettlement assessment various stakeholders were consulted through informal focus group discussions and individual interviews which aimed to promote public understanding of developmental problems such as local needs of the road, problems of resettlement, etc. This therefore, ensured that the any affected population and other stakeholders were informed, consulted, and allowed to participate actively in the development process. Stake holder participation will be continued throughout the project, both during planning, preparation, and implementation and during post implementation of the project. 79. The people consulted during the reconnaissance and other surveys viewed that the road will facilitate the development in the zone of influence of the proposed road. Most people welcomed the proposed project. In the case of resettlement and land acquisition, people viewed that they should be properly compensated. 80. During the preliminary design stage of the project regular briefings were given to RD and further consultations held with stakeholders, including people in the zone of influence during the field surveys and with transport operators through survey and discussion. (Appendix 5).

Engconsult Ltd. 17

IV THE PROPOSED PROJECT

A. Project description 1. Project Area 81. The direct project area includes the autonomous region of Ajara, including the districts (rayons) and municipalities through which the existing highway and the proposed bypasses traverse. The total population of Ajara is 380,000 (2008) of whom 122,000 were resident in Batumi municipality who will benefit from reduced congestion and safer passage on the existing road as well as faster travel for through traffic using the bypass. 82. The indirect project area covers most districts between Batumi and Tbilisi and, to a smaller extent, most of the rest of Georgia (Population 4.4 million) as they will ultimately benefit from the lower transport costs of agriculture and manufactured goods traveling over the new bypass roads, particularly to and from Turkey. 2. Components 83. Tranche 1 and 2 of the MFF aim to improve transport reliability by providing a cost- effective investment program to improve traffic flow between Poti and Sarpi, with considerable savings in both travel time and accident reduction costs. 84. It involves building a 48- kilometer (km)-long high class road between Choloki River north of Kobuleti to Chorokhi River south of Batumi in Ajara province, including 6 km of widening of existing road to four lanes; providing consulting services and training; project monitoring and evaluation (M&E); and improving road safety. The first tranche of the project comprises 28km of new road (Kobuleti Bypass) and 6 km of road widening; the second tranche comprises 14km of new road (Batumi Bypass) B. Project Rationale 85. Ajara province, located in the southwest part of Georgia, is bordered by the Black Sea to the west and Turkey to the south. The two major towns Batumi and Kobuleti are major tourist centers, especially in the summer months. Batumi is also the second largest port in Georgia. Whilst the two towns cater for extensive tourist traffic the existing main north south road also carries heavy goods and passenger traffic. Turkey is Georgia’s most important trading partner and substantially all trade with Turkey passes along this road. Traffic growth was strong in 2007-8 (averaging around 20 percent a year) but has fallen by 18 percent in 2009 north of Batumi. Travel speeds are low and traffic accidents are higher than the national rate, which itself compares badly with nearby countries. The high incidence of traffic accidents, and the associated death and injury place a heavy burden on the health and welfare systems of the area. The Project will support (i) construction of bypass roads around the two major tourist areas; (ii) improve the international road transit network; (iii) improvement of road safety; (iv) strengthening institutions and (vi) promote activities to involve the private sector. C. Impact and Outcome 86. The overall Project will help promote sustainable economic growth in Georgia as a whole and the project area in particular and improve the economic development of the project area. It will create a better trade logistic corridor and, in this manner, enhance cooperation with neighboring countries. It will achieve this by (i) improving the road transport network by building bypasses of Kobuleti (tranche 1) and Batumi (tranche 2) allowing faster through transport of goods and people from Turkey and the project area to Tbilisi and other areas of Georgia, Armenia and Azerbaijan (ii) improving trade linkages with Turkey and improving the TRACECA network (iii) reducing transport operating costs in general and, in particular, time and fuel

Engconsult Ltd. 18 savings for both through traffic and traffic remaining on the existing road, and (iv) reducing accidents and fatalities on the existing road D. Outputs 87. The output of the first tranche of the Project comprises (i) the provision of about 28 km of two lane access-controlled highway from north of Kobuleti to north of Batumi, and upgrading and widening of 6 km of existing road to four lanes (Kobuleti Bypass) (ii) safeguard measures for the project road including environmental, land acquisition, resettlement and social as required by GoG and ADB, (iii) provide road safety and pavement damage improvements by awareness campaigns, safety design enhancements and axle weight control measures (iii) capacity development of RD in the fields of bridge/tunnel design/construction, pavement overloading, contract management and PPS (iv) detail design and construction supervision consultants for the bypass road, and (v) project support to the Implementing Agency (IA). 88. The output of the second tranche of the project comprises (i) the provision of about 14 km of two lane access-controlled highway providing a bypass around Batumi (ii) safeguard measures for the project road including environmental, land acquisition, resettlement and social as required by GoG and ADB, and (iii) construction supervision consultants for the Batumi bypass road. 1. Output for Tranche 1 89. Construction of Kobuleti Bypass: The forecast traffic volumes indicate that only a two lane road is required for the new Kobuleti bypass for the foreseeable future, however the design allows for future upgrading to four lanes. The proposed design standard is Georgian two lane standard for international roads. 90. Design speed will be generally 120 kph, with some mountainous sections dropping to 100 kph. Maximum grade is generally 4 percent. A total right of way of 50m follows the GoG standard, however in some mountainous road sections and at interchanges, this width has been increased. 91. The pavement structure thickness has been calculated based on the estimated cumulative number of heavy vehicle axles in the design life. Cross road drainage is through culverts and bridges. Road and design has followed GoG specifications, tunnel design has followed international standards. 92. Since the Project Road is located in a seismic affected area anti-seismic measures have been included in the design of cross drainage structures. The tranche 1 alignment has 18 bridges with a total length of 2,984m; one bridge is over 250m. 93. There is one tunnel in tranche 1 with a length of 400m. Whilst there are existing Georgia design standards for tunnels, these are considered not suitable for the smaller length tunnels and international standards have been recommended. Tunnel lighting will be required as will traffic safety and emergency facilities. A technical and design summary is set out in Appendix 4 and a full design report in Supplementary Appendix 2. 94. Road Safety Component: Tranche 1 will provide a road safety component to improve road safety on the new and existing roads and generally in Georgia. The component will include awareness campaigns, road safety equipment and road audits (Appendix 4 and 13). 95. Project Support and Capacity Building: Tranche 1 will provide support for (i) implementation of the Kobuleti bypass road, including detail design, construction supervision, project management, monitoring, and evaluation; (ii) procurement and installation of a weighbridge station along the bypass, portable weighbridges to conduct an axle load survey, and communication and monitoring equipment, (iii) preparation of detail design of tranche 2; and

Engconsult Ltd. 19

(iv) independent external financial and technical auditing. These components will also help build the capacity of RD. This non-investment component will be implemented primarily through consulting services. 96. The new road will involve some sophisticated and modern techniques both in design and construction. Some of the skills required are not available in Georgia at the present time. This will be particularly important in tunnel design, construction and operation, latest pavement design techniques and contract management and capacity building and training will be required in this area. Training will be carried out domestically by international specialists. Training will also be needed in developing an ITS strategy for high speed roads. 97. The EA has limited experience, resources and equipment for environmental and social/resettlement management and monitoring and would experience some problems to efficiently supervise monitoring of the environmental and social safeguard parameters for tranche 1 and future projects. There is, therefore, a requirement for training RD officials at the national and regional level in monitoring environmental and social/resettlement issues with the objectives of building up the capacity and procedures of the PMU to undertake analyses of environmental and social/resettlement impacts of road projects. 98. The Government of Georgia supports the involvement of the private sector in infrastructure. Steps to involve the private sector in road construction or rehabilitation or maintenance have so far been tentative, although the proposed performance based maintenance contracts (PBC) is a welcome development. The purpose of the proposed PPP TA as part of the tranche 1 project is to move RD a step further than PBCs and prepare them for deeper involvement by the private sector in road provision. Further details are in Appendix 11 and 14 and Supplementary Appendix 7. 2. Outputs for Tranche 2 99. Construction of Batumi Bypass: Batumi bypass will be constructed to similar design standards as Tranche 1. The alignment has 5 major bridges of 500m or longer, with a total length of bridges of 3,799 m. There are five tunnels, ranging from 90m to 730m, with a total length of 1,510m. 100. Project Support and Capacity Building: Tranche 2 will provide support for (i) implementation of the Batumi bypass road, including construction supervision, project management, monitoring, and evaluation; and (ii) independent external technical and financial auditing. E. Cost Estimates 101. The total project cost estimate and financing arrangements for both bypass roads is summarized in Table 2 and for tranche 1 in Table 3. Table 4 shows a disbursement schedule for tranche 1. Details are set out in Appendix 6 and Supplementary Appendix 3. The total cost of the project, including support services, land, capacity building and contingencies is estimated at $US325 million of which $119 million (37%) is foreign exchange cost. Local currency costs, including taxes and duties, are estimated at $US205 equivalent (63%). The total cost for Tranche 1 is $163 Million of which $56 million (34%) is foreign exchange cost. Local currency costs, including taxes and duties, are estimated at $US107 equivalent (66%).

Engconsult Ltd. 20

Table 2: Project Cost Estimate and Financing – Tranches 1 and 2 $US million (at forecast exchange rate) Cost Financed by: Item ($m) ADB Government A Base cost 1. Land acquisition & resettlement 18.4 0.0 18.4 2. Civil works 177.2 177.2 0.0 3. Design & supervision 10.8 10.8 0.0 4. Capacity development 1.8 1.8 0.0 5. Road safety enhancement 0.7 0.7 0.0 6. Project management support 1.4 1.4 0.0 7. Taxes & duties 37.2 0.0 37.2 Subtotal 247.5 192.0 55.5 B Contingencies 71.2 67.6 3.6 C Interest during construction 6.1 6.1 0.0 Total 324.8 265.7 59.1

Table 3: Project Cost Estimate and Financing – Tranche 1 Cost Financed by: Item ($m) ADB Government A Base cost 1. Land acquisition & 10.1 0.0 10.1

resettlement 2. Civil works 86.1 86.1 0.0 3. Design & supervision 7.8 7.8 0.0 4. Capacity development 1.8 1.8 0.0 5. Road safety enhancement 0.7 0.7 0.0 6. Project management support 1.4 1.4 0.0 7. Taxes & duties 18.5 0.0 18.5 Subtotal 126.4 97.9 28.5 B Contingencies 33.1 31.3 1.7 C Interest during construction 3.5 3.5 0.0 Total 163.0 132.7 30.3

Table 4: Disbursement Schedule – Tranche 1 Item 2009 2010 2011 2012 2013 Total A Base cost 1. Land acquisition & 6.0 4.0 0.0 0.0 0.0 10.1 resettlement 2. Civil works 0.0 10.1 28.5 28.1 19.4 86.1 3. Design & supervision 2.4 1.6 1.4 1.4 1.1 7.8 4. Capacity development 0.0 0.5 0.5 0.5 0.5 1.8 5. Road safety enhancement 0.0 0.2 0.2 0.2 0.2 0.7 6. Project management support 0.0 0.4 0.3 0.3 0.3 1.4 7. Taxes & duties 0.5 2.3 5.8 5.8 4.1 18.5 Subtotal 9.0 19.0 36.6 36.2 25.6 126.4 B Contingencies 1.2 3.8 9.1 10.7 8.2 33.1 C Interest during construction 0.1 0.3 0.7 1.1 1.3 3.5 Total 10.3 23.1 46.4 48.0 35.2 163.0

Engconsult Ltd. 21

F. Implementation Arrangements 102. The MRDI will be the executing agency (EA) and RD the implementing agency (IA). The chairman of RD will be responsible for overall supervision of the program. RD will establish a project management unit (PMU) headed by the deputy chairman of RD. It will comprise technical staff seconded from RD and TRRC (the existing PMU for World Bank roads projects). A steering committee will be established with representatives of the cabinet office, MRDI, RD, Transport Department (of behalf of MED), the Ministry of Finance and the PMU head. The steering committee will be chaired by a deputy minister of MRDI and will meet at least every six months to ensure interagency coordination, review implementation progress, and provide approvals and guidance as necessary. G. Implementation Period 103. Preliminary design was completed in July 2009. Detail design will start in September 2009. Other preconstruction activities, preparation of bidding documents, prequalification of contractors, and tendering and awarding of contracts, will start in September 2009. Completion of the preconstruction activities for two contracts of the first tranche is targeted for December 2009, with construction scheduled to start in March 2010. Completion of preconstruction activities for the remaining contract of the first tranche is scheduled for February 2010, with construction starting in April 2010. Completion of preconstruction activities for the second tranche contracts is scheduled for July 2010, with construction starting in October 2010. The first tranche contracts (Kobuleti bypass) will be opened to traffic by March 2013. The second tranche contracts (Batumi bypass) will be opened to traffic by January 2014. The implementation schedule is in Appendix 7.

Engconsult Ltd. 22

V PROJECT BENEFITS, IMPACTS AND RISKS

104. The Project will add improvement of the overall performance of the road sub-sector, contributing to transport efficiency, and widening access to economic opportunities. The beneficiaries of the Project will be communities served by the Project road who, although not necessarily traveling on the new road, will benefit indirectly from the improved, lower cost, and faster transport services, and thereby obtain improved access to economic opportunities and social services. Transport operators, both passenger and freight transport services, will benefit from the new road in terms of lower vehicle operating cost, time savings, and improved road safety, and thereby increase productivity. Government and non-government agencies using the Project road will improve their services delivery due to improved transport efficiency. Traders both sub-regional and regional will also benefit from reduced transport costs and this should result in lower costs to consumers in Ajara and Georgia generally. 105. Those benefits and positive impacts will occur based on assumptions that sufficient funding will be obtained, that Government will implement the necessary organizational structures and that the forecast traffic volumes on the new road take place. A. Benefits 1. Traffic Analysis 106. The existing road is congested and carries unusually high volumes of heavy goods traffic: approximately 300 articulated trucks – almost all of which carry Turkish license plates – use the road daily. Average annual traffic on the existing road is currently 6,700 veh/day, although car and minibus traffic (88 percent of the total) is 30-50 percent higher in mid-summer than it is in April or October. 107. The bypass will divert long distance domestic and transit traffic, not only reducing journey times and vehicle operating costs for diverting traffic but also improving the physical and safety environment for users of the existing road. It is estimated that approximately 42 percent of vehicles will divert. Opening base case traffic on the tranche 1 and 2 project bypass will be 3,500 veh/day and grow to 8,400 veh/day in 2033. If an alternative eastern corridor between Turkey and Georgia opens, opening bypass traffic will be reduced to 2,700 veh/day, rising to 6,400 veh/day in 2033. Opening base case traffic on the tranche 1 (Kobuleti) bypass will be 3,600 veh/day, rising to 8,500 veh/day in 2033. Traffic projections take account of (i) forecast economic growth in the zones used for trip generation and (ii) the additional traffic generated by the project through reduced transport costs. For further details see Appendix 9 and Supplementary Appendix 6. 2. Economic Analysis 108. The evaluation uses 2009 prices and a 12 percent discount rate. All predicted project costs and benefits are measured in economic prices using an international price numeraire and a US dollar unit of account. A standard conversion factor of 0.97 was calculated using the ADB’s simple trade-weighted formula. 109. Bypass benefits to “normal” traffic (i.e. traffic that would have been present in the do minimum case) result from improvements in road and traffic conditions between the do minimum and project case and amount to savings in (i) vehicle operating costs (VOCs), (ii) personal travel time for vehicle occupants (other than drivers and crew, whose costs are included in VOCs), (iii) road accident costs and (iv) the value tied up in freight during transit. In addition there are the benefits of additional trips made solely as a result of the lower costs on the existing road brought about by the project (“generated” traffic). The proposed project road is

Engconsult Ltd. 23 approximately 2.3km longer than the existing road. As a result (and as is typical for bypass investments), most of the benefits arise from time, rather than VOC, savings. 110. For tranche 1 alone the estimated EIRR with base case traffic is 16.2 percent. Its net present value (NPV) is USD40.6m. Of the total benefits, 83 percent are journey time and time in transit savings, 10 percent are VOC savings, 5 percent accident cost savings and 2 percent generated traffic benefits. Using case B traffic the EIRR falls to 13.3 percent, still above the 12 percent threshold. 111. The results of sensitivity analysis confirm the robustness of the project’s economic viability: the EIRR remains above 12 percent for all costs up to 1.5 times base and for benefits above 0.7 times base. A Monte-Carlo risk analysis was used to estimate the frequency distribution of EIRRs. A negative skew was applied to costs to reflect (i) the weak construction market in 2009 and (ii) the scope for optimization of the alignment. Negative skews were also applied to all benefits to reflect the scope for loss of some corridor traffic. The results of the risk analysis showed that with these uncertainties, (i) the mean EIRR is 15.4 percent and (ii) the probability of the EIRR falling below 12 percent is 3.4 percent. 112. The EIRR of tranches 1 and 2 combined is 14.8 percent (base case traffic). The results of sensitivity analysis show that the EIRR remains above 12 percent for all costs up to 1.3 times base and for benefits above 0.8 times base. A Monte-Carlo risk analysis using the same skewed distributions of costs and benefits as for tranche 1 gave a mean EIRR of 14.0 percent; 14 percent of results fell below 12 percent. For further details see Appendix 10 and Supplementary Appendix 6. 3. Financial Analysis and Tolling 113. It is recommended that the road is not tolled, at least in the short term, but that the possibility be kept under review and re-appraised once demand for the bypass has been established. 114. Financial evaluations assessed the tolls that would be needed, should tolling be implemented, to cover all the operation and maintenance costs (including overlays but excluding reconstruction or widening) and ADB interest expenses for a period of 20 years. It was assumed that the concessionaire would include a profit element in his price for undertaking maintenance works (rather than take equity returns, as in a “full” concession) and would operate a sinking fund in order to manage the mismatch between revenue and expenditure profiles. 115. Experience suggests that a toll at around one third to one half of perceived road user cost savings produces a good compromise between economic and financial returns. On this basis the implied charge for a car using the tranche 1 (Kobuleti) bypass should be no more than $1.40 and for the entire bypass $2.10. 116. Financial evaluation modeling showed that the levels of toll needed to recover O&M and ADB interest expenses were modest by comparison. Assuming no toll indexation, (i) the total charge for using the two lane section of the tranche 1 road (Kobuleti bypass) would be GEL0.8 for a car and GEL2.30 for an articulated truck and for using the two lane section of tranches 1 and 2 (Kobuleti and Batumi bypasses) would be GEL1.8 for a car and GEL5.5 for an articulated truck. For further details see Supplementary Appendix 7. B. Safeguard Impacts 1. Environmental Impacts 117. Environmental impact assessment (EIA) has been carried out for both tranches in accordance with ADB’s Environmental Policy (2002), ADB’s Environmental Guidelines (2003) and the Government of Georgia’s Laws on Environmental Impact Permit and Ecological

Engconsult Ltd. 24

Examination (2007). Environmental issues were considered for selection of the final alignment for both tranches to avoid impacts on the (i) Ispani mire, a Ramsar Convention on Wetlands (RAMSAR) wetland site of international importance, (ii) cultural sites, such as cemeteries, and (iii) soil erosion and landslides. The EIA concludes that there will be both negative and positive impacts due to the construction activities and normal operations of both tranches. The negative environmental impacts for both tranches are similar and are broadly classified into 3 categories during preconstruction, construction and operation. Preconstruction • Pedestrian and road accidents • Land acquisition issues for new right of way (RoW) areas • Compensation issues for other fixed assets from acquired RoW land areas Construction • Increased soil erosion, landslides and/or siltation from cut and fill operations, including increased risks to downstream rivers and coastal resources • Clearing of RoW, removal of vegetation (trees and shrubs) and disposal of spoils • Extraction and transport of construction material from existing quarry sites • Temporary use of land immediately adjacent to the road for siting of contractor’s yard, asphalt plant and construction camps • Reduced air quality and visibility (Air quality impacts and/or noise pollution from construction activities, quarry sites, material storage sites, temporary diversion roads, excavations, vehicle and equipment use and asphalt mixing plant) • Reduced water quality (Water and soil pollution) at bridge rehabilitation sites from improper handling of and disposal of wastes and materials • Drainage from construction camps, material stockpiles, excavations and quarry activities • Interruption to smooth traffic flow, increased traffic congestion and public safety problems • Social conflicts due to project activities Operation • Increased traffic volume and related air, noise and public safety concerns • Possible spills from transport of hazardous cargo

118. A detailed environmental management plan (EMP) has been prepared for all the identified impacts with adequate funds for monitoring and mitigation and will be included in the contractor’s bid documents. The major positive environmental impacts will include: (i) reduction of traffic congestion on the existing road in Kobuleti and Batumi and hence improved air and noise quality, and road safety, and (ii) reduced soil erosion and landslides due to slope stabilization measures along the proposed bypass roads. 119. A Summary EIA (SEIA) report has already been circulated to the Board and was approved for public disclosure through the ADB website (Appendix 20). A Cumulative Impact Assessment (CIA) of the Project was also prepared (Appendix 21). An Environmental Assessment Review Framework (EARF) was also prepared for future projects to be financed under further tranches of the MFF and all environmental assessments of the other projects in future tranches under this MFF will be carried in accordance with the EARF (Appendix 19). 120. The EIA has been submitted to Government and ADB and is presented as a Supplementary Appendix to this Report.

Engconsult Ltd. 25

2. Social and Poverty Impacts 121. Poverty and social analysis will be conducted for all projects financed under the MFF in accordance with ADB’s Handbook for Integrating Poverty Impact Assessment in the Economic Analysis of Projects and in accordance with the Handbook on Social Analysis (2007). A summary poverty reduction and social strategy for Tranche-1 is set out in Appendix 16. The project road will traverse through two regions, Ajara and , and will pass through four Rayons, Ozurgeti, Kobuleti, Batumi and Khelvachauri with a total population of 379,472. A socio-economic survey was carried out in the project area in 20 villages with 200 samples. The survey revealed that approximately 52 percent of the people live below the poverty line. 122. The direct benefit of the Project to the poor is relatively small; however, it will provide some direct employment opportunities during the construction phase. There are also indirect benefits. The provision of transport services, including the construction and maintenance of transport infrastructure, generates demand for labor (often unskilled labor) and provides income-earning opportunities for the poor. If a transport project generates jobs for the poor who are otherwise unemployed or under-employed, it contributes indirectly to poverty reduction. The construction aspect of the transport sector development is often viewed equally as important as the service aspect of the sector in promoting economic growth. The process of poverty reduction is embedded in a broad range of socio-economic activities to which transport services provide intermediate inputs. 123. There will be both short term and long term positive impacts of the proposed Project. The short term positive impacts include access to markets for agricultural produce, availability of temporary unskilled jobs for villagers; and opportunity to provide food and restaurant services for construction workers. The long term positive impacts include access to long distance transport services with Turkey; access to long distance markets (buying and selling through regional trade); easier access to health facilities and medical treatment; increased access to agricultural extension services, increased access to education; increased opportunity for the development of tourism; and diversification of income sources. 124. Women have important economic roles in project areas and engage in a very wide range of income making activities in the agricultural and marketing sector. The Project will pay particular attention to ensure that women are the recipients of the compensation pertaining to their activities and to ensure that women who are de-facto household heads are clearly listed as beneficiaries of compensation and rehabilitation proceedings under the loan. Women will also be participants in the consultation processes to determine and negotiate compensation entitlements and implement the LARP. Special attention will be given to the impact of resettlement on women and other vulnerable groups during monitoring and evaluation of the LARP. The Project will have a positive impact on gender, because the civil works contracts will include provisions to encourage employment of women during implementation, and women will be encouraged to participate in activities to monitor program impacts. 125. All projects financed under the MFF will include appropriate measures to mitigate the potential risk of HIV/AIDS and other sexually transmitted infections, as well as drug and human trafficking including raising public awareness of the risks. Civil works contracts will include provisions requiring contractors to take measures to protect construction workers from the risks of HIV/AIDS and other sexually transmitted infections, and giving construction workers time to be tested or receive treatment. 126. No impacts on Indigenous Peoples are expected from the Project. An assessment of impact on indigenous peoples was undertaken in accordance with ADB’s Policy on Indigenous Peoples (1998). The Project will primarily affect Georgian people which are almost 93% of the total population; other ethnic groups are also present in the project area but are insignificant in

Engconsult Ltd. 26 number. These groups are mostly Armenian, Azeri, and Russian who have been fully integrated into institutional, cultural, and economic processes in the country, and they do not display sufficient features to classify them as indigenous peoples as per the ADB’s definition on IP. Therefore, the Project is classified category C. C. Resettlement Impacts 127. The impact on land acquisition and resettlement will be assessed for all projects financed under the MFF in accordance with the Land Acquisition and Resettlement Framework (LARF) prepared by the Government and agreed with ADB in compliance with ADB’s Involuntary Resettlement Policy (1995) and relevant laws and regulations in Georgia. Substantial land acquisition and resettlement impacts are expected for the tranche 1 and 2 projects which involve construction of new bypass roads. The LARF specifies compensation provisions and procedures on preparation and implementation of LARPs for the projects financed under the MFF. The summary LARF is provided in Appendix 17 and the detailed LARF is presented as Supplementary Appendix 8. 128. The tranche 1 component will entail substantial land acquisition and resettlement; hence it is categorized as “A” based on ADB’s definition on ADB’s Involuntary Resettlement. A detailed impact assessment including the census survey has been completed and is in progress of being analyzed which will give exact figures related to land acquisition and resettlement. However, a preliminary assessment shows that approximately, 163.4 hectares (Ha) of land will be required for land acquisition. A total of 894 land parcels will be affected. Additionally 154 structures/buildings will be affected. Based on preliminary estimate the total number of affected household (AH) is 881 and the total number of affected persons (AP) is 4,440. A full LARP is provided as a Supplementary Appendix to this Report. The summary LARP is provided in Appendix 18. The LARF and the LARP for Tranche 1 have been endorsed by the Executing Agency (RD) and will be disclosed to the affected people. The LARP has been submitted to ADB for review and the final LARP as agreed by the EA and ADB will be fully implemented prior to start of the civil work. 129. The LARPs for future tranches with land acquisition and resettlement impacts will be prepared in accordance with the LARF, and will be submitted to ADB for approval and fully implemented before the awards of concerned civil works contracts. Appraisal of the subsequent projects financed under the MFF will require a review of the LARF and preparing respective LARPs for those projects with land acquisition and resettlement impacts. Implementation of the LARPs will be a condition for commencement of civil works under the relevant projects. D. Project Risks 130. The Project is being carefully formulated to mitigate risks, however the construction of the project road will be in difficult terrain and operation of the road will require new techniques. Possible risks are set out below. 131. The project is being designed to reduce potential technical, economic, financial and social risks. The main technical risks are associated with the design and construction of the six tunnels and six major bridges (Tranches I and II). To minimize risks, international consultant will be included in the consulting services to review the design and contractors construction methodology of the bridges and tunnels. There is still a risk however, that further detail geological investigation will reveal adverse geological conditions which may increase cost and/or change the alignment. 132. There have been problems on major roads in Georgia with pavement damage from overloaded vehicles and there is at present little enforcement. The preliminary pavement design has taken account of this overloading. To further minimize risk an axle load survey using

Engconsult Ltd. 27 portable weighbridges is proposed under tranche 1 of the project to assess the extent of the problem. The introduction of a permanent weigh bridge on the Kobuleti bypass for freight traffic will help to reduce pavement overloading. 133. To minimize construction risks the prequalification of contractors will focus on their financial and technical capabilities in handling similar works. During the construction period, monitoring and contract management information systems will be set up and implemented, with the assistance of the international supervision consultants, for timely identification of technical problems and implementation of corrective measures. The proposed PMU will be supported by the World Bank’s PMU, the TRRC (which has had over ten years’ experience in the financial management and procurement aspects of internationally financed projects), strengthened by additional technical and financial auditing support. These arrangements will minimize contract management risks. 134. The economic and financial risk associated with the construction and operation of the road will depend on the actual level and composition of traffic and the construction costs. Traffic forecasts in the TA have been conservative and sensitivity has been tested to the loss of some long distance domestic and transit traffic to a proposed new border crossing. 135. Social risks are being identified and appropriate mitigation measures have been put in place in the Resettlement Plan and the EMP. 136. A risk of delay to the project lies with the unsettled nature of Georgia’s political situation at present. It is hoped that this risk will be minimized before the end of 2009. Government decisions, involving revised legislation regarding land registration will be required to be set in place in order that land acquisition and resettlement can be completed as scheduled.

Engconsult Ltd. Appendix 1 28

Appendix 1 DESIGN AND MONITORING FRAMEWORK (Tranche 1)

Design Performance Targets/ Indicators Data Assumptions And Risks Summary Sources/Reporting Mechanism Impact Assumptions Increased Increase in subregional trade Georgia Statistic Political and subregional through the Turkish border by 4% Yearbook macroeconomic stability trade and for 5 years after completion Region and National Implementation of the tourism revenue 30% increase in cross-border Statistic Data Transport Infrastructure freight traffic by 2015 from 1.0m ton Investment Program as in 2008 to 1.3 million ton. planned Increase in registered visitors to Ajara region by 10% per annum for 5 years after completion from 285,000 in 2008.

Outcome Assumptions Improved Increase in traffic volume for Roads Department Government's subregional Kobuleti bypass to 3,600 vehicles (RD) surveys commitment to implement road network in 2014, to 5,700 vehicles in 2024, the roadmap and the and 8,500 vehicles in 2034 Subregional Road Corridors Development Decrease in travel time from Program Choloki bridge to Chorki bridge to less than 0.6 hours after project Adequate financing for completion implementation of the Transport Infrastructure Decrease in road accident rate Investment Program from 1.3 per million veh-km on the Accident statistics existing road to 0.68 on the from the Georgian Risks bypass road in 2014. Police Rapid deterioration of the road network caused by insufficient funding for maintenance and weak axle-load control

Limited coordination among government agencies during implementation of the Subregional Road Corridors Development Program

Design Performance Targets/ Indicators Data Assumptions And Risks Summary Sources/Reporting Mechanism Outputs A 28km 2-lane IRI value for the bypass roads RD surveys Assumptions bypass road maintained at no higher than 3.0 Support from local detouring within 5 years after project governments and

Engconsult Ltd. Appendix 1 29

Design Performance Targets/ Indicators Data Assumptions And Risks Summary Sources/Reporting Mechanism Kobuleti and completion Supervision communities 6km widening of consultant's report Disputes among existing road Increased road capacity by 12,000 contractors, supervision vehicles per day after project consultants, and RD completion avoided or promptly resolved Reduced road user costs for a car from $0.42 /km to $0.32/km after Risks project completion Cost overruns caused by unexpected geological Domestic training for RD staff conditions or unexpected Road design increases in prices of fuel and and key raw materials construction and safeguard capabilities Implementation of performance strengthened. maintenance contracts

Asset Axle load equipment procured and management axle load survey completed on capacity existing roads by August 2010 strengthened Road safety publicity campaign and training provided by project Road safety with completion by 2011 improved Adoption of an intelligent traffic system strategy by project completion Activities with Milestones 1. Implementation Civil works contracts to be awarded for the 1st tranche Inputs of civil works roads by June 2010 and the construction work to be started on 1 August 2010 and be completed by • ADB: $108 million from October 2013. Asian Development Fund nd Civil works contracts to be awarded for the 2 tranche • Georgia: roads by December 2010 and the construction work to be started on 1 January 2011 and be completed by December 2013. 2. Installation of Equipment procured and installed by December 2011. equipment 3. Implementation Land acquisition and resettlement plan for the 1st of land acquisition tranche road to be completed by April 2010 and resettlement Plan Land acquisition and resettlement plan for the 2nd tranche road to be completed by June 2010 4. Implementation Consultants to be recruited by December 2009 and TA of road safety completed by August 2010 improvement and awareness campaign 5. Completion of Consultants to be recruited by January 2010 and capacity building for completed by August 2010

Engconsult Ltd. Appendix 1 30

Design Performance Targets/ Indicators Data Assumptions And Risks Summary Sources/Reporting Mechanism RD and other professionals 6. Providing Design consultants to be recruited by August 2009 and detailed design and detailed design to be completed by 31 December 2009 supervision for one contract of the 1st tranche roads, March 2010 consultancy for the remaining contract of the 1st tranche roads and by July 2010 for the 2nd tranche roads

Supervision consultants to be recruited by March 2010 and completed by December 2013

Engconsult Ltd. Appendix 2 31

Appendix 2 TRANSPORT SECTOR REVIEW

A. Economic Overview 1. Georgia borders Turkey, Armenia, Azerbaijan and Russia and forms the lynchpin in the Europe-Caucasus-Asia transport network. It is a small country, located to the south of the main Caucasus mountain range. It has a population of 4.4 million. Following independence in 1991, output declined by 70 percent, with the loss of Soviet markets, the end of large budget transfers from Moscow, and the impact of civil war and ethnic conflicts, which displaced some 300,000 people. By 2000, recorded GDP had fallen to only 30-35 percent of its 1989 level, although there is a (large) shadow economy, estimated to be up to 33 percent of GDP. Per capita income is at the lower end of that of lower middle-income countries. 2. Although a range of political and economic reforms were implemented following independence in 1991, the situation quickly worsened thereafter. By the early 2000’s, power had fragmented among competing groups, law and order had deteriorated, corruption was widespread, public salaries, pensions, and social transfers were in arrears, and the political will for reforms had slackened. The Rose Revolution in late 2003 brought into office a government of reformers led by President Saakashvili. The administration has been implementing bold reforms to fight corruption, reduce the burden of the state on the economy, move the energy sector towards efficiency and sustainability, lay down the basis for improved employment and standards of living and develop a fiscally sustainable social safety net. The reform orientation is strongly pro-market. 3. Economic performance after the Rose Revolution and until the war with Russia in August 2008 was encouraging, with strong growth, macroeconomic stability and a sound mix of fiscal and monetary policies. In the public sector, salaries and pensions are paid on time and arrears have been cleared. Public employee salaries have increased significantly. Salaries in the Roads Department (RD) have increased substantially and are now more or less on a par with the salary level in the private sector. The regulatory and administrative environment facing business has improved significantly. The World Bank’s 2006 and 2007 Doing Business Reports rated Georgia among the top reformers. In the 2008 Report Georgia moved to 18th place in the rankings. Various indicators show a marked decrease in corruption. Investment in infrastructure has been substantial, particularly for roads and energy, with major improvements in access to reliable electricity services. 4. While economic growth has generated new jobs, it has not yet been sufficient to offset labor shedding and therefore net unemployment has not reduced. Resuming growth in private investment is therefore very important. Other challenges include the need to pay greater attention to institutionalizing reforms to ensure their sustainability, protection of property rights, progress in judicial reforms and strengthening of the rule of law. 5. Georgia enjoys a strategic location it has yet to fully capitalize on. It is on the shortest route between Europe and Azerbaijan, Armenia and the Central Asian Republics, through its Black Sea ports, Poti and Batumi. It also can link Russia and Turkey. The physical location ensures that it is a key transport link on the most direct route between the Black Sea and the Caspian Sea and Central Asia, the historic “Silk Road”, which the current TRACECA initiative seeks to emulate. Trade with neighboring countries, both transit and bilateral, is thus an important feature of the economy. Wholesale and retail trade services are the second largest sector. Georgia is developing closer economic cooperation with Turkey and Azerbaijan. In February 2007, the Tbilisi Declaration on Common Vision for Regional Cooperation was signed between Azerbaijan, Turkey and Georgia aiming at developing long-term and predictable

Engconsult Ltd. Appendix 2 32

relations on the basis of regional projects to establish energy and infrastructure links. Turkey, Azerbaijan, and Georgia signed an Agreement in March 2007 to construct a railway Kars-Tbilisi- Baku, including a new 105-km section, 76 km in Turkey and 29 km in Georgia. Increasing revenue from oil exports in Central Asia and the Caucasus is likely to increase the demand for consumer and industrial goods. 6. A range of natural endowments offer strong potential for labor-intensive export oriented growth, as well as tourism activities, provided the provision of adequate infrastructure enables them to develop, which would increase employment and broaden asset-ownership, Georgia has an open trade regime. The country benefits from an educated, inexpensive labor force and it also has a long tradition of entrepreneurship. Fertile land and favorable climate enables the production of diverse agricultural products. Georgia’s scenic mountain regions, the Black Sea beaches, and its rich historical and cultural heritage offer strong tourist potential. With improving law and order, the prospect of rising incomes in the region, and greater interregional connectivity through the transit corridor, Georgia has an opportunity to revive its tourist industry. 7. Despite the successes of the Rose Revolution, Georgia is still in the process of building a modern state. GDP is not yet back to its pre-independence level. Agriculture, manufacturing and tourism, the major sources of exports and employment in the Soviet period, are only making modest recoveries. The legacy of 70 years of command economy is challenging to overcome. Georgia’s relationship with Russia remains strained, although reopening of the land border is under consideration. This has repercussions for the export of agricultural goods, especially wine, the import of gas and power and the ability of people to travel between the two countries. 8. Reducing poverty is an important challenge. Strong economic growth 2003-2008 increased incomes for many in the middle and upper ranges of the income distribution. Although the bottom 30 percent of the population has not yet experienced similar increases, important non-income dimensions of poverty have improved, including significantly improved access for the poor to electricity, natural gas, safe water, health and higher education. One of the key structural factors contributing to high poverty level is the slow economic restructuring since independence. Output contraction contributed to a major reallocation of the labor force. Agriculture became the “employer of last resort” with the share of employment in the sector more than doubling between 1992-2004. In addition, plot sizes shrank, leading to declining productivity. Reducing poverty is a key priority of the Government’s reform program. Social spending was increased by 5 percent of GDP 2003-2005, pension and wage arrears were eliminated and a targeted poverty benefit was introduced in 2006. Poverty levels remain high in rural areas, where incomes of those below the poverty line have not yet improved significantly. 9. Georgia has a comparative advantage in export-oriented agro-processing areas, such as wine, hazelnuts and fruit and vegetable processing, which are significant potential sources of employment, as well as market growth for farm products. Dense forests cover one third of the country and good potential for labor intensive export-oriented wood processing exists, if a sustainable forestry regime can be developed. Georgia also has numerous mineral resources that promise some employment creation. Integration of the rural population with the national economy and a reduction over time of over employment in agricultural activities (52 percent of employment for 16 percent of GDP) should help reduce rural poverty in the years ahead. Although the conditions necessary for poverty reduction are in place, a sustained trend toward lower poverty has not yet been established. B. Transport Overview 10. Poor infrastructure and cumbersome processes inhibit the full exploitation of transit economy potential. Traffic flows from the Caspian Sea, primarily crude oil and oil products, are rapidly expanding. The main flow in the corridor is unidirectional, consisting primarily of oil and

Engconsult Ltd. Appendix 2 33

oil products, moving west to Poti and Batumi. Non-oil traffic mostly comprises Armenian and Azeri trade goods, which account for a significant share of the turnover of Georgia’s ports. Georgia has not yet been able to realize the transit potential from Central Asian countries. Rail, road and port infrastructure require significant investments to rehabilitate and modernize an aging infrastructure, after years of maintenance neglect. Despite being the closest maritime gateway to nearly all Central Asian countries, the Georgian route presently accounts for a negligible portion of the foreign trade of these countries. Long transit times, poor road conditions, multiple borders to be crossed and the number of times cargoes need to be handled, together with formal and informal payments and delays, explain the limited use of the Caucasus route. The significant reduction in corruption should improve Georgia’s transit competitiveness. 11. Limited transport infrastructure add to the cost of doing business, deter foreign investment, add time and costs to the transit corridor, and leave large segments of the population out of mainstream economic activities. Analyses show a close correlation between poverty in rural households and the extent to which they are linked to markets. Poor transport infrastructure partly explains why almost two thirds of rural household agricultural production is for self-subsistence and that about 20 percent of rural households do not trade at all. A World Bank Rural Infrastructure survey revealed that only in 5 percent of surveyed communities were roads repaired within the last five years, while in 41 percent of communities they were last repaired over 15 years ago. Poor connectivity has also contributed to weak linkages between farmers and agro-processors, as few intermediate agents, wholesalers, or farmer-based organizations exist. 12. The Government’s transport strategy is to support market integration, to develop the infrastructure and institutional setting and to maximize the country’s potential as a transit economy. Implementation progress has been impressive. The main ports and airports have been concessioned to the private sector; the railways are developing profit centers and the adoption of modern marketing techniques; customs has been reformed, to improve trade facilitation and reduce corruption; local infrastructure has been devolved to local governments; and a massive effort is being made, with IFI support, to upgrade the road network and reduce travel costs. The sector agenda includes major challenges, including: (i) overhauling the legal and regulatory framework to incorporate the right balance in the private:public sector interface, ensuring fair competition, safety and attention to environmental concerns; (ii) strengthening institutions to fulfill their policy making, monitoring and executing roles; (iii) careful implementation of decentralization to ensure that local authorities acquire the technical and financial resources necessary to deal with their new obligations; (iv) ensuring capabilities for efficient road management, based on modern information systems; and (v) capitalizing on the benefits of a transit economy, and the ability to compete with alternative routes, requiring a concerted public private effort to design and implement a vision that significantly transcends the transport sector. 13. The transport infrastructure consists of roads, railways, ports, airports and pipelines. Road transport accounted for 89 percent of domestic passenger km in 2008 and for 8 percent of freight ton km (Table A2. 1). The transport sector was one of the fastest growing in the economy 2003-2008. Transport, storage and communications have substantially increased their contribution to the total output, from 4.6 percent of GDP in 1996 (GEL881 million) to over 10 percent in 2007 (GEL1,231 million). 14. The transit of oil through recently completed pipelines and the railways dominates the transport contribution to GDP. In physical terms the movement of goods by land modes has increased by more than 50 percent 2000-2005 from 4.3 billion ton-km to 6.7 billion, with the bulk of the change coming from an almost doubling of oil transit traffic by rail. While the railways dominate the movement of oil, the roads are the preferred mode for passengers and non-oil

Engconsult Ltd. Appendix 2 34

freight. Although the railways have the higher modal share in ton-km terms (89 percent), a greater total tonnage is hauled by road transport. The average length of haul by road is less than a tenth of that by rail, reflecting the dominance of the railways in oil transport, which accounts for more than three-quarters of Georgia’s trade. Total land transport movement amounts to only about one-third of the levels in 1990, a reflection of the civil war in 1991/92 and the change in the pattern of trade and production since the breakup of the former .

Table A2. 1: Freight and Passenger Transport by Mode Transport Mode 2000 2006 2007 2008 Freight Transport (million ton-km) Railways 3,912 7,393 6,928 6,516 Roads 475 586 594 601 Maritime 612 131 120 45 Air 3 4 4 2 Total 5,002 8,114 7,646 7,164 Passenger Transport (million passenger-km) Railways 453 809 774 675 Roads 4,500 5,269 5,417 5,568 Air 238 453 475 486 Total 5,191 6,531 6,666 6,729 km = kilometer. Source: Statistical Yearbook of Georgia 2008. Ministry of Economic Development.

1. Road Infrastructure 15. Georgia’s road network is crucial in facilitating subregional transport and trade. The east–west highway forms part of the Asian highway network connecting Georgia and Turkey to Azerbaijan and Central Asia. Georgia is a signatory to a number of international transport agreements and conventions. Continued development of an integrated and efficient road network will facilitate cross-border and transit traffic and contribute to subregional cooperation and integration. 16. The road network was primarily developed under the former Soviet Union and does not adequately cater for the current needs of the economy and of the subregion. There are only 40km of 4-lane highway and many roads remain in poor condition. Funding for secondary and local road maintenance remains inadequate. A large part of the network deteriorated rapidly after independence and became non-maintainable, resulting in high transport costs and an increase in traffic accidents, thus constituting an impediment to growth and poverty reduction. The road subsector needs to be developed to facilitate subregional trade, investment and tourism. 17. The length of the road network in 2008 was 20,329 km (Table A2. 2). Under the Organic Law on LSG Bodies, the responsibility for all local roads was transferred to LSG units effective 1 January 1 2007, after local elections in October 2006 and the implementation of territorial- administrative reform. Presidential Decree No. 554 provided that some 2,000 km of local road was reclassified as secondary road. The international and interstate roads under RD jurisdiction account for one third of the length. 18. The road network was primarily developed under the former Soviet Union. Because of insufficient funding for road maintenance following independence, about 24 percent of the RD secondary road network remained in poor condition in 2008, together with most of the local road network. The international roughness index (IRI) on subregional corridor roads is: 66 percent

Engconsult Ltd. Appendix 2 35

under IRI of 4, 28 percent an IRI of 4-6 and 6 percent an IRI above 6. Many secondary and local roads require reconstruction or rehabilitation. The poor condition of these roads results in high transport costs and low safety levels. Technical standards and specifications for road design, construction, and maintenance mostly dated from the Soviet era, with rigid criteria for upgrading road capacity based on traffic volumes, without considering level of service or economic viability. The inappropriateness of such standards for a market-based economy was widely recognized and new norms, based on European standards, have been introduced in 2009. 19. Cities remain responsible for urban roads. The main road network was 6,941km at the end-2008, of which 347km are within Ajara. The network comprises 1,495km of international roads (cross-border links), 5,446km of secondary roads (connecting major cities or regional centers) and 13,388km of local roads (connecting villages and regional centers), of which 70 per cent are unpaved and in very poor condition The local road network is approximately 13,288km (Ajara 965km). The length of both the main and local networks has changed little since 2000, the investment emphasis having been on rehabilitation, rather than network extension. All but 55km of the 1,495km of international roads are paved, but in the case of interstate roads the percentage falls to 61, while only 23 per cent of local roads are paved. The paved proportion of the national network has hardly changed since 2000. National and Ajaran road densities are 290km and 160km per 1,000km2 respectively. Densities are low by regional standards: Azerbaijan has 690km/1,000km2 (both paved and unpaved roads) and Armenia 270km/1,000km2 (but paved roads only). Table A2. 2: Road Network 2008 (in km)

Area State Road Local Total Per cent of International Interstate Road Total Ajara 46 301 965 1,312 6 Other areas* 1,449 5,145 12,423 19,017 94 Total 1,495 5,446 13,388 20,329 100 % 7 27 66 100 Source: Roads Department. Note: * Including Abkhazia and . 20. The 10 major corridors (in total 1309km) are listed in Table A2. 3. The principal subregional corridor is the East-West Highway (E-60) and the adjoining E-70, which connects Europe and Asia. The E-60 runs north from the Turkish border at Sarpi, via the Black Sea port of Batumi to the port of Poti, then the E-70 runs east via the second largest city to Tbilisi, and south-east to the border with Azerbaijan at Red Bridge. Three corridors run south from Tbilisi to the Armenian border at Sadakhlo, Guguti and south of Ninotsminda. Table A2. 3: Major Road Corridors Route Section Length Remarks (km) Principal east-west highway. Currently Tbilisi–Senaki- S-1 (E-60) 288 (510) closed to through traffic at Abkhazia (Lesleidze) border. S-2 (E-70) Senaki-Poti-Batumi-Sarpi 105 Black Sea coast highway. S-3 Mtskheta-Kazbegi-Larsi 139 Mtskheta to Russian border Continuation of S-1 to Azerbaijan S-4 (E-60) Tbilisi-Red Bridge 46 border Tbilisi-- Trunk highway east from Tbilisi towards S-5 140 Lagodekhi Azerbaijan border Secondary international route to S-6 Tbilisi-Marmeuli-Guguti 94 Armenian border at Guguti S-7 Marmeuli-Sadakhlo 34 Main international route to Armenian

Engconsult Ltd. Appendix 2 36

Route Section Length Remarks (km) border Secondary international route to Vale S-8 Khashuri-Vale 103 (Turkish border) S-9 Tbilisi Bypass 49 Eastern bypass of Tbilisi Gori--Gupta- Gori on S-1 to Roki Pass (Russian S-10 89 -Roki Border) in South Ossetia

2. Vehicle Fleet 21. Vehicle fleet statistics are of poor quality and overstate the number of vehicles in regular use. The total fleet in 2008 comprised 613,000 vehicles, of which 543,000 were passenger vehicles, 58,000 goods vehicles and 12,000 other vehicles (Table A2. 4). Average annual growth from 2000 to 2008 was 12 percent. Vehicle ownership per 1,000 population increased from 56 in 2000 to 140 in 2008. The level of car ownership is broadly consistent with that of neighboring countries, given differences in per capita GDP. The increase in vehicle ownership is attributable to sustained economic growth (GDP per head grew by an average annual rate of 7.9 percent 2001 - 2008), helped by tariff reductions following accession to the World Trade Organization in 1999. The fleet has been modernized to some extent in the process, with Soviet era vehicles replaced by western and Japanese imports, improving vehicle safety and reducing individual vehicles’ fuel consumption and their emissions. Most private vehicles are 10 to 20 years old and many people cannot afford to replace or properly maintain them. The vehicle fleet still includes many Soviet era cars, which generally consume more fuel and produce more emissions. Many buses are also in poor condition. 22. From 2005/6 annual licensing of vehicles ceased and was replaced by a requirement to register only at a change of ownership. Vehicle registration and driver licensing is handled by the police. There are no longer any vehicle roadworthiness tests and owners are not obliged to carry even third party insurance. (The abolition of annual licensing accounts for the unreliability of vehicle ownership statistics: there is no longer any mechanism for determining that a vehicle is no longer used on the road). Table A2. 4: Vehicle Fleet (in ‘000) Vehicle Type 2000 2006 2007 2008 Cars 202.0 387.9 426.9 501.2 Buses 18.2 38.5 39.2 41.7 Trucks 25.6 58.2 49.6 58.4 Other 4.5 8.9 8.7 12.0 Total 250.2 493.5 524.3 613.3 Source: Ministry of Internal Affairs. 3. Institutional Structure 23. Institutional arrangements ( Figure A2. 1) changed in February 2009 when a new ministry was created, the Ministry of Regional Development and Infrastructure (MRDI). RD and the United Transport Administration (UTA) were transferred from the Ministry of Economic Development (MED) to the new ministry. RD is solely responsible for construction and maintenance of the main road network, all of which is contracted out to private sector companies. Strategy and planning are handled by the Transport Department (whose line ministry is the MED). UTA is primarily responsible for international aspects of road haulage, aviation and maritime transport.

Engconsult Ltd. Appendix 2 37

Ministry of Economic Ministry of Regional Development (MED) Development and Infrastructure (MRDI)

Transport Department Roads Department TRRC UTA • roads strategy & • road construction, • PIU for World Bank • international policy maintenance etc projects haulage agreements etc

Figure A2. 1: Roads Department – Institutional Context

24. RD has 160 staff, nearly all based in Tbilisi. The organization chart is given in Figure A2. 2Error! Reference source not found.. There are five small regional offices: in Sagarejo, responsible for the regions of , , and Mtkheta Mtianeti; in for the regions of and Samcke Javakheti; in Kutaisi for the regions of Racha Lechkhum, Kvemo Svaneti and ; and in Poti for the regions of Guria and Samegrelo Zemo Svaneti. The fifth regional office is for the Autonomous Republic of . No formal organizational structure has been established For local roads; most local authorities lack technical staff, equipment and a secure source of funding for road maintenance and rehabilitation. 25. Responsibility for road construction and maintenance is split between the Roads Department (RD), districts and cities. RD is responsible for the main road network, comprising international roads (i.e. those used by international transit traffic) and interstate roads. 26. Intercity bus transport is completely deregulated and there is no official control of prices. The same applies to domestic freight transport. 27. Institutional capacity in the road subsector needs further strengthening to successfully implement an expanding road program. The World Bank is continuing to support necessary institutional strengthening and ADB will work in parallel to enhance resourcing for these initiatives.

Engconsult Ltd. Appendix 2

MRDI

Chairman Engconsult Ltd. Engconsult Ltd.

First deputy chairman

Financial management Legal department Administration • accounts • legal drafting (road law • HR • budgeting amendment etc) • office management • tolls • RD legal representation • legal compliance checks

Deputy chairman

Technical policy Rehabilitation • design • planning & commissioning • road standards of rehabilitation • road condition survey • R&D

Deputy chairman Deputy chairman Regional offices: Ajara Development and resettlement Sagarejo Periodic maintenance Roads operations • annual construction plan Borjomi • planning and • co-ordination with local • resettlement planning Kutaisi commissioning of routine government • bridge condition survey Poti maintenance • police liaison (blackspots,

• reports on emergencies diversions) • signing

Figure A2. 2: Roads Department Organization Chart 38

Appendix 2 39

4. Road Investment Financing 28. Consolidated budget revenues increased by more than ten percentage points of GDP between 2003 and 2006. Tax revenues increased substantially, partly owing to the sweeping tax reform adopted in 2005, but also due to the crackdown on smuggling and tax evasion since the Rose Revolution. The government acknowledges that sustaining and accelerating economic growth will require deepening integration with world markets. To that end, the Government has identified as a key priority the modernization and improvement of transport infrastructure. Reforms implemented since 2004, along with investments in infrastructure and improvements in the regulatory environment, are supporting private sector growth. In the longer term, expected sources of growth include agro-processing and small manufacturing, tourism, transit trade activities and mineral extraction and processing. 29. Up to 2004 there was a road fund, financed from the proceeds of the fuel tax, the international transit tax and a road usage tax levied on individuals and enterprises. In 2004 the government abolished special funds, including the road fund, and since then the sub-sector has been funded directly from the state budget and from donor funds. The proceeds of the remaining direct taxes on transport (fuel tax, import duties and VAT on vehicle sales, vehicle registration fees and the Rikoti tunnel tolls) are part of state revenue and are not earmarked. 30. The RD budget for 2009 is GEL521m (Table A2. 5). This is financed: GEL284m from the state budget (of which GEL102m are counterpart funds) and GEL237m from donor grants and credits. Routine maintenance accounts for GEL26m (GEL3,800 per km). GEL119m from state funds is budgeted for periodic maintenance and rehabilitation. If donor funded rehabilitation projects are included, total funding for rehabilitation and periodic maintenance amounts to GEL150million, or GEL22,000 per km. Table A2. 5 :RD Expenditure and Financing 2006-2009 2006a 2007a 2008a 2009b GEL million Staff expenses 1.2 1.2 2.0 2.2 New road construction: State budget N/A N/A 13.7 13.7 TRRC N/A N/A 53.7 205.7 Subtotal 52.6 74.0 67.4 219.4 Rehabilitation: State budget N/A N/A 122.5 121.5 TRRC N/A N/A 5.3 80.0 MCGc 0.0 0.0 34.5 50.2 Subtotal 93.9 165.4 162.3 251.7 Routine maintenance 22.0 25.5 25.5 25.8 Miscellaneous (including bridge 29.8 10.8 15.2 21.4 works) Grand total 199.5 276.9 272.4 521 Financed by: Donor and counterpartd funds N/A 63 102 339 State budget N/A 214 170 182 Source: Roads Department and Transport Department Note: (a) actual expenditure for fiscal year – 1 Jan to 31 Dec. All figures include VAT where applicable (b) budget forecast (c) the MCG compact became effective April 2006, but MCG accounts show no roads expenditure until 2008. The only MCG roads project is rehabilitation of the Samtskhe-Javakheti road (d) counterpart funds are from state budget sources

Engconsult Ltd. Appendix 2 40

5. Road Maintenance 31. Since 2004, the Government has invested some GEL900 million in road construction, rehabilitation and maintenance, with financial assistance from the World Bank, the Kuwait Fund, Germany (KfW), Japan Bank for International Cooperation (JBIC) and the US Millennium Challenge Georgia Corporation. Foreign assistance for the road subsector is increasing. Sustained financing will continue to be needed to bring the network (particularly secondary and local roads) to a maintainable standard, while adding essential additional capacity to meet the needs of trade (transit), rural development, agriculture, industry and tourism. Increased mobility and access to secondary urban centers is expected to help with job creation. 32. The condition of the network under RD responsibility improved considerably 2004-2008, with the rehabilitation and periodic maintenance of 1,722 km in this period (Table A2. 6). The percentage of the main road network in good condition increased from 45 to 66 percent, although the improvement of secondary roads was much less, from 18 to 23 percent. While only 6 percent of the main network was in poor condition in 2008, 24 percent of the secondary network was (Table A2. 7). Table A2. 6: RD Rehabilitation/Periodic Maintenance 2004-2008 (km) Year Km. 2004 276 2005 216 2006 322 2007 577 2008 331 Source: Roads Department Table A2. 7: Surface Condition of RD Roads 2004-2008 Year/Condition Main % Secondary % 2004 Good 45.0 18.0 Fair 45.5 41.5 Poor 9.5 40.5 2005 Good 55.0 20.0 Fair 37.0 50.0 Poor 8.0 30.0 2006 Good 60.0 22.0 Fair 32.0 50.0 Poor 8.0 28.0 2007 Good 64.0 20.0 Fair 30.0 55.0 Poor 6.0 25.0 2008 Good 66.0 23.0 Fair 28.0 53.0 Poor 6.0 24.0 Source: Roads Department

Engconsult Ltd. Appendix 2 41

33. Under the proposed RD maintenance program for 2009-2013, an increase in funding from $118 million to $158 million is planned, with most of the increase going on a tripling of the periodic maintenance program (Table A2. 8). Table A2. 8: Proposed RD Maintenance Program 2009-2013 Type of Work 2009 2010 2011 2012 2013 Total $ million Routine maintenance 13.3 14.5 13.3 12.1 10.9 64.2 Winter maintenance 3.6 3.8 3.9 4.1 4.2 19.8 Periodic maintenance 18.2 21.2 30.3 45.5 54.5 169.7 Rehabilitation/Strengthening 78.8 84.8 84.8 81.8 84.8 415.2 Emergency works 3.9 3.9 3.9 3.9 3.9 19.7 Total 117.9 128.4 136.4 147.5 158.5 688.5 Source: Plan for Road Infrastructure Development 2009-2013. 34. In 2007, responsibility for local roads was transferred from RD to 69 local authorities, as part of broader public sector reform aimed at decentralizing government functions. An estimated GEL1,450 million is needed to rehabilitate the local road system (90 percent in poor condition) and bring it up to a maintainable condition. Rehabilitated local roads would then require GEL1500-4000 per km annually for routine and winter maintenance. 35. RD has no direct labor force: it contracts out all its maintenance and construction activities. The country is divided into 34 routine maintenance zones. Contractors bid against a bill of quantities drawn up by the department. The last round of bidding took place in 2007 and was for a period of 27 months. These contracts will expire in 2009 and will be re-bid for a period of 12 months. Supervision and verification of quality and quantities were also contracted out to a private company but has now been brought into the department. 36. The World Bank (WB) is supporting RD in improving the efficiency of its road maintenance activities. A shift from contracted out maintenance to performance based maintenance contracts (PBCs) may produce further savings (up to 50 percent). A WB TA for developing a comprehensive framework for the introduction of PBC has been prepared. The international consultant, in collaboration with a small RD team, will apply the PBC framework to a preliminary assessment of a concession for the maintenance of the East-West Highway and an area-wide contract for rehabilitation and maintenance for 250-300km of main and secondary roads centered on a Regional RD Office. Performance based maintenance of road projects undertaken under the MFF will be standard. 37. The national contracting capacity in the road subsector includes 13 large contractors with an annual turnover of more than GEL10 million per year, about 30 medium size contactor with an annual turnover between GEL5-10 million and more than 60 smaller contractors. 6. Vehicle Axle Load Control 38. Georgia has developed regulations to limit vehicle axle load to 11.5 tons, the EU standard, but lower than the 13.0 tonnes in neighboring countries. Enforcement, however, is weak primarily because of limited institutional capacity and lack of vehicle weigh facilities along main roads. As a result, vehicle axle overloading is prevalent. If enforcement were high, however, there would be problems in dealing with transit traffic from countries with higher limits. There is a need for coordination within the Caucasus countries to agree on a common limit. 39. Informal surveys conducted during preparation of MFF Project 1 indicated that there is a problem with trucks exceeding the axle-load limit. Heavy vehicle weights are one of the primary causes of rapid pavement deterioration, therefore enforcement of vehicle axle-load control is urgently needed to prevent premature deterioration

Engconsult Ltd. Appendix 2 42

40. To control this problem there is an urgent need to establish vehicle weigh stations along international roads, where overloaded vehicles should be required to redistribute or unload excess freight and also to collect reliable data on vehicle axle loads by installing permanent “weigh-in-motion” devices at key points. ADB is proposing to establish a vehicle weigh station on the Kobuleti bypass, under MFF Tranche 1 and to carry out an axle load survey (Appendix 14). 7. Road Transport Services 41. The road freight industry is made up of private companies and is completely unregulated. Entry to the market is easy. It is difficult to judge the extent of market efficiency, but freight rates of between GEL0.20-0.27 per tonne-km for hauls of 100-500km are consistent with operating costs estimate by HDM-4, which suggests that excess profits are not being earned. 42. Intercity bus services are extensive and frequent. Although most routes are plied by both mini-buses and larger buses, there is little to choose in terms of journey time as all buses will stop to pick up passengers en route if they have room. Fares between bus stations (e.g. Tbilisi and Batumi) are controlled by bus station operators, who take a cut from the fare. Bus station operators also appear to control capacity – a new entrant would find it difficult to operate a new service from an existing bus station. This control of both prices and capacity may be hindering the development of new services (e.g. express services).

Engconsult Ltd. Appendix 3 43

Appendix 3 TECHNICAL AND DESIGN SUMMARY (Detailed in Supplementary Appendix 2)

1. Alignment. The proposed project roads bypass the Black Sea coastal cities of Kobuleti and Batumi in the province of Ajaria in western Georgia. The proposed bypasses are the combination of tranche one comprising (i) two-lane road east of Kobuleti (28km) with the first 14 km traversing mostly through plain areas and the remainder in hilly terrain, (ii) widening of the existing road from 2 to 4 lanes between Kobuleti and Batumi (6km), and tranche 2 a two-lane road that will cut through densely populated hilly terrain east of Batumi (14km); a total of 48km. The bypasses are designed as two-lane roads, although land acquisition has allowed for future upgrading to four-lanes with at least 50 m wide reserve for the right-of way. 2. Geometric Design Standards. New Georgian design standards (April 2009) were used in the preliminary design with some adjustments approved by the Roads Department summarized below. The proposed design speeds are shown in Table A3. 1. (i) A minimum design speed of 100kph was retained throughout the rolling/hilly sections. (ii) The envelope for the tunnels was reduced by omitting the 2.5m safety lane to reduce the cost. The revised envelope agrees with most international standards. (iii) The lane width was reduced from 3.75m to 3.5m over the widening section (6 km) with 1m shoulder because of restricted available land. 3.75m was used elsewhere. (iv) A slightly higher vertical crest and sag radii were adopted in flat terrain to reduce the bridge and tunnel lengths. They comply with most international standards.

Table A3. 1: Georgian Geometric Design Standard for Two Lane Road No Main Parameters Unit Flat Rolling Mountainous 1 Design speed km/h 120 100 80 2 Number of lanes 4/2 4/2 4/2 3a Lane width (4 lanes) M 3.75 3.5 3.5 3b lane Width (2 lanes) M 3.75 3.75 3.75 4a Shoulder width left M 1.0 1.0 1.0 4b Shoulder width right M 3.0 2.75 2.25 4c Shoulder Width two lane both sides M 2.5 2.25 2.0 5 Minimum width of central reserve M 4.0 4.0 4.0 6 Hard shoulder width for emergency M 2.5 2.5 2.5 stop 7 Verge for central reserve M 1.0 1.0 1.0 8 Maximum longitudinal gradient % 4 5 6 9 Minimum horizontal curvature M 700 450 250 10 Minimum radius of vertical curves Crest M 22,600 10,000 5,000 Sag M 7,700 4,900 3,200 15 Design vertical clearance bridges, M 5 5 5 overpasses Design clearance of tunnels Vertical M 4.5 Horizontal – Two lane road tunnel M 11.4 11.4 10.65 - Twin Tube (4 Lanes) M 11.4 10.65 10.65 Service walkway width M 0.75

Engconsult Ltd. Appendix 3 44

3. Pavement. An asphalt concrete pavement was adopted to allow for estimated percentage of heavy vehicles in the traffic flow. The pavement layer thicknesses were calculated using the estimated cumulative axle loads in the design life of the road and subsoil strength tests (CBR) taken along sections of the alignment. The proposed pavement varies along the road but generally consists of 120 mm asphalt concrete wearing and binding course, 230mm crushed stone base, and 300mm granual subgrade. This will be checked at the time of final design. 4. Structures and Tunnels. There are (i) 18 bridges and one tunnel along Kobuleti bypass with a total length of 2,984 meter and 400 meter, respectively; and (ii) 19 bridges and 5 tunnels along Batumi bypass with a total length of 3,799 meter and 1,510 meter, respectively. Six bridges are 500 meters or more in length. 5. The structures and tunnels are designed according to the Georgian specifications except as mentioned in para. 2(i) above. Bridges are designed to carry 80 ton loads (HK11 – HK80) with 100 years flood return frequency. Culverts and side drains are designed to carry 50 and 20 years flood return frequency, respectively. 6. The project road is located in seismic zone IV (Modified Mercalli Scale) which requires anti-seismic measures to be included in the design of structures. Most bridges will be simple span reinforced concrete on piers and spread footings. Some bridges with longer spans will be pre-stressed T-girder superstructure. 7. The Georgian standard for two way tunnels has an emergency lane on one side of 2.5m wide with a clear width of the road of 11.5m. This gives an overly large tunnel cross section compared to most international standards and would create difficulty in construction in poor ground as well as increasing construction time and cost. An internationally accepted cross section was recommended and approved by RD for the project tunnels with two 3.5m wide lanes, 1.25m wide shoulders and 0.9m wide inspection walkway one side. The total width of the tunnel is about 11 m and height is about 7.5 m. The tunnel ventilation is deemed not required due to the low traffic volume and relatively short tunnels' length. An Illumination system is included in the tunnel designs. Suitable facilities are included in the design and cost estimates for traffic safety and emergency based on the length of the tunnels and traffic volume. 8. Materials and Labor. Construction materials, such as boulders, stones, gravel and sand are locally available at a reasonable distance. Suitable sites were identified and quantities estimated. A more detailed materials source study and analysis will be required in the final design. Other materials such as bitumen, steel, some cement, tunnel equipment and spare parts would be imported. 9. For construction of the road and small/medium bridges most skilled and all unskilled labor is available in Georgia, but for the construction of tunnels and the large bridges skilled labor would need to be imported. 10. Quantity Estimates were based on the preliminary design work. These include: (i) earthworks estimated from the surveyed longitudinal profile and cross sections, (ii) retaining structures and minor drainage, (iii) pavement works calculated based on the thickness adopted in design and the pavement cross section, (iv) bridge works estimated based on the lengths of bridge spans, cross-sections, and soil type, (v) tunnel works based on the lengths of the tunnels, cross-sections and soil types, (vi) interchanges based on estimated ramp lengths and widths and overpass spans and (vii) ancillary works and road safety provisions based on the road and structures length, plans and traffic volumes.

Engconsult Ltd. Appendix 3 45

11. Unit Rates of the major cost items were developed based on recent new road and road widening projects elsewhere in Georgia and adjusted for the first quarter 2009. Rates for items not included in recent contracts were estimated based on rates in other countries, adjusted for Georgia transport costs and taxes etc. 12. Cost Estimates are based on the quantity and the unit rates of the cost of major items. Cost of special equipment was estimated using overseas rates adjusted for Georgia. Cost for some minor miscellaneous works and overheads were estimated on a percentage basis of the earthworks, pavement and drainage cost. Consulting services costs were estimated for both international design and for international/domestic supervising engineering organizations. Costs of land acquisition and resettlement were not available for the final report and were estimated from preliminary plot maps and recent rates. 13. A ten percent allowance to the civil works cost is added for physical contingencies to cover the cost of adjustment of alignments and variations in the lengths of bridges and tunnels etc in final design. 14. Limitations to Preliminary Design. The design centerline is based on the preliminary design topographic survey and cross sections taken from digitized 1:1,000 mapping. The optimization of the road geometry design will be necessary during detailed design after detailed cross section survey is carried out. It is expected that this will reduce some quantities, especially earthworks and bridge lengths. 15. Pavement thicknesses will also vary during the detail design/construction stage. It is recommended that subgrade CBRs be checked and pavement thickness design be adjusted during detail/construction stage. 16. The geotechnical investigation, whilst covering most of the major structures was limited to tomography survey because of both limited resources and time. In some cases geological conditions have been assumed based on survey at nearby sites. Additional geotechnical investigation will be required for each site during the detailed design. This could lead to some changes in final design and quantities.

Engconsult Ltd. Appendix 4 46

Appendix 4 ROAD SAFETY

A. Road Safety Trends in Georgia 1. Road safety in Georgia is an acute problem, reflected by a sharp rise in the number of traffic accidents, fatalities and injuries (Table A4. 1). The current fatality rate is about 10 times that in European countries, causing significant economic and human losses. While there are many causes of traffic accidents, inadequacy of the road network is a contributory factor. Curves and superelevation are inappropriately designed, and safety furniture is lacking on many roads. Other causes of accidents include weak enforcement of safety regulations, reckless driving, and inadequate driver education. Accidents are likely to increase, unless remedial action is undertaken. 2. In 2008 there were over 6,000 accidents involving death or serious injury and 867 fatalities, a rise of over 18 percent on 2007. The fatality rate relative to the vehicle fleet at 12 per 10,000 vehicles in 2008 is high (even higher if scrapped vehicles are removed from the fleet statistics), significantly higher than other Eastern European countries and much higher than in the EU1. Both fatalities and serious injuries show an upward trend and road-safety is thus a serious social issue. 3. Georgia as an ECMT member is, along with other members, committed to reducing road deaths by 50 percent by 2012, compared with 2000. Whereas other countries had made good progress by 2004, with West European countries reducing deaths by 5.2 percent per year and Central and Eastern European countries by 3.3 percent per year, road deaths in Georgia increased over the period by 6.4 percent a year. Besides emotional pain, traffic accidents have a direct impact on economic growth. A 1999 World Bank working paper Road Safety in Individual ECA Countries suggested that the social economic cost of road accidents for Georgia was about 1.1 percent of GDP. Estimates by the PHRD-funded consultants in 2007 indicate that the annual economic loss is now at least $100 million, 1.4 percent of GDP. 4. When Georgia enters the “explosive” phase of motorization, with rapid growth in the vehicle fleet, accidents will increase rapidly, unless effective mitigating actions are implemented. The Government is addressing road safety, the Patrol Police is now better equipped and thus more efficient, some main roads have been resurfaced, a drunk driving law has been introduced and is being enforced, traffic monitoring devices are being installed in Tbilisi. However, there is much to be done. The Patrol Police have identified the main accident causes to be the poor condition of the roads and vehicles (e.g. tire blowouts during driving), erratic behavior of pedestrians and excessive speed, which is thought to be the dominant cause of accidents on main roads. Around 66 percent of deaths occur in “urban” areas (towns or built-up areas along main roads). On rural roads, seatbelts are worn and effective police enforcement of speeding and seatbelt wearing is undertaken, and deaths are increasing by 3 percent per year. On urban roads, however, they are increasing at around 25 percent per year. Over 30 percent of deaths are pedestrians, almost double the percentage in many West European countries (e.g. Germany 14 percent, France 11 percent, Sweden 14 percent, and the Netherlands 8 percent). The fastest growth is in the 8-15 age group. Urgent actions should be taken to prevent more loss of life. Seat belts are not worn in urban areas, where vehicle speed is often excessive (both within towns and on sections of major roads which pass through). According to a 2004

1. According to a 2004 unpublished World Bank working paper, fatality rates in other ECA countries were: Bosnia 3/10,000 vehicles, Croatia 5/10,000 vehicles, Russia 12/10,000 vehicles, Serbia 5/10,000 vehicles, Turkey 8/10,000. Other European countries average 2/10,000 vehicles.

Engconsult Ltd. Appendix 4 47 unpublished World Bank working paper, fatality rates in other ECA countries were: Bosnia 3/10,000 vehicles, Croatia 5/10,000 vehicles, Russia 12/10,000 vehicles, Serbia 5/10,000 vehicles and Turkey 8/10,000vehicles. There are very few facilities to ensure pedestrian safety or to guide drivers through hazardous locations: these should be the highest priority areas for urgent attention and action. Table A4. 1: Road Traffic Accidents 2001-2008 Year Road Accidents Number of Deaths Number of Injuries 2002 2,011 515 2,509 2003 2,113 572 2,585 2004 2,936 637 4,069 2005 3,870 581 5,546 2006 4,795 675 7,084 2007 4,946 737 7,349 2008 6,015 867 9,063 Source: Ministry of Internal Affairs, Traffic Patrol Police. 5. Responsibility for road safety is spread amongst three ministries and cooperation between the three appears somewhat sketchy. The Interior Ministry, through the Patrol Police, is responsible for attending and reporting on road accidents as well as enforcing traffic regulations and laws. The Road Safety Division of RD is responsible for ensuring that the road infrastructure is designed, constructed and maintained in a safe manner. It carries out road safety improvements, sometimes at the request of the patrol police, and other times after having manually analyzed the data collected by the police. It is also responsible for installing and maintaining road signs and markings. The Transport Department of the Ministry of Economic Development was instrumental in producing a draft Road Safety Strategy for Georgia. 6. Recognizing the growing problems, the Government has established a more effective traffic policing capability, will be developing a multi sector road safety strategy and develop and implement a multi sector road safety action plan based on the final report of consultants who have been undertaking a road safety study of Georgia during 2007 and early 20082. A road safety component has been included within the World Bank Second East West Highway Improvement Project, to strengthen the Roads Department of the Ministry of Regional Development and Infrastructure (MRDIRD) road safety unit, to implement a hazardous locations program along the East-West corridor, to design a new road accident data base and to develop traffic and safety engineering standards that can be used on urban and rural roads of Georgia. A further component is proposed in the Third East-West Highway Improvement Project to develop road safety audit manuals. 7. A road safety TA is proposed under MFF Tranche 1. It will be important to coordinate this work with World Bank projects. The TA will include use of new road safety audit manuals for project roads, inclusion of historic and ongoing accident data into the new accident database being prepared under the World Bank projects, awareness campaigns for roadside communities, school children and transport operators, and road safety publicity through television and newspapers. 8. Road safety equipment will be procured for the project including surveillance cameras, variable message signs and road safety enforcement equipment for Patrol Police. Also included is the development of intelligent transport system standards. 9. This road-safety study for the Project reviewed the existing trends in road-accidents in Georgia; carried out a safety audit of the preliminary design of the project road, developed road-

2 SWEROAD – Feb 2008. Report on proposed road safety actions.

Engconsult Ltd. Appendix 4 48 safety awareness/education programs to be carried out in conjunction with the Project and discussed other aids to road safety. 10. WB has been a major force in promoting road safety in Georgia. This has been done mainly by including road safety components in their road construction projects and through advice to RD and other GoG ministries and departments. It is important that any road safety component in this project complement their work. The road safety component proposed for the project has been discussed with WB road safety specialists. B. Road Safety Audit Of Project-Road 11. A safety audit was carried out of the initial preliminary design and reviewed the road alignment and layouts of the tunnel, interchanges, service area and axle-load control station and bridges. Based on the estimated traffic volumes there are no safety issues because of capacity constraints within the design period of the Project. 12. Countermeasures have been included in the design at potentially hazardous locations including low radius curves, constrained locations owing to the combined effect of the horizontal and vertical alignment, etc. There are no visibility constraints at the tunnels or their portals, however the lighting approaching tunnel exits should be carefully monitored such that it blends in with the outside brightness, especially in the smaller tunnels. C. Intelligent Transport System 13. Intelligent Transport Systems (ITS) is a term that is used widely (and loosely) to encompass a wide range of technological solutions applied to the transport sector to improve the performance of the transport system. These improvements include safety and traveler information (speed signs, warning signs, etc), enforcement (over-speed, over-weight, non payment of tolls, etc), monitoring (traffic speeds, over-weight, traffic counting, etc), and toll and revenue collection (e.g. electronic tolling). The relative emphasis placed on ITS for these functions varies from place to place, depending on factors such as specific need, local technological skills and ability to reliably construct and maintain a relatively high technology system, and budget considerations for both construction and on-going maintenance. In Georgia ITS is in its infancy and there appear no immediate plans for an ITS strategy. 14. With the construction of high class, high speed roads, it will be important to introduce ITS facilities to maintain safe travel, to quickly identify situations as they arise, and to inform or warn the traveling public of those situations or changed conditions and regulations. 15. The ITS features of the Project will include (i) variable message displays to provide road users with information about speeding, and enforcement; (ii) a dedicated emergency phone number; (iii) surveillance technology; (iv) detection of black spot locations and property damage; and (v) monitoring of overloaded vehicles. About 0.75 percent of the Project cost will be allocated to procure road safety and ITS equipment, and training in its use will be included in capacity development component of the project. D. Proposed Road Safety and ITS Component 16. The following items and programs are to be included in the Road Safety TA for Tranche 1 of the Project: • Use of new road safety audit manuals (WB) for project road. • Inclusion of historic and ongoing accident data into the new accident database being prepared under the WB projects. • Drivers’ Education – Safe Driving Code for Access-Controlled Road & Tunnel • Roadside Community Awareness Campaign

Engconsult Ltd. Appendix 4 49

• Pedestrian Awareness Campaign for School-Children close to open access road section. • Awareness Campaign for Heavy Vehicles • Road Safety Publicity through Television and Newspaper • Surveillance Cameras • Variable Message Signs (VMS) • Truck weighbridges • Procurement of road safety enforcement equipment for Patrol Police. • Development of Intelligent Transport System standards.

17. The tentative cost for all the road-safety components above including awareness campaigns and road safety countermeasures is US$719,000. These costs are in addition to those already included in the existing road cost estimate (ToR in Appendix 13)3. E. Long-Term Measures to Achieve Tunnel and Road Safety 18. Long-term measures for tunnel and road safety include, prompt and regular maintenance of the Project road and safety-facilities; prompt monitoring of traffic operations and incidents. On the engineering side, daily safety inspections, regular safety-audits with emphasis on safety-structures, hazardous/critical locations, tunnel and ITS installations are recommended. 19. A long-term education program on code-of-behavior for road-users is crucial as traffic operations on the Project road will involve high-speeds. Strong enforcement of traffic rules in tunnels and along the road, with stiff penalties to traffic violators, is necessary. In this aspect, it is recommended that a dedicated Patrol Police unit should be formed for the project road and that it should be well equipped and adequately staffed at all times. A system of introducing a toll-free number for emergencies is also recommended 4 so that road-users can assist in identifying incidences or road-accidents along the Project road by the use of mobile phones. The number should be well publicized at frequent points along the road. 20. More details on road safety are set out in Supplementary Appendix 4.

3 Road cost estimate includes weighbridges, cameras and VMS equipment. 4 This was suggested in the Road Safety Strategy.

Engconsult Ltd. Appendix 5 50

Appendix 5 STAKEHOLDER PARTICIPATION AND CONSULTATIONS

1. A number of consultations were held during topographic, social and environmental studies with the affected people, NGOs, and other stakeholders, including national, provincial and rayon governments to discuss alignment, environmental concerns, land acquisition and resettlement issues. The list below therefore, gives only those public consultations conducted during preliminary consultations, survey and design. Location Date Participants No. of Purpose Main Issues participa Discussed/Concerns nts Batumi April 28, Affected people, 50 First Formal construction of 2009 community Public retaining structures, leaders, civil Consultation drainage structures, society, NGOs, and avoiding environmental construction close to professionals, and Kobuleti nature central and reserve regional government officials 1. Xucubani From 03 Local 20 • To describe • People mostly 2. BoboyvaTi April 2009 to Communities, villages the people welcome the 3. OCxamuri 17 April Affected (210 about the project 4. Qveda 2009 Households, participa project • They will render awyva Sacrebulo nts) • To make the their necessary 5. GiorgaZeebi Leaders, Business public aware cooperation during 6. SaCino community and about the planning and 7. Zeda awyva Farmers social and implementation 8. SuaRele resettlement • They suggested 9. Gonio issues. some feedback to 10. Salibauri • To receive the alignment and 11. AxalSeni the feedback also the problem 12. KapreSumi from the related to land 13. Ganaxleba people on registration in the 14. Sameba alignment rayon 15. Maxvilauri • To share with • Requested for 16. Sarfi the people continued 17. KvariaTi about the consultation 18. Axalsofeli draft • People viewed for 19. Maxinjauri entitlement of adequate 20. the compensation for xelvaCauri compensatio asset loss and n policy restoration of • To build the religious and local propels community aware and property. seeking their cooperation for project planning and implementati on • To make the local sacrebulo land committee aware about

Engconsult Ltd. Appendix 5 51

Location Date Participants No. of Purpose Main Issues participa Discussed/Concerns nts the project work and to seek their cooperation for the LAR activities. 1. Khelvachauri 24 April Local Government 19 Providing Land Issues, Rayon Office 2009 and 25 training on Land Compensation, 2. Kobuleti April 2009 acquisition and Verification of missing Rayon office resettlement plot numbers and 3. Batumi Rayon implementation Office mechanism for LAR activities. Batumi Mayor 26 April Property 12 Imparting Land Issues, office 2009 Recognition Training to the Compensation, Committee and working group Verification of missing the working group committee plot numbers and in the Rayon implementation mechanism for LAR activities. Batumi June 4, 2009 Ajara and Batumi 10 Discussions of All top officials were officials project benefits. very supportive of project. Concern over coordination with other Batumi infrastructure projects and possible ADB assistance. Khelvachauri April-May Landowners and 60+ Discussion of Location of alignment. Rayon 2009 villages during alignment Compensation issues. Kobuleti Rayon survey

Batumi Rayon June 25, Affected people, 43 Second Formal Avoid valued Batumi 2009 community Public agricultural land, leaders, civil Consultation adequate drainage society, NGOs, structure for the environmental peatland, and professionals, and provision of roadside central and service stations for regional selling local produce. government officials

2. A number of consultations were held during topographic, social and environmental studies with the affected people, NGOs, and other stakeholders, including national, provincial and rayon governments to discuss alignment, environmental concerns, land acquisition and resettlement issues. The consultations methods included public meetings, focus group discussions, one-on-one meetings, and interviews. Formal public consultations were held in Batumi on April 28th, and June 25th 2009 to present project concepts and seek comments and feedback from the participants. Attendees of the public consultations included affected people, community leaders, civil society, NGOs, environmental professionals, and central and regional government officials. The outcome of all these consultations was positive as people are expecting the project to bring economic benefits to the Project area. Concerns and recommendations arising

Engconsult Ltd. Appendix 5 52

from the consultations, included proper compensation plans for the affected community, construction of retaining structures, drainage structures, and avoiding construction close to Kobuleti nature reserve; these were noted by the team. 3. Additional rounds of consultations were held during the census and inventory survey in the month of June and July 2009. The officials in Ozurgeti Rayon were also consulted for their cooperation in resettlement planning as were the property recognition committees in each Rayon. Various resettlement related information were disseminated to the Affected People through consultations with them and the local administration, especially the Sacrebulos.

.

Engconsult Ltd.

Appendix 6 COST ESTIMATES 6 Appendix

Table A6. 1: Detailed cost estimate by expenditure category for tranche 1 (Kobuleti bypass) Foreign Local Total Foreign Local Total % of base A Investment costa GEL million $ million cost

En 1 Civil works gco Subgrade 6.3 21.1 27.4 3.7 12.6 16.4 13.0% Pavement 12.2 12.4 24.7 7.3 7.4 14.8 11.7% n su Structures 26.7 41.5 68.2 16.0 24.9 40.8 32.3% l t Tunnels 5.4 3.0 8.4 3.2 1.8 5.0 4.0% L

td Slope stabilization 0.0 2.3 2.3 0.0 1.4 1.4 1.1%

. Roadside facilities etc 3.9 3.8 7.7 2.4 2.3 4.6 3.6% Miscellaneous 0.0 0.6 0.6 0.0 0.3 0.4 0.3% 2 Equipment 0.2 0.1 0.3 0.1 0.1 0.2 0.2% 3 Land and resettlement 0.0 16.8 16.8 0.0 10.1 10.1 8.0% 4 Environmental protection 0.2 4.1 4.2 0.1 2.4 2.5 2.0% 5 Design & supervision 5.2 7.9 13.1 3.1 4.7 7.8 6.2% 6 TAs: Capacity building 0.7 2.3 3.0 0.4 1.4 1.8 1.4% Road safety 0.0 1.2 1.2 0.4 0.3 0.7 0.6% 7 Taxes and duties 13.7 17.1 30.8 8.3 10.2 18.5 14.6% Subtotal (A) 74.6 134.2 208.8 45.2 79.9 125.0 98.9% B Project management PMU 0.4 1.9 2.3 0.2 1.2 1.4 1.1% Subtotal (B) 0.4 1.9 2.3 0.2 1.2 1.4 1.1% Total base costs (A+B) 75.0 136.2 211.2 45.4 81.1 126.4 100.0% C Contingencies: physicalb 7.7 12.7 20.5 4.2 8.1 12.3 9.7% pricec 4.0 29.9 33.9 2.9 17.9 20.8 16.5% IDCd 5.8 5.8 3.5 3.5 2.7% Total project cost at current prices 92.5 178.8 271.3 55.9 107.0 163.0 128.9% Source: consultants’ estimates Notes: (a) at 2009 prices (b) 10 percent on all items except PMU and TAs (c) domestic inflation 7.5% in 2010, thereafter 6% pa; international inflation on foreign exchange costs at 1% in 2010, 0% in 2011, 0.3% in 2012 and 0.5% in 2013 53

(d) interest during construction at 1% per annum on ADB Special Funds resources Appendix 6 6 Appendix Table A6. 2: Detailed cost estimate by expenditure category for tranches 1 and 2 (Kobuleti and Batumi bypasses) Foreign Local Total Foreign Local Total % of base A Investment costa GEL million $ million cost 1 Civil works Subgrade 9.7 32.0 41.7 5.8 19.1 25.0 10.1% En Pavement 17.5 17.8 35.3 10.5 10.7 21.2 8.6% gco Structures 60.5 94.0 154.5 36.2 56.3 92.5 37.4%

n Tunnels 27.9 15.9 43.8 16.7 9.5 26.2 10.6% su Slope stabilization 0.0 4.1 4.1 0.0 2.5 2.5 1.0% l t

L Roadside facilities etc 5.0 4.9 9.9 3.0 2.9 5.9 2.4% td Miscellaneous 0.1 1.0 1.1 0.1 0.6 0.7 0.3% . 2 Equipment 0.3 0.1 0.4 0.2 0.1 0.2 0.1% 3 Land and resettlement 0.0 30.6 30.6 0.0 18.4 18.4 7.4% 4 Environmental protection 0.3 4.8 5.1 0.2 2.9 3.0 1.2% 5 Design & supervision 6.7 11.4 18.1 4.0 6.9 10.8 4.4% 6 TAs: Capacity building 0.7 2.3 3.0 0.4 1.4 1.8 0.7% Road safety 0.0 1.2 1.2 0.4 0.3 0.7 0.3% 7 Taxes and duties 28.8 33.3 62.1 17.3 19.8 37.2 15.0% Subtotal (A) 157.5 253.4 411.0 94.9 151.2 246.1 99.4% B Project management PMU 0.0 2.3 2.3 0.0 1.4 1.4 0.6% Subtotal (B) 0.0 2.3 2.3 0.0 1.4 1.4 0.6% Total base costs (A+B) 157.5 255.8 413.3 94.9 152.6 247.5 100.0% C Contingencies: physicalb 16.0 24.7 40.7 9.2 15.2 24.4 9.8% pricec 14.19 62.8 77.0 9.26 37.6 46.8 18.9% IDCd 10.2 0.0 10.2 6.1 0.0 6.1 2.5% Total project cost at current prices 197.9 343.2 541.1 119.4 205.4 324.8 131.2% Source: consultants’ estimates Notes: (a) at 2009 prices (b) 10 percent on all items except PMU and TAs (c) domestic inflation 7.5% in 2010, thereafter 6% pa; international inflation on foreign exchange costs at 1% in 2010, 0% in 2011, 0.3% in 2012 and 0.5% in 2013 (d) interest during construction at 1% per annum on ADB Special Funds resources 54

2009 2010 201120122013 2014 Appendix Activity Appendix 7 Tranche 1 IMPLEMENTATION SCHEDULE 7

Engconsult Ltd.

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4

Approval of Advance Procurement

Tranche I (34 km) - Kobuleti Bypass tranche I tranche II Detailed Design and Bid Documents Review of Detailed Design by the Government Construction Supervision

Recruitment of Supervision Consultants

Supervision and Preparation of Project 2 defects liability period Prequalification for Civil Works

Contract 1 & 2 - Km 0-11 & Km 28-34

Bid Documents and ADB's Approval Civil Works Procurement Civil Works Implementation

Contract 3 - Km 11-28 defects liability period

Bid Documents and ADB's Approval Civil Works Procurement Civil Works Implementation

Capacity Development for Road Network

Road Safety Improvement defects liability period

55

Appendix 8 56

Appendix 8 PROCUREMENT PLAN Basic Data Project Name: XXX-GEO MFF Subregional Road Corridors Development Program Country: Republic of Georgia Executing Agency: Ministry of Regional Development and Infrastructure Loan Amount: $118.8 million Loan (Grant) Number: XXX Date of First Procurement Plan: June 10, 2009 Date of this Procurement Plan: june 10, 2009 A. Process Thresholds, Review and 18-Month Procurement Plan 1. Project Procurement Thresholds 1. Except as the Asian Development Bank (ADB) may otherwise agree, the following process thresholds shall apply to procurement of goods and works.

Procurement of Goods and Works Method Threshold

International Competitive Bidding (ICB) for Between $1,000,000 and $10,000,000, Works1 International Competitive Bidding for Goods1 Between $500,000 and $1,000,000 Shopping for Goods Below $100,000 2. ADB Prior or Post Review 2. Except as ADB may otherwise agree, the following prior or post review requirements apply to the various procurement and consultant recruitment methods used for the project.

Procurement Method Prior or Post Comments Procurement of Goods and Works ICB Works Prior Recruitment of Consulting Firms Quality- and Cost-Based Selection (QCBS) Prior Quality-Based Selection (QBS) Prior Recruitment of Individual Consultants Individual Consultants Prior

3. Goods and Works Contracts Estimated to Cost More Than $1 Million 3. The following table lists goods and works contracts for which procurement activity is either ongoing or expected to commence within the next 18 months. General Contract Procurement Prequalification Advertisement Description Value Method of Bidders (y/n) Date (quarter/year) Comments

Engconsult Ltd. Appendix 8 57

4. Consulting Services Contracts Estimated to Cost More Than $100,000 4. The following table lists consulting services contracts for which procurement activity is either ongoing or expected to commence within the next 18 months.

Advertisement International General Contract Recruitment Date or National Description Value Method1 (quarter/year) Assignment Comments Detailed Design 3.2 QBS QII/2009 International Quality-cost ratio and Bidding 80:20. Documents for Tranche I and II Construction 10.6 QCBS QI/2010 International Quality-cost ratio Supervision of 80:20. Tranche I and II Project 1.4 QCBS, CQS QIII/2010 International Procurement Administration or individual method to be and Support consultant determined depending on the need. Capacity 1.8 QCBS, CQS QIII/2010 International Procurement Development for or individual method to be Road Network consultant determined depending on the need. Consulting 0.72 QCBS QIII/2010 International Quality-cost ratio Services for 80:20. Road Safety Improvement Assistance in 0.8 QCBS QII/2010 International Quality-cost ratio Developing a 80:20. Road Master Plan. Assistance in 0.45 QCBS or QII/2010 International Quality-cost ratio Developing PPP Individual 80:20. capacity Consultant

B. Indicative List of Packages Required Under the Project 5. The following table provides an indicative list of all procurement (goods, works and consulting services) over the life of the project. Contracts financed by the Borrower and others should also be indicated, with an appropriate notation in the comments section. Estimated Estimated Domestic General Value Number of Procurement Preference Description (cumulative) Contracts Method Applicable1 Comments Works Civil works contract 105 2 ICB No package for One-envelope construction of single stage Kobuleti Bypass with (28 km) Tranche 1 prequalification

Engconsult Ltd. Appendix 8 58

Estimated Estimated Domestic General Value Number of Procurement Preference Description (cumulative) Contracts Method Applicable1 Comments

Consulting Services Construction 10.5 1 QCBS or QBS No Quality-cost Supervision of with FTP ratio for Tranche I and II QCBS 80:20 Detailed Design for 3.5 1 QBS with FTP No Quality-cost Tranche I and II ratio for QCBS 80:20 Project 1.5 several QCBS, CQS or No Procurement Administration and individual method to be Support consultant determined depending on the need Capacity 1.8 Several QCBS, CQS or No Ditto Development for individual Road Network consultant 0.72 1 QCBS No Quality-cost Consulting Services ratio 80:20. for Road Safety Improvement

Assistance in 0.8 1 QCBS No Quality-cost Developing a Road ratio 80:20. Master Plan. 0.45 1 QCBS or No Quality-cost Assistance in Individual ratio 80:20. Developing PPP Consultant capacity

Engconsult Ltd. Appendix 9 59

Appendix 9 TRAFFIC FORECAST

A. Introduction 1. The proposed tranche 1 project is a 34km limited access bypass around the congested coastal town of Kobuleti and the tranche 2 project a 14km bypass around the Ajaran capital, Batumi. The existing road through both towns carries significant flows of transit and domestic long haul traffic. Turkey is Georgia’s most important trading partner and substantially all trade with Turkey uses the existing road. 2. In the without-project case traffic will continue to use the existing S-2 two lane international highway. With the project, long haul and transit traffic will divert to the bypass. Some additional trips will be generated on the existing road as the economy responds to lower transport costs. B. Traffic surveys 3. Roads Department (RD) carries out classified radar counts in April, July and October at three sites on the existing road: at Choloki (north of Kobuleti), at the - tunnel (between Kobuleti and Batumi) and at Chorokhi bridge (just south of Batumi). Records go back to 2006 (but with gaps). Prior to the current TA no origin and destination (OD) surveys had been carried out. Two sets of project traffic surveys were carried out. Manual classified counts were undertaken at the three RD sites on 26-27 March and 1-2 April (1x12h and 1x24h count at each). OD surveys were originally scheduled for early April but were delayed as an indirect result of the civil unrest in the country at the time. One OD survey at Chorokhi and one at Choloki were eventually carried out on 11-12 May. C. Base year traffic 4. Regional shocks in 2006 and 2008 followed by the global recession have affected traffic in the project corridor. Estimated 2009 (base year) annual average daily traffic (AADT) values are 3,655 veh/day at Choloki, 8,361 veh/day at the tunnel and 6,108 veh/day at Chorokhi. Traffic growth was strong in 2007-8 (averaging around 20 percent a year) but has fallen by 18 percent in 2009 at Choloki and the tunnel. Traffic at those sites is now at its 2007 level. Traffic at Chorokhi has fared better, perhaps as a result of continued construction activities in Batumi, and has increased by 20 percent since 2008. 5. Turkey is Georgia’s largest trading partner. Substantially all trade not only between Turkey and Georgia but also between Turkey and Azerbaijan and between Turkey and Armenia uses the Sarpi-Batumi-Kobuleti corridor. As a result long distance and transit traffic are strongly represented. Around 300 truck-semi-trailer combinations (“articulated trucks”) use the road daily of which 90 percent carry foreign (almost entirely Turkish) license plates. Large tourist buses are also predominantly foreign; 79 percent were recorded as carrying foreign license plates during the project counts at Chorokhi. 6. The Ajara coast is a popular summer tourist destination and while heavy goods traffic displays no clear seasonality, passenger vehicle traffic certainly does:;car and mini-bus traffic is typically 30-50 percent higher in July than it is in April or October. As seasonal congestion is a significant part of project rationale, project counts and historic seasonality were used to construct a traffic flow pattern that was slightly more extreme than the default HDM-4 pattern.

Engconsult Ltd. Appendix 9 60

D. Corridor diversions 7. Corridor traffic may be affected by other developments. Traffic case A assumes no corridor diversions: all traffic currently using the Sarpi-Kobuleti corridor will continue to do so. 8. Case B diverts some long distance and transit traffic using a proposed new border crossing and improved roads near Kartsakhi. Road improvements, financed by the Millennium Challenge Fund Georgia (MCG), will be complete by late 2010. There is a new border post (and improved road connections) on the Turkish side of the border but as yet no progress has been made on Georgian border facilities. Once the new road is open, the distance from Erzurum in Turkey to Tbilisi via Kartsakhi will be 575km, compared with 675km via Sarpi – a drop of 15 percent. Only some trips to and from Tbilisi and further east will switch to the shorter route: the Sarpi route is likely to be faster (especially outside the tourist season) and less prone to adverse weather conditions. 9. The project OD survey results indicate that 25 to 40 percent of all traffic at Chorokhi bridge and Choloki – and 30-50 percent of goods traffic – would find the Kartsakhi corridor attractive. For case B it is assumed that 40 percent of all trip ends east of Gori and 20 percent of those in the zone around Gori would actually divert to the Kartsakhi corridor. E. Traffic forecasts 10. Normal traffic growth is based on forecast GDP growth and elasticities of demand of 1.2 for passenger and 1.4 for goods vehicles. Conservative GDP growth percentages are used: 0% in 2009, followed by 1.5% in 2010, 2.0% in 2011, 3.0% in 2012 and 4.0% thereafter. 11. Based on changes in road user costs, the ‘with project’ generated traffic on the existing road is calculated as 10 percent of normal traffic; this was not applied to medium and heavy goods vehicles or large buses. 12. Local traffic, i.e. traffic between points within the urban centers of Batumi and Kobuleti, will not divert to the bypass but adds to congestion on existing roads and so affects journey times and vehicle operating costs. To take account of this effect, light and medium passenger and light goods traffic has been doubled on the central urban sections of the existing road. 13. Manual assignments for base case and case B traffic were made on the basis of the OD results. Base case traffic forecasts for tranche 1 and tranches 1 and 2 combined are shown in Table A9. 1 and Table A9. 2. In summary: • diversion to the Kartsakhi corridor (i.e. case B traffic) reduces project corridor traffic by approximately 20 percent • the percentage of project corridor traffic diverting to the bypass is 42-43 percent in both the base case and case B (60-61 percent in the case of goods traffic) • opening base case traffic on the new two lane section of the tranche 1 (Kobuleti) bypass in 2014 will be 3,600 veh/day (base case) and 2,800 in case B. The four lane section will carry respectively 10,200 and 8,600 veh/day • opening traffic on the tranche 2 Batumi bypass in 2014 will be 3,400 veh/day (base case) and 2,800 veh/day (case B) • in 2033, the last evaluation year, bypass traffic on the two lane sections will be 6-8,000 veh/day, depending on the traffic case, i.e. still short of the levels that would normally justify an additional lane

Engconsult Ltd. Appendix 9 61

Table A9. 1: Tranche 1 base case traffic projection, veh/day Year Without With project project Existing road Existing road Bypass Generated Total traffic 2009 6,326 0 0 0 0 2010 6,588 0 0 0 0 2011 6,862 0 0 0 0 2012 7,147 0 0 0 0 2013 7,443 0 0 0 0 2014 7,752 4,233 3,559 410 8,202 2015 8,131 4,439 3,734 430 8,602 2016 8,528 4,654 3,918 450 9,023 2017 8,945 4,880 4,111 472 9,463 2018 9,382 5,118 4,313 495 9,926 2019 9,841 5,366 4,525 519 10,411 2020 10,322 5,627 4,748 544 10,919 2021 10,787 5,879 4,963 568 11,411 2022 11,274 6,143 5,189 594 11,925 2023 11,782 6,418 5,424 620 12,462 2024 12,313 6,706 5,670 648 13,024 2025 12,868 7,006 5,928 677 13,611 2026 13,448 7,320 6,197 707 14,224 2027 14,054 7,648 6,478 739 14,865 2028 14,688 7,991 6,772 772 15,535 2029 15,350 8,349 7,080 806 16,235 2030 16,042 8,723 7,401 842 16,966 2031 16,765 9,114 7,737 880 17,730 2032 17,521 9,522 8,088 919 18,529 2033 18,311 9,949 8,455 960 19,364 Source: consultants’ calculations Notes: (a) traffic is weighted by distance; as the bypass is slightly longer than the existing road, the sum of project traffic on the bypass and the existing road slightly exceed the without project traffic (b) local traffic is not included, nor is traffic on the common alignment between km28 and km34

Table A9. 2: Tranches 1 and 2 base case traffic projection, veh/day Year Without With project project Existing road Existing road Bypass Generated Total traffic 2009 6,667 0 0 0 0 2010 6,944 0 0 0 0 2011 7,232 0 0 0 0 2012 7,532 0 0 0 0 2013 7,844 0 0 0 0 2014 8,170 4,702 3,515 457 8,674 2015 8,569 4,931 3,688 479 9,097 2016 8,988 5,170 3,870 502 9,542 2017 9,428 5,421 4,061 526 10,008 2018 9,889 5,685 4,261 552 10,497 2019 10,372 5,961 4,471 578 11,010 2020 10,879 6,251 4,691 606 11,548

Engconsult Ltd. Appendix 9 62

Year Without With project project Existing road Existing road Bypass Generated Total traffic 2021 11,369 6,531 4,904 633 12,069 2022 11,882 6,824 5,127 662 12,612 2023 12,417 7,129 5,360 691 13,180 2024 12,977 7,449 5,604 722 13,774 2025 13,562 7,782 5,858 754 14,394 2026 14,173 8,131 6,124 788 15,043 2027 14,812 8,495 6,403 823 15,720 2028 15,480 8,876 6,694 859 16,429 2029 16,178 9,273 6,998 898 17,169 2030 16,907 9,689 7,316 938 17,942 2031 17,669 10,123 7,648 979 18,750 2032 18,466 10,576 7,996 1,023 19,595 2033 19,298 11,050 8,359 1,069 20,478 Source: consultants’ calculations Notes: (a) traffic is weighted by distance; as the bypass is slightly longer than the existing road, the sum of project traffic on the bypass and the existing road slightly exceed the without project traffic (b) local traffic is not included, nor is traffic on the common alignment between km28 and km34

Engconsult Ltd. Appendix 10 63

ECONOMIC ANALYSIS

A. Introduction

1. The proposed tranche 1 project is a 34km limited access bypass around the congested coastal town of Kobuleti. The tranche 2 project is a 14km bypass around the Ajaran capital, Batumi. The existing road through both towns carries significant flows of transit and domestic long haul traffic. Turkey is Georgia’s most important trading partner and substantially all trade, not only between Georgia and Turkey but also between Turkey and Azerbaijan and Turkey and Armenia, uses the existing road. Arriving transit traffic passes through the Sarpi border, 18km south of Batumi, making its way through Batumi and Kobuleti before proceeding towards Tbilisi and further east.

2. Each tranche will be implemented over approximately 3 years. The economic appraisal covers the construction period between 2010 and 2013, and 20 years’ operation. The appraisal compares a project case (with the bypass) and a do minimum (without project) alternative that amounts to maintaining the existing road. A 12 percent discount rate is used throughout.

3. In the do minimum case traffic will continue to use the existing S-2 two lane international highway, which will become increasingly congested and dangerous as well as expensive to maintain as traffic increases. Some travel may not take place because of congestion and the resulting increased journey times and vehicle operating costs (VOCs). With the project, long haul and transit traffic will divert to the bypass, reducing congestion and accidents on the existing road. Some additional trips will be generated on the existing road as the economy responds to lower transport costs.

4. Two traffic cases are considered. The base case (case A) assumes that the Sarpi-Batumi-Kobuleti corridor continues to carry substantially all long haul and transit traffic. Case B assumes that approximately 20 percent of traffic diverts to a new route further east. The latter comprises the rehabilitation and upgrading of existing roads south west of Tbilisi, meeting the Turkish border near the town of Kartsakhi. Works on the Georgian side of the border should be complete by late 2010. There is a new border post (and improved road connections) on the Turkish side but as yet work has not started on Georgian border facilities. The new route will reduce the distance from Erzurum in Turkey to Tbilisi by 15 percent, but the existing route via Sarpi will continue to be faster, especially outside the tourist season, and less prone to adverse weather conditions.

5. The analysis was carried out using HDM-4 and using 2009 prices.

B. Evaluation of tranche 1

6. All predicted project costs and benefits are measured in economic prices using an international price numeraire and a US dollar unit of account. That is, tradable goods are expressed in world market prices plus transport costs (border parity prices), while non-tradable goods are converted from domestic prices (less taxes etc) multiplied by a

Engconsult Ltd. Appendix 10 64 standard conversion factor (SCF) and divided by the official exchange rate. An SCF of 0.97 was calculated using the ADB’s simple trade-weighted formula.

7. Bypass benefits to “normal” traffic (i.e. traffic that would have been present in the do minimum case) result from improvements in road and traffic conditions between the do minimum and project case and amount to savings in (i) VOCs, (ii) personal travel time for vehicle occupants (other than drivers and crew, whose costs are included in VOCs), (iii) road accident costs and (iv) the value tied up in freight during transit. In addition there are the benefits of additional trips made solely as a result of the lower costs on the existing road brought about by the project (“generated” traffic). The proposed project road is approximately 2.3km longer than the existing road. As a result (and as is typical for bypass investments), most of the benefits arise from time, rather than VOC, savings.

8. The estimated economic internal rate of return (EIRR) in the base case is 16.2 percent, well above the 12 percent threshold rate. Using the 12 percent discount rate it has a net present value (NPV) of USD40.6m at 2009 prices. Of the total benefits, 83 percent are journey time and time in transit savings, 10 percent are VOC savings, 5 percent accident cost savings and 2 percent generated traffic benefits (Table 2). Using case B traffic the EIRR falls to 13.3 percent, still above the 12 percent threshold (Table 3).

9. The results of sensitivity analysis (Table 1) confirm the robustness of the project’s economic viability: the EIRR remains above 12 percent for all costs up to 1.5 times base and for benefits above 0.7 times base, and is fairly robust in the face of a fall in traffic, including opening of the Kartsakhi corridor. A modest improvement is achieved by delaying opening by a year.

10. A Monte-Carlo risk analysis was used to estimate the frequency distribution of EIRRs. A negative skew was applied to costs to reflect (i) the weak construction market in 2009 and (ii) the scope for optimization of the alignment. Negative skews were also applied to all benefits to reflect the scope for loss of corridor traffic. The results of the risk analysis showed that with these uncertainties (i) the mean EIRR for the whole project was 15.4 percent and (ii) the probability of the EIRR falling below 12 percent was 3.4 percent. C. Evaluation of tranches 1 and 2

11. Table 4 shows the costs and benefits of tranches 1 and 2 combined, using base case traffic. The combined EIRR is 14.8 percent. The results of sensitivity analysis show that the EIRR remains above 12 percent for all costs up to 1.3 times base and for benefits above 0.8 times base. A modest improvement is achieved by delaying opening by a year. A Monte-Carlo risk analysis using the same skewed distributions of costs and benefits as for tranche 1 gave a mean EIRR of 14.0 percent; 14 percent of results fell below 12 percent.

Engconsult Ltd. Appendix 10 65

Table 1 Sensitivity analysis – tranche 1

Scenario EIRR percent NPV in USD Switching million factora (a) Base case 16.2% 40.6 (b) Traffic case B (Kartsakhi diversion) 13.3% 11.1 (c) Benefits less 10% 15.0% 28.3 0.67 (d) Benefits less 20% 13.7% 15.9 (e) Cost increased by 10% 15.1% 32.3 1.49 (f) Cost increased by 20% 14.2% 24.0 (g) Residual value reduced from 65 to 30% 16.0% 38.0 (h) Combination (c) and (e) 14.0% 20.0 (i) Traffic Growth + 10% 17.6% 58.8 (j) Traffic Growth -20% 13.1% 9.8 (k) Opening delayed by a year 17.0% 48.9 Notes: (a) the switching factor is the factor times the base value of the indicated item that will produce an EIRR of 12 percent

Table 2 Tranche 1 base case economic evaluation

Year Incremental cost, USDm Road user cost savings, USDm Net benefits Capital Recurrent Total VOC Time Gen Accident Total savings traffic savings USDm 2010 23.22 0.00 23.22 0.00 0.00 0.00 0.00 0.00 -23.22 2011 30.52 0.00 30.52 0.00 0.00 0.00 0.00 0.00 -30.52 2012 28.05 0.00 28.05 0.00 0.00 0.00 0.00 0.00 -28.05 2013 20.96 0.00 20.96 0.00 0.00 0.00 0.00 0.00 -20.96 2014 0.00 0.01 0.01 1.05 12.95 0.29 0.87 15.16 15.14 2015 0.00 0.01 0.01 1.16 13.69 0.31 0.91 16.07 16.05 2016 0.00 0.01 0.01 1.30 14.48 0.33 0.96 17.07 17.05 2017 0.00 0.01 0.01 1.51 15.32 0.35 1.00 18.18 18.16 2018 0.00 0.01 0.01 1.76 16.22 0.37 1.05 19.40 19.39 2019 0.00 -0.33 -0.33 2.05 17.18 0.40 1.10 20.72 21.05 2020 0.00 0.36 0.36 2.25 18.19 0.42 1.16 22.02 21.65 2021 0.00 -0.33 -0.33 2.57 19.25 0.45 1.21 23.48 23.81 2022 0.00 0.58 0.58 2.71 20.37 0.47 1.27 24.82 24.24 2023 0.00 -0.24 -0.24 3.06 21.58 0.51 1.33 26.48 26.72 2024 0.00 0.58 0.58 3.23 22.86 0.54 1.40 28.02 27.44 2025 0.00 -0.21 -0.21 3.62 24.21 0.57 1.47 29.87 30.09 2026 0.00 0.58 0.58 3.89 25.63 0.61 1.54 31.67 31.09 2027 0.00 -0.21 -0.21 4.35 27.11 0.65 1.61 33.72 33.93 2028 0.00 0.58 0.58 4.60 28.63 0.69 1.69 35.61 35.03 2029 0.00 -0.21 -0.21 5.10 30.21 0.73 1.78 37.82 38.03 2030 -0.50 0.35 -0.15 5.33 31.70 0.78 1.86 39.67 39.82 2031 -1.28 0.10 -1.18 3.20 35.75 0.75 1.95 41.65 42.83

Engconsult Ltd. Appendix 10 66

2032 0.00 0.36 0.36 3.35 41.91 0.88 2.05 48.19 47.83 2033 -65.61 0.01 -65.60 5.08 49.09 1.05 2.15 57.37 122.97 PV at 12% 82.7 0.3 83.0 12.3 102.7 2.4 6.4 123.7 40.6 Percent of total road user cost savings 9.9% 83.0% 1.9% 5.1% 100.0% EIRR 16.2%

Table 3 Tranche 1 economic evaluation, case B traffic

Year Incremental cost, USDm Road user cost savings, USDm Net benefits Capital Recurrent Total VOC Time Gen Accident Total savings traffic savings USDm 2010 23.22 0.00 23.22 0.00 0.00 0.00 0.00 0.00 -23.22 2011 30.52 0.00 30.52 0.00 0.00 0.00 0.00 0.00 -30.52 2012 28.05 0.00 28.05 0.00 0.00 0.00 0.00 0.00 -28.05 2013 20.96 -0.05 20.91 0.00 0.00 0.00 0.00 0.00 -20.91 2014 0.00 0.01 0.01 0.62 10.33 0.25 0.71 11.91 11.89 2015 0.00 0.01 0.01 0.70 10.91 0.26 0.74 12.61 12.60 2016 0.00 0.01 0.01 0.64 11.53 0.28 0.78 13.23 13.21 2017 0.00 0.01 0.01 0.75 12.18 0.29 0.82 14.04 14.02 2018 0.00 0.01 0.01 0.90 12.87 0.31 0.86 14.93 14.92 2019 0.00 -0.28 -0.28 1.07 13.60 0.33 0.90 15.90 16.18 2020 0.00 0.31 0.31 1.19 14.39 0.35 0.94 16.87 16.56 2021 0.00 -0.28 -0.28 1.41 15.22 0.37 0.99 17.99 18.27 2022 0.00 0.58 0.58 1.52 16.11 0.39 1.04 19.06 18.48 2023 0.00 -0.19 -0.19 1.75 17.06 0.42 1.09 20.32 20.51 2024 0.00 0.52 0.52 1.87 18.06 0.44 1.14 21.52 20.99 2025 0.00 -0.19 -0.19 2.11 19.11 0.47 1.20 22.88 23.08 2026 0.00 0.53 0.53 2.26 20.22 0.50 1.26 24.23 23.71 2027 0.00 -0.16 -0.16 2.59 21.40 0.53 1.32 25.84 26.00 2028 0.00 0.52 0.52 2.77 22.65 0.56 1.38 27.36 26.84 2029 0.00 -0.16 -0.16 3.19 23.96 0.60 1.45 29.19 29.35 2030 0.00 0.52 0.52 3.47 25.30 0.64 1.52 30.93 30.41 2031 -2.72 -0.07 -2.80 3.90 26.70 0.68 1.59 32.87 35.67 2032 -1.28 0.52 -0.76 -0.37 28.06 0.65 1.67 30.00 30.76 2033 -66.79 0.10 -66.69 -1.56 29.45 0.67 1.75 30.31 97.00 PV at 12% 82.5 0.3 82.8 6.6 80.2 1.9 5.2 93.9 11.1 Percent of total road user cost savings 7.0% 85.4% 2.1% 5.5% 100.0% EIRR 13.3%

Engconsult Ltd. Appendix 10 67

Table 4 Tranches 1 and 2 economic evaluation, base case

Year Incremental cost, USDm Road user cost savings, USDm Net benefits Capital Recurrent Total VOC Time Gen Accident Total savings traffic savings USDm 2010 31.43 0.00 31.43 0.00 0.00 0.00 0.00 0.00 -31.43 2011 53.09 0.00 53.09 0.00 0.00 0.00 0.00 0.00 -53.09 2012 60.88 0.00 60.88 0.00 0.00 0.00 0.00 0.00 -60.88 2013 59.94 0.00 59.94 0.00 0.00 0.00 0.00 0.00 -59.94 2014 0.00 0.02 0.02 2.36 22.02 0.35 1.35 26.09 26.07 2015 0.00 0.02 0.02 2.58 23.26 0.37 1.42 27.63 27.62 2016 0.00 0.02 0.02 2.85 24.59 0.40 1.49 29.32 29.30 2017 0.00 0.02 0.02 3.19 26.00 0.42 1.56 31.18 31.16 2018 0.00 -0.03 -0.03 3.62 27.51 0.45 1.64 33.21 33.24 2019 0.00 -0.43 -0.43 4.07 29.11 0.49 1.72 35.39 35.82 2020 0.00 0.47 0.47 4.43 30.81 0.52 1.80 37.56 37.09 2021 0.00 -0.43 -0.43 4.96 32.61 0.56 1.89 40.01 40.44 2022 0.00 0.79 0.79 5.23 34.53 0.59 1.98 42.33 41.54 2023 0.00 -0.34 -0.34 5.85 36.58 0.64 2.08 45.14 45.48 2024 0.00 0.79 0.79 6.17 38.76 0.68 2.18 47.79 47.00 2025 0.00 -0.31 -0.31 6.82 41.05 0.74 2.29 50.88 51.19 2026 0.00 0.79 0.79 7.27 43.44 0.79 2.40 53.90 53.11 2027 0.00 -0.31 -0.31 8.01 45.98 0.85 2.51 57.36 57.67 2028 -0.57 0.79 0.22 8.43 48.60 0.91 2.64 60.58 60.36 2029 0.89 -0.25 0.64 7.80 51.34 0.99 2.77 62.89 62.25 2030 -0.73 0.47 -0.26 9.11 54.08 1.07 2.90 67.16 67.42 2031 -1.28 0.10 -1.17 6.32 58.77 1.05 3.04 69.18 70.35 2032 -0.05 0.47 0.42 6.51 65.44 1.22 3.19 76.36 75.94 2033 -132.29 0.02 -132.27 8.67 74.69 1.42 3.35 88.13 220.41 PV at 12% 160.1 0.4 160.5 23.9 173.0 3.0 9.9 209.8 49.3 Percent of total road user cost savings 11.4% 82.5% 1.4% 4.7% 100.0% EIRR 14.8%

Engconsult Ltd. Appendix 11 68

Appendix 11 INSTITUTIONAL AND FINANCIAL ASSESSMENTS

A. Roads Department 1. RD’s institutional context changed in February 2009. Up to then, RD was part of the Ministry of Economic Development (MED). In February 2009 the MED was reorganized and a new ministry created, the Ministry of Regional Development and Infrastructure (MRDI) and RD and the United Transport Administration (UTA), a small organization principally responsible for the international aspects of transport, were moved from the MED to the MRDI. The Transport Department, a policy unit, remains with the MED. A body related to RD is the Transport Reform and Rehabilitation Center (TRRC), which acts as the project management unit (PMU) for World Bank projects. 2. Following re-structuring several years ago, RD is now a small, fairly non-hierarchical organization with 160 staff members. Officials at the department have indicated that its structure may still change following recent changes at the top of the organization. RD has no planning section (this being the remit of the Transport Department) and no internal financial audit section. Organization charts are shown in Appendix 2. 3. From 1995 to 2004 Georgia had a road fund. The fund performed poorly, however, and as part of a wider change of government policy regarding special funds, the fund was abolished in 2004. Since then the sector has been funded directly from the state budget and from donor funds. RD income and expenditure is summarized in Appendix 2. B. Road maintenance 4. The department has no direct labor force: it contracts out all its maintenance and construction activities. The country is divided into 34 routine maintenance zones, each of which covers 150-250km of roads. Contractors bid against bills of quantities drawn up by RD following an annual condition survey. The last round of bidding took place in 2007 and was for a period of 27 months. These contracts will expire in 2009 and will be re-bid for a period of 12 months. Supervision (i.e. checking both quality and the quantities of work undertaken) was also contracted out to a private company but has now been brought back into the department. 5. The World Bank is supporting the RD in its efforts to improve the efficiency of its road maintenance activities. It believes that a shift from contracted out maintenance to performance based road maintenance contracts (PBCs) could produce further savings. A proposed project component of the Third East-West Highway Project will assist in developing a comprehensive framework for the introduction of PBC in Georgia. C. Proposed ADB support for capacity building under tranche 1 6. A comprehensive program of capacity building is a necessary complement to increased IFI assistance for road infrastructure development. ADB is coordinating its capacity building interventions closely with those of the World Bank and other donors to maximize their effectiveness. It will target support for capacity building under tranche 1 by providing technical assistance in three areas. The ADB component of the three TAs is estimated at around $2.0 million: • assistance to RD to develop a road sector master plan ($0.8 million). The lack of a planning department in RD is noted above – the master plan will focus on the medium to long term and address concerns which are currently subject to ad hoc decisions

Engconsult Ltd. Appendix 11 69

• strengthening RD’s capacity for controlling vehicle overloading, with support to acquire necessary equipment and its capacity in tunnel and bridge engineering and in environmental and resettlement safeguards ($0.76 million) • strengthening RD’s capacity to implement PPP, including performance based maintenance of the project road ($0.45 million)

D. Financial management capacity 7. Financial management assessments of the TRRC (the World Bank roads PMU) and the RD itself were carried out. 8. The TRRC was created by presidential decree in 1995. Although physically located in RD, its twelve staff members are not departmental employees. Moves to merge the TRRC with RD are in abeyance. Most of TRRC’s operating expenses are met from World Bank project funds. Its budget for 2009 is $0.15m. 9. The TRRC has a strong financial management capability. Audits of project accounts have always been issued without qualification and while some minor deficiencies have been noted by auditors, they are being dealt with in an action plan. 10. RD has a 16 strong finance department but it has no experience of externally funded projects: both foreign and counterpart funds for such projects are managed by the TRRC. RD’s financial management capacity appears adequate for its current tasks but would need considerable strengthening if it were to handle externally financed projects. E. Proposed project management arrangements 11. Financial management of ADB projects should strengthen the long term capacity of RD, not the TRRC. It is therefore proposed that the ADB PMU is formed within RD, with procurement and financial management support from TRRC. 12. The proposed arrangement envisages the MRDI as executing agency (EA) and RD as implementing agency (IA). The chairman of RD will be responsible for overall supervision of the program. RD will establish a PMU headed by the deputy chairman of RD. It will comprise technical staff seconded from RD and TRRC. A steering committee will be established with representatives of the cabinet office, MRDI, RD, Transport Department (of behalf of MED), the Ministry of Finance and the PMU head. The steering committee will be chaired by a deputy minister of MRDI and will meet at least every six months to ensure interagency coordination, review implementation progress, and provide approvals and guidance as necessary. 13. The PMU will handle day-to-day implementation activities. It will also undertake periodic functions such as (i) preparation of project financial reports (PFRs), (ii) engagement and supervision of consultants and contractors, (iii) preparation of reports and (iv) obtainment of approvals from ADB and the government. The PMU will be supported by a team of externally contracted qualified technical, safeguards, financial, and support staff to supplement resources in RD. Figure A11. 1 and Figure A11. 2 show, respectively, the proposed organizational arrangements and the funds flow. F. Transport sector PPP 14. Public-private partnerships in the transport sector have made relatively little headway in Georgia; there are none in the roads sector although there have been two attempts. Some of the reasons for the slow progress are to be found in EBRD’s 2008 assessment of concession laws, which rated Georgia’s legal environment for private sector development as “low

Engconsult Ltd. Appendix 11 70 compliance” with international best practice. As a result a PPP capacity building TA is proposed as part of tranche 1 (see Appendix 15). 15. The Road Law of 1994 permits the imposition of tolls on an existing road and allows the creation of a private, tolled facility. The only conditions are that: • there must be an alternative, untolled route that is open to traffic • the proceeds of the toll must be used to maintain and develop the tolled facility 16. There is no explicit prohibition of making profits from tolling – although such a prohibition might be inferred from the second condition. There is no provision for the regulation of tolls. 17. RD has one tolled facility – the Rikoti tunnel. Tolls are levied by virtue of Roads Department order no 6 of January 1999. This order set the toll rates (GEL1, GEL2 and GEL3 for small, medium and large vehicles respectively), which have not been changed since. Tolls are paid in cash and are only payable when the alternative route is open. Toll proceeds are not used for tunnel maintenance; they are simply part of RD’s revenue. Annual revenue is approximately GEL2.2m. G. Tolling the Ajara bypass 18. It is recommended that the road is not tolled, at least in the short term, but that the possibility be kept under review and re-appraised once demand for the bypass has been established. Although there is an alternative, toll-free route (as required by law) and the bypass saves time rather than distance (and thus tolls would be progressive in the sense that they would apply to the better off), Georgia has, first, no experience of awarding such a concession in the transport sector, second, a proposal to toll would hinder the fast-track implementation of the project and, third, tolling would reduce the economic internal rate of return of the bypass. H. Financial evaluation 19. The basis for the evaluation is that of an operation and maintenance (O&M) concession: a private sector partner is awarded the right to toll the bypass, in return for which the company is expected to cover all the operation and maintenance costs (including overlays but excluding reconstruction or widening) for a period of 20 years. ADB interest expenses could also be met from tolls. 20. The concessionaire would include a profit element (20 percent was assumed) in his price for undertaking maintenance works (rather than take equity returns, as in a “full” concession). Revenue and expenditure profiles will differ sharply – more so than under a full concession. To handle the revenue and expenditure mismatch the operator would operate a sinking fund. Adequate financial performance is signalled by adequacy of the sinking fund, i.e. its ability to meet O&M expenses as they fall due with a nil (or negligible) closing balance at the end of the term. 21. Experience suggests that a toll at around one third to one half of perceived road user cost savings will produce a good compromise between economic and financial returns. On this basis the implied charge for a car using the tranche 1 (Kobuleti) bypass should be no more than GEL2.40 $1.40 ($1.40), and for the entire bypass GEL3.50 ($2.10). 22. Runs using Fineval, the consultants’ financial evaluation program, showed that the levels of toll needed to recover O&M and ADB interest expenses were modest by comparison. Assuming no toll indexation: • the total charge for using the two lane section of the tranche 1 road (Kobuleti bypass) would be GEL0.8 for a car and GEL2.30 for an articulated truck

Engconsult Ltd. Appendix 11 71

• the total charge for using the two lane section of tranches 1 and 2 (Kobuleti and Batumi bypasses) would be GEL1.8 for a car and GEL5.5 for an articulated truck

Steering committee (Cabinet office, MRDI, RD, MED, MoF, PMU head)

MRDI (EA)

Roads Department (IA)

TRRC: PMU Detailed design i) financial management ii) procurement

Supervision

External experts: i) tunnel & bridge engineering Capacity building TAs ii) safeguards

Road safety TA

Contract 1 Contract 2 etc

Figure A11. 1: Proposed project management arrangements

Engconsult Ltd. Appendix 11 72

ADB Treasury

Withdrawal application Application for plus certified counterpart fund claim/invoice & release RD/PMU approval

Payment advice RD/PMU approval Payment advice

Payment Payment

Certification by project engineer

Contractor’s or consultant’s claim or invoice

Figure A11. 2: Proposed funds flow arrangements

Engconsult Ltd. Appendix 12 73

Appendix 12 OUTLINE TERMS OF REFERENCE FOR CONSULTANTS FOR CONSTRUCTION SUPERVISION OF TRANCHE I AND TRANCHE II

Expected Date of Publication: Q4 2009 Tentative Date of Commencement: Q3 2010 Duration: 52 Months Selection Method: Consulting firm through Quality-and-Cost-Based Selection1 with Full Technical Proposal and 80:20 quality-cost ratio Consulting Input (person-months) International: 180 National: 1,520 professional, Outputs to be produced by the consultants: (i) Develop computer aided contract management system (ii) Monitor civil works for construction of Kobuleti and Batumi bypasses in accordance with Conditions of Contract. (iii) Certification of claims, variations, subcontracting and payments and assist in settling disputes. (iv) Conduct seminars, asses training impacts. (v) Develop risk management plan.

A. Introduction 1. International and domestic consultants will assist the Executing Agency (EA) in ensuring that (i) designs are carried out to the appropriate engineering standards; (ii) all work associated with the project are carried out in full compliance with the designs and specifications; (iii) the EA's engineers and domestic consultants receive in-country and international training in selected areas of tunnel design and construction and pavement design; (iv) resettlement, social, environmental, road safety, and monitoring are implemented in accordance with the recommendations of various studies, plans, analysis of the project. A total of 180 person months are required for international consultants and an estimated 1,520 person months for domestic engineering consultants 2. The consultants will be engaged under ADB financing and will be recruited in accordance with the EA’s procedures acceptable to ADB. B. Scope of Services and Staffing Requirements International Consultants

3. The international consultants will act as adviser to the EA, and support and assist staff in the performance of their duties. About 183 person months of international consulting services will be required to provide advisory services in construction supervision. The team of international consultants will comprise a highway construction project management specialist as Chief Engineer for 52 person months; 48 months for a highway engineer, 36 months for a bridge engineer, 18 months for a tunnel engineer, 6 month for a road pavement specialist, 3 months for a contracts specialist, 2 months for a road safety specialist, 12 months for an environmental specialist, and 3 months for a social specialist.

4. The Chief Engineer, will be appointed over the whole construction period. While being responsible for coordinating the inputs and the activities of the international and domestic specialists, he will also assist the EA in:

1 Following ADB’s Guidelines on the Use of Consultants (http://www.adb.org/Documents/Guidelines/Consulting/default.asp?p=cnsltng

Engconsult Ltd. Appendix 12 74

• developing and implementing a computer-aided contract management system for monitoring the civil works contracts and overall project implementation, and generating monthly and quarterly progress reports; • organizing, coordinating, and monitoring the construction supervision activities in accordance with the conditions of contract of Federation International des Ingenieurs- conseils (FIDIC); • assist in settling contractors’ claims; • reviewing and certifying engineering variation orders, contractor’s applications for subcontracting parts of the works, and monthly payments to the contractors for submission to the EA. Assist EA in settling disputes with contractors; • preparing documents and progress reports to be submitted to ADB; • monitor performance, deadlines, project progress, as well as conduct seminars, assess training impacts and assist in the development of a risk management plan to avoid any unexpected incidence that may have a negative impact on the project development.

5. The highway engineer will be fielded for a period of 48 person months to assist the chief engineer in the construction of the highway works. He will each be responsible for both bypass sections. 6. The bridge engineers will be fielded for a period of 36 person months to review the final bridge design, oversee the bridge construction supervision, implement quality control procedures, improve construction methods and to train the domestic site engineers on bridge construction and maintenance. 7. The tunnel engineer will be fielded for a period of 18 person months to review the final tunnel design, oversee the tunnel construction supervision, review the safety procedures set up by the contractor, set up a risk management plan, improve construction methods and to train the domestic engineers on tunnel construction and maintenance. 8. The pavement specialist consultant will be required for a period of 6 person months to review the final pavement design, review pavement construction techniques and oversee the domestic consultants in supervision of pavement construction. 9. The contracts specialist shall assist the chief engineer in reviewing the contract document and assisting in the evaluation of claims and disputes. 10. The road safety specialist will be required for a period of 2 person months and recruited upon completion of the detailed design and before the start of the works, to conduct a road safety audit, and advise on road safety issues of the expressway operations. The specialist will provide a second input to inspect the completed works before the facilities are opened and, where appropriate, recommend additional traffic engineering and ITS improvements, particularly for lighting, hazard barriers, signage, and marking. 11. The environmental specialist will be required for a period of 12 person months and will carry out environmental monitoring and management of project implementation and help ensure the implementation of environmental management practices at each stage of the construction. The specialist will develop an environmental auditing protocol for the construction period, formulate a detailed environment monitoring and management plan (EMMP), regularly supervise the environmental monitoring, and submit periodic reports based on the monitoring data and laboratory analysis reports. The specialist will also develop a program for hands on training of contractor’s staff in implementing the EMMP.

Engconsult Ltd. Appendix 12 75

12. The social/resettlement specialist will be responsible for the development and implementation of the Project Monitoring System (PMS) and monitoring and evaluating HIV/AIDS/STI prevention programs and transmittable disease detection and prevention programs. 13. In-country training programs will be submitted to ADB for review and concurrence. The international consultants will conduct workshops in their fields of expertise.

Domestic Consultants 14. The domestic consultants will be composed of graduate engineers or the equivalent, with at least 5 years of practical experience in supervision and administration of major road, bridge, and tunnel contracts.

Construction Supervision 15. The consultants will supervise all construction activities. 16. Construction supervision activities will include the following: • Assist the EA with contractors’ applications for subcontracting parts of the civil works, by reviewing ambiguities and discrepancies in contract documents, and in settling disputes with contractors. • Ensure that all required data for setting out the civil works are provided. • Help the EA with drawings quality, and advise the EA on further drawing or instructions to the contractors. • Assist the EA to review the contractors’ organizational arrangements, key personnel and construction programs, materials, and sources of materials. • Monitor the progress of the works against programmed targets, advise the EA on measures to be taken to improve progress and quality, review and advocate proposed variation orders, determine rates for works, and decide on alternatives. • Inspect the quality of the works in terms of workmanship and compliance with specifications; order, supervise, or perform tests on materials and civil works; advise on approval or disapproval of the contractors’ plan and equipment; propose, if required, the uncovering of completed civil works and the removal and replacement of improper material and works. • Check the progress of civil works, examine and review sampling measurements of any work that is about to be covered or put out of view before permanent work is placed thereon, examine the measurements of the completed civil works, and (as directed by the EA) order the initiation of additional civil works that are part of the contract. • Supervise the contractors in all matters concerning safety and care of civil works including the provision of necessary lights, guardrails, fencing, and security. • Examine the contractors’ accounts, invoices, claims, and other statements for errors in complying with the contract, and suggest and make corrections. • Advise the EA on how and when to issue certificates of payment to the contractors and certify the completion of the civil works or parts thereof. • Provide advice and help for the local government authorities to facilitate implementation of works.

Project Monitoring System

Engconsult Ltd. Appendix 12 76

17. The consultants will be responsible for the development and implementation of the Project Monitoring System (PMS) and will initially collect and analyze a set of indicators for evaluating project performance against project objectives, purposes, and outputs. The indicators set out in project documents will be reviewed and a baseline survey will be conducted at the beginning of project implementation to collect data not already supplied. Domestic consultants will be recruited to conduct follow-up surveys at project completion, and one year and three years after completion. Data collected during and after implementation will be compared with the baseline data and the target values to be established by the EA and agreed upon with the Asian Development Bank (ADB) before the works start. The main indicators to be monitored include (i) economic development and poverty indicators at the municipality level and for each county in the project area; (ii) transport costs and time for specific types of vehicles and trips; (iii) transport services and charges; (iv) air quality; (v) changes in producer prices, and (vi) jobs created during construction and maintenance. Where relevant, indicators will be differentiated by gender. The data collection method will consist of (i) reviewing secondary data from local government statistics; (ii) conducting household socioeconomic sample surveys; and (iii) developing participatory rural appraisal methods among communities affected by the bypass roads. About 24 person-months of domestic consultants specializing in socioeconomic and transport fields will be required to develop and implement the PMS.

C. Records and Reports 18. The construction supervision consultants will maintain records of implementation progress, contract variations, and design revisions for incorporation in the drawings for the final works. They will also (i) assist the EAs in preparing progress and financial status reports, and a final project completion report; (ii) certify interim and final payments; (iii) give the EAs technical and engineering advice on the day-to-day project activities; (iv) keep records of all works done, and prepare monthly and quarterly progress reports; (v) prepare, before the works start, a project implementation schedule based on the contractors’ work programs; and (vi) keep records of all payments approved and report the same in the monthly progress reports. The monthly reports will cover a description of the activities, the progress charts, the expenditure records, and implementation-related issues and the suggested remedial actions. The monthly reports will be consolidated into quarterly reports comparing the actual and originally anticipated work program. 19. Upon completion of their inputs, the individual international consultants will prepare for each relevant technical area, progress reports giving findings and recommendations. Upon completion of their services, the consultants will submit a final report covering the work performed, and highlighting findings and recommendations to the EA. Copies of these reports and actions taken should be provided to ADB. 20. A strategic plan regarding performance measures, effectiveness, training, and various activities of the Chief Engineer should be developed during the first two months of the duty commencement, and ADB should be informed. D. Support Services for the Consultants 21. The EA will make the following available to the consultants: (i) administrative assistance to international consultants in obtaining visas, customs clearance, and any other administrative permits required to perform their duties; (ii) all relevant reports; documents relating to the assignment, (iii) suitable office accommodation on site and in Tbilisi and (iv) appropriate and suitably qualified counterpart staff. Counterpart staff will be agreed upon with the EAs before the services start.

Engconsult Ltd. Appendix 12 77

Table A12. 1: Summary of International Consulting Services Consultant No. Person Months 1. Chief Engineer/Road Engineer 1 52 2. Highway Engineer 1 48 3. Bridge Engineer 1 36 4. Tunnel Engineer 1 18 5. Pavement Engineer 1 6 6. Contracts Specialist 1 3 7. Road Safety Engineer 1 2 8. Environmentalist 1 12 9. Social Specialist 1 3 TOTALS 9 180

Table A12. 2: Summary of Domestic Consulting Services Total Person Consultant No. Months 1. Deputy Chief Engineer 1 50 2. Deputy Resident Engineers 4 192 3. Highway Supervising Engineers 16 768 4. Bridge Supervising Engineer 6 180 5. Tunnel Supervising Engineer 4 144 6. Pavement Engineer 2 24 7. Geotechnical Engineer 1 12 8. Drainage Engineer 2 16 9. Quality Control Specialist 1 48 7. Road Safety Engineer 1 2 8. Environmentalist 2 72 9. Social Specialist 2 12 TOTALS 41 1,520

Table A12. 3: COST ESTIMATES AND FINANCING PLAN ($'000) Total Item Cost A. ADB Financing 1. Consultants a. Remuneration and Per Diem i. International Consultants 3,780 Number of Person-months 180 ii. Domestic Consultants 1,968 Number of Person-months 1,520 iii. Technical Support 3,184 Number of Person-months 3,264 b. International and Local Travel 384 c. Reports and Communications 41 2. Laboratory and Testing 610 3. Miscellaneous Administration and support 152 costs 4. Contingencies (5%) 506 10,62 Subtotal (A) 6

Engconsult Ltd. Appendix 13 78

Appendix 13 OUTLINE TERMS OF REFERENCE FOR CONSULTANTS FOR ROAD SAFETY ENHANCEMENT AND AWARENESS TRAINING

TA Name: Road Safety Enhancement and Awareness Training Expected Date of Publication: 2009 QIII Tentative Date of Commencement: 2010 QIII Duration: 9 Months Selection Method: Consulting firm through Quality-and-Cost-Based Selection1 with Simplified Technical Proposal and 80:20 quality-cost ratio Consulting Input (person-months) International: 11 National: 20 Outputs to be produced by the consultants: (i) Train and built capacity of the Executing Agency and other concerned agencies to improve application of road safety principles, road safety monitoring and education. (ii) Design and implement road safety awareness campaigns targeting school students, pedestrians and general public. (iii) Specify equipment to be procured for road safety enforcement (iii) Prepare an ITS strategy plan. A. Consulting Input 1. The TA will be implemented by a consultant team consisting of a team leader (international road safety specialist, 7 person-months), road safety equipment specialist (1 international for 1 person-month), ITS specialist (international for 3 person months, one national for 2 months), road safety trainers (2 national for 8 person-months each), media specialist (1 national for 2 person-months) B. Implementation Arrangements 2. The Consultant will head a team comprised of international and domestic staff along with RDMEDI and PP and other counterpart specialists who will work closely with the Consultant. These counterpart staff will be drawn from existing resources. After proper training, the Georgian counterpart staff are expected to implement the outputs from the preparation assignment under close monitoring of the Consultant. The Consultant will therefore guide the counterpart team, review and comment on outputs, and help the counterpart team to produce the outputs to the necessary standard. C. Scope of Services 3. The new Project road comprises two sections - (i) access-controlled Kobuleti bypass (28km) and (ii) the free-access six kilometer widening of the existing road. The first of these sections will be restricted, with no admittance to pedestrians, slow and non-motorized traffic, while there will be mixed traffic and pedestrians in the second section. 4. The objective of the assignment is for the Consultant to develop awareness campaigns and also establish in-country expertise for further road safety campaigns, train counterparts in road safety monitoring, specify equipment required for road safety enforcement and develop an

1 Following ADB’s Guidelines on the Use of Consultants (http://www.adb.org/Documents/Guidelines/Consulting/default.asp?p=cnsltng

Engconsult Ltd. Appendix 13 79

ITS strategy. The focus of the work shall be on ensuring that RDMEDI, PP and other professionals are fully involved in the preparation of the campaigns so that they are able to do most of the work of implementation and follow up. This ‘knowledge transfer’ to be proposed by the consultant in their proposal will be considered an essential measure of the viability of achieving the outputs. D. Scope of Work 5. The objective of the assignment is for the Consultant to develop awareness campaigns and also establish in-country expertise for further road safety campaigns, train counterparts in road safety monitoring, specify equipment required for road safety enforcement and develop an ITS strategy. The focus of the work shall be on ensuring that RDMEDI, PP and other professionals are fully involved in the preparation of the campaigns so that they are able to do most of the work of implementation and follow up. This ‘knowledge transfer’ to be proposed by the consultant in their proposal will be considered an essential measure of the viability of achieving the outputs. 1. Road Safety Audits (One person month international, 2 person months national trainers) 6. Training will be carried out to suitable counterparts from RDMRDI, PP, Technical University and NGOs in the use of road safety audits, based on the manuals prepared by WB consultants. The counterparts will then under the supervision of the consultants carry out an audit of the details design of the new road and also of the existing S2 road from Poti to Sarpi. 7. The timing of this component will depend on the completion of the Manuals by WB consultants. 2. Awareness Campaigns (6 person-months of Team Leader, 16 person months of national trainers and media specialist.) 8. The safety awareness campaign for the access-controlled section should take into consideration safety-issues related to high-speed traffic and the likely implication to both the road-users and roadside communities. The target-groups would therefore be (i) road-users and drivers of all types of vehicles and (ii) roadside communities. 9. For the former the issues which should be addressed are a safe driving code of behavior while driving on an access-controlled road including road tunnels and carrying out education about safety facilities and response in the event of incidences such as accidents. 10. For the roadside communities, the issues to be addressed would be to educate the communities on the dangers of accidentally or forcefully breaking through the safety boundary fence to the access-controlled road. This would include publicizing the construction of sufficient pedestrian and tractor underpasses and non-disruption of the existing local roads, foot-trails, etc. 11.Suggested campaigns will include (but are not limited to):

• Safety Code for Access-Controlled Roads & Tunnels. Educating drivers about the correct code of behavior while driving in the access-controlled sections of road including tunnels. It should consist of design of materials to be used in program and the actual education should be conducted to all road-users through joint collaboration with the consultants, PP and RDMRDI.

Engconsult Ltd. Appendix 13 80

In the initial period, the education should be carried out through concerned associations of transport entrepreneurs and PP. In the latter part of the Program, the Consultants should in consultation with PP, update the existing manuals for drivers incorporating the above code. With the collaboration with PP the syllabus for the written test for driving-licenses should be updated to include the safe driving code, with special reference to high speed roads.

• Roadside Community Awareness Campaign. This program will carry out safety education of roadside communities within the access-controlled section of the Project and would target school children and concerned residents within the immediate vicinity of the road. The campaign should be conducted through local community based organizations in collaboration with RDMRDI, PP and the Consultants. It should consist of preparation of brochures highlighting a code of conduct around access-controlled roads, lectures and training to local trainers (PP or NGOs), supervision of implementation of awareness campaign by local trainers to students and residents.

• Pedestrian Awareness Campaign for Open Access Section of Road. This campaign should use posters and literature related to increased danger from heavy-vehicles and safe pedestrian behavior when walking by heavily trafficked roads. This program should be conducted in schools in Kobuleti and Batumi: Activities should include, preparation of posters and other literature, conduct lectures at schools with the assistance of MRDIRD and PP.

• Awareness Campaign for Heavy Vehicles. The main message behind this campaign will be to underscore the importance of observing safe-driving habits with special consideration given to pedestrians and non-motorized vehicles on mixed traffic road sections. This campaign should be conducted at truck and bus operator’s depots and should include preparation of brochures and stickers and lectures at transport operators depots with the assistance of RDMRDI and PP.

• Publicity through Television and Newspaper. The Project should produce and broadcast a documentary on television highlighting the following issues pertaining to the Project roads, including code of driving behavior on high speed roads and tunnels and information on emergency facilities available on high speed roads.

For newspaper publicity a set of road safety advertisements/information should be designed and placed in national and regional papers.

3. Assistance in enforcement by PP by procurement of road safety enforcement equipment. (One month international) 12. The consultants will specify a list of road safety enforcement in-vehicle equipment including radar guns, breathalyzers and computerized vehicle data recoding equipment. The budget for this item would be around $25,000.

Engconsult Ltd. Appendix 13 81

4. Adoption of ITS Standards (3 person months international, 2 person months domestic) 13. With the introduction of high standard, high speed, roads in Georgia, it is appropriate to promote, at the national level, the need to develop and adopt national ITS standards. Given the amount of effort, time and cost developing and testing standards in other countries, there is considerable scope to draw upon that international experience as the basis for national ITS standards. The Consultants will collaborate with RDMRDI in developing a suitable standard for Georgia. Activities will include: (i) Workshops and lectures to road user stakeholders on the need for ITS standards, implementation of basic ITS measures and examples of ITS in other countries, and (ii) collaborate with RDMRDI on designing ITS standards suitable for Georgia, which will include a phased approach to ITS implementation. COST ESTIMATES AND FINANCING PLAN ($'000) Total Item Cost A. ADB Financing 1. Consultants a. Remuneration and Per Diem i. International Consultants 230.0 Number of Person-months 11 ii. Domestic Consultants 75.0 Number of Person-months 20 b. International and Local Travel 60.0 c. Reports and Communications 10.0 a 2. Equipment (Computer, Printer, etc.) 30.0 3. Workshops and Seminars b 20.0 4. TV/Radio/Brochures/Posters 85.0 5. Miscellaneous Administration and 10.0 Support Costs 6. Contingencies 26.0 Subtotal (A) 546.0

B. Government Financing 1. Office Accommodation and Transport 18.0 2. Remuneration and Per Diem 130.0 of Counterpart Staff 3. Venue for Workshops and Seminars 15.0 4. Previous Studies and Reports 2.0 5. Miscellaneous Administration and Support Cost 8.0 Subtotal (B) 173.0 TOTAL 719.0 a Equipment Type Quantity Cost Estimate only. To be specified during TA

b Workshops, Training/Seminars & Conferences Purpose Number of Venue Duration Cost Participants Estimate only. To be specified during TA

Engconsult Ltd. Appendix 14 82

Appendix 14 OUTLINE TERMS OF REFERENCE FOR CONSULTANTS FOR ROAD SECTOR CAPACITY BUILDING

TA Name: Road Sector Capacity Building Expected Date of Publication: 2009 QIII Tentative Date of Commencement: 2010 QIII Duration: 6 Months Selection Method: Consulting firm through Quality-and-Cost-Based Selection1 with Simplified Technical Proposal and 80:20 quality-cost ratio Consulting Input (person-months) International: 17.5 National: 18.5 Outputs to be produced by the consultants: (i) Train and build capacity of the Executing Agency and other concerned agencies in high class road design and construction techniques. (ii) Carry out axle load surveys to determine extent of overloading on Georgian roads (iii) Build up the capacity and procedures of the Environmental and Resettlement Unit (ERU) of RDMRDI and Ajara Autonomous Region officials. (iv) Develop and build on the Land Acquisition and Resettlement (LAR) capacity currently available in the RDMRDI, Ajara Autonomous Region and the local Rayons (v) Train trainers to continue building capacity. A. Consulting Input 1. The TA will be implemented by a consultant team consisting of a team leader (international road engineering specialist, 3 person-months), tunnel design/construct specialist (1 international for 1.5 person-month), bridge design/construct specialist (one international and one national for 1.5 person month each), pavement design/construct specialist (one international for 4 person months and one domestic for 6 person months) international contract management specialist (one international for 1 person-month), social/resettlement specialist (1 international for 4 and one national for 8 person months person-months), environment specialist (one international for 2 person months and one national for 3 person months). B. Implementation Arrangements 2. The Consultant will head a team comprised of international and domestic staff along with RDMEDI (the EA), Ajara Autonomous Region and other counterpart specialists who will work closely with the Consultant. These counterpart staff will be drawn from existing resources. After proper training, the Georgian counterpart staff are expected to be capable of training further groups of RDMEDI and other professional counterparts. C. Scope of Services 3. At the present time international agencies intend to or are funding a substantial number of both international and state roads in Georgia in order to improve the overall road network. Many of these roads are high speed international roads, some are limited access. There is limited experience in Georgia in the design and construction of high speed roads, and road tunnels.

1 Following ADB’s Guidelines on the Use of Consultants (http://www.adb.org/Documents/Guidelines/Consulting/default.asp?p=cnsltng

Engconsult Ltd. Appendix 14 83

4. The project roads will carry a large amount of heavy truck transit traffic. There is no information on the extent of vehicle overloading on main roads in Georgia and despite Georgia having an axle load limit of 11.5 tons there are few means of measuring actual loads. As pavement overloading can cause accelerated damage to the pavement it is important to both know the extent of truck overloading and control it. 5. The project roads will require extensive environmental and social/resettlement monitoring both during construction and in the first years of operation. Whilst there is some experience in safeguard monitoring in Georgia, it has little depth and, in particular there is no experience in ADB safeguard requirements. 6. It is proposed to engage consultants to develop and implement a human resource development plan to include domestic formal and hands on training in areas of design and construction and environment and social monitoring of impacts of high speed roads. The consultants may be engaged as part of the supervisory services for the ADB funded Kobuleti bypass road. 7. The objective of the assignment is for the Consultant to train counterparts so as to establish in-country expertise for further domestic training in high speed road design and construction, international contract management, environmental monitoring and social/resettlement monitoring. The focus of the work shall be on ensuring that RDMEDI and other professionals are fully trained such that they can become trainers themselves. The ‘knowledge transfer’ to be proposed by the consultant in their proposal will be considered an essential measure of the viability of achieving the outputs. In addition the consultants will train counterparts in the use of portable weighbridges to be procured as part of the services and assist them in carrying out an axle load survey on the existing Batumi-Kobuleti road and other international roads in Georgia. 8. The objectives of the environment training are (a) to help build up the capacity and procedures of the Environmental and Resettlement Unit (ERU) of RDMRDI and Ajara Autonomous Region officials to undertake analyses of environmental impacts of road projects and to prepare environmental management and monitoring plans, land acquisition and resettlement impacts and social monitoring in accordance with Government regulations and donor guidelines, and (b) to provide hands-on-training on environmental management to the ERU, RD, and the contractor. 9. The objective of the social/resettlement training is to allow effective execution of all Land Acquisition and Resettlement (LAR) related tasks by developing an expansion of the capacity on LAR currently available at the EA (RD/RDMRDI), Ajara Autonomous Region and in the Rayons. D. Scope of Work 10. The international and national consultants will work with domestic counterparts from MRDIRD to prepare and present a training program for staff of MRDIRD and other selected Georgian professionals from other agencies. The subjects to be selected for training are indicated below but will be finalized during negotiations. All training materials will be translated into Georgian and provision for such translation should be included in the consultants offer. 11. The training will consist of four parts (i) developing a human resource development plan, including domestic formal and hands on training, (ii) domestic training sessions to be given by the international specialists to RDMRDI counterparts, (iii) assist in carrying out an axle load survey and (iv) evaluation of the training program.

Engconsult Ltd. Appendix 14 84

12. The human resource development plan will include all in-country training programs. The formal specialist training programs will be submitted to EA and ADB for review and concurrence. Candidates trained will be required to submit a report, through the consultants, to the EA and ADB on the training received. The trainees should be able to serve as resource persons in further training seminars for dissemination of knowledge learned. 13. The international teal leader will take overall responsibility for the training including preparing an overall training program in consultation with the other international experts, assisting in all training sessions and preparing an evaluation report. 1. Road Design and Construction Training 14. Training will be carried out to suitable counterparts from RDMRDI and will consist of bridge and tunnel design and construction techniques, road pavement design and construction techniques, international bidding and contract management. Training will include: • Bridge design and construction techniques of high/long bridges • Tunnel design and construction techniques of tunnels over 0.5km in hard and soft rocks • Bridge and tunnel design in seismic active areas • The evaluation of suitable pavement surface types • International pavement design standards • Pavement construction techniques and the effect of overloading on pavements. • A review of FIDIC contract documents • Construction variations, claims and disputes • Contract management techniques 2. Environment Monitoring Training 15. The consultant’s scope of work will include the following tasks: • reviewing prevailing government regulations and donor guidelines governing the assessment and management of environmental and social impacts of road projects; • identifying the procedures and tasks required to be performed by ERU and RD to meet the requirements of these regulations and guidelines; • reviewing the skills of ERU and RD officials and assessing the need for training to establish the capability to meet environmental and social management and monitoring requirements; • preparing a short-term staff training plan and associated materials to meet immediate needs; • undertaking the training through a combination of hands-on assistance, on-the-job training, and training workshops; • evaluating the effectiveness of the training provided by other agencies measuring improvements in attitudes and skills achieved; • preparing outline proposals for the longer-term development of ERU’s organization and capabilities. 3. Social and Resettlement Training 16. The first task will be to carry out a capacity need assessment to define the capacity building activities required. Capacity building exercises will be needed at the Road Department, Ajara Regional and Rayon level. The designated team at the Rayon level for the LAR activities will be provided with training related to the LAR planning and implementation and will be made familiar with ADB’s policy on Involuntary Resettlement. This will be done through on the job training and by participation with the LAR activities.

Engconsult Ltd. Appendix 14 85

17. Training will cover the following topics: • ADB’s Policy on Involuntary Resettlement (1995) and other safeguards Policies related to IP and Gender etc. • Principles and procedures of land acquisition; • Public consultation and participation; • Entitlements and compensation & assistance disbursement mechanisms; • Grievance redress; • Implementation Mechanism of LARP and • Monitoring of resettlement operations. 4. Axle Load Survey and Analysis 18. Training in pavement design and the effect of overloading on pavement design will be covered under item I above. In addition the pavement specialist will: • Specify and assist RD in the procurement of three portable vehicle weighing equipment; • Train RD staff in the use of the equipment; • Develop a continuing program for truck axle load checking on the existing Poti-Batumi Road and also on selected locations of the E-W Highway. Obtain approval from RD on the program. • Supervise the execution of the load checking program. • Train RD staff in the analysis of the initial results and preparation of a report setting out the extent of overloading on the roads.

E. Staffing 19. The international road engineering consultants shall have a minimum of 15 years experience in their respective fields in which 5 years should be on international projects involving high speed roads. 20. The international environmental specialist will have a minimum of 10 years experience in environmental assessment and management of infrastructure projects and must possess a relevant post-graduate degree preferably doctorate in civil /environmental engineering. 21. The international social/resettlement specialist will have a post graduate in Social Science with a minimum 10 years of relevant work experiences in the social sector especially in land acquisition and resettlement planning and implementation. Up- to- date knowledge of various social safeguards policies of Asian Development Bank and Georgia Laws/regulations on Land Acquisition and Resettlement is required. Experience in implementation of LARP activities and experience in ADB funded project implementation will be desirable;

Engconsult Ltd. Appendix 14 86

COST ESTIMATES AND FINANCING PLAN ($'000) Total Item Cost A. ADB Financing 1. Consultants a. Remuneration and Per Diem i. International Consultants 387.0 Number of Person-months 17.5 ii. Domestic Consultants 62.9 Number of Person-months 18.5 b. International and Local Travel 90.0 c. Reports and Communications 5.0 2. Equipment a 75.0 3. Workshops and Seminars b 40.0 4. Translation Costs 25.0 5. Miscellaneous Administration and 10,2 Support Costs 6. Contingencies 36.3 Subtotal (A) 761.4 B. Government Financing 1. Office Accommodation 2. Remuneration and Per Diem of Counterpart Staff 3. Venue for Workshops and Seminars 4. Previous Studies and Reports 5. Miscellaneous Administration Cost

a Equipment

Type Quantity Cost Estimate only. The estimate includes portable single axle weighbridges, including indicators, computer, printer and associated software with transport

racks for testing vehicle. Other equipment will include computers, printers and presentation equipment for workshops and training sessions.

b Workshops, Training/Seminars Purpose Number of Venue Duration Cost Participant s Estimate only. To be specified during TA. Includes materials.

Engconsult Ltd. Appendix 15 87

Appendix 15 OUTLINE TERMS OF REFERENCE FOR CONSULTANTS FOR PUBLIC-PRIVATE PARTNERSHIP (PPP) CAPACITY ENHANCEMENT

TA Name: Public-Private Partnership (PPP) Capacity Enhancement Expected Date of Publication: 2010 QIII Tentative Date of Commencement: 2011 QIII Duration: 4 Months Selection Method: Consulting firm through Quality-and-Cost-Based Selection1 with Simplified Technical Proposal and 80:20 quality-cost ratio or direct appointment of individual consultants Consulting Input (person-months) International: 12 National: 12 Outputs to be produced by the consultants: (i) Train and build capacity of the Roads Department in the development of a PPP framework. (ii) A review of existing relevant legislation and drafts of any required amendments (iii) Economic and financial analyses of potential PPP projects and a participatory workshop (iv) Model PPP contracts suitable for Georgian conditions (v) A PPP procurement strategy

A. Consulting Input 1. The TA will be implemented by a consultant team consisting of a team leader (international PPP specialist, 4 person-months), deputy team leader/PPP specialist (national, 4 person-months), legal specialist (1 international for 2 person-month and one national for 2 person-months), economist/financial analyst (one international and one national for 3 person months each), contracts specialist (one international for 3 person months and one national for 3 person months). B. Implementation Arrangements 2. The Consultant will head a team comprised of international and national staff along with the Roads Department (RD) (the Implementing Agency - IA) and other counterpart specialists who will work closely with the Consultant. Counterpart staff from Ministry of Regional Development and Infrastructure (MRDI), as the expected contracting party of any PPP arrangements, should also work with the Consultant. These counterpart staff will be drawn from existing resources. After proper training, the Georgian counterpart staff are expected to be capable of training further groups of RD and other professional counterparts. The Consultant will work in close liaison with the World Bank PBC consultants and wherever possible seek ways in which delivery of PBCs can be enhanced.

1 Following ADB’s Guidelines on the Use of Consultants (http://www.adb.org/Documents/Guidelines/Consulting/default.asp?p=cnsltng

Engconsult Ltd. Appendix 15 88

C. Background 3. The Government of Georgia (GoG) supports the involvement of the private sector in infrastructure. Evidence of this in the transport sector was the early efforts by the MED to involve the private sector in the East-West Highway. These early efforts were not successful, however: significant increases in private participation in transport projects can generally only occur when there is a strong policy commitment to private approaches across a range of government functions and a defined administrative process for handling PPP project proposals. The objective of the assignment is to muster this policy commitment and put RD in a position where it could issue an RFP for a PPP arrangement and be assured of getting adequate, cost- effective bids. 4. Public-private partnerships share risks between the public and private sectors. PPP arrangements come in four main forms: management contracts (including O&M contracts and the PBCs referred to below), lease contracts, concessions for existing transport infrastructure, and concessions for new “greenfield” transport infrastructure. In all cases, the arrangement must be financially attractive to the private sector to be viable. But the degree of risk transfer to the private sector tends to increase progressively through these categories. As risk increases, the cost of debt and equity to the private sponsors will increase; the projected returns then need to be higher if the PPP is to be financeable. Public sector risk often remains substantial in transport concession agreements to facilitate the transaction at acceptable cost. The public sector risks are sometimes expressed through full or partial revenue guarantees but there are many other types and gradations of risk-sharing which can differ by project. 5. The usual rationale for PPP is that, when facing appropriate incentives, a private company will deliver a more cost-effective service to users than the public sector. In Georgia’s case there is also a transit argument: transit countries may derive little direct economic benefit from the traffic that used its infrastructure. This is undoubtedly so in the case the ADB funded Ajara bypass roads projects, where it is estimated that some 42 percent of benefits will go to transit traffic. Similar cases will arise from other bypass projects (e.g. those proposed for the EW Highway). If a way can be found to recoup some infrastructure costs from foreign beneficiaries without jeopardizing the project’s economic return, then Georgia will benefit. It can be seen as a way of capturing the negative externalities of transit traffic as a financial cash flow. 6. RD, with World Bank support, is proposing to shift from contracted out maintenance to performance based road maintenance contracts (PBCs) that could produce further savings (of up to 50 percent). A proposed project component of the Third East-West Highway Project will assist in developing a comprehensive framework for the introduction of PBC in Georgia. This will give RD valuable experience in the design of appropriate incentives. D. Scope of Services 7. The objective of this TA is to build on the policy commitment of GoG and build technical and other capacity in RD and MRDI such that they are in a position where they could issue a request for a technical and financial proposal for a PPP arrangement and be assured of getting adequate, cost-effective bids. 8. At the end of the TA the consultants should (i) have assisted the counterparts and the IA to draft any legislative changes required, (ii) have identified a pipeline of projects suitable for PPP and have discussed the projects with the IA, (iii) have drafted model PPP contracts and (iv) have, with the counterparts, developed a coherent PPP procurement strategy for the IA. It should enable the IA to move a step further than PBCs allowing a deeper involvement by the private sector in the provision and upkeep of roads in Georgia.

Engconsult Ltd. Appendix 15 89

E. Scope of Work 9. The international and national consultants will work with national counterparts from RD and MRDI. The work will consist of five parts (i) a review PPP options suitable for Georgia and of any existing contracts, (ii) a review of existing legislation relating to PPP and recommendations for new or changed legislation, (iii) economic and financial studies, (iv) preparation of standard forms of contract and (v) a procurement strategy. 1. PPP Options Overview 10. PPP projects need to meet the same criteria as fully public ones to qualify for ADB support. They should maintain or improve transport access and affordability to the poor. They must meet safeguards requirements. They should be economically justified. Properly structured private involvement should deliver risk transfer and efficiency benefits. 11. The options overview will take a critical look at RD’s operations and objectives. There is scope for PPP approaches to deliver additional capacity in the form of major highways, bridges and tunnels. It is likely that multi-year, area-wide road maintenance contracts and concessions will provide increasing opportunities for the private sector. Intelligent Transport Systems (ITS) technologies are also possible candidates for PPP as technology risks may be partly defrayed to the private sector. Thus the Consultant will identify (i) the scope for private sector involvement and (ii) the most appropriate form of PPP in each case. The Consultant will also review past efforts, whether successful or not, at transport PPPs in Georgia in order to spell out lessons learned. 2. Legal review 12. For most major transport projects PPP financing requires sophisticated legal and enforcement legislation. The review will cover: primary legislation, powers of the client agencies, rights to award concessions, constitution of client agencies and shareholders’ agreements, regulatory powers and structures, ownership of assets, labor law, rights of access and arbitration. 13. The Consultant will highlight recommended changes to the legal environment where required and draft amendments. 14. The Consultant will consider and make clear recommendations on the need for regulation. (Regulation by contract is typically used in the transport sector for specific projects rather than industry-wide regulation. Toll-road concessions are often regulated by contract. A regulatory entity is not required if there is sufficient confidence that contract law and arbitration arrangements can provide a remedy to the parties in the event of dispute). 3. Economic and financial aspects 15. It is more difficult to recover road infrastructure costs directly from user charges than it is to charge for transport services. The financial returns from infrastructure projects are often poor in the short term and only become satisfactory in the very long term and they are, therefore, risky. These risks are often not attractive to private investors without some public funding or public risk-taking, or government guarantees. Where transport infrastructure costs are not recovered directly, there are distributive consequences which may be politically significant (i.e. favor the better off). 16. The Consultant will screen potential PPP projects with the following criteria in mind: (i) economic internal rate of return, (ii) financial internal rate of return and financial ratios, (iii) need for subsidy or guarantees and (iv) fiscal and distributive consequences of a PPP arrangement. The Consultant will review and recommend appropriate payment methods: tolls may not be the

Engconsult Ltd. Appendix 15 90 most efficient method and the Consultant should consider alternatives such as shadow tolls, availability payments and annuity payments. 17. In each case the Consultant will prepare a public sector comparison case in order to calculate the incremental net benefits that may be obtained as a result of the PPP. Benefits may accrue from earlier implementation (particularly if there are public sector budgetary constraints), lower whole life costs and possibly better service. 18. The Consultant will also conduct a participatory workshop with relevant stakeholders, using appropriate models2 to demonstrate the economic and financial performance of candidate projects, the need for subsidies and the distributive consequences of PPP. 19. Subsidies and/or guarantees are likely to be required for the more capital-intensive PPPs. The Consultant will therefore establish the potential sources (both government and International Financial Institutions (IFIs)) of subsidy and the conditions required. 4. Contract Review 20. The Consultant will review suitable forms of contract and prepare model forms appropriate to potential PPP arrangements under the auspices of RD. The World Bank’s toolkit3 is a good starting point, but the Consultant should seek to base its recommendations on forms of contract common in Georgia or from their own international experience. 5. Procurement Strategy 21. The Consultant will devise a procurement strategy to cover all phases of PPP, from feasibility through to contract award. It will undertake a pre-marketing exercise with likely national and international bidders in order to ascertain the level of interest and bidders’ requirements. F. Reporting 22. The Consultant will submit (i) an inception report within four weeks of project start, (ii) a diagnostic report within eight weeks of project start, (iii) a draft final report 17 weeks after project start and (iv) a final report four weeks after receiving comments from ADB and RD. The diagnostic report will contain: (i) lessons learned from existing or attempted PPPs in Georgia, (ii) the constraints (legal, financial, institutional) that create barriers to involvement of the private sector. G. Workshops 23. Two workshops will be held. The first will be for senior officials of the EA and IA. Its aim will be to provide feedback on the Consultant’s proposed project pipeline. The second will be an analytical workshop for counterparts at which participants will work through the financial and economic principles of candidate projects. H. Staffing 24. The international PSP expert/team leader shall have a minimum of 15 years relevant experience of which 5 years should be on international PPP projects. He/she should have held a team leader position on at least two similar assignments and be familiar with IFI requirements. The other international experts should hold relevant post-graduate qualifications and a minimum

2 e.g. taken from the World Bank’s toolkit: Public-Private Options for Developing, Operating and Maintaining Highways: Toolkit for Policymakers, 2003 3 Ibid

Engconsult Ltd. Appendix 15 91 of 10 years relevant work experience. The deputy team leader (national) shall have a relevant post-graduate qualification and should have had at least five years experience of donor funded projects. The other national experts should hold relevant qualifications and a minimum of 5 years relevant work experience. All consultants should be fluent in spoken and written English. 25. The international PSP expert will be team leader and will be appointed for a period of four months. He/she will be responsible for co-coordinating the inputs and activities of the international and national specialists and will take the lead role in developing the project pipeline and facilitating the first workshop and preparing the procurement strategy. He/she will be supported by a national PSP expert as deputy team leader. 26. The international and national legal experts will each be fielded for two months and be responsible for carrying out the legal review. The international expert will be responsible for the comparison with international best practice. 27. The economist/financial international and national specialists will each be fielded for three months. They will assist the PPP specialists in screening potential projects, and will carry out economic and financial analyses for each potential project. They will facilitate the second workshop. 28. The international and national contract specialists will each be appointed for 3 months. They will review and recommend suitable contract forms suitable for Georgia and assist the Team Leader in preparing the procurement strategy.

Table A15. 1: Summary of International Consulting Services

Consultant International National 1. Team Leader/PPP Specialist 4 2. Deputy Team Leader/PPP Specialist 4 3. Legal Specialist 2 2 4. Economist/Financial Analysts 3 3 5. Contracts Specialist 3 3

Engconsult Ltd. Appendix 15 92

COST ESTIMATES AND FINANCING PLAN ($'000) Total Item Cost A. ADB Financing 1. Consultants a. Remuneration and Per Diem i. International Consultants 283.0 Number of Person-months 12 ii. National Consultants 50.0 Number of Person-months 12 b. International and Local Travel 45.0 c. Reports and Communications 5.0 2. Equipment (PCs, presentation equipment)a 5.0 3. Workshop 10.0 4. Translation Costs 10.0 5. Miscellaneous Administration and 5.0 Support Costs 6. Contingencies 41 Subtotal (A) 454.0 B. Government Financing 1. Office Accommodation 2. Remuneration and Per Diem of Counterpart Staff 3. Venue for Workshops and Seminars 4. Previous Studies and Reports 5. Miscellaneous Administration Cost

Engconsult Ltd. Appendix 16 93

Appendix 16 SUMMARY POVERTY REDUCTION AND SOCIAL STRATEGY

Country/Project Title: GEO: Sub Regional Road Corridors Development Program-Tranche-1

Department Lending/Financin Multitranche Financing Facility (MFF) / CWRD/CWID g Modality: Division:

I. POVERTY ANALYSIS AND STRATEGY A. Linkages to the National Poverty Reduction Strategy and Country Partnership Strategy

The Project under the Sub Regional Roads Corridors Development Program can be linked indirectly to the overall objective of Government’s Development Agenda as reflected in Basic Data and Directions (BDD) document- 2008-2011 and ADB’s Interim Operational Strategy (IOS) 2008-2009. Poverty Reduction is one of the Governments main agenda where the prime objectives are to reduce, over the period 2006-2015, overall poverty incidence from 31% to 15% and extreme poverty from 11% to 4%. This can be achieved through economic growth where infrastructure development plays a vital role and therefore, Infrastructure bottlenecks to growth will be removed. One of the important strategies under this is to develop transport infrastructure to strengthen market access and to realize Georgia’s transit economy potential more fully. ADB’s priorities for the IOS (2008–2009) have been identified based on the considerations such as: (i) alignment with the Government’s development agenda; (ii) selectivity in focusing limited resources on a small number of priority areas; and (iii) complementarities with other development partners. The IOS will focus on enhancing sustainable economic growth, with the crosscutting themes of governance, regional cooperation, and environmental protection, by: (i) improving service delivery in municipal infrastructure within the evolving decentralization process, (ii) reducing road transportation constraints on economic activity, and (iii) upgrading and developing energy infrastructure. ADB’s operational engagement in this area will involve financing improvements to the highway network, resulting in lower transport costs, better road safety, and reduced traffic congestion and vehicle pollution. These are high- return, in demand operations, and the Government has requested ADB assistance in this area to help address some of the remaining investment gaps.

Georgia’s road network consists of 1,495 kilometers (km) of international, 5,446 km of state, and 13,388 km of local roads. The country has five main roads and highways, including the East–West Highway connecting the ports of Poti and Batumi to the Azerbaijan border. The Government has been encouraging private sector participation in construction and maintenance of highways. Given the country’s location, transport sector development is critical to reducing the logistics costs of economic activity and realizing Georgia’s full potential as a transit economy. The difficult terrain in some sections and congestion slows traffic, increases vehicle operating costs, and reduces road safety. Road maintenance costs are also high because of the terrain and weather conditions. About 6% of the main roads, 24% of the secondary roads, and 50% of the local roads are in poor condition and need rehabilitation. The Government has prioritized improving conditions of the road network. The link between access to infrastructure services and poverty incidence has been well-established in cross-country studies. For Georgia, rural-urban differences in access to infrastructure services go some way toward explaining the pattern of poverty incidence. Because of the generally poor condition of secondary and local roads, Georgia’s rural population, including ethnic minorities living in southern Georgia, face problems of isolation, with inadequate access to markets and to social services. In rehabilitating/upgrading secondary and local roads, it will be important to build in participatory mechanisms involving local communities in design and maintenance of such roads. Therefore, the project will have indirect linkage to the overall development goal. B. Poverty Analysis Targeting Classification: General Intervention 1. Key Issues Georgia’s rank on the human development index computed by the United Nations Development Program (UNDP) is 97 out of a list of 177 countries. Rank-wise, it is better placed than Azerbaijan (99) but not so with respect to Armenia (79). Although growth will need to be more broad-based to achieve the income poverty reduction target, the policy and institutional environment for achieving the Millennium Development Goals (MDGs) has improved in recent years with higher and more efficient spending on social sectors. More than one third of Georgia’s

Engconsult Ltd. Appendix 16 94 population fall below the national poverty line with relatively higher poverty in rural areas. Poverty incidence in Georgia is close to levels in other South Caucasus countries. The incidence of Poverty in Georgia is 39.4% and the depth of poverty 13.5%. Direct benefit of the Program on poor is relatively small except the fact that it provides some direct employment opportunities during the construction phase. However, it contributes more indirectly. The provision of transport services, including the construction and maintenance of transport infrastructure, generates demand for labor (often unskilled labor) and provides income-earning opportunities for the poor. If a transport project generates jobs for the poor who are otherwise unemployed or under-employed, it contributes indirectly to the poverty reduction. The construction aspect of transport sector development is often viewed equally as important as the service aspect of the sector in promoting economic growth. The process of poverty reduction is embedded in a broad range of socio-economic activities to which transport services provide intermediate inputs. There will be both short term and long term positive impacts of the proposed Project. The short term positive impacts include access to markets for agricultural produce, availability of temporary unskilled jobs for villagers; and opportunity to provide food and restaurant services for construction workers. On the other hand, the long term positive impacts include, access to long distance transport services across the border; access to long distance markets (buying and selling through regional trade); easier access to health facilities and medical treatment; increased access to agricultural extension services, increased access to education; increased opportunity for the development of tourism; and diversification of income sources.

2. Design Features. No direct impact on poverty is envisaged to occur as result of the implementation of this Project. Nonetheless, the Project will have indirect impacts on Georgia as a whole. II. SOCIAL ANALYSIS AND STRATEGY Findings of Social Analysis The program will have positive socio-economic impact on Georgia. A social analysis has been carried out in the project area through a socio-economic household’s survey of 200 households including 20 villages. Out of the total sampled surveyed household in the project area, 70 % are covered in villages and the rest 30 % are from urban areas. Among the head of the households, about 83 % are males and 17% are females. Most of the sampled households (about 97.5 %) are of Georgian ethnicity. The sex ratio of the project area favors the females (1,004 females to 1,000 males). The average family size is 5.04 members per household. There is almost 100% literacy rate in the project area. A majority (about 59 %) have secondary level education and about 16 % have university level of education (same for both genders). Agriculture is practiced by 81 % of the households as primary economic activity followed by government services (33.17 %) and business and trading (8.91%). Out of 39 households professing secondary economic activities, nearly 9 % are practicing agriculture and 3.47 % each are either doing government jobs or are having own business. Among the household members in the age group of 15 years and above, 70 % are found to be unemployed (65% males and 75% females). About 6.43 % of the households are landless. Almost all the land is cultivable and unirrigated. The average residential land as possessed by the people is 0.022 hectares. The total annual average income per family is 6,024 Lari and the median annual income is 4,660 Lari. House hold assets like television (97.5%), LPG gas (78.2%) and refrigerator are commonly possessed by the sampled households. Large animal like cow/bulls/ buffaloes are possessed by 57 % of the households and poultry is possessed by nearly 23 % of the households. About 40 % of the households reported having loans and most of them have taken loan from the bank. Most of the structures and building are double storied and 88% of the structures are cemented structures. 98.5 % of the households are connected to central power supply. The average number of rooms available within the sampled house hold is 6.11. Only 2.5 % households reported having connection to the NG pipe line. Piped water supply is the common source of water for half of the households. About 48.51 % are having flush toilets and the other are having latrines. 94 % of the households are using wood as the fuel for heating. About 89 % of the sampled households in the project area had access to a health care center. The average distance of the health centers is 3.1 kms; for 8.23% the health center is located within 1 km from their house. 95 % had access to schools for their children. About 94.55 % households are immediately connected a to road. For 86% of the households, the immediate approach road is the village road and for 10 % it is the highway. Based on the survey and the analysis, it is derived that almost 52% of the households are living below the poverty line in the project area. B. Consultation and Participation

Engconsult Ltd. Appendix 16 95

1. Provide a summary of the consultation and participation process during the project preparation. Consultations were carried out with all the stakeholders during the program preparation. As an integral part of the safeguards planning, affected communities and APs were directly and fully involved in the project preparation and during the design. All stakeholders were informed and the stream of information will continue during the implementation of the project. Extensive consultation events were arranged at various stages of social and resettlement project preparation i.e., alternative alignment feasibility study, cadastral survey of affected lands, land census survey, socio-economic survey and targeted consultation meetings with affected people and local stakeholder organizations and individuals. Consultations were also held at various office levels such as, Department of Roads (RD) including the Resettlement Division (DRD), National Agency for Public Registry under the ministry of Justice, Local Government at Rayon Level (District Municipality), Gamgebeli, Property Recognition Committee in the Rayon Level and Sacrebulo. Community consultations were held through Focus Group Discussion at various locations, especially at 20 villages during the month of April 2009 which involved both men and women participants and totaling 210 participants. The consultation will be continued throughout the project cycle.

2. What level of consultation and participation (C&P) is envisaged during the project implementation and monitoring? Information sharing Consultation Collaborative decision making Empowerment

3. Was a C&P plan prepared? Yes No Local communities and APs were consulted throughout the preparation process to minimize impacts, ensure social acceptance and increase equitability. The process will indeed continue during the implementation of the Project. Views and concerns of all stakeholders will be taken into consideration and addressed whenever feasible.

C. Gender and Development 1. Key Issues. The Program will pay particular attention to ensure that women are the recipients of the compensation pertaining to their activities and to ensure that women who are de-facto household heads are clearly listed as beneficiaries of compensation and rehabilitation proceedings under the loan. Women will also be participants in the consultation processes to determine and negotiate for compensation entitlements and implement the LARP. Special attention will be given to the impact of resettlement on women and other vulnerable groups during monitoring and evaluation of the LARP. The Program will have a positive impact on gender, because the civil works contracts will include provisions to encourage employment of women during implementation, and women will be encouraged to participate in activities to monitor program impacts. No particular issue is expected to be rising from the implementation of the project. Nonetheless, additional compensations will be awarded to female- headed households as considered to be more vulnerable. 2. Key Actions. Measures included in the design to promote gender equality and women’s empowerment— access to and use of relevant services, resources, assets, or opportunities and participation in decision-making process: Gender plan Other actions/measures No action/measure III. SOCIAL SAFEGUARD ISSUES AND OTHER SOCIAL RISKS Issue Significant/Limited/ Strategy to Address Plan or Other Measures No Impact Issue Included in Design Significant Impact A full Land Acquisition Involuntary and Resettlement Plan Full Plan Resettlement Land acquisition will be required (LARP) has been Short Plan for the construction of the prepared for the Resettlement Bypass Road. The Project will tranche 1 components Framework entail substantial land based on the No Action acquisition and resettlement. A engineering design. detailed impact assessment The LARP is based on including the census survey is in the ADB’s policies on progress which will have exact Involuntary

Engconsult Ltd. Appendix 16 96

figures related to land Resettlement and acquisition and resettlement. Georgian laws and However, a preliminary regulations. All the assessment has been done losses have been which shows that approximately, assessed and 11.06 hectares (HA) of private provisions have been land will be required for land made to the APs for acquisition. A total of 590 land compensation and parcels will be affected. assistance including the Additionally 44 non titleholders. structures/buildings will be Additional assistances affected. The total amount of have been provided to state land required is 109.70 HA. the vulnerable Also, 7.26 HA of land will be households. required additionally whose Additionally a LARF status is yet to be confirmed has been prepared for during the survey. The the entire MFF which approximate total number of will be followed and affected household/family implemented for the (AP/AH) is 3,500. preparation of subsequent LAPs for future tranches. No Impact Not Required Indigenous Peoples No impacts on Indigenous Plan Peoples are expected for the Other Actions Program. An assessment of Indigenous Peoples impact on indigenous peoples Framework was undertaken in accordance No Action with ADB’s Policy on Indigenous Peoples (1998). The Program will primarily affect Georgian people which are almost 93% of the total population. Some of the other ethnic groups are also present in the program area, which are insignificant in number. These groups are mostly Armenian, Azeri, and Russian. These groups have been fully integrated into institutional, cultural, and economic processes in the country, and they do not display sufficient features to classify them indigenous peoples. These groups are quite mainstreamed and cannot be considered as Indigenous People as per the ADB’s definition on IP No Impact The tranche 1 work will Labor create both direct and Employment Jobs opportunities will open for indirect employment Plan opportunities skilled and unskilled labor opportunities for the Other Action Labor retrenchment during construction local people. The No Action Core labor issues related to safety

Engconsult Ltd. Appendix 16 97 standards and safeguard of laborers including employment and equal payment for women and no-employment of children in the project will be taken care by making provisions in civil work contract. Specific assurances will be added for the EA to ensure the enforcement of CLS by the contractor on the work camps/sites. No Impact. No tolls will be collected Affordability on the subproject road Action and it will lead to No Action savings on vehicle operating cost, which will lower transport costs. Other Risks and/or Limited It is expected that Vulnerabilities project is likely to bring Plan HIV/AIDS in some risks related to Other Action Human trafficking HIV/AIDS transmission No Action Others(conflict, as a result of improved political instability, etc), mobility of the people please specify and influx of labor at the time of the construction. This aspect will be studied further during the detailed design stage and mitigation measures will be developed to counter the spread of sexually transmitted diseases in the project area. IV. MONITORING AND EVALUATION Are social indicators included in the design and monitoring framework to facilitate monitoring of social development activities and/or social impacts during project implementation? Yes □ No

Engconsult Ltd. Appendix 17 98

Appendix 17 SUMMARY LAND ACQUISTION AND RESETTLEMENT FRAMEWORK (LARF)

A. Introduction 1. This Land Acquisition and Resettlement Framework (LARF) for the Sub regional Road Corridors Development Program (SRCDP) has been prepared by The Road Department of the Ministry of Regional Development and Infrastructure of Georgia (RDMRDI) and is fully endorsed by the RDMRDI. The Project, to be financed by ADB under the Multi Tranche Financial Facility (MFF) will be implemented over a period of 5 years. The objective of this document is to provide guidance in the preparation/implementation of Land Acquisition and Resettlement (LAR) tasks for the projects under each tranche and to establish the necessary covenants to ensure that this is done in compliance to ADB’s Policy on Involuntary Resettlement and procedures for the MFF. B. LAR-Related Program Processing Requirements 2. Based on ADB policy/practice appraisal of the MFF and each tranche and approval of the implementation of tranche subprojects will require the preparation of the following: (i) A LARF for the whole financial facility and applicable to all subprojects. The LARF will be reviewed, if necessary updated, and submitted for ADB approval at least annually and always at the start of the preparation of each tranche. (ii) An Initial Poverty and Social Assessment (IPSA) indicating, for each tranche, whether LAR impacts are likely to occur, type of impacts, likely magnitude, and whether there may be Indigenous Peoples (IP) affected, and; (iii) If LAR occurs, a LAR Plan (LARP) for each project under a tranche based on detailed design and commensurate to the severity of impacts1. The LARP will include detailed compensation and administration budgets and implementation schedules linking LAR tasks to the initiation of civil works.

3. Based on ADB policy/practice, the appraisal of the MFF and each specific tranche and the approval of project implementation will be based on the following LAR-related conditions: (i) MFF/first tranche appraisal: Conditional to preparation/disclosure of a LARF for the whole MFF acceptable to ADB and of the LARPs for the tranche projects requiring LAR. (ii) Following tranches appraisal: Conditional to review/update/disclosure of the LARF, and preparation/disclosure of needed LARPs consistent with the revised LARF and acceptable to ADB for projects with LAR. (iii) Contract awards signing: Conditional to the update/disclosure of the revant LARPs acceptable to ADB. The updated LARPs will reflect final impacts, final AP lists and final compensation rates and will be readily implementable. (iv) Provision of notice to proceed to contractors: Conditional to the full implementation of the relevant LARP (full delivery of compensation and rehabilitation)

1 Based on ADB Operation Manual (OM) F2/OP and BO (2006) a project is classified as Category “A” and a full LARP is needed if > 200 people suffer significant impacts (relocation or loss of >10% of productive assets). A project will instead be classified as Category “B” when less than 200 people suffer significant impacts. Category “C” projects have no LAR impacts. No subproject will affect indigenous Peoples. The WB does not have a similar resettlement classification system but its policy also envisages that for minor impacts a shorter document, called “Abbreviated Resettlement Plan”, is prepared.

Engconsult Ltd. Appendix 17 99

for the relevant project. Such a condition will be clearly spelled out in the text of the civil works contract.

C. Indigenous Peoples and Vulnerable Groups including Women 4. No impacts on Indigenous Peoples are expected for the Program. Special attention will also be given to identifying and addressing the needs of disadvantaged groups such as the landless, the poor, female-headed households, the elderly and the disabled, through measures included in the LARP to try and improve (over and above cash compensations and restoration of) their livelihoods. Women have important economic roles in project areas and engage in a very wide range of income making activities in the agricultural and marketing sector. The project will pay particular attention to ensure that women are the recipients of the compensation pertaining to their activities and to ensure that women who are de-facto household heads are clearly listed as beneficiaries of compensation and rehabilitation proceedings under the loan. D. Legal and Policy Background 5. The legal framework will be based on both Georgian Laws/ legislation and ADB’s Policy on Involntary Resettlelement (1995). In Georgia, the legislative acts that regulate the issues of obtaining State ownership rights to privately owned land parcels based on the necessary public needs caused due to road constructions activities are : (i) The Constitution of Georgia, August 24, 1995; (ii) The Civil Code of Georgia, June 26, 1997, (iii) The Law of Georgia on Protection of Cultural Heritage; (iv) The Law of Georgia on Notary Actions; (v) The Law of Georgia on Privatization of State-owned Agricultural Land, July 8, 2005’ (vi) The Law of Georgia on Ownership Rights to Agricultural Land, March 22, 1996; (vii) The Law of Georgia on Registration Ownership Rights to Immovable Property, December 28, 2005, (viii) The Law of Georgia on the Rules for Expropriation of Ownership for Necessary Public Need, July 23, 1999; and (ix) The Civil Procedural Code of Georgia, November 14, 1997. Based on the Georgian laws on land acquisition and ADB’s Policy on Involuntary Resettlement, 1995, core involuntary resettlement principles are developed for this MFF which are as follows: • Land acquisition, and other involuntary resettlement impacts will be avoided or minimized exploring all viable alternative project designs; • where unavoidable, a time-bound LARP will be prepared and APs will be assisted in improving or at least regaining their pre-program standard of living; • consultation with APs on compensation, disclosure of resettlement information to APs, and participation of APs in planning and implementing sub-projects will be ensured; • vulnerable and severely affected APs will be provided special assistance; • Non-titled APs (e.g., informal dwellers or squatters, APs without registration details) will receive a livelihood allowance in lieu of land compensation and will be fully compensated for losses other than land.; • Legalizable APs will be legalized and fully compensated for land losses. • Provision of income restoration and rehabilitation; • The LARP will be disclosed to the APs in the local language • Payment of compensation, resettlement assistance and rehabilitation measures will be fully provided prior to the contractor taking physical acquisition of the land and prior to the commencement of any construction activities on a particular package and • Establishment of appropriate grievance redresses mechanisms to solve APs grievance if occurs.

Engconsult Ltd. Appendix 17 100

E. Compensation Eligibility and Entitlements 6. LAR tasks under the Project will be implemented according to a compensation eligibility and entitlements framework in line with both Georgia laws and regulation and ADB Policy. APs entitled for compensation or at least rehabilitation provisions under the Project are: (i) All APs losing land either covered by legal title/traditional land rights, Legalizable, or without legal status; (ii) Tenants and sharecroppers whether registered or not; (iii) Owners of buildings, crops, plants, or other objects attached to the land; and (iv) APs losing business, income, and salaries. Compensation eligibility will be limited by a cut-off date to be set for each subproject on the day of the beginning of the AP Census and detailed Measurement Survey (DMS). A summary entitlements matrix is included in Table A17. 1below. Table A17. 1: Compensation Matrix Type of Loss Application Definition of APs Compensation Entitlements F. Land Permanent loss of AF losing agricultural Owner with full registration Cash compensation at agricultural land land regardless of replacement cost. impact severity Legalizable Owner These AP will be legalized and provided with cash compensation at replacement cost. Informal Settlers/ APs with One time self-relocation no registration/valid allowance in cash equal to 1 year documentation at minimum salary Non-Agricultural Land AF losing their Owner with full registration Cash compensation at commercial/ residential replacement cost. land Legalizable Owner APs will be legalized and provided with cash compensation at replacement cost.. Renter/Leaseholder Rental allowances in cash for 3 months Informal Settlers/ APs with One time self-relocation no registration/valid allowance in cash equal to 1 year documentation at minimum salary Buildings and Structures Residential and non All AFs regardless of their Full impact: Cash compensation residential legal ownership/ for loss of building/ structures at full structures/assets registration status replacement costs free of (including legalizable and depreciation and transaction costs Informal Settlers) Partial impact: compensation for repairs Loss Of Community Infrastructure/Common Property Resources Loss of common Community/Public Community/Government Reconstruction of the lost structure property resources Assets in consultation with community and restoration of their functions Loss of Income and Livelihood Crops Standing crops All AFs regardless of legal Crop compensation in cash at affected status (including market rate by default at to gross legalizable and Informal crop value of expected harvest. Settlers) Trees Trees affected All AFs regardless of legal Cash compensation at market status (including rate on the basis of type, age and

Engconsult Ltd. Appendix 17 101

Type of Loss Application Definition of APs Compensation Entitlements legalizable and Informal productive value of the trees. Settlers) N. Business/Employment Business/employment All AFs regardless of legal Owner: (i). (permanent impact) cash loss status (including indemnity of 1 year net income; (ii) legalizable and Informal (temporary impact) cash indemnity of Settlers) net income for months of business stoppage. Assessment to be based on tax declaration or, in its absence, minimum salary.. Permanent worker/employees: indemnity for lost wages equal to 3 months of minimum salary. Allowances Severe Impacts >10% income loss All severely affected AFs Agricultural income: 1 additional including informal settlers crop compensation covering 1 year yield from affected land. Other income: 1additional compensation for 3 months of minimum salary. Relocation/Shifting Transport/transition All AFs to be relocated Provision of sufficient allowance to costs cover transport expenses and livelihood expenses for the transitional period (up to 1 month). Vulnerable People AFs below poverty line, Allowance equivalent to 3 months of Allowances headed by Women, minimum salary and employment disabled or elderly priority in project-related jobs Temporary Loss Temporary impact All AFs Due compensation will be during construction assessed and paid based on this LARF during construction. Unforeseen Road Department and the resettlement impacts, construction contractor will address if any and mitigate/compensate unforeseen resettlement impact during project

O. Public Consultation, Participation, Documents Disclosure and Grievance Redress 7. Concerned officials of centre, district/rayon, municipalities and villages/sacrebolu will be informed about the Project, and their assistance will be solicited in the conduct of the inventory of affected assets and the Census of APs and the DMS. Also, prior to the finalization of the LARP and its submission to Project authorities, the APs will be thoroughly informed on the results of the Census and DMS, and their preferences on compensation or other resettlement assistance will be given due consideration. This LARF in Georgian will be disclosed on the RDMRDI website and at RDMRDI offices before Project appraisal. The LARF in Georgian will also be disclosed to the APs at the relevant Rayon office (Gamgebeli) and at village administration (Sacrebolu) once subprojects are identified. Its English version will be disclosed on the ADB website prior to Project appraisal and after the LARF is endorsed by the Executing Agency (EA) which will be RDMRDI in this case. Once a LARP for a subproject has been prepared and approved by RDMRDI and ADB it will be disclosed at relevant Rayon office (Gamgebeli) and at village administration (Sacrebolu). A pamphlet in Georgian, summarizing compensation eligibility and entitlement provisions, will be sent to all AP/AFs before the initiation

Engconsult Ltd. Appendix 17 102 of the compensation/rehabilitation process and before signing contract awards. The consultation process will be continued throughout the project cycle. 8. A grievance mechanism will be available to allow an AP appealing any disagreeable decision, practice or activity arising from land or other assets compensation. APs will be fully informed of their rights and of the procedures for addressing complaints whether verbally or in writing during consultation, survey, and time of compensation. Care will always be taken to prevent grievances rather than going through a redress process. This can be obtained through careful LAR design and implementation, by ensuring full participation and consultation with the APs, and by establishing extensive communication and coordination between the affected communities, the EA, and local governments in general P. Institutional Arrangements 9. Asian Development Bank (ADB) will be the funding agency of the MFF. The Road Department of the Ministry of Regional Development and Infrastructure of Georgia (RDMRDI) will be the Executing Agency (EA) having the lead responsibility for road construction, as well as the implementation of this LARF and subsequently the LARP. RDMRDI has overall responsibility for the MFF. This includes preparation, implementation and financing of all LAR tasks and cross-agency coordination. RDMRDI will exercise its functions through its existing resettlement division (DRD) which will be responsible for the general management of the planning and implementation of all LAR tasks. RDMRDI with the assistance of the consultants will develop and implement the LARP for each project based on the policy and procedures set out in the LARF. In addition to the RDMRDI, a number of other government departments and private agents will play an instrumental role in the design, construction and operation of the project. Pursuant to the active legislation, the Ministry of Natural Resources and Environmental Protection is responsible for environmental issues. The Ministry of Justice is responsible for legal matters regarding land ownership, and National Agency of Public Registry (NAPR) within the Ministry of Justice is in charge of the registration of land ownership and its transfer through purchase agreement from landowners to the Road Department. The local government at Rayon and village level will also be involved. To allow an effective execution of all LAR related tasks some expansion of the capacity on LAR currently available at DRD/RDMRDI may be needed. Q. Resettlement Budget, Financing and Implementation Schedule 10. All LARP preparation and implementation costs, including cost of compensation and LAR administration, will be considered an integral part of Project cost and will be contributed as a counterpart fund by the Georgia Government, in particular RDMRDI. Each LARP will include a budget section indicating (i) unit compensation rates for all affected items and allowances, (ii) methodology followed for the computation of unit compensation rates, and (iii) a cost table for all compensation expenses including administrative costs and contingencies. Costs for external monitoring tasks and for the preparation of surveys and LARPs can be allocated under the loan. Being the project owner, RDMRDI is responsible for the timely allocation of the funds needed to implement the RPs. Allocations will be reviewed twice a year based on the budget requirements indicated by the LARPs. As per the LAR finances flow the budget for compensation and rehabilitation will be directly disbursed by RDMRDI to the AP. 11. Based on experience in Georgia the Preparation and implementation of a LARP may take up to a few months. All land acquisition, resettlement, and compensation will be completed before the start of civil works All land required will be provided free of encumbrances to the contractor prior to handing over of sub-project sites and the start of civil works.

Engconsult Ltd. Appendix 17 103

R. Monitoring and Evaluation 12. LAR tasks under the Program will be subjected to both internal and external monitoring. Internal monitoring will be conducted by DRD/RDMRDI. External monitoring will be assigned to an Independent Monitoring Agency (IMA) to be hired by RDMRDI and approved by ADB. The IMA will be selected among NGOs, academic Institutions, or consulting firms. ADB will advise RDMRDI on the IMA’s terms of reference once Project implementation has begun.

Engconsult Ltd. Appendix 18 104

Appendix 18 SUMMARY LAND ACQUISTION AND RESETTLEMENT PLAN FOR PROJECT 1

A. Introduction and Project Background

1. This is the Summary Land Acquisition and Resettlement Plan (LARP) which is based on the draft Land Acquisition and Resettlement (LAR) Plan (LARP) for Project (tranche) 1 of the Ajara By-pass as has been prepared by RDMRDI as part of the feasibility study for Tranche 1 of the Program. Its objective is to assess tranche 1 impacts and plan needed compensation/rehabilitation measures. The document fits relevant Georgia laws, the ADB involuntary Resettlement Policy (1995) and the Program's Land Acquisition and Resettlement Framework (LARF). Its preparation involved: (i) detailed measurement surveys, (ii) asset valuation/documentary research on affected plots, (iii) consultation with rayons (districts), and affected parties; (iv) a 100% AH census; and (v) a socio-economic surveys of the AH. LARP preparation involved intensive consultation with all the stakeholders including the Affected Persons (APs). The LARP has been prepared for the Tranche-1 project which comprises the Kobuleti bypass (28km) and 6 km of existing road widening. The project road will fall under the jurisdiction of Ajara and Guria region. The total length of the project road under Tranche-1 is 34 Kilometers (Kms). Sufficient consideration has been given during the inception and feasibility phases of the project preparation to minimize the adverse impacts on land acquisition. The assessment of losses and land acquisition is based on the preliminary design as part of the feasibility study. If required, the LARP will need to be updated based on the detailed engineering design. The assessment has been carried out based on a 50 meters right of way (25 meters either side of the central line). According to ADB policy/practice, the appraisal of the MFF and each specific tranche and the approval of project implementation will be based on the following LAR-related conditions:

(i) Appraisal of MFF and Tranche 1: Conditional to preparation/disclosure of a LARF for the whole MFF acceptable to ADB and preparation of the LARPs for the tranche projects requiring LAR. (ii) Appraisal of subsequent projects (tranches): Conditional to review/update/disclosure of the LARF, and preparation of needed LARPs consistent with the revised LARF and acceptable to ADB for projects with LAR. (iii) Contract award for each project: Conditional to the update/disclosure of the relevant LARPs based on the final design and accepted by ADB. The updated LARPs will reflect final impacts, final AP lists and final compensation rates and be readily implementable. (iv) Provision of notice to proceed to contractors: Conditional to the full implementation of the relevant LARP (full delivery of compensation and rehabilitation) for the relevant project. Such a condition will be clearly spelled out in the text of the civil works contract.

Engconsult Ltd. Appendix 18 105

B. Impacts on Land Acquisition And Resettlement

2. The tranche 1 project will entail substantial land acquisition and resettlement. A total of 894 land parcels will be affected out of which 740 land parcels are agricultural including pasture land. The total land required for acquistion is 163.38 HA. The amount of affected agricultural land is 134.81 HA and the amount of pasture land is estimated to be 26.08. Additionally, 2.03 HA of residential land and 0.36 HA of commercial land will be affected. The total number of affected structures and buildings are 154 out of which 123 are residential structures, 18 are commercial structures and the remaining 13 are community property resources which include school, government buildings. 2,020 fruit bearing trees will be affected and 200 timber trees will be affected due to the project. The total number of households losing business is 18 whereas 141 households are eligible for relocation allowances. The total number of affected vulnerable households is 304 which include women headed households, households below poverty line etc. The total number of affected households is 881 out of which 200 households are legal title holder, 119 households belong to non titleholders and 562 households belong to the legalizable category. The total number of affected persons is estimated to be 4,440. The summary of impacts on land A summary of acquisition and resettlement is given in Table 18-1.

Engconsult Ltd. Appendix 18 106

Table 18.1: Impacts on Land Acquisition and Resettlement (Tranche 1) Impacts Number/Amount 1 Total Number of Land parcels 894 2 Agricultural and Pasture Land Parcel 740 3 Parcel for Structures and Buildings 154 4 Total area of land Required (HA) 163.38 5 Total area of Agricultural Land (HA) 134.81 6 Total area of Non Agricultural/Pasture Land (HA) 26.08 7 Total Area of Residential Land (HA) 2.03 8 Total Area of Commercial Land (HA) 0.36 9 Total number of structures/Buildings 154 10 Total number of Residential Structures 123 11 Total number of Commercial Structures 18 12 Total number of Government and community property 13 13 Total Number of Structures for Relocation 141 14 Total number of Fruit Bearing Trees 2,020 15 Total number of Non Fruit Bearing Trees 200 16 Total Number of HH losing business 18 17 Total number of severely affected HH 18 18 Total Number of Affected Households (AHs) 881 19 Total Number of Vulnerable Households 304 20 Total number of Titleholders 200 21 Total number of non-title holders 119 22 Total Number of Legalizable Households 562 23 Total Number of Affected Persons (APs) 4,440

C. Indigenous Peoples and Vulnerable Groups including Women

3. An assessment of impact on indigenous peoples was undertaken in accordance with ADB’s Policy on Indigenous Peoples (1998). No impacts on Indigenous Peoples (IP) are expected for the Project. The Project will primarily affect Georgian people which are almost 97.50% of the total population. The other ethnic group found in the project area is Armenian and their number is very small. These groups have been fully integrated into institutional, cultural, and economic processes in the country and they do not display sufficient features to be classified as indigenous peoples as per the ADB’s definition on IP. Therefore, the Program is classified category C. Special attention is given to identifying and addressing the needs of disadvantaged groups such as the poor and female-headed households through measures included in the LARP to try to improve (over and above cash

Engconsult Ltd. Appendix 18 107

compensations and restoration of) their livelihoods. Women have important economic roles in the project area and engage in a very wide range of income making activities in the agricultural and marketing sector. The project will pay particular attention to ensure that women are the recipients of the compensation pertaining to their activities and to ensure that women who are de-facto household heads are clearly listed as beneficiaries of compensation and rehabilitation proceedings under the loan.

D. Resettlement Policy, Principles and Legal Framework

4. The legal framework is based on both Government of Georgian laws/ legislation and ADB’s Policy on Involntary Resettlelement (1995). In Georgia, there are several legislative acts that regulate the issues of state's obtaining privately owned land parcels based on the necessary public needs such as public road construction: (i) The Constitution of Georgia, August 24, 1995; (ii) The Civil Code of Georgia, June 26, 1997, (iii) The Law of Georgia on Protection of Cultural Heritage; (iv) The Law of Georgia on Notary Actions; (v) The Law of Georgia on Privatization of State-owned Agricultural Land, July 8, 2005; (vi) The Law of Georgia on Ownership Rights to Agricultural Land, March 22, 1996; (vii) The Law of Georgia on Registration Ownership Rights to Immovable Property, December 28, 2005, (viii) The Law of Georgia on the Rules for Expropriation of Ownership for Necessary Public Need, July 23, 1999; and (ix) The Civil Procedural Code of Georgia, November 14, 1997. Based on the Georgian laws on land acquisition and ADB’s Policy on Involuntary Resettlement, 1995, core involuntary resettlement principles are adopted for tranche 1 components as follows:

• Land acquisition, and other involuntary resettlement impacts will be avoided or minimized through all viable alternative project designs; • where unavoidable, a time-bound LARP will be prepared and APs will be assisted in improving or at least regaining their pre-program standard of living; • consultation with APs on compensation, disclosure of resettlement information to APs, and participation of APs in planning and implementing sub-projects will be ensured; • vulnerable and severely affected APs will be provided special assistance; • non-titled APs (e.g., informal dwellers or squatters, or APs without registration details) will receive a livelihood allowance in lieu of land compensation and will be fully compensated for losses other than land; • legalizable APs will be legalized and fully compensated for land losses; • provision of income restoration and rehabilitation will be ensured; • the LARP will be disclosed to the APs in the local language; • payment of compensation, resettlement assistance and rehabilitation measures will be fully provided prior to the contractor taking physical acquisition of the land and prior to the commencement of any construction activities on a particular package; and • appropriate grievance redresses mechanisms will be established to solve APs grievance if occurs.

Engconsult Ltd. Appendix 18 108

E. Compensation Eligibility and Entitlements

5. Land acquisition and resettlement tasks under the Project (tranche) 1 will be implemented according to a compensation eligibility and entitlements in line with both Georgian laws and regulation and ADB’s policy. APs entitled for compensation or at least rehabilitation provisions under the Project 1 are: (i) all APs losing land either covered by legal title/traditional land rights, legalizable, or without legal status; (ii) tenants and sharecroppers whether registered or not; (iii) owners of buildings, crops, plants, or other objects attached to the land; and (iv) APs losing business, income, and salaries. Compensation eligibility is limited by a cut-off date on the day of the beginning of the AP Census and detailed measurement survey (DMS) which is June 2009. A summary entitlements matrix is included in Table 18-2 below.

Engconsult Ltd. Appendix 18 109

Table 18.2: Entitlement Matrix Type of Loss Application Definition of APs Compensation Entitlements Land Permanent loss of AF losing agricultural land Owner with full registration Cash compensation at agricultural land regardless of impact severity replacement cost or through replacement land equal in value/productivity to the plot lost and at location acceptable to APs where feasible Legalizable Owner These AP will be legalized and provided with cash compensation at replacement cost. Informal Settlers/ APs with no One time self-relocation registration/valid documentation allowance in cash equal to 1 year at minimum salary - Non-Agricultural Land AF losing their commercial/ Owner with full registration Cash compensation at residential land replacement cost or through replacement land equal in value to the plot lost and at location acceptable to APs where feasible Legalizable Owner APs will be legalized and provided with cash compensation at replacement cost.. Informal Settlers/ APs with no One-time self-relocation registration/valid documentation allowance in cash equal to 1 year at minimum salary

Buildings and Structures Residential and non All AFs regardless of their legal Full impact: Cash compensation residential ownership/ registration status for loss of building/ structures at full structures/assets replacement costs free of depreciation and transaction costs Partial impact: compensation for repairs Loss Of Community Infrastructure/Common Property Resources Loss of common Community/Public Assets Community/Government Reconstruction of the lost structure property resources in consultation with community and restoration of their functions Loss of Income and Livelihood Crops Standing crops affected All AFs regardless of legal Crop compensation in cash at status (including legalizable and market rate by default at the gross Informal Settlers) crop value of expected harvest.

Trees Trees affected All AFs regardless of legal Cash compensation at market status (including legalizable and rate on the basis of type, age and Informal Settlers) productive value of the trees.

Business/Employment Business/employment loss All AFs regardless of legal Owner: (i). (permanent impact) cash status (including legalizable and indemnity of 1 year net income; (ii) Informal Settlers) (temporary impact) cash indemnity of net income for months of business

Engconsult Ltd. Appendix 18 110

Type of Loss Application Definition of APs Compensation Entitlements stoppage. Assessment to be based on tax declaration or, in its absence, minimum salary..

Allowances Severe Impacts >10% income loss All severely affected AFs Other income: additional including informal settlers compensation for 3 months of minimum salary.

Relocation/Shifting Transport/transition costs All AFs to be relocated Provision of sufficient allowance to cover transport expenses and livelihood expenses for the transitional period (up to 1 month).

Vulnerable People AFs below poverty line, headed Allowance equivalent to 3 months of Allowances by Women, disabled or elderly minimum salary and employment priority in project-related jobs

F. Institutional Arrangements

6. ADB will be the funding agency of the MFF. The Ministry of Rural Development and Infrastructure (MRDI) will be the Executing Agency (EA) and Roads Department (RD) of MRDI the Implementing Agency (IA) entrusted with the lead responsibility for road construction, as well as the implementation of this LARP. This includes preparation, implementation and financing of all land acquisition and resettlement tasks and cross- agency coordination. RDMRDI will exercise its functions through its existing resettlement division (DRD) which will be responsible for the general management of the planning and implementation of all land acquisition and resettlement tasks. RDMRDI with the assistance of the construction supervision consultants will develop and implement the tasks based on the policy and procedures set out in the LARP. In addition to the RDMRDI, a number of other government departments and private agents will play an instrumental role in the design, construction and operation of the project. Pursuant to the active legislation, the Ministry of Natural Resources and Environmental Protection is responsible for environmental issues. The Ministry of Justice is responsible for legal matters regarding land ownership, and National Agency of Public Registry (NAPR) within the Ministry of Justice is in charge of the registration of land ownership and its transfer through purchase agreement from landowners to the RDMRDI. The local government at rayon and village level will also be involved. To allow an effective execution of all land acquisition and resettlement related tasks some expansion of the capacity on land acquisition and resettlement, currently weak, at DRD/RDMRDI will be needed and capacity building training will be provided. The initial level of the capacity building exercise in the relevant agencies was carried out during the preparation of the LARP. Close consultations were held with all the concerned departments. The designated official from DRD was also an active member in a leading role during the census survey. Informal training was provided by the consultant’s resettlement specialist on the requirements of ADB’s policy and how to develop a balanced

Engconsult Ltd. Appendix 18 111 compensation package which will fulfill the requirements of APs, Government, RDMRDI and ADB. Additionally, capacity building training was also initiated through a series of consultations and informal training sessions at the local administration level.

G. Complaints and Grievance

7. A grievance mechanism will be available to allow an AP appealing any disagreeable decision, practice or activity arising from land or other assets compensation. APs will be fully informed of their rights and of the procedures for addressing complaints whether verbally or in writing during consultation, survey, and time of compensation. First, complaints resolution will be attempted at the village level with the involvement of village authorities and rayon level land acquisition and resettlement teams at the sacrebolu level. Second, if the grievance is not solved at the sacrebolu level, then the AP will lodge the written complaint at the regional level RDMRDI. If, after the regional level RDMRDI intervention, no solution has been reached, a grievance can be directly lodged to DRD/RDMRDI. The AP must lodge the complaint within 2 weeks after receiving response on the original complaint from the regional level RDMRDI and must produce documents supporting his/her claim. Third, the DRD/RDMRDI at central level will provide a response within 2 weeks of registering the complaint. The DRD/RDMRDI decision must be in compliance with this land acquisition and resettlement framework provisions. Fourth, if the grievance redress system fails to satisfy the AP, they can pursue further action by submitting their case to the appropriate court of law (rayon court).

H. Public Consultation, Participation and Disclosure

8. Extensive consultations were conducted at various stages of the social and resettlement project preparation, i.e., alternative alignment feasibility study, cadastral survey of affected lands, land census survey, socio-economic survey, and targeted consultation meetings with affected people and local stakeholder organizations and individuals. Consultations were also held at various office levels such as RDMRDI including DRD, National Agency for Public Registry under the Ministry of Justice, Local Government at rayon level (District Municipality), Gamgebeli, and Property Recognition Committee in the rayon level and Sacrebulo. Community consultations were held through focus group discussion at various locations, especially at 20 villages during April 2009 which involved both men and women participants and totaling 210 participants. The concerned government offices were involved in the resettlement plan preparation. They include Roads RDMRDI, NAPR, Ministry of Justice and the office of Rayons (Ozurgeti, Kobuleti, Batumi and Khelvachauri). The consultation will be continued throughout the project cycle. Information on compensation, entitlement and resettlement management adopted for the Project will be made available to APs. Each AP will be provided information regarding specific entitlements. Information on LAR entitlement was shared at the village level through consulation and through formal public consualtion during the Public Hearing under the Environment Impact Assessement in the month of June 2009.This summary LARP will be translated into local language (Georgian) and disclosed to APs at the local level. The copy of the LARP will be available with RDMRDI. The same will be available for the APs as and when asked for. A copy of the LARP will be disclosed in ADB’s website.

Engconsult Ltd. Appendix 18 112

I. Resettlement Budget, Financing and Implementation Schedule

9. The resettlement cost estimate for the Project (tranche) 1 components includes eligible compensation, resettlement assistance and support cost for RP implementation. All LARP preparation and implementation costs, including cost of compensation and land acquisition and resettlement administration, will be considered an integral part of Project cost and will be contributed as a counterpart fund by the Government of Georgia, in particular RDMRDI. The total estimated cost for the LARP for Project 1 is approximately 9.62 million GEL equivalent to $5.83 million. Contingency provisions (@ 10% of the total cost) have also been made to take into account variations from this estimate. Incase of of any over-run in cost, RDMRDI will provide additional funds as needed in a timely fashion. RDMRDI is responsible for the timely allocation of the funds needed to implement the LARP. As per the land acquisition and resettlement fund flow, the budget for compensation and rehabilitation will be directly disbursed by RDMRDI to the APs. All land acquisition, resettlement, and compensation will be completed before the start of civil works. The LARP duration for LARP implementation schedule is estimated to be one year.

J. Monitoring and Evaluation

10. Land acquisition and resettlement tasks under the project will be subjected to monitoring. Monitoring will be the responsibility of RDMRDI. Internal monitoring will be carried out routinely by DRD/RDMRDI. The results will be communicated to ADB through the quarterly project implementation reports. External monitoring will be carried out twice a year, and its results communicated to DRD/RDMRDI and ADB through semi-annual reports. The RDMRDI (through external help) will carry out a post-implementation evaluation of the LARP about a year after completion of its implementation.

Engconsult Ltd. Appendix 19 113

Appendix 19

ENVIRONMENTAL ASSESSMENT AND REVIEW FRAMEWORK

A. Introduction 1. This environmental assessment and review framework (EARF) has been developed to guide the Roads Department (RD) of the Ministry of Regional Development and Infrastructure (MRDI) in carrying out the environmental assessment of the future projects to be financed under the multitranche financing facility (MFF): ’Subregional Road Corridor Development Program’. The EARF is based on the Georgia’s Laws on Environmental Impact Permit and Ecological Examination (2007) and ADB’s Environment Policy (2002). B. Overview of the Subprojects to be Assessed 2. Identification of projects for MFF are under progress. So far, only one project (Project 1, Adjara Bypass Roads) has been identified and this project will be implemented in two tranches. The activities included in the Project 1 are construction of 42 km of new roads and widening of 6 km of existing roads along the Poti–-Batumi–Sarpi road, bypassing Kobuleti and Batumi (Map 1). Construction of the project road will involve new construction and rehabilitation of bridges, culverts, and other associated drainage structures; tunnel construction; land acquisition and resettlement; site preparation; land clearing; disposal of gravel, soil, vegetation, and unstable material; construction of temporary access roads to construction sites; setting up of temporary construction camps to house workers; excavation and operation of borrow pits; operation of a boulder quarry; extraction of material for embankments using cut and fill procedures; protection of landslide, rockfall, and snowfall areas; measures to protect critical side-slopes; and extraction and cartage of sand for aggregate mixtures, cement works, and asphalt plants. An environmental impact assessment has been already carried out for the Project 1. C. Country’s Environmental Assessment and Review Procedures 3. The relevant national laws on environmental assessment are the ‘Law on Environmental Impact Permit’ (dated December 14, 2007), the ‘Law on Ecological Examination’ (dated December 14, 2007) and the ‘Order No. 515 of the Minister of Environmental Protection and Natural Resources (MOEPNR) on Approval of Rules on Conduction of Ecological Examination’ (dated July 7, 2008). In accordance with these laws, all proposed subprojects require Environmental Impact Permits (EIPs) from the MOEPNR under a procedure including (i) Environmental Impact Assessment (EIA) Report, (ii) Ecological Expertise (EE), and (iii) Public participation. The steps included in securing the EIPs are: • The proponent will conduct environmental assessment study and prepare an EIA report • The proponent will arrange a public hearing of EIA. The procedure involved in public hearing is o Before one week of submitting draft EIA reports to the MOEPNR and MRDI, the proponent issue advertisement in regional and national news papers about the project, and date, time and place of public consultations; and send written invitations to the local government, MOEPNR and MRDI o Arrange hearing within 50 to 60 days after publishing advertisement in the news paper • The proponent will submit an application with EIA report. minutes of meetings of public hearing and other required documents to the MOEPNR and MRDI for Environmental Impact Permit (EIP)

Engconsult Ltd.

Appendix 19 114

• MOEPNR will carry out expert examination of the EIA report, known as State Ecological Expertise (EE) and issue positive EE Conclusion (EEC). Based on the positive EEC, the MOEPNR will issue EIP. 4. A brief summary of the relevant laws applicable for EIPs are described below: • Determines the activities subject to mandatory EE and consequently to issuance of EIP, details the procedure for the issuance of EIPs, and defines the legal basis for public participation in the process of EE conduction, EIA preparation and decision making on EIP. • Activities subject to mandatory EE and consequently to issuance of EIPs inter alia include construction of automobile roads and railroads of international or inter-state importance, bridges and road tunnels thereon, as well as engineering protection structures of automobile roads, railroads and their respective territories. • View a positive EE conclusion (EEC) as mandatory for adoption of a positive decision on the issuance of an EIP. Conduction of EE and issuance of the EEC is the responsibility of the MOEPNR, which undertakes expert examination and the normative-technical and methodological guidance documents and the procedure established under law, through a special committee of experts established for each particular case.

D. Specific Procedures to be Used for Subprojects under MFF 1. Responsibilities and Authorities 5. Responsibilities of the implementing and executing agencies (RD and MRDI) include: (i) Prepare environmental screening checklist and classify projects in consultation with MOEPNR and other departments. (ii) Based on the environmental classification of projects, prepare terms of reference to conduct IEE or EIA studies. (iii) Hire an environmental consultant to prepare IEE or EIA reports including EMP and summary EIA/IEE for public disclosure. (iv) Ensure that an IEE or EIA are prepared in compliance with the requirements of the Government and ADB, and that adequate consultation with affected people is undertaken in accordance with ADB requirements. (v) Undertake review of the IEE or EIA, summary EIA/IEE, and EMP reports to ensure their compliance with the requirements of the Government and ADB. (vi) Obtain necessary permits and/or clearance, as required, from MOEPNR and other relevant government agencies, ensuring that all necessary regulatory clearances are obtained before commencing any civil work on the relevant sections. (vii) Submit to ADB the IEE or EIA, summary EIA/IEE, and EMP reports and other documents, as necessary. (viii) Ensure that any EMP including relevant mitigation measures needing to be incorporated during the construction stage by the contractor are included in the bidding documents. (ix) Ensure that contractors have access to the EIA or IEE and EMP reports of the projects. (x) Ensure that contractors understand their responsibilities to mitigate environmental problems associated with their construction activities. (xi) Ensure and monitor that an EMP including an environmental monitoring plan will be properly implemented.

Engconsult Ltd.

Appendix 19 115

(xii) In case unpredicted environmental impacts occur during the project implementation stage, prepare and implement as necessary an environmental emergency program in consultation with MOEPNR, other relevant government agencies, and ADB. (xiii) In case a project needs to be realigned during implementation, review the environmental classification, revise it accordingly, and identify whether a supplementary IEE or EIA study is required. If yes, prepare the terms of reference for undertaking a supplementary IEE or EIA and hire an environmental consultant to carry out the study. (xiv) Submit semi-annual reports on implementing EMPs, including implementation of an environmental emergency program, if any, to MOEPNR and ADB. (xv) Submit project completion environmental monitoring report to ADB after three years of completion of construction summarizing the overall environmental impacts from the projects. 6. ADB is responsible for the following: (i) Review EIA and SEIA reports prepared under supervision of EA. (ii) Review of IEE and SIEE reports if it requires ADB’s no objection approval (iii) Undertake annual environmental review missions for Category A and B sensitive projects. (iv) Public disclosure of SEIA and SIEE (for Category B sensitive projects) through ADB website 2. Environmental Criteria for Subproject Selection 7. Considering the potential environmental impacts of the future subprojects and the relevant environmental requirements of ADB and the Government of Georgia, the following criteria was agreed upon (by the Government and ADB) for selection of the future subprojects to be included in the Subregional Road Corridor Development Program: (i) The subprojects shall only involve activities that follow all the government regulations (ii) The subprojects should not pass through any wildlife sanctuaries, national parks, nature reserves, and protected areas designated by national and international regulations; (iii) The subprojects should not pass through any ecologically sensitive and significant as recognized by the Government or any area that is internationally significant (such as protected wetland and mangroves); and (iv) The subproject should as much as possible not include any stretch that passes through any cultural heritage and archaeological sites designated by UNESCO and Ministry of Culture, Monument Protection, and Sports.

3. Procedures for Environmental Assessment of Subprojects 8. ADB categorizes Road projects into two categories A and B. Projects with potential for significant adverse environmental impacts are classified into Category A. An EIA is required to address significant impacts. Project judged to have some adverse impacts, but of lesser degree and/or significance than category A is classified into Category B. An initial environmental examination (IEE) is required to determine whether or not significant environmental impacts warranting an EIA are likely. If an EIA is not needed, the IEE is regarded as the final environmental assessment report. The guidelines to prepare environmental assessment reports for a Category A project (EIA and SEIA) and for a Category B project (IEE and SIEE) in

Engconsult Ltd.

Appendix 19 116 compliance with the ADB’s Environment Policy (2002) and Environmental Assessment Guidelines (2003) are given in the following sections. a. Screening 9. All future subprojects to be included in MFF will be screened to determine its environmental category based on the ADB’s Rapid Environmental Assessment Checklist (REA). A template of the REA is given in Annex 1. Categorization is to be based on the most environmental sensitive component, which means that if one part of the project is with potential for significant adverse environmental impacts, then the project is to be classified as Category A regardless of potential environmental impacts of other aspects of the project. In general, a project will be classified as ‘Category A’ if the project: (i) is a new road alignment (ii) requires a complex mitigation measure needing to be prepared through an in-depth assessment of the impacts and detailed study for preparing mitigation measures; (iii) will generate impact on an ecologically sensitive area, particularly if the project is located less than 500 meters from any designated wildlife sanctuary, national park, other sanctuary, botanical garden, or area of international significance (e.g., an IUCN or RAMSAR site) or cultural heritage and archaeological sites designated by UNESCO. and Ministry of Culture, Monument Protection, and Sports; or (iv) passing through any ecologically sensitive areas (hilly or mountainous, forested, nearby estuarine or other area with important ecological function). 10. Road upgrading and rehabilitation subprojects that do not fall into the above category are classified as B. b. Scoping 11. Before conducting any environmental studies, a scoping document consists of the scope of the environmental surveys, methods of data collection and outputs anticipated from the study are to be prepared. In case of Category A projects, the scoping document is to be approved by the EA before taking up of detailed environmental studies. Scoping should focus on identifying those components of the environment likely significantly affected by the project based on project location, past documented experience, the potential likely geographic and time-related extent of the effects, and the measurements or thresholds to be used to assess significance. A map of the study area with a clear topographic map showing the project road(s) in relation to topography, water courses, settlement areas and preferably landuse should be enclosed with the scoping document. A sample scoping document prepared for Project 1 of the MFF is enclosed in Annex 2. c. Identifying Baseline Conditions and Impacts 12. With the screening and scoping results in hand, planning of the field program becomes relatively easy, however does need the involvement of an experienced environmental assessment practitioner. 13. The first step is to establish the baseline conditions for the components of the environment likely affected by the project. This will be usually carried through site visits and review of spatial database for all available environmental parameters such as terrain, soils, geology, rivers, forest, protected areas and landuse. This also will include collection and analysis of background noise, air and water quality. This baseline will become the measure against which any change is measured. These data must be collected in such a manner that their source can be traced by anyone who picks up the document.

Engconsult Ltd.

Appendix 19 117

14. The second step is to predict likely change as a result of major construction activities and operation of the road, by relating cause and effect such as changes in traffic volume, fleet makeup and traffic patterns to air quality and noise effects. This will be carried out by thorough review of baseline environment and proposed civil works. The locations where based data were collected, where ongoing monitoring takes place and the timing of these activities should be remain uniform or at least easily traceable, permitting analysis of technical credibility. While following strict scientific method in EIA is far too costly and time consuming, every effort should be made make the entire study transparent and traceable d. Public Consultations/Hearing 15. The third step is to present the findings on impacts and benefits during a consultation and information session to inform key stakeholders and affected community of the issues identified and to invite comments and corrections. For a full EIA consultation is required at least twice during the EIA: • first as part of the scoping stage to define the project and to get feedback in options; and, • secondly after the draft EMP has been prepared. 16. Public consultations include news paper advertisement in the regional and national news papers before 50 to 60 days of the consultations giving brief project description, location and specific contact data (including telephone numbers). Often a project website is created and link information is provided. Further, the proponent, working with the consultant should prepare a list of important participants and send emails or letters of invitation providing details including dates for both consultations. 17. Consultation sessions must have minutes and attendance sheets prepared and included as part of the EIA documentation. 18. For Category B projects, nearly all conditions as defined above are the same except there is only 1 consultation session is needed. Often, the consultation session takes place as the EMP is being prepared. e. Preparation of the Environmental Management Plan (EMP) 19. The Fourth Step is the preparation of the EMP, the most important output of an environmental assessment. The EMP must be practical, specific and systematic, such that it can be easily converted to mitigative and monitoring actions, proponents and contractors can undertakes: activities that monitors can track and activities that can be translated or simply referenced in contract specification as environmental clauses. Therefore each mitigative measure needs to be matched with a monitoring activity. 20. Good EMPs not only identify the source of the impact, the effect in the biophysical environment and the monitoring action to be taken, but also where, how often, when and who should implement each mitigative and monitoring action and who is responsible. This is the same for both EIAs and IEEs. 21. Project Management Unit of RD is required to review and update the EMP as soon as the contractor has been appointed and the mobilization date is established. f. Assessing Institutional Capacity for EMP Implementation 22. The Fifth Step involves the identification of the agencies and units at the national and provincial level that will likely be involved in the implementation and supervision of the mitigation and monitoring actions as well as the general management of the EMP from preconstruction through the operating period. The EMP is also useful in that it identifies the lead

Engconsult Ltd.

Appendix 19 118 implementing and supervising agencies (RD, MRDI and MOEPNR and their provincial/regional departments) involved in all mitigation and monitoring actions. The analysis, using mostly the interview approach, should be short and focused, identifying needs based on obvious gaps, such as lack of experience in any international-level assessments or lack of experience with preparation and implementation of EMPs. Careful interviews will almost always result in those needing assistance identifying what they need. Finally the needs are assessed in terms of longer term capacity building and short term training and workshops in relation to realistic budgetary limits and a capacity building and training is proposed. Since the contractors play such an important role in EMP implementation, they must not be left out of the analysis and a general approach to strengthening their safeguards skills must be included in the analysis. g. Estimating Mitigation, Monitoring and Training Costs 23. The Sixth Step involves costing of each of the mitigative and monitoring actions as well as the institutional capacity building. Costing details must be systematic and include rates and unit costs and an indication of actions that, while referred to as environmental, are normally found in other budget items, for example slope stabilization, revegetation, fuel handling and storage protocols and work camp waste management. h. Reporting 24. The Seventh Step is preparation of the assessment document according to a prescribed format and level of details. The templates of the reports for preparation of EIA/IEE and SEIA/SIEE reports are given in Annexes 3 to 6 and also can be found from ADB website at www.adb.org/documents/guidelines/environmental_assessment/default.asp. E. Compliance with ADB’s Environmental Policy—Due Diligence 25. RD, MRDI and MOEPNR have the responsibility to undertake environmental due diligence and monitor implementation of environmental mitigation measures for all projects under each respective responsibility. The due diligence report as well as monitoring of EMP implementation (described in the annual report) need to be documented systematically. ADB must be given access as needed to undertake environmental due diligence for all projects. 26. An EMP will be part of the overall project monitoring and supervision, and will be implemented by the contractor with oversight from the supervision consultant and PMU. Progress on the preparation and implementation of an EMP will be included in the periodic project progress reports. Specific monitoring activities defined in the IEEs or EIAs and EMPs will be carried out by the contractors and monitored by the PMU. RD will submit reports on EMP implementation to ADB for every six months to Category A projects and annually for Category B projects. 27. The PMU, with assistance of the international or national environmental consultants, will review the IEE or EIA and corresponding EMP for each project to ensure that mitigation measures and monitoring plans proposed in that document are in compliance with ADB’s and national requirements. According to the reports and reviews during its missions, ADB, in consultation with the Government, will confirm compliance. For this purpose, the PMU will provide ADB with access to information on any projects. The information on implementation of an EMP, as well as that on environmental and social safeguard compliance, will be systematically documented and reported to ADB as part of the regular progress reports F. Public Disclosure 28. RD is responsible for ensuring that all environmental assessment documents, including the environmental due diligence and monitoring reports, are properly and systematically kept as part of the project record of each project under its responsibility. All

Engconsult Ltd.

Appendix 19 119 environmental documents are subject to public disclosure. These documents should be made available to the public, if requested. In case there are category A and B-sensitive projects, the summary EIA/IEE will be disclosed to the public through ADB’s website 120 days before the project is approved. The SEIA/SIEE shall be reviewed by ADB before it is disclosed to the public. RD will also ensure that public consultations, particularly with project affected persons, are undertaken adequately during the IEE or EIA preparation in consistent with ADB requirements. G. Staffing Requirements and Budget 29. RD and MRDI will recruit environmental consultants as a part of engineering design consultants to prepare environmental assessment reports for each subproject consistent with this EARF. Terms of reference for consultants along with the budget are given in Annex 7. The estimated cost for preparation of each IEE and EIA are US$98,750 and US$156,750, respectively. Adequate funding resources will be provided under each subproject for environmental monitoring and mitigation measures for each subproject.

Engconsult Ltd.

Appendix 19 Annex.1 120

Annex 1 Rapid Environmental Assessment (REA) Checklist ROADS AND HIGHWAYS

Instructions:

‰ This checklist is to be prepared to support the environmental classification of a project. It is to be attached to the environmental categorization form that is to be prepared and submitted to the Chief Compliance Officer of the Regional and Sustainable Development Department. ‰ This checklist is to be completed with the assistance of an Environment Specialist in a Regional Department. ‰ This checklist focuses on environmental issues and concerns. To ensure that social dimensions are adequately considered, refer also to ADB checklists and handbooks on (i) involuntary resettlement, (ii) indigenous peoples planning, (iii) poverty reduction, (iv) participation, and (v) gender and development. ‰ Answer the questions assuming the “without mitigation” case. The purpose is to identify potential impacts. Use the “remarks” section to discuss any anticipated mitigation measures.

Country/Project Title:

Sector Division:

SCREENING QUESTIONS Yes No REMARKS A. Project Siting

Is the Project area adjacent to or within any of the following environmentally sensitive areas? ƒ Cultural heritage site ƒ Protected Area ƒ Wetland ƒ Mangrove ƒ Estuarine ƒ Buffer zone of protected area ƒ Special area for protecting biodiversity B. Potential Environmental Impacts

Will the Project cause… ƒ encroachment on historical/cultural areas; disfiguration of landscape by road embankments, cuts, fills, and quarries? ƒ encroachment on precious ecology (e.g. sensitive or protected areas)? ƒ alteration of surface water hydrology of waterways crossed by roads, resulting in increased sediment in streams affected by increased soil erosion at construction site? ƒ deterioration of surface water quality due to silt runoff and sanitary wastes from worker-based camps and chemicals used in construction? ƒ increased local air pollution due to rock crushing, cutting and filling works, and chemicals from asphalt processing? ƒ noise and vibration due to blasting and other civil works? ƒ dislocation or involuntary resettlement of people ƒ other social concerns relating to inconveniences in

Engconsult Ltd.

Appendix 19 Annex.1 121

SCREENING QUESTIONS Yes No REMARKS living conditions in the project areas that may trigger cases of upper respiratory problems and stress? ƒ hazardous driving conditions where construction interferes with pre-existing roads? ƒ poor sanitation and solid waste disposal in construction camps and work sites, and possible transmission of communicable diseases from workers to local populations? ƒ creation of temporary breeding habitats for mosquito vectors of disease? ƒ dislocation and compulsory resettlement of people living in right-of-way? ƒ accident risks associated with increased vehicular traffic, leading to accidental spills of toxic materials and loss of life? ƒ increased noise and air pollution resulting from traffic volume? ƒ increased risk of water pollution from oil, grease and fuel spills, and other materials from vehicles using the road?

Engconsult Ltd.

Appendix 19 Annex 2 122 Annex 2 SAMPLE SCOPING DOCUMENT

Scoping Framework for the Environmental Impact Assessment of ‘Ajara Bypass Roads Development Project’ A. Introduction 30. This scoping framework has been prepared to carryout detailed environmental impact assessment (EIA) for the ‘Ajara Bypass Roads Development Project’ in accordance with ADB’s Environment Policy (2002) and Environmental Assessment Guidelines (2003) and relevant laws and regulations in Georgia. The project involves new construction and widening of roads, bridges, overpasses, tunnels, and protection structures. The study will identify potential environmental impacts on physical, ecological, social, cultural, and economic resources of project areas during design (for different alignments), construction and operation. An EIA report will be prepared along with environmental management and monitoring plan to address all identified environmental impacts as per the work plan provided in Figure A19. 1. The study will be carried out by Consultant and the draft EIA report will be submitted along with the Final Report of the Project. B. Scope of Work 1. Baseline Studies Output 1: Memo on Georgian Legal and Administrative Procedures 31. Activity 1.1: Collection and review of relevant information regarding environmental legislation, statutory orders, by-laws, etc. connected to preparation and approval of the EIA report by Georgian Authority, and draft the memo. The memo will also consider the requirements of ADB Guidelines for ‘Category A’ Project. 32. Activity 1.2: Conduction of a series of meetings with the Ministry of Environment Protection and Natural Resources, National Environmental Agency, and the Ministry of Culture (Center of Archeological Studies, Department of Monument Protection). During these meetings appropriate legal and administrative procedures has been discussed. Discussions also includes issues such as basis for further approval/ disapproval of EIA by Georgian authorities and on the issuance of “Construction Permit” along with “Environmental Impact Permit’ (published in 2007 and entered in force in 2008) and ‘State Ecological Examination’ in accordance with Georgia’s Law on Licenses and Permits (2005). Review of other relevant environmental laws, regulations, Norms, and Standards on Air, Noise, Water, Waste, and Wildlife. 33. Activity 1.3: Conduction of discussion meeting with the Road Department of the Ministry of Regional Development and Infrastructure on the issuance of “construction permit” and “environmental impact permit” in accordance with Georgia’s Law on Licenses and Permits (2005). Output 2: Preparation of Baseline Assessment 34. Activity 2.1: Review of reports and field data collected from the Project’s pre-feasibility study; and other road projects carried out under the World Bank, JBIC, EBRD, and MCC’s funding. 35. Activity 2.2: Collection of baseline information on existing environmental condition along the project road alignments and identification of the environmental components that need detailed further study. Baseline assessment will be done based on the available secondary information, field visits, sampling and environmental monitoring including but not limited to the following: i. Physical resources: • Topography, climate, soils, geology, landuse, coastal resources, and surface and groundwater resources. ii. Natural hazards: • Seismicity, floods, landslides, and volcanic activity iii. Ecological Resources:

Engconsult Ltd.

Appendix 19 Annex 2 123

• Landscape and natural ecosystem, flora and fauna, wildlife and wetland habitats, nature reserve and Protected areas iv. Environmental quality: • Air quality (PM, CO, NOx, SOx etc.), noise quality and water quality (DO, TPH, pH, total nitrogen, BOD, total phosphorus, and suspended solid etc.). v. Cultural resources and Archaeological sites: • Structures or sites that are of historical, archaeological, paleontological, or architectural significance 2. Detailed Field Investigation to Screen Environmental Impacts Output 3: Field Investigation and Analysis of Results 36. Activity 3.1: Preparation of layout plan of the project road consisting of, but not limited to, the flowing information: • Road alignments • Sampling location for environmental parameters (air, noise, water) • Construction camp including storage of petroleum products and explosives • Asphalt and batch mixing plants, Construction sites and camps, Quarry sites and borrow pits • Water sources, Waste disposal sites, Environmentally sensitive areas 37. Activity 3.2: Collection of a cadastral land use map showing the project locations and descriptions of the surrounding activities. This is to ensure that the project road is compatible with the national regulation specified for construction sites. 38. Activity 3.3: Initiation of necessary investigations and fieldwork for gathering of following additional information about ecological and environmental parameters in the project area. i. Landscape, Geohazards and Slope Stability Identification of natural landscape along the project road. Assessment of geological and geomorphologic features of the project area, as well as any violent interference in the natural processes. Investigation and evaluation of results to predict rock falls, landslide, mudflow and debris flow, erosion, ground subsidence, floods, and banks washing-off (lateral erosion) and seashore erosion. ii. Soil erosion and slope stability Analysis of soil characteristics, moisture contents, vegetation cover etc., in conjunction with the above activity, to predict possible soil erosion and landslides due to project activities. iii. Terrestrial flora and fauna Investigation of the composition of plant species and migratory birds in the Project area. Attention should be paid to the distribution of protected plant and animal/birds species in order to ensure favorable conservation status for these species. iv. Wetland habitats and aquatic flora and fauna Investigations of occurrence of species (flora and fauna) in the identified wetlands along the project area and assess the potential influence of the proposed road alignment. Attention should be paid to the distribution of protected species in order to ensure favorable conservation status for these species. v. Protected Areas and sensitive environmental receptors Collection of protected and sensitive area maps and exact coordinates showing the boundaries and buffer zones of protected areas and project alignments and descriptions of the habitats. This is to ensure that the locations of sensitive areas and project alignment are sufficiently distant enough to maintain harmonization and avoid any potential disturbances on the habitats. Investigation will also be done for other sensitive sites along the project alignment, viz. wetlands, seashore, tourism, etc. vi. Traffic flow Traffic counts and historical traffic flow to predict the future traffic growth and the load on the project road. vii. Air quality and noise level Collection and analysis of air (PM, CO, NOx, SO2) and noise quality along the project road. viii. Water Quality Collection and analysis of water quality (DO, turbidity, pH, TOC, and dissolved solid) of the major rivers along the project road. Investigation of water quality impacts during construction and operations stages in selected river sites. ix. Monuments and Archaeological Site

Engconsult Ltd.

Appendix 19 Annex 2 124

Investigation of the impact of the project on monuments and archaeological sites along the project corridor. x. Quarry and Borrow Sites Estimation of effects on the ecological resources in the area connected to quarry and barrow pit operations needed for construction. 3. Analysis of Alternatives and Economic Assessment Output 4: Analysis of Alternative Options 39. Activity 4.1: Comparative environmental analysis of all available project alignments, including “No Project” scenario. Output 5: Economic Assessment 40. Activity 5.1: Economic analysis of all alternatives in accordance with ADB’s Handbook on Economic Evaluation of Environmental Impacts for: (i) costs and benefits of environmental impacts; (ii) costs, benefits, and cost-effectiveness of mitigation measures; and (iii) discussion of impacts that have not been expressed in monetary values, in quantitative terms where possible. 4. Forecast Future Impacts and Mitigation Measures Output 6: Forecast of Impacts 41. Activity 6.1: Forecasting of air and noise quality based on predicted traffic estimates using computer modeling software and recommendation of mitigation measures. 42. Activity 6.2: Evaluation of the project impact on all physical and ecological resources described in Activity 3.3 and recommendation of mitigation measures. 43. Activity 6.3: Evaluation of socio-economical and cultural impacts, such as: • Assessment of the status of livelihoods (agriculture, business, etc.) in the context of socio- economical impact. • Assessment of the impact on objects or areas with known archeological values in the project area. • Assessment of impacts on culturally and religiously sensitive locations (church, cemetery, etc.) • Assessment of impacts in tourism sector in Kobuleti and Batumi resort areas • Assessment of traffic safety. 44. Activity 6.4: Assessment of impact on human health and estimation of possible health impacts on construction workers and roadside residents (such as safety, HIV/AIDS, STDs, human trafficking) due to construction camps and other project activities. Output 7: Environmental Management and Monitoring Plan 45. Activity 7.1: Preparation of Environmental Management and Monitoring Plan (EMMP) for all phases of the Project for effective implementation of environmental protection and mitigation measures and monitoring of significant environmental impacts. Preparation of environmental protection measures to (i) mitigate environmental impacts, (ii) provide in-kind compensation for lost environmental resources, or (iii) enhance environmental resources. Prepare cost estimates for each mitigation measure proposed in the EMP and include all the mitigation measures in the engineering design of the Project. 46. Activity 7.2: Setting up of environmental criteria for several variables such as: • Air quality, Noise level, Water quality, Accidental spills of hazardous substances • Naturally protected areas

5. Institutional Assessment and Monitoring Mechanism Output 8: Institutional Assessment 47. Activity 8.1: Assessment of institutional capacity of the implementing agencies for effective implementation of environmental management and monitoring plan. Identification of responsible institutes for implementation and supervision of the EMMP. Assess training needs of these agencies and propose

Engconsult Ltd.

Appendix 19 Annex 2 125 capacity building measures and institutional arrangements to strengthen these agencies along with the cost estimates. Output 9: Monitoring Mechanism 48. Activity 9.1: Elaboration and specifying of “feed back monitoring” program, a tool to be used by implementing authorities in order to be able to interfere and respond quickly to activities, which during the construction and operation turn out to have a negative effect to the environment. The tool will specify the parameters, location, frequency and means of monitoring. 6. Public Consultations and Disclosure Plan Output 10: Conduct Public Consultations and Document them 49. Activity 10.1: Assisting Department of Road to conduct two public consultations (one during the inception stage and the second one after finalization of EIA report) according to ADB’s Public Communications Policy (2005) for Category A projects and the 6th Clause of the Law of Georgia on the Environmental Impact Permit. This will ensure that the consultation process will involve affected people, key agencies, NGOs, and other stakeholders and they are provided with opportunities to participate in the decision-making process and to influence decisions that will affect them. Address all the comments in the engineering designs. 50. Activity 10.2: Agreement by the appropriate authorities of specific zones where minor temporary and/or permanent impacts to the environment can be accepted during the construction and operation phase of the bypass road. The ranges of the zones will be determined on the basis of results of investigations and local conditions. Output 11: Disclosure Plan 51. Activity 11.1: In consultation with the stakeholders, preparation of information disclosure plan for dissemination of safeguard documents to the affected community and general public.

Figure A19. 1: Work Plan of Proposed EIA Study

Engconsult Ltd.

Appendix 19 Annex 3 126 Annex 3 Outline of an Environmental Impact Assessment (EIA) Report

1. Introduction 2. Description of the Project 3. Description of the Environment 4. Alternatives 5. Anticipated Environmental Impacts and Mitigation Measures 6. Economic Assessment 7. Environmental Management Plan 8. Public Involvement and Disclosure 9. Conclusions

1. Introduction 52. This section usually includes the following: (i) purpose of the report, including (a) identification of the project and its proponent, (b) brief description of the nature, size, and location of the project and its importance to the country, and (c) any other pertinent background information; (ii) stage of project preparation (i.e., pre-feasibility study, feasibility study, detailed engineering design preparation); (iii) extent of the EIA study, including the scope of the study, magnitude of effort, and persons/expertise or agency performing the study and corresponding person months; and (iv) brief outline of the contents of the report, including any special techniques or methods used for identifying issues, assessing impacts, and designing environmental protection measures.

2. Description of the Project 53. The project should be described in terms of its basic activities, location, layout, and schedule (in terms of the project cycle). This section of the EIA report should provide sufficient details on the following: (i) Type of project (ii) Need for project (iii) Location (use maps showing general location, specific location, project boundary and project site layout) (iv) Size or magnitude of operation including any associated activities required by or for the project (v) Proposed schedule for approval and implementation (vi) Description of the project including drawings showing project layout, components of project, etc. This information should be of the same type and extent as is included in feasibility reports for proposed projects, in order to give a clear picture of the project and its operations.

3. Description of the Environment 54. This section contains a description of the study area to provide a clear picture of the existing environmental resources and values within which the impacts must be considered. Detailed methodology to gather information, including data sources, should also be briefly described. As much as possible, the baseline information should be presented in maps, figures, and tables. The baseline environmental information area should include: (i) Physical Resources: e.g.

Engconsult Ltd.

Appendix 19 Annex 3 127

• atmosphere (e.g. air quality and climate) • topography and soils, • surface water • groundwater • geology/seismology. (ii) Ecological Resources: (e.g.) • fisheries • aquatic biology • wildlife • forests • rare or endangered species • protected areas • coastal resources (iii) Economic Development: (e.g.) • industries • infrastructure facilities (e.g. water supply, sewerage, flood control) • transportation (roads, harbors, airports, and navigation) • land use (e.g. dedicated area uses) • power sources and transmission • agricultural development, mineral development, and tourism facilities (iv) Social and Cultural Resources: (e.g.) • population and communities (e.g. numbers, locations, composition, employment) • health facilities • education facilities • socio-economic conditions (e.g. community structure, family structure, social well being) • physical or cultural heritage • current use of lands and resources for traditional purposes by Indigenous • Peoples • structures or sites that are of historical, archaeological, paleontological, or architectural significance. 4. Alternatives 55. The consideration of alternatives is one of the more proactive sides of environmental assessment - enhancing the project design through examining options instead of only focussing on the more defensive task of reducing adverse impacts of a single design. This calls for the systematic comparison of feasible alternatives for the proposed project site, technology, and operational alternatives. Alternatives should be compared in terms of their potential environmental impacts, capital and recurrent costs, suitability under local conditions, and institutional, training and monitoring requirements. For each alternative, the environmental costs and benefits should be quantified to the extent possible, economic values should be attached where feasible, and the basis for the selected alternative should be stated. 56. Examining alternative means of carrying out a project involves answering the following three questions: (i) what are the alternatives? , (ii) what are the environmental impacts associated with each alternative? , and (iii) what is the rationale for selecting the preferred alternative? For example, a road connecting two points can follow a number of different routes. In this case, the EIA report must describe the process taken to select the most appropriate route based on a set of pre-determined criteria. The consideration of alternatives and the selection criteria used to identify the preferred alternatives must include environmental factors. The information going into the decision and the decision-making process must be documented in the EIA report. For example, if "alternative means" refers to site selection for a

Engconsult Ltd.

Appendix 19 Annex 3 128 large hydroelectric dam, the location of each alternative would have to be described, the environmental impacts of each alternative defined and the criteria and analysis of site selection presented. 57. Since the selection of alternatives can involve detailed technical analysis that includes more than just environmental factors, it may be preferable to present the details of this analysis as an appendix and include only the results and summary of this selection process in the body of the report. For example, a table listing the alternatives on one axis, and the criteria, such as reliability, cost, performance, inherent environmental effects and necessary mitigation measures, on the other axis may provide an effective summary. 58. Alternatives to the project. In some instances it will be necessary to consider “alternatives to” the project. This situation should not arise if the project is consistent with DMC’s development strategy, ADB’s Country and Strategy Program, and has been developed based on a sector strategy and roadmap. The EIA report should describe how the project fits into this larger strategic planning context. This context helps justify the project and demonstrates the requirements that may constrain the alternatives that are feasible or permitted. 59. However, in the case of potentially controversial projects, there may public concern that the project does not represent the best way to achieve stated development objectives. In addition, segments of the public may react negatively if they perceive that the EIA report has not considered alternatives to the project, or the preferred alternative is proceeding based on flawed assumptions. Therefore, if controversy is expected surrounding the fundamental reasons for the project, the EIA report should include a discussion of alternatives to the project. 60. One alternative that should receive special attention is the "no go" alternative. In some cases, this may be the only alternative to the project that can be realistically considered. 5. Anticipated Environmental Impacts and Mitigation Measures 61. Review Characteristics of Each Environmental Impact. This section will evaluate the project’s expected impacts (in as quantified terms as possible) on each resource or value, and applicable sectoral environmental guidelines wherever any significant impact is expected (including environmental risk assessment, where appropriate. Environmental impacts to be investigated will include those due to (i) project location; (ii) caused by possible accidents; (iii) related to design; and (iv) during construction, regular operations, and final decommissioning or rehabilitation of a completed project. Where adverse effects are indicated, discuss measures for minimizing and/or offsetting these, and opportunities for enhancing natural environmental values will be explored. Both direct and indirect effects will be considered, and the region of influence indicated. This analysis is the key presentation in the report and if not sufficiently completed it may be necessary to delay the project until the analysis can be completed. It is necessary to present a reasonably complete picture of both the human use and quality of life gains to result from the project due to the utilization, alteration, and impairment of the natural resources affected by the project, so that fair evaluation of the net worth of the project could be made. 62. Mitigating Adverse Effects. For each significant adverse environmental impact, the report will carefully explain how the project plan/design minimizes the adverse effects and in addition how the project plan/design, to the extent feasible, includes provision for offsetting or compensating of adverse effects and for positive enhancement of benefits or environmental quality. Where substantial cost of mitigation measures is involved, alternative measures and costs will be explored. 63. Irreversible and Irretrievable Impacts. The EIA report will identify the extent to which the proposed project would irreversibly curtail the potential uses of the environment. For example, highways that cut through stream corridors, wetlands, or a natural estuary can result in irretrievable damage to those sensitive ecosystems. Other impacts that may be irreversible include alteration of historic sites, and expenditure of construction materials and fuels. Also, projects through estuaries, marshes, etc., may permanently impair the area’s natural ecology; or elimination of recreation areas and parklands can precipitate drastic changes in the project area’s social and economic character.

Engconsult Ltd.

Appendix 19 Annex 3 129

64. Temporary Effects During Project Construction. In the event the construction phase of the project involves special environmental impacts (to be terminated on completion of construction), these will be separately discussed including proposed remedial measures. 6. Economic Assessment 65. This section may be drawn from the economic analysis conducted as part of the project feasibility study. It should include the following elements which should be integrated into the overall economic analysis of the project: (i) costs and benefits of environmental impacts; (ii) costs, benefits, and cost-effectiveness of mitigation measures; and (iii) discussion of impacts that have not been expressed in monetary values, in quantitative terms where possible (e.g. weight of volume estimates of pollutants). 7. Environmental Management Plan 66. The EMP describes how the mitigation and other measures to enhance the benefits of environmental protection will be implemented. It explains how the measures will be managed, who will implement them, and when and where they will be implemented. The following elements should be described in the EMP (see Table A2.1 for contents of EMP): (i) implementation of mitigation measures during project design; (ii) implementation of mitigation measures by contractors, and how impacts prevention will be incorporated in the materials procurement; (iii) social development program (e.g., resettlement plan, community training); (iv) contingency response plan for natural or other disasters, and project contingencies; and (v) environmental management and monitoring costs including mitigation costs. 67. The environmental monitoring plan describes the monitoring activities to ensure that adverse environmental impacts will be minimized, and the EMP implemented. The environmental monitoring plan will cover selected parameters to indicate the level of environmental impacts. It also describes how, when, and where the monitoring activities will be undertaken; who will carry them out; and who should receive the monitoring report. More importantly, it includes a proposal to carry out environmental compliance monitoring activities. 68. The present capacity of the executing agency to implement EMP should be described and implementation costs clearly identified. 8. Public Consultation and Information Disclosure 69. This section will (i) describe the process undertaken to involve the public in project design and recommended measures for continuing public participation; (ii) summarize major comments received from beneficiaries, local officials, community leaders, NGOs, and others, and describe how these comments were addressed; (iii) list milestones in public involvement (e.g., dates, attendance, topics of public meetings), and recipients of the report and other project-related documents; (iv) describe compliance with relevant regulatory requirements for public participation; (v) if possible summarize public acceptance or opinion on the proposed project; and (vi) describe other related materials or activities (e.g., press releases, notifications) as part of the effort to gain public participation. This section will provide of summary of information disclosed to date and procedures for future disclosure. 9. Conclusions 70. The EIA report will present the conclusions of the study including: (i) gains which justify project implementation; (ii) explanation of how adverse effects could be minimized or offset, and compensated to make these impacts acceptable; (iii) explanation of use of any irreplaceable resources; and (iv) provisions for follow-up surveillance and monitoring. Simple visual presentations of the type and magnitude of the impacts may aid the decision-maker.

Engconsult Ltd.

Appendix 19 Annex 4 130

Annex 4 Outline of Summary Environmental Impact Assessment (SEIA) Report

A. Introduction (1/2 page) 71. This section will include the purpose of the report, extent of the EIA study and brief description of any special techniques or methods used. B. Description of the Project (1/2 page) 72. This section will include the type of and need for project, location, size or magnitude of operation and proposed schedule for implementation. C. Description of the Environment (2-3 pages) 73. This section will include the physical and ecological resources, human and economic development and quality of life values in the area affected by the project. Where available, environmental standards will be used as the baseline for comparative purposes. D. Alternatives (1-2 pages) 74. For each alternative, a summary of the probable adverse impacts and its relation to the project, and other alternatives will be discussed determine whether the project minimizes the environmental impact over all other alternatives and is within acceptable environmental impact limits. In most cases, environmental impacts "with" and "without" project alternatives should be examined. E. Anticipated Environmental Impacts and Mitigation Measures (4-6 pages) 75. Environmental impacts, both direct and indirect, on different environmental resources or values due to project location, as related to design, during construction and regular operation will be discussed and mitigation, offsetting or enhancement measures will be recommended. F. Economic Assessment (1-2 pages) 76. This section will include: (a) costs and benefits of environmental impacts; (b) costs, benefits and cost effectiveness of mitigation measures; and (c) for environmental impacts that have not been expressed in monetary values, a discussion of such impacts, if possible, in quantitative terms (e.g. weight or volume estimates of pollutants). This information should be integrated into the overall economic analysis of the project. G. Environmental Management Plan (1-2 pages) 77. The EMP will describe the impacts to be mitigated, and activities to implement the mitigation measures, including how, when, and where they will be implemented. The environmental monitoring plan will describe the impacts to be monitored, and when and where monitoring activities will be carried out, and who will carry them out. H. Public Consultation and Disclosure (1-3 pages) 78. This section will describe the process undertaken to involve the public in project design and recommended measures for continuing public participation; summarize major comments received from beneficiaries, local officials, community leaders, NGOs, and others, and describe how these comments were addressed; list milestones in public involvement such as dates, attendance, and topics of public meetings; list recipients of this document and other project related documents; describe compliance with relevant regulatory requirements for public participation; and summarize other related materials or activities, such as press releases and notifications. This section will provide of summary of information disclosed to date and procedures for future disclosure. I. Conclusions (1 page)

Engconsult Ltd.

Appendix 19 Annex 4 131

79. This section will describe the gains which justify implementation of the project; explain how significant adverse environmental impacts will be mitigated or offset and compensated for; explain/justify use of any irreplaceable resources and; describe follow-up surveillance and monitoring.

Engconsult Ltd.

Appendix 19 Annex 5 132

Annex 5 Outline of an Initial Environmental Examination (IEE) Report

1. Introduction 2. Description of the Project 3. Description of the Environment 4. Screening of Potential Environmental Impacts and Mitigation Measures 5. Institutional Requirements and Environmental Monitoring Plan 6. Public Consultation and Information Disclosure 7. Findings and Recommendation 8. Conclusions A. Introduction 80. This section usually includes the following: (i) purpose of the report, including (a) identification of the project and project proponent; (b) brief description of the nature, size, and location of the project and of its importance to the country; and (c) any other pertinent background information; and (ii) Extent of the IEE study: scope of study, magnitude of effort, person or agency performing the study, and acknowledgement. B. Description of the Project 81. Furnish sufficient details to give a brief but clear picture of the following (include only applicable items): (i) type of project; (ii) category of Project; (iii) need for project; (iv) location (use maps showing general location, specific location, and project site); (v) size or magnitude of operation; (vi) proposed schedule for implementation; and (vii) descriptions of the project, including drawings showing project layout, and project components. This information should be of the same type and extent as is included in feasibility reports for proposed projects to give a clear picture of the project and its operations. C. Description of the Environment 82. Furnish sufficient information to give a brief but clear picture of the existing environmental resources in the area affected by the project, including the following (to the extent applicable): (i) Physical Resources: (e.g.) • atmosphere (e.g. air quality and climate) • topography and soils, • surface water • groundwater • geology/seismology. (ii) Ecological Resources: (e.g.) • fisheries • aquatic biology • wildlife • forests • rare or endangered species • protected areas

Engconsult Ltd.

Appendix 19 Annex 5 133

• coastal resources (iii) Economic Development: (e.g.) • industries • infrastructure facilities (e.g. water supply, sewerage, flood control) • transportation (roads, harbors, airports, and navigation) • land use (e.g. dedicated area uses) • power sources and transmission • agricultural development, mineral development, and tourism facilities (iv) Social and Cultural Resources: (e.g.) • population and communities (e.g. numbers, locations, composition, employment) • health facilities • education facilities • socio-economic conditions (e.g. community structure, family structure, social well being) • physical or cultural heritage • current use of lands and resources for traditional purposes by Indigenous Peoples • structures or sites that are of historical, archaeological, paleontological, or architectural significance. D. Screening of Potential Environmental Impacts and Mitigation Measures 83. Using the checklist of environmental parameters for different sector projects, this section will screen out “no significant impacts” from those with significant adverse impact by reviewing each relevant parameter according to the following factors or operational stages. Mitigation measures, where appropriate, will also be recommended environmental problems due to project location, and related to project design, construction, and operations. Potential environmental enhancement measures and additional considerations will also be covered. E. Institutional Requirements and Environmental Monitoring Plan 84. This section should state the impacts to be mitigated, and activities to implement the mitigation measures, including how, when, and where they will be implemented. Institutional arrangements for implementation should be described. The environmental monitoring plan will describe the impacts to be monitored, and when and where monitoring activities will be carried out, and who will carry them out. The environmental management and monitoring costs should also be described. F. Public Consultation and Information Disclosure 85. This section will describe the process undertaken to involve the public in project design and recommended measures for continuing public participation; summarize major comments received from beneficiaries, local officials, community leaders, NGOs, and others, and describe how these comments were addressed; list milestones in public involvement such as dates, attendance, and topics of public meetings; list recipients of this document and other project related documents; describe compliance with relevant regulatory requirements for public participation; and summarize other related materials or activities, such as press releases and notifications. This section will provide of summary of information disclosed to date and procedures for future disclosure. G. Findings and Recommendations 86. This section will include an evaluation of the screening process and recommendation will be provided whether significant environmental impacts exist needing further detailed study or EIA. If there is no need for further study, the IEE itself, which at times may need to be supplemented by a special study in view of limited but significant impacts, becomes the completed environmental assessment for the project and no follow-up EIA will be needed. If an EIA is needed, then this section will include a brief terms of reference (TOR) for the needed follow-up EIA, including approximate descriptions of work tasks, professional skills required, time required, and estimated costs. The Bank's Environment Guidelines provides a guide for preparing the TOR for different projects.

Engconsult Ltd.

Appendix 19 Annex 5 134

H. Conclusions 87. This section will discuss the result of the IEE and justification, if any, of the need for additional study or EIA. If an IEE, or an IEE supplemented by a special study, is sufficient for the project, then the IEE with the recommended institutional and monitoring program becomes the completed EIA.

Engconsult Ltd.

Appendix 19 Annex 6 135 Annex 6 Outline of Summary Initial Environmental Examination (SIEE) Report

A. Introduction (1/2 page) 88. This section will include the purpose of the report, extent of the IEE study and brief description of any special techniques or methods used. B. Description of the Project (1/2 page) 89. This section will include the type of and need for the project; and project location, size or magnitude, operation, and proposed schedule for implementation. C. Description of the Environment (2 pages) 90. This section will include the physical and ecological resources, human and economic development, and quality of life values. D. Forecasting Environmental Impacts and Mitigation Measures (2-4 pages) 91. This section will identify "no significant impacts" from those with significant adverse impacts and will discuss the appropriate mitigation measures, where necessary. E. Institutional Requirements and Environmental Monitoring Plan (1 page) 92. This section will describe the impacts to be mitigated, and activities to implement the mitigation measures, including how, when, and where they will be implemented. The environmental monitoring plan will describe the impacts to be monitored, and when and where monitoring activities will be carried out, and who will carry them out. F. Public Consultation and Disclosure 93. This section will describe the process undertaken to involve the public in project design and recommended measures for continuing public participation; summarize major comments received from beneficiaries, local officials, community leaders, NGOs, and others, and describe how these comments were addressed; list milestones in public involvement such as dates, attendance, and topics of public meetings; list recipients of this document and other project related documents; describe compliance with relevant regulatory requirements for public participation; and summarize other related materials or activities, such as press releases and notifications. This section will provide of summary of information disclosed to date and procedures for future disclosure. G. Findings and Recommendations (1-2 pages) 94. This section will include an evaluation of the screening process, and recommendation will be provided whether significant environmental impacts exist needing further detailed study or EIA. If there is no need for further study, the IEE itself, which at times may need to be supplemented by a special study in view of some small significant impacts, becomes the completed EIA for the project and no follow-up EIA will be needed. If further additional study is needed, then this section will include a brief terms of reference (TOR) for the needed follow-up EIA, including approximate descriptions of work tasks, professional skills required, time required, and estimated costs. The Bank's Environment Guidelines provides a guide for preparing the TOR for different projects. H. Conclusions (1/2 page) 95. This section will discuss the result of the IEE and justification if any of the need for additional study or EIA. If an IEE or an IEE supplemented by a special study is sufficient for the project, then the IEE with the recommended institutional requirements and monitoring program become the completed EIA.

Engconsult Ltd.

Appendix 19 Annex 7 136

Annex 7 Outline Terms of Reference for Consulting Services for Environmental Assessment

A. Objectives 96. The objective of the services are (a) to conduct environmental impact assessment (EIA)/ initial environmental examination (IEE) of the proposed subproject to identify potential environmental impacts on physical, environmental, ecological, social, cultural and economic resources, and (b) to prepare EIA/IEE report along with environmental management plan. The duration of an EIA study is 5 months and an IEE study is 3 months. B. Scope of Work 97. The consultant’s scope of work will include the following tasks: (i) reviewing prevailing government regulations and donor guidelines governing the assessment and management of environmental impacts of road projects; (ii) prepare a scoping document for the environmental studies to be carried out under the project; (iii) undertake the EIA/IEE study to assess the direct and in-direct environmental impacts of the Project such as: (a) ecological impacts (plants and wildlife); (b) soil erosion and desertification; (c) protection of wetland habitat; (d) impact of quarry sites; (e) impact of construction camps on local environment (natural and social); (f) operational traffic safety measures; (g) areas with known archeological value; and (h) potential spills of hazardous or toxic chemicals and an appropriate response plan for the Project; (iv) prepare the EIA/IEE report and its summary in accordance with ADB’s Environmental Policy (2002) and related Environmental Assessment Guidelines (2003) as well as ADB's Public Communications Policy (2005); (v) the EIA/IEE study should address all potential direct and indirect environmental impacts of the project. The assessment of environmental impact should be presented in the order of project cycle: pre-construction, construction and operation. This EIA/IEE report should include the detailed environmental management plan that should be included in the bidding document; Analyze the likely impacts during construction and operation and suggest appropriate mitigation measures; (vi) conduct formal public consultations with affected people (two consultations for EIA and one consultation for IEE). The first consultation aims to gather environmental concerns from affected people and the final consultation aim to share the result of the assessment and the proposed mitigation measures. The list of people attended the consultation, time and locations; subject discussed during consultation should be recorded in systematic manner and should be attached in the EIA/IEE report as an appendix; (vii) solicit and incorporate comments on the draft EIA and summary EIA reports from ADB, MPW, NGOs, civil society, and other stakeholders. Finalize the report to accommodate inputs from all the stakeholders. (viii) submit the reports to MOEPNR and make presentation as required by MOEPNR to obtain a environmental impact permit;

C. Organization and Staffing 98. The services are expected to be provided a team comprising one international and one national consultant. 99. The International Environmental Specialist shall have at least 15 years experience and familiarity with all aspects of environmental management and with significant experience in environmental management and monitoring of projects, environmental assessment and / or implementation of environmental mitigation measures on construction projects. The specialist shall also have experience working in teams of multi-discipline experts and leading a national team of consultants. Candidates with higher degrees in environmental engineering or environmental science or environment management are preferred.

Engconsult Ltd.

Appendix 19 Annex 7 137

100. The National Specialists shall at least be graduates in environmental science, environmental engineering, geological science, engineering hydrology, biology or related discipline with significant experience in environmental management and monitoring of projects, environmental assessment and/or design and implementation of environmental mitigation measures. D. Budget 101. The estimated cost for preparation of each IEE and EIA are provided in Table A19. 1 and Table A19. 2, respectively. A team of International and national specialists are recommended for these studies. Generally for government funded projects, the environmental assessment team would include an environmental specialist, a geologist, and a biologist. Table A19. 1: Estimated Cost of IEE Preparation for each Subproject Item Unit Quantity Unit Cost, USD Total International Environmental Specialist Person-month 3 25,000 75,000 National Environmental Specialist Person-month 3 3,500 10,500 Land Transportation Months 3 1,250 3,750 Data collection, sample analysis Lumpsum 1 4,000 4,000 Communication Months 3 500 1,500 Report production and distribution Lumpsum 1 2,000 2,000 Public Consultation Lumpsum 1 2,000 2,000 Total 98,750 Table A19. 2: Estimated Cost of EIA Preparation for each Subproject Item Unit Quantity Unit Cost, USD Total International Environmental Specialist Person-month 4 25,000 100,000 National Environmental Specialist Person-month 4 3,500 14,000 National Hydrologist Person-month 2 3,500 7,000 National Geologist Person-month 2 3,500 7,000 National Biologist Person-month 2 3,500 7,000 Land Transportation Months 5 1,250 6,250 Communication Months 5 500 2,500 Data collection and sample analysis Lumpsum 1 7,000 7,000 Report production and distribution Lumpsum 1 2,000 2,000 Public Consultation Lumpsum 2 2,000 4,000 Total 156,750 E. Supervision 102. The team will work in association with the PMU/RD, reporting to the project director of the PMU/RD on a day-to-day basis. F. Outputs 103. The team’s outputs will include: (i) an inception report reviewing the available environmental reports and prepare preliminary assessment of impacts associated with the proposed project, (ii) an interim report; (iii) a draft final report, containing a description of project environment, anticipated environmental impacts for various alternatives and proposed environmental management, monitoring and mitigation plan, and (iv) final report incorporating comments from ADB and other stakeholders.

Engconsult Ltd.

Appendix 20 138 Appendix 20 SUMMARY ENVIRONMENTAL ASSESSMENT REPORT

I. INTRODUCTION

1. This summary environmental impact assessment (SEIA) was prepared as part of the Asian Development Bank (ADB) project preparatory technical assistance (PPTA) for the Ajara Bypass Roads Development Project in Georgia.1 The PPTA is preparing a multitranche financing facility (MFF) for the proposed Subregional Road Corridors Development Program with the Ajara bypass road (more specifically, the Kobuleti bypass road) as a key component of an ensuing project. The Executing Agency for the Project is the Ministry of Regional Development and Infrastructure and the Implementing Agency is the Roads Department under the ministry. Financing for the Project will be provided in two tranches.

2. The PPTA is to (i) package the MFF; (ii) conduct a feasibility study and preliminary engineering design for the bypass roads to Batumi and Kobuleti on the Poti–Batumi–Sarpi Highway in Ajara Autonomous Republic, including engineering, economic, social, and environmental viability tests; and (iii) undertake a road sector and institutional analysis. The 81-kilometer (km) Poti–Batumi–Sarpi road along the western coast of Georgia is a key highway and international transit route in Georgia. It connects to the major Georgian Black Sea ports, i.e., Batumi and Poti, and several holiday resorts, particularly in Batumi and Kobuleti. Heavy traffic on the existing road has resulted in a significant increase in congestion and accidents, particularly in Batumi and Kobuleti, and especially during the tourist season. The Government plans to construct two bypass roads around Batumi and Kobuleti to address these problems, and requested ADB assistance to prepare and finance these two roads.

3. As part of MFF preparation, an environmental assessment and review framework was prepared to provide a basis for preparing the environmental impact assessment (EIA) for the Project to be financed by the first and second tranche, and any EIA or initial environment examination (IEE) for projects funded by subsequent tranches. The cumulative impact assessment for the entire MFF will be prepared during MFF processing and implementation. The EIA for the Project was prepared in accordance with ADB’s Environment Policy (2002) and Environmental Assessment Guidelines (2003), as well as the Law on Ecological Examination (2007) and Order No. 515 of the Minister of MOEPNR of Georgia on Approval of Rules on Conduction of Ecological Examination (2008). The EIA is based on detailed field investigations; baseline environmental monitoring; review of proposed civil works; review of the 2005/06 prefeasibility study report; consultations with relevant stakeholders during the scoping study of the EIA; public consultation; and detailed review and analysis of available spatial databases for all environmental parameters in project areas such as terrain, soils, geology, rivers, forests, protected areas, land use, and population. The EIA was prepared with the active cooperation of the Roads Department, Ministry of Environmental Protection and Natural Resources (MOEPNR), and ADB’s project team in Manila.

4. Based on consultation with MOEPNR and Roads Department staff, the consultant understands that an environmental impact permit is not required at this time as no construction activity will be carried out based on the outcome of this feasibility study. An environmental impact permit is required for the revised EIA report, which will be based on the detailed design.

1 ADB. 2008. Technical Assistance to Georgia for Preparing the Ajara Bypass Roads Development Project. Manila.

Engconsult Ltd.

Appendix 20 139 A. II. DESCRIPTION OF THE PROJECT

5. The Project will construct 42 km of new roads and widen 6 km of existing roads along the Poti–-Batumi–Sarpi road, bypassing Kobuleti and Batumi (map). The project road is located close to Ispani Mire (Kobuleti Protected Areas), identified by the Ramsar Convention on Wetlands (RAMSAR) as a wetland of international importance and an area protected by the International Union for Conservation of Nature. Construction of the project road will involve new construction and rehabilitation of bridges, culverts, and other associated drainage structures; tunnel construction; land acquisition and resettlement; site preparation; land clearing; disposal of gravel, soil, vegetation, and unstable material; construction of temporary access roads to construction sites; setting up of temporary construction camps to house workers; excavation and operation of borrow pits; operation of a boulder quarry; extraction of material for embankments using cut and fill procedures; protection of landslide, rockfall, and snowfall areas; measures to protect critical side-slopes; and extraction and cartage of sand for aggregate mixtures, cement works, and asphalt plants.

Figure 1: Location Map of Project

Engconsult Ltd.

Appendix 20 140 6. The first tranche will cover the first 34 km (km 0–34); the second tranche will finance the remaining 14 km (km 34–48). Work financed by the first tranche is expected to commence in May 2010, and work financed by the second tranche in July 2010. Each is forecast to require 3 years to complete.

III. DESCRIPTION OF THE ENVIRONMENT

7. Topography. The first 14 km of the project road passes through flat coastal plain with elevations ranging from 2 to 20 meters (m); the remaining 34 km passes through rolling and hilly terrain with elevations ranging from 20 to 250 m. The road passes through three hill ranges: Zeda Ackva hill range from km 20 to 26; Makhinjauri hill range from km 27 to 36; and Peria hill range from km 40 to 44. Five tunnels will be constructed along the ridges of these hill ranges (600 m tunnel from km 22.8 to 23.4 on Zeda Achkva ridge; three small 500 m tunnels from km 33.7 to 34.5 on small ridges in Makhinjauri; and a 700 m tunnel from km 41 to 41.7 on Peria ridge). The hilly sections, especially between km 27 and 36, are susceptible to soil erosion, landslides, and mudslides due to steep slopes, poor geological conditions, and higher sand content of surface soils—as noticed from the newly constructed 600 m tunnel on the existing road in this section.

8. Soils. Soils in the project area consist of laterite clay and loam, originating from intense weathering of volcanic rocks; and deposits of alluvial sediments. The coastal plain comprises mainly hydromorphic and alluvial soils due to abundant rainfall and the relief of the plain. Red soil predominates in hilly areas. The soils in the project area are classified into Haplic Nitisols. Surface soils are clay loam with a sand fraction of 45%, while subsurface soils are clay. Generally soils near the coastal and river alluvium consist of coarse sand with high infiltration capacity. In terms of vulnerability to soil erosion and slope stability issues, the project road can be divided into three zones: km 0–14, low vulnerability; km 14–20, medium vulnerability; and km 20–48, high vulnerability. Because of steep slopes, high sand content, good drainage, and an active geodynamic process, soils in hilly areas are susceptible to landslides. The Batumi area is within an exceptionally strained geodynamic area, hence landslides are more frequent.

9. Climate. The project area is located in a subtropical climate—humid with prolonged rains. Average summer temperatures range from 22 degrees Celsius (0C) to 240C in lowland areas and 170C to 210C in the highlands. Average winter temperatures range from 40C to 60C along the coast, while the interior areas and mountains average around minus 20C.

10. Most precipitation in the project area is in the form of rain. Average annual precipitation in Batumi is 2,718 millimeters (mm). September is the wettest month, averaging 335 to 410 mm; May is the driest, averaging 92 mm. Batumi generally does not receive significant snow (accumulating snowfall of about 30 centimeters); 12 days of the year have snow cover. Average humidity ranges from 67% to 80%. Average wind speed is 15.8 km/hour (hr), reaching more than 20 km/hr during December–February with lower temperature. Winds are generally from the southwest. Average annual precipitation in Kobuleti is 2,320 mm; September is the wettest month and May the driest. Average wind speed is 11 km/hr, with a maximum of 14 km/hr during February. Wind is mostly from the southwest (30%) and northeast (23%).

11. Geology. The project area is located in the Ajara–Trialeti zone situated within the Meskheti, Shavshvebi, and Chakvi mountain rings. Middle Eocene-aged sediments are spread over this area. The lower part of middle Eocene is lithologically represented with tabular tuff, tuffo-gravel, argillite, and limestone. These sediments are mostly spread over the middle and upper part of the Kintrishi river and flanks of the Meskheti mountain ring, in the draw of the

Engconsult Ltd.

Appendix 20 141 Chorokhi river, and within the Chaisubani village territory. The upper part of middle Eocene is represented by a mass of fragmented volcanic breccias, tuff, and clinker stratums. In some places, andesite, basalt, tuffo-gravel, tuff, and marl can be found. Those sediments are mainly spread in the Ajara cauldron, flanks of the Chakvi and Shavshebi mountain ring, and draw of the Achara Wyali river. In the middle part of Chakvis Wyali, intrusive sediments of upper Eocene are lithologically represented by syenites and syenite-diorites. Geomorphologically, the northern part of the project area comprises the Kobuleti coastal plain crisscrossed by numerous streams; the southern part is characterized by hilly ranges with deep gorges.

12. Seismicity. The project area is located in a seismic intensity zone of Richter scale 7. The intensity of tectonic zones is calculated for 2% probability (expectation time 50 years) according to the Richter scale. On a modified Mercalli scale, the project area is in an earthquake intensity zone of IV (moderate intensity). The eastern part of the project area, in the eastern part of Kobuleti near Ozurgeti, is in a high earthquake intensity zone.

13. Water Resources. The project area has significant surface water resources. The project road crosses three major rivers (more than 15 km long), six medium-sized rivers (from 10 to 15 km long), and a several small streams and hillside drainage systems. The major rivers along the alignment are the Choloki (km 3.80), Kintrishi (km 15.00), and Chaqvistskali (km 26.10). The Chorokhi, the third largest river in Georgia, is located about 1 km away from the southern end of the project road. Discharges of all the rivers in the project areas were estimated for return periods of 10, 100, and 200 years.

14. The project area north of Kobuleti and south of Batumi is in low runoff areas of less than 1,000 mm. However the main project area is located in middle runoff areas, where annual runoff ranges from 1,000 to 2,000 mm. The eastern project area is mostly mountainous with high runoff zones. River water in the project area is extensively used for municipal drinking water supply, irrigation, industries, hydropower, fisheries, and construction. In addition, sand, gravel, and stones from these rivers are used extensively for construction.

15. Groundwater is the major source of drinking water. All towns and major villages in the project area have central water distribution systems, while smaller villages and isolated dwellings depend on private wells and springwater. Shallow aquifers of alluvial deposits are located near the major rivers and their floodlands. These aquifers generally comprise pebbles and sand with intermediate lenses of clay. Generally aquifers are in direct hydraulic interaction with the river waters and are excellent sources of drinking water. A well field comprising eight bore wells located along the Kintrishi riverbank at km 16 (on the upstream side) supplies water to the Kobuleti water supply system. In general, the water table is very shallow in flat and coastal areas with depth varying from 3 to 6 m.

16. Quarries and Borrow Pits. The quarries and borrow pits required for construction activities are located in several river sites. Two major quarry sites in the Chaqvistskali and Chorokhi rivers are located near the project road. The sand, gravel, and stones from these sites are suitable for road construction. Several stone-crushing plants are located on the banks of the Chorokhi River near Batumi. Material extraction from these sites is regulated and a permit for the Chorokhi site is issued to a local company with quantity restrictions; the Chaqvistskali site will require a permit from MOEPNR.

17. Water Quality. The quality of the water in all the rivers is generally very good, except for Choloki because of high turbidity (5.53 nephelometric turbidity units [NTU]) and low dissolved oxygen (1.8 milligrams per liter). All other rivers have turbidity of less than 0.65 NTU and

Engconsult Ltd.

Appendix 20 142 dissolved oxygen of more than 7 milligrams per liter. The rivers all have potable water with total dissolved solids ranging from 75 to 164 milligrams per liter. All the rivers are excellent sources of water for drinking and construction. No petroleum hydrocarbon or agriculture pollution was identified in the river water.

18. Noise. Noise was measured at three locations to assess background noise along the project alignment, noise along the existing road, and noise along the railway line. Background noise along the project alignment on new roads measures 30 decibel (dBA). Average hourly noise 25 m from the center of the existing road is 65 dBA. Noise 25 m from the railway line when loaded trains are moving is 77 dBA, and when empty trains are moving 73 dBA. The average hourly noise 25 m from the railway line is 58 dBA

19. Air Quality. Traffic, industries, and port operations are the major sources of air pollution in the project area. Until 1991, air quality monitoring in Georgia was carried out regularly and systematically in 11 cities, including Batumi. Dust, carbon monoxide, nitrogen dioxide, and sulfur dioxide were measured continuously three times a day. The most recent air quality monitoring data for Batumi is from 1995 to 2004, with the average yearly concentrations of dust, sulfur dioxide, and nitrogen dioxide slightly exceeding the norms. The highest dust concentration exceeded maximum permissible levels by 1.2 times. In more recent years of the data, the concentration of nitrogen dioxide increased substantially, while hydrogen sulfide pollution decreased.

20. Ispani Mire. Ispani Mire, a national protected area and RAMSAR wetland site (number 894), is located near the project road from km 6 to 12. The project road passes well beyond the safe buffer zone of the mire.2 The mire, a part of the Kolkheti lowland system of wetlands, is included in the RAMSAR list because (i) Ispani II Mire is one of two percolation bogs discovered worldwide (another one is Imnati, which also occurs in the Kolkheti lowland, about 20 km north of Ispani II Mire); and (ii) vegetative cover of the Ispani I and Ispani II peatland complex contains relict plant communities (e.g. Carex lasiocarpa, Molinia litoralis), northern Palaearctic species (e.g., Drosera rotundifolia, Sphagnum (S) imbricatum, S. palustre, S. papillosum, S. rubellum, S. auriculatum), Kolkheti elements (e.g., Rhododendron luteum, Rhododendron ponticum, Rhynchospora caucasica, and Vaccinium arctostaphyllus), as well as species such as Frangula alnus, Rubus spec., or Alnus barbata at the margin of the peatlands.

21. Flora. Natural vegetation of the project area comprises mostly grasses and bushes such as Vaccinium artostapylia and Rhododend luteum. In hilly areas, the vegetation mostly includes board-leaved trees; the main species are Fagues orientilia, Casatanea sativa, Carpinus caucasica, Cuercus dachorocalis, Tilia caucasica, Aser platanoides, Pinus sosnovsky, and Picea orientalis. The Ispani I and Ispani II peatland complex is also important because of its floristic composition. The peatland vegetation includes a high percentage of peat-moss (Sphagnum) species in the cover, and low density of vascular plant species. The existence of boreal (tundra and taiga) flora elements like Drosera rotundifolia, Sphagnum imbricatum (which is rather an unusual phenomenon for subtropical latitudes), as well as Colkhic flora elements like Rhododendron ponticum, R. Iuteum is another feature of the uniqueness of these peatlands.

22. Fauna. Despite its small size and proximity to an urbanized zone, the Ispani I and Ispani II peatland complex and its adjacent areas provide habitats to Lutra lutra (species of Georgian

2 The alignment agreed upon by the authority of the Kobuleti Protected Area runs with a 100-m distance from the buffer zone for a 500-m stretch in its narrowest gap and with a 500-m distance for a 1.5-km stretch.

Engconsult Ltd.

Appendix 20 143 Red List), Canis aureus, Felis silvestris, Myocastor coypus, Meles meles, Erinaceus concolor, Rattus norvegicus, Rattus rattus, Sorex volnuchini, Mus musculus, Myotis blythii, and Pipistrellus pipistrellus. They also serve as a good refuge for migratory birds from Eurasia and Africa: Circus cyaneus, Egretta garzetta, Merops apiaster, Lymnocryptes minimus, falco vespertinus, Grus grus, Egretta alba, and Haliaeetus albicilla. The area’s small population of reptiles (herpetofauna) includes Emys orbicularis, Elaphe longisima, Lacerta agilis, Lacerta media, Natrix tesselata, and Natrix natrix. Amphibibians include Triturus vulgaris, Triturus cristatus, Rana ridibunda, Hyla arborea, Bufo viridis, and Bufo verucozissima. Various habitats support large numbers of invertebrates.

23. Fish, Fisheries, and Aquatic Biology. The rivers in the project area are populated by a wide variety of fish. Major types in the Kintrishi and Chorokhi rivers, and the Black Sea are sturgeon (Acipenser guldenstadi Brandt); beluga; salmon; and grey mukket species including Mugi cephalu, ramada risoo, Mugli saliens risoo, and Mugil cephalus risso. Other prominent fish include barbell, bullhead, and herring. Fish species noted in the Ispani wetland system are Siluris glanis, Leuciscus cephalus, and Cyprinus carpio. No marine protected areas, mangroves, or coral reefs are located near the project area.

24. Land Use. The project area includes swamps, badlands, temporal dwellings, perennial plantations (e.g., tea, citrus), an industrial area, agricultural lands, arable lands, and forest. The road mostly passes through arable land comprising pastures, agricultural lands, and tea and citrus plantations. The locations of settlements are included on the map. Corn, the major agricultural product, is widely grown on all agricultural land. Vegetables are the second major product.

25. Population. Ajara has a population of 378,000 (2007 estimates). Its ethnic groups include Laz, Russians, Armenians, Greeks, and Abkhaz. The project area is within the administrative units of Kobuleti municipality, Batumi city, and Khelvachauri municipality. The project road passes through 16 villages and settlements, which fall under 10 sacrebulos (subdistrict administrations) and two rayons (districts).

26. Economic Resources. Tourism is the major industry in the project area; extensive resort infrastructure is located along the existing road in Batumi and Kobuleti. Other major industrial activities include an oil refinery; port; shipyard; and machine-building, zinc-plating, and furniture factories; as well as a range of light industries. In addition, several stone-crushing factories and an asphalt plant are located near Batumi. Industries and municipal infrastructure along the project road include an oil terminal, Kobuleti drinking water pumping station, and electrical substations. Agriculture, including tea and citrus plantations, is the major economic activity in the project area.

IV. ALTERNATIVES

27. During the prefeasibility study, three alignments were examined between the Cholokhi River and Sarpi to recommend an alignment for the Kobuleti and Batumi bypass roads. The present feasibility study reviewed the three alignments and made modifications to the prefeasibility-recommended alignment, considering environmental, engineering, and social issues. It also recommends a final alignment. The environmental, social, and engineering issues considered for assessing various alternate alignments are (i) location of Ispani Mire; (ii) cultural sites, such as cemeteries; (iii) land acquisition and resettlement; (vi) geological features such as stability, landslides, and cut and fills; (vi) design standards of the Roads Department and other

Engconsult Ltd.

Appendix 20 144 applicable international standards; and (vii) impact on infrastructure facilities such as the water pumping station of Kobuleti, substations, and industries along the corridor.

28. Alternatives for the Section from Kilometer 0 to 14. Various alignments were considered in this section to avoid impact on Ispani Mire. The prefeasibility alignment passes parallel to the 3.3 km long Ispani Mire (from km 2.4 to 5.7) through a narrow strip of land between a strip of residential development and a bund constraining the mire. The prefeasibility alignment is located very close to the mire (within 140–200 m) for about 1.6 km (from km 4.22 to 5.70). Road development on the narrow strip on the prefeasibility-recommended alignment will damage the peatland system and its hydrological integrity, and nullify all recent efforts to establish general awareness among the population of the natural value of the unique ecosystem of the mire. The feasibility study recommends an alternative alignment that passes to the eastern side of the mire and is well away from the safe buffer zone of the mire.

29. To meet the Project’s design standards, two possible alternatives were studied for the starting point of the Project. One alignment leaves the existing Poti–Batumi road north of the Choloki River (immediately south of the amusement park), thus starting the project road in Juria province. It then crosses the Choloki on a new bridge and turns east toward the railway line. The alternative start is similar to the prefeasibility-preferred alignment and exits the existing road just south of the Choloki River using a modified trumpet interchange. The advantage of the second option is that it is about 0.5 km shorter and requires only an additional two-lane bridge over the Choloki River. The disadvantage is that the alignment is constrained within the first kilometer and is below the geometric standard for a high class road. The consultant recommends the first alignment starting south of the amusement park.

30. Alternatives for the Section from Kilometer 14 to 28. The main oobjective in assessing alternatives for this section is to avoid impact on the physical environment such as soil and geology; and existing infrastructure facilities such as the pumping station, substation, and cemetery. The proposed alignment generally follows the prefeasibility alignment except from km 14 to 17 and km 18 to 20. In the former case, the line is moved slightly eastward to avoid a chemical factory and then a water pumping area and associated substation. In the latter case, the line is moved to avoid a new cemetery, avoid going on top of an existing minor road, and allow sufficient height to pass over a new minor access road and bridge. At km 22, the prefeasibility alignment cuts through the Zeda Achkva ridge. A closer inspection of this section indicates that the cut would be deep in erodible material, require demolishing several houses, cut a significant access road, and require steep approach grades. The consultant proposes replacing this cut with a 400 m tunnel.

31. Alternatives for the Section from Kilometer 28 to 34. At km 28.5, the new alignment joins the recently reconstructed existing road and goes through an existing 600 m tunnel to descend to the old road at km 32. This section was constructed as a two-lane road, however, the tunnel is twin bore with a total of four lanes; earthworks have been mostly completed for four lanes. Significant slope protection will be required. This section of the alignment follows the existing Kobuleti–Batumi road parallel to the coast and the railway line, and is heavily developed. The road will require widening to four lanes plus adequate shoulders as this section will carry both through and local traffic. Acquisition will be required for about 10 buildings, part of a school frontage, and a hotel parking area. An alternative alignment moving east of the existing road from km 32 to 34 would encounter steep terrain and still be in residential built-up areas; therefore it is not recommended.

Engconsult Ltd.

Appendix 20 145 32. Alternatives for the Section from Kilometer 34 to 48. In the prefeasibility study, three alignments were examined for this section. The westernmost alignment encroaches on the Batumi Municipal Development Area and passes through some heavily built-up areas; even more built-up now than when the prefeasibility study was carried out. The advantage of this alignment is that it is 0.5 km shorter and passes through easier terrain, obviating the need for a tunnel. After inspecting other two alternatives,

33. The prefeasibility-recommended alignment first rises to cross the rail line, turns southeast through three short tunnels, and descends to the Chakvis Korolistskali River, crossing the river (km 36) and some commercial structures on a 600 m bridge. It then follows the slope of the land to descend to the Bartskhana River, also crossing this and the riverside commercial development on a 600 m bridge (km 39); both bridges are about 10 m above ground level, but will need long spans. The alignment then gradually rises, following the line of the Batumi Municipal Development Area to the west of Batumi, and runs along but outside the development plan margin. At km 40.5, the alignment crosses a 200 m bridge to enter a 700 m tunnel, and then descends to cross the Batumi–Alkhaltsikhe road at km 44.5, with an interchange, to then turn west to join the Senaki–Poti–Sarpi road at km 48. This alignment ends at the existing bridge over the Chorokhi River.

34. The proposed third alignment follows the second alignment for the first 5 km and then proceeds further east, bypassing the developed area around the Bartskhana River and staying up to 2–3 km east of the Batumi Development Plan boundary. It also uses a 400 m tunnel, rejoining alignment two just before the Batumi–Alkhaltsikhe road. This alignment is 0.5 km longer and an estimated 8% more expensive.

35. Both alignments two and three were inspected. Both have some new residential development. In addition, alignment three passes through a new cemetery close to Bartskhana River; this would be difficult to avoid without demolishing residential houses. Thus the prefeasibility advantages of alignment three appear to have disappeared. Therefore, except for a minor realignment at km 45 to avoid the cement factories and another minor alignment change in the vicinity of the tunnel to minimize demolishment of houses, the consultant recommends an alignment close to the prefeasibility alignment two. The length of the tunnel will need to be increased to 700 m to reduce the tunnel grade to 3%.

36. Do-Nothing Option. The do-nothing or without-project option is not really an alternative since the objective of the Project is to construct bypass roads to divert traffic from already congested roads in Kobuleti and Batumi. Congestion and accidents on the existing road have increased significantly, especially during the summer tourist season. The without-project scenario will continue to increase the negative impacts generated by increased traffic loads on the existing road, which lacks adequate capacity (e.g., traffic jams, noise, low speed, higher emissions, accidents). With continuous growth of tourist and residential infrastructure, the existing road will continue to deteriorate affecting travel by local residents, tourists, and other transit road users.

V. ANTICIPATED ENVIRONMENTAL IMPACTS AND MITIGATION MEASURES

37. Anticipated environmental impacts of the Project, on the basis of the EIA study, are broadly classified into three categories: preconstruction, construction, and operation.

(i) Preconstruction. Anticipated impacts include

Engconsult Ltd.

Appendix 20 146 (a) pedestrian and high-speed traffic, (b) land acquisition issues for new right-of-way areas, and (c) compensation issues for other fixed assets from acquired right-of-way land. (ii) Construction. Environmental impacts could include

(a) increased soil erosion, landslides, and/or siltation, including increased risks to downstream rivers and coastal resources as a result of cut and fill operations; (b) impacts of clearing right-of-way, removing vegetation (trees and shrubs), and disposing of spoil; (c) impacts of extracting and transporting construction material from existing quarry sites; (d) impacts of temporary use of land immediately adjacent to the road for siting of contractor’s yard, asphalt plant, and construction camps; (e) reduced air quality and visibility (air quality impacts and/or noise pollution from construction activities, quarry sites, material storage sites, temporary diversion roads, excavations, vehicle and equipment use, and asphalt mixing plant); (f) water pollution (water and soil pollution) at bridge rehabilitation sites from improper handling and disposal of wastes and materials; (g) drainage from construction camps, material stockpiles, excavations, and quarry activities; (h) interruption to smooth traffic flow, increased traffic congestion, and public safety problems; and (i) social conflicts due to project activities.

(iii) Operation. Possible environmental impacts include

(a) increased traffic volume and related air, noise, and public safety concerns; and (b) possible spills from the transport of hazardous cargo.

38. Perceived benefits from the Project include

(i) reduction of traffic congestion on Kobuleti and Batumi and hence improved air and noise quality, and road safety; (ii) reduced soil erosion and landslides due to slope stabilization measures; (iii) improved access in the project area and resulting economic development; (iv) faster route for international traffic between Poti and Sarpi; and (v) improved access to tourist facilities.

39. Usually implementation of civil works does not have major or long-term impacts, but localized, short-term impacts can be addressed in detailed designs and through application of site-specific environmental management plans (EMPs). These construction-related impacts can be mitigated by (i) the contractors’ work practices, especially those related to the storage of construction materials and cleanliness of work sites; (ii) cooperation of local authorities with the contractor for traffic management and use of public space and utilities; (iii) project management’s strict enforcement of adequate construction practices and standards; and (iv) the incorporation of mitigation measures identified in the EIA. An environmental management and

Engconsult Ltd.

Appendix 20 147 monitoring plan (including cost estimates) was prepared to mitigate the negative environmental impacts of the Project (Appendix 1).

40. Soils and Materials. The project road is a new alignment with the first 22 km passing through high, fertile, arable agricultural land. The main impacts on the soil during construction will be from (i) loss of topsoil from construction sites; (ii) conversion of existing land uses such as agriculture and plantations to stockpiles of materials, and damage to temporarily acquired land; (iii) cut and fill operations; (iv) extraction of fill materials from the cut section and/or borrow pits; (v) soil erosion on mountainous slopes, side slopes, borrow pits, and uncompacted embankments; and siltation; and (vi) land contamination from hazardous and toxic chemicals and construction material spillage. In addition, site preparation activities generate spoils involving crop residue, grasses, trees, and earth.

41. Topsoil of cultivated land used for temporary work areas will be stripped off and stockpiled, to be replaced when construction is completed and the cultivated land rehabilitated. Contractors will be encouraged to minimize usage of productive agricultural land and convert it to its original state after completion of civil works. Embankments should be monitored during construction for signs of erosion; long-term material stockpiles will be covered to prevent wind erosion. A balanced cut and fill approach is designed for the Project. Deep cuts through Zeda Achkva ridge near km 22.8 and Makhinjauri ridges at km 34 will be avoided by constructing tunnels. Excess cut, if any, will be used to shape the embankment side slopes, strengthen the toe of the high slopes road bench, and fill the low-lying government lands for plantation in preapproved areas. Contractors will submit a spoil plan to the Roads Department for approval. The spoil plan should show the location of any borrow pits to be used, fill location for excess cut, and measures to rehabilitate these pits and cuts upon finalization of the Project. The spoils generated from the site preparation will be disposed of in these preapproved sites.

42. Soil Erosion and Slope Stability Problems. Construction activities in zones that are highly vulnerable for soil instability, especially in hilly areas, will intensify the effects of natural soil erosion unless precautionary measures are taken. In addition, construction activities will cause soil erosion due to cut and fill at different locations especially in hilly sections, and bridge and tunnel sites. The impacts of soil erosion and unstable side slopes are (i) increased runoff and sedimentation causing a greater flood hazard to downstream areas; (ii) loss of topsoil affecting the growth of vegetation resulting in ecological imbalances; (iii) destruction of the aquatic environment in nearby rivers, streams, and sea caused by erosion and/or deposition of sediment damaging the spawning grounds of fish; (iv) destruction of vegetation by burying or gullying; and (v) development of unsightly cuts and fills that have been riddled by uncontrolled erosion and gullying.

43. Soil erosion and landslides will be minimized by preventive measures incorporated in the engineering design, such as (i) constructing 12,250 m of retaining wall near the deep-cut sections, bridges, overpasses and tunnels; (ii) planting shrubs buffer strips and grass; (iii) ensuring appropriate compaction, rock lining, and slope walls near the proposed overpasses, tunnels, and existing tunnels; and (iv) constructing side ditches, berms, dikes, sediment basins, fiber mats, mulches, and drainage channels. Engineering design also considers cross-drainage and floodwater flows for road links on elevated fills that could obstruct natural drainage. Any cut and fill that results in a slope steeper than 45 degrees will need to be secured by a protective wall and protected by bioengineering methods. As a long-term mitigation measure, land degradation must be stopped. Maintaining a vegetative cover in mountain areas will reduce erosion through runoff, and land and mud slides. The EIA recommends planting trees and landscaping along the roads. The landscaping will beautify the

Engconsult Ltd.

Appendix 20 148 road and protect it from soil erosion. To reduce maintenance, the trees and bush species that naturally grow in the project area are recommended for plantations (Juglans regia, Populus nigra, Quercus iberica, Crataegus kyrtostyla, Spiraea hyperiafolia, Rosa canina, Jasminum fructicans, Corylus avellana).

44. Water. The project road crosses nine major rivers and several small streams, and could affect the surface runoff flow pattern. All drainage works are designed based on historical flood data and flood forecasting. Design discharges for all the rivers are prepared for return periods of 50, 100, and 200 years. The design of drainage facilities considered soil protection measures in addition to water resources management. A design discharge for a return period of 50 years is considered for culverts and 100 years for bridges. The Project will construct 18 major bridges with lengths ranging from 12 to 600 m, and 7 small 12 m bridges. In terms of water quality, the Project will increase the silt load during construction at bridge sites. Embankments and construction materials (fill, sand, and gravel) are subject to washout with rainwater. Hydrocarbon leakage and spills could occur from storage and mixing plants, discharge of sewerage from work camps to water resources, or percolation through seepage and contamination of the local water table.

45. To mitigate this, (i) small drainage structures are proposed in the engineering design to divert water flow near the bridge sites, i.e., for road sections across streams, soil and stones will be disposed of properly so that they do not block the rivers and streams; (ii) open surfaces will be covered by grasses and creepers to reduce material that can be washed away; (iii) hydrocarbons will be stored in secure, impermeable, and bounded compounds away from surface waters; and all contaminated soil will be properly handled (as a minimum, these areas will be contained such that any spills can be immediately contained and cleaned up; any petroleum products used in the preparation of bitumen mixes must be carefully managed to avoid spills and contamination of the local water table; (iv) construction and work sites will be equipped with sanitary latrines that do not pollute surface waters, and contractors will submit a simple sewage management plan; (v) cofferdams, silt fences, sediment barriers, or other devices will be included in the design to prevent migration of silt during excavation and boring operations within streams (dewatering and cleaning of cofferdams will be performed to prevent siltation by pumping from cofferdams to a settling basin or a containment unit); (vi) discharge of sediment-laden construction water (e.g., from areas containing dredged spoil) directly into surface watercourses will be forbidden (sediment laden construction water will be discharged into settling lagoons or tanks prior to final discharge); and (vii) drainage systems will be periodically cleared to ensure adequate storm-water flow.

46. Water for construction activities is available from three rivers: the Choloki (km 3.80), Kintrishi (km 15.00), and Chaqvistskali (km 26.10). Water quality of these rivers is suitable for drinking and construction. Quarterly water quality monitoring will be undertaken during construction at each major bridge site to assess the impact of bridge construction on water quality and implementation of necessary mitigation measures.

47. Air Quality. During construction, air quality is likely to be degraded by exhaust emissions from the operation of construction machinery; fugitive emissions from asphalt plants; and dust generated from haul roads, unpaved roads, exposed soil, and material stock piles. The dust will settle on trees and crops, and may cause some degree of respiratory stress for nearby residents. To mitigate these, the following will be implemented: (i) Construction equipment will be maintained to a good standard and idling of engines will be discouraged. Machinery causing excessive pollution (e.g., visible smoke) will be banned from construction sites. (ii) The contractor will submit a dust suppression program prior to construction. The plan will detail

Engconsult Ltd.

Appendix 20 149 actions to be taken to minimize dust generation (e.g., spraying of roads with water, vegetation cover in borrow sites), and will identify equipment to be used. (iii) Construction materials will be stored away from residential areas and be properly covered. (iv) Asphalt plants will be located a minimum of 500 m away from settlements, according to government Decree # 234n (Ministry of Health and Social Welfare of Georgia, 6 Oct 2003), to avoid direct impact of emissions on local settlements. This decree also specifies that the concrete production plants are to be located more than 300 m and borrow sites more than 100 m from settlements. The construction and road machinery used during construction will comply with national environmental requirements for emission and noise pollution.

48. The project road is planned in a completely new alignment with only background concentration. During road operation, emission of particulates matter, hydrocarbon, carbon monoxide, nitrogen oxides, and sulfur dioxide from diesel and gasoline engines on the highway will increase. The air quality in tunnel sections will deteriorate. However, all tunnels will have limited access and prevent the entrance of pedestrians. The annual emissions along the project roads are estimated for 2010, 2014, 2025, and 2033 for both with- and without-project scenarios. Pollutant concentration along the bypass roads will be localized and not pose any threat to the population. Project implementation should reduce emissions along the existing road considerably.

49. Noise and Vibration. During construction, the main potential impact from noise on residential areas will be temporary: noise will emanate from maintenance workshops and the operation of earthmoving equipment. Operation of heavy machinery and rock-blasting activities can generate significant noise and vibrations. Health centers, schools, and prayer halls are sensitive receptors; and residential areas, in general, are the major receptors in the project corridor. Strong vibrations resulting from compaction equipment and rock blasting can damage nearby houses and other structures. To prevent noise and vibration, these operations will be restricted between 0600 to 2100 hours within 150 m of settlements and 500 m from sensitive receptors (hospitals and schools).

50. During operation, noise along the project roads will increase due to higher traffic volume. Noise along the project road is predicted using the Federal Highway Administration Traffic Noise Model (FHWA TNM 2.5). Noise quality data exceeds standards in the flat terrain. To mitigate noise pollution in residential areas, 341 m of noise barriers are incorporated in the project design from km 8 to 22.

51. Quarry Sites. Existing quarry sites in Chaqvistskali and Choroki rivers will be used to procure construction material. Generally quarry sites are major sources of environmental impact due to dust and noise pollution, loss of biodiversity, and generation of spills. The operating procedure for borrow pits will consider the following principles: (i) maximize the amount of fill that can be effectively used from the pit, (ii) minimize erosion and sedimentation, (iii) preserve water quality of the rivers, (iv) protect air quality during excavation, (v) prevent wildlife from falling into the pit, and (vi) reinstate the site after construction.

52. The mitigation plan to be followed by the contractor at the borrow sites includes (i) use only existing borrow areas approved by the environmental authority for the Project; (ii) ensure pit management (including restoration if it follows the completion of certain works) fully complies with all applicable environmental standards and specifications; (iii) conduct excavation and restoration of borrow areas and their surroundings in an environmentally sound manner to the satisfaction of MOEPNR and the Roads Department; and (iv) grade borrow pit areas to ensure drainage and visual uniformity or to create permanent tanks and/or dams. Additional borrow

Engconsult Ltd.

Appendix 20 150 pits, if necessary, will not be opened before restoration of areas no longer in use, and without the approval of MOEPNR. Topsoil from the opening of borrow pits will be saved and reused to revegetate the pits to the satisfaction of MOEPNR. At the gravel extraction sites (quarries), the proposed mitigation plan is to (i) use resources already being exploited to prevent proliferation of extraction sites, and make control and reinstatement more manageable; (ii) if new gravel extraction sites must be opened, conduct investigations to identify possible fossil deposits at a distance from active river beds; (iii) when extraction is approved from gravel bars along existing riverbanks on the inside margins of meander curves, do not remove gravel from within 2 m of the upper water level at the time of extraction (to protect the currently active river channel); and (iv) when extraction is in areas with less sensitive, shallower river flows, consider permitting removal of gravel to the level of the existing riverbed. The existing valley grade will be maintained and the operating area should be protected by a low 1–2 m gravel bank.

53. Flora and Fauna. Construction and operation of the Project will have only a minimal effect on existing flora and fauna. Flora degradation is expected to occur marginally as a result of road construction. The work will directly cause minor degradation of the local ecology through the clearance of areas of vegetation (ground cover) at major work sites and ancillary sites. During construction, a short-term impact on ecology along the road is likely to occur in and around the quarry sites, material stockpiling areas, and worksites due to vegetation clearance. A permanent but relatively minor impact on ecology is likely to occur due to the alignment of any unstable section. Vegetative cover stripped from these locations will be kept for landslide and slope protection. Contractors will be responsible for putting new vegetation in removal sites. Construction vehicles should use temporary roads constructed to minimize damage to agricultural land and local access roads. Where local roads are used, they will be repaired to their original condition after the completion of work. Compaction around trees will be performed carefully to avoid damage to the tree drip-line.

54. Potential impacts from construction worker camps include poaching of edible animals and birds in the locality, despite prohibitions. The contractors’ environmental health and safety manager will be responsible for providing adequate knowledge to workers regarding the protection of fauna. Workers will be trained regarding nature protection and the need to avoid cutting down trees during construction. Contractors will be responsible for supplying appropriate fuel in the work camps to prevent fuelwood collection.

55. Fish, Fisheries, and Aquatic Biology. The main potential impacts on aquatic flora and fauna in the watercourses are increased suspended solids from earthworks erosion, bridge construction and quarry site operations, sanitary discharge from work camps, and hydrocarbon spills. The management plan proposed for impacts on water resources and quarry sites will also mitigate this problem.

56. Protected Areas and Sensitive Environmental Receptors. The only protected area located in the project area is Ispani Mire (para. 20). The consultant had discussions with the management authority about the boundaries of the nature reserve and the impact of the project road on the protected area. Based on the consultations, the management authority revised the boundaries of the protected areas to include buffer zones to avoid any development in the buffer zone. The project road does not pass through the buffer zone and will not have any impact on the protected area. During construction, measures will be taken to restrict encroachment into protected areas, such as (i) no construction camp will be located from km 6 to 14 in the protected area section of the road; (ii) asphalt and mixing plants will be set up outside the protected area section and transport the material to overlay the pavement in the protected section; (iii) solid waste will be strictly managed in the nature reserve, (iv) a roadside fence will

Engconsult Ltd.

Appendix 20 151 be built for the protected area section; and (v) road signs, an awareness display board, and bylaws will be developed to explain and deal with violations and encroachments.

57. Social Impacts. Road improvement under the Project requires considerable land acquisition and resettlement. A resettlement and compensation framework was prepared for the Project. Campsites for construction workers have significant impacts, such as health and safety hazards on local resources and infrastructure of nearby communities. This may lead to antagonism between residents and workers. To prevent such problems, the contractor will provide temporary worksite facilities such as health care, eating space, and praying places. In addition, a mechanism will be established to allow local residents to raise grievances arising from the construction process. The use of local labor during construction will increase benefits to the local community and resolve such conflicts. The construction sites may impact public health through the spread of malaria, HIV/AIDs, and other sexually transmitted diseases. Mitigation measures include creating a good environment at work sites and educating construction workers. Awareness campaigns for the education of road users on the importance of the culturally sensitive sites normally help with their preservation. Improved road signage and awareness display boards will be placed at sensitive religious sites: the cemetery and churches located along the project roads. These will alert construction workers and road users. During construction, extreme precautions will be taken to ensure that cemetery and sensitive religious places are protected and respected. Contractors should consult with local residents during civil works close to these sites.

VI. ECONOMIC ASSESSMENT

58. Environmental mitigation and monitoring costs for the project area estimated at $6,364,875. The main environmental mitigation costs will be associated with activities, such as landslides, soil erosion, noise, rockfalls, stripping of topsoil, and plantation and ecosystem disturbance. The budget includes environmental management and monitoring during construction and operation. Capacity-building programs are proposed to strengthen the environmental monitoring capabilities of the implementing agencies. The environmental costs will be included in the economic analysis to estimate the economic internal rate of return. The economic assessment of the Project is under way.

59. The economic benefits of the Project are expected to include

(i) reduced traffic loads on existing roads that lack sufficient capacity (e.g., traffic jams, noise, low speed, higher emissions, accidents); (ii) decreased vehicle operating cost and travel time costs due to improved road facilities, and shorter route between Poti and Sarpi; (iii) improved tourist infrastructure and facilities; (iv) improved air and noise quality in Kobuleti and Batumi as a result of fewer traffic jams due to construction of bypass roads; (v) reduced soil erosion and landslides due to slope stabilization measures; (vi) improved commercial activity in the project area resulting in economic uplift of the people of the project area; and (vii) improved access of rural residents to the markets in Kobuleti and Batumi

VII. ENVIRONMENTAL MANAGEMENT PLAN

60. An Environmental Management Plan and Environmental Monitoiring Plan were prepared to address all the anticipated impacts of the Project and are presented in Annexes A and B of

Engconsult Ltd.

Appendix 20 152 Environmental Impact Assesment Report. Locations of the impacts, mitigation measures, costs, institute responsible, and monitoring measures are listed. The following institutes are responsible for implementing and monitoring the mitigation measures:

(i) The Ministry of Regional Development and Infrastructure is responsible for planning, constructing, operating, and maintaining regional, national, and provincial infrastructure in Georgia; the Roads Department is responsible for overall management of roads. The Project Management Unit of the Roads Department will be in charge of project management to ensure that the contract provisions are properly maintained. (ii) The Roads Department and its regional authorities will undertake routine and random monitoring of specific EMPs addressed in this EIA. (iii) The supervision consultants under the Roads Department are responsible for environmental monitoring and management of project implementation, and for helping ensure implementation of environmental management practices at each stage of construction. (iv) MOEPNR will be consulted if complicated issues arise during construction and operation stages.

61. Implementation of mitigation measures presented in the EMP and monitoring plan during construction will be the responsibility of the contractor. The representative of the Roads Department and environmental specialists of the supervision consultants will supervise the monitoring of implementation of mitigation and monitoring measures during construction. The national environment specialist will coordinate with the international environment specialist to resolve complicated issues that arise in the field and provide continuously updated information for reports to be submitted to the Roads Department and ADB.

62. The Roads Department has limited experience, resources, and equipment for environmental, social, and resettlement management and monitoring. It will have difficulty efficiently supervising monitoring of the environmental and social safeguard parameters. National and regional officials of the Roads Department require environmental and social management training to increase their understanding of road-related environmental issues, implementation of mitigation measures, and subsequent monitoring and capacity building. Training for the officials is crucial for proper environmental, social, and resettlement monitoring addressed in the EIA. Environmental, social, and resettlement training will be provided to Roads Department officials (a small unit comprising one environment specialist and one land asset assessor exists in the Roads Department). In addition, the supervision consultants in association with the contractor’s environmental, health, and safety manager will provided hands- on training for the contractor at the construction site. The objectives of this training program are (i) to help build the capacity and procedures of the Roads Department to undertake analyses of environmental, social, and resettlement impacts of road construction projects including incorporation of environmental, social, and resettlement factors in engineering design and preparation of environmental, social, and resettlement impact management plans in accordance with Government regulations and development partner guidelines; and (ii) to provide training on environmental, social, and resettlement management to the Roads Department and regional road maintenance offices.

63. A budget of approximately $6,364,875 is allocated for implementing the environmental monitoring program and mitigation measures provided in the EIA. The proposed environmental mitigation cost is $4,495,275; it is included as a civil works package. The environmental management budget under the Roads Department during construction and operation is

Engconsult Ltd.

Appendix 20 153 $832,000. In addition, a budget of $119,800 is provided for a road safety campaign, procurement of office equipment for the environmental and resettlement unit of the Roads Department, and environmental and social management training. The total budget is divided into $4,508,000 for the first tranche and $1,856,875 for the second tranche. The Roads Department will retain a recognized organization for environmental monitoring and ensure that the road is monitored regularly for the first 3 years of its operation.

64. During project implementation, the supervision consultants in cooperation with the Roads Department and MOEPNR will be required to (i) develop an environmental auditing protocol for use during construction, and formulate a detailed monitoring and management plan; and (ii) regularly supervise the environmental monitoring, and submit quarterly reports based on the monitoring data and laboratory analysis report.

The contractor will be responsible for subcontracting data collection of environmental monitoring to a recognized organization. The cost for this monitoring is included in the environmental mitigation budget.

65. The Roads Department will submit the following environmental reporting documentation to ADB: (i) Baseline monitoring report. The report, to be submitted to ADB prior to commencement of civil works, will include a detailed environmental management and monitoring plan (including data collection locations, parameters, and frequency), baseline environmental data, relevant standards, and data collection responsibilities. (ii) Environmental monitoring reports. The reports will include environmental mitigation measures and monitoring activities undertaken, details of monitoring data collected, analysis of monitoring results, recommended mitigation measures, environmental training conducted, and environmental regulatory violations. The environmental monitoring reports will be submitted to ADB quarterly during construction and quarterly for 3 years after completion of construction. (iii) Project completion environmental monitoring report. Three years after completion of construction, the Roads Department will submit a project completion environmental monitoring report to ADB summarizing the overall environmental impacts of the Project.

VIII. PUBLIC INVOLVEMENT AND DISCLOSURE

66. The EIA process included stakeholder and public participation and consultation to help the Roads Department achieve public acceptance of the Project. Two types of consultations were conducted: (i) stakeholder consultation and (ii) public consultation. Stakeholder consultations are performed mainly with selected individuals and organizations with professional national and provincial expertise in the EIA process. The stakeholder consultations were held in Tbilisi and Batumi in March and April 2009. The public consultations are conducted with various stakeholders including representatives of the affected community. The consultation involved a wide range of participants representing affected people, community leaders, civil society, nongovernment organizations, environment professionals, and central and regional government officials. The first public consultation was held in Batumi on 28 April 2009; the second is scheduled for 26 June 2009. The results of the first consultation were positive; participants

Engconsult Ltd.

Appendix 20 154 considered that the bypass roads would bring significant economic benefits to the region and recommended taking cautious measures to preserve the protected nature reserve.

65. Stakeholder consultations conducted in Tbilisi and Batumi at the early stage of the PPTA involved professionals and government officials responsible for reviewing the EIA and taking decisions on construction and environmental impact permits. The objective was to share information about the Project and obtain feedback on important environmental components that would be considered in the EIA process. During the consultations, the consultant presented the scope of the EIA study, discussed potential environmental and social impacts, collected available secondary data and information on environmental parameters, identified the parameters for a baseline environmental monitoring survey, and discussed the regulatory requirements for environmental permits and public consultations.

66. The public consultation organized in Batumi on 28 April 2009 involved about 50 people mostly from project-affected areas. Key representatives of the affected communities discussed the project issues with the relevant specialists and government authorities, expressed their concerns for the environmental and social impacts in their community associated with the bypass roads, and recommended mitigation measures. The specialists obtained a better understanding of the environmental concerns of these communities and established contacts for collecting information relevant to the Project during implementation. The meetings were planned as open meetings and key informal representatives of the communities, nongovernment organizations, and provincial and rayon (district) officials were invited to attend.

67. Recommendations from the consultations are incorporated in the engineering design of the Project (e.g., construction of retaining structures and drainage structures, avoiding construction close to the Kobuleti nature reserve, and development of a proper compensation plan for the affected community). No major concerns were raised during these consultations except the issues for proper drainage, construction quality, proper engineering practices during construction, and transparency in construction work. The first public consultation was widely covered by the electronic and print media. Staff of the Ajara television station and newspaper interviewed the chairman of the Ajara Autonomous Region, Roads Department staff, and the consultant. Project details were broadcast during the day and prime time news schedule.

68. The EIA, in both English and Georgian, documenting the mitigation measures and consultation process will be made available for public review. The SEIA will be published on the ADB website and the full EIA will be available upon request from ADB as well as from the Roads Department. The affected residents and local communities expressed support for the Project during the consultations as they clearly see the benefits to the community as well as the region. Consultations and public disclosure of information should continue during project implementation through (i) preparing and disseminating a brochure in Georgian explaining the entitlements of those affected; and the procedures for obtaining compensation for temporary disturbances, trees, crops, and land for construction sites, and recording grievances; and (ii) setting up a formal grievance redress committee with representation from the affected people. The supervision consultant in association with the contractor will be responsible for managing the effective grievance redress program.

IX. CONCLUSIONS

69. The EIA identifies negative and positive impacts due to construction activities and normal operations of the bypass roads. Recommendations are made to mitigate expected

Engconsult Ltd.

Appendix 20 155 negative environmental impacts with adequate funds provided to cover environmental monitoring and mitigation cost.

70. The major positive impacts of the Project will be less air pollution and dust, less congestion, improved traffic safety along the existing road, and better accessibility. Additional positive impacts are increased economic growth in the region, substantial income and employment opportunities, improved living conditions, reduced poverty, and better access to village produce.

71. Potential negative impacts are road safety, landslides, soil erosion, rockfalls, increased air emission and noise, dumping of construction spoils into the rivers and unauthorized sites, impacts on quarries and borrow sites, impacts on the habitats of the nature reserve, and significant land acquisition.

72. Road safety will be improved by stabilizing unstable batters and installing road safety barriers including proper traffic engineering signs and display boards. Soil erosion and rockfalls will be minimized after project completion. Landslides, critical side slopes, and unstable batters will be protected by retaining structures and rockfall fences. The contractor will prepare a spoil management plan and submit it to the Roads Department for approval. Quarries and borrow materials will be collected from the preapproved sites and be properly restored after the extraction of materials. The bypass road has been realigned to a safe distance from the buffer zone to avoid any potential disturbances to the habitats of the nature reserve. In addition, strict measures are recommended in the EMP to restrict encroachment on the nature reserve. A land acquisition and resettlement framework is being developed to compensate the affected population.

73. The environmental management and monitoring program developed for both temporary and permanent works covers preconstruction, construction, and operation stages with an estimated cost of $6,364,875. Implementation of an appropriate EMP and mitigation measures during various phases will minimize the negative impacts of the Project to acceptable levels. To ensure that these plans and mitigation measures are implemented and negative impacts avoided, the EMP will be included in the contract documents of the Project with a separate line item for environmental management in the bills of quantities. The residual impact on the environment will be minor and in some cases positive where erosion has been active along the existing tunnel portal and henceforth will be controlled. The cost of the retaining structure and other protective measures for slope stability is included in the engineering design.

74. Environmental and social management training of Roads Department officials at different stages in the project cycle is required to enable them to carry out environmental, social, and resettlement monitoring and implementation of the EMP. Environmental, social, and resettlement training will require the input of 2 person-months of international and 4 person- months of national specialists with an estimated cost of about $69,800.

75. The Project will have overall beneficial impact as well as some negative impacts that will be carefully monitored and adequately mitigated. Therefore, the completion of this EIA fully meets ADB requirements and no further environmental study is required for this Project. However, an updated EIA and EMP will be produced based on the detailed engineering design, and cover the impacts of temporary work activities during the detailed design stage in early 2010. An environmental impact permit must be obtained at that time to comply with MOENPR guidelines.

Engconsult Ltd.

Appendix 21 156

Appendix 21 CUMULATIVE IMPACT ASSESSMENT

A. Introduction 1. In the context of this project, Cumulative Impact Assessment (CIA) or Cumulative Effects Assessment (CEA) involves the examination of “the impact on the environment which results from the incremental effects of action when added to other past, present, and reasonably foreseeable future actions [taking place in a defined area] stemming from projects, regardless of what agency or person undertakes such other actions” (US National Council on Environmental Quality 1997). B. Sources of Cumulative Impacts 2. The sources of cumulative impacts are implementation of ongoing or future development projects that will affect the physical, ecological and social resources in the Project areas, Various ongoing and proposed development activities near the project areas are obtained from the ‘regional economic development master plan of Adjara’ for 2007 to 2011. The major planned projects according to this master plan are: i. Improving tourism infrastructure (i) Rehabilitation and construction of water supply and sewerage system (ii) Construction of tourist objects (iii) Care and importance of historic and cultural monuments ii. Coastal strip protection works (i) Coast protection works of River Cholokhi (ii) Protecting Kobuleti coastal strip iii. Repair of Roads (i) Rehabilitation of retaining walls, gabions, bridges and forest roads iv. In addition to the above projects, specific projects targeted near Batumi are (i) Tourism development - entertainment managements, organizing sporting events, new entertainment clubs (ii) Cleaning and developing beaches (iii) Arranging sewerage and gutter systems (iv) Illuminating the city v. Projects near Kobuleti are (i) Setting up of agricultural processing industry equipped with advanced technology (ii) Rehabilitation of roads and sewerage 3. The Project 1 of MFF is a part of the road upgrading of Poti to Sarpi, a 81 km international road and is one of the sections of the Transport Corridor Europe-Caucasus-Asia (TRACECA). The rehabilitation of 35-kilometre long road section from Poti to the river Choloki (the start of Project 1) has been carried out in the last five years, the construction of Poti bypass has been completed and new bridges have been constructed across the Kaparcha and the Choloki Rivers. The section of road from Choloki Bridge to Sarpi is the subject of the Project 1. Project 1 will construct bypass roads to Kobuleti and Batumi and widen 6 km of the existing road. C. Assessment of Cumulative Impacts 4. According to World Bank’s Roads and Environment Handbook, CIA is founded on identifying broad and overarching effects that are generated by many actions (projects) affecting specific components of the biophysical and/or social/human environment, referred to as Valued Ecosystem Components (VECs).The cumulative effects of the Project 1 are evaluated by (i) compiling a list of activities that are part of the proposed project; (ii) defining the changes

Engconsult Ltd. Appendix 21 157

(impacts) that will occur using indicators that are known to be sensitive to the project activities; and (iii) estimating the effects that the project actions will have on in each of the VEC; within a pre-defined temporal and spatial boundary. 5. The methodology followed in CIA is: identification of possible potential VECs through an evaluation of the threat to each VEC was made based on the technical knowledge of the Consultant and an assessment of six impact attributes of each potential effect on each VEC. These attributes were impact direction (negative, positive or neutral), geographic extent, duration, frequency, magnitude and significance. Each VEC is assessed for three scenarios: (i) existing conditions; (ii) future conditions without the Project; and (iii) future conditions with the Project. While assessing future conditions with the Project, construction impacts during implementation of the Project were not considered as they were addressed in EIA and an environmental management and mitigation plan was already developed to mitigate those impacts. 1. VEC 1: Air Quality and Dust 6. Existing Condition: Air quality along the existing highway exceeds the national standards. Major sources of pollution are vehicles. Available data with Ministry of Environmental Protection and Natural Resources indicates that mobile sources are responsible for an increasing share to total emissions from about 70% in 1991 to 91% in 2001.1 Vehicles contribute 31% of dust, 37% of SO2, 98% of CO, and 82% of NH4. Transport is also contributing an increasing share of the CO2 from 11% to 35% in 1990-1997. The reasons for high vehicle pollution are: majority of the vehicles in Georgia today are 15 to 20 years old representing poor condition of the vehicles, poor quality of the fuels (only few laboratories exist to test the fuel quality with insufficient equipment, and the analysis is according to 1970s Soviet standards and), and poor traffic management. Georgia phased out leaded gasoline in 2000. The maximum allowable sulfur content in diesel and petrol are 350ppm and 500ppm in imported fuel. Gradual move to 50 ppm in petrol by 2011 and 50 ppm in diesel by 2010 has been planned. Georgia is planning to introduce new sulfur standard for gasoline ‐ 250ppm from 2009 and 150ppm from 2010.2 7. Future Condition without the Project: One would hope that with improved pollution control devices on vehicles, better refined fuels, reduced sulphur content in diesel fuel and legal requirements for catalytic converters, the air quality along the road corridor will as a minimum stay the same, but hopefully improve. Confounding this conclusion however will be the issue of vehicle congestion due the lack of improved roads. Future conditions without bypass or upgraded roads will mean that the roads which are already near design capacity will experience increased traffic congestion, and air pollution will rise due the excessive acceleration and deceleration cycles. The development of any new industries requiring road transport, the use of cars and trucks could quickly cause sections of the road corridor to exceed the design limits, requiring the use of more fossil fuels, thereby exhausting more pollutants into the air. 8. Future Condition with the Project: The air quality in the Kobuleti and Batumi towns will be improved due to diversion of traffic to the newly constructed bypass roads thus minimizing traffic slowdowns and moving traffic around congested urban areas. No new heavy industries are expected in this areas that cause severe air pollution. 2. VEC 2: Noise Environment 9. Existing Condition: Estimated noise levels within 25m along the existing roads are nearly 70 dB thus have already exceeded the national standard of 60dB. Noise above 60dBA at

1 http://www.thepep.org/en/workplan/urban/documents/GeorgiaEnvt.pdf 2 http://www.unep.org/pcfv/PDF/Matrix%20CEE_FuelsFeb_2009.pdf

Engconsult Ltd. Appendix 21 158 night is annoying and will result in sleep deprivation among residents, even 100m away from the roadside. The background noise levels in the undeveloped and undisturbed areas of the project areas are 30 dB. 10. Future Condition without the Project: The noise levels will continue to increase due to increased traffic. This may require landuse controls to keep sensitive activities such as schools, recreation areas, etc. out of these corridors. There will another possible scenario with a do- nothing option - the average speed of the traffic will slow down, thereby reducing the overall noise environment, unless of course there will be a large increase in the use of horns, from drivers frustrated at the slow pace. 11. Future Condition with the Project: Assuming that vehicle operating patterns remain more or less the same, e.g. the use of car horns is at a low-level, improved roads leading to higher traffic speeds will have a negative noise impact, since higher average speed generally means greater noise levels. This however will be mitigated due to the greatly improved road surface (roughness), this being an important contributor to traffic noise; and the steadily improving vehicle fleet condition and age. Secondly new roads will insure that grades are low enough to permit vehicles and trucks to move without having to use lower and noisier gears, further mitigating operating period noise. Further the new roads will include noise barriers wherever the predicted noise levels exceeds the standards. 3. VEC 3: Stabile and Roadside Communities 12. Existing Condition: The existing roadside communities from one half to three km from the edge of pavement are only marginally negatively affected by the existing road corridor; mostly in relation to occasional road safety, noise and air pollution issues, and accidents which require use of local emergency services. In fact the road provides a vital link to outside services and markets by facilitating travel, the transport of goods and provision of services. The economies of these communities have come to depend on the roads which generate the demand for services, food and fuel by car, truck and bus operators/users. The influx of temporary construction jobs should only serve to maintain and enhance a community’s prosperity. 13. Future Condition without the Project: In the absence of major developments along the roads, the change in the future will be a steady growth in traffic, generally maintaining a balance between the supply of traffic and the demand for services. In terms of physical safety, the do-nothing scenario will mean more congested slower moving traffic, therefore fewer really serious accidents, but likely more minor accidents. The congestion could lead to health problems, as air pollution grows with the frequency of traffic slow-downs. 14. Future Condition with the Project: Road construction through new alignment requires considerable land acquisition and resettlement of affected people. Highway related business along the existing road will be affected due to diversion of traffic to the bypass roads. The projects will lead to better quality roads and shorter travel times to and from markets and services, and improve the economy of the area. Tourism, the major source of economy in the project area will improved due to improvement of quality of life in the decongested cities. 4. VEC 4: Landscape Integrity 15. Existing Condition: Landscape integrity refers to the maintenance of the visual field on either side of the road free of unsightly hill cuts and piles of stored excess excavation materials. At present, at some of the hill sections, especially near Makhinjauri, there are unprotected side slopes picturing scar in the landscape view. Further, most of the hill sections near Batumi are highly susceptible to soil erosion, landslides and mudslides due to steep slopes, poor geological conditions, higher sand content of surface soils and active geodynamic process. Generally the

Engconsult Ltd. Appendix 21 159 road in these hilly sections generally provides many breathtaking views and represents an important tourist attraction. 16. Future Condition without the Project: Unless major new road construction takes place, there will be no improvement in landslide protection and control of soil erosion. The government with its limited budget cannot take up any major activities unless there will some donor funding. There is a possibility of uncontrolled in-migration along the roadsides, particularly near the towns providing obstructions to picturesque vistas. 17. Future Condition with the Project: The Project will establish new and attractive roadside corridors, ripe for settlement. Landuse planning and control is required to develop the areas near road sides. Landuse zoning enforcement permits the development of settlements and businesses along the road, degrading roadside areas likely eliminating any existing tree stands, polluting local water courses, creating congestion, as well as endangering the health of these ‘settlers’. 5. VEC 5: Vegetation and Trees 18. Existing Condition: Given that the project area is in the arable lands, there are no natural forest lands along any of the road corridors, but instead, contains natural vegetation for long stretches very old Pines and Eucalyptus. 19. Future Condition without the Project: Future losses of forested areas may continue as trees are cleared for residential and tourism development and the roadside commerce grows as the country’s population increases. Government’s ongoing tree replanting program and development of plantation forests will improve the vegetation in the project areas . 20. Future Condition with the Project: No road-related operating period impacts are predicted. In fact trees will be planted along the new road sections as well as along the upgraded roads. 6. VEC 6: Kolheti Wetlands and its Flyways 21. Existing Condition: Kolheti wetlands, along the Poti – Saripi Highway, lies along one of the main routes of migration of water-fowl and waders of Africa and Eurasia. Over 194 different bird species are found within the wetland regions including 21 species of migratory birds. Large migrating flocks can be seen in the fields beside the road from spring through early summer and again in the fall. While the main known resting and staging areas are further to the North and West of the roads, some flocks do land in the areas close to the road to feed and rest along the small wetlands and in the cultivated fields. These are not in danger of being filled or drained. 22. Future Condition without the Project: As the human population density along the corridor increases, due to new employment opportunities, the habitat will gradually deteriorate. Only conservation efforts by the national and provincial governments, protecting known resting sites, will prevent further loss of this habitat and in turn the bird populations. 23. Future Condition with the Project: The project roads will not pass through any wetlands and known bird resting areas, and hence no impact on these wetlands. 7. VEC 7: Surface Water Quality 24. Existing Condition: The water quality of the existing rivers is good except high turbidity levels in a few rivers due to soil erosion in the upstream areas. However several possible sources of pollution exist in the project areas, such as irrigation return flows, and sewerages from the municipal and industrial areas. 25. Future Condition without the Project: A future do-nothing scenario will likely have a negative impact on surface water quality since at present no storm water management exists

Engconsult Ltd. Appendix 21 160 near the project roads and as such when it rains runoff is directed straight into local streams. In the case of bridges some have actual holes in the deck permitting water to flow off the bridge into any receiving water. This permits all pollutants, including oils and metals in the rainwater and snow melt to discharge into receiving waters without being detained or diverted which would allow for percolation into surrounding soils. As traffic grows over the next decade this condition will worsen significantly. 26. Future Condition with the Project: Impact on surface water quality stemming from the roads would be minor and result primarily from improperly maintained stormwater management systems, which do not permit the storm runoff to be detained for a minimum of 12 hours before discharging into a receiving stream. D. Summary of Cumulative Impacts 1. Cumulative Impacts of Poti – Sarpi Highway 27. The cumulative impact of constructing bypass roads to Kobuleti and Batumi in the broader context of the Poti - Sarpi Highway Upgrading Program includes both positive and negative impacts. These positive impacts include: • substantial reductions in travel time - by 2013, a travel time saving of approximately 40 minutes had been achieved through the completion of Project 1 and this would result a cumulative time savings of about 80 minutes from Poti to Sarpi • improvements in road safety resulting from the diversion of traffic from cities (Poti, Kobuleti and Batumi) to bypass roads • consequential improvements in travel efficiency for local and regional businesses resulting from travel time savings • Increased opportunities for tourism development based on improved accessibility of coastal and nearby hilly resort areas. 28. Overall, the negative impacts of the Project would primarily affect the natural environment along the alignment, together with residents living close to the proposed alignment and businesses with a high reliance on highway-related trade. The design of the Project and the mitigation measures are intended to ameliorate the adverse environmental impacts on nearby residents and the natural environment. 2. Cumulative Impact with other projects or activities in the study area 29. The developments that could be expected in the Project area are mostly tourism and tourism related infrastructure. These proposed developments, though implemented in conjunction with the present project would hardly have a cumulative impact on the environment. 3. Cumulative Social Impacts 30. The Project would result in noticeable improvements in the amenity of the townships of Kobuleti and Batumi through the removal of large volumes of through traffic, including a substantial proportion of heavy traffic, especially at night. In particular, this would result in: • reduced levels of traffic noise, with associated reductions in sleep disturbance at night • reduced levels of vehicle emissions • improvements in local accessibility through the removal of traffic congestion • improvements in local road safety • opportunities to undertake urban design improvements that would not otherwise be possible with the presence of the existing highway in the centre of the township.

Engconsult Ltd. Appendix 21 161

31. When combined, these changes would noticeably improve the amenity of houses and community facilities fronting or near the existing highway. Residents located near the proposed alignment would be affected, particularly by noise. However, mitigation measures are integrated into Project design where the noise assessment has indicated that noise levels would exceed the standards and to ensure that the new road fits appropriately into the local environment. 32. A number of property owners would be affected by loss of all or part of their property to accommodate the Project. However the cumulative impact would not be such as to adversely affect the nature and supply of the housing stock in the study area or its overall demographic structure or agricultural production at a regional level. 4. Cumulative Impacts on Conservation Resources 33. The Project would result in the loss of some native vegetation and, as a result, fauna habitat. However, as discussed in EIA report, it is not considered likely that the Project would significantly reduce the biodiversity in the locality or region. The impact on conservation resources of the Project is acknowledged by the incorporation of extensive measures into the design of the road which, individually and collectively, would mitigate particular and cumulative impacts on native vegetation and fauna habitat. These measures, in particular the proposed fencing along nature reserves, would reduce the road kill of native fauna.. 5. Cumulative Economic Impacts 34. The extent of agricultural, forestry, fishing and other farming activities directly affected by the Project is minor in local and regional terms. Therefore, the Project should not cause any significant adjustments to these economic activities individually or collectively. The enhancement in road conditions, safety, efficiency and reduced travel times are likely: • to enhance access to the region for local tourism • to increase opportunities for commerce, industry and tourism between Kobuleti and Batumi, and other major centres • to attract further economic development to the cross-border trade • to increase the activity of the service industry • to improve fuel efficiency due to better horizontal and vertical road alignments • to increase accessibility to and within the region.

Engconsult Ltd. FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 1 Transport Trends and Economic Profile

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Contents

I. INTRODUCTION ------1 II. TRANSPORT TRENDS ------2 A. Vehicle ownership ------2 B. Demand for transport ------2 III. ROAD DEVELOPMENT IN GEORGIA ------4 A. Existing road network ------4 B. The bypass and the regional transport system ------5 IV. SOCIO-ECONOMIC FEATURES OF GEORGIA AND AJARA ------8 A. Georgia ------8 B. Ajara . ------9 1. Economy ------9 2. Access by train and bus ------10 C. Sarpi and Batumi trade ------11 D. Growth and inflation forecasts ------12

List of Tables

Table 1: Registered vehicles in Georgia, 2000-8 ...... 2 Table 2: Regional car ownership ...... 2 Table 3: Transport demand as tonne-km and passenger-km ...... 3 Table 4: Transport demand in tonnes and passengers ...... 3 Table 5: Road development in Georgia and Ajara, 2000-2008 ...... 4 Table 6: Rehabilitation and lengths of new road, 2004-9 ...... 5 Figure 1: Georgia’s trunk road network ...... 6 Table 7: Trunk road network ...... 6 Table 8: Population and economy of Georgia, 2000-2008 ...... 8 Table 9: Trade statistics – Georgia, 2004-2008 ...... 8 Table 10: Ajara visitors, 2005-8 ...... 10 Table 11: Passenger train services to Batumi ...... 10 Table 12: People crossing Georgia’s borders, 2007-8 ...... 11 Table 13: Transit vehicles and vehicles crossing Sarpi border, 2007-8 ...... 11 Table 14:Batumi and Poti ship calls, 2007-8 ...... 12 Table 15: Batumi rail freight traffic, 2004-8 ...... 12 Table 16: Georgia short term GDP growth forecasts ...... 13 Table 17: GDP growth forecasts ...... 13 Table 18: Inflation forecasts ...... 13

List of Figures

Figure 1: Georgia’s trunk road network ...... 6

Engconsult Ltd. 1

I. INTRODUCTION

1. This supplementary appendix is divided into several sections and is designed to give an overview of the transport system and an economic profile of Georgia and Ajara. 2. The purpose of this supplementary appendix is thus to: • summarize transport trends and the development of the road network • place the proposed bypass in the regional transport context • summarize Georgia’s and Ajara’s economies • estimate future GDP growth in order to support traffic forecasts

Engconsult Ltd. 2

II. TRANSPORT TRENDS

A. Vehicle ownership 3. Table 1 shows the growth in goods and passenger vehicles in Georgia from 2000 to 2008. The data quality is poor and there are discrepancies between data obtained from the Ministry of Internal Affairs, data from the MED reproduced in the JICA Highway Improvement Project report and data from the Transport Department. The reason for its poor quality is the administration of vehicle registration: from 2005/6 annual licensing of vehicles ceased and was replaced by a requirement to register only at a change of ownership. 4. The table below is based on MED data to 2006, followed by Ministry of Internal Affairs data for 2006-8. Table 1: Registered vehicles in Georgia, 2000-8 Year Car Truck Bus Other Total Percent change 2000 201,992 25,557 18,159 4,482 250,190 2001 225,992 29,260 22,703 4,993 282,948 13% 2002 250,358 32,545 26,487 5,479 314,869 11% 2003 276,225 36,909 29,867 6,127 349,128 11% 2004 306,211 42,580 32,460 6,752 388,003 11% 2005 342,047 48,912 35,101 7,552 433,612 12% 2006 387,866 58,237 38,549 8,891 493,543 14% 2007 426,864 49,588 39,169 8,699 524,320 6% 2008 501,248 58,363 41,684 12,002 613,297 17% Average annual change, 2000-2008 12% Source: MED and Ministry of Internal Affairs (from 2006) 5. Vehicle ownership per 1,000 people has increased from 56 in 2000 to 140 in 2008 (cars alone from 46 to 114) – faster than GDP per head (see section IV), as expected. Table 2 shows that Georgia’s car ownership is broadly consistent with that of its neighbors, bearing in mind the differences in GDP/head. Table 2: Regional car ownership Year Georgia Azerbaijan Russia Ukraine 2004 71 53 159 107 2005 79 57 GDP/head, 2005, PPP 3,570 4,648 11,861 5,583 $ GDP/head, 2005, nom $ 1,427 1,604 5,341 1,829 Sources: (i) Table 1data, (ii) EECCA CSI 29 - Passenger Transport Demand and (iii) 2005 Purchasing Power Parity International Comparison Program (World Bank et al, 2005).

B. Demand for transport 6. Tonne-km and passenger-km statistics, taken from the 2008 Statistical Yearbook for Georgia, are reproduced in Table 3. As with vehicle ownership statistics, their quality is likely to be poor – in this case because both intercity bus and freight markets are completely deregulated. Even if their quality is poor, several features stand out. Rail carries around 90 percent of goods measured by tonne-km, but only 45 percent as tonnes. This is consistent with rail’s role as the predominant carrier of bulk oil: the average trip length by rail is around 300km (approximately the distance between Tbilisi and Poti). Table 3 and Table 4 data suggest that by

Engconsult Ltd. 3 road the average trip is approximately 20km, but this appears extremely unrealistic. The apparent pattern with passenger traffic is similar: the average rail journey is 180km but by bus it is 20km (with the same caveat as for goods traffic). Table 3: Transport demand as tonne-km and passenger-km Year Million tonne-km Million passenger-km Rail Road Maritime Air Total Rail Bus Air Total 2000 3,912 475 612 2.9 5,002 453 4,500 238 5,197 2001 4,481 520 74 2.7 5,077 401 4,764 241 5,406 2002 5,075 543 148 2.9 5,769 401 4,920 297 5,618 2003 5,538 562 92 2.7 6,195 387 5,150 400 5,938 2004 4,862 570 70 3.5 5,505 615 5,200 489 6,304 2005 6,127 578 69 3.6 6,778 720 5,252 511 6,482 2006 7,393 586 131 3.9 8,114 809 5,269 453 6,531 2007 6,928 594 120 3.6 7,646 774 5,417 475 6,665 2008 6,516 601 45 1.9 7,164 675 5,568 486 6,729 Sources: (i) national statistics, Note: passenger-km data exclude journeys by urban transport (tram, trolley-bus and metro)

Table 4: Transport demand in tonnes and passengers Year Million tonnes Million passengers Rail Road Maritime Air Total Rail Bus Air Total 2000 11.5 18.5 0.063 0.0013 30.1 2.3 235.0 0.1 237.4 2001 13.2 20.0 0.031 0.0012 33.2 2.1 240.5 0.1 242.7 2002 15.0 22.5 0.036 0.0013 37.5 2.1 249.5 0.1 251.7 2003 16.6 24.5 0.021 0.0013 41.1 2.2 257.0 0.2 259.4 2004 15.4 25.7 0.024 0.0015 41.1 3.0 260.0 0.2 263.2 2005 19.0 27.0 0.024 0.0016 46.0 3.6 263.1 0.3 267.0 2006 22.6 27.3 0.040 0.0016 49.9 3.8 285.7 0.2 289.7 2007 22.2 27.6 0.038 0.0011 49.8 3.9 293.5 0.2 297.6 Sources: (i) national statistics, Note: passenger data exclude journeys by urban transport (tram, trolley-bus and metro)

Engconsult Ltd. 4

III. ROAD DEVELOPMENT IN GEORGIA

A. Existing road network 7. Table 5 shows the national and Ajaran network from 2000 to 2008. In terms of total length and surface type there has been very little change over this period. The network of international roads, i.e. those used by international transit traffic, has barely changed and similarly the interstate network did not change until 2007. In 2007 the Roads Department lost responsibility for local roads but 2,000km of what had hitherto been local roads were re- designated as interstate roads. 55km of the total network has four lanes. 8. In Ajara the 48km shown as international road is the existing Kobuleti-Batumi-Sarpi road and the 2km reduction in 2008 is the result of the realignment north of Makhinjauri. 9. Roads Department carries out an annual condition survey. The proportion of international roads that are classified as “good” (defined as IRI<4m/km) rose from 45 percent in 2004 to 66 percent in 2008. In the case of interstate roads the increase has been less impressive: from 18 to 23 percent. 10. Road network improvement is the result of investment in rehabilitation and periodic maintenance rather than network extension. Lengths of road rehabilitated and new roads are shown in Table 6. Table 5: Road development in Georgia and Ajara, 2000-2008

Inter- Interstat Inter- Interstat Local Total Local Sub-total national e national e Georgia 2000 Ajara 2000 Paved 1,414 2,328 4,411 8,153 48 140 132 320 Gravel 60 867 9,968 10,895 7 736 743 Earth 131 1,183 1,314 264 264 Total 1,474 3,326 15,562 20,362 48 147 1,132 1,327 Georgia 2001 Ajara 2001 Paved 1,414 2,328 4,284 8,026 48 140 132 320 Gravel 60 867 9,967 10,894 7 736 743 Earth 131 1,178 1,309 264 264 Total 1,474 3,326 15,429 20,229 48 147 1,132 1,327 Georgia 2002 Ajara 2002 Paved 1,414 2,529 4,284 8,287 48 140 132 320 Gravel 60 655 9,967 10,682 7 736 743 Earth 82 1,178 1,260 264 264 Total 1,474 3,326 15,429 20,229 48 147 1,132 1,327 Georgia 2003 Ajara 2003 Paved 1,414 2,459 4,101 7,974 48 140 132 320 Gravel 60 743 10,223 10,223 7 736 743 Earth 124 1,123 1,123 264 264 Total 1,474 3,326 15,447 20,247 48 147 1,132 1,327 Georgia 2004 Ajara 2004 Paved 1,414 2,358 4,039 7,811 48 140 132 320 Gravel 60 911 10,264 11,235 7 736 743 Earth 57 1,144 1,201 264 264 Total 1,474 3,326 15,447 20,247 48 147 1,132 1,327

Engconsult Ltd. 5

Inter- Interstat Inter- Interstat Local Total Local Sub-total national e national e Georgia 2005 Ajara 2005 Paved 1,432 2,395 4,027 7,854 48 140 112 300 Gravel 63 902 10,304 11,269 7 756 763 Earth 57 1,149 1,206 249 249 Total 1,495 3,354 15,480 20,329 48 147 1,117 1,312 Georgia 2006 Ajara 2006 Paved 1,432 2,395 4,027 7,854 48 140 112 300 Gravel 63 902 10,304 11,269 7 756 763 Earth 57 1,149 1,206 249 249 Total 1,495 3,354 15,480 20,329 48 147 1,117 1,312 Georgia 2007 Ajara 2007 Paved 1,442 3,379 4,821 48 120 168 Gravel 55 1,807 1,862 144 144 Earth 260 260 48 48 Total 1,497 5,446 6,943 48 312 360 Georgia 2008 Ajara 2008 Paved 1,440 3,326 4,766 46 138 184 Gravel 55 1,823 1,878 141 141 Earth 297 297 22 22 Total 1,495 5,446 6,941 46 301 347 Source: Roads Department Note: (i) local roads became the responsibility of local government from the end of 2006 (ii) Ajara roads are included in the totals for Georgia (iii) Georgia includes Abkhazia and South Ossetia (iv) urban roads (city responsibility) are excluded Table 6: Rehabilitation and lengths of new road, 2004-9 2004 2005 2006 2007 2008 2009 (planned) Km new road N/A 0.65 16.65 72.1 18.5 76.75 Km rehabilitateda 276 216 322 577 331 N/A Source: Roads Department Note: (a) includes periodic maintenance 11. The trunk road network is illustrated in Error! Reference source not found. and shown in Table 7. B. The bypass and the regional transport system 12. Georgia is a key transport link on the most direct link between the Black Sea and the Caspian. It is an important link for the transport of oil by rail and by pipeline from Azerbaijan to Supsa and Batumi. But while rail has the highest modal share when measured by tonne-km, road carries 55 percent of total tonnage. Passenger transport is dominated by bus traffic. 13. The East-West Highway (road number S-1, E60 in the European network) is the principal artery for long distance road traffic, being used by most traffic from Azerbaijan and Armenia to the Black Sea ports and Turkey as well of course as long distance traffic to and from Tbilisi. It carries 6-9,000 vehicles/day on its west-central section from Gori to Samtredia. At Samtredia the traffic splits, with roughly half using the Senaki-Poti link (S-2) and half the link to Kobuleti. The existing S-2 from Kobuleti to Sarpi carries an average of 7,000 vehicles/day. The proportion of heavy vehicles is high: 11-19 percent around Gori and Khashuri, 10 percent

Engconsult Ltd. 6 between Batumi and Sarpi. Traffic in the S-1 and S-2 corridors grew by 20 percent a year from 2004 to 2007 (but with the Russian trade embargo and global recession has grown little since).

Gantiadi

Russia

Larsi Zugdidi Roki

Kulevi Senaki Kutaisi Poti Khashuri Gori Black Sea Tbilisi Samtredia Bypass Batumi Akhaltsikhe Tbilisi Lagodekhi Red bridge Sarpi Partskhisi Trunk road network Vale Ninotsminda Railway Turkey Railway – under construction Guguti Kartsakhi Sadakhlo International airports Zhdanovi Azerbaijan Armenia

Figure 1: Georgia’s trunk road network Table 7: Trunk road network Road Numbera Length, km Tbilisi-Senaki-Zugdidi (Abkhazia border) S-1 (E60) 435 Mtskheta-Stepantsminda-Larsi (Georgian Military S-3 (E117) 139 Highway) Gori-Ergneti (S Ossetia border) S-10 25 Tbilisi-Red Bridge (Tsiteli-Khidi) (Azerbaijan border) S-4 46 Tbilisi-Bakurtsikhe-Lagodekhi (Azerbaijan border) S-5 140 Marneuli-Sadakhlo (Armenian border) S-7 34 Ponichala-Marneuli-Guguti (Armenian border) S-6 94 Akhaltsikhe-Ninotsminda (Armenian border) S-11 110 Khashuri-Vale (Turkish border) S-8 103 Senaki-Poti-Sarpi (Turkish border) S-2 (E70) 105 Tbilisi bypass S-9 49 Samtredia-Grigoleti (beach area N of Kobuleti) S-12 57 Koda-Ninotsminda No number 189 Total 1,526 Source: Transport Department Note: (a) Georgian and, where applicable, European numbering system 14. At present levels of traffic the S-1 is close to capacity over its central sections and recent projects have sought to increase its capacity by means of bypasses, road widening and lane addition. 15. The S-2 is similarly close to capacity on its central section between Batumi and Kobuleti. It is extremely congested during the busy summer period. Sarpi is the busiest border crossing in Georgia (see section IV below) and Turkey is Georgia’s most important trading partner. Trade (measured as imports plus exports) has grown faster than GDP, averaging 32 percent a year (in nominal terms) from 2004 to 2008.

Engconsult Ltd. 7

16. The S-2 is the E70 in the European network; it runs from A Coruña on the Atlantic coast of NW Spain to Poti. Regionally, it follows the Black Sea coast west from Sarpi to the port of Samsun. West of Samsun it connects to the E95 and the E80. The E80 connects Bulgaria, Istanbul and Ankara to eastern Turkey and Iran – but not (see below) to Georgia, Armenia or Azerbaijan. 17. Thus, with the present pattern of trade and the available transport links in the central Caucasus region, the bypass has a clear role, that of reducing the costs of trade with Georgia’s major trading partners. It has other important benefits: it will reduce congestion and accidents in Batumi and Kobuleti. High standard road access to Batumi will also encourage further investment in and around Batumi. 18. The regional role outlined above may be affected by several developments: • improvement of road links between Georgia and Turkey further east, in particular the Samtskhe-Javakheti road project. The road itself will be complete by late 2010, but it is less clear that border facilities will be in operation at the same time. Once completely open, it will be part of a new and shorter route for traffic between Turkey and Tbilisi and Azerbaijan – although as a low altitude alternative the E70 and S-2 will continue to attract long distance traffic during the winter months • the re-opening of the Turkey-Armenia border, closed to both road and rail traffic since 1993. This development is not certain, as it depends on ratification of a “normalization package” by the two governments

Engconsult Ltd. 8

IV. SOCIO-ECONOMIC FEATURES OF GEORGIA AND AJARA

A. Georgia 19. Population and economic growth in Georgia are summarized in Table 8 and trade statistics in Table 9. Table 8: Population and economy of Georgia, 2000-2008 Year Pop Gross domestic product CPI x Current 2008 GDP/he Real USD/hea Agric Industry Service percent 1,000 GELm GELm ad 2008 GDP d PPP s growth 2000 4,435 6,015 9,856 2,222 21.6% 22.5% 56.0% 4.0% 2001 4,401 6,646 10,389 2,361 5.4% 22.1% 22.0% 55.9% 4.7% 2002 4,372 7,457 10,960 2,507 5.5% 20.6% 24.4% 55.0% 5.6% 2003 4,343 8,565 12,177 2,804 11.1% 20.6% 25.6% 53.8% 4.8% 2004 4,315 9,970 12,895 2,988 5.9% 17.8% 25.4% 56.8% 5.7% 2005 4,322 11,621 14,133 3,270 9.6% 3,570 17.0% 27.0% 56.0% 8.2% 2006 4,401 13,790 15,461 3,513 9.4% 4,100 13.0% 25.0% 62.0% 9.2% 2007 4,395 16,999 17,379 3,954 12.4% 4,770 11.0% 24.0% 65.0% 9.3% 2008 4,382 17,900 17,900 4,085 2.1% 4,851 10.0% Average annual growth rates: 2001- - 14.7% 7.8% 7.9% 8 0.1% 2004- 0.2% 16.0% 8.1% 7.9% 8 Sources:(i) Georgia Statistics Yearbook, (ii) IMF Georgia Country Report CR2006-171, (iii) World Bank data profile for Georgia, 2008, (iv) EIU country report for Georgia, March 2009 Notes: (i) 2008 GDP and CPI values are estimates (ii) PPP values are of gross national income (=GDP+net factor income from abroad) (iii) last census was in 2002 Table 9: Trade statistics – Georgia, 2004-2008 Year Import Exports Total Trade Trade Exports by trading Imports by trading Tourist sb b trade deficit as % partner USDm partner, USDm arrival USDm USDm USD GDPa Russia Turkey Turkme Russia Turkey Azer s x m n 1,000 2004 2,008 1,092 3,100 9% 31 105 119 113 256 202 156 368 2005 2,687 1,472 4,159 10% 36 154 122 76 383 283 233 560 2006 3,686 1,667 5,353 15% 39 76 123 72 559 523 319 980 2007 4,984 2,088 7,072 17% 42 46 172 26 577 728 382 1,051 2008 6,050 2,509 8,559 20% 48 Av ann growth rate 32% -14% +14% -34% 33% 55% 35% 45% 2004-8 Sources: Georgia Statistics Yearbook Notes: (a)(exports+imports)÷GDP (b) fob basis 20. Georgia’s centralized economy had collapsed by the mid-1990s: industrial output had shrunk to a sixth of its 1989 level and inflation was high. Since then the country has pursued policies that attracted inward investment, has reformed tax collection (Georgia has historically been very poor at collecting tax revenues) and has pursued fiscal policies that helped contain inflation. The World Bank categorizes Georgia as a lower-middle-income country. When compared with other lower-middle-income countries, Georgia trades more than its peers (i.e. imports and exports are higher as a share of GDP) but it scores poorly in terms of domestic

Engconsult Ltd. 9 savings. The World Bank’s 2005 purchasing power parity comparative program put Georgia’s PPP income per head at around the median for the ten countries of the CIS. 21. In addition to its role as a transit country, Georgia’s main activities are agriculture (citrus, grapes and nuts), mining (manganese and copper) and small scale industry. It is reported that two firms have just received licenses to operate in a new free industrial zone in Poti. Tourism is locally important, particularly in Ajara – see below. 22. GDP growth has averaged 8 percent per annum over the past five years, but fell sharply in 2008 to 2 percent. In addition to the global recession, Georgia has faced regional difficulties, first with the Russian trade embargo that began to bite in late 2006 and then the conflict in South Ossetia in August 2008. 23. The trade statistics show a widening trade gap, reaching 20 percent of GDP in 2008. The changing geo-political landscape is also evident. In 2004 the Russian Federation was Georgia’s largest trading partner, representing about 12 percent of total trade. By 2007 Turkey had displaced Russia, accounting for 13 percent of total trade, while Russia had slipped to 9 percent. Russia continues as a major source of Georgian imports (especially of gas) but since the embargo, very little is now exported to Russia. The rate at which tourist arrivals grew, evidently strong up to 2006, also declined in 2007, but as there are no country of origin data it is impossible to attribute a reason for the decline. 24. Georgia is a comparatively open country measured as (imports+exports)÷GDP and its openness has grown rapidly since 2004. Trade openness was 48 percent in 2007, comparable with EU countries such as France and Italy (who would have a much larger services component in their trade) and slightly lower than Turkey. Trade has grown faster than GDP. In current USD terms, the average annual growth in trade from 2004-8 has been 32 percent, compared with 25 percent for GDP.

B. Ajara . 1. Economy

25. Although Ajara is an autonomous republic, most data relating to it are included in those for Georgia as a whole. 26. Limited data for Ajara alone appear in the Georgian statistical yearbook. Its 2002 census population was 376,000, of whom 122,000 were resident in Batumi (down from 137,000 in 1989). GDP in 2007 is recorded as GEL518m (current prices), implying a GDP per head of GEL1,380, compared with GEL3,900 for Georgia as a whole. The principal economic activities are tourism, fruit growing (especially citrus) and transport. 27. The bare statistics above understate Ajara’s importance to Georgia. Although Batumi is only Georgia’s third largest city (with a population of 122,200 in 2008 after Tbilisi, with 1,106,500, and Kutaisi, with 188,600) it is a focus for development. Nearly all trade with Turkey, Georgia’s largest trading partner, takes place through the Sarpi border post, 10km south of Batumi. Batumi port is a natural deepwater harbor. 28. Tourism along the coast from Sarpi to Grigoleti (between Kobuleti and Poti) is an extremely large business. Table 10 summarizes Ajara government data for registered arrivals for 2005-2008. Foreign, especially Turkish, visitors have accounted for much of the impressive growth between 2005 and 2008. Equally, it is predominantly foreigners who stayed away in 2008.

Engconsult Ltd. 10

Table 10: Ajara visitors, 2005-8

Year Total Domestic Foreign Foreigners by nationality

Turkey Azerbaij Armenia Russia Others an 2005 147,000 120,000 27,000 2,369 1,700 20,000 820 2,111 2006 250,000 182,523 67,477 10,062 2,498 46,273 1,713 6,931 2007 352,085 239,786 112,299 39,588 4,628 54,996 2,595 10,492 2008 285,000 208,782 76,218 30,804 3,757 26,130 2,371 13,156 Ann growth, 2005-7 34% 26% 61% 155% 40% 40% 47% 71% Change, 2007-8 -19% -13% -32% -22% -19% -52% -9% 25% Source: Ajara government; consultants’ calculations

29. Table 10 data omit visitors who do not register (i.e. either do not spend the night or stay with friends and relatives or in other unregistered accommodation). This stretch of coast may receive around half a million tourist arrivals a year. 30. Ajara is known for its mandarin oranges and was at one time the principal source for the Soviet Union. In 2005 Ajara accounted for 5,800ha out of a national total of 8,700ha of citrus. However, this market disappeared in late 2006 as a result of the Russian embargo. Ajara has had to find alternative markets, but has found this difficult: grading and packaging have been below western European standards. 2. Access by train and bus

31. Passenger rail services are shown in Table 11. Passenger numbers provided by are unreliable, but based on observed ridership and train frequencies total arrivals in Batumi are estimated at perhaps 100,000 per year. Table 11: Passenger train services to Batumi Service Single fare, Approx distance, Duration, h GEL km

Tbilisi-Batumi/Makhinjauria 1 x night & 1 x day service dailyb 15-40 340 7h-9h30 Yerevan (Armenia)- Batumi/Makhinjauri Seasonal service, odd days 30-90 700 13h (Yerevan-Tbilisi) Kutaisi- Batumi/Makhinjauri Daily 4h20 Ozurgeti- Batumi/Makhinjauri Daily 2h Total Source: Georgian Railways Notes: (a) services now terminate at Makhinjauri, 5km north of Batumi (b) day services are every other day outside the tourist season

32. It is difficult to estimate the numbers who arrive in the project area by bus. Outside the summer holidays approximately 16 minibuses and 2 large buses leave Tbilisi daily for Batumi.

Engconsult Ltd. 11

These buses are “registered”, i.e. they are affiliated to a bus station. In addition there are unregistered vehicles, probably another 10 minibuses and two large buses. In summer there may be another 15 minibuses and two buses. Assuming eight passengers per minibus and 30 per large bus suggests a total of 136,000 arriving passengers. However, this is bound to be an underestimate as it omits buses starting from towns en route and international buses. The single fare from Tbilisi to Batumi is GEL15-18, increasing to GEL18-20 in summer, i.e. about the same as a second class rail fare for the same journey. 33. The Batumi airport operator’s website (www.tavyatirimciiliskileri.com) reports 39,637 passengers in 2007 and 69,354 in 2008, relatively small compared with those arriving by surface modes. (Batumi international airport has a new terminal, opened in 2007, with a capacity of 600,000 passengers).

C. Sarpi and Batumi trade 34. Sarpi border crossing is by most definitions the busiest in Georgia. Approximately a third of all individuals crossing Georgia’s international borders cross at Sarpi (Table 12). Although nationalities are not reported by border crossing, around 600,000 people holding Turkish nationality cross Georgia’s borders annually, overwhelmingly, it is assumed, through Sarpi, evidence of the popularity of Ajara with Turkish visitors. There is only one other border with Turkey currently open. This is at Vale, 100km east of Sarpi. Road conditions remain poor at Vale and only 27,000 people crossed the border there in 2007. Table 12: People crossing Georgia’s borders, 2007-8

Period Sarpi Tsiteli Khidi Lagodekhi Tbilisi Sadakhlo All borders (Turkey) (Azerbaijan) (Azerbaijan) airport (Armenia) 2007 1,601,776 1,126,020 384,443 622,779 682,191 5,007,537 Q1 2008 385,411 274,786 102,738 141,587 125,123 1,215,329 Q2 2008 538,540 299,124 152,232 205,962 159,576 1,578,388 Q3 2008 600,526 346,210 115,518 224,274 210,531 1,774,156 Source: Georgia Border Agency (www.gbg.gov.ge) Notes: (a) only the five most frequented border points are shown (b) Sadakhlo data include those crossing by rail (c) Q3 2008 data are affected by the conflict in S Ossetia 35. Table 13 shows the numbers of vehicles crossing at Sarpi in the direction of Turkey. Approximately 40 percent are goods vehicles (compared with 15 percent in the 2009 Batumi traffic counts), indicating the significance of merchandise trade. The proportion of truck traffic in transit is comparable with the 14 percent carrying foreign license plates at Batumi (see Supplementary Appendix 6). Table 13: Transit vehicles and vehicles crossing Sarpi border, 2007-8 Period Vehicles crossing border Transit vehicles crossing border Cars Trucks Buses Total Cars Trucks Buses Total 2007 125,023 114,394 22,757 262,174 8,051 27,854 676 36,581 Q1 2008 40,880 30,045 6,433 77,358 902 4,970 131 6,003 Q2 2008 50,492 40,932 7,723 99,147 1,337 6,294 106 7,737 Q3 2008 49,717 41,334 8,712 99,763 1,056 3,160 84 4,300 Percent 49% 42% 9% 100% 21% 77% 2% 100%

Proportion of each vehicle type in transit: 4.3% 18.6% 2.2% Source: Georgia Border Agency (www.gbg.ge)

Engconsult Ltd. 12

36. Batumi port is one of the deepest ports on the Black Sea (at 11-15m it is considerably deeper than Poti, with 7-11m). It has 11 berths: three for oil tankers, two for containers, one for rail, two passenger berths and four for general cargo. It is predominantly an oil exporting facility, however: Batumi shipped 11.2m tonnes in 2007, of which 9.6m were oil or refined product. 37. Following a modernization program Batumi’s oil terminal is capable of handling tankers of up to 130,000dwt and has a capacity of 240,000b/d. Poti, Georgia’s other port, is limited to tankers of up to 45,000dwt and has a capacity of 60,000b/d. (Regional oil pipeline capacity is however much larger: Baku-Tbilisi-Ceyhan has a capacity of 1m b/d and with upgrades and additional pumping stations could carry 1.8m b/d. The older Baku-Supsa pipeline has a capacity of 120,000b/d). 38. The oil terminal is owned by KazTransGas, a company owned by the Kazakhstan government. KazTransGas also manages the port on behalf of the state. Early in 2009 KazTransGas abandoned a plan to construct a refinery at Batumi. 39. Table 14 shows the number of ship calls at Batumi and Poti. While Batumi remains an important oil port, Poti dominates container traffic. (Poti remains a small container port, however. It handled 127,000 TEU in 2007, well below its capacity). Table 14:Batumi and Poti ship calls, 2007-8 Period Batumi Poti 2007 1,349 3,354 Q1 2008 761 1,061 Q2 2008 516 938 Q3 2008 371 882 Source: Georgia Border Agency (www.gbg.gov.ge) Notes: Q3 2008 data are affected by the conflict in S Ossetia 40. Table 14 shows rail freight traffic from Batumi for 2004-8. Oil and oil products account for 80-90 percent of tonnage carried and are overwhelmingly carried to and from international destinations. Table 15: Batumi rail freight traffic, 2004-8 Batumi-Potib Batumi-Tbilisib Batumi-otherbc t-km m Oil%a t-km m Oil% t-km m Oil% 2004 0.659 8% 7.807 20% 2,865 91% 2005 0.484 11% 16.278 34% 4,166 93% 2006 0.492 12% 10.614 20% 5,062 92% 2007 2.622 41% 11.203 11% 4,228 89% 2008 2.647 13% 23.904 8% 3,277 84% Source: Georgian Railway Notes: (a) percentages are based on tonnes carried (b) traffic is in both directions (c) includes international traffic (average trip length is 390km) D. Growth and inflation forecasts 41. Growth of GDP or of GDP per head is needed in order to forecast traffic (Supplementary Appendix 6). Inflation forecasts, needed for the financial evaluation and cost estimates (Supplementary Appendix 7), are included here for convenience. 42. As population is essentially stable (see Table 8), the traffic forecasts use GDP as the explanatory variable. Table 16 shows three GDP growth forecasts for 2009-11 (forecasts beyond 2011 are not currently available).

Engconsult Ltd. 13

Table 16: Georgia short term GDP growth forecasts

Source 2009 2010 2011 EIU, March 2009 country report 0.5% 1.8% - Georgia government, April 2009 economic overview 2.5% - - World Bank office, Tbilisi, April 2009 -2% 1% 2% IMF, World Economic Outlook, April 2009 1% 3% - Georgia Ministry of Economic Developmenta 2.5% 6% 8% Note: (a) letter from the MoED dated 15.04.09 in response to a project request 43. There is evidently, as yet, little consensus on the effects of the global recession on Georgia. For the purposes of traffic forecasting it is assumed that there is no growth in 2009, followed by 2.5 percent in 2010 and 3 percent in 2011. 44. Past growth in Georgia was greatly helped by FDI: net FDI was 19.8 percent of GDP in 2007 and 10.1 percent in 2008. An early return to these levels of investment – and the 8 percent average annual growth rate experienced during recent years – seems unlikely. Also, much lower long term GDP growth rates have been used to forecast traffic on other recent Georgian roads projects: 2.8 percent in the case of the Samtskhe-Javakheti Roads Rehabilitation Project (final feasibility study report for Millennium Challenge Georgia Fund, February 2007) and 3 percent in the case of the World Bank’s Second East-West Highway Project (project report, November 2007). Given much higher historic growth rates and the significant regional impact of trade, a slightly higher rate is justified here, 4 percent is adopted as the long term growth rate. 45. Transit traffic in Ajara will respond to regional growth rates in Turkey, the South Caucasus and Central Asia. Recent forecasts for growth in these economies are slightly higher than those for Georgia (for example, ADB’s 31 March 2009 GDP forecast for Central Asia was 3.9 percent for 2009 and 4.8 percent in 2010). However, a recent World Bank appraisal (Second East-West Highway Project, project report, Nov 2007) put long term regional growth rates no higher than those for Georgia, and in any event the goods vehicle elasticity used in this project (see Supplementary Appendix 6) implicitly factors in regional growth. For these reasons it was decided not to use a separate set of growth forecasts for transit vehicles. Table 17 summarizes the assumptions used in the traffic forecasting. Table 17: GDP growth forecasts

2009 2010 2011 2012 2013 2014 2015 on GDP growth, % per yr 0% 2.5% 3.0% 3.5% 4.0% 4.0% 4.0% Source: consultants’ estimates 46. Inflation forecasts were made available by ADB and are shown in Table 18. In practice, devaluation of the GEL, which cannot be ruled out, would have a significant impact on inflation. Table 18: Inflation forecasts Series 2009 2010 2011 2012 2013 on Georgia 7.0% 7.5% 6.0% 6.0% 6.0% Internationala 1.9% 1.0% 0.0% 0.3% 0.5% Source: ADB Notes: (a) Jan 2009 manufacturers’ unit value (MUV) index (proxy for the price of developing country imports of manufactures in USD terms)

Engconsult Ltd. FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 2 Road Standards, Road, Structures and Tunnel Design

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Contents I. INTRODUCTION ...... 1 II. ALIGNMENT ...... 2 A. Introduction ...... 2 1. Location of the Project ...... 2 2. Principles for Route Selection ...... 3 3. Recommended Alignment ...... 4 III. HORIZONTAL ROAD Design Standards ...... 7 A. Design Criteria ...... 7 1. Design Standard ...... 7 2. Embankment and Cut Slopes ...... 10 3. Retaining Structure ...... 11 4. Slope Stabilization ...... 11 5. Design Drawing ...... 11 IV. GEOLOGICAL SURVEY ...... 12 A. Introduction ...... 12 B. Terrain and Climate conditions ...... 12 C. Geomorphologic Characteristics ...... 12 D. Engineering geological conditions of the rocks ...... 13 E. Geological risks ...... 15 V. PAVEMENT DESIGN ...... 17 A. Introduction ...... 17 B. Pavement Surface Analysis ...... 17 C. Subgrade strength ...... 17 D. Traffic ...... 18 1. Determination of damage potential ...... 19 2. Distribution of traffic between the lanes ...... 19 E. Reliability ...... 20 F. Serviceability ...... 20 G. Proposed Pavement Structure ...... 20 H. Summary of Recommended Pavement Structure ...... 21 1. Pavement Depth ...... 21 2. Wearing Course ...... 22 3. Binder Course ...... 22 4. Base Course ...... 22 5. Sub-base ...... 23 6. Shoulders ...... 23 VI. HYDROLOGICAL STUDY ...... 24 A. Introduction ...... 24 B. Methodology...... 24 1. Culvert Opening Sizes ...... 24 2. Bridge Span ...... 24 3. Freeboard ...... 25 C. Hydrological Characteristics of Main Rivers ...... 25 1. River Descriptions ...... 25 2. Maximum Flow Rate ...... 27 3. Chorokhi’s Maximum Flowrate ...... 29 VII. STRUCTURES ...... 30 A. Design Standards ...... 30 1. General ...... 30

Engconsult Ltd. ii

2. Bridge Cross Sections: ...... 31 B. Bridges in the Project ...... 31 1. Large Bridges ...... 31 C. Technical and Design Summary: Bridges ...... 32 1. General Design Principles ...... 32 2. Other General Bridge Design Issues ...... 34 D. Interchanges ...... 34 1. Introduction ...... 34 2. Interchange Types ...... 35 3. Interchange locations ...... 35 VIII. OTHER PROJECT ROAD ELEMENTS ...... 37 A. At Grade Intersections ...... 37 B. Roadside Facilities ...... 37 1. Signs and Markings ...... 37 2. Roadside Service/Information Stations ...... 37 IX. Tunnel Design ...... 38 A. Introduction ...... 38 B. Standards and dimension of tunnels ...... 38 C. Tunnel Design ...... 38 1. Tunnel Cross Section ...... 38 2. Tunnel Support Design ...... 40 3. Project Tunnels ...... 42 4. Method of Tunnel Excavation ...... 42 5. Tunnel Portals ...... 43 X. Facilities for Tunnels ...... 44 A. Ventilation System ...... 44 B. Illumination ...... 44 C. Emergency Facilities ...... 44 D. Construction and Operation Risks ...... 46 1. Risks in Tunneling ...... 46 2. Risks and Targets in Operation and Maintenance ...... 47 E. Site condition and recommended method of construction of tunnels ...... 47 1. Introduction ...... 47 2. Zeda Achkva Ridge Tunnel (km 23.1-23.5) ...... 47 3. Three short tunnels (Makhinjauri I, II, III) between km 33.9 and 34.7 ...... 48 4. Peria RidgeTunnel (km 41.3 - 42.) ...... 48 5. SalibauriTunnel (Km 37.6-38.1) ...... 49 6. Existing twin tunnels at km 30 ...... 49

Annex A Tomograhpy Profile List of Tables Table 1: Major Civil Works’ Quantities ...... 6 Table 2: Georgian Road Standards ...... 7 Table 3: Additional Proposed Design Criteria ...... 8 Table 4: Conditions of Alluvial Sediments ...... 14 Table 5: Equivalency factors ...... 19 Table 6: Vehicle % in inside lane (4 lane road) ...... 19 Table 7: Cumulative Axle loads ...... 19 Table 8: Pavement Design Calculations ...... 21 Table 9: Bituminous base / Granular base ...... 21

Engconsult Ltd. iii

Table 10: Pavement Depth ...... 22 Table 11: Freeboard Allowances ...... 25 Table 12: Flow rates of Rivers and Streams...... 28 Table 13: River Chorokhi maximal flow rates ...... 29 Table 14: Summary of bridges ...... 31 Table 15: Large Bridges ...... 32 Table 16: Interchange Names and Locations ...... 34 Table 17: Approximate Tunnel Cross Sectional Area ...... 39 Table 18: Particulars of Tunnels ...... 42 Table 19: Design Target of Ventilation ...... 44 Table 20: Standard of Emergency Facilities of Road Tunnel ...... 46

List of Figures

Figure 1: Two lane Road Cross section ...... 9 Figure 2: Passing Lane Cross Section ...... 9 Figure 3: Four Lane Cross Section (Flat) ...... 10 Figure 4: Four Lane Cross Section ...... 10 Figure 5:Two Lane Bridge ...... 31 Figure 6: Cross Section Georgian Standard (Type 1)...... 39 Figure 7: Adopted Cross Section Type 1 ...... 40 Figure 8: Support Pattern Type 3 ...... 41 Figure 9: Support Pattern Type 1 ...... 41 Figure 10: Guideline for natural ventilation (two way traffic) ...... 45 Figure 11: Tunnel Classification ...... 45

Engconsult Ltd. iv

Glossary

ADT Annual Daily Traffic BoQ Bill of Quantities CBR California Bearing Ration CL Center Line DTM Digital Terrain Model E&M Electrical and Mechanical ESAL Equivalent Standard Axle Load FHA Federal Highway Administration (US) GoG Government of Georgia HFL/HWL High Flood (Water) Level JHPC Japan Highway Public Corporation RC/RCC Reinforced (Cement) Concrete RD Roads Department, Ministry of Regional Development and Infrastructure SPT Standard Penetration test

Engconsult Ltd. 1

I. INTRODUCTION

1. This Supplementary Appendix describes the selection of the alignment of the Ajara bypass roads and the preliminary design of the chosen alignment. It sets out the design standards adopted and design concepts for the major components, including major bridges and tunnels. The alignment and design standards are based on Georgian Road standards1 and have been discussed and agreed with the Roads Department of the Ministry of Regional Development and Infrastructure (RD). 2. The original preliminary design of the alignment was based on field inspection, a GPS centerline survey and the use of digital mapping to a scale of 1:10,000. The agreed alignment was designed based on field walk over surveys, topographical surveys and geotechnical investigation of tunnel and major bridge locations.

1 February 2009

Engconsult Ltd. 2

II. ALIGNMENT

A. Introduction

1. Location of the Project

3. The proposed Ajara Bypass Road Project is located in Ajara province2 in the western region of Georgia, and is contained in four districts, Batumi, Kobuleti, Khelvachauri in Ajara and Ozergeti in Guria. 4. The project road will provide a bypass of the two tourist resorts of Batumi and Kobuleti, reducing travel time for through traffic from nearly two hours to less than one hour, reducing congestion on the existing road and reducing road accidents, particularly pedestrian accidents. a. The Existing Road

5. Starting at Choloki Bridge the 10 km long Choloki-Kintrishi section of the existing coastal road passes through the main street of Kobuleti resort. Resort infrastructure and houses are located continuously on both sides of the street. During the resort season these installations and houses are filled with tourists. The road is straight with a road pavement of 8-10 cm of asphalt concrete. The pavement is constructed on a 35-40 cm thick gravel base. The width of road formation is 21-22 m, and the width of carriageway is 12-14m with sidewalks on both sides. The condition of the road pavement is not satisfactory, with many potholes. The road carries both the transit and urban traffic and during the resort season is congested with traffic flows exceeding 10,000 vehicles per day producing high air and noise pollution. 6. The 4.0 km long road section from Kintrishi-Tsikhisziri is also located in flat terrain and lies between the Samtredia-Batumi railway and the Black Sea, limiting the scope of any widening. The width of the road formation is 11-12 m and the width of the carriageway 8-9 m. The road pavement consists of 10cm of asphalt concrete, with 35-40 cm thick gravel base. 7. The 9 km section of road between Tsikhisziri and Chakvi traverses a section where the hills come down close to the coastline. The section has a road formation of 11-12 m, and the width of carriageway is 7.0 m. This road section is situated in difficult terrain and geological conditions and generally follows the contours using small radii curves (20-50m) with gradients as high as 8%. Traffic accidents are high as design of the section does not meet either the international or national interstate road standards. To improve the existing road it would be necessary to construct deep cuts and high fills in poor geological conditions with an area prone to landslides. The best illustrative example of landslide conditions is the deformations and slides originated in the cuts at the access to the new tunnel (see below). 8. The 7 km long Chakvi-Makhinjauri road section is identical to the previous road section; however, 4 km of new road have been recently constructed at the southern end, including construction of an 800 m long twin tube tunnel under the Green Cape. Although the road pavement on the approaches to the tunnel is two lanes, earthworks and drainage have been completed to accommodate four lanes. 9. On the existing Choloki-Makhinjauri section there are 28 culverts, 12 box culverts and 9 reinforced concrete simply supported span bridges.

2 The first 2 kilometers are located in the adjacent Guria province.

Engconsult Ltd. 3

10. AADT (2008) is: Choloki Bridge 4,300 vehicles, Choloki-Makhinjauri section, 10,200. Existing safety and service levels are very low especially in the tourist season. 11. The existing section of road through Batumi starts from Km 96 at Makhinjauri and ends at Km 111.5 at the Chorokhi River. This section of the existing road mainly consists of Batumi town streets thus the road conditions only meet city requirements. At present all traffic, including transit traffic goes through Tbilisi Highway, Gogebashvili, Chavchavadze and Abuseridze streets and leaves Batumi on the Nara Highway. There is little possibility of upgrading these roads on their present alignment because of the extensive city roadside development. 12. AADT at Chorokhi bridge (2008) is 4,400, however, traffic is much higher, and congestion greater on the major city streets. b. Alternative New Alignments

13. The road alignments proposed in the pre-feasibility studies, including some of the alternative alignments, were transferred to available topographic sheets and the suitability and condition of the road alternative corridors were assessed. 14. Combined with field inspections and preliminary geological investigation the possible alternatives to be evaluated were identified with regard to engineering, environmental and resettlement feasibility and impacts. Some of the original proposed alignment and alternatives were found to be infeasible and were dropped from consideration. 15. The alignments to be considered were plotted on 1:10,000 maps and a 500 m alignment corridor was digitized from these maps. Initial GIS defined control points where located on the ground and plotted on the digitized maps, and used as initial control points for the following topographic surveys. Finally the points were plotted on orthophoto maps to as an aid in examining the terrain, land use and structures in the alignment corridor. 16. Prior to the topographic survey twelve permanent control points were constructed close to the alignment and coordinated to the UTM system. It is intended that these points for the basis for the detail design and construction survey controls. 17. For the purpose of the initial alignment selection it was assumed that the road would be four lanes, except for the Batumi bypass. Further information on estimated traffic volumes on the bypass roads indicates that two lanes will be sufficient for the next ten to twenty years. It is assumed that land would be acquired (or frozen) throughout for four lanes (minimum 50m Right of Way (RoW). As the two lane road would be designed to be a future four lane road, four lane road design standards would be used throughout. 2. Principles for Route Selection

18. In general, a proposed high speed bypass alignment should meet the following principles: • It should fit in with the existing Road Network; • It should provide easy access from the major towns in the region; • It should take account of future traffic distribution; • It should be coordinated with local and national development plans; • It should minimize civil works quantities and avoid sections with bad geological conditions or difficult engineering works such as deep cut or high fill so as to minimize construction cost;

Engconsult Ltd. 4

• It should minimize environment impact along the corridor; • It should minimize structural and agriculture land acquisition; • It should provide ease of construction, for instance, by choosing economical hauling distance for local construction material, and avoiding disturbance with existing roads and utilities as much as possible; and • It should take account of opinions from local people in the selection of proposed interchanges, and the location and arrangement of pedestrian bridges and underpasses. 19. The above principles were used in the selection of the alignment for the project road. 3. Recommended Alignment

20. In the following sections the kilometer measurements assume that the new alignment starts at the southern abutment of the Natanebi River Bridge, Km 63.37 on the Senaki-Poti- Sarpi road, about 1.5 km north of the Choloki River Bridge and uses our recommended alignment (i.e. the alignment is longer than the prefeasibility alignment). The alignments discussed are set out in Maps 1-3. a. Kilometer 0-14

21. The prefeasibility alignment started at the Choloki River Bridge and used a modified trumpet interchange to connect to the existing road via an existing access road. It then briefly swung east and then south to parallel the existing road and traverse low lying ground between residential strip development and a bund constraining a wet land area. At about kilometer 6 the wetland area becomes a declared wetland protected area. The alignment paralleled this protected area at about 200mt distance. At the end of the protected area (Km 10) the alignment turned south-east to cross the main railway line to Batumi and the Sajavakho-Ozurgeti-KobuIeti state road at kilometer 14. 22. Under ADB and Government environmental regulations this alignment would go too close to the protected area. In addition, in April 2009 the Ministry of Environment extended the wetland protected area buffer zone to include parts of this alignment. 23. An alternative alignment in the prefeasibility was briefly considered which left the existing road close to Natanebi River, crossed the Choloki River 1.2-1.5 km upstream of the existing bridge, bypassed (but close to) the protected area, crossed the Samtredia - Batumi railway near to station, running parallel to the rail line to link with the first alternative at Km 14. This alignment alternative was rejected in the pre-feasibility. 24. After examining both alignments on the ground it was found that the second eastern alignment could be moved so as to avoid the new protected area buffer zone, minimize residential land and would need only two additional bridges over 50 m. 25. Two possible alternatives to the start were examined. One leaves the existing Poti - Batumi road north of the Choloki River (immediately south of the amusement park), thus starting the project in Guria province. It then crosses the Choloki on a new bridge and swings east towards the railway line. The other alternative start is similar to the prefeasibility preferred alignment and leaves the existing road just south of Choloki River using a modified trumpet interchange. The advantage of the second option is that it is about 0.5 km shorter and does not require an additional two lane bridge over the Choloki River. The disadvantage is that the alignment is constrained over the first kilometer and below the standard for a high class road. We recommend the first alignment starting south of the amusement park.

Engconsult Ltd. 5

26. Interchanges in this section are proposed at the start of the project and at the Sajavakho- Ozurgeti-KobuIeti road at kilometer 11.5. 27. The alignment is in flat terrain over this section and includes bridges over the Choloki and Ochkhamuri Rivers, one minor river bridge, one rail and three road underpasses. b. Kilometer 14-32

28. Over this section the proposed alignment generally follows the prefeasibility alignment except between Kilometer 14-17 and 18-20. In the former case the line is mover slightly eastwards to avoid first a chemical factory and then a water pumping area and associated sub- station and in the latter case the line is moved to avoid a new cemetery, to avoid going on top of an existing minor road and finally to allow sufficient height to overpass a new minor access road and bridge. 29. At Km 22 the prefeasibility alignment cut through the Zeda Achkva ridge. A closer inspection of this section indicates that the cut would be deep, in erodible material, would require the demolishment of a number of houses, would cut a significant access road and would require steep approach grades. It is proposed to replace this cut with a 400m tunnel 30. At Km 28.5 the new alignment joins to the recently reconstructed existing road and goes through an existing 700 m tunnel to descend to the old road at kilometer 32. This section was constructed as a two lane road, however, the tunnel is twin bore four lanes total and earthworks have been mostly completed for four lanes, although significant slope protection will be required. 31. Over this section one interchange is proposed at Kilometer 28 where the new alignment meets the existing (new) road. The section includes major bridges over the Kinikishi and Kinitrishi Rivers, 3 minor river bridges, eleven road underpasses (box culverts or bridges), one duplication of an existing overpass and one 400m tunnel. c. Kilometer 32-33.8

32. This section of the alignment follows the existing Kobuleti-Batumi road. It runs parallel to the coast and the railway line and is heavily developed. The road will require widening to four lanes plus adequate shoulders as this section will carry both through and local traffic. Acquisition will be required for about 10 buildings, part of a school frontage and a hotel parking area. 33. An alternative alignment moving east of the existing road from Kilometer 32-34 would encounter steep terrain and still be in residential built up areas and is therefore not recommended. d. Kilometer 33.8-47.8 Batumi Bypass

34. In the prefeasibility three alignments were examined. The westernmost alignment encroached on the Batumi municipal area and passed through some heavily built up areas; even more built up now than when the pre-feasibility was carried out. The advantage of this alignment was that is was 0.5 kilometers shorter and passed through easier terrain, obviating the need for a tunnel. 35. The pre-feasibility recommended alignment first rises to cross the rail line, turns south- east through three short tunnels, and descends to the Chakvisi Korolistskali River, crossing the river (Km 36) and some commercial structures on a 600m bridge. It then follows the slope of the land to descend to the Bartskhana River, also crossing this and the riverside commercial

Engconsult Ltd. 6 development on a 600m bridge (Km 39); both bridges are about 15m above ground level but will need long spans. The alignment then gradually rises, following the line of the Batumi Development Plan (BDP) area to the west of Batumi and runs along but outside the BDP margin. At Km 40.5 the alignment enters crosses a 200m bridge to enter a 400m tunnel (Peria Ridge) and then descends to cross the Batumi-Alchaltsikhe road at kilometer 44.5, with an interchange, to then swing west to join the Senaki-Poti-Sarpi road at kilometer 48. This alignment ends just before the existing bridge over the Chorokhi River. 36. A third alignment followed the second alignment for the first 5 kilometers and then went further east, bypassing the developed area around the Bartskhana river and keeping up to 2-3 kilometers east of the BDP boundary. It also used a 400m tunnel, rejoining alignment two just before the Batumi-Alchaltsikhe road. This alignment was 0.5 kilometers longer and an estimated 8 percent more expensive. 37. Both alignments 2 and 3 were inspected on the ground and it was found that, whilst there was some new residential development along alignment 2, this was also the case for alignment 3. In addition alignment 3 went through a new cemetery close to Bartskhana River, which would be difficult to avoid without demolishing a considerable number of residential houses. Thus the pre-feasibility advantages of alignment 3 appear to have disappeared. Therefore, except for a realignment at km 38 to avoid a historic house and land, minor realignment at kilometer 45 to avoid the cement factories and another minor alignment change in the vicinity of the tunnel to minimize demolishment of houses and avoid two small cemeteries, our recommended alignment is close to the pre-feasibility alignment 2. It has also been necessary to provide an additional 400m tunnel at the realignment at km 38, to increase the length of the Peria tunnel to 730m so as to reduce the grade in the tunnel to 3 percent and, because of the higher standards in the new Georgian road standards relating to vertical grades and curve radii, to lengthen some bridges and provide some new bridges across valleys in the vicinity of the tunnels. 38. A total of 15 bridges are required, not including interchange bridges. Major bridges are required over the Chakvisi Korolistskali and Bartskhana rivers (including road and building overpass) and over valleys approaching portals of both long tunnels, nine bridges/overpasses are required over minor roads. Interchanges are required at Khelvachauri Road (Km44.6) and the existing Batumi-Sarpi road (Km 47.8). A partial interchange is required at km 39, close to Bartskhanan River, (two ramps) to allow access to north Batumi. There are 5 tunnels in this section, three small tunnels (Km 34) total 550m, one 440m tunnel cutting through Salibauri ridge at 38km and one 730m tunnel cutting through the Peria ridge at 41km. 39. The major quantities of the proposed alignment are as shown in Table 1. Table 1: Major Civil Works’ Quantities

Earth/Rock Pavement Retaining Culverts Bridges Tunnels Chainage Walls Cut Fill Box Pipe < 100 Long Interchange Km m3 m3 m2 ‘000 m2 ‘000 No No no/m n/m no. no/m ‘000 ‘000 1-14 23 1,350 221 0.4 14 23 5/370 1/100 2 14-28 644 2,057 187 3.0 33 12 4/98 7/2,180 1 1/440 28-34 9 28 77 0.4 4 - 1/100 1 34-47.8 947 1,309 210 3.3 18 7 12/449 7/3,350 3 5/1,540

Engconsult Ltd. 7

III. HORIZONTAL ROAD DESIGN STANDARDS

A. Design Criteria

1. Design Standard

40. We have been requested by RD to use the new Georgian national road design standards; these are set out in Table 2 for a four lane road. These standards are comparable with other international standards; except for tunnel width standards which are not suitable for small tunnels (see Section IX below). In addition, the lane width (3.75m in flat terrain) is considered to be generous; 3.5m is often used, particularly if costs are a consideration. The additional lane width has not been shown to provide much additional capacity or safety. It is recommended that for the road widening in urban areas (Km 32-34 below) the shorter width be used to reduce expensive land acquisition. Note that using 3.75m lane width and two shoulders of 2.5m then the total carriageway width (not including median) for a two lane road in flat terrain is 12m, wider than that required for one carriageway of a four lane road. 41. Based on road class and terrain conditions, the design speed proposed for initial analysis of alternatives will be 120km/h for flat sections, dropping to 100km/h for rolling terrain and 80km/h for mountainous sections, although it is hoped to be able to provide at least 100km/h design for these sections. To reduce earthworks, however, vertical curve radii have been reduced to 100kph standard (see below).

Table 2: Georgian Road Standards

No Main Parameters Unit Flat Rolling Mountainous 1 Design speed km/h 120 100 80 2 Number of lanes 4/2 4/2 4/2 3a Lane width (4 lanes) m 3.75 3.5 3.5 3b lane Width (2 lanes) m 3.75 3.75 3.75 4a Shoulder width left m 1.0 1.0 1.0 4b Shoulder width right m 3.0 2.75 2.25 4c Shoulder Width two lane both sides m 2.5 2.25 2.0 5 Minimum width of central reserve m 4.0 4.0 4.0 6 Hard shoulder width for emergency m 2.5 2.5 2.5 stop 7 Verge for central reserve m 1.0 1.0 1.0 8 Maximum longitudinal gradient % 4 5 6 9 Minimum horizontal curvature m 700 450 250 10 Minimum radius of vertical curves Crest m 22,600 10,000 5,000 Sag m 7,700 4,900 3,200 11 Width of acceleration and m 3.75 3.5 3.5 deceleration lane 12 Minimum stopping distance For straight section m 250 200 140 13 Normal cross fall of carriageway % 2.5 2.5 2.5 14 Maximum gradient of super elevation % 7 7 7 15 Design vertical clearance bridges, m 5 5 5

Engconsult Ltd. 8

No Main Parameters Unit Flat Rolling Mountainous overpasses Design clearance of tunnels Vertical m 4.5 Horizontal – Two lane road tunnel m 11.4 11.4 10.65 - Twin Tube (4 Lanes) m 11.4 10.65 10.65 Service walkway width m 0.75 42. Whilst we have designed to these standards (except for tunnel design – see below) we consider that they are excessive in places. In particular the minimum vertical curve radii are high compared to other international standards, including TEM standards, and have led to heavy fill, longer bridges and one additional tunnel. We recommend that these be reconsidered in detail design to lower the cost. 43. For other concepts of the road design criteria not specified in the Georgian standards, it is proposed to use TRACECA standards; these are shown in Table 3. Table 3: Additional Proposed Design Criteria

No Main Parameters Unit Flat Rolling Mountainous 1 Minimum stopping distance for curves m 250 250 125 with radius < 500m 2 Design loads For bridges AK-11,HK-80 For Roads Axle 100 KH 3 Pavement structure . For roads To be decided For tunnels Cement concrete 4 Technical parameters of interchanges and junctions Design speed km/h 40 40 30 Min. horizontal curves m 50 50 30 Max. grade upward % 7 7 8 Max. grade downward % 8 8 9 One-way carriage way width m 4 4 4 Two-way carriage way width m 7 7 7 Shoulder width m 1.5 1.5 1.5 44. Other design concepts include: • For 2 lanes road sections, a passing lane will be provided on sections over 3 percent gradient, depending on length of grade to allow passing of heavy slower vehicles. The lane width of this lane will be 3.0 meters. • Where there is a passing lane the width of shoulder on the passing lane side will be 1.5meters. • The maximum gradient of tunnel sections will be 3 percent so as to allow heavy vehicles to maintain speed and also to reduce vehicle exhaust fumes in the tunnel section. The minimum grade will be 1 percent to allow for drainage. 45. Figure 1, Figure 2, Figure 3 and Figure 4 shows proposed typical cross sections3.

3 The cross-sections are diagrammatic and do not show side drains or allowance for guardrail.

Engconsult Ltd. 9 C/L

12.5m (11.0m)*

2.5 3.75 3.75 2.5

(2.0m) (2.0m)*

* In mountainous sections

Figure 1: Two lane Road Cross section

C/L

13.5m

2.0m 3.5m 3.5m 3.0m 1.5m

(Passing Lane)

Figure 2: Passing Lane Cross Section

Engconsult Ltd. 10

C/L

26.0m

4.0m 3.0m 3.75m 3.75m 1.0m 3.75m 3.75m 3.0m

(Median)

Figure 3: Four Lane Cross Section (Flat)

C/L

25.5m

(24.5m)***

4.0m 2.75m 3.5m 3.5m 1.0m 1.0m 3.5m 3.5m 2.75m

(2.25m)* (2.25m)* (Median)

* Mountainous

Figure 4: Four Lane Cross Section

Rolling/Mountainous 46. The right of way (ROW) is generally 50m and includes allowance for boundary fencing. In areas with steep cross slope the RoW is increased. 2. Embankment and Cut Slopes

47. The cut slope is designed based on its type (earth or rock) and follows national standards for natural slopes in stable situations. In general the average ratio of side slopes is 1:0.5 to 1:1. For short stretches in steep rocky terrain on the main line (between km 14-44) a 1:0.5 cut slope

Engconsult Ltd. 11 has been adopted. For each 5-8 m of cut slope height there is a debris-catching berm 1.5m in width. The slope surface will be stabilized by grass sodding, mortared rubble facing wall, anchor rod and reinforcing mat shotcrete, or anchor cable as appropriate. 48. The filling of embankments will mainly make use of soil and rock generated from cutting. The ratio of the side slopes is designed as 1 Vertical: 1.5 Horizontal for 0-5m high slopes and 1V:1.8H for 5-20m high slopes. At a slope changing point, a 1.5-m wide berm will be provided. The surface of the slopes shall be protected with sod revetment when the height is less than 4 meters and sodded rubble grids or arches when the height exceeds 4 meters. Where the alignment is close to rivers solid slope protection or retaining walls will be provided to ensure the stability of the subgrade under flood levels. In sites with wet and soft grounds the groundwater or retained surface water may result in a soft base beneath the subgrade, therefore, blind drains, crushed-stone or band drains will be installed. This generally agrees with international standards. 49. In sections alongside rivers, riverside slopes will be protected, based on design flood level and local conditions, with retaining walls. 50. In the preliminary design it has not been possible to provide the optimum balance between cut and fill and initial estimates show an excess of excavated material because of the tunnel excavation and cuts through the noses of ridges. The alignment will require further examination in detail design to provide a better balance and some bridges may need to be replaced by culverts or spill through abutments. 3. Retaining Structure

51. The retaining structures on the valley side of the alignment have been designed as RCC or wall anchor wall. In general RCC walls for up to 10m height and anchored wall and reinforced earth walls, where wall heights are greater than 10m in height. The requirement of the valley side retaining wall was determined by viewing the natural slope and design cross section. If the outer road edge was above the surface and natural terrain hill slope was greater than 1V:1.5H then it was considered that a retaining wall was necessary. 4. Slope Stabilization

52. Upon construction of the road formation, exposed cut and batter will need to be stabilized using bioengineering and other appropriate stabilization techniques. Preliminary quantities for bioengineering and other stabilization works are included in the Bill of Quantities (BOQ). The detail stabilization (including bioengineering, cement pitching, stone pitching etc.) requirements at specific locations will be made during detail design and construction stage. 5. Design Drawing

53. The geometric design of the project roads plan and profile was undertaken on the basis of the preliminary engineering surveys. The design drawings will be presented in Volume 3 of the Final Report. The plan drawings are presented at a scale of 1:2,000 for all road sections. The vertical profiles are presented at a horizontal scale of 1:2,000 and a vertical scale of 1:500. 54. The topographic contours shown on the plans and vertical natural surface profile along the design centerline are based on preliminary topographic survey and digitized 1:10,000 plans and should be reviewed at the time of detail design when more detail cross section surveys are carried out.

Engconsult Ltd. 12

IV. GEOLOGICAL SURVEY

A. Introduction 55. The Project area belongs to the area of wrinkled system of Ajara-Trialeti which is situated within the ring of the Meskheti, Shavshvebi and Chakvi mountain ranges. Geologically the area consists largely of middle Eocene age sediments. The lower part of the middle Eocene is lithologically represented with tabular of tuff, tuffo-gravel, argillite and limestone. Those sediments are mostly spread over middle and upper part of the Kintrishi River, the flanks of the Meskheti mountain ring, in the lower reaches of the Chorokhi River and the area of Chaisubani’s village. The upper part of the middle Eocene is represented by a mass of fragmented volcanic breccias, tuff and clinker stratums. In some places andesite, basalt, tuffo-gravel, tuff and marl can be found. Those sediments are mainly spread in the Ajara cauldron, the flanks of the Chakvi and Shavshebi mountains and the lower reaches of the Achara Wyali river. In the middle part of Chakviskali river intrusive sediments of upper Eocene are found, lithologically represented by syenites and syenite-diorites. Geomorphologically, north of Project Area comprises the Kobuleti coastal plain crisscrossed by numerous streams, while southern part is characterized by hilly ranges with deep gorges. B. Terrain and Climate conditions 56. The terrain and climate conditions of the project area are dictated by its orographic location directly close to Black Sea. There are two main areas: • A valley plain terrain with humid subtropical air, over humid soil and mires and single- deck mesophile vegetation; • A hilly terrain with subtropical air with well developed red soil and partially transformed vegetation of Kulkheti. The project area climate falls completely into the West Georgian humid subtropical area, which is characterized by high humidity, a large volume of atmospheric sediments (more than 1.800- 2.700 m), low fluctuation of temperatures and monsoon winds. The average annual temperature around +13°. The lowest temperature during the coldest month in Kobuleti and Batumi is +4.2 and 6.4° respectively and during August, the warmest month, 22.9 and 23.2°. C. Geomorphologic Characteristics 57. The geomorphologic nature of the area, which falls under the Ajara-Trialeti range tectonic system and the settling Kolkheti area, is still directly dependent on the morphostructures and young tectonic profiles. It is notable that the tectonic profile is interrupted and is of a graded nature, which is shown by the layer nature of the sea and river terraces and absence of denudation. 58. There are two sharply designated morphological units within the project area: • plain-accumulative valley; • hilly and gentle mountain terrain. 59. The plain-accumulative valley terrain includes the south part of the Kolkheti valley, which from the North-East side is bordered by the Kobuleti and Chakvi mountain rings and from South by the Black Sea and Shavsheti (Lazistani) mountain ring. Morphologically it occupies the Kobuleti and Kakhaberi accumulative plains, which is framed by deltaic river-sea type sediments whose maximum capacity varies from 140 m (Kobuleti plain) to 300 m (Kakhaberi plain).

Engconsult Ltd. 13

60. The Kobuleti plain valley is bounded by the Tsikhisdziri volcanic rock hillocks in the south and by the Natanebi River in the north and represents the newest tectonic sag block, which incurs settlement of 2 mm annually. The settlement of the structure block was followed with quarter transgressions and strong river and sea sediment accumulation. Cobble and Eocene period volcanic rock masses are represented by XXX massif crystallized rock, which proves that the Chorokhi river deposits were spread out as far as the outfall of the Natanebi River. 61. The Kobuleti plain valley morphology is designated lagoon-mire terrain, which runs along the shelf zone and sea coast in a narrow line. In the 8-9m depth the sea-alluvial cross section of the lagoon-mire terrain is substructed. 62. An old bank dike is spread along the coast stretching for 10 km with an average width of 250-300 m and height of 7-12 m. The coastal dike is represented by Fanagori regressive relict, which is erected with the capacity up to 19 m sand-break stone-gravel sediments. 63. The Kakhaberi accumulative plain belongs to the Chorokhi-Batumi structural block, which during the whole quarter period was incurring and still incurs settlement from 0.8 m (Batumi) to 1.3 m (Ghorokhi river delta). The Chorokhi River divides the Kakhaberi plain into two unequal parts. On the southern wide part, where there are several relict lakes, Batumi is located. The narrow northern part of the Kakhaberi plain is represented by low grove terraces. 64. The gentle mountainous hilly morphological area is the part of the Ajara-Guria highlands and the Chaqvis and Kobuleti Ridges are its branches and its extreme south part is the Shavsheti Range western end. The height of this morphological area ranges from 200 – 600 m. The morphological conditions of this terrain are mostly specified by Middle Eocene volcanic rocks and the constant intensive chemical process of laterization. In the hilly coastal terrain area residual fragmentarily sea terrace steps can be found – from Chauduri, including Black Sea and in water dividing sinciput area – denudation surfaces. The elevated hilly terrain is significantly cut by the Kintrishi, Chakvistskali, Korolistskskali, Ajaristskali and Chorokhi rivers, which begin from the high mountains Small Kavkasioni and Pontidebi rise and cross low mountain hilly areas where the rate of erosive partition equals to 2,0 – 2,5 m/cm2. Most of river water dividing crowns are represented by undulant-denudative surface. In the spole of the big river’s ravines two to three step terraces have been developed and small rivers and ravines are creating V- profile erosive formations with inclined with frequent flood and landslides. D. Engineering geological conditions of the rocks 65. The tectonic and geological frame of project area has few hard conditions. There are two main ranges – Ajaristskali syncline and Chakvistavi anticline, which spread over the whole Ajara region and are complicated with small sized brekiformal ranges. 66. The Ajaristskali syncline occupies the whole upper part of the Ajaristskali river basin. The Syncline’s south wing is sloping (15-30°) and the north wing is steeper (40-75°); occasionally rocks are overturned to the south. The Syncline’s widest part (30 km) is in the east part of the region, the west part narrows to a minimum at the down belt of the Chorokhi River and then it is cut by the Black Sea coast. 67. The Chakvistavi anticline is located north of the Ajarastskali River. In the west part it has almost crossing location and in the East part it goes in a North-East direction. The Anticline has an unsymmetrical structure with a sharp slope at the south wing (40-65°) and gentler slope at the North (10-30°). These structures have fault dislocations and breaks, which are reflected in the terrain morphology. The largest tectonic structure of project area is the Middle Eocene volcanic-sedimentary slopes, which are represented by massive and thick-foliated volcanic breccias, tufas, with andesite face embedment, tufa conglomerates, and greenstone intrusive rocks. In the upper part are occasional tuffs, gravel stones, tuff-sandy-stones and margels. 68. All these rocks from an engineer-geological point of view are in rocky mountainous and partially rocky mountainous conditions. The density of the rocks varies from 2.6-2.9 gr/cm3,

Engconsult Ltd. 14 water-absorption of 0.4 to 5.44 percent, natural condition resistance of 72-143 mpa, and water- absorption condition 51-130 mpa. In weathered areas the characteristics are lower: density is 2.3-2.4 gr/cm3, water-absorption 2.8-5.22 percent, resistance 47 water absorption 27-50 mpa. 69. Structured by the volcanic rocks the hilly terrain area is almost covered by alluvial-diluvia valley embedment and therefore despite the high resistance and stability of the rock the area is significantly damaged by geological processes. 70. The biggest part of the project area is covered by valley formation and litho-geochemical structure conditions and the capacity of the foundation fully depends on the weathering- disintegrating, geo-morphological location and climate conditions of the main rocks. 71. In the whole territory of the mountainous Ajara region mechanical destruction processes of the rocks are occurring, timing intensiveness of its disintegration, creation of exhaustion areas and capacities are dependent on the petrologic composition, resistance, tectonic disorder and terrain morphology. In this terrain situation in some places there is a fully weathered profile (areas of full destruction or highly dispersed and cleft boulders) in a frame of one meter to one meter capacity. Sometimes there are places of clay sedimentation from the upper to lower horizons which strongly decreases the filtration processes. In the valley formation there two main complexes: alluvial-diluvia and hypergene-lateritic. 72. Sediments of alluvial-diluvia complexes are characterized by widely dispersed, complicated engineer-geological conditions and fully developed multilateral landslides phenomenon. When at the high slope valleys the sediments capacity are no more than meters, at the sloping and flat surface this capacity comes up to tens meters. These sediments are represented by big-fractional, clay-clay lines, clay- RorRovani, RorRovani da dresvis rows. 73. Often fossil soil layers occur and become a divisor of alluvial layers or are replacing these layers. Mostly at a flat and sloping terrain different percentage of fragmented inclusions are contained in clay peciaes. sediments. 74. Close to the zones of deep hypergenic there are layers of high porosity clay and claylines approximately 51 to 56 percent. In other areas porosity is no more than 39-42 percent. Natural humidity also varies from 36-42 to 11-20 percent. 75. The Table 4 shows the main common conditions of alluvial-diluvia sediments formed by the volcanic-sedimentary rocks: Table 4: Conditions of Alluvial Sediments

Relative Rock Natural Volume Level of Plasticity Density Porosity Natural Humidity Type Humidity weight fluctuation No. g/cm3 Sida Level of Plasticity natural % Zvris Compression fluctuation Number humidity kuTxe Clayline 36.32 1.78 2.73 51.3 42 12 2,251 0.045 Clay 45.28 1.67 2.74 56.5 62 23 1,921 0.085 76. Clay and clay-line mostly are macro-porous. They have weak structural connections and in cuts no lithiphication and solidification is found. Levels of fluctuation vary from 30-38 to 42-64 percent. They have low water stability; usually dissolving in water from several seconds to 5 hours. 77. In 7-13 m deep cuts their volume weight changes which can be a result of weak zones of deformed layers. These rocks, both in natural humidity or water impregnated conditions, are characterized with plastic and fluctuate -plastic consistency.

Engconsult Ltd. 15

78. In the project area alluvial-diluvia rock complexes have sporadic liquidation. In most areas they create variable water containing underground areas. 79. These areas lose water through a flat surface and valley bottoms which lead to springs, whose flow varies from 0.1 to 1.0 l/sec. The water is fresh with mineralization of 0.277 m/l, including Sodium hydrocarbon or Calcium hydrocarbon sulphates. 80. Some valley locations have complex laterite sediments, which are created in the low mountainous/hilly areas as a result of volcanic rocks deep hyper geneses during the Pliocene quarter period. 81. Complex hypergenic–laterite is characterized with deep weathering of volcanic rocks during the Pliocene-quarter period. 82. The alkali volcanic rocks are mostly shaped reddish and yellowish-brown laterites, which in literature are mentioned as “Batumi laterites”. The Laterites weathered capacities are measured by tens of meters and in some places reaches 100-150 m. Active relive capacity of dispersed and disintegrated areas reaches 20-30 m. This area is rich with aluminous soil, ferrous and hydrochloride; these can be seen in some cuts. They contain up to 20 percent of E2O3, 40-50 percent of SiO2, and are acidic (Ph-4.7-5.1) and there is an active process of kaolinization. 83. Laterites are represented with clay-lines and cooperatively rarely by clays; lower layers of cuts are characterized by strongly weathered surface rocks fragments inclusions up to 10-12 percent. 84. Clay and clayline rocks are characterized with high porosity (60-70 percent), a high range of natural humidity (from 10-15 to 52-61 percent), and a grain crisp structure. Mostly they have a low density (1.13-1.63 g/cm3) and a low plasticity number (18-21). The index of its full water absorption is as high as 65-68 percent and is higher than limit (59-65). They have a high filtration rate, in a nature humidity condition the solid consistency and increased firm indexes (internal friction angle) decreases to 3-5 and power of resistance is 0.05-0.45° kg/m2. 85. Usually these rocks are in a natural humid condition in valleys at 5-15 m height and have high angles (45-60° and more). They have high water stability and therefore there is a less possibility for deep and big landslip development. However, for a large part of the area a high consistency of ferrous hydroxides significantly increases plasticity indices and, as a result, in laterite areas in a condition of abundant atmospheric sediments freely infiltrated waters quickly change the clay soil consistency parameters and are often close to the water-absorption limit condition where they change to an unstable condition, leading to landslips and large ravines development. E. Geological risks 86. Historically formation and development of natural geological processes always had a place in Ajara and this development was mostly connected with a wide spectrum of natural factors, in particular: • vertical arrangement contrast of terrain elements, deep and intensive fragmentation of the terrain and high valley slopes, 65 percent of which have slopes more than 30°; • intensive overturn of structural rocks, fissure development, process of deeply weathered and strong disintegration-friability, significant fall of main conditions and in total low physical-mechanical indexes of the surface quarter sediments, all these lead to the hard engineer-geological conditions in the region; • an abundance of atmospheric sediments and high humidity, up to 2,000—3,000 mm.

Engconsult Ltd. 16

• Close location of Ajara’s young mountain range system to Javaxeti and Pontidebi highly intensive seismic regions, which results in some energetically high earthquakes and strong tensile growth. For example, 4-5 degree earthquake in 1986 in Kobuleti and Khelvachauri region caused activation of landslip and gravity processes, which resulted in damage to 450 houses. 87. During last ten-years significant development in the region caused disturbance to the geological environment. 88. The situation has become more c complicated because of the formation mechanisms and intensiveness of the erosive, landslip and flood water processes which have acted together. For example, 80 percent of cases of flood waters landslip activation. 89. The best proof of this, that at the territory of Kobuleti and Khelvachauri regions the 1990 1:10.000 scale engineer-geological survey identified 124 potential landslip areas and 11 potential flood water sources over 39.6 ha. The total area of the region with geologic processes causing intensive damage area is 5.1 percent and activating critical risk is 11.7 percent. 90. The National Environmental Agency registered up to 2008 in Kobuleti and Khelvachauri region: Kobuleti landslips – 78, rock fall areas – 6, river banks ablation – 26 km. Khelvachauri landslips – 85, flood water ravines – 19, gravity zones – 29, river banks ablation – 21 km.

Engconsult Ltd. 17

V. PAVEMENT DESIGN

A. Introduction 91. In order to determine an adequate technical and economical pavement, a study has been made using AASHTO Guide for Design of Pavement Structures, 1993, Volume 1, Design Procedures for new construction or reconstruction. The AASHTO methodology is based on precise input numbers for material properties, performance, reliability and traffic. Based on available results of the geotechnical and pavement investigations a preliminary pavement design has been determined. The road pavement design is based on the following main relevant input data: • Subgrade Strength • Traffic • Reliability • Servicability

B. Pavement Surface Analysis 92. The recommended pavement design is for an asphalt concrete pavement. The reasons for this are: • The alignment passes through areas of poor geology and soils, especially in the initial peaty area and in some mountainous terrain, There is, therefore, a concern about the lower subgrade leading to settlement. Normally this would be mitigated by waiting for a period after the subgrade and base was laid and compacted before applying the surface layer to allow for any settlement. Given the urgency to complete the project, this is not possible. If there are some areas of settlement it is much easier and cheaper to repair asphalt than cement concrete. • An anti-skid layer can be incorporated in asphalt concrete surfaces. This is important in the mountainous areas. • The use of asphalt pavement is common in Georgia as is maintenance of asphalt pavements. • Asphalt pavements are now cheaper to construct in Georgia. 93. During detail design, however, a whole of life comparison of both asphaltic and cement concrete should be carried out. C. Subgrade strength 94. The subgrade strength has been evaluated based on the California Bearing Ratio (CBR). The subgrade design CBR has been evaluated for similar road sections. 95. The AASHTO design method uses the Resilient Modulus as a main input factor to characterize the roadbed soil. Determination of the Resilient Modulus is given in AASHTO Test

Engconsult Ltd. 18

Method T274, however, a direct determination of the resilient Modulus could not be done in Georgia as the equipment and laboratories for the execution of the required test are not locally available. As this is not an unusual case, correlations and equations have been developed which can be used to estimate the Resilient Modulus (MR) from the standard CBR value. In addition to the correlation given by AASHTO there are various other relationships that are used around the world: AASHTO (Heukelom and Klomp)

MR (psi) = 1,500 CBR, valid for values of CBR < 10 Transportation and Road Research Laboratory (TRRL)

MR (psi) = 2,555*(CBR)^0.64 U.S. Army Corps of Engineers (Green and Hall):

M R (psi) = 5,409*(CBR)^0.71 96. The equation suggested in the AASHTO Design Guide is considered reasonable for fine grained soils with a soaked CBR of 10 or less. The TRRL and AASHTO correlation provide results within a reasonable range where the U.S. Army Corps of Engineers equation gives results almost twice of the AASHTO correlation. The TRRL quotation will therefore be used for the determination of the Resilient Modulus from CBR values for this project.

0.64 MR (psi) = 2,555*(CBR) This is considered a slightly conservative approach. D. Traffic 97. Studies carried elsewhere have shown that overloading is a serious problem on unregulated roads. Overloading leads to: 1. Increased /accelerated potential damage to the pavement which increases exponentially with the actual increase in the axle loads and determined in terms of equivalent factor or damage factor according to the following empirical formula:

5.4 ⎛W ⎞ EF = ⎜ o ⎟ ⎝ SA ⎠

Where

EF -Equivalent damage factor. Wο - actual axle load. SA - Standard axle load equal to 10.0 t. 98. Higher structural thickness of the pavement needed to accommodate much higher stresses applied at the pavement surface and ultimately transmitted to the sub-grade through various pavement layers. 2. Increased damage to the vehicles and increased cost for maintaining the vehicles. 3. Increased cost for maintaining the road.

Engconsult Ltd. 19

4. Increased potential for road accidents and vehicle breakdown. 99. Although overloading can be and should be controlled at least to some extent by appropriate regulatory measures it is unlikely that overloading will be controlled completely. It is therefore necessary to design the pavement to cater for expected or regulated overloading. 1. Determination of damage potential

100. There are no studies on major roads in Georgia examining overloading and therefore the equivalency factor is unknown. The project road will carry a large percentage of heavy truck traffic. For this traffic an equivalency factor of 4.6 has been assumed based on HDM4 recommendations and experience elsewhere. Other equivalency factors are set out in Table 5 Table 5: Equivalency factors

Large Car Mini-bus Med bus LGV MGV HGV Artic bus 0 0.01 0.7 0.8 0.1 1.25 2.3 4.6 LGV = Light Goods vehicle, MGV = Medium Goods Vehicle, HGV = Heavy Goods Vehicle Artic = Articulated Goods Vehicle

2. Distribution of traffic between the lanes

101. Since most of the road is to be initially built as a two lane facility it has been assumed that all traffic will use the same lane. For the widening of the existing road to four lanes it has been assumed that most of the heavy goods vehicles will use the inside lanes. The percentages assumed are set out in Table 6. Table 6: Vehicle % in inside lane (4 lane road)

Large Car Mini-bus Med bus LGV MGV HGV Artic bus 40% 50% 75% 90% 60% 80% 90% 100%

102. The traffic load is based on the forecasts of future traffic, vehicle damage factors and cumulative standard axles for a design life of 20 years. For the pavement design the cumulative equivalent standard axles for a 20 year (2009-2029) design life are used for the relevant road sections as set out in Table 7. Table 7: Cumulative Axle loads

Design life ESAL *106 Road Section Km (years) Total traffic Design lane 0 15 20 11.3 11.3 15 28 20 11.3 11.3 28 34 20 13.8 13.3 34 40 20 13.1 13.1 40 48 20 13.1 13.1

103. Detailed traffic data are presented in Supplementary Appendix 6.

Engconsult Ltd. 20

E. Reliability 104. The reliability design factor accounts for chance variation in both traffic prediction and the performance prediction, and therefore provides a predetermined level of assurance that pavement sections will survive the period for which they are designed. 105. For a given Reliability level (R), the reliability factor is a function of the overall standard deviation (So) that accounts for both chance variation in the traffic prediction and normal variation in pavement performance prediction for a given W18.

106. Values of So developed at the AASHO Road Test corresponds to a total standard deviation for traffic of 0.35 for rigid pavements and 0.45 for flexible pavements 107. Suggested Levels of Reliability for Various Functional Classifications are given in the AASHTO Guide. For the project road, which can be classified as a Principal Arterial passing through rural and urban areas, a Level of Reliability of 95% is recommended. F. Serviceability 108. The serviceability of a pavement is defined as its ability to serve the type of traffic which use the facility. The primary measure of serviceability is the Present Serviceability Index (PSI), which ranges from 0 (impossible road) to 5 (perfect road). Selection of the lowest allowable PSI or terminal serviceability index (pt) is based on the lowest index that will be tolerated before rehabilitation, resurfacing, or reconstruction becomes necessary. An index of 2.5 or higher is suggested for design of major highways and 2.0 for highways with lesser traffic volumes.

109. The original or initial serviceability value po has been observed at the AASHTO Road Test at 4.2 for flexible pavements and 4.5 for rigid pavements. 110. The total change in the serviceability index is defined as:

∆ PSI = po - pt 111. Given the projected traffic volumes on the project roads it is recommended to use an overall terminal serviceability level of 2.0 for the complete length of the road.

112. With a terminal serviceability index pt=2.0 and an initial serviceability value po=4.2 for the project road the total change in the serviceability index is: ∆ PSI = 4.2 – 2.0 = 2.2

G. Proposed Pavement Structure 113. Taking into consideration the traffic loads and subgrade material alternative pavement structures have been evaluated. The main reason for the evaluation of the pavement structure in the Feasibility Study was to find the technical and economical viable solution for use in the cost estimates and economic evaluation. The currently high costs of bituminous products which have to be imported are a major cost factor which is reduced with reduction of the thickness of bituminous layers. 114. The following three alternatives for each road section for the pavement structure have been evaluated in the feasibility study: (A) Flexible surface with Asphalt Surface and hydraulic bound base (B) Flexible surface with Asphalt Surface and bituminous base (C) Flexible surface with Asphalt Surface and granular base 115. The construction costs of the pavement structures have been estimated for each of the three options construction with hydraulic bound base (Variant A), bituminous road base (Variant

Engconsult Ltd. 21

B) and construction with granular base (Variant C). Construction cost estimate comprise the pavement courses, including granular sub-base, only and it does not take account costs of earth works which does nor differ between the three alternatives. 116. The cost estimate presented in the Feasibility Study indicated clearly lower construction costs for the pavement structure with granular base. Based on the result of the cost estimate and cost comparison of the three pavement structure alternatives the pavement design with a granular base had been recommended. Only this recommended pavement alternative is further detailed in the following. 117. The first section from . Natanebi River-Road to Qveda Kvirike Village (km 0.0 – km 15.0) goes through an area of soft peaty soil, which is confirmed with the CBR tests conducted. However, the section has slightly lesser traffic load (11.3 msa) compared to the other sections of the design road. 118. Along the road section, Qveda Kvirike road – Chaqvi (km 15 – km 28), the design road begins to cross river sedimentation geotechnical zone and then enters rolling terrain in cut. The section has the same traffic load as the first section (11.3 msa). . 119. In the Subsection Chaqvi-Makhinjauri tunnel (km 28 – km 34) the existing road granular sub base layer has a thickness of not less than 200mm, which is reflected in the CBR test data obtained. This material should be used for the capping layer. 120. Within the Makhinjauri tunnel- Akhalsheni River road (km 34 – km 40) section of the design road subgrade soil is represented as similar to that of the section 2, Qveda Kvirike road – Chaqvi 121. Akhalsheni River road–Chorokhi River (km 40 – km 48) passes the valleys of rivers and is represented by stronger soil conditions. 122. For each of the above subsections pavement design calculations has been carried out and are summarized Table 8: Table 8: Pavement Design Calculations

Length # Section Chainage R S0 W CBR MR ∆PSI SN km Natanebi River —Road to 1 0-15 15 0.95 0.45 11.3 3 5.16 2.2 6 Qveda-Kvirike Road to Qverda Kvirike- 2 15-29 13 0.95 0.45 11.3 10 11.15 2.2 4.5 Caqvi 3 Chavqi-Makhinjauri Tunnel 28-34 6 0.95 0.45 13.3 15 14.46 2.2 4.2 Makhinjauri Tunnel- to 0.95 0.45 4 34-40 6 13.1 10 11.15 2.2 4.5 Akhaisheni 5 Akhalsheni-Chorokhi River 40-48 6 0.95 0.45 13.1 15 14.46 2.2 4.2 123. Because of the fact, that bituminous materials are imported in Georgia and gravel materials are available in the project area, the depth of the asphalt concrete has been assigned as minimal in the design of pavement structure, giving importance to gravel materials to option for cost effectiveness. H. Summary of Recommended Pavement Structure

1. Pavement Depth

124. Table 9: Proposed thicknesses of pavement layers, Table 9: Bituminous base / Granular base

Engconsult Ltd. 22

Road Section base Chainage Granular Sub Granular Base Capping Layer Asphalt concrete concrete Asphalt (Sand-gravel soil) Bituminous Binder km Mm mm mm mm mm

1 0-15 50 70 220 300 500 2 15-28 50 70 230 300 28-34 50 70 180 300 3

4 34-40 50 70 250 300 40-48 50 70 180 300 5

125. Assigned structure of the pavement has been checked to meet AASHTO requirements for the value of Structural Number, represented in Table 10:

Table 10: Pavement Depth

Pavement Structure Depth Factor 1 2 3 4 5 Asphalt wearing Course 0.44 50 50 50 50 50 Asphalt Concrete Binding Course 0.4 70 70 70 70 70 Granular base Course 0.14 220 230 180 250 180 Sand-Gravel Sub-base 0.11 300 300 300 300 300 Capping Layer (Sand-Gravel Soil) 0.08 500 Conditional Depth 153.8 115.2 108.2 118 108.2 Overall required Depth for SN value (mm) 25.4 152.4 114.3 106.7 116.8 106.7

2. Wearing Course

126. A continuously graded asphalt concrete based on Marshall test criteria is recommended for the construction of the structural surfacing. On pavements having a granular base emphasis should be given to a flexible bituminous mix. The binder for the bituminous mix shall be a 50/70 penetration grade bitumen. Crushed river gravel will meet the requirements for aggregate production. 3. Binder Course

127. Between the wearing course and Bituminous/Granular base course the binder course of an asphalt mix based on Marshall test criteria is recommended. The binder for the bituminous mix should be 50/70 penetration grade bitumen. A nominal mix proportion of 5.0% bitumen in average is assumed for this binder course. Crushed material from borrow areas with natural river gravel will meet the main requirements for aggregate production. Recommended maximum nominal aggregate size is 20mm. Detailed testing of selected material sources is required to confirm and verify material properties prior to start of construction work. 4. Base Course

Granular Base Course Material

Engconsult Ltd. 23

128. According to the pavement design, graded crushed stone base is recommended for most of the project sections. It is technically justified and the use of such bases is accepted practice in Georgia. If it is necessary, adjustment of the grading might be made by mixing in non plastic fine grained natural material. In any case the blend should be none or slightly plastic and should produce laboratory soaked CBR values in excess of 80 %. 5. Sub-base

Sub-base of Natural Granular Material 129. This type of sub-base material can be selected from road excavation and/or from borrow areas. The material consists of river gravel and sand and gravel of glacial and alluvial origin. Selection within the material sources will be required in some cases. The recommended plasticity should generally not exceed a plasticity Index of 6. 130. River gravel is often coarse grained and a nominal treatment of the excavation material by means of screening may be necessary in order to produce a grading to fulfil sub-base requirements. 131. For all sub-base of natural granular material a minimum CBR value of 30 % is required. Sub-base of Crushed Stone Material 132. Material for this purpose will be produced from fresh stone and shall comply with the requirements of graded crushed stone base as described above for crushed base course except that a minimum CBR value of 30% is required. 6. Shoulders

133. For better internal drainage of the pavement layers it is advisable to extend the base and sub-base course across the shoulders to the edge of the embankment or drainage ditches as the case may be. This has the further advantage of simplifying the construction. As the shoulders are then made of low or non-plastic material they should be protected against erosion, wear and the ingress of water. This can easily be achieved by proper compaction and regular maintenance or the application of topsoil with grassing.

Engconsult Ltd. 24

VI. HYDROLOGICAL STUDY

A. Introduction 134. The rivers in Georgia generally rise in steep mountainous areas and flow to gentler slopes in the central plain and coastal fringe. Due to variability, both temporally and spatially, in the rainfall pattern and intensity, the resulting flood flows and the consequent sediment flows are also variable. The available hydrological and meteorological records of Georgia show that the range and the consequent standard deviation of the flood flows and the 24-hour maximum rainfall are very high. 135. The proposed bypass road alignment crosses 27 rivers and rivulets. The culvert and bridges over the rivers should be designed to be safe during an occurrence of floods within the expected design life of the structures. At the same time, the design of the river crossing structures should be economical and within an acceptable risk limit. To determine the different hydrological parameters required for the safe, efficient and economical design of the river crossing structures a hydrological study of all the rivers and rivulets which cross the proposed road alignment of the project was conducted. B. Methodology 136. The basic data for the hydrological study were the topographical maps published by the Survey Department, and the project topographical survey. The catchment of the rivers at the river crossings was manually drawn on the topographical maps of scale 1:25,000 and the catchment area was estimated by the square-count method. The cross section and the river bed slope of the rivers at the river crossing were obtained from the topographical survey. 1. Culvert Opening Sizes

137. A road culvert is a river crossing structure having a gross opening of 6 m or less between its abutments. The openings required for the pipe and box culverts, where the proposed road alignment crosses the river, were estimated based on the design discharge, the roughness coefficient of the proposed culvert structure, and the topographical slope in the vicinity of the river crossings. The slope of the culvert should be kept as close to the topographical slope as possible to minimize excavation and/or filling. When the topographical slope is high, the slope of the designed culvert should be reduced to avoid very high flow velocity (velocity higher than 3 m/second). A free board of 20% should be provided, i.e., the size of the opening should be determined so that the design discharge can pass through 80% or less of the cross-sectional area of the opening. 2. Bridge Span

138. The total span of the bridges over the rivers were determined by the cross sectional area of the river at the crossing, the bed slope, and the roughness of the bed materials at the proposed bridge location. The cross-sectional area and the bed slope were obtained from the topographical survey. The roughness of the bed material was estimated from the composition of the materials and their approximate sizes. 139. The flood level which matched the design discharge at the river cross-section was taken as the HFL. The free board was added on the HFL to determine the minimum elevation of the bridge superstructure. The opening in the cross-section at the minimum elevation of the bridge superstructure was taken as the effective span of the bridge. The total bridge span was obtained by adding the width of the piers to the effective bridge span.

Engconsult Ltd. 25

3. Freeboard

140. The difference between the elevation of the lowest part of the bridge super structure and the HFL is the free board. The freeboard in a bridge allows the safe passage of flowing debris during flooding. The permissible values of free board for bridges for different discharges commonly used are set out in Table 11. Table 11: Freeboard Allowances

Discharge Vertical Clearance (mm) (m3/s) < 200 600 201-500 800 501-2,000 1,000 2,001-5,000 1,500 >5,000 >2,000

C. Hydrological Characteristics of Main Rivers

1. River Descriptions

141. The Choloki River flows from the confluence of various streams at an altitude of around 780 m. on the north-west slope of the Ilia Tsikhe Mountain and joins to the Natanebi River 0.6 km. from its mouth (outfall). The length of the river is 24 km., the fall of water surface is 780 m., average inclination 32 percent and the catchment area is 159 km2. The average height of the river basin is 210 m. 142. The main tributaries are the Ghela (12 km. length), Sharistskali (16 km. length) and Ochkhamuri (19 km. length). 143. The river basin is located on the north-west slopes of Adjara-Imereti mountain ridge. The upper part of river basin is mountainous relief which is separated by tributaries, ravines and gorges. The components of the mountainous part of River basin are marl, sandstones, shale and conglomerates. The lower part of the basin is conglomerate with sandstone, loam and newer alluvial sediments; the main rocks are covered by loamy soil. The upper basin is Kolkheti type forest while the lower part is characterized by cultivated lands and plantations of subtropical plants. Swamps cover 15-20 percent of the lower basin. 144. The river-bed in the upper and middle regions is moderately sinuous, but on the lowland it is meandering. The flow width changes from 2-4m. to 25 m., the depth from 0.2 m. to 1-3 m., and the speed from 2.5 m/s to 0.3-0.5 m/s. 145. River aqueous regime is characterized by water inflows all year round. Water inflows levels, caused by rains exceed the levels of inflow caused by snow melting in spring. The river flow reduces in the summer. River’s monthly annual flow distribution shows major changes from year to year. The river does not ice over. 146. The Achkva River flows from the confluence of various streams at 1,000 m. on the north-west slopes of Ilia Tsikhe Mountain and flows into the Black Sea near Kobuleti. The length of the river is 19 km., the fall of the water surface 999 m., the average inclination is 52.6 percent the catchment area 37 km2, and the average height of river basin is 156 m. The river has 79 tributaries with a total length 80 km. 147. The upper part of the river basin is located on the north-west slopes of Adjara-Imereti mountain ridge and is separated by tributaries, ravines and gorges. The middle part of the river basin is hilly; the lower part is situated on seashore lowland.

Engconsult Ltd. 26

148. The geological structure is basically conglomerate with sedimentary rocks of tertiary and quaternary periods. The main rocks are generally covered by mountainous and forested loamy soils. The vegetation cover of the river basin is Kolkheti type forest. 149. The river bed is moderately sinuous. The width of flow changes from 2m to 12 m., the depth from 0.2 m. to 1.5 m., and the speed from 1.1 m/s to 0.2 m/s. 150. The river’s aqueous regime is characterized by water inflows all year round. The river does not ice over. 151. The Kintrishi River flows from 2,320 m. on the Adjara-Imereti mountain ridge, near the Khino Mountain (2,599 m.), and flows into the Black Sea one km. south of Kobuleti. The length of river is 45 km., average inclination 52 percent, the catchment-area is 254 km2 and the average height of river basin is 835 m. The main tributaries of river are Magalakhevisghele (12 km. length) and Kinkisha (15 km. length). 152. The whole river basin is mountainous; its geological structure is made up of tufa, andesite, and basalts, with alluvial, diluvia and eluvia sediments. The main rocks are covered by loamy soil and 70 percent of the river basin is covered by thick mixed forest. 153. The river bed is sinuous and, after the village of Khutsubani, is divided into bayous. The bayous form little, low islands, (length 50-1000 m., width 50-200 m.). The width of the river flow changes from 1 to 80 m., the depth from 0.2 to 2 m., and the speed from 1.8 m/s to 0.7 m/s. 154. The aqueous regime of the river is characterized by inundations in spring and water inflows all over the year. The level of water inflows exceeds the level of inundation caused by snow melting in spring. The river flow reduces in the summer. River’s monthly annual flow distribution shows major changes from year to year. The river does not ice over. 155. The Chakvistskali River flows from the Kobuleti mountain ridge, from the southern slope of Titati Mountain (1,379 m.) at a height 1,300 m. and flows into the Black Sea to the south of village of Chakvi. 156. The length of the river is 23 km., the fall of water surface 1,300 m., average inclination 56.5 percent, and the catchment-area 176 km2. The river has 496 various tributaries and the whole length is 337 km. 157. The upper river basin is in mountainous terrain; after the village of Khala it changes into hilly relief. The river’s flow is moderately sinuous and up to the village of Gorgadzeebi has no bayous. After the village the river forms several islands, which are covered by 1 m. height water during high water. 158. The aqueous regime of the river is characterized with floodwaters in spring and water inflows all year caused by rains, the level of these water inflows exceeds the level of inflow caused by snow melting in spring. The river flow reduces in the summer. The river’s monthly annual flow distribution shows major changes from year to year. The river does not ice over. 159. The Chorokhi River flows from Turkey’s, Okus-Badatsagi Mountains, 20 km. south-west of Isfiri Mountain, at a height of 2,700 m., and flows into the Black Sea in Georgia, 6 km. south- west from Batumi. 160. The length of the river is 438 km., catchment area 22 100 km2, the lower section of the river is located in Georgia with a length of 26 km., and the fall of water surface of this section is 780 m., the average inclination 30 percent. In Georgia the river has 3 main tributaries: Machakhelistskali (length 37 km.) Acharistskali (90 km.) and Charnali (13 m.). 161. The river basin relief is mountainous. It covers the northern slopes of the Shavsheti mountain ridge, western the slopes of Arsiani mountain ridge and the southern slopes of the Adjara-Imereti mountain ridge. The lower part of the river basin, about 10 km. in length, is

Engconsult Ltd. 27 located in the Kakhaberi lowland. The mountainous part of the river basin is separated by deep ravines of the tributaries Machakhelastskali and Acharistskali. 162. The geological structure of river basin is a conglomerate of tufa, shale, new andesite and basalt lavas. The main rocks are covered by layers of loamy soil. The vegetative layer is mainly deciduous and conifer forest. In the Kakhaberi lowland are cultivated lands. 163. The river bed from the state border to the village of Kapandiba is moderately sinuous and has 2-3 bayous. After the village the river bed is very sinuous and branched. The bayous have formed islands, with widths from 20 to 100 m. and lengths from 100 to 300 m. Part of the islands are covered by plants and grass. The river’s swift flowing and stagnant flow sections are changing every 500 meters. In the Kakhaberi lowland the river changes river bed frequently. 164. The width of the flow changes from 50 m. (near village of Maradidi) to 120 m. (near village of Makho), the depth changes from 1.5 m to 4.8 m., and speed changes from 0.7 m/s to 2.5 m/s. The riverbed is stony-gravelly. 165. The river is fed by snow, rain and ground waters. Its aqueous regime is characterized by floodwaters in spring and, water inflows in autumn and with low river flows in winter and summer. Heavy inflows in spring starts at the beginning of March, rises to a maximum in May and ends in late July. In August and September flow reduces substantially, but there are rain water inflows occasionally during this period. In autumn there are rainwater inflows which are greater than summer. At the end of November the river flow reduces and stays stable through until March. In spring the flows are 45 percent of annual flows, in summer 25 percent, in autumn 17 percent and in winter 13 percent. 166. In periods of inundations and water inflows the sedimentation of river changes from 3,700 to 11,000 g/m3. The maximum stable (steady) volume flow rate, observed in May, is 3100kg/s, the minimum is in September at 3.0 kg/s. There are short periods of icing. 167. In Georgia the river Chorokhi is not used for cultivation, however there are weirs and the Muratli Reservoir is located near Georgian-Turkish border. 168. Other small rivers and streams have similar aqueous characteristics as the rivers described above. 2. Maximum Flow Rate

169. Hydrological studies have not been carried out for most of the rivers which cross the bypass alignments. Up until 1990 only flows of the Chakvistskali, Kintrishi and Chorokhi rivers had been done. Because of this the maximum volume has been calculated based on the methodology, given in “Technical guideline for calculating rivers maximum flow in Caucasus”4. 170. Using this methodology the water’s maximum volume flow rate with maximal catchment area of up to 400 km2, is calculated using the formula:

125,038,035,13/2 ⎡ τ ⋅⋅⋅ iKF ⎤ 3 RQ ⋅= ⎢ 44,0 ⎥ ⋅Π⋅ λ m /s ⎣ ()L +10 ⎦

Where, R = District parameter; 1.35 for Western Georgia; F= Catchment area river (km2);

4

Engconsult Ltd. 28

K= Climate coefficient of district. For Ajara this is 9; τ = Time period (years) i = River’s flow’s balanced inclination in units, from the source to bridge; L= length of river from source to bridge(km); Π = The coefficient characterizing soil in river basin taken from special maps and tables; λ = Coefficient of river basins tree cover, calculated by the formula: 1 λ = F 2,01 ⋅+ t F

where Ft is area of woody part of river basin. 171. Topographic maps (1:25,000 and 1:50,000) have been used to calculate catchments areas, inclination and length. The resultant maximal volume flow rates are given in Table 12

Table 12: Flow rates of Rivers and Streams.

Maximal volume flow rate m3/sec F L i λ River 2 K Π τ = 200 τ = 10 km km cal. percent τ = 100 years years years

1. Choloki River 83.0 20.2 0.039 9.0 1.19 60 510 440 185

2.Ochkhamuri 38.0 15.0 0.015 9.0 1.19 55 295 255 105 River 3.Shavi Ghele-1 2.97 3.40 0.019 9.0 1.19 98 60.0 50.0 21.5 4. Shavi Ghele 1.92 3.80 0.017 9.0 1.19 98 41.5 35.5 15.0 5.Shavi Ghele -2 0.86 1.90 0.019 9.0 1.19 99 22.0 19.0 8.00 6.Achkva river 30.0 13.5 0.058 9.0 1.19 80 295 255 105 7. stream 1.47 1.88 0.019 9.0 1.19 65 38.0 32.5 13.5 8. Kintrishi River 208 42.8 0.054 9.0 1.19 70 765 660 275 9. Kinkishi River 36.0 14.0 0.089 9.0 1.19 65 360 310 130 10.Dekhva-qveda 41.0 15.0 0.076 9.0 1.19 70 370 320 135 River 11.Dekhva-zeda 39.0 13.7 0.083 9.0 1.19 70 370 320 135 River 12.Shua Ghele 11.2 5.98 0.079 9.0 1.19 75 190 165 68.5 13. stream 4.47 3.50 0.116 9.0 1.19 60 110 95.5 40.0 14. Chakvistskali 167 18.5 0.069 9.0 1.19 75 880 760 320 River 15. stream 2.44 2.80 0.101 9.0 1.19 50 68.0 58.5 24.5 16. stream 2.42 2.65 0.123 9.0 1.19 80 65.5 56.5 23.5 17. stream 1.16 2.12 0.101 9.0 1.19 70 39.5 34.0 14.5 18. stream 3.97 4.40 0.083 9.0 1.19 90 91.0 78.5 33.0 19. stream 1.17 1.65 0.102 9.0 1.19 75 40.0 34.5 14.5 20.Korolistskali 49.5 11.4 0.086 9.0 1.19 55 475 410 170 River 21. stream 5.00 3.75 0.060 9.0 1.19 50 120 105 43.0 22. stream 3.24 4.00 0.055 9.0 1.19 45 77.0 66.5 28.0

Engconsult Ltd. 29

Maximal volume flow rate m3/sec F L i λ River 2 K Π τ = 200 τ = 10 km km cal. percent τ = 100 years years years

23.Bartskhana 14.3 7.75 0.119 9.0 1.19 55 235 200 83.5 River 24.Mejnistskali 7.62 4.80 0.075 9.0 1.19 45 165 140 60.0 River 25.Choroxi river 22100 436 0.030 ______26. stream 0.61 0.95 0.270 9.0 1.0 95 20.0 18.0 8.00 27.KvariaTi Ghele 0.91 1.10 0.364 9.0 1.0 98 30.0 25.0 10.0

3. Chorokhi’s Maximum Flowrate 172. As the River Chorokhi’s catchment area is more then 400 km2, the above methodology cannot be used. 173. Several weirs were built at the end of 20th century on the river Chorokhi in Turkey; one, the Muratali, situated near Georgian state border, is still used. The weir regulates the river’s flow, and changes its flow characteristics in the lower reaches in Georgian territory. 174. In order to make substantial hydrological calculation and to make optimal decisions for project structures it is necessary to know the conditions of weir use. Unfortunately, despite many requests to the operators of the weir, this data has not been forthcoming. Therefore, the hydrological calculations have been carried out based only on data provided through observations on River Chorokhi in Georgian territory. 175. On projected sections of the river the maximum volume flow rate is calculated based on data of historical hydrological measurements based on a 63 year observation period of natural river flow, (1930-1992). In this period the maximal volume flow rate varied from 3,840 m3/s (1942) to 628 m3/s (1986) . 176. Using data from the hydrographical station at Erges, close to Batumi, the maximum water flow was estimated as: ΣQ Q = i = 1,422 m3/s-s; Q0 where Cv and Cs are the coefficients of variation and asymmetry and λ2 and λ3 are factors for vegetation coverage which are defined based on nonographs: Σlg K Σ lg KK where λ = −= 0.028 and λ = = +0.029-s, 2 n −1 3 n −1

Cv = 0.38, and 4 ⋅= CCs v =1.52. 177. Based on the distribution curve’s parameters and the parameters of the distribution data, the maximum volume flow rate at Erges hydrological station is set out in Table 13 for different return periods (years) for a catchment area of 22,000 km2 Table 13: River Chorokhi maximal flow rates

Return period 200 100 50 20 10 5 (Years) QQ m3/sc 3840 3215 2945 2435 2115 1790

Engconsult Ltd. 30

VII. STRUCTURES

A. Design Standards

1. General

178. Lateral highway drainage is mainly through culverts and bridges. The size of the flood opening is determined by the catchment area parameters and consideration of existing nearby structures. 179. There are 111 culverts crossing the bypass roads and these will be designed in compliance with standard design practice for highways using pipe, or box culverts. They are used not only for stream and flood water disposal but also for farmland irrigation where necessary. Sufficient up and down stream training works will be constructed to allow smooth flow. Box culverts shall also be used for minor road crossings and pedestrian underpasses. 180. For the purpose of easy maintenance circular culverts are designed of 1,500 mm diameter. Two size cross section of box culverts are suggested by the preliminary design: 2.5 * 4.5 m and 4.5 * 6.0 m according to the Georgian Road Design Standard. The former is designated for pedestrians and cattle under passing in agricultural areas as well as for small streams in steep valleys, while the latter is for minor roads to enable the local transport underpass the design road. 181. The following standards for bridge design have been used for the preliminary design. These generally follow Georgia and international practice. • Design Load: HL-93 loading for the road way according to the AASHTO standard (1993). • Design Flood Frequency: A frequency of 1/100 years is adopted for bridges; 1/50 is used for small bridges and culverts and 1/20 years for side rains. • Earthquake: The project road is located in seismic 8 score seismic zone by Richter’s Scale (as specified by the decree #42 of Ministry of Architecture and Construction, July 7, 1991), which requires necessary anti-seismic measures in the design of all structures in this section. • Navigation Requirement: There are no navigable rivers on the alignment. • Deck Width: In line with Georgian road design standard for design speed 100 kph, the deck width of all the bridges, both small and large, has to be the same as the roadway width (sub-grade): the bridge deck width is two lanes of 3.5m, 2.5 m shoulders plus a railing edge beam of 0.5 m on both sides, giving a total width of 13.0 m.. As this is an access controlled road the width of footpath has been considered oversized and unnecessary and has been reduced by 0.5 meter. So geometric parameters for all the bridges are assigned as follows: o carriageway – 2 * 3,5 = 7,0 m o safety lane both sides – 0,5 m o footpath both sides – 0,5+1,5+0,5 = 2,0 m (see Fig. 2.1)

Engconsult Ltd. 31

• Vertical Clearance: Minimum vertical clearance is 5.0m. (Georgia Road Standards (2009)).

2. Bridge Cross Sections:

182. The Bridge cross-section is shown in Figure 5

4.0% 4.0% 2.50% qano bi /SLOPE 2.50% qano bi /SLOPE

Figure 5: Two Lane Bridge

B. Bridges in the Project 183. The project has 37 bridges with a total length of 6,643 m (Table 14); the overall bridge length is about 16 % of the total alignment, which means that any refinement to the bridge design and construction in detail design could achieve significant cost savings. Table 14: Summary of bridges

Item Length (m) Number

Small 643 19 Large (>50m) 1,800 12 Extra Large (>250m) 4,200 6 Total 6,643 37

1. Large Bridges

184. Table 15 lists the large bridges in the project with span length/s, super structure types, abutments and pier types. A General Arrangement Drawing of the proposed bridge sites and the span arrangement for most of these large bridges are provided in the design drawings.

Engconsult Ltd. 32

Table 15: Large Bridges

RC Pier RC Super Start Maximum Length Abutments No. Crossing Type/Height Structure Chainage Span (M) Type/Height (m) (M) Type River and

connection Two Circular PC, 1 23.90 45 U-type/10 road to Zeda columns /25 2 web 600 Achkva Riv. Two Circular PC, 2 26.2 45 800 Spill-through /15 Chaqvistskali columns /15 2 web Riv. Two Circular PC, 3 35.85 45 770 Spill-through /15 Korolistskali columns /25 2 web Riv. Two Circular PC, 4 38.5 45 750 Spill-through /15 Bartskhana columns /25 2 web Valley north of Two Circular PC, 5 40.35 45 580 U-type /15 Peria Tunnel columns /35 2 web Valley south of Two Circular PC, 6 42.3 45 700 Spill-through /15 Peria Tunnel columns /35 2 web Total 4,200 Note: PC 2 web – Reinforced concrete continuous girder with 2 webs and integral reinforced concrete deck prestessed in longitudinal and transverse directions. C. Technical and Design Summary: Bridges

1. General Design Principles

185. The proposed expressway traverses both flat and rolling/mountainous areas. The proposed bridges will mainly be aligned in the general direction of the road. Large bridges were subject to visual geological condition investigation and comparative alignment studies to determine their final location. 186. Normally bridge structures are used in preference to fill for those alignment sections with a fill height of more than 20m. However, some bridges in easy terrain fill height of over 20m could be changed to culverts with fill sub-grade so as to reduce the waste material from cuts. This should be examined during detail design. 187. Generally bridge location and structure type have been designed following the principle of “suitability, economy, safety and landscaping”. Bridge design has also taken account of environmental-friendly design and ease of construction and maintenance. 188. The spans and length of the various bridges have been determined by integrating the topographical, geological and hydrological conditions with the bridge elevation, longitudinal grade and radius of horizontal curves. The span of bridges in mountainous areas was also, where appropriate, determined by flood levels. The elevations of the bridge decks in the flatter areas of the road are mostly governed by the clearance required for debris and the existing bridge levels. The location of the large bridges have also been chosen where the watercourse is straight, where possible in a stable geological area, and where the maximum designed flow can be contained in its channel. a. In order to save investment and improve the efficiency of assembly and construction, bridge types have been standardized where possible.

Engconsult Ltd. 33

a. Super-structure:

189. The superstructure for the bridges has been selected depending mainly on the height of substructure, use of the local materials, economy, and local expertise in design and construction. Taking into account the topographic and geologic conditions of the alignment, bridge superstructures are proposed as: b. prefabricated reinforced concrete T-girders – for single span bridges (12 meter or 18 meters) c. prefabricated, free supported pre-stressed T-girders – for small bridges up to 250 meter in length d. Cast in situ pre-stressed concrete continuous girder with 2 webs and integrated deck slab – for long bridges over 250 meter in length and over 20 meter in height. 190. The length of a bridge is mainly dictated by the width of the natural drainage channel and the height of fill. These superstructures have been standardized so as to facilitate pre-cast construction, reduce construction cost and also speed up the construction progress. 191. The type of bridges used for the preliminary design is also based so as to maximize the use of local materials and thus economize on the cost of the bridges. The span used for super- structure of the bridges is mainly based on the height of the substructure from the foundation level. The height of the bridge deck level is fixed based on the High Water Level (HWL) plus freeboard based on the Georgian Road Design Standard. The Bridge deck level is the higher of the deck level as obtained above or the design level of the road demanded by the vertical design requirement of the road. 192. For the Preliminary Design, the depth of the T Beam and Box Girders are taken as 1/16 of the Span. The width of Bridge is as shown in the Bridge Cross-Section in the report. b. Substructure:

193. For pier substructures, columnar piers are proposed. For abutments up to 10 meters in height inverted U-type of abutment is proposed; for those higher than 10 meters – spill through type of abutment. 194. The Foundation type for substructures is decided on the geo-technical investigation carried out. The soil investigation suggests that Spread Footing for the foundation can be used for most bridges except in the first 14 km. For the preliminary design purpose spread footings have been used for the bridges from km 14-44. For long and high bridges the foundation level is fixed where the N-value is more than 50 and below the Scour level. For small bridges foundation level is taken respective to the N=30 value. For the bridges where the soil investigation has not been carried out, the assumption is based on a similar type of investigation data. The Allowable bearing pressure based on the N value of 30 for non-granular soils for spread footing is taken as 30-60 t/m2 and for granular soil it is taken as 30 t/m2. For first 14 km of the design road, in peaty soil bridges are designed with pile foundations with 15 -20 meters in depth based on preliminary calculation of geotechnical capacity of piles. 195. Reverse T type of Abutment has been used for a height of up to 12 m. Box(spill-through) type of abutment has been used from 12 m and up to 20m height. 196. Wall type of columnar piers have been assumed for the piers of the Bridges.

Engconsult Ltd. 34

2. Other General Bridge Design Issues

197. As there are more than 50 bridges in the Project, for design drawing and cost estimation purposes they have been grouped broadly by span length and arrangement. General arrangement drawings for each span length and arrangement have been produced at a scale of 1:500 and 1:1,000 for profile and plan respectively and at 1:100 and 1:200 for cross-sections. Typical drawings for small cross drainage such as circular culverts, box culvert, pedestrian crossings and overpasses have not been included in the drawings. 198. In the detail design stage, additional detail geological condition investigation will be required for all bridges, particularly large and extra-large bridges before finalizing the bridge type and span layout. 199. Bridge deck drainage system design and pavement improvement are important for high class road operation, as poor design or construction can lead to additional maintenance and cost. Early damages have occurred of deck drainage and pavement on some highway bridges in Georgia. Most designers and contractors correctly concentrate on structural safety; however, this does not mean that careful proper design and construction of accessorial parts of bridge should be ignored. These details need to be carefully considered in detail design. D. Interchanges

1. Introduction

200. In the 48 Kilometers from the northern end of the project road at Natanebi River to the Chorokhi River connection with the existing road, there are five full and two partial interchanges proposed. The interchange names and locations are given in Table 16. Spacing of interchanges ranges from 3 - 17 km. 201. The interchanges will be designed to connect to existing major roads and also to provide access to areas with development or possibility of future, industrial or residential development. 202. During design account has been taken of: • Traffic flows leaving and entering the bypass road; • Intervals between interchanges; • Local traffic needs; • Classification of the interchange in terms of design speed and ramp widths; • Land use, interference with existing facilities and costs; Table 16: Interchange Names and Locations

Type of Connecting Road Chainage (K) Interchange Interchange Road Name Road Class 0 Natanebi Kobuleti-Poti International Trumpet Kobulheti- 11.3 Ochkhamuri State Diamond Azurgeti Modified 28.3 28km Existing S2 State Trumpet Partial 33.8 Batumi Existing S2 Town access Directional 39 Bartskhana Batumi access Town Road Partial Batumi- Modified 44.6 Khelvachauri State Khelvachauri Cloverleaf 47.8 Chorokhi Batumi-Sarpi International Directional

Engconsult Ltd. 35

203. In all cases, the taper lengths, acceleration and deceleration lengths are at or above those required by international standard for the selected design speeds. The width of one-way ramp is 4m with 1.5m shoulder. The interchanges have been design for a two lane main road but with allowance for future upgrading to four lanes. Designed ramp widths for and entry and exit lanes have adequate capacity for the road design life. All interchange bridges will have a vertical clearance of 5m. 204. Landscaping is proposed to be integrated with the design of interchanges and service areas. 205. Interchanges have been designed to cater for forecast traffic volumes and the existing and future planned road network and development. Land for all future interchanges upgrading should be reserved. 2. Interchange Types

206. In order to minimize land acquisition, where ramp traffic volumes are small diamond interchanges have been selected (Ochkhamuri). For the major connections with the existing road at Natemebi, 28km and Chorokhi a modified trumpet and a directional interchange have been selected. At the junction of the Bartskhana where the terrain is unsuitable and land developed a partial interchange is proposed with just two ramps. At Khelvachauri road where traffic volumes are higher or are expected to grow because of development, a modified partial cloverleaf interchange is planned. At the junction with the existing road at km 28 and at the start of the Batumi bypass where the existing road will split to enter Batumi a partial directional interchange is planned. 3. Interchange locations

a. Natanebi

207. The interchange is at the intersection of the project road (Kobuleti bypass) and the existing Poti-Kobuleti road and will provide uninterrupted traffic flow at the intersection of these two major roads. It will cater for traffic from the north wishing to travel via the bypass or go directly to Kobuleti and vice versa. b. Ochkhamuri

208. Ochkhamuri interchange is located at 11.3 km from the northern end of the bypass. The purpose of this interchange is to cater the traffic coming and going to Azurgeti and central Kobuleti. It is anticipated that interchange will initially have light traffic and so a diamond interchange is planned with stop signs or traffic signals at the end of the ramps on the Kobulheti-Azurgeti road. Traffic coming from the south from Batumi or beyond to the center of Kobuleti would also use this interchange. c. 28km

209. The interchange is located at 28.3 km where the bypass road joins the existing, improved, road. This is the start of the existing road widening to four lanes. A modified trumpet type of interchange is proposed and will provide uninterrupted traffic flow for traffic from Batumi to Kobuleti as well as bypass traffic.

Engconsult Ltd. 36

d. Batumi Interchange

210. An interchange is required at km 33.8 where the proposed Batumi bypass leaves the (four lane) existing road. It will allow uninterrupted traffic flow for traffic wishing to enter Batumi or for through traffic using the bypass. A semi directional interchange is proposed; traffic from Batumi will not be allowed to turn back onto the bypass and traffic from the bypass will not be able to turn back to Batumi. This is necessary because of limited space. A bridge is required on the bypass road as it climbs to enter the first of the small tunnels and the Batumi road (northbound) would go under this bridge. e. Bartskhana Interchange

211. A partial interchange is proposed at 39km at the Bartskhana River road for traffic to and from the south wishing to travel from or to north Batumi. the rail yards and port. This is a busy congested road and the area is being developed so traffic volumes will rise. Because of the terrain and existing development, however, this has been designed with just two ramps to cater for the traffic to and from the south only and may be upgraded later to a partial cloverleaf. However, this may be difficult unless land is resumed or frozen. With improvement o the existing road from the interchange west towards the rail yards (1.6 km) a connection to the proposed Batumi relief road can be made which would attract more traffic from the port and railyard areas to the bypass. f. Khelvachauri Interchange

212. A cloverleaf interchange is proposed at 44.6km at the Batumi-Khelvachauri road giving access to east Batumi and to the Khelvachauri road for traffic to Keda, Suakhevi and beyond. This is a busy major road with probable future growth thus a diamond interchange is not suitable. It may be possible to construct a partial cloverleaf at first and to upgrade to full cloverleaf as traffic volumes increase. 213. The preliminary layouts of the interchanges are shown in the drawings in Volume 3.

Engconsult Ltd. 37

VIII. OTHER PROJECT ROAD ELEMENTS

A. At Grade Intersections 214. Two fully access controlled bypass roads are planned with the only access via grade separated interchanges. The ramp roads to two of these interchanges will connect into the local road network at grade intersections. The design of these intersection need care since the traffic from the project road will have been traveling at high speeds and must be given clear warning of the lower standard of the road and intersection ahead. B. Roadside Facilities

1. Signs and Markings

215. All traffic signs and road markings will conform to GoG or Vienna Convention standards for international roads. Detailed schedules for signs and markings are not shown in the preliminary design but would normally be shown in detail design. 216. Typical road marking layouts for the main line and ramps should be detailed in the final design. Both road marking and signs should be painted with reflectorized paint in accordance with international standards. 217. The project road will require continuous communication and monitoring facilities based on international specification and suitable communication and monitoring facilities for the tunnel and variable message signs. 2. Roadside Service/Information Stations

218. Service areas are usually but not always provided at interchange locations on high class roads. It is proposed to include a service area close to Ochkhamuri interchange. This should provide a good level of service for motorists and therefore unauthorized stopping on the Project road should not be necessary. The service area should contain the full range of services, including fuel, food, parking and toilet facilities and possibly overnight accommodation for truck drivers. Space should be reserved for a future weigh bridge area. Road maintenance facilities should also be located at the service area.

Engconsult Ltd. 38

IX. TUNNEL DESIGN

A. Introduction 219. The preliminary tunnel design has been developed using the preliminary topographical and geotechnical investigations. The magnitude and cost of ventilation and illumination has been estimated from past experience. In the Detail Design stage additional geotechnical investigation and further study for exhaust gas emission properties of heavy vehicles will be required to upgrade the tunnel design and E&M design. B. Standards and dimension of tunnels 220. The Georgian design standard for two way tunnels, initially required by RD to be used in this project, has an emergency lane on one side of 2.5m wide with a clear width of the road of 11.5m (Figure 9). 221. This is an overly large tunnel cross section compared to most international standards and will create difficulty in construction in poor ground as well as far more construction time and cost (43 percent greater) than normal international standards. 222. The twin tube tunnels on the existing road, which will be part of the Kobuleti by-pass road have a cross section which is close to normal international standards and differ significantly from the Georgian Standard because the tunnel does not have the extra 2.5m lane. One of the twin tunnels has 3.5m wide two lanes with 0.7m wide shoulders and inspection walkways at both sides and the other has 3.75m wide two lanes with 0.5m wide shoulders and 1.1m and 1.2m wide inspection walkways. 223. When there are different tunnels in the same section of road whose cross sections differ too much, traffic safety cannot be maintained and this may cause a serious risk. For the safety of the traffic and for safe and economical construction, a smaller internationally accepted cross section is recommend for the project tunnels, which, for two lane two way tunnels, has two 3.5m wide lanes, 1.25m wide shoulders and 0.9m wide inspection walkway at a side. After further discussion with RD it was agreed to use this cross section. C. Tunnel Design

1. Tunnel Cross Section

224. The ttypical Standard cross section for a 2 lane two way tunnel is shown in Figure 6. For the portal areas and for sections of weak geotechnical conditions, where unconsolidated material develops, invert concrete has been specified as shown in Figure 7. 225. The width of the road in tunnels is 2 x 3.5 m lanes with a 1.25m wide shoulders at both sides (for the safety of two way traffic; this can be decreased to 75cm wide when one way tunnels are designed) and a walkway, 0.9m wide and 2.0m high, for inspection, maintenance and emergency. Tunnels have a clear height of 4.5m from the road surface with an extra 20cm allowed for pavement overlay. 226. No footpath is designed, because for safety reasons non-motorized traffic and pedestrians are not allowed to enter the tunnel. Thus, the total horizontal clearance of the tunnel is 9.5 m and height is about 7 m. 227. Outside tunnel dimension to be excavated is approximately 12 m wide and 8 m high. 228. When the length of the tunnel is more than one kilometre, emergency parking zones are generally designed. These are located at intervals of about 500 m to 1 km alternately on both

Engconsult Ltd. 39 sides. However, in this project tunnels are less than 700m and no emergency parking area is required. 229. Approximate cross sectional areas for respective cross sections are shown in Table 17. Table 17: Approximate Tunnel Cross Sectional Area

Tunnel type Excavation area m2 Type 1 and 2 81 Type 3 93.5

Figure 6: Cross Section Georgian Standard (Type 1)

Engconsult Ltd. 40

Single tube two Lane two way Two Lane one way (for double tube)

Figure 7: Adopted Cross Section Type 1

2. Tunnel Support Design

230. Tunnel supports are generally designed in accordance with the quality of rock mass by adopting empirical design methods such as RMR, Q-values or Japan Highway Public Corporation’s Classification (JH system). These systems are correlated to each other. Design for support will be re-examined during the detailed design stage, however, for the preparation of this preliminary study, simplified support design is specified by assuming three types of rock class; rock mass in shallow depth (Type 3); rock mass of moderately weathered to moderately hard with fractures (Type 2); and moderately hard to hard and massive rock mass (Type 1). 231. Rock mass in shallow depth (Type 3) corresponds to a highly to completely weathered tuffaceous rock mass. Rock mass of moderately weathered to moderately hard with fractures (Type 2) corresponds to moderately weathered tuff breccia to moderately hard tuff breccia/basaltic rocks with fractured natures. Moderately hard to hard and massive rock mass (Type 1) corresponds to tuff breccia/basaltic rocks. 232. In accordance with the classes of above rock mass typical Support Patterns are predetermined as shown in Figure 8 (Type 3) and Figure 9 (Type 1) 233. Major tunnel support components are rock bolts, shotcrete and steel arch support, and secondary concrete lining is placed in-situ for the safety, serviceability and long term durability. Shotcrete is used with or without wire mesh. 234. Type 3 Pattern consists of 20cm thick shotcrete, forepoling rockbolts in the arch section, 6m long rockbolts in side walls and H200x200 steel arch support and 40cm thick secondary lining. Type 1 Pattern consists of 10cm thick shotcrete, rockbolts and 40cm thick secondary lining. Type 2 Pattern is an intermediate support pattern to be used in transitional portion from Type 3 to Type 1, which may or may not use steel arch support depending on the actual ground condition. 235. In cases when tunneling encounters difficult ground, a combination of extra auxiliary supporting methods such as fore poling along with arch, face bolting, long span fore-poling, chemical injection and other particular methods are applied to stabilize the excavation periphery. In this study fore-poling is designed for certain lengths of tunnels and is included in cost estimates. 236. Due to the aggressive nature of the rock mass and groundwater, anti-sulphate cement shall be used for the secondary lining concrete.

Engconsult Ltd. 41

Figure 8: Support Pattern Type 3

Figure 9: Support Pattern Type 1

Engconsult Ltd. 42

3. Project Tunnels

237. The particulars of tunnels are shown in Table 18 and geology and geotechnical condition of respective tunnels are set out in Section E below.

Table 18: Particulars of Tunnels

max. Vol. of Location Length gradient Tunnel Portal elevation overburden excavation km to km m % m m3 Zeda Achkva 22.1-23.5 400 N:83.96 S:81.85 0.05 53 22,045 Makhinjauri I 33.9-34.1 130 N:14.0 S:17.15 2.4 21 10,222 Makhinjauri II 34.1-34.3 90 N:20.32 S:22.0 1.9 23 5,082 Makhinjauri III 34.3-34.5 150 N:22.0 S:24.53 1.7 18 9,672 Salibauri Ridge 37.6-38.1 440 N: 63.34 S:67.35 1.0 41 36,265 Peria Ridge 41.2-42.0 700 N:104.4 S:85.06 2.6 72 62,382 Sum 1,910 146,667

4. Method of Tunnel Excavation

238. Construction method is selected taking into account the geotechnical condition of the rock mass in the tunnel. Generally speaking, a Drill & Blasting method provides fastest progress where it is applicable and where there is no regulatory difficulty in using explosives. 239. Mechanical excavation by Road-Headers is a safe excavation method. When hardness of rock mass doesn’t exceed the capacity of Road-Headers, fairly good progress rate can be expected. Tunnel Boring Machine is not applicable in the project due to the variable nature of rock mass and uneconomical shape of the tunnel being excavated. 240. Most tunnels in the project may be in moderately hard to moderately soft volcanic to pyroclastic rocks except in portal areas and may be excavated either by Road-Headers or by Drill & Blasting methods depending on the hardness of rock mass in particular tunnels, by adopting short-bench-cut excavation method. Due to the large cross section of excavation full face excavation method is not applicable and in portal areas where rock mass is completely weathered and weak, excavation shall be carried out by top heading-and- bench-cut method. 241. Road-headers are recommended to use for the Zeda Achkva Ridge tunnel and the short tunnels A, B and C because the rock mass is somewhat soft consisting of tuffaceous rock. Giant breakers may be required for some portion of the cross section and for some distances along the tunnel to cope with harder rock mass which will probably be encountered. The soft rock portion in the portal areas can be excavated by power shovels. Tunneling may be done by short bench excavation method to stabilize the excavation faces. However, the construction method should be further examined when further geotechnical investigations are carried out during detailed design. 242. The Salibauri and Peria Ridge longer tunnels may be excavated by Drill & Blasting method due to the hard nature of the rock mass, with the help of power shovels for the excavation of portal areas. Tunneling may be done by short bench excavation method except portal areas.

Engconsult Ltd. 43

5. Tunnel Portals

243. The portal zone of a tunnel shall be designed with due consideration of the ground condition, condition of the slopes, surface drainage system and topographic condition. When tunnel portal is located at an acute angle to the slope, filling of the side of the tunnel portal on the valley side becomes necessary in order to avoid the heterogeneous loading action of the ground. 244. Tunnel portals may be designed as concrete shell structures extending 3 to 7 m out from the slope to protect the portal area from landslides and rock falls. Selection of exact locations and types of portals shall be carried out during detailed design stage.

Engconsult Ltd. 44

X. FACILITIES FOR TUNNELS

A. Ventilation System 245. Tunnel ventilation design is controlled by the volume and vehicle mix of traffic, the length and gradient of tunnel, and climate and natural condition. 246. Generally a ventilation system is required when the length of the tunnel is longer than 500 m. However, if the traffic volume is small it will not be required for tunnels of up to 1 km long. 247. The purpose of ventilation is to reduce the concentration of harmful gasses and particles such as CO, C, HC, NOx etc. When ventilation to reduce CO concentration is carried out to the designed target level, other harmful gasses and particles also decrease to a safe level. On the other hand, soot generated by heavy vehicles generates visibility problems and ventilation design should also satisfy visibility requirements. Design targets for ventilation are given in Table 19 and are based on international experience. Table 19: Design Target of Ventilation

Object of ventilation Design target CO concentration 100ppm Visibility reduction 50% at 100m

248. Ventilation is normally done by a series of jet fans which are fixed to the concrete arch lining at adequate intervals. However, taking into account the maximum traffic volume of between 500 and 1000 vehicles/hour after 20 years of operation and that the length of the tunnels are less than 700m, ventilation is generally not required. This is shown in Figure 10 which sets out the guideline for the effect of natural ventilation in a two way traffic tunnel1. Beneath the dotted line in Fig.3.7 no mechanical ventilation is required except some in long tunnels. With one way traffic, natural ventilation is much more effective due to the traffic movement effect and mechanical ventilation may not be required up to three times longer than two way traffic. This is well observed in the existing twin tube tunnel. One tunnel has a set of ventilation fans in two locations but the other has no ventilation equipment but the air is still clean. The decision as to whether to install ventilation should be further investigated in detail design when tunnel length is finalized and traffic volumes updated. 249. Due to the relatively short length of tunnels and small volume of traffic ventilation system has not been or costed in this preliminary study. B. Illumination 250. For the safety of the traffic illumination is designed to be installed along the tunnel side walls at about 10 m interval with the addition of transitional illuminations at both tunnel portals. Illumination is generally required for the tunnels longer than 50 m. The detail required will be determined in the detailed design stage but the magnitude of illumination required in this preliminary design has been estimated from past experiences. C. Emergency Facilities 251. For traffic safety, emergency facilities are planned in accordance with the length and gradient of the tunnel and volume of traffic. Tunnels are categorized with respect to length and volume of traffic as shown in Figure 11. This has been based on the probability of accidents and fires based on past experience in a number of countries. Using the tunnel classes

Engconsult Ltd. 45 categorized in Figure 3.8 standard for emergency facilities is determined and is shown in Table 202.

Figure 10: Guideline for natural ventilation (two way traffic)

Figure 11: Tunnel Classification

Engconsult Ltd. 46

Table 20: Standard of Emergency Facilities of Road Tunnel

252. In this study maximum traffic volume is less than 5,000 vehicle/day and tunnel length is less than 1 km. Thus, tunnels are classified into D which requires no emergency facilities. D. Construction and Operation Risks

1. Risks in Tunneling

a. Risks Common in Tunneling

253. Tunneling is an underground activity where geotechnical conditions generally are not foreseen precisely in the design stage and tunneling encounters various changes of underground conditions during implementation, which leads to changes in cost and time problems as well as safety problems. 254. The stability of the excavated faces generally depends on the quality of rock mass and influence of groundwater. In specific ground where rock mass is decomposed by hydrothermal alteration, squeezing may occur and extra time and cost may be required to cope with the phenomenon. 255. While the quality of rock mass differs in places, when no serious phenomena are anticipated such as squeezing, rock burst, gas emission and sudden groundwater inflow, then risks due to change of rock quality is not a big issue provided it is assessed adequately at the design stage. Misjudgement of quality of rock mass leads to the shortage of cost and time. b. Particular Risks in Project

256. In the project, particular risks are the danger of changed underground conditions which may be accompanied by the appearance of hydrothermally altered rock mass. Although it cannot yet be determined as to whether such condition exists in the project tunnels it should become more apparent when further study is carried out during the detailed design stage.

Engconsult Ltd. 47

2. Risks and Targets in Operation and Maintenance

a. Operation and Maintenance of Facilities and Emergency Response

257. A permanent reliable electric power supply for sole use of the tunnel is necessary for the operation of the ventilation system (if required) and illumination. The capacity must cover that of existing tunnels. An emergency generator system is also required for minimum operation of facilities in the event of main power failure. Tunnel facilities require proper maintenance immediately after tunnels are opened for traffic. In cases of emergencies, fire and accidents, immediate responsive actions are required. b. Safety of Traffic

258. Many vehicles in Georgia are defective, poorly maintained and old. If vehicles accidentally stop inside the tunnel then both head on and rear end collisions may occur. To maintain traffic safety, strict control of unacceptable vehicles from entering the toll road is required. E. Site condition and recommended method of construction of tunnels

1. Introduction

259. The geology along the by-pass road mainly consists of Tertiary (Eocene) igneous rocks. In the northern part of the project tuffaceous to brecciate basic rocks are distributed, which are deeply weathered with gentle slopes. In these area slope failures have been observed in places, however, many of them are minor. 260. On the other hand the southern area of the project consists mainly of basaltic rocks and hard tuff breccia, where rock masses are exposed in places with some are forming steep cliff faces, in particular seen along the coast line near Sarpi. In these areas rock falls are a particular phenomenon and adequate slope stabilization measures need to be applied at portal areas. 261. All the tunnel sites except the Salibauri tunnel5 were visited based on the preliminary alignment. Topographic conditions of the portal areas as well as geotechnical conditions were examined. Further examination was carried out using the geotechnical investigation result (tomography) and topographical survey result so far available. 262. Followings are the results of observation for each tunnel and recommendation on the method of tunnel excavation; 2. Zeda Achkva Ridge Tunnel (km 23.1-23.5)

263. The north portal is located in a gentle hillside slope at km 22.9. The portal faces normal to the topography and no specific difficult is foreseen in construct the portal structure. 264. The south portal is located in the hillside with the portal apparently located at an acute angle to the slope. It is recommended that, in order to construct a stable portal, a minor cut in the hill side and embankment in the valley side should be designed during the detailed design stage.

5 The need for the Salibauri tunnel was not realized until late into preliminary design after surveys were finished and the tunnel specialist had completed his field work. Later site visits by the road engineer ascertained that conditions were similar to the nearby Peria Ridge tunnel.

Engconsult Ltd. 48

265. The geology of the tunnel seems to consist of tuffaceous rock which is weathered to deep into the ground and both portal areas may require a longer tunnel section equipped with invert concrete. The remaining tunnel section may require a support pattern corresponding to that of good to poor rock mass condition. The tunnel can be excavated by Drill & Blasting method or by Road-Headers with a short bench cut excavation method. The final decision as to whether to use Road-Headers should be made in the detailed design stage after obtaining further geotechnical information. The portal areas may be excavated by power shovels. 3. Three short tunnels (Makhinjauri I, II, III) between km 33.9 and 34.7

266. All three tunnels are located in gentle hilly area and portals are located in small valleys. Overburdens of these tunnels are quite small and tunnel construction may significantly affect the stability of the surface ground. Geology seems to be of tuffaceous rock which is deeply weathered and tunnel construction in such geotechnical conditions requires heavy tunnel support which leads to high construction cost. The alignment should be further reviewed during detail design to minimize the environmental hazards and construction cost. 267. The North portal of Makhinjauri I is located in the middle of the slope where the tunnel penetrates into the hill at a nearly normal angle. The South portal of the tunnel is located in a small gentle valley. Because the depth of weathering of the rock mass seems to be very thick and the overburden is quite small, the tunnel level may need to be lowered by several meters to maintain the surface vegetation. Moreover, there is a ruined church close to the tunnel alignment under thin overburden therefore the tunnel support should be rigid to minimize any possible settlement of the church. For this purpose long span fore-piling is recommended. Tunneling may be done by short bench cut method with Road-Headers or power shovels. Due to small overburden and soft nature of the rock mass expected, more than half length of the tunnel should be equipped with invert concrete lining. 268. Makhinjauri II is located in a gentle slope. The North portal is located in the middle of the slope and the south portal is also located in a small valley slope. The rock mass seems to be deeply weathered and tunnelling may be done by short bench cut method with Road-Headers with the assistance of power shovels in portal areas. Short bench cut excavation method can be used. Due to small overburden and soft nature of the rock mass expected, about half length of the tunnel shall be equipped with invert concrete lining. 269. Makhinjauri III is located in a gentle hill where a stable slope extends west towards the sea. The rock mass can hardly be seen in this area and seems to be completely weathered to a significant depth. Tunnelling may be done by Road-Headers by short bench cut method with the assistance of power shovels in portal areas. Occasional appearance of harder rock mass can be anticipated and rock-breakers may be helpful in such occasion. Due to small overburden and soft nature of the rock mass expected, tunnel shall be equipped with invert concrete lining for most of its length. 4. Peria RidgeTunnel (km 41.3 - 42.)

270. The north portal is located in the middle of a gentle slope where the ridge extends. The south portal is located in a steep slope below the saddle. Both portals initially faced nearly normal to the slope. However, the south portal was shifted to the east in order to avoid houses on the alignment. 271. Geology here consists of tuff breccia which is weathered to a depth of several meters. Except in the portal areas the geotechnical condition of the tunnel should be good because of the massive nature of the rock mass and tunneling may be done by full-face excavation method using Drill & Blast.

Engconsult Ltd. 49

5. SalibauriTunnel (Km 37.6-38.1)

272. The need for the Salibauri tunnel arose because of a realignment of the road to avoid a historic house and land. This was not realized until late into preliminary design after surveys were finished and the tunnel specialist had completed his field work. Later site visits by the road engineer ascertained that geotechnical conditions were similar to the nearby Peria Ridge tunnel and the same construction methods should apply. 273. The north and south portals are located at the bottom of steep slopes. 6. Existing twin tunnels at km 30

274. In the midway of the project there are existing twin tunnels, which were opened to traffic in 2007. Both tunnels are nearly 800m long. The north portals are located in the cut slope where final protection of the slope is not yet finished. The south portals are located beneath the slope which has been protected by earth anchors; further protection is required. Overburden is relatively small and the tunnels seem to have been excavated in the soft rock mass without much difficulty. 275. There are inspection walkways at both sides of tunnels and illumination is installed. 276. Ventilation fans are fixed in south bound tunnel only. Two sets of jet fan are fixed in two locations and are working well. However, the other tunnel is without jet fans and seems to be unpolluted by gasses – although this may change in the summer tourist season. Because each tunnel is one way traffic natural ventilation works well and, considering the length of the tunnel, mechanical ventilation generally should not be required. The mechanical ventilation system in south bound tunnel, therefore, may be redundant. 277. There appear to be no other installations or emergency facilities.

Engconsult Ltd. 50

ANNEX A

TOMOGRAHPY PROFILES

Engconsult Ltd.

ANNEX A: TOMOGRAPHY PROFILES A nnex A Geoelectric Profile of the Design Tunnel- Legend Scale: H- 1:1000; V- 1:500

p=120-130 Specific Electric Resistance of the Geophysical Layer(GPL), ρ - Om.m

Engconsult Ltd.

p=120-130 I GPL- Loam and Clay with volcanogenic-sediment fragmental inserts of various size; intensively weathered out zone (GPL-1) of main strata. Water saturated

p=20-50 II GPL- Splintered, weathered and saturated volcanogenic and sediments strata (tuff breccias, Tuff sandstones, tuffs) (GPL-11)

P=85-110 III GPL- sounder saturated volcanogenic and sediments strata (tuff breccias, Tuff sandstones) (GPL-III)

Layout legend

I I’

Geophysical Profile with numerical indication

v-7

Point of Vertical Electrical Profiling (VEP)

Project Road The following pages set out the tomography results. 51

A

nnex A

A. Zeda Ackva Engconsult Ltd. 52

A

nnex A

Engconsult Ltd.

53

A

nnex A

Engconsult Ltd.

54

A nnex A

Engconsult Ltd.

55

A nnex A

Engconsult Ltd.

56

A

nnex A

Engconsult Ltd. B. Makhinjauri 57

A nnex A

Engconsult Ltd.

58

A nnex A

Engconsult Ltd.

59

A nnex A

Engconsult Ltd.

60

A nnex A

Engconsult Ltd.

61

A nnex A

Engconsult Ltd.

62

A

nnex A

C. Peria Ridge Engconsult Ltd. 63

A nnex A

Engconsult Ltd.

64

A nnex A

Engconsult Ltd.

65

A nnex A

Engconsult Ltd.

66

A nnex A

Engconsult Ltd.

67 FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 3 Project Preliminary Cost Estimates and Schedules

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Contents

I. INTRODUCTION ...... 1 II. COST ESTIMATES ...... 2 A. Introduction ...... 2 III. Costing Details ...... 3 A. Materials, Labor and Equipment...... 3 1. Materials ...... 3 2. Labor & Equipment ...... 3 B. Quantity Estimate ...... 4 C. Unit Rates ...... 4 D. Land and Resettlement ...... 8 E. Consulting Services ...... 8 F. Maintenance and Operation...... 8 1. Road Maintenance ...... 8 2. Tunnel Operation and Maintenance ...... 9 IV. Pricing Components ...... 10 A. Foreign Component ...... 10 B. Taxes and Duties...... 10 V. Cost Estimates ...... 11 VI. PROCUREMENT PLAN ...... 16 A. Introduction ...... 16 B. Contract Packages ...... 16 C. Mode of Procurement ...... 16 VII. Implementation Schedule ...... 18 A. Introduction ...... 18 B. Comments on Schedule ...... 18

List of Tables

Table 1: Unit Construction Cost for Tunnel Types ...... 5 Table 2: Total Construction Cost for each Tunnel ...... 5 Table 3: Unit rates (GELs) ...... 8 Table 4: Operation Cost of Tunnel Facilities ...... 9 Table 5: Maintenance Cost of Tunnel Facilities ...... 9 Table 6: Detail Cost Estimate – Total Project ...... 11 Table 7: Detail Cost Estimate – Tranche 1 ...... 13 Table 8: Proposed Contract Packages Tranche 1 ...... 17

List of Figures

Figure 1: Tranche 1 Schedule ...... 19

Engconsult Ltd. 1

I. INTRODUCTION

1. This Supplementary Appendix sets out the methodology for calculating the preliminary design cost estimate. It discusses the derivation of unit rates and quantities and the source and suitability of local materials where available. Based on the overall cost estimate suitable contract packages have been suggested and a construction schedule given. 2. This report set out costs based on quantities based on the final preliminary design cross sections and bridge locations and type and are prepared for the economic and financial analysis (Supplementary Appendix 7 and 8). They are based on an analysis of prevailing unit rates for each cost item and an estimate of quantities based on completed preliminary design. 3. Cost and schedules have been estimated for a two lane road from 0-28km (Kobuleti Bypass), widening to four lanes from 28-34km and two lanes from 34-48km (Batumi Bypass). Land acquisition quantities have assumed a future widening to four lanes.

Engconsult Ltd. 2

II. COST ESTIMATES

A. Introduction 1. The break down of unit costs of the major measurable items such as earth excavation, sub- base, base, asphalt, side drains, minor structures of various types etc have been estimated based on prevailing unit rates quoted and used in various contracts of similar projects in Georgia and the quantities of these items estimated from the design drawings prepared based on the survey data. 2. Unit rates and costs for items new to Georgia, such as construction of tunnels have been derived from rates and costs in other countries, adjusted, where feasible for local conditions. 3. Cost of items which are, at this stage, immeasurable; slope stabilization, bio- engineering, insurance, design consultancy services etc. have been estimated on a lump sum basis, based on the experience of other similar projects. Bio-Engineering and slope stabilization in themselves include many smaller cost items including tree plantation, mulching / brushing, grass planting, shotcreteing slopes, retaining walls, boulder pitching, peripheral drains, horizontal drains, scour protection etc. The final appropriateness and magnitude of these items depend on the site condition, which will be apparent only after roadway excavation/embankment filling is carried out. Therefore, these items are estimated for the preliminary design from site inspection, survey and plotted cross-sections. 4. Overhead items such as insurance of works, plant, construction equipment etc are also immeasurable until final design and have been estimated from similar works elsewhere on a lump sum basis. 5. During detailed design and project implementation it is probable that refinements of the alignment will be made leading to a variation in quantities and cost. Whilst this may reduce the overall estimate it has been assumed that it would lead to an increase in cost and a contingency of 10 percent has been included in the cost estimate. 6. Costing has been sub-divided into categories of Civil Works (earth works, subgrade, pavement, structures, tunnels (including ventilation and lighting), interchanges and roadside facilities), Miscellaneous Items, Equipment, Land and Resettlement Costs, and Consulting Services. Equipment costs include monitoring and communication equipment, tunnel lighting equipment, and also include the cost of a weighbridge to be installed at the service area. Preliminary environmental mitigation costs include those preliminary costs recommended in the Environmental Assessment that are not included in other items of the estimate. Consulting Services for detail design and supervision include overseas and domestic training as recommended in Supplementary Appendix 5B. Consulting services for detail design and supervision have estimated assuming an international firm associating with domestic consultants. All costs are first quarter 2009.

Engconsult Ltd. 3

III. COSTING DETAILS

A. Materials, Labor and Equipment.

1. Materials

1. For the calculation of unit rates, after examining the results of the geological and material survey it has been assumed that boulders/rocks, stone aggregates, gravels and sands will be locally available and at a reasonable haul distance. Steel reinforcement and cement for common construction are available in the markets at Batumi and, the nearest markets for the project. Probable sources of common construction materials are described below: • Rock: - The alignment passes some rocky sections, therefore, rock fragments may be obtained directly from roadway and tunnel excavation. • Boulders: - Boulders are available in some of the river beds crossed by the alignment. • Gravel and Sand: - Gravel and sand are available in the Chaqvistskali and Chorki Rivers. Smaller quantities are available from other rivers crossed by the alignment. 2. Gravel and sand obtained from these sources will require to be sieved and washed before use. For most of the route haul distances of sand and gravels will be from 20-30 km. 3. Crushed stones / Aggregates: - It has been assumed that crusher plants will be installed at suitable locations along the alignment for crushing rocks and boulder obtained from tunnel excavation and the river bed. Suitable locations for installation of crusher plants would be near Km 14 and near the mouth of Choroki river (Km 48) Other plants could be erected at suitable locations in flat areas. 4. Lime, Cement and Bitumen: - Georgia produces lime & cement, but high grade cement is in limited quantities. High grade cement mainly for bridges and tunnel will be imported. It has assumed that bitumen would also be imported in bulk quantity from Turkey. 5. Timber: - Timber is a scarce material in Georgia. Therefore, use of timber in construction should be avoided as much as possible. Timber is available in … in limited quantities. Locally produced plywood is also available in local markets in limited quantity. 6. Structural Steel: - Structural steel for construction of bridges and other structures is not produced in Georgia and will be imported. 7. For the preliminary design initial evaluation only has been carried out of material sources and quantity. A more specific materials source study and analysis will be required in the Final Design.

2. Labor & Equipment

8. Skilled and unskilled labor for general works such as earth excavation, form works, re- bar bending and binding, placing concrete, landscaping etc. are available in the area of the project in sufficient quantity. However, specialist skilled labor for works such as erection of major bridges, construction of tunnels etc will need to be imported. 9. General construction equipment such as excavators, dozers, road rollers, tippers, trucks etc may be obtained in limited quantities in the local markets, however they may not be in suitable condition for a project of this scale and new equipment will be imported. All specialized equipment will be imported.

Engconsult Ltd. 4

B. Quantity Estimate 10. Quantity estimation of earth and rock excavation has been calculated from the measured cross sections and geological investigations. 11. Adopted road formation widths for different terrain based on the typical cross section are:

• 14.5m– for two lane road in Plain and Rolling terrain • 13.5m – for two lane road in Mountainous & Steep terrain 12. In the preliminary design cost estimates a two lane road has been assumed except for the section km28-34 where the existing road will be widened to four lane. A passing lane has been included on the uphill side where longitudinal gradients exceed 3 percent. 13. During the topographical survey road cross sections taken only at bridge and intersection locations. Other cross sections were extrapolated from the 1:10,000 digitised mapping. The road was designed, as much as feasible, so as to balance cut and fill, however, particularly in the first 14 km this was not possible and a significant amount of additional fill will be required.. To try to balance cut and fill suitable earth retaining structures, buried culverts, high fill in bridge approaches etc were applied where possible; this will need further refinement in detail design. Using the geotechnical report the roadway excavation was classified as excavation in common soil, hard soil, soft rock and/or hard rock.. Quantities of retaining structures were also computed based on typical design cross sections for differing terrain. 14. Quantities for both bypass roads are shown in Tables 3.7 and 3.8. 15. Pavement quantities were estimated from the pavement design (Supplementary Appendix 2. Design). Quantities of slope stabilization, scour protection etc. were estimated from the design drawings for typical sections of road. 16. Bridges and culverts sizes were calculated based on calculated design flood discharge and the road geometry for bridge or culvert locations as set out in Supplementary Appendix 2. Most of the major bridges are high enough above the natural terrain so that flood levels are not an issue. Unit quantities of retaining walls, culverts and road side drains were estimated from cross sections and typical drawings. 17. Quantities of tunnel works were taken off the details in the preliminary design drawings. 18. Quantities related to special items like the service area , interchanges, pedestrian underpasses etc are computed based on the design drawings. Quantities of bio-engineering, roadside facilities and other items which are not available at the preliminary design stage were estimated on a lump sum or kilometer basis for sections of similar terrain based on costs of similar work.

C. Unit Rates 19. There has been little recent major road work construction in Ajara province except for the road improvement and tunnel at Green Cape (Km28-32 on the Project roads alignment. Unfortunately, for various reasons, the contract costs for this road are unreliable and cannot be used. There has, however, been recent and ongoing road improvement on the main East-West highway. Where feasible the rates for this work have been used in the calculation of unit rates for preliminary design standard estimate cost items. To account for impact of the economic downturn which has reportedly lowered bid prices by over 20 percent below engineers estimates, some of the rates have been adjusted downward.

Engconsult Ltd. 5

20. In particular historic unit rates were derived from four recent projects in Georgia and adjusted to present prices: • D • B • R, and • U 21. Unit rates for minor structural items, pipe culvert, box culverts, slab culvert, and side drains etc. were calculated from base item unit rates. The unit costs of retaining walls, road side drains, culverts include all auxiliary woks such as excavation, formwork, filter materials, backfill etc as needed to complete the item in all respects. 22. Bridges were divided into 3 types, small reinforced concrete (RC) bridges (total length less than 50mt), small pre-stressed concrete girder bridges (spans to 30mt), and large PC girder bridges (spans up to 40m). Costs for RC bridges were based estimated costs for recently constructed RC bridges, adjusted for average span length and pier height. Cost for simply supported girder and box girder pre-stressed concrete bridges have been based on unit rate of similar bridges in Georgia, adjusted for average span length and pier height. 23. Unit rates and costs for special items, most of which will be imported, are based on the rates and costs of recently completed similar toll road projects in other countries adjusted where feasible for Georgian conditions, taxes, duties etc. 24. There is no significant experience to date of road tunnel construction in Georgia. For this reason hardly any equipment or materials for use for road tunnel construction and operation exists in Georgia. Almost all the equipment and materials will need to be imported from other countries. Given this situation tunnel costs have been estimated after considering costs for tunnels in other countries in Europe. 25. Unit costs for tunnels are assumed for each type of tunnel and facilities per meter. Cost estimates for each tunnel have been calculated based on these unit costs and length of each tunnel type shown in the preliminary design plan and profile drawings. The estimates are set out in Table 1. 26. Tunneling costs are estimated based on the assumption that the quality of rock mass is relatively poor so as to avoid underestimation. Further detailed study of tunnel costs will be required in the detailed design stage when final alignment is fixed and geologic survey for tunnels are completed. 27. Total construction costs for each tunnel are given in Table 2. Table 1: Unit Construction Cost for Tunnel Types Unit cost / m Tunnel Type GEL Type 3 42,000 Type 2 27,500 Type 1 19,000 Extra support 2,400 Tunnel Illumination 2,000

Table 2: Total Construction Cost for each Tunnel Zeda Achkva Tunnel (L=400m)

Engconsult Ltd. 6

Tunnel Type/ Length Unit cost Total cost Facility m 1000 US$ 1000 US$

Type 3 50 25,000 1,250,000 Type 2 150 16,369 2,455,357 Type 1 200 11,310 2,261,905 Extra support 50 1,429 71,429 North portal 60 South portal 100 Illumination 400 1,190 476,190 Total 6,515,041

Makhinjauri I (L=150m) Tunnel Type/ Length Unit cost Total cost Facility m 1000 US$ 1000 US$ Type 3 100 25,000 2,500,000 Type 2 30 16,369 491,071 Type 1 0 11,310 0 Extra support 10 1,429 14,286 Long span pipe 90 0 roof* North portal 59880 60 South portal 60 Illumination 130 1,190 154,762 Total 3,160,239 *: To protect church above Makhinjauri II (L=90m) Tunnel Type/ Length Unit cost Total cost Facility m 1000 US$ 1000US$ Type 3 50 25,000 1,250,000 Type 2 40 16,369 654,762 Type 1 0 11,310 0 Extra support 20 1,429 28,571 North portal 60 South portal 60 Illumination 90 1,190 107,143 Total 90 2,040,596

Makhinjauri III (L=150m)

Engconsult Ltd. 7

Tunnel Type/ Length Unit cost Total cost Facility m 1000 US$ 1000US$ Type 3 50 25,000 1,250,000 Type 2 100 16,369 1,636,905 Type 1 0 11,310 0 Extra support 40 1,429 60 North portal 60 South portal 60 Illumination 150 1,190 178,571 Total 150 3,065,656

Salibauri Ridge (L=440) Tunnel Type/ Length Unit cost Total cost Facility m 1000 US$ 1000US$ Type 3 50 25,000 1,250,000 Type 2 280 16,369 4,583,333 Type 1 370 11,310 4,184,524 Extra support 60 1,429 90 North portal 60 South portal 60 Illumination 440 1,190 523,810 Total 10,541,877

Peria Ridge Tunnel (L=730m) Tunnel Type/ Length Unit cost Total cost Facility m 1000 US$ 1000US$ Type 3 50 25,000 1,250,000 Type 2 280 16,369 4,583,333 Type 1 400 11,310 4,523,810 Extra support 60 1,429 90 North portal 60 South portal 60 Illumination 730 1,190 869,048 Total 18,646

28. Unit rates and overall costs for the project road are high, reflecting he high costs of imported materials in Georgia (Table 3) and the need to import some technology. The overall direct cost for both bypass roads is $5.6 million per kilometer, although for the Batumi bypass, with it’s high number of tunnels and bridges it is $9.4 million/km. As a comparison of direct construction cost per kilometer for recent ADB project estimates it was $10 million per kilometer in Nepal for a two lane access controlled road in mountainous territory and for toll roads (4 lane) in PRC it ranges from $6 million per kilometer in Guangxi province, $7 million in Central Sichuan and $9.5 million in Gansu Province. All these projects are in mountainous terrain with many bridges and tunnels.

Engconsult Ltd. 8

Table 3: Unit rates (GELs) Georgia Item Unit 2008/09 Labor (unskilled) Person-day Reinforcement Bar ton Steel mesh m2 Cement ton Asphalt ton Sand m3 Rubble m3 Crushed stone m3 Diesel Fuel Kg D. Land and Resettlement 29. As of the time of production of this Report detail land and resettlement quantities were not available. The quantity of land and structures assumed is based on preliminary measurements from the affected land plots..

E. Consulting Services 30. Consulting services for the Project include detail design, supervision of construction and training of local road personnel. It has been assumed that both detail design and supervision would be carried out by international consultants assisted by domestic consulting firms. 31. The scope of the supervising consulting services have been based on similar Bank funded projects in other countries and is described in detail in Supplementary Appendix 5. Given the high numbers of domestic professional staff required it may be necessary to use foreign staff for some of the domestic engineering positions. The cost of design and supervision staff are based on recent contracts in Georgia. 32. Training for domestic staff of the Roads Department will be required in various areas (Supplementary Appendix 5). A total of 87 person months is recommended, the estimated cost has been included in the construction cost estimate.

F. Maintenance and Operation.

1. Road Maintenance

33. Maintenance costs are required for both the economic and financial analyses. Maintenance on the existing and Project roads consist of three operations a) Routine & Recurrent Maintenance, b) Periodic Maintenance and c) Emergency Maintenance. In addition there is maintenance of tunnels and service areas on the Project road. The cost of maintaining the existing road is required only for the economic evaluation. Costs for Routine, Periodic and Emergency maintenance for the existing road were taken from recent actual costs incurred by Roads Department and are set out in Supplementary Appendix 7, Economic Evaluation. Costs for maintenance of the new road are based on a preventative maintenance program. The costs of Periodic Maintenance, except for pavement rehabilitation, have been averaged as an annual cost for the analysis, although this cost occurs periodically. Minor maintenance cost of bridges and cross drainage is also included in this cost. Maintenance costs for the Project Road will be small in the early years of operation, rising as cumulative traffic volumes increase. It was assumed that a full reseal of the asphalt paved toll road is carried out every 15 years and this is costed separately.

Engconsult Ltd. 9

2. Tunnel Operation and Maintenance

34. Operation cost of tunnels is estimated based on the past experience and is shown in Table 4 for each 10 year period. It is based provision of electricity from the national grid with a present electricity cost of NR7.55/KWH although it may prove cheaper to install a dedicated power source. 35. During the initial stage of operation, say the first 5 years, the volume of traffic is relatively small and the ventilation fans will operate in a restricted mode; thus the cost for the first 10 years is reduced. For the second 10 years ventilation operation gradually increases to the designed load and for the third 10 years ventilation operation is at full capacity. 36. Illumination operation is required from the start of operation and constant operation cost is assumed. Table 4: Operation Cost of Tunnel Facilities 1st 10 years 2nd 10 years 3rd 10 years Usage unit price Lengt ‘000 ‘000 ‘000 Tunnel (MWH/y (MWH) MWH MWH MWH h US$ US$ US$ ear/m) US$’000 Zeda Ackva 400 0.25 0.1 100.0 10.0 100.0 10.0 100.0 10.0 Makhinjauri I 150 0.25 0.1 37.5 3.8 37.5 3.8 37.5 3.8 Makhinjauri 90 0.25 0.1 22.5 2.3 22.5 2.3 22.5 2.3 II Makhinjauri 130 0.25 0.1 32.5 3.3 32.5 3.3 32.5 3.3 III Salibauri 440 0.25 0.1 110.0 11.0 110.0 11.0 110.0 11.0 Peria 730 0.25 0.1 182.5 18.3 182.5 18.3 182.5 18.3 Total 1800 485.0 48.5 485.0 48.5 485.0 48.5 Source: Consultants Estimates 37. Maintenance cost is also estimated based on past experience elsewhere and includes labor, spare parts and replacement of equipment and is shown in Table 5. Table 5: Maintenance Cost of Tunnel Facilities Usage 2nd 10 3rd 10 1st 10 years $US'000/year/m years years 0-10 Tunnel length 10-20 20-30 000 US$ 000 US$ 000US$ years Zeda Ackva 400 0.012 0.012 0.12 4.8 4.8 48 Makhinjauri I 150 0.012 0.012 0.12 1.8 1.8 18 Makhinjauri 90 0.012 0.012 0.12 1.08 1.08 10.8 II Makhinjauri 130 0.012 0.012 0.12 1.56 1.56 15.6 III Salibauri 440 0.012 0.012 0.12 Peria 730 0.012 0.012 0.12 8.76 8.76 87.6 Total 1800 18 18 180 Source: Consultants Estimates

Engconsult Ltd. 10

IV. PRICING COMPONENTS

A. Foreign Component 38. Foreign and local component of the cost items are estimated based on the use of labor, materials, and equipment in for each item. It is assumed that local materials and labor would be used as much as possible in the works. However, full or partial import of materials such as cement, high tensile and rebar steel and bitumen will be required, as will most machinery and equipment, spare parts and particular materials for construction of long span bridges, tunnels and slope stabilization Cost of the imported materials have been included in the foreign component. However, cost of the materials such as petrol diesel, oil, etc is included in the local component as it is assumed that, although Georgia imports these items they would be purchased locally. An overall foreign component of 36 percent of total base line costs has been estimated.

B. Taxes and Duties. 39. There are two components to taxes and duties included in the total cost estimates, VAT and import duties. An estimate has been made of each of these items. As the contracts will be open to international bidders it has been assumed that a foreign contractor would be awarded the contract and the import items have been estimated under this assumption. 40. For infrastructure construction work the contractor pays a VAT which is charged at 18 percent whenever a goods or service is sold. VAT has been assumed to be incurred on the total contract price. 41. Import duties may be levied on imported materials and equipment for the project road. The import tariffs are small in Georgia and have been estimated at 3 percent. For ADB work these duties are normally not payable and have therefore been deleted from the total costs. This is discussed further in supplementary Appendix 6.

Engconsult Ltd. 11

V. COST ESTIMATES

42. The total estimated baseline construction cost of the Project including both bypasses and 6 km of road widening is estimated at GEL 445 million ($ 2661 million) in March 2009 prices, including, land, design and supervision and physical contingencies of 10 percent but not including price contingencies or interest during construction. The costs for the components of the project are: 1. Kobuleti Bypass GEL 207 million ($124 million) 2. Existing Highway Widening GEL 15 million ($9 million) 3. Batumi Bypass GEL 222 million ($133 million)

43. The detail project cost estimate of the total project is set out in Table 6 and for the Tranche 1 components in Table 7. Table 6: Detail Cost Estimate – Total Project Unit Cost Cost Unit Quantity GEL Gel Million

I. Investment Costs A. Subgrade Works General site clearance Ha 114 1,485 0.17 Tree felling and removal No 17,100 18 0.31 Clearing and Grubbing m3 228,000 4 0.91 1,178,6 Earth Excavation- Common Soil m3 5 5.89 65 -Soft Rock m3 270,951 5 1.35 - Hard Rock m3 174,670 10 1.75 Embankment Filling – Common 4,744,3 m3 7 33.21 Soil 65 Removal of Unsuitable Materials m3km 159,600 2 0.36 Dismantle Existing Structures No 300 8,000 2.40 Roadside Drain – Masonry Lined m 30,400 19 0.58 Concrete Lined m 3,040 49 0.15 Unlined m 45,600 7 0.32 Catch Drains m 4,000 9 0.04 Sub-Surface Drains m 11,020 58 0.64 Chute and Scour Protection m2 4,829 60 0.29 River Training Works m2 9,000 140 1.26 Subtotal Subgrade Works 49.62 B. Pavement Preparation of Subgrade m2 696,000 1 0.70 Sub Base m2 696,000 6 4.18 Upper Base m2 696,000 12 8.35 Asphalt Concrete surface course m3 69,600 420 29.23 Subtotal Pavement 42.46 C. Structures Circular Culverts, 1500 mm No 42 15,500 0.65 Box Culverts, 2,5X4,0 m No 42 55,000 2.31

1 At March 2009 exchange rate of 1.67, the $US cost does not allow for differential domestic/international inflation during the construction period

Engconsult Ltd. 12

Unit Cost Cost Unit Quantity GEL Gel Million Box Culverts, 4,5X6,0 m No 27 105,000 2.84 Small Bridges RC < 50m m 643 16,150 10.38 Large Bridges RC Box Girder m 4,200 27,000 113.40 Large Bridges RC Prestressed m 1,800 23,500 42.30 Interchanges No 6 1,500,000 9.00 Separated Intersections No 1 240,000 0.24 Over/Under passes No 12 255,000 3.06 Pedestrian crossings No 10 75,000 0.75 (Over/under) Retaining Walls m 2,660 750 2.00 Subtotal Structures 186.93 D. Tunnel Works –Single Tube Civil Works Type DI m 720 17,000 12.24 Type DII m 870 25,000 21.75 Type DIII m 350 38,000 13.30 Extra Support m 250 2,200 0.55 Long Span Pipe Roof m 90 2,200 0.20 Tunnel Portals No 12 95,000 1.14 Tunnel Illumination m 1,940 1,800 3.49 Emergency Generator LS 4 35,000 0.14 Provision of Power Supply LS 4 22,000 0.09 Subtotal Tunnel Works 52.90 E. Slope Stabilization/ Bio Engineering Dry Stone Masonry Walls m2 13,300 350 4.66 Landscaping m2 12,500 40 0.50 Subtotal Stabilization 5.16 F. Roadside Facilities 0.00 Service/weighbridge Area No 1 80,000 0.08 Signing and Road Marking KM 48 10,000 0.48 Steel Guardrails m 76,000 105 7.98 New Jersey Barriers m 2,000 155 0.31 Boundary Fencing m 76,000 40 3.04 Subtotal Roadside facilities 11.89 G. Miscellaneous Relocation of km 1 50,000 0.05 Transmission/service Lines Environmental protection LS 1 650,000 0.65 Noise barrier m 341 1,044 0.36 Temporary Works Amount 1 200,000 0.20 Realignment of Existing Roads Amount 1 300,000 0.30 km. Maintenance of Existing Roads 1 500,000 0.50 year. Subtotal Miscellaneous 2.06 I. Equipment Monitoring, Comms, Enforcement Ls 1 417,000 0.42 Weigh Bridges No 1 60,000 0.06 Subtotal Equipment 0.31 J. Land and resettlement 0.00 2,357,7 Land Acquisition ha 10 23.58 95 Structures m2 23,571 300 7.07

Engconsult Ltd. 13

Unit Cost Cost Unit Quantity GEL Gel Million Subtotal Land and resettlement 30.65 K. Consulting 0.00 Survey and Design Amount 1 7,000,000 7.00 Supervision Amount 1 15,350,000 15.35 Subtotal Consulting 22.35 L. Over Heads Insurance Etc LS 1 250,000 0.25 Subtotal Over Heads 0.25

Total, GEL 404.6 Contingencies 40.5 Grand Total GEL 445.1 Total, $ 266.5

Table 7: Detail Cost Estimate – Tranche 1

Unit Cost Cost Unit Quantity GEL Gel Million

I. Investment Costs A. Subgrade Works General site clearance Ha 80 1,485 0.12 Tree felling and removal No 11,942 18 0.21 Clearing and Grubbing m3 159,223 4 0.64 Earth Excavation- Common Soil m3 496,832 5 2.48 -Soft Rock m3 128,902 5 0.64 - Hard Rock m3 51,561 10 0.52

Embankment Filling – Common 3 3,435,6 m 7 24.05 Soil 79 Removal of Unsuitable Materials m3km 79,800 2 0.18 Dismantle Existing Structures No 180 8,000 1.44 Roadside Drain – Masonry Lined m 21,230 19 0.40 Concrete Lined m 2,123 49 0.10 Unlined m 31,845 7 0.22 Catch Drains m 2,793 9 0.03 Sub-Surface Drains m 7,696 58 0.45 Chute and Scour Protection m2 3,372 60 0.20 River Training Works m2 6,285 140 0.88 Subtotal Subgrade Works B. Pavement Preparation of Subgrade m2 486,050 1 0.49 Sub Base m2 486,050 6 2.92 Upper Base m2 486,050 12 5.83 Asphalt Concrete surface course m3 48,605 420 20.41 Subtotal Pavement C. Structures

Engconsult Ltd. 14

Unit Cost Cost Unit Quantity GEL Gel Million Circular Culverts, 1500 mm No 35 15,500 0.54 Box Culverts, 2,5X4,0 m No 34 55,000 1.87 Box Culverts, 4,5X6,0 m No 17 105,000 1.79 Small Bridges RC < 50m m 298 16,150 4.81 Large Bridges RC Box Girder m 1,400 27,000 37.80 Large Bridges RC Prestressed m 1,150 23,500 27.03 Interchanges No 3 1,500,000 4.50 Separated Intersections No 1 240,000 0.24 Over/Under passes No 9 255,000 2.30 Pedestrian crossings No 10 75,000 0.75 (Over/under) Retaining Walls m 1,575 750 1.18 Subtotal Structures D. Tunnel Works –Single Tube Civil Works Type DI m 200 17,000 3.40 Type DII m 150 25,000 3.75 Type DIII m 50 38,000 1.90 Extra Support m 60 2,200 0.13 Long Span Pipe Roof m 0 2,200 0.00 Tunnel Portals No 2 95,000 0.19 Tunnel Illumination m 400 1,800 0.72 Emergency Generator LS 1 35,000 0.04 Provision of Power Supply LS 1 22,000 0.02 Subtotal Tunnel Works E. Slope Stabilization/ Bio Engineering Dry Stone Masonry Walls m2 7,878 350 2.76 Landscaping m2 8,125 40 0.33 Subtotal Stabilization F. Roadside Facilities Service/weighbridge Area No 1 80,000 0.08 Signing and Road Marking KM 34 10,000 0.34 Steel Guardrails m 62,000 105 6.51 New Jersey Barriers m 2,000 155 0.31 Boundary Fencing m 50,000 40 2.00 Subtotal Roadside facilities G. Miscellaneous Relocation of km 1 26,000 0.03 Transmission/service Lines Environmental protection LS 1 338,000 0.34 Noise barrier m 341 1,044 0.36 Temporary Works Amount 1 104,000 0.10 Realignment of Existing Roads Amount 1 144,000 0.14 km. Maintenance of Existing Roads 1 240,000 0.24 year.

Engconsult Ltd. 15

Unit Cost Cost Unit Quantity GEL Gel Million Subtotal Miscellaneous I. Equipment Monitoring Comms, Enforcement Ls 1 342,000 0.34 Weigh Bridges No 1 60,000 0.06 Subtotal Equipment J. Land and resettlement 1,505,1 Land Acquisition ha 10 15.05 77 Structures m2 5,906 300 1.77 Subtotal Land and resettlement K. Consulting Survey and Design Amount 1 7,000,000 7.00 Supervision Amount 1 9,000,000 9.35 Subtotal Consulting L. Over Heads Insurance Etc LS 1 172,500 0.17 Subtotal Over Heads

Total, GEL 202.3 Contingencies 20.2 Grand Total GEL 222.5 Total, $ 133.2

Engconsult Ltd. 16

VI. PROCUREMENT PLAN

A. Introduction 44. The Project road is made up of 48 km consisting of 28 km of Kobuleti two lane road, 6 km of existing road widening and 14 km of Batumi by pass two lane road. The first two sections make up the first tranche of the project, while Batumi bypass forms the second tranche. It is proposed that detail design start in August 2009, that 2 contracts of the first tranche be let by July 2010, the remaining contracts of tranche 1 by October 2010. Tranche 2 contracts will start in 2011. For the purpose of this analysis only tranche 1 contracts have been calculated 45. It should be possible to complete tranche 1 contracts (Kobuleti bypass and existing road widening the end of first quarter 2013. Three separate civil works contract packages are proposed for the 32 kilometers, including one for the road widening.. It should be noted, however, that these packages are based upon preliminary design and could be amended during detail design.

B. Contract Packages 46. The contract packages proposed for Tranche 1 comprise 3 packages of civil works, 2 for landscaping, one for signs and road marking, one for boundary fencing and guard rail and 3 packages for equipment. The packages are based on the preliminary design cost estimates. 47. The price for the two new road civil works contracts is $$23 million and $74million. This would attract international bidders, especially as qualified bidders could bid for more than one section. 48. The contract to widen the existing road is $4.6 million and could be done by a domestic contractor or by a bidder bidding for two contracts together. 49. It is assumed that vertical slicing of packages will be used (i.e. include pavement surfacing in the civil works packages). 50. Four contracts are specified to supply and install road furniture (signs, markings, posts etc) and to carry out landscaping. These contracts would be carried out immediately the paving work was complete.

C. Mode of Procurement 51. It is proposed that two larger civil works packages be International Competitive Bidding (ICB) and the existing road widening be National Competitive Bidding (NCB). 52. The contracts for landscaping, fencing, guard rail and signs, marking have been proposed as National Competitive Bidding (NCB). Equipment packages would be by ICB/IS unless locally sourced. 53. The sections and packages are numbered from the zero point (Km 0) at the northern end of the project south towards Batumi. Contractors may be allowed to bid for more than one package. The actual geographical limits of each package were determined from considerations of size, road connections and access to site.

Engconsult Ltd.

Table 8: Proposed Contract Packages Tranche 1 I- Earth/Roc Retaining Contra No Chainage Small Large chang Tunnel Cost Mill k cut Walls ct e Cut Fill Bridge Bridge mGEL $US Type Km m3 ’000 m3 ’000 m m m No Engconsult Ltd. C1 0.0-10.0 - 1,350 141 212 170 2 43.3 25.9 ICB C2 10.0-28.0 644 2,057 996 86 2,380 1 440 123.6 74.2 ICB C3 28-34.0 9 28 438 100 8.4 5.0 NCB Signs & Road marking 0.34 0.2 NCB Boundary Fence &Guard 8.5 5.0 NCB Rail LS1 Landscaping 0.32 0.19 NCB E1 Communication, Monitoring and Enforcement Equipment 0.417 0.25 IS E2 Tunnel M&E 0.735 0.44 IS E3 Weigh Bridge 0.06 0.36 IS

17 18

VII. IMPLEMENTATION SCHEDULE

A. Introduction 54. The project area is affected by seasonal rains and some small snowfalls; work may be limited over two months of the year. Otherwise is not expected that weather conditions will seriously affect work progress. 55. All sections of the road are accessible by existing roads although most roads will need improvement for heavy equipment. There is an existing minor road which would give access to the vicinity of the tunnel. 56. Schedules for both preconstruction activities and construction work are set out in Figure 3.1 and 3.2. These assume a very quick period for pre-construction procedures and delays by either Government or ADB would delay start of construction.

B. Comments on Schedule 57. The schedule assumes the construction of a two lane road with completion in four years. The timing will be dictated by the bridge and tunnel works, however there are no specific construction items which would slow overall progress and he three year construction period should be ample time to complete the road sections. 58. The progress of the tunnel excavation, which is followed by secondary lining being carried out at about 200 m apart, is assumed to be about 20 m/month for Type 3, 45 m/month for Type 2 and 60 m/month for Type 1. This means that tunnel excavation should be complete within a year after providing access. Necessary niches for facility installation should be prepared during the secondary lining concrete activity and the installation of facilities can start immediately after completion of the tunnel lining. As a relatively long access road is required it is estimated that the tunnel construction will be 20 months. 59. The pre-construction activity schedule for the project road is shown in Figure 1. This shows that mobilization should be feasible for the first contracts by July 2010. 60. Figure 1 shows the construction schedule assuming a July 2010 start and a 3 year construction period; it is unlikely that this can be shortened. An October 2013 opening date is assumed for the complete Kobuleti bypass, although some construction activities (service area and some signing etc) may continue later into 2013. Assuming that technically expert construction and supervision companies are employed, especially for the bridge and tunnel works, it is considered that the implementation schedule is achievable.

Engconsult Ltd.

2009 2010 2011 2012 2013 2014 Activity Figure 1: Tranche 1 Schedule

Engconsult Ltd. Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4

Approval of Advance Procurement

Tranche I (34 km) - Kobuleti Bypass tranche I tranche II Detailed Design and Bid Documents Review of Detailed Design by the Government Construction Supervision

Recruitment of Supervision Consultants Supervision and Preparation of Project 2 defects liability period Prequalification for Civil Works

Contract 1 & 2 - Km 0-11 & Km 28-34

Bid Documents and ADB's Approval Civil Works Procurement Civil Works Implementation

defects liability period Contract 3 - Km 11-28

Bid Documents and ADB's Approval Civil Works Procurement Civil Works Implementation

Capacity Development for Road Network defects liability period Road Safety Improvement 19

FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 4 Road Safety Component

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Contents I. Introduction ...... 1 A. Background ...... 1 B. Framework of Road Safety Improvement ...... 1 II. Road Safety in GEORGIA ...... 3 A. Responsibility for Road Safety in Georgia...... 3 B. Other Road Safety Work in Georgia ...... 3 C. Present Conditions and Analysis of Road Accident Data ...... 3 1. Number of Road Accidents & Trends ...... 3 2. Accident History on the Existing Road ...... 4 3. Accident Database System ...... 5 D. Road Safety Strategy and Organization ...... 5 1. Georgia Traffic Safety Strategy ...... 5 III. Road Safety audit of project road ...... 9 A. Background ...... 9 B. Review of Typical Project Layouts & Design Plans ...... 10 1. Design Planning ...... 10 2. Design Speed ...... 10 3. Roadway Cross-section ...... 10 4. Safety-barriers & Edge-delineators ...... 11 5. Sign and Marking ...... 11 6. Visibility ...... 11 7. Bridges & Culverts ...... 11 8. Road Tunnels ...... 12 9. Climbing Lanes ...... 12 10. Axle-Load Control Complex ...... 12 IV. ROAD SAFETY COMPONENT ...... 13 A. Coordination with World Bank projects ...... 13 B. Awareness Campaigns for Road Safety ...... 13 1. Safety Code for Access-Controlled Roads & Tunnels ...... 14 2. Roadside Community Awareness Campaign ...... 14 3. Pedestrian Awareness Campaign for Open Access Section of Road...... 14 4. Awareness Campaign for Heavy Vehicles ...... 15 5. Publicity through Television and Newspaper ...... 15 V. Intelligent Transport Systems ...... 16 A. Introduction ...... 16 B. The Current Situation ...... 16 1. Review of ITS in Georgia ...... 16 2. International Practice ...... 16 C. ITS for the Roads in the project ...... 17 1. Project Road Proposed ITS Arrangements ...... 17 2. Recommended ITS for the Project ...... 17 3. Future ITS proposals ...... 18 VI. Long-Term Measures to Achieve Tunnel and Road Safety ...... 21 VII. Recommendations ...... 22

Engconsult Ltd. ii

List of Tables Table 1: Recorded Number of Road Accidents ...... 3 Table 2: Accidents on project road ...... 4 Table 3: ITS Recommendations ...... 19 Table 4: Cost Estimate for Road Safety Components ...... 22

List of Figures Figure 1: Trend of Road Accidents in Nepal ...... 4

Engconsult Ltd. iii

GLOSSARY AADT Average Annual Daily Traffic AASTHO American Association of State Highway and Transportation Official ADB Asian Development Bank GIS Geographical Information System GOG Government of Georgia ITS Intelligent Transport System MFF Multi Facility Financing RD Road Department, Ministry of Regional Development and Infrastructure. RoW Right of Way RSA Road Safety Audit TD Transport Department of the Ministry of Economic Development VMS Variable Message Sign WB World Bank

Engconsult Ltd. 1

I. INTRODUCTION

A. Background 1. There were 867 fatalities from road-traffic accidents in Georgia in 2008, a rise of over 18% from 2007. The fatality rate relative to the vehicle fleet at 16 per 100,000 vehicles in 2007 is significantly higher than other Eastern European countries and much higher than other developed countries in EU 1 . Both the fatalities and serious injuries resulting from traffic accidents in Georgia over the past few years show an upward trend and road-safety is thus becoming a serious social issue in Georgia. 2. World Bank review of statistics for 19962 showed that about 66% of deaths from road accidents occur in "urban" areas and 34% occur on "rural" roads. Whereas on rural roads in Georgia (where wearing of front seat seatbelts is compulsory and there is visible and effective police enforcement of speeding) deaths were only increasing at 3% per year, deaths on "urban" roads were increasing at around 25% per year; in 2006 the high increases and a significant proportion of road deaths occurred in towns or built up areas along major roads. Over 30% of those killed nationally in 2006 were pedestrians. This is almost double the percentage in many West European countries3. The fact that seatbelts are not worn and vehicle speeds are often excessive in "urban" areas and that there are very few facilities to ensure pedestrian safety or to guide drivers through hazardous locations on the road network are major causes of these accidents and fatalities. This project will provide a bypass to two congested areas, Kobuleti and Batumi which, at present, incur traffic and pedestrian congestion, especially in the summer holiday season. 3. Asian Development Bank (ADB) considers road safety an important issue and has made an effort to promote road safety by providing “Road Safety Guidelines for the Asian and Pacific Region”. Similarly, the Government of Georgia (GoG) through the Transport Department of Ministry of Economic Development (TD) has recently produced a Georgian National Traffic Safety Strategy (2008). Both the ADB and GoG guidelines on road-safety therefore apply to the Project Road. 4. This Supplementary Appendix: (i) describes present state of road safety conditions in Georgia and issues relevant to the Project Road; and (ii) discusses the safety audit carried out on the preliminary design and the need for further audits, (iii) recommends a road safety component for the project, and (iv) discusses a possible Intelligent Transport System (ITS) component to be included in the project and the overall MFF.

B. Framework of Road Safety Improvement 5. For the safe design of any road, it is necessary to formulate a road safety component to effectively prevent or reduce the road accidents along the road. As the Project Road is a new construction along a new alignment, there are no existing accident blackspots4 to consider and therefore the objective is to prevent road-accidents by incorporate adequate countermeasures. Successful road-safety practices generally require improvements to engineering, education and enforcement (the principle of three “Es”). With the above in mind, there are three aspects outlined in the safety component for this Project road to effect comprehensive road safety measures.

1 World Bank working paper. 2 World Bank Appraisal Report 2nd E-W Highway Improvement 3 e.g. Germany 14%, France 11%, Sweden 14%, Netherlands 8% 4 Accident blackspots are locations where accidents are concentrated

Engconsult Ltd. 2

6. A safety audit of the preliminary design of the proposed project road, combined with a study of general trends in road safety in Georgia, has led to minimization of potentially hazardous or critical locations related mainly to alignment and road profile. 7. To ensure smooth traffic control efficient enforcement is required by the Traffic Police (Patrol Police). This enforcement aspect is important and will be crucial during the opening years of road operations when many road-users and roadside communities will be unfamiliar with the concept of an access controlled, high speed highway. Effective and strict enforcement are also necessary to control non-motorized/slow-moving traffic, to regulate axle-loads and to mitigate traffic disruptions due to accidents (incidence management). 8. Education of road-users is an important part of the road safety component to assure the desired effect of other parts of the component. Successful road safety will largely depend on how effective the education programs or safety campaigns are. This is a nationwide need and is therefore best included in the overall MFF. 9. The incorporation of some elements of an ITS will help alert drivers and road authorities of changes to road conditions or dangers ahead. 10. Finally, an effective emergency response system is important to safeguard road-users during emergencies once the project road opens. The emergency response system will require coordination amongst concerned authorities, with logistic support to deal with emergencies related to road accidents, vehicle breakdown or tunnel incidents.

Engconsult Ltd. 3

II. ROAD SAFETY IN GEORGIA

A. Responsibility for Road Safety in Georgia. 11. Responsibility for road safety in Georgia is spread amongst three ministries and cooperation between the three appears somewhat sketchy. 12. The Interior Ministry, through the Patrol Police is responsible for attending and reporting on road accidents as well as enforcing traffic regulations and laws. The police record all accidents to which they attend and the data is entered into a computerized data base; however, analysis of this data appears to be minimal. 13. The Road Safety Division of the Roads Department of the Ministry for Regional Development and Infrastructure is responsible for ensuring that the road infrastructure is designed, constructed and maintained in a safe manner. It carries out road safety improvements, sometimes at the request of the patrol police, and other times after having manually analyzed the data collected by the police. It is also responsible for installing and maintaining road signs and markings. 14. The Transport Department of the Ministry of Economic Development was instrumental in producing a Road Safety Strategy for Georgia (D.1 below).

B. Other Road Safety Work in Georgia 15. Little work has been carried out in road safety in Georgia prior to 2002. A road sign audit was carried on all international roads in 2003 by Roads Department and some improvements have been made to infrastructure since then. 16. A PHRD funded road safety study in 2007 led to a summary Georgian Traffic Safety Strategy but this is only being spottily implemented, mainly because it does not include a time bound action plan. 17. World Bank (WB) has ongoing road safety projects in Georgia, mainly as part of the E-W Highway upgrading and is proposing a new road safety component in its 3rd East West highway Improvement. This component will develop road safety audit guides, train MRDIRD staff in their use and use the E-W Highway upgrading as a pilot to use the audit guides. It also intends to update the traffic sign and marking audit for all main roads; it is important that any road safety component for the MFF project should complement those studies. .

C. Present Conditions and Analysis of Road Accident Data

1. Number of Road Accidents & Trends

18. The number of the road accidents in Georgia including the number of the fatalities and injuries are shown in Table 1 and Figure 1. 19. The table shows that both accidents and fatalities have been rising steadily over the last eight years, and have risen by 20 percent between 2007 and 2008 (18 percent for fatalities). The overall annual growth since 2001 has been 17.6 percent for accidents and 6.5 percent for fatalities. The growth rate for alcohol related accidents has grown at the same rate as accidents Table 1: Recorded Number of Road Accidents Year Accidents Alcohol related Injuries Fatalities 2001 1,937 123 2,370 558 2002 2,011 na 2,509 515

Engconsult Ltd. 4

Year Accidents Alcohol related Injuries Fatalities 2003 2,113 123 2,585 572 2004 2,936 86 4,069 637 2005 3,870 329 5,546 581 2006 4,795 225 7,084 675 2007 4,946 311 7,349 737 2008 6,015 378 9,063 867 Source: Ministry of Internal affairs of Georgia, Department of Police

10,000

8,000

6,000

4,000 Number 2,000

0 2001 2002 2003 2004 2005 2006 2007 2008

Year

Accidents Injuries Fatilities

Figure 1: Trend of Road Accidents in Nepal

20. The Patrol Police identify the major causes of accidents to be the poor condition of the roads and vehicles, erratic behavior of pedestrians and speeding, which is thought to be the dominant cause of accidents on the main roads. However other reasons, easily observed on both urban and rural roads, are driver recklessness (speeding, dangerous overtaking) inadequate consideration for pedestrians, and inadequate safety facilities.

2. Accident History on the Existing Road

21. There were 220 accidents along the existing Poti to Sarpi road in 2008 with 406 injuries and 43 fatalities. This is a rise of 59 percent in accidents and 72 percent in fatalities, an alarming rate of growth. The very high rate of injuries to accidents would indicate that many accidents that do not cause injury are not officially reported. A summary of these accidents and the reasons given by the police for them is set out in Table 2. Table 2: Accidents on project road Cause of Accident % Speeding 33 Maneuvering 19 Violation Technical Fault 12 Crossing Center Line 11

Engconsult Ltd. 5

Cause of Accident % Not Keeping Distance 6 Alcohol 3 Passing unlawfully 2 Tiredness 1 Other 13

22. Little work has been carried out on the cost of accidents in Georgia but PHRD consultants in 2007 indicated that economic losses were at least US$100 million per year (around 1.4 % of annual 2007 GDP), although further work has indicated that this figure is very low. 23. In the economic evaluation for this study (Supplementary Appendix 6) the economic cost of road accidents has been estimated as: • Average cost per accident along Project bypass road : GEL.57,600 (US$ 34,500) • Average cost per accident along existing roads: GEL.46,000 (US$ 27,600)

3. Accident Database System

24. A comprehensive database system for road accidents is an essential tool to investigate causes behind road-accidents and develop effective countermeasures to mitigate future accidents at accident blackspots. It is also an essential aid in outlining broad road safety strategies and countermeasures to prevent accidents at hazardous locations at proposed or new roads. 25. In Georgia a simple database containing accident records was installed by the police in 2007 and accidents appear to be reported in sufficient detail that they could be used to provide a reasonable analysis of the cause and type of accidents on main roads and to identify black spots. World Bank, in their Second East West Highway Project have formulated a plan for a new accident data base system and, in the next phase will migrate historic data and train staff in its use. It is important that data from the existing project road be included and that new road data be input as it becomes available.

D. Road Safety Strategy and Organization

1. Georgia Traffic Safety Strategy

26. GoG has begun to realize the problem of poor road safety and has made some attempt to alleviate the problem. For instance, the Patrol Police are now better equipped and thus more efficient in enforcement, some major roads have been resurfaced or rehabilitated, a drunken driving law has been introduced and enforced, and traffic monitoring devices have been installed in Tbilisi. However, there is much to be done. 27. In July 2008, GoG endorsed the National Traffic Safety Strategy presented by the Transport Department of Ministry of Economic Development (TD). The main goals were: • Improve infrastructure in accordance with revised safety standards; • Complete the legislative base for road safety reform; • Provide the institutional framework for road safety; and • Increase public awareness of road safety.

Engconsult Ltd. 6

a. Strategy Actions

28. The main actions to achieve those goals were stated as: • Patrol Police o Police Patrolling Instead of Police Posts (implemented 2004) o Installing Video-cameras on some road sectors to prevent and identify traffic infringements (ongoing). o Additional equipment for enforcement -radar speed traps and alcohol breath testing (ongoing) o "One Call" service (Fire, Rescue, Emergency and Police services) – 022 (proposed). • Legislation o Restrictions on use of Mobile phones in vehicles o Proportional Penalties for speeding and drunk driving (at present there is one fixed penalty) o Increase penalties for drunk drivers and aggressive driving. o Introduce compulsory seat belt wearing for back seats on rural roads and for front seats in urban areas. o Regulate design of Entrances to Commercial premises • Infrastructural Measurements o Revise road standards (introduced March 2009) o Road construction/Reconstruction (ongoing) ƒ East-West Highway widening. ƒ Chuberi-Sakeni Road ƒ rehabilitation of Tbilisi-Senaki-LeseliZe Highway sectors: Agaiani-Igoeti (13 km) and lgoeti-Sveneti (25 km). Total cost of the reconstruction process is 41.4 million USD; ƒ Rehabilitated Nataxtari-Agaiani (16 km) - sector of Tbilisi-Senaki-LeseliZe Highway. ƒ Rehabilitation of Urban Roads (50m GEL proposed) ƒ rehabilitative of parts of Chumburi-Senaki Highway and entrance road to Batumi International Airport • Action Plan o Preparation of prioritized road rehabilitation and maintenance action plan • Pedestrian Safety o More Pedestrian Tunnels. o Pedestrian Overpasses at places of high intensity traffic o Pedestrian walkway on bridges. o Provision of walkways and crossings at all urban tunnels. o Barriers on footpaths of central city streets to stop parking. o Improve Bus Stops and Entrance Roads placement to reduce congestion o Provision of central island at pedestrian crossings. o Provide facilities for disabled persons at pedestrian tunnels and overpasses. o Provision of Speed Reduction Devices near schools o Providing Islands and Kerbs instead of painted lines, at U-turns, and raised central islands at all roundabouts. • Equip road and railway intersections with essential facilities and traffic signs • Provide animal and man proof fencing along RoW of major highways • Road Signs o More road signs to be provided in urban and rural roads

Engconsult Ltd. 7

o Provide essential temporary road signs during road reconstruction. • Road Lightning o Improve lightening facilities on international highways, and minor roads; o Improve lighting on high traffic roads. • Maintenance, design research, scientific and expert activities. o Improve periodic maintenance and rehabilitation of roads; o Research into minimization of the effect of natural disasters; o River-bed and coastal protection methods; • Institution Framework o Expert Group: creation of a special expert group of representatives of competent authorities, which will have a coordination role and prepare needed information and data for the Georgian Transport and Communication Policy Development Commission. The Working Group will be made up of representatives of: Ministry of Economic Development of Georgia; Ministry of Internal Affairs of Georgia; Ministry of Education and Science; Ministry of Health, Labour and Social Protection; Ministry of Finances and other Interested Authorities. • Improve public awareness on traffic safety o Civil Information: The Ministry of Internal Affairs of Georgia and TV and radio media provide information on traffic safety. This should be continued. o Devise appropriate informational campaigns. o Pilot Project in some schools regarding traffic safety rules to be continued and enlarged. o Consider publishing fliers and brochures and distribute them to schools and other educational institutions. b. Responsibility for Strategy

29. Under the new strategy the following authorities will be the main entities in road safety. a. The supervisory authority would be the Georgian Transport and Communication Policy Development Commission. The main functions of the Commission would be to separate the economical and technical regulation fields of transport and communication sectors policy directions, and facilitation of infrastructure improvement. They would ensure appropriate monitoring and coordination of Georgian National Road Safety Strategy. b. Georgian National Road Safety Strategy Implementation Authority: The Ministry of Economic Development of Georgia: Department of Transport, Road Department would ensure data collection of road accidences and road facilities from the relevant organizations, analyze information, and prepare recommendations for the Georgian Transport and Communication Policy Development Commission for determination of future activities. Ministry of Internal Affairs of Georgia would prepare and transfer information regarding road accident to the Ministry of Economic Development of Georgia and would provide special training and instruction for (telephone) 022 staff for effective perception of information and coordination with proper authorities. c. Local Self-Government Authorities would harmonize roads construction, rehabilitation and maintenance in accordance with "Road Projection Geometrical and Structural Standards". 30. As can be seen there are significant differences in responsibilities under the strategy as compared to the existing situation.

Engconsult Ltd. 8

31. Whilst the strategy is wide ranging it lacks substance and, in particular, it does not set out a time bound action plan for implementation. Consequently little progress has been achieved in implementing the strategy. It is understood that such an action plan is being proposed.

Engconsult Ltd. 9

III. ROAD SAFETY AUDIT OF PROJECT ROAD

A. Background 32. A Road Safety Audit (RSA) is a systematic process for checking the road safety implications of highway improvements and new road schemes. The main objective of an audit is to minimize future road accident frequency and severity once the road has been built and comes into use. The audit needs to consider all road users, particularly pedestrians and cyclists. Having identified potential road safety problems, the auditor makes recommendations for improvement. 33. In the pre-construction phase, the road safety audit has the greatest potential to improve safety as it examines the design before a road facility is built. Pre-construction phase RSAs can be conducted on any proposed road project that is likely to alter interactions between different road users or between road users and the road environment. 34. Preliminary design RSAs are conducted on the preliminary design drawings which normally show only general details regarding the road alignment and profile, with typical cross- sections only. Design drawings for interchanges are normally at a larger scale which allow for closer examination. A base map can show vertical contours, watercourses, and existing roads, structures, and property lines. In this project the base map was at 1:10,000 scale on which the initial alignment was plotted. The alignment and profile was then developed from the topographic survey of the alignment corridor. 35. The main purpose of the preliminary design RSA is to: • Avoid wasting time and effort in redesigning basic elements during detailed design; • Ensure safety isn't compromised when project elements which meet only the minimum design standards interact; • Evaluate whether any departures from standards will significantly impact safety; • Evaluate horizontal and vertical alignments (roadway and roadside cross-sections), interchange configuration, intersection layout, to ensure they do not present safety hazards; • ensure the needs of all road user groups are being met. 36. During the preliminary design RSA, the safety engineer should see how the planned road improvement will tie in to the existing road network and examine adjacent roadways to determine how consistent the design will be from the perspectives of different road users. 37. Whilst fundamental decisions regarding route choice, the overall design, and layout of the project are being decided the safety engineer (in this case also the design engineer) can suggest significant physical changes, such as horizontal and vertical alignment, provision of a median, lane and shoulder width, footpaths, and channelization. Interchanges should be reviewed for upstream/downstream effects, potentially conflicting movements, and sight distances. In this project many of these elements were set through the use of new Georgian road standards5. When alternative options were being considered, each one was reviewed from a road safety viewpoint. 38. It is important to carry out this preliminary audit when significant land acquisition is required, as in this project, before acquisition is finalized in the event that the RSA leads to recommendations for significant changes to horizontal alignment that would require additional land.

5 2009

Engconsult Ltd. 10

39. Further RSAs should be carried out at the detail design stage, the construction stage and the pre-opening stage.

B. Review of Typical Project Layouts & Design Plans

1. Design Planning

40. The Project Road is planned to be fully access-controlled, except for a six kilometer section of existing road widening, with the prime objective to provide a quick bypass of Kobuleti and Batumi. To ensure safety of the vulnerable road-users and as a pre-requisite for an access-controlled road, non-motorized, slow-moving vehicles and pedestrians should be restricted along the road. In order to enforce this restriction, boundary fencing is proposed at the boundary of the road right-of-way. In addition, police enforcement and education is also recommended to educate roadside communities. 41. Access to the Project road is restricted through interchanges (which will be grade- separated) proposed at Natanebi River (Km 0), Ozurgeti Road (Km 14), existing road (Km 28), start of Batumi bypass (Km 34), Khelvachauri (km 44) and at the end of the project (km 48) with the existing road. 42. Average Annual Daily Traffic (AADT) at the opening year (2013/4) is forecasted to vary from 2,400 on Kobuleti bypass to 1,600 vehicles on the Batumi bypass, rising to 5,700 and 3,700 respectively in 2034. Based on assumed design capacity of the Project road (AADT of 12,000 vehicles), the two-lane carriageway will be adequate. In the unlikely case that the alternative border post between Turkey and Georgia is not opened or the border between Turkey and Armenia remains closed, the volumes by 2034 would require the road to be widened to four lanes. There are therefore no capacity constraints foreseen in the design period of the Project Road and potential conflicts due to traffic congestion will be minimized

2. Design Speed

43. Following the Georgia road standard for four lane roads, a design speed of 120 kph for flat terrain, reducing to 80 kph in mountainous terrain has been proposed for the two lane road. The design speed is justified and should be conducive to safe driving.

3. Roadway Cross-section

44. Georgian standards stipulate, 3.75m lanes in each direction with 2.5m shoulders in flat terrain, dropping to 2.0m in mountainous terrain (two lane road). This is adequate to allow vehicles to pull completely off the driving lane in the event of a breakdown or an emergency. 45. With a two lane high speed road, head-on collisions between opposing traffic could be a serious safety issue, especially in flatter terrain where speeds will be high, although the wide shoulder and controlled access should help to minimize such accidents. However it is recommended that reflective thermoplastic road-marking be used along the whole road and concrete delineator posts 6 be placed on the valley-side for hill-cut sections in rolling or mountainous terrain and both sides for embankment sections in flat terrain.

6 Prismatic concrete delineator posts basically improve delineation of the formation edge and cannot act as safety- barrier.

Engconsult Ltd. 11

4. Safety-barriers & Edge-delineators

46. Steel guard-rails are proposed along the valley-side of all sections of the Road in steep terrain. It is also recommended that single-beam steel guard-rails be placed on all sections of road having vertical drop of 3 m or more. The start and end of all guard-rails should be splayed away and buried into the ground so as not to pose a safety-hazard itself. 47. At road-section with a vertical-drop of less than 3 m concrete post-delineators should be provided at both edges and on the valley-side on sections cut/fill sections. 48. “New Jersey” type median barriers are proposed for a four kilometer section of the existing road widening where land restrictions make the provision of a wide central median infeasible.

5. Sign and Marking

49. At this stage of design, no detailed sign and marking plan for the road is provided. However, it is recommended that the new Road Traffic Signs and Marking standard be used as this generally follows the Vienna Convention UNECE on road traffic signs which is internationally accepted. It is understood that the new Georgian Signs/markings standard is to be based on this.

6. Visibility

50. The potential locations where visibility or sight-distance could be constrained would be sharp corners (blind corners), locations obscured by summits and tunnel entrance/exits. Given that the design has retained a minimum horizontal radius of >600m wherever possible there should be no visibility problems along the road.

7. Bridges & Culverts

51. The full carriageway including the shoulder will be continued along all culverts. Georgian standards have a small (0.5m) shoulder across bridges but have a footpath on both sides of up to 2.5m (minimum 0.6m). For a fully access road two wide footpaths are inappropriate as pedestrians are not allowed in the RoW. It would be safer to widen the shoulder width to 2.0m and to have a minimum width footpath on one side only for maintenance or emergency use. 52. Based on the design plans, the following interchanges are proposed: • Modified trumpet interchanges are proposed at Natanebi River (Km 0), and at the end of the project (km 48) with the existing road. • Aligned trumpet-interchange at existing road (Km 28), start of Batumi bypass (Km 34). • Partial Cloverleaf interchange at Ozurgeti Road (Km 14), and Khelvachauri (km 44). 53. Deceleration and acceleration lanes are proposed before exit and after entry ramps, respectively, to all interchange. Georgian standards require these to be 3.75m in flat sections which is satisfactory. 54. Visibility at interchanges is generally based on criteria satisfying decision sight-distance. Decision sight-distance is the distance required for a driver to detect a complex road environment, perceive hazard, analyze it and take action. 55. While a detailed assessment is not currently feasible, preliminary study shows that there is adequate visibility from all access-roads to interchange entry-ramps.

Engconsult Ltd. 12

Based on AASHTO (and TRACECA) Standard, design-speed for ramps would be 30 kph and 40 kph for mountainous and flat terrain, respectively.

8. Road Tunnels

56. Five tunnels are proposed along the project road, two are medium size (600m and 700m) and three are small (<250m). The three small tunnels are connected by short bridges or fill, about 150m in length. All tunnels have straight alignments. 57. The Project proposes single two lane, two way tunnels. Cross-sectional drawings agree with Georgian standards and are larger than international standards with 3.5m lanes and 0.7m shoulders on each side with an emergency lane of 2.5 in one direction only and with an effective vertical-clearance of 4.5 m. In addition, a 0.75m wide maintenance-footpath is proposed. The overall width is excessive given the short length of the tunnels but is satisfactory from a road safety view point. 58. Because of the length of the tunnels and the low traffic volumes over the first ten years of operation no ventilation or other safety-facilities are required. This should be reassessed during detail design. 59. Based on current profile, there should be no visibility issues at the longer tunnel portals despite the combined effect of both horizontal and summit curve approaching some portals. The impact on visibility because of transition from dark to light or vice-versa while entering and exiting the tunnels should be assessed in detail in detail design, especially in the three small connected tunnels. 60. While there will be illumination and provision for emergency power, all vehicles should put their headlights on while traveling through tunnels to avoid accidents during sudden power- outages 61. Other tunnel safety aspects are discussed in Supplementary Appendix 2, Design.

9. Climbing Lanes

62. Climbing-lanes are proposed on grades greater than four percent to allow overtaking of slow underpowered vehicles and minimize dangerous overtaking maneuvers which are common on Georgian roads.

10. Axle-Load Control Complex

63. It is recommended that weighbridge stations be included on both bypasses to ensure truck overloading does not occur as this not only damages the pavement but is dangerous to other traffic. For the project road these would be placed in the proposed service areas on each bypass.

Engconsult Ltd. 13

IV. ROAD SAFETY COMPONENT

A. Coordination with World Bank projects 64. WB has been a major force in promoting road safety in Georgia. This has been done mainly by including road safety components in their road construction projects and through advice to MRDIRD and other GoG ministries and departments. 65. WB plans to include a specific road safety component in their Third East West Highway Improvement project which will, amongst other items develop road safety audit guides which they will use as a pilot application on the E-W Highway improvement work. 66. In order to complement this work it is recommended that the guides also be used on this project during and after construction7. An audit of RDMRDI sections of the existing road should also be carried out.

B. Awareness Campaigns for Road Safety 67. Education is an integral part of any road-safety strategy and is of significant importance for this project as it is a controlled-access road with significant lengths of tunnel. 68. The Project road will comprise of three sections - (i) access-controlled Kobuleti bypass (28km), (ii) the free-access six kilometer widening of the existing road, and (iii) the access controlled Batumi bypass (14km). The first and second of these sections comprise Tranche 1, while Tranche 2 consists of the third section. The first and third of these sections will be restricted, with no admittance to pedestrians, slow and non-motorized traffic, while there will be mixed traffic and pedestrians in the middle widened section. 69. The following paragraphs suggest awareness campaigns which could be included in the scope of the first tranche of the MFF. 70. The safety awareness campaign for the access-controlled section should take into consideration safety-issues related to high-speed traffic and the likely implication to both the road-users and roadside communities. The target-groups would therefore be (i) road-users and drivers of all types of vehicles and (ii) roadside communities. 71. For the former the issues which should be addressed are safe driving code of behavior while driving on an access-controlled road including road tunnels and education about safety facilities and response in the event of incidences such as accidents. 72. For the roadside communities, the issues to be addressed would be to educate the communities on the dangers of accidentally or forcefully breaking through the safety boundary fence to the access-controlled road. In order to develop good public relations with the roadside communities, it is also important to consult with them during detail design about, and to later publicize the design of, sufficient pedestrian and tractor underpasses and non-disruption of the existing local roads, foot-trails, etc. 73. Suggested campaigns should be designed by the detail design or supervision engineers and should include:

7 Also for detail design if the guides are ready.

Engconsult Ltd. 14

1. Safety Code for Access-Controlled Roads & Tunnels

74. This program should consist of educating drivers and Patrol Police (PP) about the correct code of behavior while driving in the access-controlled sections of the road including the tunnels. The program should consist of design of materials to be used in the training and the actual education/training should be conducted to all road-users through joint collaboration with Traffic Police and MRDIRD. 75. In the initial period, the education should be carried out through concerned associations of transport entrepreneurs and PP. In the latter part of the Program, consultants should update the existing manuals for drivers incorporating the above code. It is recommended that the syllabus for the written test for driving-licenses should include special reference to high speed roads. The following actions are therefore proposed for this Program: • Preparation of code of behavior for access controlled roads. o This will comprise introduction of various signs and marking for access-controlled roads, driving conduct in tunnels in case of fire-hazard or accidents o Provide information on various emergency facilities available • Provide trainer training to PP. • Conduct training/discussions at regional associations of transport operators. • Incorporate into and update existing driving manuals • Coordinate with PP to introduce the updated driving manual in written test for driving licenses. 2. Roadside Community Awareness Campaign

76. This program would carry out safety education of the roadside communities within the access-controlled section of the Project and would target school children and concerned residents within the immediate vicinity of the road. The campaign should be conducted through local community based organizations in collaboration with MRDIRD and PP. This Program could consist of the following: • Preparation of brochures highlighting code of conduct around access-controlled roads; safety implications; instructions regarding non-access for pedestrians to the project road and information about emergency services available along or close to the road. • Lectures and training to local trainers (PP or NGO). • Supervision of implementation of awareness campaign by local trainers to students and residents 77. Road safety brochures and posters should be posted in schools and village community centers.

3. Pedestrian Awareness Campaign for Open Access Section of Road.

78. This campaign should use posters and literature related to increased danger from heavy-vehicles and safe pedestrian behavior when walking along heavily trafficked roads. This program should be conducted in schools in Kobuleti and Batumi: • Prepare posters and other literature • Conduct lectures, mock sessions at schools with the assistance of MRDIRD and PP. • Evaluate effectiveness of campaigns

Engconsult Ltd. 15

4. Awareness Campaign for Heavy Vehicles

79. The main message behind this campaign will be to underscore the importance of observing safe-driving habits with special consideration given to pedestrians and non-motorized vehicles on mixed traffic road sections. This campaign should be conducted at truck and bus operator’s depots. The following activities should be considered: • Design and printing of brochure and stickers. • Conduct lectures at transport operators depots with the assistance of MRDIRD and PP. • Distribute brochures and stickers to transport operators. • Evaluate effectiveness of campaign. 5. Publicity through Television and Newspaper

80. The Project should produce and broadcast a documentary on television highlighting the following issues pertaining to the Project roads: • Code of driving behavior on high speed roads and tunnels. • Information on emergency facilities available on high speed roads. 81. The documentary could be aired several times a year and the possibility of airing the documentary on national television on a free basis should be explored. 82. For newspaper publicity a set of road safety advertisements/information should be designed and placed in national and regional papers. 83. It is estimated that a total of 6 person months will be required for an international road safety expert plus 16 person months of domestic road safety experts and two person months of a domestic media specialist. 84. In addition to he training and awareness programs set out above it will be important to increase and improve enforcement measures on the new road. As part of the World Bank road safety program for the E-W highway support is proposed to PP with training and equipment to deliver the enforcement element of the overall road safety plan. It is proposed that PP from Ajara who would patrol the Project roads be included in this training and that additional road safety enforcement equipment be procured through the project for use on the project and other roads in Ajara. This would include radar equipment for speed control, breathalyzer equipment to control drunk driving and onboard computer equipment to allow quick recording of accidents at the accident scene.

Engconsult Ltd. 16

V. INTELLIGENT TRANSPORT SYSTEMS

A. Introduction 85. Intelligent Transport Systems (ITS) is a term that is used widely (and loosely) to encompass a wide range of technological solutions applied to the transport sector to improve the performance of the transport system. These improvements include some or all of the following (and overlapping) functions: • Safety and traveler information (speed signs, warning signs, etc) • Enforcement (over-speed, over-weight, non payment of tolls, etc) • Monitoring (traffic speeds, over-weight, traffic counting, etc) • Toll and revenue collection (e.g. electronic tolling) 86. The relative emphasis placed on ITS for these functions varies from place to place, depending on factors such as specific need (arising from topographical and environmental, traffic volumes, delays and vehicle speed considerations), local technological skills and ability to reliably construct and maintain a relatively high technology system, and budget considerations for both construction and on-going maintenance.

B. The Current Situation

1. Review of ITS in Georgia

87. At the present time, other than for the simplest facilities such as fixed signs and traffic signals and the introduction of variable message signs (VMS) on a section of the widened E-W Highway, there are no ITS facilities in Georgia. There are also no immediate plans for ITS introduction. 88. With the construction of the project road, one of the first limited access high speed roads in Georgia, it is appropriate to consider the introduction of further ITS and to develop a plan and standards for future ITS facilities on other roads, including major urban roads.

2. International Practice

a. ITS Standards

89. With the wide variety of ITS hardware and software functions and the ever-increasing number of system suppliers, ITS standards are becoming ever more important. 90. In the United States, the US National ITS Architecture provides a common framework for planning, defining and integrating ITS, and defines: • Functional requirements (e.g. what monitoring, data collection, reporting is required); • Physical components (e.g. video surveillance cameras); and • Information flows and physical sub-systems of the overall integrated system. 91. Other countries that are introducing ITS are also developing and/or adopting similar ITS standards (e.g. European countries under common European standards). By specifying how systems and components inter-connect, the standards promote inter-operability. The standards do not attempt to specify or limit the type or supplier of any component: what the standards do is ensure that, no matter what component is used, it will talk to and be able to be understood by all other components.

Engconsult Ltd. 17

92. In Europe ITS standards have been developed and are frequently reviewed. These standards allow inter-operability between systems and between countries. The main focus of ITS in Europe now is the actual implementation of ITS. 93. The ‘Trans-European Network for Transport’ project is intended to allow a number of existing different and separate ITS to inter-connect. This is seen as an important step in developing a homogeneous ITS network across Europe. 94. Typical features of the pan-European system are: • Automatic vehicle location and tracking; • Traveler information, both before trip for trip planning and during trip, including Variable Message Signs (VMS), web cameras, internet road, traffic and weather status and on-board route planning systems; • Vehicle tolling / user charges systems; • Highway management, including incident detection and control; • Vehicle speed and weight monitoring and enforcement. 95. ITS in North America is well developed, and includes a wide range of applications across all transport sectors, using a wide range of communication systems and providing a wide range of services. 96. Common communication systems include GPS, internet, dedicated short range communications and FM broadcasting. Common services include: • Traveler information, • Traffic management; • Electronic toll collection; • Commercial vehicle management; and • Advanced vehicle safety systems (e.g. vehicle to vehicle proximity detection) C. ITS for the Roads in the project

1. Project Road Proposed ITS Arrangements

97. The specific ITS proposals for the Project road and other projects in the MFF should be defined during later stages of the project, however the following proposals could be considered: • Surveillance cameras on project roads at key locations (such as interchanges, tunnel portals and other ‘high risk’ areas) to give warning of incidents (potential or actual accidents etc; • Advisory maximum speed change display VMS; • Gantry-mounted driver information VMSs; • Vehicle weight detection on all international roads; and • Development of an ITS national standard for Georgia. 2. Recommended ITS for the Project

98. The specific features which could be incorporated on the Project are discussed below. Use of VMS and cameras will require a control building located which could be in Batumi or close to the project road in the proposed service areas.

Engconsult Ltd. 18

a. Surveillance Cameras

99. The main use of surveillance cameras is to quickly identify changed or emergency traffic conditions. In particular well supervised surveillance systems can pinpoint traffic accidents and alert emergency services and note deterioration of traffic conditions due to either weather or landslides etc. Cameras should be placed at all ‘high risk’ locations such as tunnel portals, major bridges, high cut areas and high accident risk locations.

b. Use of Variable Message Signs

100. VMSs need to be spaced close enough to provide timely information to road users, particularly in ‘high risk’ areas, on road conditions, accidents, weather and speed limitations. Where it is not practicable to provide this frequency of fixed location VMSs, then mobile VMSs should be used (and moved from location to location as required). 101. It is recommended that non-road related information not be displayed on VMS, including commercial advertising. This distracts the driver and its frequent use will detract from the urgency of important road, speed or weather information. VMS speed limit information is also important to advise drivers of reduced speeds either for incidents, road works or weather or change in design speed. This is particularly important at tunnel locations for incident management. VMS should be used in conjunction with the surveillance cameras.

c. Vehicle Weight Detection

102. A Weigh Bridge station is recommended to be installed on Kobuleti bypass as part of the Tranche 1 works. All loaded trucks should be weighed to avoid overloading of the road pavement and consequent accelerated deterioration. Overweight vehicles should not be allowed on the project road. Combine with the weigh station at Sarpi border post this should help ensure safeguarding the road pavement from overloading.

d. Adoption of ITS Standards

103. With the introduction of high standard, high speed, roads in Georgia, it would be appropriate to promote, at the national level, the need to develop and adopt national ITS standards. Given the amount of effort, time and cost developing and testing standards in other countries (e.g. United States) there may be considerable scope to draw upon that international experience as the basis for national ITS standards. This needs to be further discussed with Government.

3. Future ITS proposals

104. As a second step other ITS technologies could gradually be introduced.

a. Weather Condition Monitoring

105. ITS technology allows weather conditions that impact on road safety / operations to be automatically detected and to automatically update VMSs accordingly. Since the project roads will pass through mountainous areas that experience fog, snow and ice, this should also be incorporated into the ITS system.

Engconsult Ltd. 19

b. Speed Limit Monitoring and Enforcement

106. Speed limits can be monitored and enforced very easily with speed cameras. These can be at fixed locations or mobile. The key to enforcement is for motorists to consider that there is sufficiently high probability of being ‘caught’ (coupled with a sufficiently high penalty) so as to deter speeding. In order to incorporate this technology there needs to be good coordination between the road authorities and the patrol police. 107. The initial ITS installation should allow for fixed location speed detection to be later incorporated into some overhead VMS gantry, enforcing speed limits every 20-30 miles of travel.

c. Automatic Toll Collection

108. It has been assumed that initially if any tolls are levied on project roads they will be collected manually. Plans should be prepared, however, for future electronic system of toll collection and the system initially installed should allow for an easy transition.

d. Management Information Reporting

109. The ITS should be designed to so that it can readily produce key management information reports, such as vehicles using the roads, by direction, by vehicle type and by time period, number of overloaded vehicles and number of speeding vehicles. A series of simple standard reports should be designed to clearly show key trends (e.g. changes in traffic volumes and changes in the proportion of overloaded vehicles). Non-standard reports should also be able to be easily specified and produced. Selected reports should be made available to planners, maintenance companies, enforcement agencies and prospective investors in the roads. 110. ITS recommendations are summarized in Table 3. Table 3: ITS Recommendations Item Recommendation Comment Immediate VMS Displays Every 20-30 minutes Fixed location variable (depending on speed), with 20-30 kms apart on average message signs particular attention to location near ‘high risk’ locations Every 20-30 minutes Fixed location variable speed (depending on speed), with 20-30 kms apart on average signs particular attention to location near ‘high risk’ locations Able to be rapidly deployed in 3-4 in each 50-70 km of Mobile VMSs case of major incidents / lane international road. restrictions, etc Video Surveillance On average around 3 km To cover as much as possible apart, subject to sight lines of the project road. Later Cameras with variable focal extension to full coverage. Fixed location video length. Could also later be integrated surveillance. Cameras able to be rotated / with vehicle speed, weight and raised / lowered toll payment monitoring and Cameras to provide full enforcement at selected

Engconsult Ltd. 20

Item Recommendation Comment coverage within tunnel / on locations. major bridges / at Tunnel cameras should adjust interchanges for variable brightness. Vehicle Weight Monitoring and

Enforcement Overweight vehicles Loaded trucks weighed at Fixed location photographed or each end of project road. apprehended. Future Weather Monitoring In locations where roads are Would update preceding Visibility subject to fog, snow, ice VMSs. Should be taken in to Wind speed and direction In location where roads are account in determining subject to high cross winds location of VMSs Speed Monitoring and

Enforcement Speeding vehicles Fixed location At each overhead VMS gantry photographed. By Police, focusing on high Variable location speed / high risk areas

Engconsult Ltd. 21

VI. LONG-TERM MEASURES TO ACHIEVE TUNNEL AND ROAD SAFETY

111. In order to provide long term improvements in road safety improvements must be made and maintained in engineering, education and enforcement. 112. In engineering, daily safety inspections and regular safety-audits should be conducted throughout the project road with particular emphasis on safety-structures such as safety-barriers, road marking and signing, bridge and culverts, hazardous or critical locations and tunnels. 113. The education aspects outlined above should be sustained on a long-term basis. This is crucial for roads which allow high-speeds and introduces new concepts. 114. The ITS measures outlined above should be gradually incorporated on all international roads. 115. Lastly, strong enforcement of traffic rules along the road and in tunnels with stiff penalties for traffic violators is necessary. In this aspect, Patrol Police should be well equipped with all necessary logistics (patrol vehicles, equipment) and be adequately staffed. A system of introducing a toll-free number for emergencies with direct communication with Patrol Police should also be introduced8 so that road-users themselves can assist in identifying incidences or road-accidents. The number to be used should be well publicized at intervals along the road.

8 This has been suggested in the Georgian Road Safety Strategy

Engconsult Ltd. 22

VII. RECOMMENDATIONS

116. The previous sections discussed all issues pertaining to the project including RSAs, safety implications during operations awareness campaigns and implementation of ITS. Due to importance and complexity involved, it is pertinent that all aspects identified in the RSAs are backed by implementation and that awareness and educations aspects be implemented. 117. RSAs should be conducted during construction and operation to monitor whether previous audit recommendations have been implemented or not. 118. For effective traffic and incidence management, adoption of various ITS technologies particularly variable message signs and speed cameras, etc, should be considered in the future. 119. The project should assist Police to revise the safe driving manual so as to include safety aspects, traffic rules and code of behavior on high speed roads and road tunnels and they should include these subjects in the syllabus for written examination while applying for driving licenses. 120. National (ITS) standards for effective and efficient operation of major roads should be developed and ITS components gradually introduced on international roads. Table 4 summarizes tentative total cost of all road-safety components except cost of following structures and components: Table 4: Cost Estimate for Road Safety Components Total Description ('000 US$)

1 Safety Code – High speed road 70

Roadside Community Awareness - Access-Controlled 2 Section 50

3 Pedestrian Awareness for School-Children 15 4 Awareness for HV Drivers 15 5 Road-Safety Publicity in TV 60

6 Road-Safety Publicity in Newspaper 15 7 Enforcement equipment for PP 100 ITS

7 Provision of weigh bridges on bypass roads 60 8 Provision of static and mobile VMS on bypass roads 100 9 Installation of surveillance cameras on bypass roads 150 10 Development of National ITS standards 80 11 Road Safety Enforcement Equipment 100 Total for Safety Component 715

Engconsult Ltd. FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 5 Terms of Reference for Detailed Design

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Content

I. Consultants Terms of Reference ...... 1 A. Background ...... 1 B. Project Road Description ...... 1 C. Objective and Outcome ...... 2 D. Scope of work ...... 2 E. Terms of Reference ...... 3 1. Engineering Design ...... 3 2. Contracts ...... 5 3. Traffic Engineering and Road Safety ...... 6 4. Financial ...... 6 5. Social/Resettlement ...... 6 6. Environment ...... 7 7. Project Framework and Monitoring of Impacts ...... 7 8. Supervision Consultants ...... 7 F. Reports and Timing...... 8 1. Resettlement Action Plans ...... 8 2. Draft Engineering Drawing and Tender Documents ...... 9 3. Final Engineering Drawings and other Documents ...... 9 4. Penalties...... 9 G. The Consultants ...... 9 1. The Required Expertise ...... 10 H. Requirements for key staff...... 10

Engconsult Ltd. 1

I. CONSULTANTS TERMS OF REFERENCE

A. Background 1. Georgia is located south of the Caucasus mountain range, with Russia to the north, Armenia and Turkey to the south, Azerbaijan to the east, and the Black Sea to the west. It has a population of 4.5 million. Georgia, due to its geographic location, provides the shortest transit link between Central Asia and Europe. Hence, transport plays a pivotal role in supporting the national economy, and development of the transport sector is vital to increasing economy of the region through reduced transport costs and increased transit revenues. 2. The 81 km Poti – Batumi – Sarpi road along the western coast of the country, mostly located in the Ajara Autonomous Republic, is a key highway and international transit route in Georgia. It is connected to the major Black Sea ports of Georgia, viz. Batumi and Poti, and a number of beach resorts including Batumi and Kobuleti. Due to heavy traffic on this road there has been significant increase in traffic congestion and accidents particularly at Batumi and Kobuleti during the tourist season. The Government plans to construct two bypass roads around Batumi and Kobuleti to address these problems and with ADB assistance is carrying out a feasibility study and preliminary design. The study, (TA-7059-GEO), has identified a 48.4 km alignment between Choloki River north of Kobuleti to Chorokhi River south of Batumi, to bypass Kobuleti, Makhinjauri and Batumi. The Government requested ADB’s help to prepare the detail design and construction drawings and finance the construction of the bypass roads. The road will be constructed with the help of ADB financing. The Roads Department of the Ministry of Regional Development and Infrastructure (RD) will be the executing agency. B. Project Road Description 3. A full description of the alignment proposed in the Preliminary Design is set out in he attached alignment report and maps. 4. It is proposed to construct the road with two lanes only on all sections except for six kilometers where the existing two lane road will be widened to four lanes. The design cross sections should, however, show both two and four lane design. All horizontal design parameters shall conform with a four lane road, except that on the two lane sections the road pavement crossfall shall conform to two lane road standards. The right of way of the road and land acquisition shall be sufficient for a four lane road and shall be not less than 50m, except on the urban section of road (approximate km 32-34), where the road widening is constrained by existing railway right of way and buildings. 5. The proposed road can be spilt into four sections: • Section 1. The first 14 kms is in flat terrain with resettlement required. It includes three low level river bridges, one road overpasses, one rail overpass and two grade separated interchanges. • Section 2. The next section from Km 14 to 28 is in rolling terrain with five river bridges (two major), eleven road overpasses, one 400m tunnel and one grade separated interchange. • Section 3. The next section from Km 28 to 33.8 involves widening of the existing two lane road to four lanes.. • Section 4. The final section (Batumi bypass) from Km 33.8 to 48.4 is in mountainous terrain. It includes 3 small (<200mt) and two medium (400m and 700m) tunnel. It also includes 14 bridges including two major river and two long valley crossings, and two grade separated interchanges. Box culverts are recommended for minor road crossings .

Engconsult Ltd. 2

C. Objective and Outcome 6. The main outcome of the consulting services will comprise • final detail design and construction drawings of the road based on the feasibility/preliminary design alignment (shown on the plan provided with the bid documents) to provide an access controlled two and four lane high class highway from Choloki Bridge to Chorokhi Bridge, with detail construction drawings and full BOQs and Contract Documentation suitable for international bidding; • updating of the preliminary EIA (based on preliminary design) based on final design alignment following ADB and Government of Georgia (GoG) requirements and preparation of an Environmental Management Plan; • preparation of updated social and resettlement action plans based on the final design alignment; and • preparation of a road safety program along the road corridor relating to particular risks of high speed roads. D. Scope of work 7. The broad scope of work includes: 8. Preparation of detail design construction drawings and documentation of about 48km of high standard road from Choloki to Chorokhi, including 6 (about 1.95 km total) of single tube tunnels and about 37 (6,640m) small and large bridges. Preparation of BoQs suitable for international construction contracts and full bidding documents. More details of the proposed bridges and tunnels are set out in the alignment report. 9. Assist RD throughout the contract tendering process in preparation of tendering documents and procurement. 10. Carry out social, resettlement, road safety and environmental assessment, based on the final alignment and complete all statutory documentation to achieve compliance under ‘The Laws on Environmental Impact Permit and on Ecological Examination, 2008’and the Georgian laws/ legislation on land acquisition and resettlement, ADB’s Policy on Involuntary Resettlement, 1995 and ADB’s Policy on Indigenous Peoples, 1998 . 11. Ensure that the detailed design minimizes the land acquisition and resettlement by avoiding sensitive habitat areas and by adopting best engineering solutions. Ensure that social impact assessment is carried out and the Land Acquisition and Resettlement Plan (LARP) is updated using a detailed measurement survey based on the final alignment. Ensure that subproject-specific social mitigation measures are incorporated into contract documents. Ensure that necessary input is provided to the RD during the implementation of LARP for the initial contract packages. 12. Ensure, through consultation with affected people, that all existing rights of way across the alignment are maintained. In particular ensure that sufficient pedestrian and agricultural vehicle under or overpasses are included. 13. Ensure that all environmental mitigation measures as recommended in the EIA are incorporated in the final project design and that the recommendations for environmental management plans are included in the tender documentation and that all the necessary resources and capacity building to implement the environmental mitigation measures, as recommended in the EIA and EMP, are identified and costed appropriately in the final project design. 14. Preparation of a road safety improvement program; and

Engconsult Ltd. 3

15. Review benchmark indicators and monitoring system prepared during preliminary design in line with ADB's Guidelines1 for Preparing a Design and Monitoring Framework and collect necessary remaining baseline monitoring data. E. Terms of Reference 16. The terms of reference (TOR) for the consultants include but are not limited to the following; 1. Engineering Design

a. Detail Design of the Project Road:

(a) Carry out topographical and hydrological surveys/studies for detailed design and quantity computation purposes. Establish survey control points giving horizontal coordinates relative to national grid and bench marks giving levels relative to the national datum. The control points, which shall also be tied to the UTM coordinate system, shall be permanent and shall be located in pairs with an interval of 3-5 km. Each pair shall be located 200-250m apart with unrestricted visibility between them. Twelve control points have been placed during detail design and tied to the UTM coordinate system. The coordinates will be supplied to the consultant and use can be made of these points. Cross sections should be measured not more than 20 m apart and will range from 50m wide in the flat areas to up to 100m in the hilly/mountainous areas. (b) Review the preliminary design alignment and proposed highway design standards and suggest alternatives if necessary. Detail design shall use the Georgian RD standards where appropriate and other agreed international standards. (c) Review available data relating to traffic and pavement strength. Carry out additional classified traffic counts and axle load surveys as required and estimate present and future traffic and axle loads to a level of accuracy necessary for designing appropriate pavement structures in accordance with best international practice2 as may be agreed. The consultant will prepare a “whole of life” comparison of asphalt concrete and cement concrete and recommend which pavement type to use. Detail pavement design should not be completed until agreement is obtained from RD. The consultant is encouraged to investigate alternate lower cost environmentally friendly pavement designs for consideration by RD. (d) Carry out a detailed study of the availability of construction materials such as aggregates, water, bitumen, cement, steel etc. and possible access to the existing quarry sites. Investigate, test and define new sources and available quantities of construction materials and prepare a material sources map and report with indicative properties and quantities. (e) Identify acceptable spoil areas to be clearly designated in the contract documents; and quantify the haulage requirements based on the material and spoils locations. (f) Survey any existing services, irrigation systems and oil pipelines that may cross the project road or are located within the road reserve and liaise with service owners and authorities to determine requirements for service diversions, improvements, and/or protection works. The Consultants shall prepare detailed plans and proposals for the relocation of the services affected by the proposed road work, including volume and quantities of required work. (g) Review existing hydrological and drainage data and collect such further data as may be required. Check all stream beds and cross drainage channels above and below the road alignment for possible erosion effects; design and quantify any necessary protective works;

1 www.adb.org/Documents/guidelines/guidelines-preparing-dmf/default. 2 e.g. UK Transport Research Laboratory, Overseas Road Note 31

Engconsult Ltd. 4

Specify, design and quantify side drainage and line drains where necessary to eliminate scour and erosion and to carry surface and subsurface water from the road formation; (h) Assess the requirement for slope protection measures adjacent to the road, and design cost- effective remedial works. (i) Prepare detailed design construction drawings for all elements of road works. Plan and longitudinal profile will be at the scale of 1:1,000 showing existing roads, tracks, rivers, ground levels, formation and design levels, gradient etc. Cross-sections will be drawn at 20 meter interval at the scale of 1:100. b. Tunnel Design

(j) Carry out a geological and hydrological survey of the rock mass comprising the tunnel corridor. (k) Review the preliminary design of the 6 tunnels (approximate lengths ranging from 100m to 700m) and, in co-operation with the Highway Design Engineers, establish final alignments for the tunnels. (l) Carry out a detailed design and prepare construction drawings of tunnel civil works including, but not limited to, tunnel support design, secondary lining for various selected tunnel types, design of tunnel portals including slope cut, slope protection and surface drainage, design of drainage behind the secondary lining and invert, (m) Carry out detailed design for ventilation3 and illumination for each tunnel including fixing details and preparation of particular specification and BOQ. The selection of the ventilation system should include an automatic monitoring and management systems to ensure normal climate conditions in the tunnel. Design power-saving systems for each tunnel, considering management of the transition between the outside daylight and artificial lighting in the tunnel. Design should include independent transformers. Design and report on a suitable fire fighting system and specify suitable equipment for analyzing toxic gases, traffic lights, video system, devices for dimension control and loudspeaker communication and warning systems for tunnels over 400m. c. Bridge Structures:

(n) Georgia has experienced seismic activities with earthquakes registering magnitudes up to 8 on the Richter scale; the design shall take this constraint into consideration with particular attention to specifications for bridges. Review all available seismic data and information relevant to the bridges and recommend seismic design acceleration based on the Georgian Seismic Standard. (o) Examine possible bridge formats considering both structural steel and concrete and obtain the RD’s statement of no objection to the recommended formats. (p) Review the work done during preliminary design. Determine maximum river flood levels and review possible scour risk for each bridge. (q) Prepare detailed design calculations for each bridge or other highway structure in accordance with the agreed relevant design codes and standards and prepare detail drawings suitable for construction. Prepare a Bill of Quantities for each bridge. (r) State of the art investigation is expected for the foundations of bridges and structures to be constructed.

3 Ventilation has not been specified in the preliminary design but this should be reviewed after traffic forecasts and tunnel lengths are updated.

Engconsult Ltd. 5

(s) The design of all structures shall consider site access such that costs of any improvements to existing roads and road connections from existing roads to the bridge sites are not excessive; the cost of such improvements and new construction when combined with the bridge construction costs shall be minimized. (t) Prepare a cost estimate of each bridge and other structure including the cost of roads improvements and temporary roads. (u) Drawings for structures will use various scales between 1:100 and 1:10 depending on the structure and requirement for readability of the drawings. d. Testing and Survey

(v) The Consultants shall review any relevant data and undertake additional soils and materials investigations as required to analyze the material properties of base and sub base layers, and subgrade/roadbed soil. The Consultants shall investigate suitable materials for road embankment, pavement and structures and propose using the most economical solutions. The Consultants should confirm or defined the locations for borrow pits and asphalt plants installation if needed. The list of drilling, sampling and test activities to be performed during the contract at all bridge, tunnel and other sites shall be described in the technical offer of the consultant and costs should be detailed. The list may be adjusted during the project and reported upon by the Consultants. (w) The Consultants shall incorporate any temporary works and/or diversions required during the construction period. All temporary works or diversion designed or proposed should be able to cater for the uninterrupted flow of traffic for the period concerned. The Consultants shall use agreed Georgian or international standards for traffic control at temporary road works and diversion sites. The Consultants shall ensure that adequate provisions are made for access to all adjacent properties during the construction. The Consultants shall prepare proposals for traffic management and obtain the RD’s agreement to these proposals. They shall ensure that adequate provision is made in the bidding documents for contractors to be responsible for all aspects of traffic management during construction, including operations of sections of single lane working, construction and maintenance of temporary diversion roads to an acceptable standard and maintenance of any existing secondary roads utilized as diversion roads 2. Contracts

(x) For a time schedule for production of contract documentation and drawings see ‘F. Reports and Timing’ below. (y) Review the preliminary contract packages set out in the preliminary design. Prepare a schedule of appropriate contract packages suitable for international bidding and detailed project implementation schedules showing anticipated progress of works and expenditures for each contract package; (z) Produce complete detail drawings, design and construction specifications and detailed BOQ and other tender documentation suitable for procurement under ICB in accordance with ADB's Guidelines for Procurement; (aa) Produce a comprehensive Engineer's Estimate for the cost of works in each contract and assist Government with procurement of civil works; in particular review Tenders, Contractor’s design proposals and construction supervision; (ab) Prepare a construction technical specification defining materials used in construction and minimum workmanship standards for all aspects of the work.

Engconsult Ltd. 6

3. Traffic Engineering and Road Safety

(ac) Review the preliminary design forecast for normal, generated and induced traffic flows for 20 years and amend if necessary taking into account domestic and foreign/transit trade and any new development projects which might affect traffic on the Project Road. Carry out any additional traffic survey as required. (ad) Carry out traffic safety studies as required and design and carry out a road safety audits of the detail design and review the road safety component for the project, including road safety awareness campaigns prepared during preliminary design and to be implemented as a separate TA project. (ae) Coordinate safety aspects for tunnel construction, operation and maintenance point of view; Propose long term measures to achieve tunnel safety standards 4. Financial

(af) Prepare an overall project cost estimate table for the proposed investment, taking into account all relevant financial costs and benefits. Prepare a project financing plan, including proposed ADB lending, counterpart funds and other loans. (ag) Review possible toll levels for an Operation and Maintenance (O&M) concession for the Batumi bypass as discussed in the preliminary design and amend if necessary. Identify project revenue and cost risks and conduct relevant sensitivity analyses. If required by RD/ADB prepare a model O&M concession contract with a discussion paper for Government setting out full explanation of the concession and advice on areas of negotiation. (The financial analysis will be guided by, and outputs prepared in accordance with, ADB's Guidelines for the Financial Governance and Management of Investment Projects, ADB's Operations Manual, ADB's Project Administration Instructions, and the ADB Loan Disbursement Handbook). 5. Social/Resettlement

(ah) For the detailed design phase the consultants will review the Land Acquisition and Resettlement (LAR) work carried during the preliminary design and, following the finalization of the alignment (to be achieved by the first 40 days of detailed design work), they will carry out the following tasks: • Review the Land Acquisition and Resettlement Plan (LARP).. If some sections of the road are realigned during detailed design, resurvey those sections and update the data base on impacts and Affected Persons (APs) as needed and prepare final (LARP).. • Carry out a final consultation with all APs. • Jointly with the local government provide assistance to the land owners (preparation of land parcels cadastrial drawings together with electronoic version) during land registration process to ensure that all APs are legalized. • Assist EA during land acquisition process to ensure that compensation is paid to the land owners and provisions of Land Acquisition and Resettlement Framework (LARF) are followed. • Update/re-scope the LARPs already prepared including figures on expropriation. • Assist the external monitoring agency authorized by ADB in the preparation of a final compliance report proving that all APs have either been compensated as required by the

Engconsult Ltd. 7

LARPs or have been expropriated in accordance with the procedures defined under the Law on Expropriation.. (ai) Update the social impact assessment in the influence zone as required and assess the impact of project on poverty reduction. Assist RD to implement information campaigns and stakeholder participation (aj) Prepare of updated (final) LARP in for RD and government, also for the disclosure to APs based on ADB’s requirements; (ak) Coordinate with NGOs and various agencies during RP implementation in terms of availability of information regarding ongoing procedures and disclosure to public in line with the current legislation. (al) Prepare training modules and dissemination materials and conduct workshops with RD and Rayon officials and NGOs on the issues concerning principles and procedures of land acquisition, entitlement and compensation disbursements, grievance readdressing and monitoring of resettlement operations; (am) Assist RD and liaise with ADB during LARP implementation. In particular prepare a scheduled action plan detailing all tasks for the compensation program and monitor the implementation of compensation. Coordinate with various agencies for timely land acquisition and disbursement of compensation to AP. 6. Environment

(an) Conduct further detailed field and desk studies that will lead to the submission of all necessary statutory and other required documents to achieve environmental clearance for the Project. Ensure that all potential contractors are made aware of requirements for all environmental mitigation measures including EIA and EMP through the tender documentation. (ao) Review and verify the assumptions, assessments and recommendations made in the EIA made under TA 7059 –GEO and design and supervise surveys and updating of all necessary documentation based on the final detailed designs. Environmental Management Plans will be prepared for all roads sections to guide the management of environmental impacts during construction.

7. Project Framework and Monitoring of Impacts

(ap) review and/or edit the draft project framework in the preliminary design report following ADB's Project Performance Management System (PPMS). Collect/develop additional baseline indicators for monitoring the impact of the project as required. 8. Supervision Consultants

(aq) The Consultants shall prepare ToR for the supervision consultants based on the draft ToR prepared during preliminary design. e. Procurement

(ar) The consultants shall assist RD throughout the tendering processes up to selection of contractors in the preparation of bid documents, and pre-qualification (PQ) documents, and in the evaluation of PQ and bid proposal.

Engconsult Ltd. 8

F. Reports and Timing 17. The detail design and all associated work will be completed within eleven (11) months of the commencement of consulting services (CC). Full documentation for two construction contracts of approximately $60 million total construction cost will be completed within four (4) months of the CC. These contracts will be agreed with RD and ADB but it is anticipated that they will be at the northern area of the alignment, one in flat terrain (0-14km) with three low level two lane river bridges, road and rail overpasses and two interchanges, and the second upgrading of the existing road to four lanes (28-34km) with one two lane bridge. A further contract in rolling/mountainous terrain (14-28km), including one single bore (two-way) tunnel and two major and three minor low level bridges will be completed in seven (7) months of the CC. The detail time schedule will be: Time from CC Item 3 months Updated LARP for two contracts Draft contract documents and draft 3.5 months final drawings for two contracts 4 months Final Design Drawings and BQ for two contracts 5 months Updated LARP for third contract Draft contract documents and draft 6 months final drawings for third contract 7 months Final Design Drawings and BQ for third contract 9 months Updated LARP for remaining projects 10 months Draft contract documents and draft drawings and BQ for remaining construction contracts 11 months Final design Drawings and BQ for remaining contracts

18. Unless otherwise specified below the consultants will submit 2 copies of all reports, technical working papers, and progress reports to ADB in English, except for the technical drawings, of which only one copy should be submitted to ADB; and 5 copies of each to the Government in English and Georgian. The timetable for submitting the reports will be: • An inception report setting out the initial findings, review of preliminary design and alignment alternative, and detailed work schedule and plan, within 4 weeks of the commencement of consulting services; • Monthly progress reports at the end of each month, outlining the progress of work during the previous month, the work program for the subsequent month, and major issues to be addressed. 1. Resettlement Action Plans

19. The Consultant shall submit one set of Resettlement Action Plans suitable for land acquisition on reproducible stable medium and 6 copies thereof to the Government, whenever any change in horizontal alignment is involved. The plans shall be made available to the Government as the work progresses to facilitate timely action for acquisition of the necessary right-of -way.

Engconsult Ltd. 9

2. Draft Engineering Drawing and Tender Documents

20. One month in advance of the date on which final engineering designs and tender documents are due, the consultants shall submit draft final plans and bid documents (6 copies of the complete set to the Government and 3 copies of bid documents and 1 copy of engineering drawing (A3 size with A1 size soft copy) to ADB) suitably bound and indexed. The Government will review these designs and documents in consultation with ADB and furnish their comments to the consultants within 21 days after receipt. The consultants will also submit in English and Georgian (5 copies to the Government and 2 copies in English only to ADB) the following documents suitably bound: • detailed cost estimates for each contract with work item rate analysis. • complete design calculations for road works, structures and tunnel presented in a form which enables calculations to be checked independently; and; • Geo-technical and materials report. 3. Final Engineering Drawings and other Documents

21. The consultant shall furnish 15 copies of all engineering drawings (A3 size with A1 size soft copy), specifications, pre-qualification documents and bid documents including geo- technical and material reports to the Government in English and Georgian and one copy (A3 size drawings) to ADB. In addition, the consultants will submit all original engineering drawings together with 5 sets of contract prints, detailed cost estimates and design calculation to the Government and two sets of detailed cost estimates of design calculations to ADB. In addition, the consultant will also submit all engineering design drawings, estimates, design calculations and reports to the Government in two sets in a digital format (CD ROM). 4. Penalties.

22. In the event that the consultant does not keep to the contract terms and conditions penalties will be applied, such penalties will be specified in the contract between the consultant and the Government. In particular there will be penalties if the consultants exceeds the time schedule set out in these Terms of Reference.

G. The Consultants 23. The Consultants are expected to have extensive experience in high standard road design to international standards, including tunnels and major bridges, and relevant previous experiences in highway detailed design in the region. The consultant shall demonstrate capability in designing to major international design codes (British, American, etc. as appropriate). 24. The Consultant will work closely with the staff of the Ministry of Regional Development and Infrastructure, Road Department, local officials, and other related ministries/agencies, as well as Asian Development Bank (ADB)'s Georgia Resident Mission. The consulting firm will be responsible for all aspects of the project.

Engconsult Ltd. 10

1. The Required Expertise

25. The team proposed by the consultant shall be composed of skilled and experienced specialists, who will carry out necessary engineering works that may be required to perform the services. The team should be, at a minimum, composed of the following:

Required Expertise a. International Experts Team Leader/Senior Highway Engineer Bridge / Structural Engineer Tunnel Engineer Geotechnical / Pavement Engineer Tunnel Ventilation & Lighting Specialist Contracts and Procurement Specialist Financial Specialist Environmental Specialist Social Development/ Resettlement Specialist Geologist1 Hydrologist / Drainage Engineer1 Traffic/Road Safety Engineer1

b. Domestic Specialists Highway Engineer Bridge / Structural Engineer Tunnel Engineer Survey Engineer Contract & Procurement Engineer Hydrologist / Drainage Engineer Geologist / Geotechnical Engineer Traffic Engineer Environmental Specialist Resettlement / Social Specialist 1. May be replaced by domestic specialists if suitably qualified and experienced

26. The team should also include cost estimator, CAD draftsmen, translators (with knowledge of technical terminology), and other counterpart staff. 27. It is anticipated that up to 100 person months of professional international and 170 person months of professional domestic staff may be required, not including technical and administrative support staff, however, the consultants should note that the required expertise and total person months are indicative and the consultant is responsible in ensuring that the staff nominated are sufficient to complete the works successfully. H. Requirements for key staff. f. Team Leader/ Senior Highway Engineer

28. The team leader must have expertise in all aspects of high speed road design with experience in leading a team of multi-discipline experts. The team leader will have overall responsibility for implementing the project and managing the international and domestic team of

Engconsult Ltd. 11 consultants with expertise in tunnel, bridge and road engineering, cost estimation and contract documentation, BOT concessionary contracts, geology, traffic engineering, road safety, social development, and environment. 29. He/she should be a senior engineer with a minimum of 15 years of relevant experience covering road design and construction projects in various countries and preferably having experience in Georgia. He should have good communication and reporting skills. Highway Engineer 30. Should be familiar with the design of urban and rural highways or other major road schemes incorporating major bridges, tunnels and major grade separated interchanges, should have a minimum of 12 years experience and preferable having experience in the region. He/she should be familiar with highway design software as well as detailed electronic designs for contract drawings. Bridge/Structural Engineer 31. The Bridge/Structural Engineer should be a qualified structural engineer with at least 15 years experience in bridge and structure engineering, including at least 10 years of assignments in developing countries including the Caucasus region. The candidate should have a thorough understanding and experience with international standards and “best practices”, and of modern bridge construction. Tunnel Engineer 32. Should have a minimum of 15 years experience in the design of road and other tunnels in mountainous areas including preparation of specifications and evaluation of construction methods. Geologist 33. Should have a minimum of 10 years experience in geological investigation for roads, bridges and tunnels. Ventilation Specialist 34. Should have a minimum of 10 years experience in detail design of tunnel ventilation systems and fixing details. Tunnel M&E Specialist 35. Should have a minimum of 10 years experience in detail design of tunnel lighting and other mechanical and electrical details and fixing details. Contracts Specialist 36. Should have a minimum of 10 years experience in the preparation of Contract Documents and Specifications for major road or infrastructure projects using FIDIC and evaluation of PQ and bid proposals. He/she should be familiar with ADB standard bidding documents and procurement guidelines. Geotechnical / Pavement Engineer 37. The Geotechnical/Pavement Engineer should be a senior engineer with university degree in his/her field, or higher with a minimum of 15 years of relevant experiences in pavement design of road projects with extensive knowledge in materials of roads and pavement and materials investigations. He/she should be familiar with international pavement design guidelines and state-of-the art pavement construction technologies. He/she should be familiar with the preparation of Contract Specifications for materials and testing.

Engconsult Ltd. 12

Hydrologist/Drainage Engineer 38. The Hydrologist/Drainage Engineer should have a minimum of 15 years relevant major highway and bridge experience designing drainage systems using international codes and methods. He/she should be familiar with local condition, calculation and design methods. Financial Specialist 39. Should have a minimum of 15 years experience. Should have proven experience in financial analysis of major road or infrastructure projects using loan financing and experience in drafting infrastructure concession contracts. Traffic/Road Safety Engineer 40. Should have 15 years experience in traffic analysis and forecasting and road safety analysis and design of road safety programs. Social Development/ Resettlement Specialist 41. The specialist should have a master’s degree in social science with a minimum 10 years of work experiences. Up- to- date knowledge of ADB’s safeguards policies and procedures, particularly on social impact assessment, poverty assessment, land acquisition, and resettlement, and its implementation are desirable. Experiences in ADB funded projects or projects funded by multilateral agencies in the transport sector will be preferable. The specialist shall also have experience in working in multidisciplinary teams with good communication skill. 42. The National social development and resettlement specialist should be a graduate in social science and should have 5 years of working experience. The specialist should be familiar with the Georgian laws/legislations/regulations and procedures related to land acquisition and resettlement in Georgia. Experience in similar projects will be preferable. Environmental Specialists 43. The Environmental Specialist (ES) shall have at least 15 years experience and familiarity with all aspects of environmental management and with significant experience in environmental management and monitoring of projects, environmental assessment and / or implementation of environmental mitigation measures on construction projects. The ES shall also have experience working in teams of multi-discipline experts and leading a national team of consultants. Candidates with higher degrees in environmental engineering or environmental science or environment management are preferred. 44. The Environmental Specialist National (ESN) shall at least be a graduate in environmental science, environmental engineering or a related discipline with significant experience in environmental management and monitoring of projects, environmental assessment and / or design and implementation of environmental mitigation measures.

Engconsult Ltd. FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 6 Traffic Forecasts and Economic Evaluation of the Project

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Content

I. TRAFFIC SURVEYS 1 II. EXISTING TRAFFIC 1 A. Historic traffic on the existing road 1 B. Base year traffic 3 C. Traffic flow pattern 4 III. FUTURE TRAFFIC DEMAND 5 A. Introduction 5 B. Growth of Normal Traffic 5 1. Introduction 5 2. Elasticities 5 3. GDP and traffic forecasts 7 C. Corridor diversions 7 1. Introduction 7 2. The Samtskhe-Javakheti road project 8 3. Adopted corridor diversions 8 D. Adopted traffic 8 E. Diversion from other modes 14 IV. ECONOMIC EVALUATION 15 A. Approach 15 1. Introduction 15 2. Alternatives considered 16 3. Evaluation period and residual value 16 4. Pricing assumptions 17 5. Project benefits 18 B. Evaluation of project 1 – the Kobuleti bypass 26 1. Road sections 26 2. Costs 28 3. Evaluation results 29 C. Evaluation of tranche 2: the Batumi bypass 31 1. Road network 31 2. Costs 31 3. Evaluation results 33 D. Benefits distribution 34 E. Combined project evaluation (tranches 1 and 2) 35 1. Base case 35 2. Case B traffic 36

Engconsult Ltd. ii

F. Sensitivity and risk 37 1. Tranche 1 (Kobuleti) 37 2. Tranches 1 and 2 40

List of Tables

Table 6-1 Traffic surveys 1 Table 6-2 Historic traffic at project sites 2 Table 6-3 Foreign registered traffic 2 Table 6-4 Base year traffic on existing road sections 3 Table 6-5 Seasonal traffic flow pattern 5 Table 6-6 GDP and traffic growth rates 7 Table 6-7 Base case traffic (case A): no corridor diversion, veh/day 11 Table 6-8 Case B traffic: diversion to the Kartsakhi corridor, veh/day 11 Table 6-9 Tranche 1 traffic projection, case A (base case), veh/day 12 Table 6-10 Tranche 1 traffic projection, case B, veh/day 13 Table 6-11 Tranches 1 and 2 traffic projection, case A (base case), veh/day 14 Table 6-12 Standard conversion factor calculation 17 Table 6-13 Typical road user costs 20 Table 6-14 Basic vehicle characteristics 21 Table 6-15 Economic values of vehicles, labor and time 22 Table 6-16 Fuel and oil prices 23 Table 6-17 Accident statistics, Sarpi-Poti road 24 Table 6-18 With and without project accident rates 25 Table 6-19 With and without average numbers of accidents per year 25 Table 6-20 Kobuleti: existing and bypass road sections 27 Table 6-21 Kobuleti bypass economic costs 28 Table 6-22 Kobuleti bypass maintenance costs 28 Table 6-23 S-2 maintenance expenses, 2007-8 29 Table 6-24 Do minimum maintenance costs 29 Table 6-25 Tranche 1 (Kobuleti) base case: economic evaluation 30 Table 6-26 Tranche 1 (Kobuleti) base case: costs and benefits 30 Table 6-27 Kobuleti traffic case B: discounted costs and benefits 31 Table 6-28 Batumi bypass economic costs 31 Table 6-29 Tranche 2 – Batumi bypass: existing and bypass road sections 32 Table 6-30 Batumi bypass maintenance costs 33 Table 6-31 Tranche 2 (Batumi) base case economic evaluation 33

Engconsult Ltd. iii

Table 6-32 Tranche 2 (Batumi) base case: costs and benefits 34 Table 6-33 Benefits distribution, tranche 1 35 Table 6-34 Journey time savings 35 Table 6-35 Tranches 1 and 2 base case: costs and benefits 36 Table 6-36 Tranches 1 and 2: costs and benefits for case B traffic 37 Table 6-37 Sensitivity tests: tranche 1 38 Table 6-38 Input ranges for risk analysis 38 Table 6-39 Results of risk analysis, tranche 1 39 Table 6-40 Sensitivity tests: tranches 1 and 2 40 Table 6-41 Results of risk analysis 41 Table 6.42 Sensitivity results for tranches 1 and 2, 30 percent RV 42

List of Figures Figure 6-1 Daily traffic pattern, Chorokhi 4 Figure 6-2 Elasticities of vehicle numbers 6 Figure 6-3 Traffic forecasts 9 Figure 6-4 OD survey results 10 Figure 6-5 Frequency distribution of tranche 1 EIRR 39 Figure 6-6 Frequency distribution of EIRR (tranches 1 and 2) 41 Figure 6-7 Frequency distribution of EIRR (tranches 1 and 2, 30% RV) 43

Engconsult Ltd. 1

I. TRAFFIC SURVEYS

1. Roads Department carries out classified counts at three sites on the existing road three times a year (see tabulated values for 2006-8 below). No origin and destination (OD) surveys have been carried out.

2. Two sets of project traffic surveys were carried out: • manual classified counts on 26-27 March and 1-2 April • OD surveys on 11-12 May (originally to have been carried out in early April but delayed as an indirect result of the civil unrest in the country at that time)

3. Table 6-1 shows details of the surveys. Table 6-1 Traffic surveys

Location Type Days Dates Manual classified counts Choloki bridge 1 x 12h & 1 x 24h classified counts 2 24-25 Mar (Tues-Wds) Chakvi-Makinjauri tunnel 1 x 12h & 1 x 24h classified counts 2 24-25 Mar (Tues-Wds) Batumi south (Chorokhi br) 1 x 12h & 1 x 24h classified counts 2 24-25 Mar (Tues-Wds) OD survey and classified counts Choloki bridge Approx 1,100 interviews & 12h 2 11-12 May classified count Batumi south (Chorokhi br) 2 11-12 May

II. EXISTING TRAFFIC

A. Historic traffic on the existing road

4. Roads Department radar counts are available at three sites on the existing project road (road S-2). Allocation to vehicle class is based on vehicle speed and length. Classification is probably fairly reliable for articulated trucks and small vehicles (cars and mini-buses) but less so for intermediate sizes. From north to south on the Senaki-Poti-Sarpi road the three sites are (i) the Choloki river crossing (km64), (ii) the Chakvi-Makhinjauri (C-M) tunnel at km95 and (iii) the Chorokhi river crossing (km115).

5. Counts take place in April, July and October. Heavy goods traffic shows no discernible seasonality, but car and minibus traffic certainly does and is typically 30- 50 percent higher in July than it is in April or October. In order to estimate AADT from the 24h data derived from the counts, July values were taken to apply to 180 days, April to 90 days and October to 95 days. It is assumed that additional weekend holiday traffic offsets the usual reduction at religious festivals.

6. Table 6-2 summarizes counts and estimated AADT at the three sites, and the project counts in late March 2009. Original counts were obtained from the Roads Department and re-analysed; as a result the figures in Table 6-2 differ slightly from those in the JICA Highway Improvement Study report (2008).

Engconsult Ltd. 2

7. There have been two regional shocks during this period – the Russian embargo on imports from Georgia in October 2006 and the conflict in S Ossetia in August 2008 – and of course the global recession that started during 2008. Any effect of the Russian embargo is not immediately apparent in the data, but the effects of the recession and the regional conflict are apparent in the fall in traffic growth in 2008. Table 6-2 Historic traffic at project sites

Site Year April July October Average AADT veh/day % change Choloki bridge 2006 2,214 3,345 2,927 2,829 2,957 (km64) 2007 4,184 3697 3,640 3,840 3,802 29% 2008 3,819 4952 4,030 4,267 4,433 17% 2009 3,302a 2,847b 3,655c -18% C-M tunnel 2006 6,614 7,467 N/A N/A 7,035a (km95) 2007 8,098 10,142 8,595 8,945 9,235 31% 2008 9,545 10,840 9,552 9,979 10,185 10% 2009 7,596a 8,361c -18% Chorokhi bridge 2006 3,600 3,433 N/A N/A 3,518d (km115) 2007 3,489 7,563 3,560 4,871 5,517 57% 2008 4,224 5,814 4,483 4,840 5,076 -8% 2009 5,644a 6,132b 6,108c 20% Sources: roads department counts and consultants’ assumptions Note: (a) project count, March 2009 (b) project count, early Mary 2009 (c) adjusted using historic seasonality (d) estimated assuming

8. Numbers of vehicles with foreign registration plates were noted during the project counts and are summarized in Table 6-3. The table shows: • nearly all articulated trucks have foreign registrations and the absolute numbers are similar at all three sites • 70-80 percent of large buses at the northern and southern sites are foreign but only 30 percent are recorded between Batumi and Kobuleti • the substantial percentages of foreign registered LGVs and medium buses at Chorokhi – but not further north – suggests a significant trade in goods and passengers between the border and Batumi

Table 6-3 Foreign registered traffic

Car Mini- Med bus Large LGV MGV HGV Artic Total (2 bus bus way) Choloki 16 1 3 23 0 1 11 238 294 1% 0.2% 7% 71% 0% 1% 18% 89% 9% C-M tunnel 43 9 4 13 1 0 1 239 309 1% 0.4% 8% 31% 1% 0% 1% 80% 4% Chorokhi 216 17 51 47 51 4 14 266 665

Engconsult Ltd. 3

Car Mini- Med bus Large LGV MGV HGV Artic Total (2 bus bus way) 6% 2% 30% 79% 20% 4% 7% 93% 12% Sources: consultants’ traffic counts, March 2009

B. Base year traffic

9. Traffic in the base year (2009) is based on project counts. At Choloki bridge March 2009’s count was 14 percent lower than that recorded a year earlier, while at the C-M tunnel it was 20 percent lower. At both sites March counts were expanded to full year values using seasonal factors for July and October observed in 2007-8. At Choloki bridge and the tunnel this produced AADTs of 3,655 and 8,361 veh/day – around the levels of traffic experienced in 2006-7 and 18 percent below 2008’s AADT.

10. At Chorokhi bridge, however, the March 2009 count was 34 percent higher than that recorded a year earlier and the later count in May supported this increase. Most of the increase in March was attributable to cars, buses and medium goods vehicles, while mini-bus and heavy goods vehicle counts were consistent with records from 2008. Knowing of no local factors that could produce such an anomalous result it was decided to use historic seasonality to give an AADT of 6,108. (Anecdotally, city officials in Batumi confirm that traffic has not reduced).

11. Base year traffic on the existing road is shown in Table 6-4. Traffic north of Kobuleti (sections K1 and K1a) is taken as equal to the seasonally adjusted 2009 Choloki AADT in Table 6-2. Traffic south of Kobuleti and north of Batumi center is taken as equal to the seasonally adjusted C-M tunnel AADT. South of the A306 junction (section B3) traffic is equated to the Chorokhi bridge AADT. Traffic on section B2 is the average of the tunnel and Chorokhi bridge AADTs. Table 6-4 Base year traffic on existing road sections

Sections Car Mini-bus Med busLarge LGV MGV HGV Artic Total bus Project 1: Kobuleti bypass K1-K1a veh/day 2,057 890 48 32 153 86 73 316 3,655 PCU/day 2,057 1,157 72 64 199 129 146 790 4,614 K2-K6 veh/day 5,455 2,120 56 46 207 106 63 308 8,361 PCU/day 5,455 2756 84 92 269 159 126 770 9,711 Average veh/day 4,247 1,683 53 41 188 99 67 311 6,688 % 63.5 25.2 0.8 0.6 2.8 1.5 1.0 4.6 100% PCU/day 4,247 2,187 80 82 244 148 132 777 7,899 Project 2: Batumi bypass B1 veh/day 5,455 2,120 56 46 207 106 63 308 8,361 PCU/day 5,455 2756 84 92 269 159 126 770 9,711 B2 veh/day 4,708 1,556 118 54 236 96 119 318 7,205 PCU/day 4,708 2,022 177 108 307 144 238 795 8,499 B3 veh/day 4,000 1,001 182 63 268 87 177 331 6,108

Engconsult Ltd. 4

Sections Car Mini-bus Med busLarge LGV MGV HGV Artic Total bus PCU/day 4,000 1,301 273 125 348 130 353 828 7,359 Average veh/day 4,794 1,611 113 54 235 97 115 319 7,337 % 65.3 22.0 1.5 0.7 3.2 1.3 1.6 4.3 100% PCU/day 4,794 2,095 170 107 305 146 229 796 8,642 Combined project Average veh/day 4,408 1,662 71 45 202 98 81 313 6,871 % 64.1 24.2 1.0 0.7 2.9 1.4 1.2 4.6 100% PCU/day 4,408 2,160 106 89 262 148 162 783 8,118 Source: consultants’ estimates Notes: (a) PCU equivalents are in Table 6-13 (b) details of road sections are in sections B and C (c) averages are weighted by distance

C. Traffic flow pattern

12. The daily flow pattern for Chorokhi (Figure 6-1) shows a pattern typical for an inter-urban road carrying a high proportion of long distance goods and passengers. Instead of urban morning and evening peaks the traffic is fairly uniform from 09h00 to 22h00. Most of the night-time traffic is heavy goods and passenger vehicles. Figure 6-1 Daily traffic pattern, Chorokhi

9%

8%

7%

c 6%

5%

4%

Percent of daily traffi daily of Percent 3%

2%

1%

0% 09h00- 11h00- 13h00- 15h00- 17h00- 19h00- 21h00- 23h00- 01h00- 03h00- 05h00- 07h00- 10h00 12h00 14h00 16h00 18h00 20h00 22h00 00h00 02h00 04h00 06h00 08h00

13. The annual flow pattern is strongly seasonal. Concerned that the HDM-4 seasonal flow pattern option might not adequately reflect conditions on the Ajara coast, a slightly more seasonal pattern was adopted, based on project traffic counts and historic evidence of seasonality.

Engconsult Ltd. 5

Table 6-5 Seasonal traffic flow pattern

Pattern Total Hours per year 950 1,100 1,720 1,430 3,560 8,760 Percent of AADT in period 25% 25% 25% 15% 10% 100% Sources: consultants’ estimates

III. FUTURE TRAFFIC DEMAND

A. Introduction

14. The demand for travel on project corridor roads comprises: • normal traffic – travel that would arise without the bypass; • diverted traffic – traffic choosing to use the bypass in preference to travel on the existing roads • generated traffic – travel that only arises because of the existence of the bypass

15. This section deals with the first two of these: generated traffic is discussed in section IV.A.5 below.

B. Growth of Normal Traffic

1. Introduction

16. The demand for transport is related to the output of the economy. Traffic is almost invariably positively correlated with economic activity, measured in various ways, and traffic growth similarly with growth in economic activity. In the most general terms the relationship is as follows:

= YkQ )( e where Q is some measure of demand for transport, k is a constant and e is the elasticity of demand for transport with respect to some measure of real economic activity, Y. Elasticity is the proportional change in demand per unit change in economic activity, so an elasticity of 0.2 implies a 2 percent growth in travel demand in response to a 10 percent increase in real economic activity.

17. In the case of passenger traffic, the measure of economic activity would typically include measures of population or household as well as income growth. As the population of Georgia is essentially static (see Supplementary Appendix 1), forecasts of passenger traffic will depend solely on GDP.

18. For goods traffic GDP is the usual measure of economic activity is GDP. In this case the trade component of GDP is especially significant.

2. Elasticities

19. The following statistics are both indicative of transport demand and are available:

Engconsult Ltd. 6

• numbers of registered vehicles • tonne-km and passenger-km data • vehicle counts

20. Registered vehicle statistics are not of the best quality. Annual licensing is no longer required. Instead the only requirement is registration on change of ownership. As a result the numbers of registered vehicles almost certainly include vehicles that are no longer used. They therefore probably overstate the total number of vehicles, though not necessarily annual growth rates. There are also inconsistencies between data obtained by this project (from the Ministry of Internal Affairs), data published in the 2008 Statistical Yearbook of Georgia and data reproduced in the JICA Highway Improvement Project. For goods vehicles and all vehicles the most plausible series comprise JBIC data from 2000 to 2006 for trucks and other goods. For passenger vehicles a hybrid series is used: JBIC data from 2000 to 2006 followed by Ministry of Internal Affairs data for 2007-8. (For more detail see Supplementary Appendix 1).

21. Figure 6-2 shows a plot of indices of registered vehicles against GDP, also expressed as an index. Best fit power curves indicate elasticities as follows: (i) 1.34 for all vehicles, (ii) 1.30 for passenger vehicles and (iii) 1.69 for goods vehicles. Figure 6-2 Elasticities of vehicle numbers

200

180

160

140

120 Reg vehicles (2002 = 100) = (2002 vehicles Reg 100 100 110 120 130 140 150 160 170 GDP (2002 = 100)

All vehicles Passenger veh Goods veh

22. Tonne-km and passenger-km data are also available from the statistical yearbook but are considered too unreliable to use.

23. Project area traffic data – both historic data from the roads department and project counts – are available at three stations from 2004 to March 2009. Historic data for 2002-7 are also available at six other stations in Georgia. Both sets produced elasticities (for all vehicles) of about 2.2 – well above the range usually used.

24. Lower elasticities have been used in other recent studies in Georgia. For the MCG Samtskhe-Javakheti road project feasibility study the consultants used 1.2 for passenger traffic and 1.0 for goods (based on TRACECA assumptions rather than project analysis). The JICA study referred to above varied elasticities over time, but

Engconsult Ltd. 7

for the early years in the evaluation used 1.0 for goods traffic, approximately 1.05 for cars and 0.95 for buses.

25. Project corridor traffic is heavily influenced by trade, both in services (tourism) and goods. Trade grows faster than the economy as a whole: data in Supplementary Appendix 1 indicate that the growth in Georgian trade has in recent years been roughly 5 percent faster than GDP and justifies the use of elasticities that are higher than usual.

26. Taking the above into account, adopted elasticities are 1.2 for passenger traffic and 1.4 for goods traffic.

3. GDP and traffic forecasts

27. GDP forecasts are in Supplementary Appendix 1. Applying the elasticities in section 2 above gives forecast increases in traffic shown below. Table 6-6 GDP and traffic growth rates

2009 2010 2011 2012 2013 2014 2015 on GDP growth 0% 1.5% 2.0% 3.0% 4.0% 4.0% 4.0% Passenger vehicles 0% 1.8% 2.4% 3.6% 4.8% 4.8% 4.8% Goods vehicles 0% 2.1% 2.8% 4.2% 5.6% 5.6% 5.6% Sources: Supplementary Appendix 1 and consultants’ calculations

C. Corridor diversions

1. Introduction

28. Currently, almost all road traffic between Turkey and Georgia, Azerbaijan, Armenia and Central Asia is obliged to use the S-2 between Sarpi and Kobuleti. (There is another Turkish/Georgian border further east, at Vale, but connecting roads are poor and very little traffic uses it). This situation will not persist indefinitely and the economic returns to the project are likely to be affected by two developments: • improvement of road links between Georgia and Turkey further east (see below). This development appears certain but the timing of the Turkey-Georgia border opening is not • the re-opening of the Turkey-Armenia border, closed since 1993. This development is not certain, as it depends on ratification of a “normalization package” by the two governments

29. Traffic case A assumes no corridor diversions: all traffic currently using the Sarpi-Kobuleti corridor will continue to do so. This case acknowledges the high degree of uncertainty inherent in forecasts of border openings. It is the base case. Traffic case B (discussed below) assumes that a proportion of traffic for which the new route across the Turkey-Georgia border would be shorter will in fact divert to the new route. Relatively few vehicles currently travel between Turkey and Armenia via Georgia and so the effect of opening of the Turkey-Armenia border is not considered as a separate traffic case. (See Figure 6-4: Azerbaijan is by far the most

Engconsult Ltd. 8

significant transit origin/destination. Armenia only accounts for 5-6 percent of traffic to or from origins/destinations from Tbilisi and beyond).

2. The Samtskhe-Javakheti road project

30. The Samtskhe-Javakheti road project is being financed by the Millennium Challenge Georgia Fund (MCG). When complete it will provide an improved transport connection between Tbilisi and the region to the south-west of the capital and ultimately between Tbilisi and Turkey. It will also shorten journey times between Turkey and Azerbaijan and Turkey and Armenia. Road improvements will be complete by late 2010. There is a new border post (and improved road connections) on the Turkish side of the border but as yet no progress has been made on the Georgian side.

31. Once the new road is open, the distance from Erzurum in Turkey to Tbilisi via Kars and Ninotsminda will be 575km, compared with 675km via Sarpi – a drop of 15 percent. The distance from Erzurum to Gori will drop by 40km; for destinations further west in Georgia the Sarpi route will remain shorter. Even trips to and from Tbilisi and further east will not immediately switch to the shorter route. The Sarpi route is likely to be faster (especially outside the tourist season) and less prone to adverse weather conditions.

32. This new route is referred to as the Kartsakhi corridor, Kartsakhi being the Georgian village closest to the border.

3. Adopted corridor diversions

33. The OD survey results (see Figures 6-3 and 6-4) indicate that 25 to 40 percent of all traffic at Chorokhi bridge (Batumi S) and Choloki (Kabuleti N) – and 30-50 percent of goods traffic – would find the Kartsakhi corridor attractive. It is assumed that 40 percent of all trip ends east of Gori and 20 percent of those in the zone around Gori actually divert to the Kartsakhi corridor. The impact of corridor diversions is discussed below in section D.

D. Adopted traffic

34. Manual assignments for traffic cases A and B were made on the basis of the OD results, shown in Figure 6-4. In general, assignment was straightforward and left little scope for doubt. Only some of the traffic using the existing C-M tunnel is recorded by the Choloki OD survey, however: trips purely between Kobuleti and Batumi were not recorded. A conservative assumption was made, namely that 20 percent of the AADT at the tunnel less the AADT north of Kobuleti would divert to the bypass. (This does not apply to existing road section K6, which will be widened to become bypass section KB3 under the project and will take all corridor traffic).

35. Some of the OD results for Choloki, especially the southbound trips originating in Tbilisi and zones east of Tbilisi, appear too low and if used as presented would have overstated the proportion of case B traffic diverting to the bypass. To offset this, the overall proportion of traffic using the Kobuleti bypass was reduced to make it consistent with the proportion using the Batumi bypass.

36. Adopted traffic on the existing and bypass road sections for cases A and B are shown in Tables 6-7 and 6-8, and graphically in Figure 6-3. Tables 6-9 and 6-10

Engconsult Ltd. 9

show tranche 1 traffic for each year for the two traffic cases and Table 6-11 the base case for tranches 1 and 2.

Figure 6-3 Traffic forecasts

12,000

10,000

8,000

6,000 AADT (vehicles) 4,000

2,000

0 2005 2010 2015 2020 2025 2030 2035

Existing rd, case A Existing rd, case B Bypass, case A Bypass, case B

37. To summarize: • diversion to the Kartsakhi corridor (i.e. case B traffic ) reduces project corridor traffic by approximately 20 percent. Goods traffic diverting to the bypass is reduced by 30 percent • the percentage of project corridor traffic diverting to the bypass is 42-43 percent in both case A and case B. Goods traffic diverting to the bypass is 60-61 percent of project corridor goods traffic • opening traffic on the two lane section of the Kobuleti bypass in 2014 will be 3,600 veh/day (no diversion to the Kartsakhi corridor) and 2,800 if corridor diversion takes place. The four lane section will carry respectively 10,200 and 8,600 veh/day • opening traffic on the Batumi bypass in 2014 will be 3,400 veh/day (case A) and 2,800 veh/day (case B) • in 2033, the last evaluation year, bypass traffic on the two lane sections will be 6-8,000 veh/day, depending on the traffic case, i.e. still short of the levels that would normally justify an additional lane

38. Local traffic, i.e. traffic between points within the urban centers of Batumi and Kobuleti, is outside the scope of diversion to the bypass but adds to congestion on existing roads and so affects journey times and vehicle operating costs. To take account of this effect, light and medium passenger and light goods traffic has been doubled on the urban sections of the existing road (sections K1a and B2 in Tables 6-16 and 6-22).

Engconsult Ltd. 10

Figure 6-4 OD survey results

Northbound at Chorokhi (Batumi south) Destination All vehicles Turkey Batumi Kobuleti Ozurgeti Poti Swaneti Kutaisi Akhaltsikhe Gori Tbilisi Telavi Larsi Yerevan Baku 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Total Turkey 1 1,332 19 19 69 5 531 21 52 390 2,438

Origin Batumi 2 196 79 42 79 19 183 27 19 19 663 Total 0 1,332 215 79 61 0 148 24 0 714 48 0 71 409 3,101 Passenger vehicles Turkey 1 1,173 19 19 61 445 21 39 257 2,034

Origin Batumi 2 196 79 39 79 19 165 19 19 615 Total 0 1,173 215 79 58 0 140 19 0 610 21 0 58 276 2,649 Goods vehicles Turkey 1 159 8 5 86 13 133 404

Origin Batumi 2 3 18 27 48 Total 0 159 0 0 3 0 8 5 0 104 27 0 13 133 452 Northbound at Choloki (Kobuleti north)

All Turkey Batumi Kobuleti Ozurgeti Poti Swaneti Kutaisi Akhaltsikhe Gori Tbilisi Telavi Larsi Yerevan Baku 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Total Turkey 1 6 15 3 19 65 4 15 127 Batumi 2 4 276 522 9 475 13 56 390 33 3 1,781

Origin Kobuleti 3 135 30 10 37 3 45 260 Total 4 0 0 411 552 19 512 13 59 435 33 0 3 0 2,041 Passenger Turkey 1 0 0 0 6 15 3 14 0 0 61 4 0 0 3 106 Batumi 2 4 0 0 234 427 6 438 13 45 350 29 0 0 0 1,546

Origin Kobuleti 3 0 0 0 127 30 10 32 0 3 42 0 0 0 0 244 Total 4 0 0 361 457 16 470 13 48 392 29 0 0 0 1,790 Goods Turkey 1 0 0 0 0 0 0 5 0 0 4 0 0 0 12 21 Batumi 2 0 0 0 42 95 3 37 0 11 40 4 0 3 0 235

Origin Kobuleti3 0 0 0 80 0 5 00 300 0 016 Total 0 0 0 50 95 3 42 0 11 43 4 0 3 0 251

Southbound at Chorokhi (Batumi south) Destination Destination Destination Turkey Batumi Turkey Batumi Turkey Batumi All vehicles 12Total Pax 12Total Goods 12Total Turkey 1 0 0 0 1 0 0 0 1 0 0 0 Batumi 2 1,190 0 1,190 2 1,035 0 1,035 2 155 0 155 Kobuleti 3 18 178 196 3 18 178 196 3 0 0 0 Ozurgeti 4 0 68 68 4 0 68 68 4 0 0 0 Poti 5 18 38 56 5 18 37 55 5 0 1 1 Swaneti 6 0 0 0 6 0 0 0 6 0 0 0 Kutaisi 7 47 68 115 7 45 68 113 7 2 0 2 Akhaltsikhe 8 2 18 20 8 0 18 18 8 2 0 2 Origin Gori 9 0 0 0 9 0 0 0 9 0 0 0 Tbilisi 10 466 180 646 10 401 141 542 10 65 39 104 Telavi 11 13 47 60 11 13 0 13 11 0 47 47 Larsi 12 0 0 0 12 0 0 0 12 0 0 0 Yerevan 13 41 18 59 13 37 18 55 13 4 0 4 Baku 14 349 18 367 14 236 18 254 14 113 0 113 2,144 633 2,777 1,803 546 2,349 341 87 428

Southbound at Choloki (Kobuleti north) Destination Destination Destination Turkey Batumi Kobuleti Turkey Batumi Kobuleti Turkey BatumKobuleti All vehicles 12 3TotalPax 1 2 3 Total Goods 1 2 3 Total Turkey 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 Batumi 2 0 0 0 0 2 0 0 0 0 2 0 0 0 0 Kobuleti 3 0 0 0 0 3 0 0 0 0 3 0 0 0 0 Ozurgeti 4 0 214 99 313 4 0 193 86 279 4 0 21 13 34 Poti 5 0 396 92 488 5 0 357 84 441 5 0 39 8 47 Swaneti 6 0 6 0 6 6 0 6 0 6 6 0 0 0 0 Kutaisi 7 7 259 56 322 7 6 234 45 285 7 1 25 11 37 Akhaltsikhe 80160168 01601680000 Origin Gori 9 3 17 0 20 9 3 13 0 16 9 0 4 0 4 Tbilisi 10 6 129 38 173 10 6 121 36 163 10 0 8 2 10 Telavi 11 0 12 3 15 11 0 10 3 13 11 0 2 0 2 Larsi 12 0 0 0 0 12 0 0 0 0 12 0 0 0 0 Yerevan 13 0 0 00 13 0 0 00 13 00 00 Baku 14 3 0 0 3 14 0 0 0 0 14 3 0 0 3 19 1,049 288 1,356 15 950 254 1,219 4 99 34 137

Notes: (i) main towns within each zone identified for convenience (ii) totals do not necessarily represent adopted daily traffic

Engconsult Ltd. 11

Table 6-7 Base case traffic (case A): no corridor diversion, veh/day

Without bypass, veh/day 2009 2014 2020 2033 Pax Goods Total Pax Goods Total Pax Goods Total Pax Goods Total K1 3,027 628 3,655 K1 3,697 792 4,489 K1 4,897 1,099 5,996 K1 8,596 2,113 10,709 K2-K5 7,677 684 8,361 K2-K5 9,375 863 10,238 K2-K5 12,421 1,197 13,618 K2-K5 21,801 2,301 24,102 K6 7,677 684 8,361 K6 9,375 863 10,238 K6 12,421 1,197 13,618 K6 21,801 2,301 24,102 B1 7,677 684 8,361 B1 9,375 863 10,238 B1 12,421 1,197 13,618 B1 21,801 2,301 24,102 B2 6,461 773 7,234 B2 7,891 976 8,867 B2 10,454 1,353 11,807 B2 18,349 2,601 20,950 B3 5,246 862 6,108 B3 6,406 1,088 7,494 B3 8,487 1,509 9,996 B3 14,897 2,901 17,798

With bypass, veh/day 2009 2014 2020 2033 Pax Goods Total Pax Goods Total Pax Goods Total Pax Goods Total Existing roads Existing roads Existing roads Existing roads K1 1,141 232 1,373 K1 1,394 292 1,686 K1 1,847 405 2,252 K1 3,241 780 4,021 K2-K5 4,785 261 5,046 K2-K5 5,843 330 6,173 K2-K5 7,741 457 8,198 K2-K5 13,587 879 14,466 K6 7,677 684 8,361 K6 9,375 863 10,238 K6 12,421 1,197 13,618 K6 21,801 2,301 24,102 B1 5,430 260 5,690 B1 6,631 328 6,959 B1 8,785 455 9,240 B1 15,420 876 16,296 B2 3,883 304 4,187 B2 4,741 383 5,124 B2 6,282 532 6,814 B2 11,026 1,022 12,048 B3 3,195 363 3,558 B3 3,902 458 4,360 B3 5,169 636 5,805 B3 9,074 1,222 10,296

Bypass sections Bypass sections Bypass sections Bypass sections KB1 1,886 396 2,282 KB1 2,303 500 2,803 KB1 3,051 693 3,744 KB1 5,355 1,333 6,688 KB2 2,892 423 3,315 KB2 3,532 533 4,065 KB2 4,680 740 5,420 KB2 8,214 1,422 9,636 KB3 7,677 684 8,361 KB3 9,375 863 10,238 KB3 12,421 1,197 13,618 KB3 21,801 2,301 24,102 BT1 2,247 424 2,671 BT1 2,744 534 3,278 BT1 3,635 741 4,376 BT1 6,381 1,425 7,806 BT2 2,579 469 3,048 BT2 3,149 592 3,741 BT2 4,172 821 4,993 BT2 7,323 1,579 8,902 BT3 2,050 499 2,549 BT3 2,504 630 3,134 BT3 3,317 873 4,190 BT3 5,823 1,679 7,502

Ex roads 3,574 267 3,842 4,365 337 4,702 5,783 468 6,251 10,150 900 11,050 Bypass 2,437 427 2,864 2,977 539 3,515 3,943 748 4,691 6,921 1,437 8,359 Diversion 41% 62% 43% 41% 62% 43% 41% 62% 43% 41% 61% 43%

Notes: (i) local and generated traffic not included (ii) percentages diverted do not include sections K6/KB3, which follow the same alignment with and without the project

Table 6-8 Case B traffic: diversion to the Kartsakhi corridor, veh/day

Without bypass, veh/day 2009 2014 2020 2033 Pax Goods Total Pax Goods Total Pax Goods Total Pax Goods Total K1 2,165 456 2,621 K1 2,644 575 3,219 K1 3,503 797 4,300 K1 6,149 1,532 7,681 K2-K5 6,512 513 7,025 K2-K5 7,953 647 8,600 K2-K5 10,536 897 11,433 K2-K5 18,494 1,724 20,218 K6 6,512 513 7,025 K6 7,953 647 8,600 K6 10,536 897 11,433 K6 18,494 1,724 20,218 B1 6,021 454 6,475 B1 7,353 573 7,926 B1 9,742 795 10,537 B1 17,099 1,528 18,627 B2 4,625 521 5,146 B2 5,648 658 6,306 B2 7,483 912 8,395 B2 13,135 1,754 14,889 B3 3,760 553 4,313 B3 4,592 698 5,290 B3 6,084 967 7,051 B3 10,678 1,860 12,538

With bypass, veh/day 2009 2014 2020 2033 Pax Goods Total Pax Goods Total Pax Goods Total Pax Goods Total Existing roads Existing roads Existing roads Existing roads K1 829 171 1,000 K1 1,012 216 1,228 K1 1,341 299 1,640 K1 2,354 575 2,929 K2-K5 4,069 196 4,265 K2-K5 4,969 247 5,216 K2-K5 6,583 343 6,926 K2-K5 11,554 659 12,213 K6 6,512 513 7,025 K6 7,953 647 8,600 K6 10,536 897 11,433 K6 18,494 1,724 20,218 B1 4,261 183 4,444 B1 5,204 231 5,435 B1 6,894 320 7,214 B1 12,100 615 12,715 B2 2,796 214 3,010 B2 3,414 270 3,684 B2 4,523 374 4,897 B2 7,940 720 8,660 B3 2,300 249 2,549 B3 2,808 314 3,122 B3 3,721 436 4,157 B3 6,531 838 7,369

Bypass sections Bypass sections Bypass sections Bypass sections KB1 1,336 285 1,621 KB1 1,632 359 1,991 KB1 2,162 498 2,660 KB1 3,795 958 4,753 KB2 2,444 317 2,761 KB2 2,984 400 3,384 KB2 3,953 554 4,507 KB2 6,939 1,066 8,005 KB3 6,512 513 7,025 KB3 7,953 647 8,600 KB3 10,536 897 11,433 KB3 18,494 1,724 20,218 BT1 1,760 285 2,045 BT1 2,150 359 2,509 BT1 2,848 498 3,346 BT1 4,999 913 5,912 BT2 1,829 307 2,136 BT2 2,234 388 2,622 BT2 2,960 538 3,498 BT2 5,195 1,034 6,229 BT3 1,460 304 1,764 BT3 1,783 383 2,166 BT3 2,363 532 2,895 BT3 4,147 1,022 5,169

Ex roads 2,858 194 3,052 3,490 245 3,735 4,623 340 4,963 8,115 654 8,768 Bypass 1,908 302 2,210 2,329 381 2,710 3,086 528 3,614 5,417 1,010 6,427 Diversion 40% 61% 42% 40% 61% 42% 40% 61% 42% 40% 61% 42%

Notes: (i) local and generated traffic not included (ii) percentages diverted do not include sections K6/KB3, which follow the same alignment with and without the project

Engconsult Ltd. 12

Table 6-9 Tranche 1 traffic projection, case A (base case), veh/day

Year Without project With project Existing road Existing road Bypass Generated traffic Total 2009 6,326 0 0 0 0 2010 6,588 0 0 0 0 2011 6,862 0 0 0 0 2012 7,147 0 0 0 0 2013 7,443 0 0 0 0 2014 7,752 4,233 3,559 410 8,202 2015 8,131 4,439 3,734 430 8,602 2016 8,528 4,654 3,918 450 9,023 2017 8,945 4,880 4,111 472 9,463 2018 9,382 5,118 4,313 495 9,926 2019 9,841 5,366 4,525 519 10,411 2020 10,322 5,627 4,748 544 10,919 2021 10,787 5,879 4,963 568 11,411 2022 11,274 6,143 5,189 594 11,925 2023 11,782 6,418 5,424 620 12,462 2024 12,313 6,706 5,670 648 13,024 2025 12,868 7,006 5,928 677 13,611 2026 13,448 7,320 6,197 707 14,224 2027 14,054 7,648 6,478 739 14,865 2028 14,688 7,991 6,772 772 15,535 2029 15,350 8,349 7,080 806 16,235 2030 16,042 8,723 7,401 842 16,966 2031 16,765 9,114 7,737 880 17,730 2032 17,521 9,522 8,088 919 18,529 2033 18,311 9,949 8,455 960 19,364 Source: consultants’ calculations Notes: (a) traffic is weighted by distance; as the bypass is slightly longer than the existing road, the sum of project traffic on the bypass and the existing road slightly exceed the without project traffic (b) local traffic is not included, nor is traffic on the common alignment between km28 and km34

Engconsult Ltd. 13

Table 6-10 Tranche 1 traffic projection, case B, veh/day

Year Without project With project Existing road Existing road Bypass Generated traffic Total 2009 5,120 0 0 0 0 2010 5,332 0 0 0 0 2011 5,553 0 0 0 0 2012 5,784 0 0 0 0 2013 6,023 0 0 0 0 2014 6,273 3,491 2,826 337 6,654 2015 6,579 3,661 2,965 353 6,978 2016 6,900 3,838 3,110 370 7,319 2017 7,237 4,025 3,263 388 7,676 2018 7,590 4,220 3,423 407 8,050 2019 7,961 4,425 3,591 427 8,443 2020 8,349 4,640 3,767 447 8,854 2021 8,725 4,848 3,938 467 9,253 2022 9,117 5,065 4,116 488 9,669 2023 9,528 5,291 4,303 510 10,104 2024 9,957 5,528 4,498 532 10,558 2025 10,405 5,776 4,701 556 11,033 2026 10,873 6,034 4,914 581 11,529 2027 11,362 6,304 5,137 607 12,048 2028 11,874 6,586 5,370 634 12,590 2029 12,408 6,881 5,613 662 13,156 2030 12,966 7,189 5,868 691 13,748 2031 13,550 7,511 6,134 722 14,366 2032 14,160 7,847 6,411 754 15,013 2033 14,797 8,198 6,702 788 15,688 Source: consultants’ calculations Notes: (a) traffic is weighted by distance; as the bypass is slightly longer than the existing road, the sum of project traffic on the bypass and the existing road slightly exceed the without project traffic (b) local traffic is not included, nor is traffic on the common alignment between km28 and km34

Engconsult Ltd. 14

Table 6-11 Tranches 1 and 2 traffic projection, case A (base case), veh/day

Year Without project With project Existing road Existing road Bypass Generated traffic Total 2009 6,667 0 0 0 0

2010 6,944 0 0 0 0

2011 7,232 0 0 0 0

2012 7,532 0 0 0 0

2013 7,844 0 0 0 0

2014 8,170 4,702 3,515 457 8,674

2015 8,569 4,931 3,688 479 9,097

2016 8,988 5,170 3,870 502 9,542

2017 9,428 5,421 4,061 526 10,008

2018 9,889 5,685 4,261 552 10,497

2019 10,372 5,961 4,471 578 11,010

2020 10,879 6,251 4,691 606 11,548

2021 11,369 6,531 4,904 633 12,069

2022 11,882 6,824 5,127 662 12,612

2023 12,417 7,129 5,360 691 13,180

2024 12,977 7,449 5,604 722 13,774

2025 13,562 7,782 5,858 754 14,394

2026 14,173 8,131 6,124 788 15,043

2027 14,812 8,495 6,403 823 15,720

2028 15,480 8,876 6,694 859 16,429

2029 16,178 9,273 6,998 898 17,169

2030 16,907 9,689 7,316 938 17,942

2031 17,669 10,123 7,648 979 18,750

2032 18,466 10,576 7,996 1,023 19,595

2033 19,298 11,050 8,359 1,069 20,478 Source: consultants’ calculations Notes: (a) traffic is weighted by distance; as the bypass is slightly longer than the existing road, the sum of project traffic on the bypass and the existing road slightly exceed the without project traffic (b) local traffic is not included, nor is traffic on the common alignment between km28 and km34

E. Diversion from other modes

39. Air traffic through Batumi airport is currently negligible (see section IV of Supplementary Appendix 1), leaving rail as the only other significant mode in the corridor.

Engconsult Ltd. 15

40. The railway line to Batumi is predominantly a means of transporting oil and refined products from Baku to Batumi port. Rail is far more efficient than road at carrying commodities over long distances, especially when the fixed investment costs are all sunk. Bypass improvements would not change rail’s cost advantage.

41. Roughly 20 percent of Kobuleti-Batumi’s visitors arrive by train. The trains are seldom full; those using them doubtless prefer their safety and comfort over the faster but less comfortable and more hazardous bus services. The bypass will make little difference to this assessment for the overwhelming proportion of rail passengers who are long distance travelers. It is concluded that diversion from rail would be negligible.

IV. ECONOMIC EVALUATION

A. Approach

1. Introduction

42. The main objective of the project is to divert transit traffic away from the existing coastal road. The latter, especially through the urban areas of Batumi and Kobuleti, is congested, accident-prone and in a poor condition.

43. HDM-4 is used to evaluate the project. The (combined) project is divided into two projects: • project 1: Kobuleti to Makhinjauri (the Kobuleti bypass) • project 2: Makhinjauri to Sarpi (the Batumi bypass)

44. Dividing the project up in this way makes it easier to identify optimum solutions in terms of timing and construction standard. It also matches the tranches of financing envisaged under the proposed multi-tranche financing facility (MFF). Each project is evaluated separately before combining them as a single project evaluation.

45. This section explains how the economic feasibility of each project has been established by comparing its costs and benefits. It is arranged in three parts: after this introductory section, sections B and C deal with the evaluation of projects 1 and 2 respectively, section D with the combined project and section E with risk and uncertainty, confined to project 1.

46. All predicted project costs and benefits are measured in economic prices1 using an international price numeraire and a US dollar unit of account. That is, tradable goods are expressed in world market prices plus transport costs (border parity prices), while non-tradable goods are converted from domestic prices (less taxes etc) multiplied by a standard conversion factor (SCF) and divided by the official exchange rate (OER). Costs and benefits include: (i) the economic costs of

1 Economic prices measure the value of project inputs and outputs in terms of the consumption of real resources. They differ from financial (retail market) prices. For goods that are not tradable this means excluding taxes and subsidies and making adjustments where market prices do not reflect the value of resources consumed (see note below on shadow wage rates). For tradable goods it means using world market (border) prices and adding transport costs.

Engconsult Ltd. 16

construction, (ii) the economic costs of maintaining the new road, (iii) savings in vehicle operating costs (VOCs) and road user time costs2, (iv) savings in the economic costs of accidents and (v) an estimate of the benefits of “generated” traffic (i.e. traffic that is currently suppressed by the relatively high costs of travel on existing roads).

47. The direct impacts of the bypass – the savings in VOCs, user time costs etc – are relatively easy to quantify and value. Less so are the longer-term impacts of the project on regional incomes and growth – Ajara Autonomous Republic is heavily dependent on tourism receipts and citrus fruit exports. The government is also keen to foster more diversified growth around Batumi. Longer-term changes may be non- transport responses to the changed transport system or multiplier effects arising from purchases by firms and individuals. These effects are impossible to forecast with any reasonable accuracy. Given this uncertainty the best available measure of the changes brought about by the project alone is given by its direct impacts on transport costs, travel times and accident costs. These direct savings represent a conservative proxy for the broader benefits of improved transport.

48. The usual 12 percent discount rate is adopted.

2. Alternatives considered

49. For each evaluation a single project case (traffic case A – see above) are compared with a do minimum alternative. For section K6 of project 1 – the 5.6km section from Chakvi to Makhinjauri where all traffic will use the bypass – opening traffic will be around 10,000 veh/day and a four lane standard is proposed. Elsewhere, however, early runs of HDM-4 demonstrated that the economic return on a four lane bypass fell well below 12 percent and a two lane standard has therefore been adopted. For the same reason the original proposal of a new alignment from Chorokhi to Sarpi has been dropped.

3. Evaluation period and residual value

50. Each project is evaluated over 20 years following opening. Although each project will be phased slightly differently, the first full year of benefits will be 2014.

51. To account for the value of the project remaining at the end of the evaluation period, a negative cost is included in 2033 equivalent to the remaining unused portion of the project’s life (i.e. its residual value (RV) – referred to by HDM-4 as a salvage value). A weighted life of 60 years is used for both projects, based on the following assumptions about individual project components: land, earthworks, consultants’ services etc (25 percent of total costs) 100 years; structures (61 percent of costs) 50 years; pavement (14 percent of costs) 25 years. After 20 years’ operations an asset with an expected life of 60 years should have an economic salvage value of 65 percent of initial cost.

52. At ADB’s request an evaluation of the combined tranches was undertaken using a residual value of 30 percent of the initial cost. The results of this appear in section IV.F.

2 Measured, for passengers, in terms of the value of travel time saved and, for freight, in terms of the savings in inventory cost tied up during transport.

Engconsult Ltd. 17

4. Pricing assumptions

53. The economic evaluation uses first quarter 2009 prices. Cost estimates are already at 2009 prices and therefore require no adjustment for inflation.

54. A standard conversion factor (SCF) is used to convert the domestic market prices (less taxes) to international economic prices. This is calculated using the simple trade weighted formula presented in ERD Technical Note no 11, Feb 2004. The calculation is set out in Table 6-22; the SCF for 2007, 0.97, is adopted. It is approximate: in the table both imports and exports are expressed on an fob basis (whereas imports should be expressed cif) and export subsidies – small and probably declining in accordance with WTO rules – are not published in standard sources.

Table 6-12 Standard conversion factor calculation

2005 2006 2007 2008 Imports, USD m -2,687 -3,686 -4,984 -6,050 Exports, USD m 1,472 1,667 2,088 2,509 Import duties, GEL m 123 132 34 n/a Excise taxes, GEL m 286 336 374 n/a Export subsidies, GEL 10 10 10 - m Av GEL/USD 1.81 1.78 1.67 1.49 SCF 0.95 0.95 0.97 - Sources: IMF country reports and EIU country report for March 2009 (see also Supp App 1) Notes: (i) values in GEL converted to USD using tabulated average exchange rates (ii) export subsidies are not published but are believed to apply to agricultural commodities

55. The official exchange rate (OER) in April 2009 is USD1 = GEL1.67. Adopting an SCF of 0.97 this means that financial market prices of non-tradable goods are converted to USD by first multiplying their domestic prices (less taxes and other adjustments – see below) by 0.97 and then using the OER to convert to USD.

56. Other adjustments are made as follows: • unskilled labor is priced using a shadow wage rate factor of 0.70 (i.e. 70 percent of the estimated wage rate on the project)3. This is low, but reflects the low productivity of unskilled labor drawn specifically from poor areas. The same factor has been used in recent work on the Masalli-Astara expressway in Azerbaijan (see appendix 10 of the RRP for the Azerbaijan MFF, Sep 2007). Unskilled labor costs amount to an estimated 15 percent of the project’s total financial costs • the principal tradable input is reinforcing steel, which is imported from Ukraine at a CIF price of GEL900/t. This differs little from the border parity price implied by the world price for reinforcing bars in Jan-Mar

3 A “shadow” price represents the opportunity cost to the economy of using resources. A shadow wage rate factor of 0.7 implies that hiring an additional unskilled person on a project only displaces output worth 70 percent of the wage paid. That is, without the project, the unskilled person would be producing relatively little (e.g. working on the land) – his or her opportunity cost to the economy would thus be lower than the going rate for unskilled labor.

Engconsult Ltd. 18

2009 of USD470/t (see World Bank commodity price data for April 2009) and therefore does not merit an adjustment • contracting companies face a profits tax of 15 percent; it is assumed that a contractor’s net margin is 5 percent of total financial cost

57. At the time of reporting land and resettlement costs are estimated at a financial cost of GEL34m. In principle the financial costs of land acquisition and resettlement should not differ greatly from economic costs and for the moment, therefore, no adjustment is made.

5. Project benefits

a) Road user cost savings – VOCs

58. Savings in road user costs (RUCs, i.e. VOCs and user time costs) constitute the major proportion of bypass benefits. Traffic diverting to the bypass enjoys higher speeds on a smoother riding surface with better vertical and horizontal alignment, less congestion and a shorter journey. Traffic continuing to use the existing road also benefits from a reduction in the volume of traffic and hence level of congestion. All these changes affect vehicle speeds and VOCs.

59. VOCs are estimated for all the vehicle types for which traffic forecasts were prepared. For this study the following eight vehicle types are used: • large cars (medium cars plus SUVs – see below) • minibuses • medium buses • large buses • light goods vehicles (LGVs) • medium goods vehicles (MGVs) • heavy goods vehicles (HGVs) • articulated trucks

60. Historically most vehicles were manufactured within the Former Soviet Union (FSU) but the proportion of vehicles of FSU or Russian origin is declining rapidly and vehicle characteristics, shown in Table 6-14, are therefore based on western or Japanese models. SUVs form a significant part of the fleet but were not counted separately from cars during the project traffic counts. A hybrid category of “large car”, representing 25 percent SUVs and 75 percent medium cars, is therefore used in the economic analysis.

61. Table 6-15 shows the economic prices of new vehicles, tires, labor and time. Vehicle and tire prices are quoted US dollar prices in Georgia less import duty (in the case of cars and SUVs) and VAT at the standard rate of 18 percent in the case of all other vehicles. Crew costs vary widely. Those in Table 6-15 are believed to be representative and include social costs. Overheads are the costs of insurance and (for commercial vehicles) administration, training etc.

62. Fuel costs are shown in Table 6-16. Two approaches are used. The first takes the world market price for oil (taken as USD70 per barrel) and adds margins

Engconsult Ltd. 19

for refining, transport from Azerbaijan, distribution and retail. The other deducts taxes from retail prices in Georgia and converts the result to USD using the SCF and the OER. The former gives an economic price of USD0.59/liter and the latter USD0.47 for petrol. The values given by the border price approach are used.

b) Road user cost savings – journey time savings

63. Values adopted for working and non-working journey time savings are also shown in Table 6-15. The values placed by travelers on travel time savings are best established using revealed or stated preference surveys. In the absence of survey results there are two approaches to valuing time savings: • use of an empirical relationship between VOT and GDP per head derived using regression between pairs of values from stated and revealed preference studies; or • use of appropriate income levels.

64. The latter approach is used here. The national average GDP per head at purchasing power parity is approximately USD5,000 (see Supplementary Annex 1), with an urban value of around USD5,500, or USD4/h. Accordingly the range adopted for working time ranges from USD3/h for a working bus passenger to USD4/h for a car passenger. These are close to the values adopted in the JICA study in 2007, though somewhat higher than those in the Samtskhe-Javakheti feasibility study.

65. Further justification for adopting higher than usual values of time can be adduced from tourism. Batumi is the third largest city in Georgia and up until 2007 the growth in tourist arrivals was very strong (see Supplementary Appendix 1, Table 1-10). Even in 2008, which experienced much lower than average economic growth, Ajara tourist arrivals were 285,000 (209,000 domestic plus 76,000 foreign). This is large compared with a resident Batumi population of around 122,000. Fairly high tourist income multipliers are to be expected (definitely in excess of unity) as most goods and services consumed by tourists will be regionally produced and not subject to “leakage”. The same reasons support higher than usual values for bus passengers making working trips, although working trips only represent 10-20 percent of the total.

66. A non-work trip is assigned 40 percent of the value of a working trip. Although it is conventional to take 0-25 percent, recent work suggests that non- working travelers place much higher values than this on travel time savings. For example, stated preference work by IT Transport for the UK Department for International Development (Valuation of Travel Time Savings: Empirical Studies in Bangladesh, Ghana and Tanzania: IT Transport for DFID, July 2005) recommends two approaches to the valuation of travel time savings by non-working adults: (i) 0.55 x the weighted average wage rate per hour adjusted by the SCF, or (ii) 0.37 x household consumption per hour.

67. The real value of passenger travel time will increase as income rises. Unfortunately, HDM-4 does not acknowledge this. A way around this difficulty is to calculate the equivalent increase in the average value that will produce the same present value as an annually increasing amount. The adopted increase, 25 percent, is equivalent, on average, to an annual growth of 3 percent.

68. Goods in transit are unproductive – they represent inventory costs. Thus there is a value in reducing travel times. This value can be substantial for that

Engconsult Ltd. 20

portion of goods that are time sensitive, i.e. where the shipper or recipient bears substantial costs arising from late collection or delivery. Such detail is not available here. For a load of 1 tonne with a value of USD10,000 and a discount rate of 12 percent, saving one hour is worth USD0.14, or USD0.28 for a 2t payload. This is close to HDM-4’s default value of USD0.3/hour, which is adopted. In practice goods in transit savings are negligible compared with other benefits.

c) Road user costs – typical values

69. Table 6-13 shows road user costs for typical conditions on the bypass and on existing roads. The largest savings accrue to large goods vehicles and medium and large buses. Table 6-13 Typical road user costs

Section Car Mini- Med Large LGV MGV HGV Artic bus bus bus Existing roads without bypass (operating speeds 33-51km/h; IRI=5; high side friction): VOC, USD/km 0.22 0.33 0.33 0.89 0.24 0.49 0.79 1.24 Travel time cost, USD/km 0.20 0.37 0.74 1.22 0.03 0.06 0.07 0.01 Total RUC 0.42 0.69 1.07 2.10 0.27 0.55 0.86 1.24 Bypass (operating speeds 80-105km/h; IRI=2; low side friction); VOC, USD/km 0.22 0.30 0.30 0.72 0.22 0.44 0.71 1.11 Travel time cost, USD/km 0.10 0.20 0.45 0.72 0.02 0.04 0.04 0.00 Total RUC 0.32 0.50 0.75 1.43 0.24 0.48 0.75 1.12 Source: consultants’ HDM-4 runs

Engconsult Ltd.

Table 6-14 Basic vehicle characteristics Vehicle class Large car Minibus Medium bus Bus Light goods Medium goods Heavy goods 4x2 tractor + tandem trailer

En HDM-4 base type Car large Minibus Bus medium Bus heavy Light truck Medium truck Heavy truck Articulated truck

g PCUs 1.0 1.3 1.5 2.0 1.3 1.5 2.0 2.5 consult Ltd. Fuel type Petrol Diesel Diesel Diesel Diesel Diesel Diesel Diesel No of tyres 4 4 6 6 6 6 10 14 Tyre type Radial Radial Radial Bias Radial Bias Bias Bias No of axles 2 2 2 3 2 2 3 4 Km/year 10,000 80,000 100,000 100,000 100,000 100,000 100,000 100,000 Hours driven/year 200 2,000 2,000 2,000 2,600 2,600 2,600 2,600 Working time, h/yeara 2,000 2,200 2,200 2,200 3,300 3,300 3,300 3,300 Service life (yr) 15 10 10 10 8 8 8 8 Av no of passengers 3 9 18 30 1 2 2 0 % time private use 100 10 0 0 20 0 0 0 Work related passenger 50 20 20 10 0 0 0 0 trips, % Gross vehicle weight, t 1.4 3.20 7.5 15.0 3.00 13.00 20.00 30.00 Source: manufacturers’ and consultant’s estimates Notes: (a) working time = driving time + time spent loading, unloading and waiting for work. Working time has to be specified for HDM-4. Large values used for cars and motor-cycles in order to avoid spurious savings in time related VOCs

21

Table 6-15 Economic values of vehicles, labor and time

Unit Large car Minibus Medium bus Bus LGV MGV HGV Articulated truck

En Typical model: Renault Temsa + Temsa + Renault Isuzu + Tegeta MAN + Tegeta MAN + Tegeta (4x2 Traffic Mitsubishi Mitsubishi Master tractor + tandem) g consult Ltd. Vehicle cost USD 18,000 23,000 46,000 90,000 24,000 54,000 100,000 100,000

Replacement tyre USD 125 80 170 330 80 170 330 330

Maintenance labour USD/h 2.00 2.00 2.50 3.00 2.50 3.00 3.00 3.50

Crew USD/h 0.00 2.00 2.00 3.00 2.00 4.00 4.00 5.00

Annual overhead USD 900 1,100 2,300 10,000 1,200 2,700 6,500 6,500

Passenger working time USD/h 4.00 3.00 3.00 3.00 0.0 0.0 0.0 0.0

Passenger non-working timea USD/h 1.60 1.20 1.20 1.20 1.20 1.20 1.20 0.0

Source: local dealers, recent construction contracts, consultants’ estimates Note: (a) 40 percent of value of working time. Although it is conventional to take 0-25 percent, recent work (e.g. Valuation of Travel Time Savings: Empirical Studies in Bangladesh, Ghana and Tanzania: IT Transport for DFID, July 2005) suggest that non-working travellers place much higher values than this on travel time savings

22 23

Table 6-16 Fuel and oil prices

Unit Petrol Diesel Lubricating oil

Using border price of oil

Crude oila USD/litre 0.44 0.44 Refiningb USD/litre 0.05 0.02 Rail/road transportb USD/litre 0.05 0.05 Distribution and retailb USD/litre 0.05 0.05 Economic price at the pump USD/litre 0.59 0.56 Using financial retail pump prices

Economic price at the pumpc USD/litre 0.47 0.50 6.00e Financial pump priced GEL/litre 1.12 1.18 Notes: (a) based on FOB spot price of $70/bbl (1bbl = 159 litre) (b) refining and distribution margins deduced from GTZ: International Fuel Prices 2005, 4rd edition, 2005 and other published sources (c) applying the OER of GEL1.67 = USD1 to the financial price less taxes (32 percent plus a 5 percent retailer’s margin) and multiplying by the SCF of 0.97 (d) typical prices in April 2009 (petrol varied from GEL1.05-1.20/liter) (e) based on a retail price range of GEL70-80 per 5l container in Tbilisi in April 2009

d) Generated traffic

70. Generated traffic is additional trips made by existing users of the corridor because journeys are now cheaper or quicker and trips made by others who, in the absence of the project, are deterred from making them because the cost or time is perceived to be prohibitive. The benefit of generated traffic is calculated as 0.5*(T2- T1)*(C1-C2), where T1 and T2 are the number of trips made without and with the bypass and C1 and C2 the road user costs without and with the bypass.

71. Construction of the bypass will reduce RUCs on the existing roads. Estimates of generated traffic are made by calculating the RUC savings on the existing roads and then applying price elasticities. A wide range of elasticities is in use, ranging from -0.2 to -1.0 (a price elasticity of -1.0 means that a 20 percent drop in RUCs generates a 20 percent increase in traffic). The Kobuleti and Batumi bypasses change RUCs by 5-20 percent, depending on vehicle type and road conditions. Assuming an elasticity in the range -0.5 to -1.0 implies generated traffic of 2.5-20 percent of normal traffic. For evaluation purposes 10 percent is taken, but is only applied to light-medium passenger vehicles and light goods vehicles.

e) Accidents

72. Accident rates change following a change in road geometry. In general, a limited access bypass will reduce accident rates but, because speeds are higher, may increase accident severity. HDM-4 models the change in accident costs by comparing the costs and rates of three classes of accident: (i) a fatal accident, (ii) an accident involving injuries but no fatalities and (iii) an accident causing only damage. While damage only data are not to hand, injury and fatality data are available for 2007-9 on the existing Sarpi-Poti road and are summarized in Table 6-17.

Engconsult Ltd. 24

Table 6-17 Accident statistics, Sarpi-Poti road

2007 2008 Jan-Apr Average 2009 Total accidents 139 222 42 Fatalities 25 43 5 Injuries 264 406 73 No of fatal accidents 19 32 5 No of non-fatal accidents 120 190 37 All accidents: av no of fatalities per accident 0.18 0.19 0.12 0.18 av no of injured persons per accident 1.90 1.83 1.74 1.84 Fatal accidents: av no of fatalities per fatal accident 1.3 1.3 1.0 1.3 av no injured persons per fatal accident 2.2 1.4 0.8 1.6 No of injured persons per non-fatal accident 1.9 1.9 1.9 1.9 Sources: police records and consultants’ calculations Note: injured persons are assumed to include pedestrians

73. The estimated 2008 vehicle-km for the Sarpi-Poti road is 170 million. This implies accident rates per million veh-km as follows: • total accidents: 1.30 • fatalities: 0.26 • fatal accidents: 0.18 • non-fatal accidents: 1.12

74. In the absence of local or regional data or research findings the impact of the bypass on accident rates and accident severity has to be inferred from other projects. For example, in 2001 accident rates per million PCU-km were 0.34 and 0.84 on Chinese expressways and class III highways respectively. South African research (reported in Predicting Changes in Accident rates in Developing Countries following Modifications in Road Design, TRL, 1995) found a base accident rate of 0.24 “personal injury collisions” per million veh-km on a multi-lane highway, increasing to 0.34 for a two lane road, increasing to 0.64 if junctions were at-grade (rather than grade-separated). The case for assuming increased severity on the bypass rests on the link between speed and severity, which has been the subject of much research. For example, a Swedish international risk model for predicting accidents (see TRL report cited above) relates the frequency of fatalities to the fourth power of the relative increase in speed. Thus, if the bypass increased the average operating speed from 60 to 80km/h, the frequency of fatalities would be predicted to rise threefold.

75. Table 6-18 shows (i) a summary of the accident analysis for the existing road taken from the column headed “average” in Table 6-17 and (ii) the consultants’ assumptions for the bypass. Overall, the number of accidents per million veh-km is assumed to drop by about 50 percent, from 1.3 (=0.18+1.12) to 0.68 (=0.08+0.6), whereas the number of fatalities per average accident rises by 50 percent, from 0.18 to 0.24. The first five columns represent assumptions. Columns 6 and 7 are results calculated using the assumptions. For example, if the bypass carries 100m veh-km,

Engconsult Ltd. 25

the expected number of accidents is 68 (=100x(0.08+0.6)), of which 8 will be fatal, with a total of 8x2.0=16 deaths, giving a severity per accident of 16÷68=0.24. The expected number of injuries is 8x1.9 arising from fatal accidents and 60x2.3 from non-fatal incidents, i.e. a total of 153.2 injuries, equivalent to a severity per accident of 153.2÷68=2.25. Table 6-18 With and without project accident rates

Road type Accidents/million veh-km Severity per average accident Fatal accident Non-fatal All Fatal Non-fatal Fatalities Injuries Injuries Fatalities Injuries accidents accidents Existing road 0.18 1.12 1.3 1.6 1.9 0.18 1.84 Bypass 0.08 0.6 2.0 1.9 2.3 0.24 2.25 Sources: consultants’ assumptions

76. The table below shows with and without project average numbers of accidents per year over the evaluation period. Values are for combined tranches 1 and 2, base case traffic case. Table 6-19 With and without average numbers of accidents per year

Fatal accidents Non-fatal accidents Total Without project 42 259 301 With project 23 162 185 Sources: consultants’ HDM-4 runs for combined tranches 1 and 2, base case traffic.

77. The “human capital” approach to valuation is to estimate the sum of (i) an appropriate proportion of income foregone, (ii) marginal emergency and medical costs and (iii) marginal damage costs.

78. A representative range of incomes is from USD2,400 (2008 GDP/head converted to USD at the OER) to USD5,500 (PPP estimate of urban income used for journey time savings). Thus the cost of a fatality based on a 12 percent discount rate and 20 years’ foregone income ranges from USD18,000 to USD41,000. If marginal emergency services and damage costs add a further 20 percent, the total cost of a fatality is USD22-49,000. The mid-point value of USD36,000 is adopted. For comparison, a recent evaluation for ADB in Azerbaijan (appendix 10 of RRP for a Proposed MFF for the Republic of Azerbaijan: Road Network Development Program, Sep 2007) valued a fatality at USD25,000.

79. Injury costs vary widely. Work in India (Dinesh Mohan: Social Cost of Road Traffic Crashes in India, Proc First Safe Community Conference on Cost of Injury, Viborg, 2002) estimated that the social cost of a serious injury was 20-45 percent of that of a fatality, while a minor injury was only worth 4 percent. Minor injuries are usually under-reported. Assuming that also to be true in Georgia and taking the mid- point of 32 percent, a serious injury would be valued at USD11,500.

80. Average accident costs can now be calculated using Table 6-18 data: • on the existing road: USD27,600 • on the bypass: USD34,500

Engconsult Ltd. 26

81. The percentage of GDP lost to road accidents can be estimated using national data in Supplementary Appendix 4. In 2008 there were 867 fatalities and 9,063 injuries. Applying the values adopted here gives a total social cost of USD135m, or 1.3 percent of GDP. This is at the lower end of the usually accepted range of 1 to 3 percent.

B. Evaluation of project 1 – the Kobuleti bypass

1. Road sections

82. For economic evaluation purposes the existing S-2 road (from Choloki bridge in the north to the start of the Batumi bypass at Makhinjauri) is divided into six homogeneous sections. Section K1a, around Kobuleti town, carries additional local traffic (i.e. traffic that would never use the bypass – see section III.D).

83. The proposed bypass is divided into three sections. Each bypass section is defined by interchanges either with the S-2 or (at Kobuleti town) or with the Ozurgeti road.

84. HDM-4 section definitions for the existing road and proposed bypass are shown in Table 6-20.

Engconsult Ltd.

Table 6-20 Kobuleti: existing and bypass road sections

Section km Lanes Width Shoulder Friction Rise + fall Rises & Horiz curve Speed Surface m m m/km falls o/km limit km/h no/km thickness mm En XMT XFRI XNMT g consult Ltd. Existing road sections: K1: Choloki br-Kabuleti center 7.1 2 12 1 0.7 0.7 0.9 10 2 50 50 80 K1a: Choloki-Kabuleti 4.1 2 12 1 0.6 0.6 0.6 10 2 50 25 80 K2: Kabuleti-railway br 4.0 2 8 1 0.7 0.7 0.9 10 2 50 50 100 K3: Railway br- Chakvi N 4.5 2 8 1 0.7 0.7 0.9 100 20 500 50 100 K4: Chakvi N-Chakvi 2.3 2 7 1 0.7 0.7 0.9 10 2 50 50 100 K5: Chakvi-bypass IC 3.9 2 10 1 0.7 0.7 0.9 10 10 100 50 100 K6: Bypass IC-Makhinjauri 5.6 2 10 1 0.7 0.7 0.9 10 2 50 50 100 Total 31.5 Bypass road sections: KB1: km0-11.3 11.3 2 7 2 1.0 1.0 1.0 15 2 75 120 125 KB2: km11.3-28.2 16.9 2 7 2 1.0 1.0 1.0 15 2 75 120 125 KB3: km28.2-33.8 5.6 4 14 2 1.0 1.0 1.0 15 2 75 120 125 Total 33.8 Notes: (a) K1a is the urban section of K1 to which local traffic is added (see section D); K3 is steep and has many hairpin bends (b) XNMT = non-motorized traffic friction factor (range 0.6-1.0); XFRI = roadside friction (range 0.6-1.0); XMT = friction arising from roadside activities and motorized traffic (range 0.4-1.0). A friction factor of 0.9 means speeds are 90 percent of their undisturbed values (c) existing road: IRI = 6m/km at the end of 2008, 5 potholes/km (also end 2008). Bypass has IRI=2 on completion (d) existing road: seasonal traffic flow pattern adopted (see section II.C); HDM-4 default two lane road speed flow curve adopted. Bypass: HDM-4 free flow traffic pattern and default two lane wide speed flow curve adopted (e) K6 includes the Chakvi-Makhinjauri tunnel and has the same alignment as KB3 (f) all other characteristics assume HDM-4 default values. Pavement has an asphaltic concrete surface with an asphalt mix pavement on a granular base

27 28

2. Costs

a) Project costs

85. Project financial costs are taken from Supplementary Appendix 3. Table 6-21 shows the derivation of the cost at economic prices. The residual value is 65 percent of the economic cost in Table 6-21. Table 6-21 Kobuleti bypass economic costs Item USD m, 2009 prices Financial base cost + physical contingencies 133.2 Financial base cost + physical contingencies per km 3.94 Adjustment to economic prices Shadow wage adjustment: 0.3 x 15% x $133m financial construction cost -5.99 Adjustment for VAT (financial base cost + physical contingencies)x(1-1/1.18) -20.32 Adjustment for contractor’s company taxes: 15% x net margin of 5% x fin cost less VAT -0.85 SCF applied to non-tradable inputs: 0.03 x 80% x financial cost -3.20 Total adjustment -30.36 Cost at economic prices 102.8 Cost at economic prices per km 3.04 Source: consultants’ estimates Notes: costs/km based on 33.8km

86. The cost of maintaining the bypass road comprises routine maintenance of the highway (drain cleaning, crack sealing etc) and the periodic cost of an overlay once pavement condition has deteriorated to the point where routine maintenance is no longer adequate. In addition there are small costs involved in operating the tunnels (both the existing 800m C-M tunnel, which has ventilation equipment, and the 600m of new tunnel, which will have lighting but no ventilation).

87. The Roads Department generally budgets GEL3,400/km (excluding VAT) (USD2,000) for routine maintenance of a two lane road and have indicated that they would consider this appropriate for the project roads, at least initially. HDM-4 calculates maintenance costs in response to modeled deterioration, not as an annual amount. If overlay and tunnel maintenance costs are excluded, the unit costs in Table 6-22 generate maintenance costs that approximate those budgeted by the Roads Department. Table 6-22 Kobuleti bypass maintenance costs Item Unit Cost Intervention Clearing snow etc USD/km 150 Annual Crack sealing USD/m2 10 Wide structural cracking>=10% Pothole patching USD/m2 12 At >=10 potholes/km 50mm overlay USD/m2 15a At IRI=6 Tunnel maintenanceb USD/km 250 Annual Source: current Ajara road maintenance contract and consultants’ estimates Note: (a) plus preparatory works: edge repair at USD14/m2 and patching at USD12/m2

Engconsult Ltd. 29

(b) includes existing 800m C-M tunnel as well as proposed 400m tunnel. Costs are per highway km.

b) Do minimum costs

88. Roads Department has supplied the current road maintenance contract for Ajara and extracted the actual expenditure for 2007-8 on the Choloki-Sarpi section of the S-2. Table 6-23 S-2 maintenance expenses, 2007-8 Item 2007 2008 Winter maintenance, USD/km 290 220 Routine maintenance, USD/km 1,420 1,800 Maintenance of C-M tunnel, USD 22,300 32,800 Source: Roads Department Note: values exclude VAT

89. Essentially, routine maintenance costs approximately USD2,000/km and maintaining the tunnel approximately USD500 per highway km. For modeling using HDM-4, the unit costs in Table 6-24 were used: if overlay costs are excluded, HDM- 4 generates approximately the same costs as those actually incurred. Table 6-24 Do minimum maintenance costs Item Unit Cost Intervention Clearing snow etc USD/km 250 Annual Crack sealing USD/m2 10 Wide structural cracking>=10% Pothole patching USD/m2 12 At >=10 potholes/km 50mm overlay USD/m2 15a At IRI=6 Tunnel maintenance USD/km 500 Annual Source: current Ajara road maintenance contract and consultants’ estimates Note: (a) plus preparatory works (see Table 6-20)

3. Evaluation results

a) Case A (base case) traffic

90. Table 6-25 shows the summary of base case discounted costs and benefits from HDM-4. The striking feature is the high proportion of time saving benefits. This arises because: • the existing road is highly congested • the bypass is 2.3km (7 percent) longer than the existing road, leading to a small proportion of VOC savings

91. Evidence of the importance of journey time savings is available from calculated operating speeds. Using HDM-4 average speeds, use of the bypass around Kobuleti approximately halves total journey time, from 0.8h to 0.4h. This is an average for the entire year and as such probably underestimates savings during the summer holiday season. Under free flow traffic conditions the consultants’

Engconsult Ltd. 30

vehicle took 0.67h (implying an average speed of 47km/h), suggesting that 0.8h (39km/h) is reasonable for an annual average. Table 6-25 Tranche 1 (Kobuleti) base case: economic evaluation PV, incremental Benefits, USDm NPV, cost, USDm Time VOC Accident Total EIRR % USDm savings savings savings 82.7 104.7 12.6 6.3 123.6 40.6 16.2 84.7% 10.2% 5.1% Source: consultants’ HDM-4 runs Note: all PVs discounted at 12 percent; totals may not agree as the result of rounding

92. Table 6-26 shows incremental costs and benefits and is compiled from HDM- 4 output. Benefits from generated traffic are lower than usual; this is because the drop in road user costs on the existing road sections are not great and so the resulting increase in traffic is also not large (see also section IV.5.d). Table 6-26 Tranche 1 (Kobuleti) base case: costs and benefits

Year Incremental costs, USDm Road user cost savings, USDm Net benefits Capital Recurrent Total VOC Time savingsGenerated Accident Total USDm traffic savings 2010 23.22 0.00 23.22 0.00 0.00 0.00 0.00 0.00 -23.22 2011 30.52 0.00 30.52 0.00 0.00 0.00 0.00 0.00 -30.52 2012 28.05 0.00 28.05 0.00 0.00 0.00 0.00 0.00 -28.05 2013 20.96 0.00 20.96 0.00 0.00 0.00 0.00 0.00 -20.96 2014 0.00 0.01 0.01 1.05 12.95 0.29 0.87 15.16 15.14 2015 0.00 0.01 0.01 1.16 13.69 0.31 0.91 16.07 16.05 2016 0.00 0.01 0.01 1.30 14.48 0.33 0.96 17.07 17.05 2017 0.00 0.01 0.01 1.51 15.32 0.35 1.00 18.18 18.16 2018 0.00 0.01 0.01 1.76 16.22 0.37 1.05 19.40 19.39 2019 0.00 -0.33 -0.33 2.05 17.18 0.40 1.10 20.72 21.05 2020 0.00 0.36 0.36 2.25 18.19 0.42 1.16 22.02 21.65 2021 0.00 -0.33 -0.33 2.57 19.25 0.45 1.21 23.48 23.81 2022 0.00 0.58 0.58 2.71 20.37 0.47 1.27 24.82 24.24 2023 0.00 -0.24 -0.24 3.06 21.58 0.51 1.33 26.48 26.72 2024 0.00 0.58 0.58 3.23 22.86 0.54 1.40 28.02 27.44 2025 0.00 -0.21 -0.21 3.62 24.21 0.57 1.47 29.87 30.09 2026 0.00 0.58 0.58 3.89 25.63 0.61 1.54 31.67 31.09 2027 0.00 -0.21 -0.21 4.35 27.11 0.65 1.61 33.72 33.93 2028 0.00 0.58 0.58 4.60 28.63 0.69 1.69 35.61 35.03 2029 0.00 -0.21 -0.21 5.10 30.21 0.73 1.78 37.82 38.03 2030 -0.50 0.35 -0.15 5.33 31.70 0.78 1.86 39.67 39.82 2031 -1.28 0.10 -1.18 3.20 35.75 0.75 1.95 41.65 42.83 2032 0.00 0.36 0.36 3.35 41.91 0.88 2.05 48.19 47.83 2033 -65.61 0.01 -65.60 5.08 49.09 1.05 2.15 57.37 122.97

PV at 12% 82.7 0.3 83.0 12.3 102.7 2.4 6.4 123.7 40.6 Percent of total road user cost savings 9.9% 83.0% 1.9% 5.1% 100.0% EIRR 16.2% Source: consultants' HDM-4 runs Note: generated traffic includes both VOC and time savings, hence there are differences between the PVs in this table and those in the evaluation summary

b) Case B traffic

93. Table 6-27 shows the effect of diversion to the Kartsakhi corridor. As expected, it reduces the economic returns significantly, although the project remains worthwhile at a 12 percent discount rate.

Engconsult Ltd. 31

Table 6-27 Kobuleti traffic case B: discounted costs and benefits PV, incremental Benefits, USDm NPV, cost, USDm Time VOC Accident Total EIRR % USDm savings savings savings 82.8 81.8 6.9 5.2 93.9 11.1 13.3 87.1% 7.4% 5.5% Source: consultants’ HDM-4 runs Note: all PVs discounted at 12 percent

C. Evaluation of tranche 2: the Batumi bypass

1. Road network

94. The existing S-2 road (from the start of the Batumi bypass at Makhinjauri to the interchange with the approach road to the Chorokhi bridge) is divided into three homogeneous sections. Section B2, in the center of Batumi, carries additional local traffic (see section D). There is also an inner transit traffic route within Batumi whose surface is poorly laid concrete slabs; they have been given an IRI of 7m/km.

95. The proposed bypass is likewise divided into three sections, each linking tow interchanges. HDM-4 section definitions for the existing and proposed roads are shown in Table 6-28.

2. Costs

a) Project costs

96. Project financial costs are taken from Supplementary Appendix 3. Table 6-28 shows the derivation of the cost at economic prices. The residual value is 65 percent of the economic cost in Table 6-28. Table 6-28 Batumi bypass economic costs Item USD m, 2009 prices Financial base cost + physical contingencies 133.0 Financial base cost + physical contingencies per km 9.57 Adjustment to economic prices Shadow wage adjustment: 0.3 x 15% x $133m financial construction cost -5.99 Adjustment for VAT (financial base cost + physical contingencies)x(1-1/1.18) -20.31 Adjustment for contractor’s company taxes: 15% x net margin of 5% x fin cost less VAT -0.85 SCF applied to non-tradable inputs: 0.03 x 80% x financial cost -3.20 Total adjustment -30.35 Cost at economic prices 102.6 Cost at economic prices per km 7.38 Source: consultants’ estimates Notes: costs/km based on 13.9km

Engconsult Ltd.

Table 6-29 Tranche 2 – Batumi bypass: existing and bypass road sections

Section km Width Shoulder Friction Rise + fall Rises Horiz Speed Surface m m m/km & falls curve o/km limit thickness mm En no/km km/h g XMT XFRI consult Ltd. XNMT

Existing road sections: B1: Makhinjauri-Batumi cr 5.3 10 2 0.6 0.6 0.6 10 2 50 50 80 B2: Batumi cr-A306 jn 3.8 10 2 0.6 0.6 0.6 10 2 50 50 80 B3: A306 jn - Chorokhi IC 4.1 10 2 0.6 0.6 0.7 10 2 50 50 100 Total 13.2 Bypass road sections: BT1: km33.8-39.2 5.4 7 2 1.0 1.0 1.0 15 2 75 120 125 BT2: km39.2-44.6 5.4 7 2 1.0 1.0 1.0 15 2 75 120 125 BT3: km44.6-47.7 3.1 7 2 1.0 1.0 1.0 15 2 75 120 125 Total 13.9 Notes: (a) B2 is the urban section to which local traffic is added (see section D) (b) XNMT = non-motorized traffic friction factor (range 0.6-1.0); XFRI = roadside friction (range 0.6-1.0); XMT = friction arising from roadside activities and motorized traffic (range 0.4-1.0). A friction factor of 0.9 means speeds are 90 percent of their undisturbed values (c) IRI = 6m/km at the end of 2008 (except for B2, IRI=7), 5 potholes/km (also end 2008); bypass: IRI = 2m/km on completion (d) existing road: seasonal traffic flow pattern adopted (see section II.C); HDM-4 default two lane road speed flow curve adopted. Bypass: HDM-4 free flow traffic pattern and default two lane wide speed flow curve adopted (e) A306 is the Batumi-Akhaltsikhe road (interchange with Batumi bypass) (f) all other characteristics assume HDM-4 default values. Pavement has an asphaltic concrete surface with an asphalt mix pavement on a granular base

32

33

97. Maintenance costs are compiled on the same basis as that adopted for the Kobuleti bypass – see Table 6-30 below. Do minimum costs are calculated on exactly the same basis as they are for Kobuleti, but omitting tunnel costs. Table 6-30 Batumi bypass maintenance costs Item Unit Cost Intervention Clearing snow etc km 150 Annual Crack sealing m2 10 Wide structural cracking>=10% Pothole patching m2 12 At >=10 potholes/km 50mm overlay m2 15 At IRI=6 Tunnel maintenancea km 130 Annual Source: current Ajara road maintenance contract and consultants’ estimates Note: (a) illumination costs per highway km for 1,540m of new tunnel

3. Evaluation results

98. Table 6-31 shows the summary of base case discounted costs and benefits from HDM-4. As in the case of tranche 1, and for the same reasons, savings are dominated by time savings. In this case the bypass is 5 percent longer than the existing road.

99. Evidence of the importance of journey time savings is available from calculated operating speeds. Using HDM-4 average speeds, use of the bypass around Batumi saves approximately 15 minutes and reduces the without project journey time of 25 minutes by about 60 percent. This is an average for the entire year and as such probably underestimates savings during the summer holiday season. (The consultants’ vehicle, under free flow traffic conditions, took 22 minutes).

100. Table 6-31 shows that the economic case for tranche 2 is quite weak. This arises from high initial costs and low traffic: economic investment costs are USD7.4m/km and opening traffic is only 3,400 veh/day, whereas tranche 1’s cost is USD3.0m/km and opening traffic 3,600 veh/day. Table 6-31 Tranche 2 (Batumi) base case economic evaluation PV, incremental Benefits, USDm NPV, cost, USDm Time VOC Accident Total EIRR % USDm savings savings savings 77.4 70.8 11.7 3.5 86.0 8.7 13.1% 82.3% 13.6% 4.1% Source: consultants’ HDM-4 runs Note: all PVs discounted at 12 percent

101. Table 6-32 shows incremental costs and benefits and is compiled from HDM-4 output.

Engconsult Ltd. 34

Table 6-32 Tranche 2 (Batumi) base case: costs and benefits

Year Incremental costs, USDm Road user cost savings, USDm Net benefits Capital Recurrent Total VOC Time savingsGenerated Accident Total USDm traffic savings 2010 8.21 0.00 8.21 0.00 0.00 0.00 0.00 0.00 -8.21 2011 22.57 0.00 22.57 0.00 0.00 0.00 0.00 0.00 -22.57 2012 32.83 0.00 32.83 0.00 0.00 0.00 0.00 0.00 -32.83 2013 38.98 0.00 38.98 0.00 0.00 0.00 0.00 0.00 -38.98 2014 0.00 0.00 0.00 1.32 9.08 0.06 0.48 10.93 10.93 2015 0.00 0.00 0.00 1.42 9.58 0.06 0.51 11.57 11.56 2016 0.00 0.00 0.00 1.54 10.11 0.07 0.53 12.25 12.25 2017 0.00 0.00 0.00 1.69 10.68 0.07 0.56 13.00 13.00 2018 0.00 -0.04 -0.04 1.86 11.29 0.08 0.59 13.81 13.85 2019 0.00 -0.10 -0.10 2.03 11.93 0.09 0.61 14.66 14.76 2020 0.00 0.10 0.10 2.18 12.63 0.10 0.64 15.55 15.44 2021 0.00 -0.10 -0.10 2.39 13.36 0.11 0.68 16.53 16.63 2022 0.00 0.21 0.21 2.53 14.15 0.12 0.71 17.51 17.30 2023 0.00 -0.10 -0.10 2.78 15.00 0.13 0.74 18.66 18.76 2024 0.00 0.21 0.21 2.94 15.91 0.15 0.78 19.77 19.56 2025 0.00 -0.10 -0.10 3.20 16.83 0.16 0.82 21.01 21.11 2026 0.00 0.21 0.21 3.38 17.82 0.18 0.86 22.24 22.02 2027 0.00 -0.10 -0.10 3.67 18.87 0.20 0.90 23.64 23.74 2028 -0.57 0.21 -0.36 3.83 19.97 0.23 0.94 24.97 25.33 2029 0.89 -0.04 0.86 2.70 21.12 0.26 0.99 25.07 24.22 2030 -0.23 0.12 -0.11 3.78 22.38 0.29 1.04 27.48 27.60 2031 0.00 0.00 0.00 3.12 23.03 0.29 1.09 27.53 27.52 2032 -0.05 0.11 0.06 3.16 23.53 0.34 1.14 28.17 28.11 2033 -66.68 0.00 -66.67 3.59 25.60 0.37 1.20 30.76 97.44

PV at 12% 77.4 0.1 77.4 11.6 70.3 0.6 3.5 86.1 8.7 Percent of total road user cost savings 13.5% 81.7% 0.7% 4.1% 100.0% EIRR 13.1% Source: consultants' HDM-4 runs Note: generated traffic includes both VOC and time savings, hence there are differences between the PVs in this table and those in the summary evaluation table

D. Benefits distribution

102. Table 6-33 shows the distribution of benefits for tranche 1 (Kobuleti), traffic case A. The striking feature is the proportion of benefits accruing to foreign registered vehicles, nearly of which will be transit traffic. The policy implication is that Georgia should explore ways of recouping some of its infrastructure costs from transit traffic by means of tolls or transit fees.

Engconsult Ltd. 35

Table 6-33 Benefits distribution, tranche 1 Benefit Passenger Goods vehicles Total Percent vehicles PV in USDm VOC (normal traffic) 1.7 10.6 12.3 10.0% Time savings 101.5 1.1a 102.6 83.0% Generated trafficb 2.3 0.1 2.4 1.9% Accident savingsc 5.1 1.2 6.3 5.1% Total 110.6 13.0 123.6 100%e Foreign registered vehiclesd 52 42% Source: consultants’ calculations Notes: (a) some goods vehicles carry one or two passengers (b) HDM-4 splits generated traffic by VOC and time savings: in this table passenger vehicle benefits are HDM-4 time savings (c) split using the fleet composition on the existing roads (d) based on the proportions of foreign registered vehicles observed at the C-M tunnel (see Table 6-3) (e) percentages may not appear to add up to 100% as a result of rounding

E. Combined project evaluation (tranches 1 and 2)

1. Base case

103. The combined evaluation makes use of incremental costs and road user cost savings exported from HDM-4 for each tranche. The combined streams of incremental costs and benefits are shown in Table 6-35. The base case EIRR of 14.8 percent is above the 12 percent threshold but is vulnerable to lower than forecast traffic growth (section E) and the possibility of corridor diversion (see case B below).

104. Reducing the residual value to 30 percent of initial cost (see section IV.A.3) reduced the EIRR from 14.8 to 14.3 percent.

105. Journey time savings, taken from sections B and C are shown in Table 6-34. Table 6-34 Journey time savings Existing roads Bypass Hours km Hours km Tranche 1 (Kobuleti) 0.8 32 0.4 34 Tranche 2 (Batumi) 0.4 13 0.17 14 Total 1.2 45 0.57 48 Source: current Ajara road maintenance contract and consultants’ estimates

Engconsult Ltd. 36

Table 6-35 Tranches 1 and 2 base case: costs and benefits

Year Incremental costs, USDm Road user cost savings, USDm Net benefits Capital Recurrent Total VOC Time savings Generated Accident Total USDm traffic savings 2010 31.43 0.00 31.43 0.00 0.00 0.00 0.00 0.00 -31.43 2011 53.09 0.00 53.09 0.00 0.00 0.00 0.00 0.00 -53.09 2012 60.88 0.00 60.88 0.00 0.00 0.00 0.00 0.00 -60.88 2013 59.94 0.00 59.94 0.00 0.00 0.00 0.00 0.00 -59.94 2014 0.00 0.02 0.02 2.36 22.02 0.35 1.35 26.09 26.07 2015 0.00 0.02 0.02 2.58 23.26 0.37 1.42 27.63 27.62 2016 0.00 0.02 0.02 2.85 24.59 0.40 1.49 29.32 29.30 2017 0.00 0.02 0.02 3.19 26.00 0.42 1.56 31.18 31.16 2018 0.00 -0.03 -0.03 3.62 27.51 0.45 1.64 33.21 33.24 2019 0.00 -0.43 -0.43 4.07 29.11 0.49 1.72 35.39 35.82 2020 0.00 0.47 0.47 4.43 30.81 0.52 1.80 37.56 37.09 2021 0.00 -0.43 -0.43 4.96 32.61 0.56 1.89 40.01 40.44 2022 0.00 0.79 0.79 5.23 34.53 0.59 1.98 42.33 41.54 2023 0.00 -0.34 -0.34 5.85 36.58 0.64 2.08 45.14 45.48 2024 0.00 0.79 0.79 6.17 38.76 0.68 2.18 47.79 47.00 2025 0.00 -0.31 -0.31 6.82 41.05 0.74 2.29 50.88 51.19 2026 0.00 0.79 0.79 7.27 43.44 0.79 2.40 53.90 53.11 2027 0.00 -0.31 -0.31 8.01 45.98 0.85 2.51 57.36 57.67 2028 -0.57 0.79 0.22 8.43 48.60 0.91 2.64 60.58 60.36 2029 0.89 -0.25 0.64 7.80 51.34 0.99 2.77 62.89 62.25 2030 -0.73 0.47 -0.26 9.11 54.08 1.07 2.90 67.16 67.42 2031 -1.28 0.10 -1.17 6.32 58.77 1.05 3.04 69.18 70.35 2032 -0.05 0.47 0.42 6.51 65.44 1.22 3.19 76.36 75.94 2033 -132.29 0.02 -132.27 8.67 74.69 1.42 3.35 88.13 220.41

PV at 12% 160.1 0.4 160.5 23.9 173.0 3.0 9.9 209.8 49.3 Percent of total road user cost savings 11.4% 82.5% 1.4% 4.7% 100.0% EIRR 14.8% Source: consultants' HDM-4 runs 2. Case B traffic

106. The combined streams of incremental costs and benefits for the case where some traffic takes the Kartsakhi corridor are shown in Table 6-36. The combined project just achieves an acceptable EIRR. If the residual value is 30 percent of the initial cost the EIRR falls from 12.4 to 12.1 percent.

Engconsult Ltd. 37

Table 6-36 Tranches 1 and 2: costs and benefits for case B traffic

Year Incremental costs, USDm Road user cost savings, USDm Net benefits Capital Recurrent Total VOC Time savings Generated Accident Total USDm traffic savings 2010 23.22 0.00 23.22 0.00 0.00 0.00 0.00 0.00 -23.22 2011 61.00 0.00 61.00 0.00 0.00 0.00 0.00 0.00 -61.00 2012 58.53 0.00 58.53 0.00 0.00 0.00 0.00 0.00 -58.53 2013 61.61 -0.05 61.55 0.00 0.00 0.00 0.00 0.00 -61.55 2014 0.00 0.02 0.02 1.85 17.86 0.28 1.11 21.09 21.08 2015 0.00 0.02 0.02 2.01 18.85 0.30 1.16 22.33 22.31 2016 0.00 0.02 0.02 2.04 19.89 0.32 1.22 23.47 23.45 2017 0.00 0.02 0.02 2.26 20.99 0.34 1.28 24.86 24.84 2018 0.00 0.02 0.02 2.53 22.15 0.36 1.34 26.38 26.36 2019 0.00 -0.42 -0.42 2.84 23.38 0.38 1.40 28.01 28.43 2020 0.00 0.46 0.46 3.06 24.69 0.41 1.47 29.63 29.18 2021 0.00 -0.42 -0.42 3.46 26.09 0.43 1.54 31.53 31.95 2022 0.00 0.79 0.79 3.68 27.58 0.46 1.62 33.34 32.54 2023 0.00 -0.29 -0.29 4.10 29.16 0.49 1.70 35.45 35.74 2024 0.00 0.74 0.74 4.35 30.84 0.53 1.78 37.49 36.75 2025 0.00 -0.29 -0.29 4.81 32.60 0.56 1.87 39.84 40.13 2026 0.00 0.74 0.74 5.16 34.47 0.60 1.96 42.19 41.45 2027 0.00 -0.26 -0.26 5.76 36.47 0.64 2.05 44.92 45.18 2028 0.00 0.70 0.70 6.10 38.53 0.69 2.15 47.46 46.76 2029 -0.57 -0.18 -0.75 6.80 40.69 0.74 2.26 50.49 51.23 2030 0.00 0.70 0.70 6.12 42.94 0.79 2.37 52.21 51.52 2031 -3.34 -0.05 -3.38 6.68 45.29 0.85 2.48 55.30 58.69 2032 -0.94 0.66 -0.27 1.82 47.62 0.82 2.61 52.86 53.14 2033 -132.83 0.15 -132.69 0.76 49.60 0.84 2.73 53.93 186.62

PV at 12% 157.9 0.4 158.3 16.8 137.4 2.3 8.1 164.5 6.2 Percent of total road user cost savings 10.2% 83.5% 1.4% 4.9% 100.0% EIRR 12.4% Source: consultants' HDM-4 runs

F. Sensitivity and risk

1. Tranche 1 (Kobuleti)

a) Sensitivity

107. Table 6-37 shows how tranche 1’s EIRR reacts to changes in: • benefits • traffic case • construction costs • traffic growth • delayed opening

Engconsult Ltd. 38

108. The EIRR remains above 12 percent for all costs up to 1.5 times base and for benefits above 0.7 times base, and is fairly robust in the face of a fall in traffic, including opening of the Kartsakhi corridor. A modest improvement is achieved by delaying construction and opening by a year.

Table 6-37 Sensitivity tests: tranche 1

Scenario EIRR percent NPV in USD Switching million factora (a) Base case 16.2% 40.6 (b) Traffic case B (Kartsakhi diversion) 13.3% 11.1 (c) Benefits less 10% 15.0% 28.3 0.67 (d) Benefits less 20% 13.7% 15.9 (e) Cost increased by 10% 15.1% 32.3 1.49 (f) Cost increased by 20% 14.2% 24.0 (g) RV reduced from 65% to 30% 16.0% 38.0 (h) Combination (c) and (e) 14.0% 20.0 (i) Traffic Growth + 10% 17.6% 58.8 (j) Traffic Growth -20% 13.1% 9.8 (k) Opening delayed by a year 17.0% 48.9 Notes: (a) the switching factor is the factor times the base value of the indicated item that will produce an EIRR of 12 percent

b) Risk analysis

109. Table 6-38 shows the ranges used for variables included in the risk analysis. The negative skew applied to costs reflects (a) the weak construction market in 2009 and (b) the scope for optimization of the alignment. The negative skews applied to all benefits reflect the scope for loss of some corridor traffic. Table 6-38 Input ranges for risk analysis

Variable Triangular distributions Multiplying factors Minimum Maximum VOC benefits 0.5 1.2 Generated traffic 0.5 1.2 Time savings 0.5 1.2 Accident savings 0.5 1.2 All investment & operating costs 0.8 1.1 Source: consultants’ assumptions

110. The results of 1,000 simulations using RiskAMP are shown graphically in Figure 6-5 and summarized in Table 6-39. As the input distributions are mostly skewed against the base case EIRR, the resulting frequency distribution of EIRRs is itself slightly negatively skewed. The result is mean and median EIRRs that are lower than

Engconsult Ltd. 39

the base case value. However, in only 3.4 percent of cases is the EIRR below the 12 percent threshold. Table 6-39 Results of risk analysis, tranche 1

Statistic Value, percent

Base case EIRR (Table 6-21) 16.2 RiskAMP results: Mean EIRR 15.4 Median EIRR 15.5 Range of EIRRs 10.2-20.2 Percentage of EIRRs < 12 percent 3.4

Source: consultants’ estimates

Figure 6-5 Frequency distribution of tranche 1 EIRR

12.0%

10.0%

8.0%

6.0%

4.0% Percent falling inclassinterval

2.0%

0.0% 10% 12% 13% 15% 16% 18% 20% EIRR

Engconsult Ltd. 40

2. Tranches 1 and 2

a) Sensitivity

111. See Table 6-40. The EIRR remains above 12 percent for all costs up to 1.3 times base and for benefits above 0.8 times base, but is vulnerable to a fall in traffic, including opening of the Kartsakhi corridor. The improvement arising from a delayed opening is proportionately greater than it is in the case of tranche 1 alone, indicating in the particular benefits of a delay to tranche 2 (Batumi). Table 6-40 Sensitivity tests: tranches 1 and 2

Scenario EIRR percent NPV in USD Switching million factora (a) Base case 14.8% 49.3 (b) Traffic case B (Kartsakhi diversion) 12.4% 6.2 (c) Benefits less 10% 13.7% 28.3 0.77 (d) Benefits less 20% 12.4% 7.4 (e) Cost increased by 10% 13.8% 33.3 1.31 (f) Cost increased by 20% 12.9% 17.2 (g) Combination (c) and (e) 12.7% 12.3 (h) RV reduced from 65% to 30% 14.3% 45.0 (i) Traffic Growth + 10% 16.6% 81.5 (j) Traffic Growth -20% 11.3% -6.4 (k) Opening delayed by a year 15.5% 61.4 Notes: (a) the switching factor is the factor times the base value of the indicated item that will produce an EIRR of 12 percent

b) Risk analysis

112. Input ranges are the same as those adopted for tranche 1 (Table 6-38). The results of 1,000 simulations using RiskAMP are shown graphically in Figure 6-5 and summarized in Table 6-41. As the input distributions are mostly skewed against the base case EIRR, the resulting frequency distribution of EIRRs is itself slightly negatively skewed. The result is mean and median EIRRs that are lower than the base case value. However, in 14 percent of cases the EIRR is below the 12 percent threshold.

Engconsult Ltd. 41

Table 6-41 Results of risk analysis

Statistic Value, percent Base case EIRR (Table 6-31) 14.8 RiskAMP results: Mean EIRR 14.0 Median EIRR 14.1 Range of EIRRs 8.7-18.5 Percentage of EIRRs < 12 percent 14.4 Source: consultants’ estimates

Figure 6-6 Frequency distribution of EIRR (tranches 1 and 2)

12.0%

10.0%

8.0%

6.0%

4.0% Percent falling in class interval class Percent in falling

2.0%

0.0% 10% 12% 13% 15% 16% 18% 20% EIRR

Engconsult Ltd. 42

c) Tranches 1 and 2 with 30 percent residual value

113. At ADB’s request an additional set of evaluations was prepared, this time using a residual value (RV) equal to 30 (rather than 65) percent of the initial cost. The results are summarised here as well as being shown in Tables 6.37 and 6.39. In general, reducing the residual value in this way reduces the EIRR by around 0.5 percent.

114. Table 6.42 shows the sensitivity table and Table 6.43 the risk analysis results.

Table 6.42 Sensitivity results for tranches 1 and 2, 30 percent RV

Scenario EIRR percent NPV in USD Switching million factora (a) Base case 14.3% 45.0 (b) Benefits less 10% 13.3 23.9 0.79 (c) Benefits less 20% 12.2 2.9 (d) Cost increased by 10% 13.4 28.4 1.27 (e) Cost increased by 20% 12.5 11.9 (f) Combination (c) and (e) 12.4 7.4 (g) Traffic Growth -10% 12.9 15.5 0.84a (h) Traffic Growth -20% 11.0 -10.3 Note: (a) traffic growth for an EIRR of 12% is -16% times the base case

Engconsult Ltd. 43

Figure 6-7 Frequency distribution of EIRR (tranches 1 and 2, 30% RV)

12.0%

10.0%

8.0%

6.0%

4.0% Percent falling in class interval class in Percent falling

2.0%

0.0% 8% 10% 12% 13% 15% 16% 18% EIRR

Engconsult Ltd. FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 7 Institutional and Financial Assessment and Financial Evaluation of the Project

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

i

Table of Contents

I. INTRODUCTION 1 II. ROADS DEPARTMENT INSTITUTIONAL ANALYSIS 2 A. Institutional arrangements 2 C. Road maintenance 3 D. Capacity building requirements 4 III. ROADS DEPARTMENT FUNDING 6 A. Road fund 6 B. Transport taxes 6 C. Roads Department expenditure 6 IV. FINANCIAL MANAGEMENT CAPACITY OF THE TRRC 10 A. Assessment 10 V. FINANCIAL MANAGEMENT CAPACITY OF ROADS DEPARTMENT 13 VI. PROJECT MANAGEMENT ARRANGEMENTS FOR ADB PROJECTS 14 A. Structure of PMU 14 B. PMU costs 15 1. Financial management and procurement staff 15 2. External expert support for the PMU 16 3. TA costs 16 VII. TRANSPORT SECTOR PPP AND TOLL ROADS IN GEORGIA 17 A. Transport sector PPP 17 1. Introduction 17 2. Roads sector 18 B. Rikoti tunnel tolls 18 C. Prospects for tolling the Ajara bypass 19 D. Proposed PPP TA 19 1. Rationale 19 2. Objective 20 3. Activities 20 4. Cost estimate 21 VIII. FINANCIAL EVALUATION OF THE AJARA BYPASS 23 A. Introduction 23 B. Investment cost and financing plan 23 1. Tranche 1: Kobuleti 23 2. Tranches 1 and 2 25 C. O&M and interest expenses 26 D. Tolls 27 1. Collection systems 27 2. Vehicle classes 27 3. Toll levels 27 E. Financial evaluation 28 1. Approach 28 2. Evaluation of tranche 1: Kobuleti 29 3. Evaluation of tranches 1 and 2 29 F. Incremental maintenance expenses 30

Annex A TRRC’s Response to Financial Management Questionnaire

Annex B RD’s Response to Financial Management Questionnaire

Engconsult Ltd. ii

List of Tables

Table 1: Road fund income 6 Table 2: Roads Department expenditure and financing, 2006-9 7 Table 3: Roads Department budgeted expenditure, 2009 8 Table 4: Roads Department expenditure, 1996-2009 8 Table 5: Pipeline of trunk road projects 9 Table 6: TRRC documentation 10 Table 7: TRRC’s 2009 budget 11 Table 8: TRRC project status 12 Figure 3: RD’s finance department 13 Figure 4: Proposed project management arrangements 14 Figure 5: Proposed funds flow arrangements 15 Table 9: PMU financial management staffing costs 16 Table 10: External support for the PMU 16 Table 11: TA costs 16 Table 12: Transport sector PPP projects in Georgia 17 Table 13: Rikoti tunnel tolls and exemptions 18 Table 14: PPP TA activities 20 Table 15: PPP TA cost estimate 21 Table 16: Rikoti tunnel tolled traffic records 22 Table 17: Tranche 1 investment cost by financier 23 Table 18: Financing plan, tranche 1 24 Table 19: Weighted average cost of capital 25 Table 20: Bypass investment cost by financier 25 Table 21: Financing plan, tranches 1 and 2 26 Table 22: Vehicle classes for tolling 27 Table 23: Fineval assumptions 29 Table 24: Tranche 1 evaluation results 29 Table 25: Tranches 1 and 2 evaluation results 30

List of Figures

Figure 1: Roads Department: institutional context ...... 3 Figure 2: Roads Department organization ...... 5 Figure 3: RD’s finance department ...... 13 Figure 4: Proposed project management arrangements ...... 14 Figure 5: Proposed funds flow arrangements ...... 15

Engconsult Ltd. 1

I. INTRODUCTION

1. The purpose of this supplementary appendix is (i) to review the institutional and financial structure of the implementing agency (the Roads Department), (ii) to report on the financial management capacity of the department and of the TRRC, (iii) to present the legislative context and experience of tolling in Georgia and (iv) to undertake a financial evaluation of the proposed bypass.

Engconsult Ltd. 2

II. ROADS DEPARTMENT INSTITUTIONAL ANALYSIS

A. Institutional arrangements 2. Institutional arrangements changed in February 2009. Up to February 2009 the Roads Department (RD) was part of the Ministry of Economic Development (MED). In February 2009 the MED was reorganized and a new ministry created, the Ministry of Regional Development and Infrastructure (MRDI) and RD and the United Transport Administration (UTA) were moved from the MED to the MRDI. The Transport Department, a policy unit, remains with the MED. A body related to RD is the Transport Reform and Rehabilitation Center (TRRC). 3. The functions of these roads sub-sector bodies are as follows:

• the Transport Department is responsible for strategy and policy (and covers the other transport modes with the exception of rail)

• RD is responsible for implementing road construction, maintenance and rehabilitation of international and interstate roads (but not local or urban roads)

• the UTA is principally responsible for the international aspects of road haulage, aviation and maritime transport

• the TRRC is the project implementation unit (PIU) set up to implement World Bank road projects. It is physically located within the Roads Department but its staff are not government employees 4. In addition to the above there are several organisations that have looser ties to the Roads Department but are nonetheless of importance. Principal amongst the latter is the Millennium Challenge Fund Georgia (MCG), responsible for the Samtskhe-Javakheti road project. 5. Figure 1 shows RD’s institutional context.

B. Roads Department organization

6. Following restructuring several years ago, RD is now a small, fairly non-hierarchical organisation with 160 staff members (down from 195 in 2007), mostly based in Tbilisi. See Figure 2. Although its parent ministry changed earlier in 2009 its internal structure appears to have changed little when compared with that reported by the JBIC Study for Project Formulation for Highway Improvement Project, 2008. Officials at the department have however indicated that its structure may still change following some changes at the top of the organisation. The department has no planning section (this being the remit of the Transport Department) and no internal financial audit section. 7. There are five small regional offices: in Sagarejo, responsible for the regions Kakheti, Kvemo Kartli, and Mtkheta Mtianeti; in Borjomi for the regions of Shida Kartli and Samcke Javakheti; in Kutaisi for the regions Racha Lechkhum, Kvemo Svaneti and Imereti; in Poti for the regions Guria and Samegrelo Zemo Svaneti; and in the Autonomous Republic of Ajara. No formal organizational structure has yet been established for local roads; most local governments lack technical staff, equipment and a secure source of funding for road maintenance and rehabilitation. Municipalities are responsible for urban roads.

Engconsult Ltd. 3

Ministry of Economic Ministry of Regional Development (MED) Development and Infrastructure (MRDI)

Transport Department Roads Department TRRC UTA • roads strategy & • road construction, • PIU for World Bank • international policy maintenance etc projects haulage agreements etc

Figure 1: Roads Department: institutional context

C. Road maintenance 8. The department has no direct labor force: it contracts out all its maintenance and construction activities. 9. The country is divided into 34 routine maintenance zones, each of which covers 150- 250km of roads. (Zone 33 includes the S-2 in Ajara plus some other roads). Contractors bid against a bill of quantities drawn up by the department. Maintenance requirements are established from an annual condition survey. The last round of bidding took place in 2007 and was for a period of 27 months. These contracts will expire in 2009 and will be re-bid for a period of 12 months. Supervision (i.e. checking both quality and the quantities of work undertaken) was also contracted out to a private company. This function has now been brought back into the department; the department has 20 people who will act as supervisors. 10. Contractors are quite small: each zone has a separate contractor. They provide their own plant, labor and materials, although some hire plant and vehicles from a government agency (not the Roads Department). 11. The World Bank is supporting the RD in its efforts to improve the efficiency of its road maintenance activities. It believes that a shift from contracted out maintenance to performance based road maintenance contracts (PBCs) could produce further savings (up to 50 percent). A proposed project component of the TEWHP will assist in developing a comprehensive framework for the introduction of PBC in Georgia. The international consultant, in collaboration with a small RD team, will apply the PBC framework to a preliminary assessment of a concession for the maintenance of the East-West Highway and an area-wide contract for rehabilitation and maintenance for 250-300km of main and secondary roads centered around a regional RD Office. 12. Maintenance of those sections of main roads that pass through municipalities is the responsibility of the municipality. (In the case of the S-2 in Ajara this exclusion only applies to 8km within Batumi). 13. Periodic maintenance contracts are also awarded on a bill of quantities basis.

Engconsult Ltd. 4

D. Capacity building requirements 14. Roads Department discharges its responsibilities – technical and financial – satisfactorily. However, despite some progress on policy and institutional reforms, carried through with considerable World Bank support, institutional capacity in the road subsector remains limited. It does for example face an ever increasing works program (see Supplementary Appendix 1: RD expenditure is expected to quadruple 2008-10) and this may raise quality issues: there is limited staff experience in the management of roads projects, especially in the areas of bridge and tunnel engineering and safeguards. The proposed tranche 1 and tranche 2 projects contain particularly challenging bridge and tunnel works, and resettlement issues, and it is therefore proposed to strengthen RD’s capacity in these areas (see below). 15. It lacks PPP capacity (see section VII). The other area where capacity building is proposed is public-private partnerships (PPPs). As discussed in section VII, Georgia’s record in the roads sector is limited. It is therefore proposed to strengthen its PPP capacity as part of tranche 1 activities (see below). 16. The country’s road safety record is poor. The problems include poor interagency co- operation, inappropriate and unenforced speed limits, ubiquitous use of mobile phones while driving and a complete lack of regard for pedestrians. The MED is developing an action plan for road safety. A road safety TA is also proposed as part of this project: see Appendix 4.

Engconsult Ltd.

Figure 2: Roads Department organization

Engconsult Ltd. MRDI

Chairman

First deputy chairman

Financial management Legal department Administration • accounts • legal drafting (road law • HR • budgeting amendment etc) • office management • tolls • RD legal representation • legal compliance checks

Deputy chairman

Technical policy Rehabilitation • design • planning & commissioning • road standards of rehabilitation • road condition survey • R&D

Deputy chairman Deputy chairman Regional offices: Ajara Development and resettlement Sagarejo Periodic maintenance Roads operations • annual construction plan Borjomi • planning and • co-ordination with local • resettlement planning Kutaisi commissioning of routine government • bridge condition survey Poti maintenance • police liaison (blackspots,

• reports on emergencies diversions) • signing 5

6

III. ROADS DEPARTMENT FUNDING

A. Road fund 17. From 1995 to 2004 Georgia had a road fund. In 2003 its income was GEL40m. It was financed by a portion of fuel tax (30 percent of the fund’s 2003 income), the proceeds of Rikoti tunnel tolls, the international transit tax and a road usage tax levied on individuals and enterprises. 18. According to a World Bank project appraisal report (Proposed Credit for a Secondary and Local Roads Project, May 2004) the fund did not perform particularly well. It terms of the use of funds, the Bank concluded that routine maintenance and rehabilitation were the principal beneficiaries; little was spent on periodic maintenance. The fund’s income was prone to fluctuation and decline (see Table 1) and leakage. The Bank also commented on its inadequacy – the Bank’s 2004 estimate of annual routine and periodic maintenance requirements was GEL172m.

Table 1: Road fund income 1999 2000 2003 GEL million Fuel tax & Rikoti toll 14 4.6 12 revenue Vehicle license fees 5.6 N/A 2.0 Transit fees 10.7 N/A 9.6 Source: World Bank project appraisal report (Proposed Credit for a Secondary and Local Roads Project, May 2004) Note: table does not show all sources of income 19. The World Bank considered that the prospects for increasing road fund income were promising, provided road user charges were raised to regional norms and collection made more effective. In 2004, however, the government abolished special funds, including the road fund, and since then the sub-sector has been funded directly from the state budget and from donor funds.

B. Transport taxes 20. Along with the special funds, most transport related taxes were also abolished in 2004. The international transit tax and road usage fees were abolished. Vehicles are now subject to a small registration fee (of GEL56) on change of ownership, collected by the police, and new vehicles are subject to import duty and VAT, but there is no longer an annual fee. Fuel tax plus VAT amounts to 32 percent. The proceeds of the remaining transport taxes, as well as toll receipts from the Rikoti tunnel, are now part of state revenue.

C. Roads Department expenditure 21. Table 2 shows RD expenditure from 2006-9. Expenditure shown in this table excludes local and urban roads, which ceased to be a RD responsibility in 2006. There is no separate contribution from AR Ajara.

Engconsult Ltd. 7

Table 2: Roads Department expenditure and financing, 2006-9 2006a 2007a 2008a 2009b GEL million Staff expenses 1.2 1.2 2.0 2.2 New road construction state budget N/A N/A 13.7 13.7 TRRC N/A N/A 53.7 205.7 Subtotal 52.6 74.0 67.4 219.4 Rehabilitation state budget N/A N/A 122.5 121.5 TRRC N/A N/A 5.3 80.0 MCGc 0.0 0.0 34.5 50.2 Subtotal 93.9 165.4 162.3 251.7 Routine maintenance 22.0 25.5 25.5 25.8 Miscellaneous (including bridge 29.8 10.8 15.2 21.4 works) Grand total 199.5 276.9 272.4 521 Financed by: donor and counterpartd funds - 63 102 339 state budget - 214 170 182 Source: Roads Department and Transport Department Note: (a) actual expenditure for fiscal year – 1 Jan to 31 Dec. All figures include VAT where applicable (b) budget forecast (c) the MCG compact became effective April 2006 but MCG accounts show no roads expenditure until 2008. The only MCG roads project is rehabilitation of the Samtskhe-Javakheti road (d) counterpart funds are from state budget sources 22. The striking feature of Table 1 is the sub-sector’s dependence on donor funding. Of the forecast GEL521m expenditure in 2009, GEL336m is either from donor sources or represents counterpart funds. 23. A more detailed breakdown of 2009’s budgeted expenditure is given below. Unfortunately the budget does not distinguish between rehabilitation and maintenance – and nor do the statements of actual expenditure in 2007-8. Similarly there is no breakdown by class of road. Routine maintenance in the budget amounts to GEL26m, or about GEL3,800 (USD2,300) per km. This is consistent with the department’s approach to budgeting, which is as follows:

• routine maintenance for main roads is budgeted at GEL4,000/km (including bridges) • routine maintenance for secondary roads is budgeted at GEL2,600/km (including bridges) • the repair of winter damage is budgeted separately and accounts for approximately GEL5m in a typical year (All costs include VAT). 24. The department’s budget is lower than conventionally expected: for example, using the World Bank’s ROCKS database, a network with Georgia’s characteristics would be expected to need approximately USD4,000 (GEL7,000) per km. 25. GEL119m from state funds is budgeted for periodic maintenance and rehabilitation. This understates total rehabilitation expenditure, however, as both the World Bank’s secondary roads and the MCG’s Samtskhe-Javakheti project are rehabilitation projects. If these are included, funds for rehabilitation and periodic maintenance amount to GEL150m, or GEL22,000 (USD13,000) per km. For a network of 6,900km in a lower-middle-income country this appears

Engconsult Ltd. 8 reasonable, i.e. should stabilise road condition within a reasonable horizon. Support for this comes from Roads Department figures showing that 1,722km of roads have been rehabilitated or received periodic maintenance over the period 2004-8, equivalent to about 5 percent of the network per year.

Table 3: Roads Department budgeted expenditure, 2009 GEL million Total 521 State funds 284.2 284.2 Staff expenses 2.2 Other admin expenses 1.7 Periodic maintenance and rehabilitation 119.1 Routine maintenance 25.8 Minor new construction 4.0 Bridge works 9.7 Miscellaneous 19.5 State counterpart funds 102.2 102.2 East-West Highway (WB) 61.8 Secondary Roads (WB) 24.0 MCG (Samtskhe-Javakheti)a 11.1 Tbilisi-Telavi (WB) 4.5 BTC etc 0.8 Grants and credits 237.1 237.1 East-West Highway (WB) 128.6 Secondary Roads (WB) 56.0 MCG (Samtskhe-Javakheti) 39.2 Tbilisi-Telavi (WB) 10.5 BTC etcb 2.8 Source: Roads Department 2009 budget (there are minor differences between this table and Table 1) Note: (a) counterpart funding for the MCG project amounts to payment of VAT (b) includes roads funded by BP as part of the BTC (Baku-Tbilisi-Ceyhan) pipeline project

26. The position in 2008 was similar: of GEL272m, GEL94m were from donor funds. Allowing that a separate Transport Department source shows GEL102m from external sources leaves around GEL175m from the state budget (of which some will be earmarked as counterpart funds). The same Transport Department source shows GEL63m in 2007 from external sources and GEL214m from the state budget. 27. Table 4 shows that In both nominal and real GEL terms road expenditure has risen dramatically since 1996, principally as a result of access to donor funds.

Table 4: Roads Department expenditure, 1996-2009 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Current GELm 31 30 32 35 32 61 88 100 95 141 200 277 272 521 Const 2009 91 71 72 71 58 106 146 158 142 197 258 327 292 521 GELm Sources: Transport Department to 2005; Roads Department thereafter 28. RD does not forecast its expenditure beyond 2009. Table 4 shows expected project starts in 2009-11 and was provided by the Transport Department.

Engconsult Ltd. 9

Table 5: Pipeline of trunk road projects Road Donor Length km Start year Est cost, USD million Sveneti-Rikoti (E-60): Sveneti-Ruisi 15 2009 127 Ruisi-Agara 12 2010-11 46 World Bank Agara-Rikoti 33 2010-11 116 Rikoti tunnel 4 2009 rehabilitation -Samtredia JICA 77 2009-10 Vaziani-Gombori-Telavi State budget 64 2009 Tbilisi- State Budget10 2010 Ajara bypass ADB 45 2010 200 Akhalkalaki-Turkey border State budget 36 2009-11 18 Sources: Transport Department; this project

Engconsult Ltd. 10

IV. FINANCIAL MANAGEMENT CAPACITY OF THE TRRC

A. Assessment 29. The TRRC is the project implementation unit (PIU) for World Bank projects in the transport sector. Its functions, as set out in a typical World Bank loan agreement, are “to carry out [at the request of Georgia – the borrower] the financial management for the project and assist the Roads Department of the Ministry of Regional Development and Infrastructure in the procurement under the project in accordance with the implementation agreement”. In practice it advises the Roads Department on procurement and is responsible for documenting the procurement process. It is then responsible for financial management during contract implementation. 30. At the consultants’ request the TRRC completed ADB’s financial management assessment questionnaire. The completed questionnaire is at Annex A to this supplementary appendix. 31. The TRRC also supplied relevant documentation as listed below. In addition to the documents listed in Table 6 the consultants obtained a World Bank financial supervision report dated January 2009.

Table 6: TRRC documentation No Title Author Date 1 Operations manual for First East-West Highway Improvement Dec 2005 Project 2 Financial statements for quarter ending 31.12.08 for grant H204GE 3 Funds flow diagram Undated 4 Correspondence between World Bank and MoF re IBRD loan Undated 7671-GE 5 Loan agreement for Secondary and Local Roads Additional Financing 6 Interim financial report (IFR) for Second East-West Highway Project to 31.12.08 7 IFR for First East-West Highway Project to 31.12.08 8 Operations manual for the Infrastructure Pre-Investment Facility Dec 2005 project 9 Operations manual for the Second East-West Highway project Dec 2007 10 Financial and accounting management manual TRRC Jan 2009 11 Operations manual for the Secondary and Local Roads project Feb 2009 12 Terms of reference for the financial accounting department TRRC 13 Terms of reference for the audit of project financial statements of… 14 Charts of accounts (three sets) TRRC 15 CV of financial manager, TRRC 16 CV of finance assistant, TRRC 17 Audited accounts of First East-West Highway Improvement project Deloitte & June 2008 to 31.12.07 Touche 18 Management letter related to doc no 17 Deloitte & June 2008 Touche 19 Minutes of negotiation between Georgia and World Bank for World Feb 2008 additional financing of Secondary and Local Roads project Bank Source: TRRC Note: all the above are in English

Engconsult Ltd. 11

32. The TRRC was created by presidential decree no 119 of 16 April 1995. It was originally known as the “Europe-Asia Transport Corridor Problem Center”. The name was changed to “investment center” by decree no 161 of 21 April 2000. Strictly, the name TRRC only applies to its role as the World Bank PIU, but as this is its only role at present, the name TRRC will be used here. It has a staff of twelve. Although physically located in the Roads Department its staff are not departmental employees. It is understood that the government has for some time been anxious to merge the TRRC with the Roads Department. While this view was not shared by TRRC staff whom the consultants met, the World Bank’s stance appears to be fairly neutral. 33. The TRRC is headed by a director. Including the director, six professional staff members are trained in financial management and accounting; the remainder are procurement specialists. The chief professional officer has been with the TRRC since it was founded. The World Bank’s project document for the Second East-West Highway Project, however, commented in late 2007 on high staff turnover. This was taken up with the director, who stated that there had been no resignations in the past five years. The chief accountant has been in post since the TRRC started and the current director was appointed in 1997. 34. The TRRC’s operating expenses were originally met from the First East-West Highway project. With the completion of this project, USD0.72m was included in the Second East-West Highway project to cover TRRC’s operating expenses, hiring of consultants and project audit fees for 3.5 years. TRRC’s compensation packages are subject to monitoring by the Roads Department. The 2009 budget is summarized in Table 7.

Table 7: TRRC’s 2009 budget Item GEL Salaries (including social costs) 191,810 Utilities, postage, travel, fuel, spare parts etc 45,960 Translation, office equipment maintenance etc 15,660 Total 253,430 Financed by: IDA (70 percent) 177,400 Government of Georgia (30 percent) 76,030 Source: TRRC 35. The TRRC uses a modified cash basis for accounting and runs ORIS software. The World Bank (in its January 2009 financial management supervision report) considers the software satisfactory for accounting and reporting purposes. Excel is used to finalize reports submitted to the World Bank. There is a written financial and accounting management manual, most recently updated in January 2009 (document 10 in Table 6). 36. Audits of project accounts have always been “clean” (i.e. issued without qualification) as far as the consultants can ascertain from information supplied by the TRRC and the World Bank. Management letters issued by the TRRC’s auditors, Deloitte & Touche have (in document no 18 in Table 5) pointed to deficiencies in (i) data back-up, (ii) reconciliation of expenditure financed by the government with the Treasury Department and (iii) documentation of tender files. The World Bank’s supervision report confirms that auditors’ recommendations concerning (i) and (ii) have been implemented. The action plan for the TRRC in the same report covers (i) timely submission of auditors’ reports, (ii) timely submission of financial management reports (FMRs) and interim financial reports (IFRs) and (iii) implementation of the auditors’ recommendation regarding project documentation. 37. The TRRC has a substantial current workload, however, listed in Table 8, and has (of course) no experience of handling ADB projects.

Engconsult Ltd. 12

Table 8: TRRC project status Project Status as at Feb 2009 First East-West Highway Improvement Project Almost completely disbursed Second East-West Highway Improvement Project Approx 50% of SDR22.4m disbursed Infrastructure Pre-Investment Facility Approx 75% of SDR3.5m disbursed Secondary and Local Roads Project Almost completely disbursed Second East-West Highway Improvement Project – No disbursements yet of SDR12.8m additional financing Secondary and Local Roads Project – additional No disbursements yet of USD70m financing Source: World Bank and TRRC

Engconsult Ltd. 13

V. FINANCIAL MANAGEMENT CAPACITY OF ROADS DEPARTMENT

38. At the consultants’ request RD staffs were able to complete some of ADB’s financial management assessment questionnaire. Completed sections are shown in Annex B to this supplementary appendix. 39. The structure of RD’s finance department is shown in Figure 3. Excluding staff working on Rikoti tolls, it has 16 members. The financial management and analysis section appears to be responsible for budget planning. Staffing is said to be adequate for the tasks in hand. There are no unfilled posts and turnover is said to be low (no resignations in the past 18 months).

Financial management department (3 managers)

Accounts (6 staff) Financial management Toll section (14 staff) & analysis (7 staff)

Figure 3: RD’s finance department 40. The finance department has no experience of externally funded projects: both foreign and counterpart funds for such projects are managed by the TRRC (counterpart funding is transferred directly from the Ministry of Finance to contractors and does not pass through RD). It does not operate commercial bank accounts; its funds are held in a Ministry of Finance account at the National Bank of Georgia. RD is funded quarterly in advance. 41. The department runs ORIS accounting software (as do the TRRC and other Georgian government departments) and claims to produce accruals based accounts. 42. The ability to control project costs appears limited. Activities are coded by purpose and geography rather than contract. Thus, expenses associated with a road whose maintenance was contracted out to several companies would appear against a single code. 43. RD is subject to occasional scrutiny by Georgia’s supreme audit institution, the Chamber of Control. The last such scrutiny took place in 2008. RD is also subject to external audit, auditors being appointed annually by competitive tender. RD has no internal audit department. 44. In summary, RD’s financial management capacity appears adequate for its current tasks but would need considerable strengthening if it were to handle externally financed projects.

Engconsult Ltd. 14

VI. PROJECT MANAGEMENT ARRANGEMENTS FOR ADB PROJECTS

A. Structure of PMU 45. The TRRC was set up as a temporary organization to undertake the financial management of World Bank road projects. Although it may in future become part of RD, it remains for the moment a separate body. Financial management of ADB projects should strengthen the long term capacity of RD, not the TRRC. It is therefore proposed that the ADB PMU is formed within RD, with procurement and financial management support from TRRC. 46. The proposed arrangement envisages the MRDI as executing agency (EA) and RD as implementing agency (IA). The chairman of RD will be responsible for overall supervision of the program. RD will establish a PMU headed by the deputy chairman of RD. It will comprise technical staff seconded from RD and TRRC. A steering committee will be established with representatives of the cabinet office, MRDI, RD, Transport Department (of behalf of MED), the Ministry of Finance and the PMU head. The steering committee will be chaired by a deputy minister of MRDI and will meet at least every six months to ensure interagency coordination, review implementation progress, and provide approvals and guidance as necessary. Figure 4 illustrates the proposed set-up.

Steering committee (Cabinet office, MRDI, RD, MED, MoF, PMU head)

MRDI (EA)

Roads Department (IA)

TRRC: PMU Detailed design i) financial management ii) procurement Supervision

External experts: i) tunnel & bridge engineering Capacity building TAs ii) safeguards

Road safety TA

Contract 1 Contract 2 etc

Figure 4: Proposed project management arrangements

Engconsult Ltd. 15

47. The PMU will handle day-to-day implementation activities. It will also undertake periodic functions such as (i) preparation of PFRs, (ii) engagement and supervision of consultants and contractors, (iii) preparation of reports and (iv) obtainment of approvals from ADB and the government. The PMU will report to the chairman of RD. The PMU will be supported by a team of externally contracted qualified technical, safeguards, financial, and support staff to supplement resources in RD. 48. Proposed funds flow arrangements are shown in Figure 5.

ADB Treasury

Withdrawal application Application for plus certified counterpart fund claim/invoice & release RD/PMU approval

Payment advice RD/PMU approval Payment advice Payment Payment Payment

Certification by project engineer

Contractor’s or consultant’s claim or invoice

Figure 5: Proposed funds flow arrangements

B. PMU costs

1. Financial management and procurement staff 49. In order to undertake procurement and financial management in accordance with ADB procedures, the TRRC would need to recruit and train additional staff. The PMU would also have to hire support staff and budget for financial audit. Based on the TRRC’s 2009 budget the cost for six additional professional staff, three support staff, additional overheads and training for the professionals in the first year is shown in Table 9. The four year (2010-13) total implied by Table 9 is USD0.60m.

Engconsult Ltd. 16

Table 9: PMU financial management staffing costs Item USD 6 x professional staff at GEL20,000 (USD12,000)/yr 72,000 3 x support staff at GEL10,000 (USD6,000)/yr 18,000 Office etc overheads at 30% x staff costs 27,000 Audit fees (say) 20,000 Annual total 137,000 Initial training (say) 50,000 First year total 187,000 Source: consultants’ estimates 2. External expert support for the PMU 50. In addition to financial management, the PMU will require external expert support in the following areas:

• construction • safeguards 51. Construction experts will be required to monitor construction implementation and supervision. Their task will be to ensure that implementation complies with the detailed design and meets Georgian and international standards as appropriate. It is envisaged that intermittent inputs by international bridge and tunnel engineers would be required. 52. The role of the environmental and resettlement experts will be to ensure compliance by the independent consultants hired by the IA.

Table 10: External support for the PMU Item Total USD over 4 years Bridge engineer, 1 month/year, at USD25,000/mon 100,000 (international) Tunnel engineer, 1 month/year, at USD25,000/mon 100,000 (international) Environmentalist, 6 months/year, at USD2,500/mon 300,000 (domestic) Resettlement expert, 6 months/year, at USD2,500/mon 300,000 (domestic) Total 800,000 Source: consultants’ estimates 3. TA costs 53. Table 11 lists the proposed TAs and their estimated costs.

Table 11: TA costs TA Total USD Road safety (see Supplementary Appendix 4 and Appendix 725,000 13) Capacity building TAs: Transport strategy study (Appendix 13) 600,000 Training and capacity development (Appendix 13) 750,000 PPP TA (this supplementary appendix) 454,000 Total 2,529,000 Source: consultants’ estimates

Engconsult Ltd. 17

VII. TRANSPORT SECTOR PPP AND TOLL ROADS IN GEORGIA

A. Transport sector PPP

1. Introduction 54. Public-private partnerships in the transport sector have made relatively little headway in Georgia; there are none in the roads sector. Table 12 lists some transport sector examples.

Table 12: Transport sector PPP projects in Georgia Project Sector Details Tbilisi airport, 2008 Aviation The project company (TAV Georgia) has an 11.5 yr lease on the airport from 2006 with investment responsibility for a new terminal, taxiways and runway widening (est cost $76.5m). The project company is owned by Turkish shareholders. IFC and EBRD have loan participation. Batumi airport Aviation 20 year lease with investment responsibility for new terminal (now complete). $28.5m payment to government. Project company controlled by TAV Airports Holding Co. Batumi container Maritime 48 year concession held by ICTSI of the Philippines terminal Batumi port Maritime Management contract held by KazTransGas; GoG is contracting party. Poti port oil terminal Maritime Construction and operation of a new oil terminal. Project company is Channel Energy (Poti) Ltd, a JV company formed in 2000 between Channel Energy Ltd (Eire) and Poti Sea Port (Georgia). EBRD loan participation up to $11.6m in parallel with $8m from the Black Sea Trade Development Bank. Source: EBRD and World Bank PPI websites; local knowledge

55. Some of the reasons for the slow progress are to be found in EBRD’s annual assessments of concession laws. Its 2008 assessment (Concession Laws Assessment 2007/8) rated Georgia’s legal environment for private sector development as “low compliance” with international best practice. It noted the following:

• the scope of application (of the relevant law) is unclear • there are few provisions for the selection of concessionaires • the contracting authority is not clearly defined

56. The EBRD did however note some positive features, for example concessionaires’ right to bring claims against public agencies for abuse of power.

Engconsult Ltd. 18

2. Roads sector 57. The Road Law of 1994 permits the imposition of tolls on an existing road and allows the creation of a private, tolled facility. The only conditions are that:

• there must be an alternative, untolled route that is open to traffic • the proceeds of the toll must be used to maintain and develop the tolled facility 58. There is no explicit prohibition of making profits from tolling – although such a prohibition might be inferred from the second condition. There is no provision for the regulation of tolls. 59. In 2007 the MED launched an expression of interest for a design and build concession for the Natakhtari-Rikoti section of the East-West Highway. The World Bank was in dialogue with the MED over this proposal and its Second East-West Highway project appraisal report states that the Bank and the government agreed that the Bank could withdraw if it were not satisfied with the proposed concession agreement. 60. MED’s discussions with a potential investor in the East-West Highway broke down, or were cancelled, during 2007. Earlier, in 2005, a structural survey of the Rikoti tunnel, preparatory to a pilot PPP transaction, was proposed by the Public Private Infrastructure Advisory Facility (PPIAF), but was eventually cancelled. Thus, despite PPP activities in other sectors, no market for the provision of privately funded roads in Georgia has developed. 61. In view of the lack of progress in roads PPP, a capacity building technical assistance (TA) project is proposed – see section D below.

B. Rikoti tunnel tolls 62. The Rikoti tunnel is 1.75km long, has two lanes in a single bore. It was built in 1983 and is in poor condition. It is on the East-West Highway a few kilometers west of Khashuri. It is in poor condition and has no ventilation. Average tolled traffic through the tunnel is around 5,700 vehicles/day. 63. Tolls are levied by virtue of Roads Department order no 6 of January 1999. This order set the toll rates, which have not been changed since. Order no 6 was amended by order no 74 of 2000, which listed categories of individuals who are exempt from paying the charge. Both orders apply only to the Rikoti tunnel. Tolls are paid in cash and are only payable when the alternative route is open. (The alternative route has a very poor alignment and is frequently closed in winter). Tolls and categories of exempt persons are set out in Table 13. 64. Toll proceeds are not used for tunnel maintenance; they are simply part of RD’s revenue. RD employs 14 people in its toll section.

Table 13: Rikoti tunnel tolls and exemptions Item Tolls 1. Passenger vehicles with up to 30 seats GEL1.00 2. Trucks carrying a load of up to 10t GEL2.00 3. Large vehicles not in category (1) or (2) GEL3.00 Exempt categories Police Traffic police Employees of the prosecution service Employees of the intelligence service 65. Tunnel traffic by toll class is summarised in Table 11. In addition to traffic, the table also shows, for 2008-9, the number of days on which tolls were collected. During 2008 tolls were

Engconsult Ltd. 19 only collected on 75 percent of possible days. Tolls are not collected when the alternative route is closed (usually by snow), but for most of August 2008 the reason was the conflict in South Ossetia. 66. Annual revenue based on 2007’s tolled traffic records would be GEL2.2m. Revenues are strongly seasonal: monthly receipts in December-February are about 60 percent of the annual average.

C. Prospects for tolling the Ajara bypass 67. In section VI the tolls needed to cover defined operating and maintenance expenses are calculated. There remains however the broader question of whether the bypass is suitable for tolling. 68. In favor of tolling are the following:

• as required by law there is an alternative, toll-free route • the bypass saves time rather than distance and so is particularly attractive to those with high values of time. Thus the toll would be progressive in the sense that it would apply to the better off • the Rikoti tunnel shows that tolling of a single facility is feasible 69. However, against tolling must be reckoned:

• Georgia has no experience of awarding such a concession in the transport sector (Rikoti tolls do not cover defined expenses) • a proposal to toll would hinder the fast-track implementation of the project, as would the legal process required to establish toll rates • tolling would reduce the economic internal rate of return of the bypass, both by deterring traffic that would use it if it were free and by adding the cost of toll plazas and toll operations to project costs 70. On balance it is considered that the arguments against tolling outweigh those in favor of it. It is therefore recommended that tolling not be introduced, at least in the short term, but that the possibility be kept under review and re-appraised once demand for the bypass has been clearly established.

D. Proposed PPP TA

1. Rationale 71. The Government of Georgia supports the involvement of the private sector in infrastructure. Evidence of this in the transport sector was the early efforts by the MED to involve the private sector in the East-West Highway. 72. The usual rationale for PPP is that, when facing appropriate incentives, a private company will deliver a more cost-effective service to users than the public sector. In Georgia’s case there is also a transit argument: transit countries may derive little direct economic benefit from the traffic that used its infrastructure. This is undoubtedly so in the case the Ajara roads projects, where it is estimated that some 42 percent of benefits go to transit traffic (see Supplementary Appendix 6). Similar cases will arise from other bypass projects (e.g. those proposed for the EW Highway). If a way can be found to recoup some infrastructure costs from foreign beneficiaries without jeopardising the project’s economic return, then Georgia will

Engconsult Ltd. 20 benefit. It can be seen as a way of capturing the negative externalities of transit traffic as a financial cash flow. 73. RD, with World Bank support, is proposing to shift from contracted out maintenance to performance based road maintenance contracts (PBCs) that could produce further savings (of up to 50 percent). A proposed project component of the Third East-West Highway Project will assist in developing a comprehensive framework for the introduction of PBC in Georgia. This will give RD valuable experience in the design of appropriate incentives. 74. The purpose of this TA is to move RD a step further than PBCs and prepare them for deeper involvement by the private sector in road provision.

2. Objective 75. The objective will be to put RD in a position where it can invite bids for PPP contracts without having to embark on time consuming preparatory studies. It should also complement and improve delivery of PBC contracts. 76. At the end of the TA it should be able to (i) identify and if possible implement any legal changes that will need to be made, (ii) identify those types of asset that are best suited to PPP (bridges, tunnels, bypasses etc), (iii) devise an appropriate payment vehicle (toll, shadow toll, performance payments etc), (iv) devise suitable standard forms of contract and (v) obtain clear guidance on the forms of guarantee, tax break etc that may be available from government.

3. Activities 77. Table 14 lists the principal activities.

Table 14: PPP TA activities Activities Legal: review existing legislation and draft amendments as necessary review existing contracts and identify lessons learned Economic/financia l: identify assets for private sector involvement recommend payment vehicle at an outline level, undertake financial evaluations undertake a review of the market for finance for infrastructure in Georgia (donors, commercial banks etc) and the guarantees that may be available Contractual: devise appropriate standard forms of contract devise suitable procurement strategies Other: undertake a market review amongst potential bidders (attitude to Georgia country risk etc)

Engconsult Ltd. 21

4. Cost estimate

Table 15: PPP TA cost estimate Individual Rate, USD/mon Months Total USD International: Team leader/PPP expert (international) 30,000 4 120,000 Legal expert (international) 50,000 2 100,000 Economist/financial analyst 25,000 3 75,000 Contracts expert 25,000 3 75,000 Domestic: Deputy team leader/PSP expert 4,000 4 16,000 Lawyer 10,000 2 20,000 Economist/financial analyst 3,000 3 9,000 Contracts expert 3,000 3 9,000 Support staff, translators accommodation LS 30,000 etc Total 454,000 Source: consultants’ estimates

Engconsult Ltd.

Table 16: Rikoti tunnel tolled traffic records Year Toll class Vehicles tolled in month by toll category Total Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 2006 1 187,687 134,022 111,901 96,429 81,024 611,063 2 9,400 9,954 7,968 7,196 8,406 42,924 Engconsult Ltd. Engconsult Ltd. 3 16,154 15,234 14,544 12,208 13,282 71,422 Total 213,241 159,210 134,413 115,833 102,712 725,409 2007 1 39,484 68,916 86,307 119,670 109,540 115,334 160,569 226,903 161,857 143,009 106,128 44,289 1,384,014 2 3,028 5,994 7,666 7,492 8,637 8,756 10,612 11,497 12,146 11,924 8,849 5,655 102,258 3 6,807 13,226 16,749 17,091 19,489 19,577 20,142 21,222 22,282 23,452 19,035 9,188 208,263 Total 49,319 88,136 110,722 144,253 137,666 143,667 191,323 259,622 196,285 178,835 134,012 59,132 1,692,972 2008 1 32,208 44,778 106,270 29,992 135,997 123,722 179,558 52,877 149,498 140,087 145,787 111,027 1,253,810 2 2,676 4,176 8,396 2,556 10,231 8,461 10,972 2,685 9,898 10,047 10,042 9,160 89,302 3 6,374 9,575 19,519 5,170 24,095 21,552 24,667 5,802 25,386 25,127 26,515 19,458 213,243 Total 41,258 58,529 134,185 37,718 170,323 153,735 215,197 61,364 184,782 175,261 182,344 139,645 1,554,341 Daysa 10 12 28 9 31 29 31 7 30 31 30 26 274 2009 1 57,389 115,695 112,888 287,982 2 3,519 5,787 5,603 14,911 3 9,800 18,279 18,813 46,895 Total 70,708 139,761 137,304 347,773 Daysa 16 27 29 72 Source: Roads Department Note: (a) number of days in month during which tolling took place (alternative route usually closed during other days) 22

23

VIII. FINANCIAL EVALUATION OF THE AJARA BYPASS

A. Introduction 78. Although tolls are ruled out in the short term (see section V), they remain a longer term possibility. The basis for the evaluation is that of an operation and maintenance concession: a private company is awarded the right to toll the bypass. From the proceeds of the tolls the company is expected to cover all the operation and maintenance costs (including overlays but excluding reconstruction or widening) for a period of 20 years. ADB interest expenses might also be met from tolls. 79. It is assumed that a conventional, traffic-based toll arrangement is implemented. In the absence of a transit fee, traffic-based tolls provide a way for Georgia to extract some of the producer surplus that Turkish (and other foreign) operators will derive from use of the bypass. On the other hand, it will expose the concessionaire to traffic risk – a risk that he cannot manage. Other arrangements, such as payments for road condition, lane availability and safety performance, might well provide better value for money and should be considered before contracts are drafted. 80. The purpose of this section is twofold:

• to assess the levels of toll that would be required to cover, at most, operation, maintenance and interest expenses • to compare the incremental costs of maintenance with the Roads Department’s budget 81. By law, tolls may only be applied if there is an alternative road that is not tolled. This means that the four lane section of project 1 (the Kobuleti bypass), where both the existing and the bypass alignments are the same, cannot be tolled. Deducting the four lane section leaves a 42km tolled bypass.

B. Investment cost and financing plan

1. Tranche 1: Kobuleti 82. Table 17 shows the investment cost for tranche 1 split between ADB and Georgia.

Table 17: Tranche 1 investment cost by financier Total ADB Georgia USD Amount % of Amount % of cost m USD m cost USD m category category A Investment cost (2009 prices) 1 Civil works Subgrade 16.4 16.4 100.0 0.0 0.0 Pavement 14.8 14.8 100.0 0.0 0.0 Structures 40.8 40.8 100.0 0.0 0.0 Tunnels 5.0 5.0 100.0 0.0 0.0 Slope stabilization 1.4 1.4 100.0 0.0 0.0 Roadside facilities etc 4.6 4.6 100.0 0.0 0.0 Miscellaneous 0.4 0.4 100.0 0.0 0.0 2 Equipment 0.2 0.2 100.0 0.0 0.0 3 Land and 10.1 0.0 0.0 10.1 100.0

Engconsult Ltd. 24

Total ADB Georgia USD Amount % of Amount % of cost m USD m cost USD m category category resettlement 4 Environmental 2.5 2.5 100.0 0.0 0.0 protection 5 Design & supervision 7.8 7.8 100.0 0.0 0.0 6 TAs: Capacity building 1.8 0.0 100.0 0.0 0.0 Road safety 0.7 0.7 100.0 0.0 0.0 7 Taxes and dutiesa 18.5 0.0 0.0 18.5 100.0 Subtotal (A) 125.0 96.5 77.2 28.5 22.8 B Project management PMU 1.4 1.4 100.0 0.0 0.0 Subtotal (B) 1.4 1.4 100.0 0.0 0.0 Total base costs 126.4 99.3 78.5 28.5 22.8 (A+B) C Contingencies: physicalb 12.3 11.2 91.8 1.0 8.2 pricec 20.8 20.1 96.5 0.7 3.5

IDCd 3.5 3.5 100.0 0.0 0.0 Total at current prices 163.0 132.7 81.4 30.3 18.6 Source: consultants’ estimates (see Supplementary Appendix 3 for base costs) Notes: (a) VAT at 18 percent on all costs except land and TA and resettlement and 5 percent duty on all foreign costs (b) 10 percent on all items except PMU and TAs (c) see Supplementary Appendix 1 for domestic and international inflation assumptions (d) interest during construction on ADB loan 83. The financing plan in Table 18 is predicated on counterpart funding from the Georgia state budget only for land and resettlement costs and taxes and duties. The remainder is financed by an ADB Special Funds loan, which carries the following terms:

• interest rate 1 percent per annum during the grace period of eight years, 1.5 percent thereafter • the loan is amortised over 32 years with equal semi-annual repayments of principal • there are no front-end or commitment fees • borrowed funds are denominated in Special Drawing Rights

Table 18: Financing plan, tranche 1 Year ADB, USDm Georgia, USDm Total USDm 2009 3.3 7.0 10.3 2010 17.1 6.0 23.1 2011 41.3 5.0 46.4 2012 41.2 6.8 48.0 2013 29.7 5.5 35.2 Total 132.7 30.3 163.0 Share 81.4% 18.6% 100% Source: consultants’ estimates

Engconsult Ltd. 25

84. The real weighted average cost of capital (WACC) is estimated at 2.3 percent, using the ADB methodology (Table 18). In this table government funds are assumed to have the same opportunity cost as National Bank of Georgia 91 day certificates of deposit (trading with an average 13 percent yield on 30 March 2009). (Sovereign eurobond yields to maturity (in 2013) were 18 percent on the same date).

Table 19: Weighted average cost of capital

Item ADB loan Georgia Total grant A Proportion of capital 0.81 0.19 1.00 B Nominal rate 1.5% 13% C Tax rate 15% 0% D Tax adjusted cost of capital, Bx(1-C) 1.28% 13% E Inflation rate 0.5% 6% F Real cost adjusted for tax, 1.28% 6.6% (1+D)÷(1+E)-1 G Weighted cost of capital = AxF 1.04% 1.23% 2.3% WACC 2.3%

Source: consultants’ estimates 85. The relevance of WACC calculated as above is limited. Commercial investors would price their claims on the basis of asset risk and return, whereas a WACC of 2.3 percent is an imputed price for public sector borrowers. For a commercial investor a necessary condition is that the asset’s long term financial internal rate of return (FIRR) exceeds its WACC.

2. Tranches 1 and 2 86. Table 20 shows the investment cost for tranches 1 and 2.

Table 20: Bypass investment cost by financier Total ADB Georgia USD Amount % of Amount % of cost m USD m cost USD m category category A Investment cost (2009 prices) 1 Civil works Subgrade 25.0 25.0 100.0 0.0 0.0 Pavement 21.5 21.5 100.0 0.0 0.0 Structures 93.7 93.7 100.0 0.0 0.0 Tunnels 26.5 26.5 100.0 0.0 0.0 Slope stabilization 2.5 2.5 100.0 0.0 0.0 Roadside facilities etc 5.9 5.9 100.0 0.0 0.0 Miscellaneous 0.7 0.7 100.0 0.0 0.0 2 Equipment 0.2 0.2 100.0 0.0 0.0 3 Land and 18.4 0.0 0.0 18.4 100.0 resettlement 4 Environmental 3.0 3.0 100.0 0.0 0.0

Engconsult Ltd. 26

Total ADB Georgia USD Amount % of Amount % of cost m USD m cost USD m category category protection 5 Design & supervision 10.8 10.8 100.0 0.0 0.0 6 TAs: Capacity building 1.8 1.8 100.0 0.0 0.0 Road safety 0.7 0.7 100.0 0.0 0.0 7 Taxes and dutiesa 37.2 0.0 0.0 37.2 100.0 Subtotal (A) 246.1 190.6 77.4 55.5 22.6 B Project management PIU 1.4 1.4 100.0 0.0 0.0 Subtotal (B) 1.4 1.4 100.0 0.0 0.0 Total base costs 247.5 192.0 77.6 55.5 22.4 (A+B) C Contingencies: physicalb 24.4 22.5 92.5 1.8 7.5 pricec 46.8 45.1 96.3 1.7 3.7

IDCd 6.1 6.1 100.0 0.0 0.0 Total at current prices 324.8 265.7 81.8 59.1 18.2 Source: consultants’ estimates (see Supplementary Appendix 3 for base costs) Notes: (a) VAT at 18 percent on all costs except land and resettlement and 5 percent duty on all foreign costs (b) 10 percent on all items (c) see Supplementary Appendix 1 for domestic and international inflation assumptions (d) interest during construction on ADB loan 87. The financing plan shown in Table 21 has the same basis as that for tranche 1 in Table 18.

Table 21: Financing plan, tranches 1 and 2 Year ADB, USDm Georgia, USDm Total USDm 2009 3.4 6.9 10.3 2010 17.6 15.7 33.3 2011 72.8 7.6 80.4 2012 87.0 12.8 99.8 2013 84.8 16.0 100.8 Total 265.7 59.1 324.8 Share 81.8% 18.2% 100% Source: consultants’ estimates 88. WACC is also the same as for tranche 1: 2.3 percent.

C. O&M and interest expenses 89. O&M expenses are generated by HDM-4 as part of the economic evaluation (Supplementary Appendix 6). The same values are used here, converted to financial prices by adding VAT. In addition, toll administration and operation expenses are added at USD7,000/km (not assessed rigorously: this is a typical value taken from work in China and Africa).

Engconsult Ltd. 27

D. Tolls

1. Collection systems 90. A variety of toll collection systems is in operation around the world. These are the broad categories:

• plazas on the mainline: drivers pay as they pass through each plaza. Charges reflect distance travelled but mainline queuing reduces bypass travel time savings. Each plaza has to be sized to accommodate the full road capacity • pay on entry: a fixed fee is paid irrespective of distance travelled. Reduces queuing and simple to operate and audit but likely to dissuade those making short trips from using the bypass • pay on exit: as above • closed tolling: collect ticket on entry, pay according to distance travelled on exit. Seen as “fair” but less simple to operate and audit than other systems 91. The assumption here is that closed tolling is operated, as the principal aim of the project is to reduce congestion on the existing roads by encouraging use of the bypass.

2. Vehicle classes 92. Pavement wear is largely a function of axle load and weight based tolling has been adopted in some countries in order to reflect this. It requires use of electronic weighbridges and is comparatively sophisticated. Adopting weight based tolling on this project would be counterproductive if it led to heavy goods vehicles using existing roads (without weighbridges) instead. 93. The adopted alternative is the conventional one of increasing tolls with vehicle size, on the assumption (only roughly true) that the value of a new road is greater for larger vehicles. A typical set of 4 classes is shown in Table 22, together with the ratios of each toll rate to the toll for a passenger car or SUV. The ratios are typical of those adopted in many countries.

Table 22: Vehicle classes for tolling 1 2 3 4 Definition Cars & Minibuses and Buses HGVs and SUVs LGVs and articulated MGVs trucks Ratio to class 1 vehicle toll 1 1.5 2.0 3.0

Source: consultants’ estimates

3. Toll levels 94. Tolls have to strike a balance between achieving a high economic internal rate of return (EIRR) and a high FIRR. The former is achieved at zero toll; the latter is achieved at a (much higher) toll that just persuades the marginal vehicle to divert to the bypass. Experience suggests that a toll at around one third to one half of perceived savings will produce a good compromise, i.e. it will generate revenues close to their maximum without jeopardising an acceptable EIRR. 95. HDM-4 calculates an average journey time saving of about 0.7h, which is worth approximately USD6 for a car carrying three people, implying a toll rate of 4-6US¢ per veh-km

Engconsult Ltd. 28

(GEL0.07-0.10). This in turn implies that the charge for a car should not exceed around 5¢ (GEL0.08) per veh-km, with charges for other vehicles scaled up using Table 18 ratios. The implied charge for a car using the Kobuleti bypass would thus be no more than USD1.40 (GEL2.40), and for the entire bypass USD2.10 (GEL3.50) – 2 to 3 times the Rikoti tunnel charge for a small vehicle. 96. The consultants’ financial evaluation model, Fineval, requires input tolls expressed per PCU-km. For case A traffic on the bypass (see Supplementary Appendix 6), the weighted average toll for all vehicles that is equivalent to a car toll of 5¢ is 5.55¢ (GEL0.093) per PCU/km. 97. Rikoti tunnel tolls have never been increased (nor is there any such provision in existing legislation). This is typical of many jurisdictions: tolls are left unchanged at their initial nominal levels for several years. They may then be increased in order to capture some or all past inflation. Complete indexation is unusual (and is not necessarily appropriate as debt service expenses, often a large part of total costs, are seldom linked to inflation). Scenarios run here are:

• no indexation – the real value of tolls falls from year to year • partial indexation: tolls are indexed every five years. In the meantime they are left unchanged • complete indexation E. Financial evaluation

1. Approach 98. Fineval works on the premise that a road is financed by a mixture of debt, equity and grant, and toll income is used to operate the road, service the debt and distribute any remaining surplus to shareholders. This arrangement is referred to here as a “full” concession. 99. The premise here is different in three respects:

• an O&M concession would not normally imply recovery of initial costs • the concessionaire would include a profit element in his price for undertaking maintenance works (rather than take equity returns) • revenue and expenditure profiles will differ sharply and more so than under a full concession: income rises as traffic rises, but interest expenses fall over time, while maintenance costs tend to be both “lumpy” and uncertain as overlays are required towards the end of the concession period 100. To handle the revenue and expenditure mismatch the operator would probably have to deposit net proceeds in a sinking fund. When in credit the fund earns interest, when in deficit the operator has to borrow to make up the shortfall. Adequate financial performance is signalled by adequacy of the sinking fund, i.e. its ability to meet O&M expenses as they fall due with a nil (or negligible) closing balance at the end of the term. 101. Fineval assumptions are shown in Table 23.

Engconsult Ltd. 29

Table 23: Fineval assumptions Item Assumption O&M expenses From corresponding HDM-4 run + 18% VAT Toll collection efficiency 90% Accounts receivable Nil Accounts payable 1% x operating costs Interest expenses on ADB loan 1.5% x outstanding principal (as Table 12), repaid over 32 years, converted to GEL at 1.67 Contractor’s margin on O&M 20% expenses Nominal GEL deposit interest ratea 8% (sinking fund) Nominal GEL borrowing rate (to 15% cover short term sinking fund deficits)a

Source: consultants’ estimates Note: (a) rates for corporate borrowers in early June 2009; time deposits can attract up to 10 percent, but it is unlikely that a weighted average greater than 8 percent could be achieved in practice 2. Evaluation of tranche 1: Kobuleti 102. Results for tranche 1 are shown in Table 24. The range of tolls is solely the result of alternative approaches to indexation. In terms of FIRR, each scenario produces sensibly the same result: an FIRR of -1.0 percent.

Table 24: Tranche 1 evaluation results Toll scenario Requireda base toll Base charge per km per PCU-km Car Articulated truck No indexation GEL0.049 (2.9¢) GEL0.027 (1.6¢) GEL0.080 (4.8¢) Partial indexation GEL0.034 (2.0¢) GEL0.019 (1.1¢) GEL0.056 (3.3¢) Full indexation GEL0.029 (1.7¢) GEL0.016 (1.0¢) GEL0.047 (2.8¢) Source: consultants’ estimates Note: (a) weighted average toll at 2009 prices required to cover O&M and ADB interest expenses 103. Tolls to achieve an FIRR equal to the WACC would have to be approximately five times those in Table 19. For example, a fully indexed toll of GEL0.14 would be required (compared with GEL0.029 in Table 23). 104. Tolls to recover O&M, but not interest expenses, would be approximately a quarter of those in Table 24. For example, a fully indexed toll of GEL0.01 would be required (compared with GEL0.029 in Table 24). 105. Using Table 24 results, the total charge for using the two lane section of the Kobuleti bypass would be GEL0.8 for a car and GEL2.30 for an articulated truck (assuming no indexation). These are modest charges relative to the journey time saving benefits of using the bypass.

3. Evaluation of tranches 1 and 2 106. Results for tranches 1 and 2 are shown in Table 25. The range of tolls is solely the result of alternative approaches to indexation. In terms of FIRR, each scenario produces sensibly the same result: an FIRR of -0.8 percent.

Engconsult Ltd. 30

Table 25: Tranches 1 and 2 evaluation results Toll scenario Requireda base toll Base charge per km per PCU-km Car Articulated truck No indexation GEL0.078 (4.7¢) GEL0.042 (2.5¢) GEL0.13 (7.8¢) Partial indexation GEL0.053 (3.2¢) GEL0.024 (1.4¢) GEL0.08 (4.8¢) Full indexation GEL0.046 (2.8¢) GEL0.018 (1.1¢) GEL0.05(3.0¢) Source: consultants’ estimates Note: (a) weighted average toll at 2009 prices required to cover O&M and ADB interest expenses 107. Tolls to achieve an FIRR equal to the WACC would have to be approximately four times those in Table 25. For example, a fully indexed toll of GEL0.19 would be required (compared with GEL0.046 in Table 25). 108. Tolls to recover O&M, but not interest expenses, would be approximately a third of those in Table 25. For example, a fully indexed toll of GEL0.015 would be required (compared with GEL0.046 in Table 25). 109. Using Table 25 results, the total charge for using the two lane section of the bypass would be GEL1.8 for a car and GEL5.5 for an articulated truck (assuming no indexation). As in the case of tranche 1 these are modest charges relative to the journey time saving benefits of using the bypass.

F. Incremental maintenance expenses 110. Total maintenance expenses over 20 years from HDM-4 for tranches 1 and 2 are USD9.3m (at 2009 financial prices), or USD0.5m (GEL0.8m) per year. These values include overlay costs of USD5m towards the end of the evaluation period. 111. Total maintenance expenses represent 3.0 percent of RD’s 2009 routine maintenance budget of GEL25.8m (Table 2). Overlays would not normally be considered part of routine maintenance. If they are deducted the burden falls to 1.3 percent, a small and easily tolerated increase. This is the long term position: for the first few years there should be very little additional expenditure at all.

Engconsult Ltd. 31

ANNEX A: TRRC’s RESPONSES TO FINANCIAL MANAGEMENT QUESTIONNAIRE

Topic Response Remarks 1. Implementing Agency

What is the entity’s legal status / Public Law Legal Entity registration? 1.2 Has the entity implemented an externally- Yes (7 WB Projects and 4 financed project in the past (if so, please provide Grants) details)? 1.3 What are the statutory reporting The World Bank, Ministry requirements for the entity? of Finance of Georgia, Line Ministry, Roads Department, State Treasury Service 1.4 Is the governing body for the project Yes independent? 1.5 Is the organizational structure Yes appropriate for the needs of the project?

2. Funds Flow Arrangements 2.1 Describe (proposed) project funds flow See attached Chart 1 arrangements, including a chart and explanation of the flow of funds from ADB, government and other financiers. 2.2 Are the (proposed) arrangements to Yes transfer the proceeds of the loan (from the government / finance ministry) to the entity satisfactory?

2.3 What have been the major problems in the N/A There are no problems past in receipt of funds by the entity? with regard to receipt of funds after opening Project Accounts for co- financers and Government funds at the State Treasury Service 2.4 In which bank will the Imprest Account be At present all Project PIU has experience in opened? Accounts are opened at the cooperating with local State Treasury Service and foreign commercial banks 2.5 Does the (proposed) project implementing No PIU has experience in unit (PIU) have experience in the management of management of disbursements from ADB? disbursements from IDA, IBRD and Trust Funds 2.7 Does the entity have/need a capacity to Yes manage foreign exchange risks?

Engconsult Ltd. 32

Topic Response Remarks 2.8 How are the counterpart funds accessed? Through the State Treasury Service 2.9 How are payments made from the There are sub-accounts counterpart funds? with special treasury codes for each Project are opened at the State Treasury Service and payments are made based on submitted liabilities and payment orders 2.10 If part of the project is implemented by No PIU has enough communities or NGOs, does the PIU have the reporting and necessary by such agencies? monitoring features built into its systems to track the use of project proceeds by the Roads Department 2.11 Are the beneficiaries required to contribute Yes, contribution is made to project costs? If beneficiaries have an option to only in form of 30% co- contribute in kind (in the form of labor), are proper financing of all project guidelines formulated to record and value the labor costs. Therefore, there are contribution? no proper guidelines formulated to record and value the labor contribution 3. Staffing 3.1 What is the (proposed) organizational Accounting Department structure of the accounting department? Attach an consists of Chief organization chart. accountant and Accountant, Finance Department consists of Finance Manager and Finance Assistant 3.2 Identify the (proposed) accounts staff, See attached Terms of including job title, responsibilities, educational Reference of Finance background and professional experience. Attach Department and CVs of job descriptions and CVs of key accounting staff. finance staff 3.3 Is the project finance and accounting Yes function staffed adequately? 3.4 Is the finance and accounts staff Yes adequately qualified and experienced? 3.5 Is the project accounts and finance staff No trained in ADB procedures? 3.6 What is the duration of the contract with the Automatically renewable 6- finance and accounts staff? months contract 3.7 Indicate key positions not contracted yet, N/A and the estimated date of appointment. 3.10 Does the project have written position Yes, please see attached descriptions that clearly define duties, TRRC FMM responsibilities, lines of supervision, and limits of authority for all of the officers, managers, and staff?

Engconsult Ltd. 33

Topic Response Remarks 3.11 At what frequency are personnel N/A transferred? 3.12 What is training policy for the finance and Finance and accounting accounting staff? staff has opportunity to attend ILO Training Courses and World Bank workshops 4. Accounting Policies and Procedures 4.1 Does the entity have an accounting system Yes that allows for the proper recording of project financial transactions, including the allocation of expenditures in accordance with the respective components, disbursement categories, and sources of funds? Will the project use the entity accounting system? 4.2 Are controls in place concerning the Yes preparation and approval of transactions, ensuring that all transactions are correctly made and adequately explained? 4.3 Is the chart of accounts adequate to Yes, see attached Chart of properly account for and report on project activities Accounts and disbursement categories? 4.4 Are cost allocations to the various funding Yes sources made accurately and in accordance with established agreements? 4.5 Are the General Ledger and subsidiary Yes ledgers reconciled and in balance? 4.6 Are all accounting and supporting Yes documents retained on a permanent basis in a defined system that allows authorized users easy access? Segregation of Duties 4.7 Are the following functional responsibilities Yes performed by different units or persons: (i) authorization to execute a transaction; (ii) recording of the transaction; and (iii) custody of assets involved in the transaction? 4.8 Are the functions of ordering, receiving, Yes accounting for, and paying for goods and services appropriately segregated? 4.9 Are bank reconciliations prepared by Yes someone other than those who make or approve payments? Budgeting System 4.10 Do budgets include physical and financial Yes targets? 4.11 Are budgets prepared for all significant Yes activities in sufficient detail to provide a meaningful tool with which to monitor subsequent performance?

Engconsult Ltd. 34

Topic Response Remarks 4.12 Are actual expenditures compared to the Yes budget with reasonable frequency, and explanations required for significant variations from the budget? 4.13 Are approvals for variations from the Yes, in advance budget required in advance or after the fact? 4.14 Who is responsible for preparation and PIU Director and Finance approval of budgets? Manager 4.15 Are procedures in place to plan project Yes activities, collect information from the units in charge of the different components, and prepare the budgets? 4.16 Are the project plans and budgets of project Yes activities realistic, based on valid assumptions, and developed by knowledgeable individuals? Payments 4.17 Do invoice-processing procedures provide Yes for: (i) Copies of purchase orders and receiving reports to be obtained directly from issuing departments? (ii) Comparison of invoice quantities, prices and terms, with those indicated on the purchase order and with records of goods actually received? (iii) Comparison of invoice quantities with those indicated on the receiving reports? (iv) Checking the accuracy of calculations? 4.18 Are all invoices stamped PAID, dated, Yes reviewed and approved, and clearly marked for account code assignment? 4.19 Do controls exist for the preparation of the Yes payroll and are changes to the payroll properly authorized? Policies And Procedures 4.20 What is the basis of accounting (e.g., cash, Modified cash based accrual)? method 4.21 What accounting standards are followed? IPSAS 4.22 Does the project have an adequate policies Yes and procedures manual to guide activities and ensure staff accountability? 4.23 Is the accounting policy and procedure Yes manual updated for the project activities? 4.24 Do procedures exist to ensure that only Yes authorized persons can alter or establish a new accounting principle, policy or procedure to be used by the entity? 4.25 Are there written policies and procedures Yes covering all routine financial management and related administrative activities?

Engconsult Ltd. 35

Topic Response Remarks 4.26 Do policies and procedures clearly define Yes conflict of interest and related party transactions (real and apparent) and provide safeguards to protect the organization from them? 4.27 Are manuals distributed to appropriate Yes personnel? Cash and Bank 4.28 Indicate names and positions of authorized Mr. George Tsagareli signatories in the bank accounts. Ms. Marina Majagaladze 4.29 Does the organization maintain an N/A PIU has enough adequate, up-to-date cashbook, recording receipts experience in and payments? performing cash operations and maintaining cash-books 4.30 Do controls exist for the collection, timely N/A See attached FMM deposit and recording of receipts at each collection location? 4.31 Are bank and cash reconciled on a monthly N/A basis? 4.32 Are all unusual items on the bank N/A reconciliation reviewed and approved by a responsible official? 4.33 Are all receipts deposited on a timely N/A basis? Safeguard over Assets 4.34 Is there a system of adequate safeguards Yes to protect assets from fraud, waste and abuse? 4.35 Are subsidiary records of fixed assets and Yes stocks kept up to date and reconciled with control accounts? 4.36 Are there periodic physical inventories of Yes fixed assets and stocks? 4.37 Are assets sufficiently covered by Yes insurance policies? Other Offices and Implementing Entities 4.38 Are there any other regional offices or No Roads Department is executing entities participating in implementation? involved in project implementation 4.39 Has the project established controls and Yes procedures for flow of funds, financial information, accountability, and audits in relation to the other offices or entities? 4.40 Does information among the different Yes offices/implementing agencies flow in an accurate and timely fashion? 4.41 Are periodic reconciliations performed Yes among the different offices/implementing agencies?

Engconsult Ltd. 36

Topic Response Remarks Other 4.42 Has the project advised employees, Yes WB, MOF, MORDI, beneficiaries and other recipients to whom to report RDMRDI if they suspect fraud, waste or misuse of project resources or property? 5. Internal Audit 5.1 Is there an internal audit department in the N/A See attached FMM for entity? internal control procedures 5.2 What are the qualifications and experience N/A of audit department staff? 5.3 To whom does the internal auditor report? N/A 5.4 Will the internal audit department include N/A the project in its work program? 5.5 Are actions taken on the internal audit N/A findings? 6. External Audit 6.1 Is the entity financial statement audited Yes Deloitte & Touch regularly by an independent auditor? Who is the auditor? 6.2 Are there any delays in audit of the entity? No When are the audit reports issued? 6.3 Is the audit of the entity conducted Yes according to the International Standards on Auditing? 6.4 Were there any major accountability issues No brought out in the audit report of the past three years? 6.5 Will the entity auditor audit the project Deloitte & Touch accounts or will another auditor be appointed to audit the project financial statements? 6.6 Are there any recommendations made by No the auditors in prior audit reports or management letters that have not yet been implemented? 6.7 Is the project subject to any kind of audit Yes Tax Inspection, from an independent governmental entity (e.g., the Chamber of Control, supreme audit institution) in addition to the external Finance Police, etc. audit? 6.8 Has the project prepared acceptable terms Yes of reference for an annual project audit? 7. Reporting and Monitoring

7.1 Are financial statements prepared for the Yes, IPSAS entity? In accordance with which accounting standards? 7.2 Are financial statements prepared for the N/A implementing unit?

Engconsult Ltd. 37

Topic Response Remarks 7.3 What is the frequency of preparation of Yes Weekly, monthly, financial statements? Are the reports prepared in a quarterly, annual timely fashion so as to useful to management for Financial Statements decision making? for different organizations 7.4 Does the reporting system need to be No adapted to report on the project components? 7.5 Does the reporting system have the Yes PIU produces following capacity to link the financial information with the quarterly reports - project's physical progress? If separate systems PMR ( Project are used to gather and compile physical data, what Monitoring Report) and controls are in place to reduce the risk that the its relevant FMR physical data may not synchronize with the (Financial Monitoring financial data? Report) 7.6 Does the project have established financial Yes Formats of Financial management reporting responsibilities that specify Reports differ what reports are to be prepared, what they are to depending on the contain, and how they are to be used? addressee – World Bank, Georgian Government, etc. 7.7 Are financial management reports used by Yes management? 7.8 Do the financial reports compare actual Yes expenditures with budgeted and programmed allocations? 7.9 Are financial reports prepared directly by Yes the automated accounting system or are they prepared by spreadsheets or some other means? 8. Information Systems

8.1 Is the financial management system Yes computerized? 8.2 Can the system produce the necessary Yes project financial reports? 8.3 Is the staff adequately trained to maintain Yes the system? 8.4 Does the management organization and Yes processing system safeguard the confidentiality, integrity and availability of the data?

Supporting documents – see list in main text

Engconsult Ltd. 38

ANNEX B RD’S RESPONSES TO FINANCIAL MANAGEMENT QUESTIONNAIRE

Note : Roads Department was unable to complete the questionnaire. Only those sections that were completed are shown here.

Topic Response Remarks 4. Accounting Policies and Procedures

4.1 Does the entity have an accounting system that Yes Accounting allows for the proper recording of project financial software "ORIS transactions, including the allocation of expenditures in Accounting" accordance with the respective components, disbursement categories, and sources of funds? Will the project use the entity accounting system? 4.2 Are controls in place concerning the preparation and Yes approval of transactions, ensuring that all transactions are correctly made and adequately explained? 4.3 Is the chart of accounts adequate to properly Yes account for and report on project activities and disbursement categories? 4.4 Are cost allocations to the various funding sources Yes made accurately and in accordance with established agreements? 4.5 Are the General Ledger and subsidiary ledgers Yes reconciled and in balance?

4.6 Are all accounting and supporting documents Yes retained on a permanent basis in a defined system that allows authorized users easy access?

Segregation of Duties 4.7 Are the following functional responsibilities Yes performed by different units or persons: (i) authorization to execute a transaction; (ii) recording of the transaction; and (iii) custody of assets involved in the transaction? 4.8 Are the functions of ordering, receiving, accounting Yes for, and paying for goods and services appropriately segregated? 4.9 Are bank reconciliations prepared by someone other Yes than those who make or approve payments? Budgeting System 4.10 Do budgets include physical and financial targets? 4.11 Are budgets prepared for all significant activities in sufficient detail to provide a meaningful tool with which to monitor subsequent performance? 4.12 Are actual expenditures compared to the budget with reasonable frequency, and explanations required for significant variations from the budget?

Engconsult Ltd. 39

Topic Response Remarks 4.13 Are approvals for variations from the budget required in advance or after the fact? 4.14 Who is responsible for preparation and approval of budgets? 4.15 Are procedures in place to plan project activities, collect information from the units in charge of the different components, and prepare the budgets? 4.16 Are the project plans and budgets of project activities realistic, based on valid assumptions, and developed by knowledgeable individuals? Payments 4.17 Do invoice-processing procedures provide for: (i) Yes Copies of purchase orders and receiving reports to be obtained directly from issuing departments? (ii) Comparison of invoice quantities, prices and terms, with those indicated on the purchase order and with records of goods actually received? (iii) Comparison of invoice quantities with those indicated on the receiving reports? (iv) Checking the accuracy of calculations? 4.18 Are all invoices stamped PAID, dated, reviewed Yes is not used the and approved, and clearly marked for account code stamp "payed" assignment? 4.19 Do controls exist for the preparation of the payroll Yes and are changes to the payroll properly authorized? Policies And Procedures 4.20 What is the basis of accounting (e.g., cash, Bank transfer accrual)? 4.21 What accounting standards are followed? International Standards 4.22 Does the project have an adequate policies and procedures manual to guide activities and ensure staff accountability? 4.23 Is the accounting policy and procedure manual updated for the project activities? 4.24 Do procedures exist to ensure that only authorized persons can alter or establish a new accounting principle, policy or procedure to be used by the entity? 4.25 Are there written policies and procedures covering all routine financial management and related administrative activities?

4.26 Do policies and procedures clearly define conflict of interest and related party transactions (real and apparent) and provide safeguards to protect the organization from them? 4.27 Are manuals distributed to appropriate personnel? Cash and Bank

Engconsult Ltd. 40

Topic Response Remarks 4.28 Indicate names and positions of authorized signatories in the bank accounts.

4.29 Does the organization maintain an adequate, up-to- Yes Cash payment date cashbook, recording receipts and payments? operations are provided very rarely 4.30 Do controls exist for the collection, timely deposit Yes and recording of receipts at each collection location?

4.31 Are bank and cash reconciled on a monthly basis? Yes

4.32 Are all unusual items on the bank reconciliation Yes reviewed and approved by a responsible official?

4.33 Are all receipts deposited on a timely basis? Yes Safeguard over Assets 4.34 Is there a system of adequate safeguards to protect Yes assets from fraud, waste and abuse? 4.35 Are subsidiary records of fixed assets and stocks Yes kept up to date and reconciled with control accounts?

4.36 Are there periodic physical inventories of fixed Yes assets and stocks? 4.37 Are assets sufficiently covered by insurance Yes policies? Other Offices and Implementing Entities

4.38 Are there any other regional offices or executing entities participating in implementation?

4.39 Has the project established controls and procedures for flow of funds, financial information, accountability, and audits in relation to the other offices or entities? 4.40 Does information among the different offices/implementing agencies flow in an accurate and timely fashion?

4.41 Are periodic reconciliations performed among the different offices/implementing agencies?

Other

Engconsult Ltd. 41

Topic Response Remarks 4.42 Has the project advised employees, beneficiaries and other recipients to whom to report if they suspect fraud, waste or misuse of project resources or property?

5. Internal Audit 5.1 Is there an internal audit department in the entity?

5.2 What are the qualifications and experience of audit department staff?

5.3 To whom does the internal auditor report? 5.4 Will the internal audit department include the project in its work program?

5.5 Are actions taken on the internal audit findings? 6. External Audit 6.1 Is the entity financial statement audited regularly by Yes an independent auditor? Who is the auditor? 6.2 Are there any delays in audit of the entity? When are the audit reports issued? 6.3 Is the audit of the entity conducted according to the International Standards on Auditing?

6.4 Were there any major accountability issues brought out in the audit report of the past three years? 6.5 Will the entity auditor audit the project accounts or will another auditor be appointed to audit the project financial statements?

6.6 Are there any recommendations made by the No auditors in prior audit reports or management letters that have not yet been implemented?

6.7 Is the project subject to any kind of audit from an independent governmental entity (e.g., the supreme audit institution) in addition to the external audit?

6.8 Has the project prepared acceptable terms of reference for an annual project audit? 7. Reporting and Monitoring 7.1 Are financial statements prepared for the entity? In International accordance with which accounting standards? Standards

7.2 Are financial statements prepared for the N/A implementing unit?

Engconsult Ltd. 42

Topic Response Remarks 7.3 What is the frequency of preparation of financial quarterly; statements? Are the reports prepared in a timely fashion timely so as to useful to management for decision making?

7.4 Does the reporting system need to be adapted to report on the project components? 7.5 Does the reporting system have the capacity to link the financial information with the project's physical progress? If separate systems are used to gather and compile physical data, what controls are in place to reduce the risk that the physical data may not synchronize with the financial data? 7.6 Does the project have established financial management reporting responsibilities that specify what reports are to be prepared, what they are to contain, and how they are to be used? 7.7 Are financial management reports used by management? 7.8 Do the financial reports compare actual expenditures with budgeted and programmed allocations? 7.9 Are financial reports prepared directly by the automated accounting system or are they prepared by spreadsheets or some other means?

Engconsult Ltd. FINAL REPORT

Asian Development Bank Roads Department

Supplementary Appendix 8 Land Acquisition and Resettlement Framework

TA No. 7059-GEO Preparing the Ajara Bypass Roads Development Project

August 2009

ENGCONSULT LTD. 21 Queen Street E., Suite 302 Brampton, Ontario, L6W 3P1 Canada BT LTD. ECL CONSULTANTS LTD. Road No 2, House No. 18, (2nd Floor) 11 Brothers Kakabadze St., 0108 Tbilisi, Georgia Mirpur 11, Dhaka 1221, Bangladesh. www.eng‐consult.com

Land Acquisition and Resettlement Framework (LARF)

Sub Regional Road Corridors Development Program

July 2009

Roads Department (RDMRDI) Ministry of Regional Development and Infrastructure of Georgia

ii

TABLE OF CONTENTS

ABBREVIATIONS iii DEFINITIONS OF LAR TERMS iv A. INTRODUCTION 1 B. LAR-RELATED PROGRAM PROCESSING REQUIREMENTS 1 C. INDIGENOUS PEOPLES AND VULNERABLE GROUPS 2 D. LARP PREPARATION AND IMPLEMENTATION 2 E. LEGAL AND POLICY BACKGROUND 3 E.1 Georgia Law and Regulation on Land Acquisition and Resettlement 3 E.2 ADB’s Involuntary Resettlement Policy 4 E.3 Comparison of Georgia Law on LAR and ADB Resettlement Policy 4 E.4 LARF Principles and Entitlements adopted for this MFF 5 F. COMPENSATION ELIGIBILITY AND ENTITLEMENT FOR THE PROJECT 6 F.1 Eligibility 7 F.2 Compensation Entitlements 7 F.3 Assessment of Compensation Unit Values 8 F.4 Conditions for Expropriation 9 G. GENDER IMPACTS AND MITIGATION MEASURES 9 H. PUBLIC PARTICIPATION AND DOCUMENTS DISCLOSURE 9 I. INSTITUTIONAL ARRANGEMENTS 10 I.1 RDMRDI 10 I.2 Consultants 11 I.3 Local Governments 11 I.4 Other Agencies and Institutions 11 J. COMPLAINTS AND GRIEVANCES 13 K. MONITORING AND EVALUATION 13 L. CAPACITY BUILDING AND TRAINING IN LAR IMPLEMENTATION 14 M. RESETTLEMENT BUDGET AND FINANCING 15 N. LARP IMPLEMENTATION PROCESS 15 Tables: Table 1: Comparison of Georgian Laws on LAR and ADB Resettlement Policy 5 Table 2: Compensation Matrix 6 Table 3: Grievance Resolution Process 13 Figures: Figure 1: LAR Organogram 12 Boxes Box 1 : LAR Task Process 15 Annexes: Annex A: Synopsis of Selected Georgia Laws and Regulations on LAR 17 Annex B: Summary of ADB’s Policy on Involuntary Resettlement 21 Annex C: Outline of a Standard LARP 23

iii

ABBREVIATIONS

ADB Asian Development Bank AF Affected Family AH Affected Household AP Affected Person

DMS Detailed Measurement Survey DRD Development and Resettlement Division

IA Implementing Agency IPSA Initial Poverty and Social Assessment LAR Land Acquisition and Resettlement LARF Land Acquisition and Resettlement Framework LARP Land Acquisition and Resettlement Plan LRAG Land Registration Agency of Georgia MFF Multi Tranche Financial Facility NAPR National Agency of Public Registry NGO Non-Governmental Organization PPTA Project Preparatory Technical Assistance PSA Poverty and Social Assessment RDMRDI Road Department of the Ministry of Regional Development and Infrastructure of Georgia

iv

DEFINITIONS OF LAR TERMS

Beneficiary Community - All persons and households situated within the government- owned or acquired property who voluntarily seeks to avail and be part of the Project and represented by a community association that is duly recognized by the community residents, accredited by the local government, and legally registered with the appropriate institutions.

Compensation - Payment in cash or in kind of the replacement cost of the acquired assets.

Entitlement - Range of measures comprising compensation, income restoration, transfer assistance, income substitution, and relocation which are due to affected people, depending on the nature of their losses, to restore their economic and social base.

Host population - Community residing near the area where the Project beneficiaries propose to voluntarily resettle as part of the Project.

Improvements - Structures constructed (dwelling unit, fence, waiting sheds, animal pens, utilities, community facilities, stores, warehouses, etc.) and crops/plants planted by the person, household, institution, or organization.

Land Acquisition - The process whereby a person is compelled by a government agency to alienate all or part of the land a person owns or possesses to the ownership and possession of the government agency for public purpose in return for a consideration.

Affected Person/People - Any person affected by Project-related changes in use of land, water, natural resources, or income losses.

Affected Family - All members of a household residing under one roof and operating as a single economic unit, who are adversely affected by the Project, or any of its components. It may consist of a single nuclear family or an extended family group.

Rehabilitation - Compensatory measures provided under the Policy Framework on involuntary resettlement other than payment of the replacement cost of acquired assets.

Relocation - the physical relocation of a AP/AF from her/his pre-Project place of residence.

Replacement Cost - The value determined to be fair compensation for land based on its productive potential, the replacement cost of houses and structures (current fair market price of building materials and labor without depreciation or deductions for salvaged building material), and the market value of residential land, crops, trees, and other commodities.

Resettlement - All measures taken to mitigate any and all adverse impacts of the Project on AP’s property and/or livelihood, including compensation, relocation (where relevant), and rehabilitation of the damaged/removed infrastructure and installations.

Sacrebolu: This is the village administration

Gamgebeli: This is the administrative unit of each Rayon

Rayon: It is the local government

1

LAND ACQUISITION AND RESETTLEMENT FRAMEWORK (LARF)

A. INTRODUCTION

1. This Land Acquisition and Resettlement Framework (LARF) for the Southern Corridor Roads Development Program (the Program) has been prepared by The Road Department of the Ministry of Regional Development and Infrastructure of Georgia (RDMRDI) and is fully endorsed by the RDMRDI. The Project, to be financed by ADB under the Multi Tranche Financial Facility (MFF) will be implemented over a period of 5 years and will cover three tranches with various projects in each tranche. The Program will be implemented by RDMRDI and will finance the civil works for construction of various roads.

2. The objective of this document is to provide guidance in the preparation/implementation of Land Acquisition and Resettlement (LAR) tasks for the projects under each tranche and to establish the necessary covenants to ensure that this is done in compliance to ADB’s Policy on Involuntary Resettlement and procedures for the MFF. This LARF sets out the procedures for the preparation of Land Acquisition and Resettlement (LAR) tasks for projects under the Program and sets the objectives, principles, compensation eligibility/entitlement criteria, legal/institutional frameworks, participation/consultation procedures and grievance redress mechanisms to be employed to compensate, resettle and rehabilitate the living standards of Affected Families (AF) and People (APs.) This LARF also sets out the steps for preparing LAR Plans (LARP) in case they are needed.

3. Before Program appraisal this LARF will be translated into Georgian and distributed to all interested stakeholders. The English version will be posted on the ADB website.

B. LAR-RELATED PROGRAM PROCESSING REQUIREMENTS

4. Based on ADB policy/practice appraisal of the MFF and each tranche and approval of the implementation of tranche subprojects will require the preparation of the following:

(i) A LARF for the whole financial facility and applicable to all subprojects. The LARF will be reviewed, if necessary updated, and submitted for ADB approval at least annually and always at the start of the preparation of each tranche. (ii) An Initial Poverty and Social Assessment (IPSA) indicating, for each tranche, whether LAR impacts are likely to occur, type of impacts, likely magnitude, and whether there may be Indigenous Peoples (IP) affected, and; (iii) If LAR occurs, a LAR Plan (LARP) for each project under a tranche based on detailed design and commensurate to the severity of impacts1. The LARP will include detailed compensation and administration budgets and implementation schedules linking LAR tasks to the initiation of civil works.

1 Based on ADB Operation Manual (OM) F2/OP and BO (2006) a project is classified as Category “A” and a full LARP is needed if > 200 people suffer significant impacts (relocation or loss of >10% of productive assets). A project will instead be classified as Category “B” when less than 200 people suffer significant impacts. Category “C” projects have no LAR impacts. No subproject will affect indigenous Peoples. The WB does not have a similar resettlement classification system but its policy also envisages that for minor impacts a shorter document, called “Abbreviated Resettlement Plan”, is prepared.

2

5. Based on ADB policy/practice, the appraisal of the MFF and each specific tranche and the approval of project implementation will be based on the following LAR-related conditions:

(i) Appraisal of MFF and Project 1: Conditional to preparation/disclosure of a LARF for the whole MFF acceptable to ADB and preparation of the LARPs for the tranche projects requiring LAR. (ii) Appraisal of subsequent projects: Conditional to review/update/disclosure of the LARF, and preparation of needed LARPs consistent with the revised LARF and acceptable to ADB for projects with LAR. (iii) Contract award for each project: Conditional to the update/disclosure of the relevant LARPs based on the final design and accepted by ADB. The updated LARPs will reflect final impacts, final AP lists and final compensation rates and be readily implementable. (iv) Provision of notice to proceed to contractors: Conditional to the full implementation of the relevant LARP (full delivery of compensation and rehabilitation) for the relevant project. Such a condition will be clearly spelled out in the text of the civil works contract.

C. INDIGENOUS PEOPLES AND VULNERABLE GROUPS

6. No impacts on Indigenous Peoples are expected for the Program. Special attention will also be given to identifying and addressing the needs of disadvantaged groups such as the landless, the poor, female-headed households, the elderly and the disabled, through measures included in the LARP to try and improve (over and above cash compensations and restoration of) their livelihoods.

D. LARP PREPARATION AND IMPLEMENTATION

7. LARP preparation activities will be initiated as part of the preparation of each tranche project starting with LAR screening tasks and ending with a readily implementable a LARP based on detailed design. Following the finalization of road alignment and identification of the land parcels, cross-sections design and land acquisition requirements a, Detailed Measurement Survey (DMS), a census of all AF and AP and a valuation of all affected asset will be carried out to assess the AF/P losses and relative compensation budgets. In addition, a sample socio-economic survey will be conducted based on the 20% of the AFs. The Socio-economic Survey will cover their major socio-economic features of the affected population (ethnicity, education level, modes of livelihood, and sources of income, poverty/income levels, and house type/value and land tenure types. The DMS and census survey include the following:

• Inventory of the 100% loss of land parcel and property • Categorization and measurements of loss; • Measurements of the affected assets/structures, including their replacement valuation; • AF characteristics, including social, economic and demographic profile; • The AP/AF Census will identify all AF and their members by number, gender and age; all AF by land occupancy status; all severely AF (losing >10% of their land and income; and all vulnerable AFs (women headed family or family under poverty line). • Identification of non titleholders

8. LARP preparation/implementation Process. LARP preparation for the first

3

tranche of the Program will be carried out by the PPTA consultants; the same tasks for the following tranches will be carried out by the detailed design consultants. LARP implementation will be overseen by the supervision consultants. LARP preparation/implementation will be carried out according to the following process:

a. LARP preparation. This phase will be carried out in parallel with the preparation of the projects feasibility study (See Annex C for a standard outline of a LARP). b. LARP update/finalization. This phase will be carried out during the preparation of detailed design and will include eventual updates of the impacts/AP data, possible revisions of compensation rates and the administrative work done to legalize legalizable APs. After the final LARP is approved by ADB, the civil works contracts will be signed. c. LARP implementation. This phase will be carried out after the final LARP is approved. In principle civil works will start only after all compensation and rehabilitation is provided to the APs, However, if in a road there are sections without impacts and sufficiently long to economically justify construction, the civil works can initiate immediately after the signing of contract awards. Following the preparation of a due diligence report fully documenting the absence of impacts,

E. LEGAL AND POLICY BACKGROUND

E.1 Georgia’s Laws and Regulations on Land Acquisition and Resettlement

9. In Georgia, the legislative acts given below regulate the issues of obtaining State ownership rights to privately owned land parcels based on the necessary public needs caused due to road constructions activities:

• The Constitution of Georgia, August 24, 1995 • The Civil Code of Georgia, June 26, 1997 • The Law of Georgia on Protection of Cultural Heritage • The Law of Georgia on Notary Actions • The Law of Georgia on Privatization of State-owned Agricultural Land, July 8, 2005 • The Law of Georgia on Ownership Rights to Agricultural Land, March 22, 1996 • The Law of Georgia on Registration Ownership Rights to Immovable Property, December 28, 2005 • The Law of Georgia on the Rules for Expropriation of Ownership for Necessary Public Need, July 23, 1999 • The Civil Procedural Code of Georgia, November 14, 1997

10. The existing Laws provide that compensation for lost assets, including land, structures, trees and standing crops, should be based on the current market price without depreciation. Overall the above laws/regulations provide that the principle of replacement cost compensating at market value is reasonable and legally acceptable. The laws also identify the types of damages eligible to compensation and indicate that compensation is to be given both for loss of physical assets and for the loss of incomes. Finally, these laws place strong emphasis on consultation and notification to ensure that the APs participate in the process. Income loss due to loss of harvest and business closure will be compensated to cover net loss. The above-listed laws and regulations give the possibility of applying the following mechanisms for legal application of the property rights:

4

• Obtaining the right on way without expropriation through the payment of due compensation (on the basis of negotiations or a court decision) prior to commencement of the activities. • Expropriation which gives the possibility of obtaining permanent right to land and/or other real estate property on the basis of Eminent Domain Law or a court decision through the payment of due compensation;

11. Attempts should first be made to acquire private land on the basis of negotiation with individual affected entities. Should the negotiation fail, the power of eminent domain will be sought, and expropriation process will start. Under the existing Law in Georgia, the president will issue an order for expropriation based on the request from relevant state agencies. Relevant regional court will assess the presidential order and determine the case of public needs, and grant the expropriation entity rights to obtain land. The court will also appoint a third party to assess the market value of lost assets and determine the compensation payable to relevant land owners accordingly to the value of assets thus found.

E.2. ADB’s Involuntary Resettlement Policy

12. The ADB Policy on Involuntary Resettlement is based on the following principles: • Involuntary resettlement is to be avoided or at least minimized. • Compensation/Rehabilitation provisions will ensure the maintenance of the APs’ pre-project standards of living. • APs should be fully informed and consulted on LAR compensation options. • APs’ socio-cultural institutions should be supported/used as much as possible. • Compensation will be carried out with equal consideration of women and men. • Lack of legal title should not be a bar to compensation and/or rehabilitation. • Particular attention should be paid to households headed by women and other vulnerable groups, such as Indigenous Peoples, and appropriate assistance should be provided to help them improve their status. • LAR should be conceived and executed as a part of the project, and the full costs of compensation should be included in project costs and benefits. • Compensation and resettlement subsidies will be fully provided prior to clearance of right of way/ ground leveling and demolition.

E.3 Comparison of Georgia Law/Regulation on LAR and ADB Resettlement Policy

13. Overall, the legislation of Georgia adequately reflects the major provisions of the ADB Resettlement Policy but a few differences are to be noted (for a short summary of these Georgian laws see Annex A). The most significant of these differences is that under Georgian legislation/regulation, emphasis is put on the definition of formal property rights and on how the acquisition of properties for public purposes is to be implemented and compensated while in the case of ADB policy emphasis is put both on the compensation of rightfully owned affected assets and on the general rehabilitation of the livelihood of Affected People (AP) and Households (AH). Because of this, ADB policy complements the Georgian legislation/regulation with additional requirements related to (i) the economic rehabilitation of all AP/AF (including those who do not have legal/formal rights on assets acquired by a project); (ii) the provision of indemnities for loss of business and income, (iii) and the provision of special allowances covering AP/AH expenses during the resettlement process or covering the special needs of severely affected or vulnerable AP/Ahs (See Annex B for a synopsis on ADB’s Policy on Invloumatry Resettlement). Also, in addition, the legislation of Georgia does not require any specific measure regarding the need to

5 prepare LARPs based on extensive public consultations. The differences between Georgia law/regulation and ADB policy are outlined in Table1 below

Table 1: Comparison of Georgian Laws on LAR and ADB Resettlement Policy Georgia Laws and Regulations ADB Involuntary Resettlement Policy Land compensation only for titled landowners In Lack of title should not be a bar to compensation and/or practice legaliizable land owners are also rehabilitation. Non-titled landowners receive rehabilitation. compensated after they are issued with the necessary papers Only registered houses/buildings are compensated All affected houses/buildings are compensated for for damages/demolition caused by a project buildings damages/demolition caused by a project Crop losses compensation provided only to Crop losses compensation provided to landowners and registered landowners. sharecrop/lease tenants whether registered or not Land Acquisition Committee is the only pre-litigation Complaints & grievances are resolved informally through final authority to decide disputes and address community participation in the Grievance Redress complaints regarding quantification and assessment of Committees (GRC), Local governments, and NGO and/or compensation for the affected assets. local-level community based organizations (CBOs). Decisions regarding LAR are discussed only Information on quantification, affected items value between the landowners and the Land Acquisition assets, entitlements, and compensation/financial Authorities. assistance amounts are to be disclosed to the APs prior to appraisal. No provision for income/livelihood rehabilitation, ADB policy requires rehabilitation for income/livelihood, allowances for severely affected or vulnerable APs, severe losses, and for expenses incurred by the APs or resettlement expenses. during the relocation process. No specific plan for public consultation is provided Public consultation and participation is the integral part under the Georgian laws of ADB’s policy which is a continuous process at conception, preparation, implementation and finally at post implementation period

14. To reconcile the gaps between Georgia laws/regulations and ADB Policy, RDMRDI has drafted this LARF for the Project, ensuring compensation at replacement cost of all items, the rehabilitation of informal settlers, and the provision of subsidies or allowances for AFs that may be relocated, suffer business losses, or may be severely affected.

E.4 LARF Principles and Entitlements adopted for this MFF

15. Based on the Georgian laws on land acquisition and ADB’s Policy on Involuntary Resettlement, 1995, core involuntary resettlement principles are developed for this MFF which are as follows:

• Land acquisition, and other involuntary resettlement impacts will be avoided or minimized exploring all viable alternative project designs; • where unavoidable, a time-bound LARP will be prepared and APs will be assisted in improving or at least regaining their pre-program standard of living; • consultation with APs on compensation, disclosure of resettlement information to APs, and participation of APs in planning and implementing sub-projects will be ensured; • vulnerable and severely affected APs will be provided special assistance; • Non-titled APs (e.g., informal dwellers or squatters, APs without registration details) will receive a livelihood allowance in lieu of land compensation and will be fully compensated for losses other than land.; • Legalizable APs will be legalized and fully compensated for land losses. • Provision of income restoration and rehabilitation; • The LARP will be disclosed to the APs in the local language

6

• Payment of compensation, resettlement assistance and rehabilitation measures will be fully provided prior to the contractor taking physical acquisition of the land and prior to the commencement of any construction activities on a particular package and . • Establishment of appropriate grievance redresses mechanisms to solve APs grievance if occurs. F. COMPENSATION ELIGIBILITY AND ENTITLEMENTS FOR THE PROJECT

16. LAR tasks under the Project will be implemented according to a compensation eligibility and entitlements framework in line with both Georgia laws and regulation and ADB Policy. A summary entitlements matrix is included in Table 2 below.

Table 2: Compensation Matrix Type of Loss Application Definition of APs Compensation Entitlements Land Permanent loss of AF losing agricultural land Owner with full registration Cash compensation at replacement cost or agricultural land regardless of impact severity through replacement land equal in value/productivity to the plot lost and at location acceptable to APs where feasible Legalizable Owner These AP will be legalized and provided with cash compensation at replacement cost. Informal Settlers/ APs with no One time self-relocation allowance in cash registration/valid documentation equal to 1 year at minimum salary Non-Agricultural Land AF losing their commercial/ Owner with full registration Cash compensation at replacement cost or residential land through replacement land equal in value to the plot lost and at location acceptable to APs where feasible Legalizable Owner APs will be legalized and provided with cash compensation at replacement cost.. Renter/Leaseholder Rental allowances in cash for 3 months Informal Settlers/ APs with no One time self-relocation allowance in cash registration/valid documentation equal to 1 year at minimum salary Buildings and Structures Residential and non All AFs regardless of their legal Full impact: Cash compensation for loss of residential ownership/ registration status building/ structures at full replacement costs free of structures/assets (including legalizable and depreciation and transaction costs Informal Settlers) Partial impact: compensation for repairs Loss Of Community Infrastructure/Common Property Resources Loss of common Community/Public Assets Community/Government Reconstruction of the lost structure in consultation property resources with community and restoration of their functions Loss of Income and Livelihood Crops Standing crops affected All AFs regardless of legal Crop compensation in cash at market rate by status (including legalizable and default at to gross crop value of expected harvest. Informal Settlers) Trees Trees affected All AFs regardless of legal Cash compensation at market rate on the basis status (including legalizable and of type, age and productive value of the trees. Informal Settlers) Business/Employment Business/employment loss All AFs regardless of legal Owner: (i). (permanent impact) cash indemnity of 1 status (including legalizable and year net income; (ii) (temporary impact) cash indemnity Informal Settlers) of net income for months of business stoppage. Assessment to be based on tax declaration or, in its absence, minimum salary.. Permanent worker/employees: indemnity for lost wages equal to 3 months of minimum salary. Allowances Severe Impacts >10% income loss All severely affected AFs Agricultural income: 1 additional crop compensation including informal settlers covering 1 year yield from affected land. Other income: 1additional compensation for 3 months of minimum salary. Relocation/Shifting Transport/transition costs All AFs to be relocated Provision of sufficient allowance to cover

7

Type of Loss Application Definition of APs Compensation Entitlements transport expenses and livelihood expenses for the transitional period (up to 1 month). Vulnerable People AFs below poverty line, headed Allowance equivalent to 3 months of minimum salary Allowances by Women, disabled or elderly and employment priority in project-related jobs Temporary Loss Temporary impact All AFs Due compensation will be assessed and paid during construction based on this LARF during construction. Unforeseen Road Department and the construction contractor resettlement impacts, will address and mitigate/compensate unforeseen if any resettlement impact during project

F.1 Eligibility

17. APs entitled for compensation or at least rehabilitation provisions under the Project are:

• All APs losing land either covered by legal title/traditional land rights, Legalizable, or without legal status; • Tenants and sharecroppers whether registered or not; • Owners of buildings, crops, plants, or other objects attached to the land; and • APs losing business, income, and salaries.

18. Compensation eligibility will be limited by a cut-off date to be set for each subproject on the day of the beginning of the AP Census and DMS. APs who settle in the affected areas after the cut-off date will not be eligible for compensation. They, however will be given sufficient advance notice, requested to vacate premises and dismantle affected structures prior to project implementation. Their dismantled structures materials will not be confiscated and they will not pay any fine or suffer any sanction.

F.2 Compensation Entitlements

19. Entitlement provisions for APs losing land, houses, and income and rehabilitation subsidies will include provisions for permanent or temporary land losses, buildings losses, crops and trees losses, a relocation subsidy, and a business losses allowance based on tax declarations and/or lump sums. These entitlements are detailed below:

• Agricultural land impacts will be compensated at replacement value in: (i) cash at current market rates or based on an assessment of the reproduction cost of the affected land, or (ii) through replacement land equal in value/productivity to the plot lost and at location acceptable to APs if alternate land is feasible and available. When >10% of an AP agricultural land is affected, AP (owners, leaseholders and sharecroppers) will get an additional allowance for severe impacts equal to the market value of a year’s gross yield of the land lost. In case of severe impact on other income, the APs will be paid additional compensation for 3 months income. Transaction taxes and fees will be paid by the EA or waived. Legalizable APs will be legalized and paid as titled owners. Non-legalizable APs will be compensated with one time self-relocation allowances in cash equal to 1 year of minimum salary. If the remaining part of a particular plot becomes inaccessible or unviable for cultivation or for any use after the acquisition, then the same can be compensated if the owner offers.

• Non agricultural land (Residential/commercial land). Legal settlers will be compensated at replacement rate either (i) in form cash at current market rates free of transaction costs and depreciation or (ii) through replacement land equal in value to the

8

plot lost and at location acceptable to APs if alternate land is feasible and available. Renters/leaseholders will receive a 3 months rent allowance. Non-titled and non- legalizable land users will be compensated with one time self relocation allowances in cash equal to 1 year of minimum salary.

• Houses, buildings, and structures will be compensated in cash at replacement cost free of deductions for depreciation, and transaction costs irrespective of the registration status of the affected item. In case of partial impacts (<15%) and unwillingness of the owner to relocate, compensation will cover only the affected portion of a building and its full rehabilitation to previous use. Full compensation will be paid if partial impacts imperil the viability of the whole building.

• Crops: Cash compensation at current market rates for the gross value of 1 year’s harvest by default. Crop compensation will be paid both to landowners and tenants based on their specific sharecropping agreements.

• Trees: Cash compensation at market rate based on type, age and productivity of trees.

• Businesses: If business is lost permanently it will be compensated in cash equal to a 1- year income based on tax declaration or, if unavailable, based on the official minimum salary; temporary business losses will be compensated in cash for the business interruption period based on tax declaration or, if unavailable, official minimum salary.

• Permanent business workers and employees: Indemnity for lost wages for the period of business interruption up to a maximum of 3 months.

• Relocation subsidy: APs forced to relocate will receive a relocation subsidy sufficient to cover transport costs and living expenses for 1 month.

• Community Structures and Public Utilities: Will be fully replaced or rehabilitated so as to satisfy their pre-project functions.

• Vulnerable people Livelihood: Vulnerable people (APs below poverty line and widow or elder headed households) will be given priority in employment in project-related jobs.

F.3 Assessment of Compensation Unit Values

20. The methodology for assessing unit compensation values of different items is as follows: (i) Agricultural Land will be valued at replacement rates according to two different methodologies depending on whether in affected areas active land markets exist or not. a. Where active land markets exist land will be compensated at replacement rate based on a survey of land sales in the year before the impact survey. b. Where active land markets do not exist land will be compensated based on the reproduction cost of a plot with equal features, access and productivity to the plot lost. A clear valuation methodology for these cases will be detailed in the LARPs. (ii) Houses/buildings will be valued at replacement value based on construction type, cost of materials, type of construction, labor, transport and other construction costs. No deduction for depreciation and transaction costs will be applied. For the partial impact (if the loss is less than 15% then compensation is paid for the repair of the affected structure) (iii) Annual crops will be valued at net market rates at the farm gate for the first year

9

crop. In the eventuality that more than one-year compensation is due to the APs the crops after the first will be compensated at gross market value. (iv) Trees will be valued according to different methodologies depending whether the tree lost is a wood tree or a productive tree. a. Wood trees will be valued based on age category (a. seedling; b. medium growth and c. full growth) and wood value and volume. b. Fruit/productive trees will be valued based on age (a. seedling; b. adult-not fruit bearing; and c. fruit bearing). Stage (a) and (b) trees will be compensated based on the value of the investment made; stage (c) trees will be compensated at net market value of 1 year income x number of years needed to grow a new fully productive tree.

21. The unit compensation rates will be assessed by Project consultants or by the authorized independent evaluator based on clear and transparent methodologies acceptable to ADB. The assessed compensation rates will then be verified and certified by the resettlement division in RDMRDI.

F.4 Conditions for Expropriation 22. Acquisition of land through expropriation proceedings entails lengthy procedures often may be resisted. Such an approach will thus be pursued under the Program only in extreme cases when negotiations between APs and RDMRDI fail and no alternative land is available to implement a subproject. In these cases, however, RDMRDI will not occupy the needed plots until: (i) the proper judicial process as defined by the law is initiated; (ii) a court injunction has been obtained and properly communicated to the APs; and (iii) the compensation/rehabilitation amounts are deposited in an escrow account.

G. GENDER IMPACT AND MITIGATION MEASURES

23. Women have important economic roles in project areas and engage in a very wide range of income making activities in the agricultural and marketing sector. The project will pay particular attention to ensure that women are the recipients of the compensation pertaining to their activities and to ensure that women who are de-facto household heads are clearly listed as beneficiaries of compensation and rehabilitation proceedings under the loan. In order to ensure the above the following actions will be considered: • Include women in the impact enumerators. • Impact assessment of AFs/APs indicating the total number of families and people must be gender-disaggregated to pinpoint how many women are likely to be affected by the Project and establish their pre-Project conditions. • Women will be major participants in the consultation processes to determine and negotiate for compensation entitlements and implement the RP. • Special attention will be given to the impact of resettlement on women and other vulnerable groups during monitoring and evaluation of the RP.

H. PUBLIC CONSULTATION, PARTICIPATION AND DOCUMENTS DISCLOSURE

24. Concerned officials of centre, district/rayon, municipalities and villages/sacrebolu will be informed about the Project, and their assistance will be solicited in the conduct of the inventory of affected assets and the Census of APs and the DMS. Also, prior to the finalization of the LARP and its submission to Project authorities, the APs will be thoroughly informed on the results of the Census and DMS, and their preferences on compensation or other resettlement assistance will be given due consideration. The

10 processes and mechanisms ensuring the active involvement of APs and other stakeholders will be detailed in the LARPs which will also include an appendix with date, list of participants, and minutes of consultation meetings.

25. This LARF in Georgian will be disclosed on the RDMRDI website and at RDMRDI offices before Project appraisal. The LARF in Georgian will also be disclosed to the APs at the relevant Rayon office (Gamgebeli) and at village administration (Sacrebolu) once subprojects are identified. Its English version will be disclosed on the ADB website prior to Project appraisal and after the LARF is endorsed by the Implementing Agency (IA) which will be RDMRDI in this case. Once a LARP for a subproject has been prepared and approved by RDMRDI and ADB it will be disclosed at relevant Rayon office (Gamgebeli) and at village administration (Sacrebolu). A pamphlet in Georgian, summarizing compensation eligibility and entitlement provisions, will be sent to all AP/AFs before the initiation of the compensation/rehabilitation process and before signing contract awards. The consultation process will be continued throughout the project cycle.

I. INSTITUTIONAL ARRANGEMENTS

26. Asian Development Bank (ADB) will be the funding agency of the MFF. The compensation/rehabilitation program described in this LARF involves distinct processes and dynamics and different actors. The Road Department of the Ministry of Regional Development and Infrastructure of Georgia (RDMRDI) will be the Implementing Agency (IA) having the lead responsibility for road construction, as well as the implementation of this LARF and subsequently the LARP. RDMRDI with the assistance of the consultants will develop and implement the LARP for each project based on the policy and procedures set out in the LARF. In addition to the RDMRDI, a number of other government departments and private agents will play an instrumental role in the design, construction and operation of the project. Pursuant to the active legislation, the Ministry of Natural Resources and Environmental Protection is responsible for environmental issues. The Ministry of Justice is responsible for legal matters regarding land ownership, and National Agency of Public Registry (NAPR) within the Ministry of Justice is in charge of the registration of land ownership and its transfer through purchase agreement from landowners to the Road Department. The local government at Rayon and village level will also be involved. The role of each of these actors is detailed below.

I.1. RDMRDI

27. RDMRDI has overall responsibility for the MFF. This includes preparation, implementation and financing of all LAR tasks and cross-agency coordination. RDMRDI will exercise its functions through its existing resettlement division (DRD) which will be responsible for the general management of the planning and implementation of all LAR tasks.

28. DRD. The DRD staffed with a sufficient number of LAR specialists, will be tasked with all LAR coordination tasks at central and local government level and will be responsible for: (i) screening the projects and ensuring that the LARPs are properly prepared and sent to ADB for review, (ii) supervising the consultants that prepare the LARPs and assist in their implementation; (iii) establishing needed LAR capacity at each regional level office where LAR is relevant; (iv) ensuring proper internal monitoring; and (v) hire, following ADB recommendation the external monitoring agency. DRD will also provide all needed documentation to ensure the prompt allocation of LAR budgets to the APs and will maintain the coordination of all LAR related activities.

11

29. Local RDMRDI Offices The Regional RDMRDI offices will assist the activities of the DRD with one dedicated officer who will facilitate the communication between the DRD, the local governments and the APs and assist in implementing LAR tasks related to the local administration.

I.2 Consultants. 30. Different types of consultants will be involved in LAR tasks:

a. PPTA consultants: These will include international and local LAR capacity and needed survey teams and will carry out all field-surveys and prepare the needed LARPs for the first tranche. b. Design consultants: These will include international and local LAR capacity and needed survey teams carry out the same activities for updating/finalizing the LARPs of first tranche based on the detailed engineering design and will also prepare LARPs for the following tranches. c. Supervision consultants: These will include international and local LAR capacity and needed survey teams and will assist in the overall supervision of the projects. The supervision consultants will also oversee LARP Implementation and carry out external monitoring and evaluation of the implementation of LARP for following tranches of the Program. e. Independent Asset valuators. These will be accredited private firms to be hired by the PPTA or Design consultants to carry out the evaluation of affected assets.

I.3 Local Governments

31. Local government especially at Rayon level has direct jurisdiction for land administration, valuation, verification and acquisition. To confirm the surveys and the asset valuations carried out by the LAR consultants RDMRDI through its consultants will establish in Rayon where LARPs are to be implemented a Rayon Level LAR team which will have designated officials from the Rayon administration. The LAR team at Rayon level will have close coordination with the village administration for the LAR activities. Effective inter-agency coordination at Rayon and village level will be assigned to the regional level RDMRDI.

I.4 Other Agencies and Institutions

32. Several other institutions will participate to the preparation and implementation of LAR tasks. These are:

(i) Ministry of Finance. The budgets for the implementation of the LARPs will be provided to RDMRDI by the Ministry of Finance following the official approval of the final LARPs.

(ii) Ministry of Justice. The Ministry of Justice is responsible for legal matters regarding land ownership, and National Agency of Public Registry within the Ministry of Justice is in charge of the registration of land ownership and its transfer through purchase agreement from landowners to the Road Department.

12

(iii) Local Courts. In case of expropriation issues RDMRDI will have to rely on the Rayon court which based on due legal process will have to review the expropriation cases, carry out a hearing and decide whether the land can be expropriated and at what price. In order to expedite the expropriation process RDMRI will negotiate with the courts a fast-tracked action plan.

(iv) ADB. Beside supervising periodically the Project ADB will review all LARPs and provide clearance to contract awards signing and initiation of civil works to all subprojects with LAR.

33. The LAR organogram is provided below in Figure 1:

Figure 1: LAR Organogram and Action

Government of ADB Georgia

RDMDRI

DRD

rvision Consultant Regional Level RDMDRI Supe Monitoring Agency/RDMRDI Agency/RDMRDI Monitoring

Team Consultant PPTA /Design

Rayon Level LAR Team

Local Government (Village/Sacrebolu)

Affected People

13

J. COMPLAINTS AND GRIEVANCES

34. A grievance mechanism will be available to allow an AP appealing any disagreeable decision, practice or activity arising from land or other assets compensation. APs will be fully informed of their rights and of the procedures for addressing complaints whether verbally or in writing during consultation, survey, and time of compensation. Care will always be taken to prevent grievances rather than going through a redress process. This can be obtained through careful LAR design and implementation, by ensuring full participation and consultation with the APs, and by establishing extensive communication and coordination between the affected communities, the EA, and local governments in general. Complaint & Grievances will be addressed through the process described below in Table 3.

Table 3: Grievance Resolution Process Land/ Crops Compensation Issues 1. First, complaints resolution will be attempted at village level with the involvement of village authorities and Rayon level LAR team at the sacrebolu level. 2. If the grievance is not solved at the sacrebolu level, then the AP will lodge the written complaint at the regional level RDMRDI. If after the regional level RDMRDI intervention no solution has been reached a grievance can be directly lodged to DRD/RDMRDI. The AP must lodge the complaint within 2 weeks after receiving response on the original complaint from the regional level RDMRDI and must produce documents supporting his/her claim. 3. The DRD/RDMRDI at central level will provide a response within 2 weeks of registering the complaint. The DRD/RDMRDI decision must be in compliance with this LARF provisions. 4. Should the grievance redress system fail to satisfy the AP, they can pursue further action by submitting their case to the appropriate court of law (Rayon Court).

K. MONITORING AND EVALUATION

Land acquisition and resettlement tasks under the Program will be subjected to monitoring. Monitoring will be the responsibility of RDMRDI. Internal monitoring will be carried out routinely by DRD/RDMRDI. The results will be communicated to ADB through the quarterly project implementation reports. Indicators for the internal monitoring will be those related to process and immediate outputs and results. This information will be collected directly from regional RDMRDI level and reported monthly to DRD/RDMRDI to assess the progress and results of LARP implementation, and to adjust the work program, if necessary. The monthly reports will be quarterly consolidated in the standard supervision reports to ADB. Specific monitoring benchmarks will be: (i) Information campaign and consultation with APs; (ii) Status of land acquisition and payments on land compensation; (iii) Compensation for affected structures and other assets; (iv) Relocation of APs; (v) Payments for loss of income; (vi) Selection and distribution of replacement land areas; and (vii) Income restoration activities

35. The above information will be collected by DRD/RDMRDI which is responsible for monitoring the day-to-day resettlement activities of the project through the following instruments:

a) Review of census information for all APs; b) Consultation and informal interviews with APs; c) In-depth case studies; d) Sample survey of APs; e) Key informant interviews; and f) Community public meetings.

14

36. External monitoring will be carried out twice a year, and its results communicated to DRD/RDMRDI and ADB through semi-annual reports. Subprojects whose implementation time frame will be under six months will be monitored only once. RDMRDI will need external help for the monitoring of first tranche project LAR activities. However, the external monitoring for the following tranches will be carried out by the supervision consultant. The external monitoring will also assess the status of project affected vulnerable groups such as female-headed households, disabled/elderly and families below the poverty line. Indicators for External Monitoring tasks will include:

(i) Review and verify internal monitoring reports prepared by DRD/RDMRDI; (ii) Review of the socio-economic baseline census information of pre-displaced persons; (iii) Identification and selection of impact indicators; (iv) Impact assessment through formal and informal surveys with the affected persons; (v) Consultation with APs, officials, community leaders for preparing review report; and (vi) Assess the resettlement efficiency, effectiveness, impact and sustainability, drawing lessons for future resettlement policy formulation and planning.

37. The RDMRDI will carry out a post-implementation evaluation of each LARP about a year after completion of its implementation. The compelling reason for this study is to find out if the objectives of the LARP have been attained or not. The benchmark data of socioeconomic survey of severely affected APs conducted during the preparation of the LARP and Poverty Social Assessment (PSA) will be used to compare the pre and post project conditions. The post- implementation evaluation will recommend appropriate supplemental assistance for the APs should the outcome of the study show that the objectives of the LARP have not been attained. The following will be considered as the basis for indicators in post implementation evaluation of the project:

(i) Socio-economic conditions of the APs in the post-resettlement period; (ii) Communications and reactions from APs on entitlements, compensation, options, alternative developments and relocation timetables etc.; (iii) Changes in housing and income levels; (iv) Rehabilitation of informal settlers; (v) Valuation of property; (vi) Grievance procedures; (vii) Disbursement of compensation; and (viii) Level of satisfaction of APs in the post resettlement period.

L. CAPACITY BUILDING AND TRAINING IN LARP IMPLEMENTATION

38. To allow an effective execution of all LAR related tasks some expansion of the capacity on LAR currently available at DRD/RDMRDI may be needed. As soon as the project will become effective RDMRDI will carry out a capacity need assessment and will define the capacity building activities and if needed the additional experts required. Financing for these capacity building initiatives will be included under the capacity building component of the Project.

39. Capacity building exercise will be needed at the Rayon level. The designated team at the Rayon level for the LAR activities will be provided with the capacity enhancement related to the LAR planning and implementation and will be made familiar with ADB’s policy on Involuntary Resettlement. This will be done through on the job training and by

15

participation with the LAR activities. The consultants responsible for the LAR activities in various stages, i.e., feasibility study, detailed design and supervision will work closely with the Rayon level LAR team. A continuous consultation process and informal training through discussion with the Rayon level LAR team and its due involvement in the planning and implementation of LAR activities will be facilitated by the consultant’s team.

40. All concerned staff both at DRD, Rayon and field level involved in LAR activities will undergo a week-long orientation and training in ADB resettlement policy and management. At the very beginning of Project implementation the training will be provided by a consultant hired by RDMRDI, later it will be provided by DRD. Training will cover the following topics:

i). Principles and procedures of land acquisition; ii). Public consultation and participation; iii). Entitlements and compensation & assistance disbursement mechanisms; iv). Grievance redress; and v). Monitoring of resettlement operations.

M. RESETTLEMENT BUDGET AND FINANCING

41. All RP preparation and implementation costs, including cost of compensation and LAR administration, will be considered an integral part of Project cost and will be contributed as a counterpart fund by the Georgia Government, in particular RDMRDI. Each LARP will include a budget section indicating (i) unit compensation rates for all affected items and allowances, (ii) methodology followed for the computation of unit compensation rates, and (iii) a cost table for all compensation expenses including administrative costs and contingencies. Costs for external monitoring tasks and for the preparation of surveys and LARPs can be allocated under the loan.

42. Being the project owner, RDMRDI is responsible for the timely allocation of the funds needed to implement the RPs. Allocations will be reviewed twice a year based on the budget requirements indicated by the LARPs. As per the LAR finances flow the budget for compensation and rehabilitation will be directly disbursed by RDMRDI to the AP.

N. LARP IMPLEMENTATION PROCESS

43. Based on experience in Georgia the Preparation and implementation of a LARP may take up to a few months. The basic LAR-related steps for the preparation and implementation of a LARP are summarized on Box 1 below.

BOX 1: LAR TASKS PROCESS

Step Action Responsibility A) LARP PREPARATION 1 Assessment of Project’s Poverty and Social Impacts RDMRDI Finalization of Detailed design 2 Design consultants

Prepare surveys forms for Census and DMS, train local Census and DMS teams, and establish 3 Consultant coordination with relevant local government agencies. 4 Collection of cadastral and land parcel maps of the project area Consultants/DRD/NAPR Verify land records in affected areas, update cadastral maps and carry out impacts and valuation surveys Consultant/DRD/Rayon level 5 – Detailed Measurement Survey (DMS) LAR Team/ Consultant/DRD/Rayon Level 7 Conduct public consultations LAR Team

16

Consultant/DRD/Regional 8 Negotiations with APs RDMRDI 9 Integrate data from Census into the RP. Consultant 10 Submission of LARP to RDMRDI, local governments and ADB for approval. Consultant/DRD/ADB 11 Disclosure Consultant/DRD B) LARP Finalization (Detailed Design) 1 Detailed final alignment fixed Design Consultant Consultant/ Regional RDMRDI/ 2 Review of impacts and AP lists based on detailed design DRD/Rayon Level LAR Team Consultant/ Regional RDMRDI/ 3 Review of prices based on the updated rate DRD/Rayon Level LAR Team Consultant/ Regional RDMRDI/ 4 Legalization of legalizable DRD/Rayon Level LAR Team 5 Preparation of the final LARP Consultant/DRD 6 ADB Approval ADB RP disclosure: Distribution of RP and information pamphlets in Georgia in the affected communities; DRD/Consultant/Rayon Level 7 posting of RP in English on the ADB website LAR team/ADB 8 Signing of civil contract award ADB/RDMRDI C) RP IMPLEMENTATION 1 Approval of Contract awards Signing ADB 2 Detailed Schedule for compensation action plan DRD/RDMRDI RDMRDI/ DRD/Regional 3 Distribution of Relocation Notices to APs RDMRDI/Rayon Level LAR team RDMRDI/DRD/Regional Level 4 Award of Cheques for Land Compensation RDMRDI RDMRDI/DRD/ Regional Level 5 Award of Cheques for other Compensation & Assistance/ Rehabilitation RDMRDI 6 Demolishing/ Relocation of Affected Structures/Assets RDMRDI 7 Review of RP Implementation through a compliance report RDMRDI/DRD/IMA/ADB 8 If RP Implementation found satisfactory, Notice to proceed for Civil works is issued ADB/RDMRDI D) POST-IMPLEMENTATION TASKS RDMRDI (through External 1 Independent evaluation of LARP Help) E) CYCLICAL/CONTINUOUS TASKS 1 Internal monitoring. Quarterly reporting to ADB DRD RDMRDI/Supervision 2 External Monitoring. Semi-annual reporting to ADB Consultant DRD/Regional level 3 Grievances Redress/Law Suites RDMRDI/Court RDMRDI/DRD/Regional Level 4 Inter-agency coordination and Communication with AP RDMRDI

17

ANNEX A

SYNOPSIS OF SELECTED GEORGIA LAWS AND REGULATIONS ON LAR

A. The Constitution of Georgia

The Constitution determines the essence of private ownership and defines presumption of inviolability however also regulates the issues related to compensation and expropriation of land and immovable property for necessary public need. The Constitution of Georgia ensures the publicity of information. Pursuant to the Article 21 of the Constitution of Georgia “the right of ownership and inheritance is declared and secured”. No body is eligible cancel the universal right of ownership and legacy. Throughout of the necessary public need or if the urgent necessity has emerged, the Article 21.3 of the Constitution allows the expropriation of the private ownership however, only according to the Court Decision or under the rules identified in the organic law2 on basis of the appropriate and fair reimbursement. Other articles of the Constitution of Georgia also create legislative basis in respect with Land Acquisition and resettlement issues. These considers the State actions for expropriation of land for urgent public need, i.e. exercising the right of expropriation (power of eminent domain), also information disclosure and public consultations, protection of cultural heritage and grievance redress related to land acquisition and resettlement of population. The Constitution ensures the right of a citizen to live in safe and healthy environment and use natural and cultural environment. The State undertakes environment protection measures to secure safe environment for people. People have the right to obtain "full, true and timely information" in regard with their work place and residential environment. The Article 42 of the Constitution makes the citizens eligible to claim, in particular protects them and encourages appealing to the court for protection of their rights and freedom.

B. Civil Code of Georgia The Civil Code of Georgia regulates private civil relationships and evolves property rights, obligations, family law and the law on inheritance. Those regulations of the Civil Code that describes ownership right to property and considers right to build, servitude and other rights directly apply to the given project. Ownership Rights The ownership right entitles its beneficiary to freely possess and use property. It may be limited under the legislation or contractual basis. Ownership on the land parcel gives implicit right to land owner to implement construction activities if it is not restricted by any agreement or law. Alienation of real property is not limited under the Civil Code of Georgia. Pursuant to the article 183 of the Civil Code of Georgia “in order to purchase real property the agreement shall be made in a written format and the ownership right shall be registered on the name of the buyer at the Public Register”. The agreement - based on which one person purchases and the other sells the real property may be notarized. The agreement also may be proved by the person identified under the law (Article 69 of the Civil Code of Georgia). Presently the agreement of sales transaction of real property may be proved by the representative of the Public Register. The presumption of veracity and completeness of entries operate with respect to the Public Register, pursuant to the paragraph 1, Article 312 “an entry in the Public Register shall be deemed to be accurate until its inaccuracy is proven Right to Build. The owner is allowed to transfer a land plot to another person in temporary usage (not to exceed 59 yeMDFG) for charge or free of charge. The transferee obtains the right to build a building/construction on or under the land plot, as well as to assign and transfer this right under inheritance or tenancy, borrowing or renting. The construction right may cover such part of a land plot that is not necessary for the actual construction but allows a better use of the facility constructed on the basis of the construction permit. Termination of the construction right requires

2 In the hierarchy of the laws the Organic Law stands between the Constitution and other laws, what highlights the significant importance of the latest.

18 consent from the landowner. Based on the Article 180 of this Code, if a land parcel lacks the access to public roads that are necessary for its adequate use, the other owner may claim from a neighbour to tolerate the use of his land parcel by the owner for the purpose of providing the necessary access. The mentioned article may be used for road construction, though the determination of necessary right of way is rather complicated procedure and in case of road construction evolves the obligations to prove the existence of the elements of such rights. In case of necessary right of way, the implementer of road project shall have the right to undertake road construction notwithstanding the owner’s will. Right of Servitude. This right shall also be noted that according to the Civil Code of Georgia means the restriction imposed onto a land parcel or real property in favor of the owner (beneficiary) of another land parcel or real property. The Beneficiary is granted the right to use land parcel under restriction with some conditions and /or restricts undertaking specific activities or prohibits land owner to exercise some rights against this land parcel. However, in regard with this project, terms and conditions for transfer of any right (among them ownership, construction, necessary right of way or servitude) shall be defined against each land parcel in accordance to the identified rule and on the basis of the agreement entered between the land owner and the party holding appropriate right to act so. This agreement shall be registered at the Public Register.

C. Law of Georgia on the Protection of Cultural Heritage In addition to the Constitution of Georgia affirming the State's obligation to protect cultural heritage and requiring each citizen to care for, protect and preserve cultural heritage the Law of Georgia on Protection of Cultural Heritage defines the legislative principals for protection of existing cultural heritages in Georgia. According to the Law, State protection of cultural heritage is undertaken by the Ministry of Culture, Monuments Protection and Sport, Ministry of Justice of Georgia, local self-government bodies, as well as other State Institutions, Public and Legal Persons of Private Law; On the territories of Abkhazia and Adjara autonomous republics the corresponding bodies of Abkhazia and Adjara autonomous republics within the scope of authority defined under the legislation of Georgia. It is worth to be mentioned that the State and local self-government bodies exercise their authorities in the sphere of protection of cultural heritage in accordance to the Constitutional Agreement between the State and the Orthodox Church of Georgia. The Ministry of Culture, Monuments Protection and Sports of Georgia provides general coordination and manages the activities undertaken in this sphere. In respect with the ownership rights, the Law identifies some differentiations. Namely, the alienation of the State-owned land parcel - located within the zone of State-owned monument, considered as cultural valuables, or located within archeologically protected area - with the right to possess and use the Law considers the agreement with the Ministry of Culture, Monuments Protection and Sports with the terms and conditions of protecting and care being identified ahead. On the other hand, the Law directly restricts alienation of the monuments under private ownership that can only be alienated under the right to possess, and use and with the terms and conditions to care-and protect.

D. The Law of Georgia on Notary Actions The stated law defines the types of notary actions and rule of their implementation. Also the law defines which insittutions and authorized persons except the Notaries have right to conduct Notary actions within the territory of Georgia and beyond it. On the basis of the Article 42 of the Law the local self-governments have right to implement Notary actions related to inheritance, accuracy of the copy to the original document, proving the fact that a citizen is alive, proving the fact of a citizen’s ceratin location. Rural population often apply to local self-governments to condact certain notary actions. Especially, when it is required to identify a person and a document, or the notary actions are required to replace the deceased head of the household with a new member. This rule is often utilized in regsitration of the land parcels when as the owner of the land parcel the other member of the household is registered in place of the deceased member. The representatives of the Consulates of Georgia (consuls) also other key persons at the Consulates are eligible to

19 conduct notary actions on behalf of the State of Georgia beyond the territory of the country. (Article 43). Citizens being abroad may apply to the Consulate of Georgia in the county of their location.

E. The Law of Georgia on Privatization of State-owned Agricultural Land The Law regulates the privatization of State-owned agricultural land. On the basis of this law the leased or non leased State-owned agricultural land subject to privatization. However, the categories of agricultural lands listed below do not subject to privatization: - grazing lands except the grazing lands which before law enactment were leased; grazing lands attached to existing structures being under ownership of legal and/or physical persons or state ownership in accordance to the rule refined by the Law; - Cattle-driving routs; - water fund land, except fish breeding artificial ponds and the lands of common water use category utilized as agricultural lands in accordance to the Law of Georgia on Water. - Forest fund land used under agricultural designation; - Recreation lands; - Lands allocated to Historical monuments, nature and religious monuments; - Land of protected areas; - Agricultural lands assigned as reform land in Adjara Autonomous republic; - Agricultural lands being used by Budgetary Institutions and legal entities of public law in the form of usufruct. Privatization of the two categories (forest fund and recreation land) of agricultural land is still allowed, although only for development of resort-recreation infrastructure in accordance to the decision of the Government of Georgia. F. The Law of Georgia on Ownership to Agricultural Land The current law is completely different from the initial version adopted in 1996. The changes made in this law in different times (among them the amendments on the basis of the Law # 389 as of July 14, 2000) have significantly changed its initial format and simplified to maximum extend the procedures regulated by this Law. Article 3.1. Defines that "a land parcel with or without household structure that is registered at the public register and used for cattle-breeding and plant cultivation produces is considered as an agricultural land parcel" with existing household and additional structures or without them. Also the share of a member of household community within the shared hay fields, grazing lands or forestry areas and the part of the agricultural land that may be the object "of separate ownership right" (Article 3.2). The same Law determines that the ownership right to agricultural land is granted to the State, citizen of Georgia, household (komli) and legal person registered in accordance to the legislation of Georgia, which carries out his activities in agricultural sphere. Besides, the Law declares the State, private and community ownership right to grazing lands in the high mountain regions (Article 43). Also, according to the limitations determined under this law, a foreigner and legal person registered abroad held ownership right only to the bequeathed agricultural land parcels and foreigners also in case when as citizens of Georgia they reasonably possessed agricultural land parcels. Besides, it is worth to be considered that foreigners and legal persons registered abroad are obliged to alienate privately owned agricultural land parcels to the citizen of Georgia, Komli and/or legal person registered in Georgia according to the legislation of Georgia within six months period since they obtained private ownership to the given land parcel. Besides, in case this legal requirement is neglected, private ownership to the agricultural land parcels privately owned by foreigners and legal persons registered abroad shall be taken away under the Court Decision and in return of due compensation. (The standards identified by the Law of Georgia on the Rules for Expropriation of Ownership for Necessary Public Need shall also be exercised in given case). According to Articles 6 and 8, acquisition of agricultural land is allowed on the basis of ordinary rules and general restrictions. Ordinary rule considers land alienation without any permits and other limitations, and general restrictions consider land alienation only on the basis of the consent of co- owner of shared property. In case of agricultural land acquisition the lessee has the priority right to

20 purchase the land. (Article10). Alienation is restricted if the area after this action will be less then 5 hectare (Article 9). The Law defines Tax sanctions if land has not been cultivated for 2 yeMDFG and for non-payment of land tax and non transmission to the other person in lease condition. In such cases the law does not directly state any type of penalty and only refers that in described cases shall be exercised the sanctions under the Tax legislation (Article 20).

G. Law of Georgia on Registration of Rights to Immovable Property The Law defines the rules, terms and conditions for registration of rights to immovable property (things), rights and obligations of the subjects participating in registration procedures. The goal of this Law is to declare and verify ownership rights for immovable property (things) within the territory of Georgia (so as some other rights out of subjective, guarantee and liability relationships) through registration of these rights into the Public Register. The Law describes the rules set for organization and functioning of Public Register. Pursuant to the Law ownership right to real property, mortgage, right to build, usufruct, servitude, lease, sub-lease, rent, sub-rent, lending subject to registration (Article 13.2). This law ensures successful process of expropriation and obtaining of necessary right of way since in case of purchasing immovable property from an owner, it is required that land and real property is registered into the public register to provide legal validity to the sales agreement. Pursuant to the active legislation of Georgia, acquisition (purchase) of private property is legally valid and ownership rights are declared only after its registration into the Public Registry.

G. The Law of Georgia on the Rules for Expropriation of Ownership for Necessary Public Need

The Law of Georgia of “Rules for Expropriation of Ownership for Necessary Public Need” (July, 23, 1999) specified the expropriation procedures, liabilities and implements the rules. The Rules for Expropriation of the ownership which has a possible usage in Georgia in some occasions including Road construction will be issued Regional (civil) Court verdict on the basis of the Presidential Decree. According to the Constitution of Georgia the expropriation of the property in the process of construction of the magisterial motor road is permitted for necessary public need. The process of Expropriation is to be undertaken only by the Court Decision of the President of Georgia and corresponding instances. Expropriation is undertaking by means of compensation payment which is to be corresponding to market value and honest. The State of Georgia has a constitutional right to carry out acquisition property through Expropriation instead of payment of legislative compensation (Clause 21).

H. Procedural Civil Code of Georgia The general courts of Georgia consider the cases according to the rules identified under the Procedural Civil Code of Georgia. The requirements of the procedural law are exercised during the lawsuit, during implementation of separate procedural actions or execution of the court decision. The Procedural Civil Code of Georgia also regulates those cases when determination of the defendant is impossible. This may be important for the Project in the cases when the landowner is not found and correspondingly ownership to his/her land parcel cannot be obtained in legally valid manner, i.e. it is impossible to enter corresponding agreement with the landowner or him/her cannot sign other type of document. The above-listed laws and regulations give the possibility of applying the following three mechanisms for legal application of the property rights: Obtaining the road right of way without expropriation through the payment of due compensation (on the basis of negotiations or a court decision) prior to commencement of the activities; • Expropriation which gives the possibility of obtaining permanent right to land and/or necessary road on the basis of Eminent Domain Law or a court decision through the payment of due compensation;

21

• Expropriation of private properties for urgent public necessity, which gives the possibility of obtaining permanent rights on land and/or necessary road for the purpose of national security or accident prevention. Expropriation is to be made on the basis of the Presidential Decree on Expropriation through the payment of due compensation to affected people. If applied adequately the above listed mechanisms can ensure the appropriate consideration of lawful interests of all parties and the due observation of the existing legislations.

ANNEX B

SYNOPSIS ON ADB POLICY ON INVOLUNTARY RESETTLEMENT

• Involuntary resettlement should be avoided whenever feasible.

• Where population displacement is unavoidable, it should be minimized by providing viable livelihood options.

• Replacing what is lost. If individuals or a community must lose all or part of their land, means of livelihood, or social support systems, so that a project might proceed, they will be compensated and assisted through replacement of land, housing, infrastructure, resources, income sources, and services, in cash or kind, so that their economic and social circumstances will be at least restored to the preproject level.3 All compensation is based on the principle of replacement cost.

• Each involuntary resettlement is conceived and executed as part of a development project or program.4 ADB and executing agencies or project sponsors, during project preparation, assess opportunities for affected people to share project benefits. The affected people need to be provided with sufficient resources and opportunities to reestablish their livelihoods and homes as soon as possible, with time-bound action in coordination with the civil works.

• The affected people are to be fully informed and closely consulted. Affected people are to be consulted on compensation and/or resettlement options, including relocation sites, and socioeconomic rehabilitation. Pertinent resettlement information is to be disclosed to the affected people at key points, and specific opportunities provided for them to participate in choosing, planning, and implementation options. Grievance redress mechanisms for affected people are to be established. Where adversely affected people are particularly vulnerable groups, resettlement planning decisions will be preceded by a social preparation phase to enhance their participation in negotiation, planning, and implementation.

• Social and cultural institutions. Institutions of the affected people, and, where relevant, of their hosts, are to be protected and supported. Affected people are to be assisted to integrate economically and socially into host communities so that adverse impacts on the host communities are minimized and social harmony is promoted.

• No formal title. Indigenous groups, ethnic minorities, pastoralists, people who claim for such land without formal legal rights, and others, who may have usufruct or customary rights to affected land or other resources, often have no formal legal title to their lands. The absence of a formal legal title to land is not a bar to ADB policy entitlements.

3 If the residual of an asset taken is not economically viable, compensation and other assistance are provided as for the entire asset. In this case, affected people have the option to retain their assets. Nonland based options may be used where land is not the preferred option of the affected people; or where land of similar quality and quantity is not available. 4 ADB may treat resettlement either as part of the main investment or as a free-standing resettlement project that is prepared, financed, and implemented in association with the main investment.

22

• Identification. Affected people are to be identified and recorded as early as possible in order to establish their eligibility through a population record or census that serves as an eligibility cutoff date, preferably at the project identification stage, to prevent a subsequent influx of encroachers or others who wish to take advantage of such benefits.5

• The Poorest. Particular attention must be paid to the needs of the poorest affected people,6 and vulnerable groups that may be at high risk of impoverishment. This may include those without legal title to land or other assets, households headed by females, the elderly or disabled and other vulnerable groups, particularly indigenous peoples.7 Appropriate assistance must be provided to help them improve their socio-economic status.

• The full resettlement costs are to be included in the presentation of project costs and benefits. This includes costs of compensation, relocation and rehabilitation, social preparation and livelihood programs as well as the incremental benefits over the without- project situation (which are included in the presentation of project costs and benefits). The budget also includes costs for planning, management, supervision, monitoring and evaluation, land taxes, land fees, and physical and price contingencies. Where loans include subprojects, components or investments prepared only after project approval and loans through financial intermediaries that are likely to cause involuntary resettlement, sufficient contingency allowance must be allocated for resettlement prior to approval of the loan. Similarly, resettlement plans should also reflect the timeframe for resettlement planning and implementation.

• Eligible8 costs of compensation. Relocation and rehabilitation may be considered for inclusion in ADB loan financing for the project, if requested, to assure timely availability of the required resources and to ensure compliance with involuntary resettlement procedures during implementation.

5 An eligibility cutoff date should be established as soon as possible in the project cycle. 6 The resettlement planning documents will, in each case, define the poorest and vulnerable groups, using, as appropriate, the poverty line as defined in the poverty partnership agreement with the DMC concerned, or other accepted ADB documents. A range of other documents may also provide information on poverty in the project area. 7 When significant indigenous peoples or ethnic minority issues are identified, special attention will be paid to exploring viable alternative designs that will reduce or eliminate such impacts. An Indigenous Peoples Development Plan may be required in addition to a resettlement plan. If the indigenous people issues are judged to be less than significant, specified “indigenous people actions” within the resettlement plan may suffice to meet the indigenous people policy objectives. 8 Involuntary resettlement costs eligible for loan financing may include, for example, income restoration, relocation, site development, social preparation, monitoring, and evaluation.

23

ANNEX C

OUTLINE OF A STANDARD LARP

1. LAR issues for the scheme This chapter describes the scheme activities and items requiring LAR; alternative options, if any, considered to minimize land acquisition and its effects; and why remaining effects are unavoidable.

2. Scope of land acquisition and resettlement This chapter describes the preparation of the impacts (who carried it out and when it was initiated) and provides a full assessment of each type of impact and a census of affected peoples as described in the LARF. The chapter also includes a description of the methodology followed to determine unit-compensation rates for each affected item and subsidy/allowance.

3. Objectives, policy framework, and entitlements Based on the LARF, this chapter outlines the eligibility and compensation framework for the scheme.

4. Consultation and grievance redress participation This chapter summarizes procedures for redress of grievances by people affected described in LARF and describes the consultation/participation process and grievance redress that occurred in the subproject at hand.

5. Compensation, relocation, and income restoration This chapter outlines the income restoration measures to be implemented.

6. Institutional framework This chapter outlines the institutional arrangements for the scheme based on this LARF. It includes the following issues: responsibilities for main tasks and for planning, negotiating, consulting, approving, coordinating, implementing, financing, monitoring, and evaluating land acquisition and resettlement.

7. Resettlement budget and financing This chapter provides the unit compensation rate for each affected item and assesses the LAR budget for the scheme. The LAR budget will include land acquisition and eventual land acquisition costs, amounts due for crop compensation and for the subsidies and allowances, monitoring and evaluation costs, and administrative costs and will be adjusted for inflation and applicable taxes.

8. Implementation schedule This chapter provides a time schedule showing the LAR process and linking LAR tasks with civil works implementation.

9. Monitoring and evaluation This chapter specifies arrangements for routine and independent monitoring and evaluation activities.