Encrypted D3.5 SYNER-G Fragility Functions for Water Waste Water

D 3.5

DELIVERABLE

PROJECT INFORMATION

Systemic Seismic Vulnerability and Risk Analysis for Project Title: Buildings, Lifeline Networks and Infrastructures Safety Gain

Acronym: SYNER-G

Project N°: 244061

Call N°: FP7-ENV-2009-1

Project start: 01 November 2009

Project end: 31 October 2012

DELIVERABLE INFORMATION

D3.5 - Fragility functions for water and waste-water Deliverable Title: system elements

Date of issue: 31 October 2010

Work Package: WP3 – Fragility functions of elements at risk

Deliverable/Task Leader: Aristotle University of Thessaloniki (AUTH)

Reviewer: Norwegian Geotechnical Institute (NGI)

REVISION: Final

Project Coordinator: Prof. Kyriazis Pitilakis Institution: Aristotle University of Thessaloniki e-mail: [email protected] fax: + 30 2310 995619 telephone: + 30 2310 995693

Abstract

This deliverable provides the technical report on the assessment of fragility functions for water and waste-water system elements. This deliverable comprises four main parts. A short review of past earthquake damages on water and waste-water system elements is provided in the first part, including the description of physical damages, the identification of main causes of damage and the classification of failure modes. The following two parts deal with the identification of the main typologies of water and waste-water system components and the general description of existing methodologies, damage states definitions, intensity indexes and performance indicators of the elements. In the next part the validation of the available vulnerability functions for pipes is provided based on damage data from recent European earthquakes (Düzce and Lefkas). Finally, improved vulnerability functions for the individual components are proposed along with the coding and digital description of fragility functions.

Keywords: fragility functions, vulnerability, water system, waste-water system

Acknowledgments

The research leading to these results has received funding from the European Community's Seventh Framework Programme [FP7/2007-2013] under grant agreement n° 244061.

iii

Deliverable Contributors

[AUTH] Alexoudi Maria, Dr. Civil Engineer, MSc Kyriazis Pitilakis, Professor Argyro Souli, Civil Engineer, MSc

Table of Contents

1 Introduction...... 1 2 Damages from past earthquakes ...... 3 2.1 PHYSICAL DAMAGES / MAIN CAUSES OF DAMAGE OF WATER SYSTEM ELEMENTS...... 3 2.1.1 Tanks ...... 6 2.1.2 Water Treatment Plant...... 7 2.1.3 Canals...... 7 2.1.4 Tunnels ...... 8 2.1.5 Pipes...... 9 2.1.6 Pumping Stations...... 9 2.2 CLASSIFICATION OF FAILURE MODES / DIRECT LOSSES OF WATER SYSTEM ELEMENTS...... 9 2.2.1 Pipes...... 9 2.3 PHYSICAL DAMAGES / MAIN CAUSES OF DAMAGE OF WASTE-WATER SYSTEM ELEMENTS...... 13 2.3.1 Waste-Water Treatment Plant ...... 15 2.3.2 Tunnels ...... 15 2.3.3 Pipes...... 15 2.3.4 Lift Station...... 15 3 Methodology for the vulnerability assessment of water and waste-water system elements...... 16 3.1 IDENTIFICATION OF THE MAIN TYPOLOGIES OF WATER SYSTEM ELEMENTS...... 16 3.1.1 Water Source...... 17 3.1.2 Water Treatment Plant...... 17 3.1.3 Pumping Station ...... 18 3.1.4 Storage ...... 19 3.1.5 Supervisory Control and Data Acquisition (SCADA) ...... 20 3.1.6 Conduits...... 21 3.2 SYNER-G TYPOLOGIES OF WATER SYSTEM ELEMENTS ...... 26 3.3 IDENTIFICATION OF THE MAIN TYPOLOGIES OF WASTE-WATER SYSTEM ELEMENTS...... 28 3.3.1 Conduits...... 29

vii

3.3.2 Waste-water Treatment Plant ...... 30 3.3.3 Lift station ...... 31 3.3.4 Supervisory Control And Data Acquisition (SCADA) ...... 32 3.4 SYNER-G TYPOLOGIES OF WASTE-WATER SYSTEM ELEMENTS ...... 33 3.5 GENERAL DESCRIPTION OF EXISTING METHODOLOGIES...... 34 3.6 DAMAGE STATES OF WATER SYSTEM ELEMENTS ...... 34 3.6.1 Water Source...... 34 3.6.2 Water Treatment Plant...... 35 3.6.3 Pumping Station ...... 35 3.6.4 Storage tanks...... 35 3.6.5 Canal ...... 35 3.6.6 Pipes...... 35 3.6.7 Tunnels ...... 36 3.7 DAMAGE STATES OF WASTE-WATER SYSTEM ELEMENTS ...... 36 3.7.1 Waste-Water Treatment Plant ...... 36 3.7.2 Conduits...... 36 3.7.3 Lift station ...... 36 3.8 INTENSITY INDEXES ...... 36 3.8.1 Water System Elements ...... 37 3.8.2 Waste-Water System Elements...... 39 3.9 PERFORMANCE INDICATORS...... 39 3.9.1 Water System/ component performance indicators...... 40 3.9.2 Waste-Water System/ component performance indicators...... 45 4 Fragility functions for water and waste-water system elements...... 49 4.1 STATE-OF-THE-ART FRAGILITY CURVES PER COMPONENT OF WATER SYSTEM ...... 49 4.2 STATE-OF-THE-ART FRAGILITY CURVES PER COMPONENT OF WASTE- WATER SYSTEM ...... 61 4.3 VALIDATION / ADAPTATION / IMPROVEMENT...... 63 4.3.1 Validation of vulnerability models for pipes...... 64 4.4 FINAL PROPOSAL ...... 77 4.4.1 WATER SYSTEM ELEMENTS...... 77 4.4.2 WASTE-WATER SYSTEM ELEMENTS...... 91 5 Coding and digital description of fragility functions...... 98

viii

List of Figures

Fig. 2-1 Tank failure in Izmit (Kocaeli 《arthquake,1999)...... 5 Fig. 2-2 Water pipe failure of north part of Anatolian fault (3.7 m- Kocaeli 《arthquake, 1999)...... 5 Fig. 2-3 Failure of Rinconada Water Treatment Plant (Loma Prieta, 1989)...... 5 Fig. 2-4 Total collapse of 750,000-gallon tank near Castaic Junction (Northridge, 1994) ...... 5 Fig. 2-5 Different types of seismic response of pile foundation tanks (ASCE, 1997)...... 6 Fig. 2-6 Failure modes of segment pipes for wave propagation (O’Rourke and Liu, 1999).. 10 Fig. 2-7 Basic failure modes for ductile pipes ...... 11 Fig. 2-8 Failures modes of pipelines as result of liquefaction (O’Rourke and Palmer, 1996) 12 Fig. 2-9 Failures modes of pipelines as result of landslide (O’Rourke et al., 1998)...... 12 Fig. 2-10 Failures modes of pipelines as result of fault crossing (O’ Rourke et al., 1998).... 13 Fig. 2-11 Plenary view of waste-water treatment plant of Lefkas (Greece) ...... 14 Fig. 2-12 No damage observed in waste-water lift station during the 2003 Lefkas earthquake in Greece (from in-situ inspection Alexoudi and Argyroudis 2003)...... 15 Fig. 3-1 Breakdown of potable water system components ...... 16 Fig. 3-2 Breakdown of potable water conduits...... 22 Fig. 3-3 Breakdown of waste-water system...... 29 Fig. 3-4 Breakdown of waste-water conduits...... 29 Fig. 4-1 Location of Düzce and Lefkas island ...... 63 Fig. 4-2 Düzce. Analyzed method: 1D linear equivalent, Local Soil Condition: Based on Soil Profiles, a) Earthquake: Kocaeli, 1999, PGA (g) [a(1)], PGV (m/sec) [a(2)], b) Earthquake: Düzce, 1999, PGA (g) [b(1)], PGV (m/sec) [b(2)]...... 65 Fig. 4-3 Mahallas that present low, moderate and extensive failures as result of Kocaeli earthquake and O’Rourke and Ayala (1993) (a) and Eidinger and Avila (1999) (b) relationships. The points represent the well documented damages shown earlier. Earthquake: Kocaeli 1999, Microzonation study (Alexoudi et al. , 2007) ...... 67 Fig. 4-4 Mahallas that present low, moderate and extensive failures as result of Düzce earthquake and O’Rourke and Ayala (1993) (a) and Eidinger and Avila (1999) (b) relationships. The failures collected are illustrated with points. For each mahalla, ID is corresponded. Earthquake: Düzce 1999, Microzonation study. (Alexoudi et al., 2007)...... 67 Fig. 4-5 Mahallas that presents low, moderate and extensive failures (a) before the two earthquakes, (b) after Kocaeli earthquake, (c) after Düzce earthquake (d)

present research as result of both earthquakes. Points illustrate the failures collected while the ID corresponds to each mahalla...... 68 Fig. 4-6 Digitized Waste- Water network (left) in Düzce and distribution of waste-water pipe/ conduits diameters (up)...... 69 Fig. 4-7 Estimated damages of waste-water network as percentage of the total length of the network for Kocaeli (a) and Düzce (b) earthquake (Alexoudi et al., 2008) ...... 69 Fig. 4-8 Spatial distribution of waste-water pipe damages in Düzce network for Kocaeli (a) and Düzce (b) earthquake (Alexoudi et al., 2008)...... 70 Fig. 4-9 Estimated waste-water pipe damages per mahalla for Kocaeli earthquake (a), for Düzce earthquake (c) and recorded water pipe damages per mahalla after Kocaeli earthquake (b) and Düzce earthquake (d) - (Alexoudi et al., 2008)...... 71 Fig. 4-10 Water distribution network of old city of Lefkas and the location of main water system failures and secondary connections (p-primary network, sec-secondary network-connections with customers...... 72 Fig. 4-11 Vulnerability assessment of potable water system (Fragility curve: O’ Rourke and Ayala, 1993, Earthquake: Lefkas 2003) ...... 75 Fig. 4-12 Vulnerability assessment of potable water system (Fragility curve: Eidinger and Avila, 1999, Earthquake: Lefkas 2003)...... 75 Fig. 4-13 Vulnerability assessment of potable water system (Fragility curve: Isoyama et al., 1998, Earthquake: Lefkas 2003) ...... 76 Fig. 4-14 Vulnerability assessment of potable water system (Fragility curve: ]LA, 2001, Earthquake: Lefkas 2003) ...... 76 Fig. 4-15 Fragility curves for wells (Anchored components, low – rise R/C building with low seismic code design) subjected to ground shaking...... 78 Fig. 4-16 Fragility curves for wells (Anchored components, low – rise R/C building with advanced seismic code design) subjected to ground shaking ...... 79 Fig. 4-17 Fragility curves for Water Treatment Plant (Anchored components) subjected to ground shaking...... 82 Fig. 4-18 Fragility curves for pumping station (Anchored components, low-rise R/C building with low seismic code design) subjected to ground shaking ...... 84 Fig. 4-19 Fragility curves for pumping station (Anchored components, low -rise R/C building with advanced seismic code design) subjected to ground shaking ...... 85 Fig. 4-20 Fragility curves for above ground R/C tanks (wave propagation)...... 88 Fig. 4-21 Fragility curves for above ground R/C tanks (permanent ground deformations) ... 88 Fig. 4-22 Fragility curves for Waste- Water Treatment Plant (Anchored components) subjected to ground shaking (low-rise R/C building with low seismic code design)...... 92 Fig. 4-23 Fragility curves for Waste- Water Treatment Plant (Anchored components) subjected to ground shaking (low-rise R/C building with advanced seismic code design)...... 93

Fig. 4-24 Fragility curves for lift station (Anchored components, low-rise R/C building with low seismic code design) subjected to ground shaking...... 95 Fig. 4-25 Fragility curves for lift station (Anchored components, low-rise R/C building with advanced seismic code design) subjected to ground shaking ...... 96

List of Tables

Table 2-1 Brief presentation of water system damages as result of Loma Prieta, Northridge and Hyogo-ken Nanbu (Kobe) earthquake...... 3 Table 2-2 Restoration time of water system and number of customers influenced [1989 Loma Prieta, 1994 Northridge, 1995 Hyogo-ken Nanbu (Kobe) earthquakes].... 6 Table 2-3 Main failure modes of water treatment plants (ASCE, 1987)...... 7 Table 2-4 Tunnel failures for different earthquakes ...... 8 Table 2-5 Possible failure modes for pipes as result of wave propagation...... 10 Table 2-6 Brief presentation of waste-water system damages as result of Loma Prieta, Northridge and Hyogo-ken Nanbu (Kobe) earthquake ...... 14 Table 3-1 Typology of water storage tanks...... 20 Table 3-2 Typology of tunnels (ALA 2001a)...... 25 Table 3-3 Comparison of the typologies of potable water elements provided in NIBS 2004, ALA 2001a,b and SYNER-G ...... 27 Table 3-4 Comparison of the typologies of potable water elements provided in HAZUS (NIBS, 2004) and SYNER-G ...... 33 Table 3-5 Intensity measures for the vulnerability assessment potable water system elements...... 38 Table 3-6 Intensity measures for the vulnerability assessment waste- water system elements ...... 39 Table 3-7 Summary of Water Component Performance Indicators (WCPIs)...... 41 Table 3-8 Summary of Water System Performance Indicators (WSPIs) ...... 42 Table 3-9 Summary of Waste-Water Component Performance Indicators (PPIs)...... 47 Table 3-10 Summary of Waste-Water System Performance Indicators (WWSPIs) – ALA (2004) ...... 47 Table 4-1 Review of existing fragility functions for potable water elements...... 51 Table 4-2 Review of existing fragility functions for waste-water system elements...... 61 Table 4-3 Computed water pipe failures in the water network of Düzce due to ground shaking for different fragility expressions, and input motions (Alexoudi et al., 2010) ...... 66 Table 4-4 Estimated number of repairs for Lefkas earthquake using different fragility curves ...... 73 Table 4-5 Comparison of Repair Rate/km (wave propagation) with the recorded damages of water network of Lefkas...... 74 Table 4-6 Comparison of the number of failures (wave propagation) for water system of Lefkas...... 74

xii

Table 4-7 Description of damage states for water source subject to ground shaking...... 77 Table 4-8 Parameters of fragility curves for water source (wells) ...... 78 Table 4-9 Subcomponent Damage Algorithms for Wells with Anchored Components (SRM- LIFE, 2003-2007)...... 79 Table 4-10 Description of damage states for Water Treatment Plant subjected to ground shaking ...... 80 Table 4-11 Parameters of fragility curves for Water Treatment Plant...... 81 Table 4-12 Subcomponent Damage Algorithms for Water Treatment Plants with Anchored Components ...... 81 Table 4-13 Description of damage states for Pumping Station subjected to ground shaking83 Table 4-14 Parameters of fragility curves for pumping station...... 83 Table 4-15 Subcomponent Damage Algorithms for Water Treatment Plants with Anchored Components ...... 85 Table 4-16 Fragility curves for anchorage R/C at grade tanks (wave propagation)- ALA (2001a,b) ...... 86 Table 4-17 Fragility curves for unanchorage R/C at grade tanks (wave propagation)- ALA (2001a,b) ...... 86 Table 4-18 Fragility curves for Open reservoirs with or without seismic design code (wave propagation) ALA (2001a,b) ...... 87 Table 4-19 Fragility curves for unanchorage R/C at grade tanks (permanent deformations)- ALA (2001a,b) ...... 87 Table 4-20 Fragility curves for at-grade R/C tanks (wave propagation)- (HAZUS; NIBS, 2004) ...... 87 Table 4-21 Fragility curves for buried R/C tanks (permanent ground deformation)- (HAZUS; NIBS, 2004)...... 87 Table 4-22 Description of damage states for Canals (ALA, 2001a,b)...... 89 Table 4-23 Vulnerability of canals (wave propagation, ALA, 2001a, b) ...... 90 Table 4-24 Vulnerability of canals (permanent deformations, ALA, 2001a, b)...... 90 Table 4-25 Description of damage states for Waste-Water Treatment Plant subjected to ground shaking...... 91 Table 4-26 Parameters of fragility curves for Water Treatment Plant...... 92 Table 4-27 Subcomponent Damage Algorithms for Waste- Water Treatment Plants with Anchored Components...... 93 Table 4-28 Description of damage states for Lift Station subjected to ground shaking ...... 94 Table 4-29 Parameters of fragility curves for lift station...... 95 Table 4-30 Subcomponent Damage Algorithms for Lift Station with Anchored Components96

xiii

D3.9 - Fragility functions for water and waste-water system elements

1 Introduction

The present report reviews the damages sustained by water and waste-water system elements during past earthquakes, with special emphasis to European earthquakes. Different failure modes are identified and classified respectively. The following components are proposed to be studied within SYNER-G: For Water System o Water source o Treatment plant o Pumping station o Storage o Supervisory Control and Data Acquisition (SCADA) o Conduits (pipes, tunnel, canals) For Waste-Water System o Conduits (pipes, tunnels) o Treatment plant o Lift station o Supervisory Control and Data Acquisition (SCADA) The description of the European typology for the different components is performed. A review of existing methodologies for the vulnerability assessment of water and waste-water system elements is followed by the definition, for each component, of some key parameters: o Damage states scales. o Intensity index (indices) (intensity-measure parameter). o Performance indicators that can help specify the link between the damage state of the component and its serviceability / functionality. Finally, based on the review of state-of-the-art fragility curves for each component, and the validation of some methods based on damage data from recent European earthquakes, improved fragility functions for the individual components are proposed along with their coding and digital description. For the proposed vulnerability functions, the following parameters are provided: o Typology classification of each component. o Damage scale definition. o Intensity index used. o Fragility curve parameters, for each damage state and each typology.

D3.9 - Fragility functions for water and waste-water system elements

2 Damages from past earthquakes

Water and waste-water systems are prone to damage from earthquakes, not only under severe levels of shaking but under moderate levels as well. Furthermore, as shown by the experience during past earthquakes, seismic damage to water system elements can cause extended direct and indirect economic losses, while environmental pollution is the main result of waste-water failure.

2.1 PHYSICAL DAMAGES / MAIN CAUSES OF DAMAGE OF WATER SYSTEM ELEMENTS

The main damages in water network were observed in water pipes (Table 2-1); secondarily in pumping stations, tanks and water treatment plants. The pipeline damages can be mainly attributed to permanent ground deformation and less to wave propagation. Rigid pipe body, connections, age and corrosion are some of the factors that influence the seismic response of water network.

Table 2-1 Brief presentation of water system damages as result of Loma Prieta, Northridge and Hyogo-ken Nanbu (Kobe) earthquake

Earthquake/ Loma Prieta, 1989, Northridge, 1994, Hyogo-ken Nanbu System Mw=6.9, max. MMI=IX Mw=6.7, max. MMI=IX (Kobe), 1995, Mw=6.9, max. JMA=VII Water The 350 repairs in water More than 1400 damages 86 reservoirs that give System system mains of San were observed in Los water to Kobe were Francisco, Oakland and Angeles water network. empty in 24 hours. The Berkeley were observed in The most of the damages, damages in water cast iron and in asbestos- 100 were observed in network influence the cement pipe with water transmission pipes. operability of fire-fighting diameters 100-200mm. In The three transmission system. 1.610 repairs of Santa Cruz area, 240 systems of San Fernando the main water system failures in the water were broken. The seismic and 71.235 repairs in network were observed, response of dams, water customer’s connection mainly concentrated in drills, pumping station as result of building areas with large was very good although damages and permanent permanent deformations. the electric power was ground deformation were The electric power loss disrupted. Minor damages recorded. Electric Power had a great impact in were observed in water losses were responsible water distribution system treatment plants while for the malfunction of 3 of San Francisco. extensive damages were emergency valves in Moreover, the pipes that recorded in above ground reservoirs and several were damaged influenced tanks with no code pumping stations the reliability of fire- design. fighting system. References EERI (1990), NRC (1994) EERI (1995), TCLEE NIST (1996), NCEER (1995), NIST (1994) (1995) Shrestha (2001)

D3.5 -Fragility functions for water and waste-water system elements

Europe and especially Balkan and Mediterranean countries have experienced several large earthquakes, although limited records are available. In Bucharest earthquake (1977), water system experienced extensive damages both in transmission and distribution system. In the water treatment plant, the pumps in the pumping station were dislocated, causing immediately the periodical stop of the treatment process. In three locations, transmission pipes (diameter: 1200-2200mm, total length: 200km) were severely damaged with extensive breaks in water supply. Moreover, water blow as result of water failures and extensive electric power losses provoke the break of 120 connections in water distribution system. Approximately 10% of 120 connections were steel pipes, while the remaining 90% were concrete pipes. Asbestos-cement and cast iron pipes experienced no damages (Aldea et al., 2002). In Kocaeli (1999) earthquake, minor damages were recorded in the water treatment plant and in the dam while 2 buried tanks of R/C were cracked, 70% of water distribution system in Adapazari (500km) was destroyed while the rest had extensive leaks. Moreover, in the same earthquake, water transmission system had experienced important damages especially in areas close to the surface trace of the fault (Izmit). Extensive damages were occurred in Golcuk area in the water distribution system. About 45% of the water network was destroyed while the rest experienced important leaks. It is important to mention that Turkey’s water systems are very old with extensive water loss even before the earthquake. Greek experience is also limited in lifeline system damages. No major damages have been observed in Greece. Although, failures were observed in pipes in Thessaloniki earthquake

(1978, Mw=6.4, R=29km, PGA= 0.15g, PGV= 16.7cm/sec, PGD=3.4cm). For a few days, water supply in Thessaloniki stopped when the main pumping station was out of order. Water was polluted with oil as a result of an oil pipe break nearby. In Kozani- Grevena earthquake (1995, Mw= 6.6, R= 19km, PGA= 0.21g, PGV= 8.8cm/sec, PGD=1.5cm), the damages were limited and localized. Water supply stopped in the majority of villages, the cause of the supply interruption was never confirmed. Several assumptions were made that include pipeline break and electric power failure. No important damages were occurred in the water system after Kalamata earthquake (1986, Mw= 6.0, R= 12km, PGA= 0.27g, PGV= 32.3cm/sec, PGD=7.2cm). Many times, due to lack of experience, the results of earthquakes are noticed later. The recording of water pipeline failures was performed after the Lefkas earthquake (2003, ‒w=6.4, PGAtrans=0,42g, PGAlong= 0.34g, PGAvert=0.19g). In particular, more than 30 failures in water customer’s connection were recorded and 10 failures in water mains (Alexoudi, 2005). Water system failures were observed in 1989 Loma Prieta, 1994 Northridge and 1999 Kocaeli earthquakes (Fig. 2-1 - Fig. 2-4) The water failures are closely connected with restoration times and number of customers. An important factor of restoration time is the interactions between the systems and the available personnel after the earthquake. After the 1995 Hyogo-ken Nanbu (Kobe) earthquake, restoration lasted 14 days. The restoration and design personnel were 450 people. About 1757 failures of main water system were fixed; in the secondary network the fixed repairs were about 62.300. Restoration time of water system and number of customers influenced as result of 1989 Loma Prieta, 1994 Northridge and 1995 Hyogo-ken Nanbu (Kobe), earthquakes are illustrated in Table 2-2.

D3.9 - Fragility functions for water and waste-water system elements

Fig. 2-1 Tank failure in Izmit (Kocaeli Fig. 2-2 Water pipe failure of north part 《arthquake,1999). of Anatolian fault (3.7 m- Kocaeli 《arthquake, 1999).

Fig. 2-3 Failure of Rinconada Water Fig. 2-4 Total collapse of 750,000-gallon Treatment Plant (Loma Prieta, 1989) tank near Castaic Junction (Northridge, 1994)

D3.5 -Fragility functions for water and waste-water system elements

Table 2-2 Restoration time of water system and number of customers influenced [1989 Loma Prieta, 1994 Northridge, 1995 Hyogo-ken Nanbu (Kobe) earthquakes].

Earthquake/ Loma Prieta, 1989 Northridge, 1994. Hyogo-ken Nanbu Water Kobe, 1995 References: NIST References: NIST 871 system GCR 97-719 (1994), NISTIR 539 (1994), References: NCEER NCEER (1994) (1995), NIST 901 (1996)

Restoration Immediate Max … 7 days 14 days days

N.b of customers 50.000 people 1.300.000 people - influenced

According to international experience, for the repair of a point failure of the main potable water pipes, 3- 6 people are needed for total recovery, while for the rest water system failures about 1.5 people are needed.

2.1.1 Tanks

According to ]SCE (1997) there are six main failure modes for tanks: shell buckling mode, roof and miscellaneous steel damage, anchorage failure, foundation failure, support system failure, hydrodynamic pressure failure, connecting pipe failure and manhole failure. The basic failure modes for tanks under seismic loads are presented in Error! Not a valid link..

Fig. 2-5 Different types of seismic response of pile foundation tanks (ASCE, 1997)

D3.9 - Fragility functions for water and waste-water system elements

2.1.2 Water Treatment Plant

The main failure modes of water treatment plants and its elements are described briefly in Table 2-3.

Table 2-3 Main failure modes of water treatment plants (ASCE, 1987)

2.1.3 Canals

Canal failure is often closely connected to the increased friction between the water and the liner, as the result of debris residue that is lowering hydraulic capacity. Debris may have entered into the canal causing higher sediment transport, which could cause scour of the liner or earthen embankments. Damage to overcrossings may have also occurred. Overcrossing damage could include the collapse of highway bridges and leakage of non- potable material pipelines such as oil, gas, etc. Damage to bridge abutments could cause constriction of the canal's cross-section to such an extent leading to significant flow restriction which warrants immediate shutdown and repair (ALA, 2001a).

D3.5 -Fragility functions for water and waste-water system elements

2.1.4 Tunnels

Ground shaking will induce stresses in the liner system of tunnels. If the level of shaking is high, the liner can crack, which can result in tunnel collapse. For water tunnels, the impact of liner failure may or may not be immediate. Small cracks in liners will not generally directly impact the flow of water through the tunnel, although there may be some minor increases in head loss. Over time, small cracks will allow water from the tunnel to enter the native materials behind the liner, which could cause erosion of the materials and ultimately could lead to more damage to the liner. For the most part, the factors that lead to the major damage state are fault offset through the tunnel itself or landslide at the tunnel portals. Table 2 4 provides tunnel failures depending on the different coating for 10 different earthquakes.

Table 2-4 Tunnel failures for different earthquakes

Coating Earthquake Timber Reinforced (after Power ‒w Or Concrete Concrete Unknown Without Total et al., 1998) Masonry Liner or Steel Liner Pipe Liner San Francisco, CA 7.8 - 1 7 - - 8 (1906) Kanto, Japan 7.9 - 7 4 2 - 13 (1923) Kern Country, 7.4 - 4 - - - 4 CA (1952) Alaska (1964) 8.4 - 8 - - - 8 San Fernando, 6.6 - 8 - - 1 9 CA (1971) Loma Prieta, 7.1 3 - 2 11 6 22 CA (1989) Petrolia, CA 6.9 - - - 11 - 11 (1992) Hokkaido, 7.8 - - - 1 1 Japan (1993) Northridge, CA 6.7 6 - - 5 20 31 (1994) Hyogo-ken Nanbu (Kobe), 6.9 3 - 1 87 6 97 Japan (1995) Sum 204

D3.9 - Fragility functions for water and waste-water system elements

2.1.5 Pipes

Damage to segmented pipes (e.g., cast iron pipe having caulked bell-and-spigot joints) will be heavy when crossing surface ruptured faults according to ALA (2001). Moreover, pipe breaks occur due to relative vertical (differential) settlements at transition zones from fill to better soil, and in areas of alluvial soils prone to localized liquefaction. Breaks can also occur where pipes enter tanks or buildings. Landslides may also produce localized, severe damage to buried pipe. Experience has also shown that welded pipelines with bends, elbows and local eccentricities will concentrate deformation at these features, especially if permanent ground deformations develop compression strains. Segmented pipe with somewhat rigid caulking cannot tolerate much movement before leakage occurs. Pipeline damage tends to concentrate at discontinuities such as pipe elbows, tees, in-line valves, reaction blocks and service connections. Such features create anchor points or rigid locations that promote force/stress concentrations. Locally high stresses can also occur at pipeline connections to adjacent structures (e.g., tanks, buildings and bridges). Age and corrosion will accentuate damage, especially in segmented steel, threaded steel and cast iron pipes. Age effects are possibly strongly correlated with corrosion effects caused by the increasing impact of corrosion over time. Corrosion weakens pipe by decreasing the material’s thickness and by creating stress concentrations.

2.1.6 Pumping Stations

Pumping stations are complex components. Damages in pumping stations are closely connected with the failure modes of their sub-components. The major subcomponents are presented in the next section of this report.

2.2 CLASSIFICATION OF FAILURE MODES / DIRECT LOSSES OF WATER SYSTEM ELEMENTS

In general, water system failure may include damages in all water system components. According to the redundancy and the importance of water elements, the failure of some components has more impact than others. The definition of water system failures is defined according to the operation period of the system (normal, crisis and recovery). Specifically, water system failure can include disability: o To supply the available water and pressure for fire-fighting purposes in the end- point node. o To serve customers’ needs in summer days with maximum daily consumption. o To supply water to all customers independent of the region and the floor.

2.2.1 Pipes

The basic failure modes of pipes are presented in Table 2-5 and in Fig. 2-6 and Fig. 2-7 for the case of wave propagation.

D3.5 -Fragility functions for water and waste-water system elements

Table 2-5 Possible failure modes for pipes as result of wave propagation

Continuous pipes Segmented pipes (Singhal, 1984, O’Rourke ALA (2001a) (O’Rourke and Liu, 1999) and Liu, 1999)

- Tensile failure - Axial pull-out - Axial pull-out - Wrinkling - Crushing of bell and - Joint rotation spigot joints - Beam buckling - Tensile and bending - Joint rotation deformations of the pipe - Welded slip joint barrel. - Round flexural cracks

Fig. 2-6 Failure modes of segment pipes for wave propagation (O’Rourke and Liu, 1999)

D3.9 - Fragility functions for water and waste-water system elements

Fig. 2-7 Basic failure modes for ductile pipes

D3.5 -Fragility functions for water and waste-water system elements

For liquefaction, landslides and fault crossing the pipeline failure modes are illustrated in Fig. 2-8 to Fig. 2-10, respectively.

Fig. 2-8 Failures modes of pipelines as result of liquefaction (O’Rourke and Palmer, 1996)

Fig. 2-9 Failures modes of pipelines as result of landslide (O’Rourke et al., 1998)

D3.9 - Fragility functions for water and waste-water system elements

Fig. 2-10 Failures modes of pipelines as result of fault crossing (O’ Rourke et al., 1998)

2.3 PHYSICAL DAMAGES / MAIN CAUSES OF DAMAGE OF WASTE-WATER SYSTEM ELEMENTS

The main damages in waste-water network were observed in waste-water pipes (Table 2-6); secondarily in lift stations and waste-water treatment plants. The pipeline damages can be attributed mainly to permanent ground deformation and less to wave propagation. Rigid pipe body, connections, age and corrosion are some of the factors that influence the seismic response of waste-water network. In Europe, very limited data are available. In Bucharest earthquake (1977), no damages were observed in waste-water pipeline network (total length: 1400km)- (Aldea et al., 2002). No damages occurred in waste-water treatment plant as a result of Lefkas earthquake (2003) in Greece. Two damages were recorded in the main waste-water system in the coastline of Lefkas as a result of permanent deformations, although in several areas of the city a smell of wastes was intense. Moreover, it must be mentioned that no damage was induced to the pumping station despite the occurrence of 11cm settlement (Alexoudi, 2005).

D3.5 -Fragility functions for water and waste-water system elements

Table 2-6 Brief presentation of waste-water system damages as result of Loma Prieta, Northridge and Hyogo-ken Nanbu (Kobe) earthquake

Earthquake/ Loma Prieta, 1989, Northridge, 1994, Hyogo-ken Nanbu (Kobe), System 1995, Mw=6.9, max. Mw=6.7, max. MMI=IX MMI=IX Mw=6.9, max. JMA=VII

Waste- As result of electric Minor to moderate 3 of the 8 waste-water Water power loss damages were treatment plant were totally System (commercial and observed in waste- destroyed. Extensive back-up power if any) water treatment plant damages were observed in in lift stations, lead as result of wave Higashi-mada Plant in wastes to San propagation and cracks Kobe area as result of Francisco area and in the tanks. Moreover, permanent deformations. polluted Monterey waste-water process The direct impact of the Bays. Moreover, was also interrupted by Higashi-mada Plant failure extensive damages electric power loss. All was the dismissal of wastes were occurred in the lift stations lost their without any treatment to main waste-water connection with electric Osaka Bay. Waste-water system of San power system in LA system mains, presented Francisco Bay and in region. The waste- total failure in areas with Watsonville. Minor to water network was large permanent moderate damages destroyed by deformations. The loss of were observed in permanent ground electric power influence the waste-water treatment deformations. operability of pumping plant in the area of stations. San Francisco.

References EERI (1990), NRC EERI (1995), TCLEE NIST (1996), NCEER (1994) (1995), NIST (1994) (1995) Shrestha (2001)

Fig. 2-11 Plenary view of waste-water treatment plant of Lefkas (Greece)

D3.9 - Fragility functions for water and waste-water system elements

Fig. 2-12 No damage observed in waste-water lift station during the 2003 Lefkas earthquake in Greece (from in-situ inspection Alexoudi and Argyroudis 2003)

2.3.1 Waste-Water Treatment Plant

The main failure modes of waste-water treatment plants are the same as the potable water treatment plants.

2.3.2 Tunnels

The main failure modes of tunnels are the same as in potable water system.

2.3.3 Pipes

The main failure modes of pipes are the same as in potable water system.

2.3.4 Lift Station

The main failure modes of lift stations are the same as in potable water system.

D3.5 -Fragility functions for water and waste-water system elements

3 Methodology for the vulnerability assessment of water and waste-water system elements

3.1 IDENTIFICATION OF THE MAIN TYPOLOGIES OF WATER SYSTEM ELEMENTS

A potable water supply is necessary for drinking, food preparation, sanitation, fire- extinguishing etc. Water (which may be non-potable) is also required for cooling equipment. A potable water system consists of transmission and distribution systems: o Transmission system stores “raw” water and delivers it to treatment plants. Such a system is made up of canals, tunnels, elevated aqueducts and buried pipelines, pumping plant and reservoirs. o Distribution system delivers treated water to customers.

Various components comprise potable water system according to ALA (2001a); RISK-UE (2001-2004) and LESSLOSS (2004-2007). The same components are also proposed in SYNER-G (Fig. 3-1) as listed below: o Water source o Treatment plant o Pumping station o Storage o Supervisory Control and Data Acquisition (SCADA) o Conduits (pipes, tunnel, canals)

POTABLE WATER SYSTEM

Water Source Water Treatment Pumping station Pipes Building facilities Plant - Springs Tunnels - System control - Wells Canals - Storage Storage - Rivers - Administrative, customer service - Lakes - Impounding reservoirs

Fig. 3-1 Breakdown of potable water system components

D3.9 - Fragility functions for water and waste-water system elements

3.1.1 Water Source

The typical water sources are springs, shallow or deep wells, rivers, natural lakes, and impounding reservoirs. Wells are used in many cities as both a primary and supplementary source of water. Wells include a pump to bring the water up to the surface, various electromechanical equipments and a building to enclose the well and the equipment.

Typology Wells, springs or river catchments are different types of water sources. Wells are described according to HAZUS (NIBS, 2004) with respect to: o Anchored / Unanchored; o The subcomponent. The subcomponents of wells that are considered in SYNER-G are the same as in HAZUS (NIBS, 2004): o Electric power (commercial power) o Well pump o Building o Electric equipment.

3.1.2 Water Treatment Plant

Water treatment plants are complex facilities, generally composed of a number of connected physical and chemical unit processes, whose purpose is to improve the water quality. Treatment processes used depend on the raw-water source and the quality of finished water desired. A conventional water treatment plant consists of a coagulation process, followed by a sedimentation process, and finally a filtration process. Components in the treatment process include pre-sedimentation basins, aerators detention tanks, flocculators, clarifiers, backwash tanks, conduit and channels, coal sand or sand filters, mixing tanks, settling tanks, clear wells, and chemical tanks. Alternatively, a water treatment plant can be regarded as a system of interconnected pipes, basins, and channels through which the water moves, and where the flow is governed by hydraulic principles.

Typology Water Treatment Plant may be described (HAZUS; NIBS, 2004) with respect to: o Its size (small, medium or large); o Anchored / Unanchored; o The subcomponent (equipment and backup power) considered. The size of the water treatment plant may be considered as a typological parameter, due to its increasing redundancy and importance factor for design (HAZUS; NIBS, 2004).

D3.5 -Fragility functions for water and waste-water system elements

Small water treatment plants (~50 M Gallons å 189.500 m3/day), are assumed to consist of a filter gallery with flocculation tanks (composed of paddles and baffles) and settling (or sedimentation) basins as main components, chemical tanks (needed in the coagulation and other destabilization processes), chlorination tanks, electrical and mechanical equipment, and elevated pipes. Medium water treatment plants are simulated by adding more redundancy to small treatment plants (i.e. twice as many flocculation, sedimentation, chemical and chlorination tanks) and large water treatment plants (i.e., three times as many flocculation, sedimentation, chemical and chlorination tanks/basins) – (between or ‡200 M Gallons å 758.000 m3/day). In SYNER-G, in order to account for the uncertainty in their final response as a result of the different European practices used for Water Treatment Plants of different sizes and the semi- anchorage of the subcomponents, only one fragility curve for Water Treatment Plant is proposed independently of the size. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario). The following subcomponents that may be considered in SYNER-G for water treatment plant are the same as in HAZUS (NIBS, 2004) except for the back-up power. o Electric Power (commercial power); o Chlorination equipment; o Sediment floculation; o Basins; o Baffles, Paddles, Scrapers; o Chemical Tanks; o Electric equipment; o Elevated pipe; o Filter Gallery.

3.1.3 Pumping Station

A Pumping station is a facility that boosts water pressure in both transmission and distribution systems. In general, pumping stations include larger stations adjacent to reservoirs and rivers, and smaller stations distributed throughout the water system intended to raise head. Pumping stations typically comprise buildings, intake structures, pump and motor units, pipes, valves, and associated electrical and control equipment (ATC-25, ALA 2001a).

Typology Pumping Station may be described (HAZUS; NIBS, 2004) with respect to: o Its size (small, medium or large); o Anchored / Unanchored; o The subcomponent (equipment and backup power) considered.

D3.9 - Fragility functions for water and waste-water system elements

A small pumping station boost less than 10 M Gallons (37.900 m3/day) to transmission and distribution systems, according to HAZUS (NIBS 2004). In SYNER-G, in order to account for the uncertainty in their final response as a result of the different European practices used for Pumping Station of different sizes and the semi- anchorage of the subcomponents, only one fragility curve for Pumping Station is proposed independently of the size for different building categories. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario). The following subcomponents that may be considered in SYNER-G for a pumping station are the same as in HAZUS (NIBS, 2004) except for the back-up power. o Electric Power (backup, commercial power); o Vertical/ Horizontal Pump; o Building; o Equipment. Comment: See also D3.1 “Fragility functions for common RC building types in Europe” and D3.2 “Fragility functions for masonry buildings in Europe”.

3.1.4 Storage

Storage tanks can be located at the start, along the length or at the end of a water transmission/distribution system. Their function may be to hold water for operational storage, provide surge relief volumes, provide detention times for disinfection, and other uses. Most water systems include various types of storage reservoirs in their transmission/ distribution systems. Storage reservoirs can be either tanks or open cut reservoirs. Open Cut Reservoir simply means that the reservoir is built by creating a reservoir in the natural lie of the land, often with one side of the reservoir made up of an earthen embankment dam. Many open cut reservoirs are enclosed by adding a roof so that treated water inside is protected from contamination from outside sources. A tank is a vessel that holds water. Water tanks are usually built of steel, concrete or wood (most often redwood). Tanks can be elevated by columns, built “at-grade” to rest directly on the ground or on a foundation on the ground, or buried. Also, in some smaller parts of distribution systems, water can be stored in pressure tanks, which are small horizontal pressure vessels on supports, at grade.

Typology Storage typology parameters may be the following: o Material (wood, steel or concrete); o Size; o Anchorage; o Position (at grade or elevated); o Type of roof;

D3.5 -Fragility functions for water and waste-water system elements

o Seismic design. o foundation type o Construction technique

Table 3-1 Typology of water storage tanks.

Element Tanks

Elevated tanks have capacities ranging between 750-53000m3, they are generally from steel or concrete and founded on piles or with surface foundations. They are usually located in small cities or rural areas. Elevated steel Steel tanks typically have lateral load resistant capacity for wind or earthquakes. In many cases they do not have any seismic design. The roofs of steel tanks are either made of steel or wood. It is also possible, to have steel tanks without roofs.

In general, tanks in Europe and in Greece are reinforced concrete (R/C) with roof from concrete. Concrete tanks can be either at-grade or buried, anchored or un- Concrete anchored. Many reinforced concrete tanks are post-tensioned. In urban areas in Greece, tanks are reinforced concrete or post-tensioned, with surface foundation or supported on piles.

Wood tanks are generally at-grade, they have limited capacity less than 1500m3 and they are not anchored. Elevated tanks are rarely used and they are usually Wood constructed from sekou wood. In Greece, we do not find any, in contrast to e.g. Scandinavian countries where they are still in use.

There are only masonry and masonry with reinforced concrete structures. This Masonry kind of tanks is still in use in some parts of the water system.

An open cut reservoir is made by cutting into the ground. They usually not Open cut include roof structures. In rare cases, a roof structure is installed to protect water reservoirs from external pollution.

3.1.5 Supervisory Control and Data Acquisition (SCADA)

Various types of in-line components exist along water transmission pipelines, including portions of the supervisory control and data acquisition (SCADA) system located along the conveyance system and various flow control mechanisms (e.g., valves and gates). In-line SCADA hardware includes a variety of components, including: o Instrumentation; o Power Supply (normal, backup); o Communication components (normal, backup); o Weather enclosures (electrical cabinets and vaults).

D3.9 - Fragility functions for water and waste-water system elements

SCADA system components in water transmission systems are the followings. o Instruments attached to the pipeline may include flow and pressure devices that are sometimes installed in a venturi section of pipeline. o Instruments attached to a canal may include various types of float instruments, which are used to assess the water level in the canal. o Remote Terminal Units (RTUs) and Programmable Logic Controllers (PLCs) are most commonly solid state devices. An RTU device picks up the analog signals from one or more channels of SCADA system devices at one location. The RTU converts these signals into a suitable format for transmission to a central SCADA computer, often at a location remote from the devices. A PLC can control when pumps are turned on or off, based on real time data or pre-programmed logic. o Most water systems have used manual recorders to track pressures, flows and gradient information. These recorders are still in use in many water systems. The recorders sometimes report on the same information as the automated SCADA system, often using the same instruments. Also, since the installation of automated SCADA system hardware is often relegated to a few locations in the water system, the manual recorder may be the only recording device at a location. o SCADA Cabinet is a metal enclosure that is mounted to a floor or bolted to a wall. o Most SCADA systems include battery backups. o Communication Links. The remote SCADA system is connected in some manner to the central location SCADA computer system. The most common links are radio, leased landlines and, to a lesser extent, microwaves; the use of public switched landlines is rare. o Canal gate structures.

Typology The location of the valves is often important when deciding how a pipeline system performs as a whole; damage to a pipeline between two valves will need to be isolated by closing the valves. Thus, typology depends on the following parameters: o Intervals between valves on conduits; o Anchorage of SCADA cabinet and inside equipments; o Number and type of communication links.

3.1.6 Conduits

Transmission conduits are typically large size pipes (more than 400mm in diameter) or channels (canals) that convey water from its source (reservoirs, lakes, rivers) to the treatment plant. Transmission pipelines are commonly made of concrete, ductile iron, cast iron, or steel. These could be elevated/at grade or buried. Elevated or at grade pipes are typically made of steel (welded or riveted), and they can run in single or multiple lines.

D3.5 -Fragility functions for water and waste-water system elements

Canals are typically lined with concrete, mainly to avoid excessive loss of water by seepage and to control erosion. In addition to concrete lining, expansion joints are usually used to account for swelling and shrinkage under varying temperature and moisture conditions. Distribution of water through conduits can be accomplished by gravity, or by pumps in conjunction with on-line storage. Except for storage reservoirs located at a much higher altitude than the area being served, distribution of water would necessitate, at least, some pumping along the way. Typically, water is pumped at a relatively constant rate, with flow in excess of consumption being stored in elevated storage tanks. The stored water provides a reserve for fire flow and may be used for general-purpose flow should the electric power fail, or in case of pumping capacity loss. Conduits are artificial channels made for the conveyance of fluids (Fig. 3-2). They fall into two categories: o Free-flow conduits guide the fluid as it flows down a sloping surface. o Pressure conduits confine and guide fluid movement under pressure. Free-flow conduits may be simple open channels or ditches, or pipes or tunnels flowing partially full. A pressurized conduit can be a pipeline or tunnel flowing under internal pressure.

Fig. 3-2 Breakdown of potable water conduits.

Typology Beyond the nature of the conduits (see suitable sections), typology depends mainly on the flowing (gravity or pumped systems) and secondarily to the appurtenances along the aqueduct. o Gravity system aqueducts deliver the flow from higher elevations to lower elevations, and do not need any pumping to move the water. o Pumped-system aqueducts require pumps along the length of the aqueduct to keep the water moving. Appurtenances along the length of the aqueduct includes various turnouts, gates, valves, etc. Often ignored for a simplified earthquake loss estimate, these may be important if there are particular component vulnerabilities, or if a system model that includes connectivity is to be used.

D3.9 - Fragility functions for water and waste-water system elements

3.1.6.1 Pipes (Common to potable water and waste-water systems)

Pipes can be free-flow or pressure conduits, buried or elevated. Several materials can be used. In order to avoid contamination of treated water, potable water pipes are most of the time pressurized.

Typology Pipe typology depends on the following parameters: o Location (buried or elevated); o Material (type, strength); o Geometry (diameter, wall thickness); o Type of joints, continuous or segmented pipes; o Appurtenances and branches; o Corrosiveness (age and soil conditions). The selection of material type and pipe size are based on the desired carrying capacity, availability of material, durability and cost.

Location: Elevated pipes are large-diameter pipes supported on bents. They are often used in areas that traverse poor soils, and the bents are often supported on piles that extend to competent materials. Pile supports can be made of wood, concrete or concrete-encased steel. Buried pipes are buried 1 to 5 m or deeper in the ground.

Material: For detailed diagnostics of pipe failure, mechanical characteristics of material will be required. Otherwise, pipeline material allows simplified assessment. Pipes are commonly made of: o Asbestos Cement (AC), o Concrete (C), o Cast Iron (CI), o Ductile Iron (DI), o Welded Steel (S), o PolyVinyl Chloride (PVC), o High Density PolyEthylene (HDPE). o Vitrified Clay; o Brick; o Bituminised fibre;

D3.5 -Fragility functions for water and waste-water system elements

Geometry: The diameter of distribution pipe is important both in terms of pipe damage algorithms and post-earthquake performance of the entire water system. For more detailed study, wall thickness is also required. Pipe diameters are generally greater than 4 inches and one should consider the following classes:

o Small diameter means 4 to 12 inches (…100 to 300 mm); o Large diameter mean 16 inches and more (@400 mm).

Type of joints: A jointed pipeline consists of pipe segments coupled by relatively flexible (or weak) connections (e.g., a bell-and-spigot cast iron piping system). Continuous pipelines are those having rigid joints, such as continuous welded steel pipelines.

Appurtenances and branches: Pipeline damage tends to concentrate at discontinuities such as pipe elbows, tees, in-line valves, reaction blocks and service connections. Such features create anchor points or rigid locations that promote force/stress concentrations. Locally high stresses can also occur at pipeline connections to adjacent structures (e.g., tanks, buildings and bridges), especially if there is insufficient flexibility to accommodate relative displacements between the pipe and the structure. Corrosiveness: Corrosion will accentuate damage, especially in segmented steel, threaded steel and cast iron pipes. Older pipes appear to have a higher incidence of failure than newer pipes. Age effects are possibly strongly correlated with corrosion effects caused by the increasing impact of corrosion over time. Soil conditions can also influence corrosion. Experience has also shown that continuous pipelines with bends, elbows and local eccentricities will concentrate deformation at these features, especially if permanent ground deformations develop compression strains. Other pipe attributes that may be developed when collecting inventory data include: leak history, encasement, corrosion protection systems, location of air valve and blow-offs, etc. These attributes may yield some extra information as to the pipeline's fragility, but they may not be available to the analyst in all cases.

Functionality: Distribution pipes represent the network that delivers water to consumption areas. Distribution pipes may be further subdivided into primary lines, secondary lines and small distribution mains. The primary mains carry flow from the pumping station to and from elevated storage tanks, and to the consumption areas, whether residential, industrial, commercial, or public. These lines are typically laid out in interlocking loops. Secondary lines have smaller loops within the primary mains and run from one primary line to another. They serve primarily to provide a large amount of water for fire fighting without excessive pressure loss. Small distribution lines represent the mains that supply water to the user and to the fire hydrants.

D3.9 - Fragility functions for water and waste-water system elements

3.1.6.2 Tunnels (common to roadway, railway, potable water and waste-water systems)

Whatever the content (potable or waste-water, road or railway), tunnels are confined structures. They are often not redundant, and major disruption to the utility or transportation system is likely to occur should a tunnel become non-functional.

Typology Tunnels may be described according to (Table 3-2): o Construction technique, o Liner system o Geologic conditions. For a more detailed assessment, the shape of the section, the depth, the length and the diameter of the tunnel, the liner thickness might be a useful information.

Table 3-2 Typology of tunnels (ALA 2001a).

Typology Poor-to-average construction Good construction

Tunnels in very sound rock and designed for geologic conditions (e.g., special Tunnels in average or poor rock, support such as rock bolts or stronger either unsupported masonry or liners in weak zones); unreinforced, strong Rock timber liners, or unreinforced concrete liners with contact grouting to conditions concrete with frequent voids assure continuous contact with rock; behind lining and/or weak average rock; or tunnels with reinforced concrete. concrete or steel liners with contact grouting.

Tunnels that are bored or cut and cover box-type tunnels and include Tunnels designed for seismic loading, Alluvial tunnels with masonry, timber or including racking mode of deformation for soil or unreinforced concrete liners, or any cut and cover box tunnels. These also Cut and liner in poor contact with the soil. include tunnels with reinforced strong Cover These also include cut and cover concrete or steel liners in bored tunnels in conditions box tunnels not designed for good contact with soil. racking mode of deformation.

A more detailed description can be found in D3.7 “Fragility functions for roadway system elements” where it presents the final proposal for SYNER-G.

D3.5 -Fragility functions for water and waste-water system elements

3.1.6.3 Canals

Canals are free-flowing conduits, usually open to the atmosphere, and usually at grade. They tend to be larger than pipelines operated under pressure. The advantages of using a canal include the possibility of construction with locally available materials, longer life than metal pipelines, and lower loss of hydraulic capacity with age. The disadvantages include the need to provide the ultimate flow capacity initially and the likelihood of interference with local drainage. Flumes are open-channel sections that carry water in elevated structures.

Typology Canals can be formed by cutting a ditch into the ground, building up levees, or a combination of the two. Most often, canals are concrete-lined to reduce water losses. Canals can traverse both stable and unstable geologic conditions. Thus, canal typology may consider whether the canal is: o Open cut or built up using levees; o Reinforced, unreinforced liners or unlined embankments. Flumes sections are commonly made of wood or metal. The support systems can be built of wood, concrete or steel. The support structures might be a few feet high where the flume runs along a contour, or very tall where the flume crosses a creek or river. Flumes are specialized structures and are not specifically addressed here.

3.2 SYNER-G TYPOLOGIES OF WATER SYSTEM ELEMENTS

In summary, Table 3-3 provides a comparison of the typologies of potable water elements provided by HAZUS (NIBS, 2004) and ALA (2001a,b). The third column provides SYNER-G proposal for potable water elements. In Greece, the typology of potable water systems’ elements is based on international practice, although some features do not exist. In particular, components’ anchorage is not based on any specifications, despite the fact that some measures are taken for their seismic support. Usually, it depends on the workers’ expertise and the local experience from earthquakes. Thus, there is not a standard level of anchorage and the water system components cannot be considered as anchored. Regarding potable water treatment in Greece, there are some treatment facilities in water sources or even in central pumping stations. Transmission conduits from water sources are in general closed-type, but there are also some open parts. The reason for this, except from the cost, is because these canals were initially used for irrigation. In general, distribution systems are comprised from pipes with different materials, connection types and diameters. Construction codes for water systems do not exist in Greece until nowadays; although, at the end of the 70’s, specifications for the pipelines’ materials started to be applied, while special references are made to technical reports for the best construction practices. Nevertheless, large parts of the water systems in Greece have not been constructed using specific studies, resulting in lack of data for their characteristics. Storage tanks are usually constructed from concrete with concrete roof. They are half-full and not anchored. In big urban centres, they are anchored

D3.9 - Fragility functions for water and waste-water system elements with surface foundation or sited on piles according to the soil type. Pumping stations are reinforced concrete buildings, designed according to the current seismic codes. They usually have one part elevated, with the largest part being below ground where the tank and electric and mechanical equipment are located. In Greece (Thessaloniki), SCADAs exist in about 40% of the water pumping stations and in 3 points in water transmission pipeline in Thessaloniki.

Table 3-3 Comparison of the typologies of potable water elements provided in NIBS 2004, ALA 2001a,b and SYNER-G

Element ｠』〈S, 2003 ALA (2001a,b) SYNER-G

Water Components’ Components’ Sources - anchorage anchorage (wells) Water Size Size Treatment Components’ - Components’ Plant anchorage anchorage Soil type See D3.7 “Fragility

Tunnels Quality of functions for roadway - construction system elements” Material of Material of construction construction Canals - Amount of debris that Amount of debris that may enter the canal may enter the canal after an earthquake after an earthquake Material Material Material Diameter Type of joints/ Pipes Type of joints/ Type of joints/ connection connection connection

Soil type Size Size Material Material Material Tanks Foundation type Anchorage Anchorage Anchorage Foundation type Foundation type Seismic design Seismic design Pumping Size Size station Components’ - Components’ anchorage anchorage

D3.5 -Fragility functions for water and waste-water system elements

In France water is managed through dedicated plans called SDAGE (Schéma Directeur d’Aménagement et de Gestion des Eaux). These plans enable the protection of water resources from natural hazards. Up to now water management is very local and involves a great number of actors (more than 36.000 municipalities and 30 000 services). This explains the lack of harmonized national database. However a national observatory for water and waste-water systems was launched in 2009 (according to the new law on water, 30/12/2008). This observatory has defined a number of descriptive and performance indicators and aims at harmonizing data formats on the territory. These data should become available in the following years (www.services.eaufrance.fr) and should enable a targeted improvement of the systems. In France the sources of drinking water are mainly underground water tables (2/3) and surface water (1/3). The distribution network for drinking water represents about 878.000 km. The leaks are assumed to represent about 20% and to be due to corrosion, ground modification, old joints and individual connexion. The priority concerning the replacement of pipes is the following: grey cast-iron/steel (<1960) and cement-asbestos, PVC (<1975), grey cast-iron/steel (>1960), PVC (>1975), ductile cast-iron. In 2002 there were about 27.514 distribution units, but there are few standardised information at national level on drinking water units. A typology of potable water systems in Austria has not been available. Instead of that a brief description of water system of Vienna is given here. The potable water system in Vienna can principally be divided into two main water lines. These are the first and second Hochquellenleitung. The capacity of the first conduit is 220.000 m3/day and that of the second conduit is 217.000 m3/day. The first conduit is mostly made out of brickwork and concrete canals. The down-grade is sufficient enough, so that there are no pumping stations needed. The second conduit has a total length of about 200 km and the down grade is so high that there are no pumping stations needed. There are about 30 high-level tanks in Vienna.

3.3 IDENTIFICATION OF THE MAIN TYPOLOGIES OF WASTE-WATER SYSTEM ELEMENTS

Waste-water system can alternatively be called sewer network. Sewer network is comprised of components that work together to: o Collect o Transmit o Treat o Dispose of sewage

Various components comprise waste-water system according to RISK-UE (2001-2004) and LESSLOSS (2004-2007). The same components were also proposed in SYNER-G (Fig. 3-3). o Conduits (pipes, tunnels) o Treatment plant o Lift station o Supervisory Control and Data Acquisition (SCADA) 28

D3.9 - Fragility functions for water and waste-water system elements

WASTE WATER SYSTEM

Waste-Water Lift station Conduits Building Facilities Treatment - System control Plant - Storage - Administration/ customer service

Fig. 3-3 Breakdown of waste-water system.

3.3.1 Conduits

Conduits are artificial channels made for the conveyance of fluids. Mainly free-flow conduits that guide the fluid as it flows down a sloping surface are present in waste-water system. Free-flow conduits may be pipes or tunnels flowing partially full. Collection sewers are generally closed conduits that carry normally sewage with a partial flow. They could be sanitary sewers, storm sewers, or combined sewers. Interceptors are large diameter sewer mains, usually located at the lowest elevation areas.

Fig. 3-4 Breakdown of waste-water conduits.

Typology In general, mains in the sanitary sewer system are underground conduits that normally follow valleys or natural streambeds. Waste-water conduits are usually designed as free flow channels except where lift stations are required to overcome topographic barriers. Sometimes the sanitary sewer system flow is combined with the storm water system prior to treatment.

3.3.1.1 Pipes (Common to potable water and waste-water systems)

Waste-water pipes are most of the time free flow conduits.

D3.5 -Fragility functions for water and waste-water system elements

Typology The typology of waste-water pipes is the same as in potable water pipes. More specific pipe materials used for collection sewers and interceptor sewers are similar to those for potable water. The most commonly used sewer material is clay pipe manufactured with integral bell and spigot end. Concrete pipes are mostly used for storm drains and for sanitary sewers carrying non corrosive sewage (i.e. with organic materials). For the smaller diameter range, plastic pipes are also used.

3.3.1.2 Tunnels

The typology of waste-water tunnels is the same as in potable water tunnels. A more detailed description can be found in D3.7 “Fragility functions for roadway system elements”.

3.3.2 Waste-water Treatment Plant

Waste-water treatment plants in the sanitary sewer system are complex facilities which include a number of buildings and underground or on ground reinforced concrete tank and basins. Common components at a treatment plant include trickling filter, clarifiers, chlorine tanks, recirculation and waste-water pumping stations, chlorine storage and handling, tanks, and pipelines. Concrete channels are frequently used to convey the waste-water from one location to another within the complex. Within the buildings there are mechanical, electrical, and control equipment, as well as piping and valves. Conventional waste-water treatment consists of: o preliminary processes (pumping, screening, and grit removal), o primary settling to remove heavy solids and floatable materials, o secondary biological aeration to metabolise and flocculate colloidal and dissolved organics. Preliminary treatment units vary but generally include screens to protect pumps and prevent solids from fouling grit-removal units and flumes. Additional preliminary treatments (flotation, flocculation, and chemical treatment) may be required for industrial wastes. Primary treatment typically comprises sedimentation, which removes up to half the suspended solids. Secondary treatment removes remaining organic matter using activated-sludge processes, trickling filters or biological towers. Chlorination of effluents is commonly required.” Waste sludge may be stored in a tank and concentrated in a thickener. Raw sludge can be disposed of by anaerobic digestion and vacuum filtration, with centrifugation and wet combustion also currently used.

D3.9 - Fragility functions for water and waste-water system elements

Typology Waste-Water Treatment Plant may be described (HAZUS; NIBS, 2004) with respect to: o Its size (small, medium or large); o Anchored / Unanchored; o The subcomponent (equipment and back-up power) considered. The size of the waste-water treatment plant may be considered as a typological parameter, due to its increasing redundancy (HAZUS; NIBS, 2004) and importance factor for design.

Small treatment plants (~50 M Gallons å 189.500 m3/day) are assumed to consist of a filter gallery with flocculation tanks (composed of paddles and baffles) and settling (or sedimentation) basins as main components, chemical tanks (needed in the coagulation and other destabilization processes), chlorination tanks, electrical and mechanical equipment, and elevated pipes. Medium treatment plants (50 < x <200M Gallons) are simulated by adding more redundancy to small treatment plants (i.e. twice as many flocculation, sedimentation, chemical and chlorination tanks).

Large treatment plants (‡200 M Gallons å 758 000 m3/day) are simulated by adding even more redundancy to small treatment plants (i.e., three times as many flocculation, sedimentation, chemical and chlorination tanks/basins). Whether the subcomponents (equipment and back-up power) are anchored or not is another typological parameter. In order to account for the uncertainty in their final response as result of the different European practices used for Waste-Water Treatment Plants of different sizes and the semi-anchorage of subcomponents, only one fragility curve for Waste-Water Treatment Plant is proposed independently of the size. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario). The following subcomponents that may be considered in SYNER-G for waste-water treatment plant are the same as in HAZUS (NIBS, 2004) except for the back-up power: o Electric Power (commercial power); o Chlorination equipment; o Sediment floculation; o Chemical Tanks; o Electric equipment; o Elevated pipe. o Building Also, the treatment level could be considered (primary, secondary, tertiary).

3.3.3 Lift station

Lift or pumping stations serve to raise sewage over topographical rises or to boost the disposals. They are typically used to transport accumulated waste-water from a low point in

D3.5 -Fragility functions for water and waste-water system elements the collection system to a treatment plant. If the lift station is out of service for more than a short time, untreated sewage will either spill out near the lift station, or back up into the collection sewer system. Pumping stations consist primarily of a wet well, which intercepts incoming flows and permit equalization of pump loadings and a bank of pumps, which lift the waste-water from the wet well. The centrifugal pump finds widest use at pumping stations. Thus, a plant is usually composed of a building, one or more pumps, electrical equipment, and, in some cases, back-up power systems. Lift stations are often at least partially underground.

Typology Lift station may be described (HAZUS; NIBS, 2004) with respect to: o Its size (small, medium or large); o Anchored / Unanchored; o The subcomponent (equipment and backup power) considered.

Small lift stations transport less than 10 M Gallons (37 900 m3/day) of disposal according to HAZUS (NIBS, 2004) while medium/large lift station transfer more than 10 M Gallons. In SYNER-G in order to account or for the uncertainty in their final response as a result of the different European practices used for lift stations of different sizes and the semi- anchorage of subcomponents, only one fragility curve for Pumping Stations is proposed independently of the size for different building types. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario). The following subcomponents may be considered in a pumping station (HAZUS; NIBS, 2004) expect for the back-up power: o Electric Power (commercial power); o Vertical/ Horizontal Pump; o Building; o Equipment.

3.3.4 Supervisory Control And Data Acquisition (SCADA)

Various types of in-line components exist along waste-water transmission pipelines, including portions of the supervisory control and data acquisition (SCADA) system located along the conveyance system and various flow control mechanisms (e.g., valves and gates). In-line SCADA the hardware includes a variety of components, including: o Instrumentation; o Power Supply (normal, backup); o Communication components (normal, backup); o Weather enclosures (electrical cabinets and vaults).

D3.9 - Fragility functions for water and waste-water system elements

SCADA system components in waste-water transmission systems are the following: o Instruments attached to the pipeline may include flow devices that are sometimes installed in a venturi section of pipeline. o Remote Terminal Units (RTUs) and Programmable Logic Controllers (PLCs) are most commonly solid state devices. An RTU device picks up the analog signals from one or more channels of SCADA system devices at one location. The RTU converts these signals into a suitable format for transmission to a central SCADA computer, often at a location remote from the devices. A PLC can control when pumps are turned on or off, based on real time data or pre-programmed logic. o SCADA Cabinet is a metal enclosure that is mounted to a floor or bolted to a wall. o Most SCADA systems include battery backups. o Communication Links. The remote SCADA system is connected in some manner to the central location SCADA computer system. The most common links are radio, leased landlines and, to a lesser extent, microwaves; the use of public switched landlines is rare.

3.4 SYNER-G TYPOLOGIES OF WASTE-WATER SYSTEM ELEMENTS

In summary, Table 3-4 provides a comparison of the typologies of waste-water elements provided by HAZUS (NIBS, 2004) and the proposal within SYNER-G.

Table 3-4 Comparison of the typologies of potable water elements provided in HAZUS (NIBS, 2004) and SYNER-G

Element ｠』〈S, 2003 SYNER-G Waste-Water Size Size Treatment Plant Components’ anchorage Components’ anchorage See D3.7 “Fragility functions for Tunnels - roadway system elements” Material Material Pipes Type of joints/ connection Type of joints/ connection Lift station Size Size Components’ anchorage Components’ anchorage

D3.5 -Fragility functions for water and waste-water system elements

In Greece, collection sewers (sanitary, storm or combined sewers) are usually closed conduits. The older storm sewers are constructed from clay, while the newer ones are made of concrete. Sanitary sewers carrying non-organic materials are also constructed from concrete; they usually are large-diameter, gravity pipes. Smaller conduits are constructed from PVC. Pressure (usually steel) pipes are used for the conveyance of sewage from pumping stations of Central Sewage Conduits to areas of higher elevation. Central Sewage Conduits have diameters >1.000mm and constructed from reinforced concrete (sometimes pre-stressed). In Greece (Thessaloniki), SCADA exists in all lift stations. In France the waste-water collection system represents about 280.000 km of pipes. Among these pipes 10% were assumed to be older than 60 years in 2002, and some pipes were not installed correctly in the 70s, which makes replacement necessary. These pipes feed about 17.300 waste-water treatment plants with a total capacity of 76 millions Equivalent-Human (75% of these plants were built after 1990). Plants with capacity >100.000 EH represent only about 113 plants, whereas plants with capacity < 500 EH are numerous (about 6.225). The most used technique is the activated sludge process for waste-water treatment. In Austria, collection sewers (sanitary, storm or combined sewers) are closed conduits. Older sewers can be constructed from clay or brickwork. The younger ones are constructed from concrete or PVC. The waste-water system in Vienna is roughly 2.400 km long and takes all sewage in Vienna to one main sewage treatment plant.

3.5 GENERAL DESCRIPTION OF EXISTING METHODOLOGIES

Fragility relationships are a critical component of seismic impact assessment. The fragility, or vulnerability, functions relate the severity of shaking to the probability of reaching a level of damage (e.g. light, medium, extensive, near-collapse) to various infrastructure items. The level of shaking can be quantified using numerous shaking parameters, including peak ground acceleration, velocity, displacement, spectral acceleration, spectral velocity or spectral displacement. Each infrastructure item requires a corresponding set of fragilities to determine damage level likelihoods (probability). In general, fragility functions relate a level of shaking, or system demand, to the conditional probability of a specific system reaching or exceeding a limit state response. A deterministic response, or the vertical line, indicates a lack of uncertainty in the system response. Fragility curves close to vertical indicate a low level of uncertainty, while those with a much higher uncertainty are spread over a much wider range of shaking values.

3.6 DAMAGE STATES OF WATER SYSTEM ELEMENTS

3.6.1 Water Source

Parameters defining damage states of water sources are: o Type and extent (level) of structural damage (HAZUS; NIBS, 2004). o Serviceability state (HAZUS; NIBS, 2004).

D3.9 - Fragility functions for water and waste-water system elements

3.6.2 Water Treatment Plant

Parameters defining damage states of water treatment plant are: o Type and extent (level) of structural damage (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007). o Serviceability state (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007) o Functionality level (Ballantyne et al., 2009) o Restoration Cost (% of replacement cost) - (Ballantyne et al., 2009)

3.6.3 Pumping Station

Parameters defining damage states of pumping station are: o Type and extent (level) of structural damage (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007). o Serviceability state (HAZUS, NIBS, 2004; SRM-LIFE, 2003-2007). o Reliability index (Scawthorn, 1996)

3.6.4 Storage tanks

Parameters defining damage states of storage tanks are: o Description of the type and extent (level) of structural damage (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007; O’Rourke and So, 1999). o Loss of context (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007; ALA, 2001; O’Rourke and So, 1999)

3.6.5 Canal

Parameters defining damage states of canals are: o Hydraulic performance of a canal

3.6.6 Pipes

Parameters defining damage states of pipes are: o Repair rate per km (Katayama et al., 1975; ATC-13,1985; Isoyama and Katayama, 1982; Memphis, Tennessee, 1985; O’ Rourke and Ayala, 1993; Eidinger et al., 1995; Eidinger, 1998; Isoyama, 1998; O’Rourke et al.,1998; O’Rourke and Leon, 1999; Eidinger and Avila, 1999; Isoyama et al., 2000; Toprak, 1998; Hung, 2001; Honegger and Eguchi, 1992; Heubach, 1995; Eidinger et al.,1999; ]LA, 2001a,b; Yeh et al., 2006) o Break/ 1000 feet (Eguchi , 1983; Wang et al., 1991; O’Rourke and Deyoe, 2004) o Vulnerability class (Ballantyne and Heubach, 1996)

D3.5 -Fragility functions for water and waste-water system elements

3.6.7 Tunnels

See D3.7 “Fragility functions for roadway system elements”

3.7 DAMAGE STATES OF WASTE-WATER SYSTEM ELEMENTS

3.7.1 Waste-Water Treatment Plant

Parameters defining damage states of waste-water treatment plant are: o Type and extent (level) of structural damage (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007). o Serviceability state (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007)

3.7.2 Conduits

Parameters defining damage states of conduits are: o Level of ground strain (Mataki et al., 1996)

Moreover, for the case of waste-water pipes, the parameters defining damage states are the same as in potable water system while for the case of tunnels are the same as for roadway system elements

3.7.3 Lift station

Parameters defining damage states of lift station are: o Type and extent (level) of structural damage (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007). o Serviceability state (HAZUS; NIBS, 2004; SRM-LIFE, 2003-2007).

3.8 INTENSITY INDEXES

The characteristics of ground motions that influence the seismic performance and integrity of lifelines are intensity, frequency content and duration of the motions. Each of these characteristics of ground motion at a given site is influenced by the nature of the fault rupture process, the travel path followed by the resulting seismic waves as they propagate from the ruptured fault to the site, the “site effects” including the effects of local soil conditions, the basin effects and topography. The intensity of the shaking has been typically represented using the parameter given below: o Peak Ground Horizontal Acceleration (PGAH). o Peak Ground Vertical Acceleration (PGAV). o Acceleration time history (ies) a(t). 36

D3.9 - Fragility functions for water and waste-water system elements

o Peak Ground Horizontal Velocity (PGVH). o Peak Ground Vertical Velocity (PGVV). o Peak Ground Displacement (PGD). o Acceleration, Velocity and Displacement Response Spectrum SA(T, つ) for a suitable range of periods. (normally <10 Hz but recently up to 20 Hz as well especially for the displacement spectra) o Transient ground strains o Arias Intensity o Fourier Spectrum o Fundamental period of the ground motion (it is related to the site effects as well) o Duration (Bracketed Duration, Significant Duration). o Equivalent number of uniform cycles, Neq. o In case of slope movement, fault crossing and liquefaction induced phenomena (lateral spreading and subsidence), the Permanent Ground Deformations (displacements, PGD) - total and differential - are the key parameters.

The main issue is to define the appropriate ground motion intensity parameter that best captures the response of each element, minimizes the dispersion of that response and is related to the approach that is followed for the derivation of fragility curves. As a general apposition, the empirical fragility curves relate the observed damages with the seismic intensity and so PGA and PGVs are the more suitable parameters with lower uncertainties. For linear lifeline systems like pipelines it has been proved that peak ground velocity is better correlated to the observed damages, and thus the vulnerability assessment should be based on ground velocity estimates. An alternative approach may be the use of ground strains (longitudinal and transversal) or/and differential ground displacements, which are directly correlated to the ground velocity. For other lifeline components it may be peak ground acceleration (i.e. buildings, tanks, water treatment plant). Of course permanent ground deformations are also a key parameter.

3.8.1 Water System Elements

The following is a comprehensive list of the different descriptors used for the components in potable water system (Table 3-5).

D3.5 -Fragility functions for water and waste-water system elements

Table 3-5 Intensity measures for the vulnerability assessment potable water system elements

Element Intensity Reference Comments at risk Measure Complex components including several NIBS (2004) subcomponents. The overall performance of the Wells SRMLIFE (2003- PGA component is based on the subcomponents. 2007) Fragility curves based on PGA are given for each subcomponent. Complex components including several Water NIBS (2004) subcomponents. The overall performance of the Treatment SRMLIFE (2003- PGA component is based on the subcomponents. Plants 2007) Fragility curves based on PGA are given for each subcomponent. Complex components including several NIBS (2004) subcomponents. The overall performance of the Pumping SRMLIFE (2003- PGA component is based on the subcomponents. Stations 2007) Fragility curves based on PGA are given for each subcomponent. Water NIBS (2004) PGA Storage PGA, ALA (2001a,b) Tanks PGD* Barenberg (1988) PGV Empirical fragility curve Eguchi (1991) MMI Mercalli Intensity Empirical fragility curve for wave propagation. Good correlation with damages (Alexoudi , 2005; O’ Rourke and Ayala PGV Alexoudi et al., 2007; Pitilakis et al., 2005) for (1993) Düzce (Turkey), Lefkas island (Greece) earthquake. Eidinger and Avila PGV, Empirical fragility curves for wave propagation and (1999) PGD* for permanent ground deformation. Hwang and Lin PGA Empirical fragility curve for wave propagation (1997)

Pipes Isoyama et al. (1998) PGV Empirical fragility curve for wave propagation O’Rourke and Jeon Vscaled Empirical fragility curve for wave propagation (1999) PGA, Empirical fragility curve for wave propagation and ALA (2001a,b) PGD* for permanent deformation Empirical fragility curve for permanent ground Porter et al. (1991) PGD* deformation Honegger and Empirical fragility curve for permanent ground PGD* Eguchi (1992) deformation Empirical fragility curve for permanent ground Heubach (1995) PGD* deformation Empirical fragility curve for permanent ground Terzi et al. (2006) PGD* deformation PGD*: Permanent Ground Displacements 38

D3.9 - Fragility functions for water and waste-water system elements

3.8.2 Waste-Water System Elements

The following is a comprehensive list of the different descriptors used for the components in potable water system (Table 3-6).

Table 3-6 Intensity measures for the vulnerability assessment waste- water system elements

Intensity Element at risk Reference Comments Measure Complex components including several subcomponents. The overall performance Waste-Water NIBS (2004) of the component is based on the Treatment PGA SRMLIFE (2003- subcomponents. Fragility curves based Plants 2007) on PGA are given for each subcomponent. Complex components including several NIBS (2004) subcomponents. The overall performance of the component is based on the Lift Stations PGA SRMLIFE (2003- subcomponents. Fragility curves based 2007) on PGA are given for each subcomponent. Tunnel as tunnels in Roads (Interceptors) as potable water Pipes (Sewer) pipes

3.9 PERFORMANCE INDICATORS

In general, the performance measures used to assess the performance of water, waste- water system can be defined by: o Inventory Functions: physical characteristics, numbers of facilities. o Engineering: structural integrity, deterioration. o Operational Reliability: Connectivity/ Serviceability/ Operability/ Functionality. o Direct/ Indirect consequences in economy (e.g Cost/Benefit Analysis, capital and financial resources). o Demand: e.g. pressure and flow (for water system). o Safety and Security. Water system and waste-water system are very complex systems comprised by several individual components (e.g Water System å water source, water treatment plants, pipelines, tunnels, canals, storage tanks, pumping stations and SCADA; Waste-Water System å waste-water treatment plants, lift stations, pipelines and tunnels). The overall performance of a system depends on the individual performance of its components. For that reason, some specific performance measures can be defined for each component and for the whole system. 39

D3.5 -Fragility functions for water and waste-water system elements

3.9.1 Water System/ component performance indicators

ALA (2002) proposes some performance metrics for water system that are related to: o Percent (%) served (in total or by sector) within a specific number of days with raw water with adequate fire flow pressures, and/or o Percent (%) served (in total or by sector) within a specific number of days with fully treated water These metrics could be measured alternatively in terms of number of service connections, populations served, or volume of water served (i.e., cubic feet or gallons) for the whole water system. Each water component, according to ALA (2002), can have different importance with respect to a set of performance objectives. Their importance can be accounted according to Component Criticality Rating (CCR), that is: LSR = Life Safety Rating (based on fraction of time occupied) FFR = Fire Flow Rating (significance to fire fighting) DWR = Drinking Water Rating (significance to drinking water supply) DPR = Damage Potential Rating (potential for causing damage to adjacent facilities)

Essential to the evaluation of water system performance is a system vulnerability model. In such a system vulnerability model, the basic issues to be addressed are if the final nodes (service zones, service connections, fire hydrants) have (a) flows with adequate fire flow and pressures or (b) potable water supply that meets stringent safe drinking water health standards. Simpler, water system performance indicators can be described by water flow [m3/h], discharge / pressure [bar] / number of people supplied [people/km2 supplied] (or ratio of zones [%]) / drinkability / ratio of critical facilities supplied [%]. In a case of water system components: o Water source: water flow [m3/h] and drinkability, reserve [m3] o Treatment plant: treatment capacity (qualitative and quantitative [m3/h]) o Pumping station: flow capacity [m3/h] o Storage tanks: reserve [m3] o Tunnels: water flow [m3/h], o Pipes: water flow [m3/h], repair rates [repair per km] o Canals: water flow [m3/h] o SCADA A summary of water component performance indicators is given in Table 3-7. Furthermore, a summary of water system performance indicators is provided in Table 3-8.

D3.9 - Fragility functions for water and waste-water system elements

Table 3-7 Summary of Water Component Performance Indicators (WCPIs).

A/A Approach Component Description Reference Functionality Certain critical pipelines serving critical facilities remain 1 Pipeline ALA (2005) analysis operational during and following an earthquake An acceptable damage rate should be about 0.03 to 0.06 Acceptable damage 2 Pipeline breaks per 1,000 feet of equivalent 6-inch diameter pipe, in ALA (2005) rate evaluation order to confirm with the service restoration target. Redundancy Especially for transmission pipelines (Function Class II – 3 Pipeline ALA (2005) analysis pipes) Estimation of the performance of pipelines after the Pipelines, Storage comparison of the condition of existing pipeline with the 4 Operability facilities, Pumping ideal pipe with appropriate design and construction practice. ASCE 7-02 provisions station Water storage facilities and pump structures needed to supply water pressure to rest network. JWWA defines important facilities and for them defines the Acceptable damage 5 All components damage state that complies with the acceptable 1997 JWWA Guidelines states performance criterion

D3.5 -Fragility functions for water and waste-water system elements

Table 3-8 Summary of Water System Performance Indicators (WSPIs)

A/A Analysis Type Description Comments Reference Examines: - “Damage ratio”: the degree of physical damage to the system (defined as the Propose a diagram between damage expected number of failures per unit length ratio (km) and Service ratio (%). 1a Connectivity or per link) Application for the restoration Kawakami (1990) - “Service ratio: indicates the ratio of process of water transmission normally supplied houses to the total system in the City of Tokyo number in the system. This value increases as restoration proceeds Uses: Formal graph theoretic notions to define Dueñas-Osorio et al. (2007a) Dueñas-Osorio et al. (2007a, 1b Connectivity characteristic measures of the network, such as examine the loss of connectivity of a 2007b, 2009) an importance ordering of the vertices, the water distribution system characteristic path length and redundancy Examines: - The “Reachability” of water to certain key Application for the water distribution Moghtaderi-Zadeh et al. 1c Connectivity nodes system in the East of San Francisco. (1982) - The probability that a certain amount of water flow would reach key locations Estimate: For simplified evaluations, a 1d Connectivity - Connectivity matrix graphical portrayal of the system is ALA (2002) - Reachability matrix adequate. Examines: Application for the water distribution Adachi and Ellingwood 2a Serviceability Probability distribution of the percentage of customers who would lose their service after an system in Shelby County, TN (2008) earthquake

D3.9 - Fragility functions for water and waste-water system elements

A/A Analysis Type Description Comments Reference The water system operating Evaluate: conditions are defined below: Their particular systems ability to meet ̇ Pre-Natural Hazard Water hydraulic requirements including existing and System Condition 2b Serviceability future water needs (i.e. fire flow, maximum day ̇ Post Natural Hazard Water ALA (2002) or MD and maximum hour or MH domestic System Condition needs, storage needs, etc) and to properly size ̇ Water System Restoration future facilities ̇ Water System Start Up Condition Examines: - Life cycle cost minimum criterion (minimum Application for water supply lifeline Investment cost expected costs on seismic investment) network located in the metropolitan Imai and Koike (2010) 3 for upgraded - Cost benefit ratio criterion area of Japan - Positive value balance criterion. Accounts: ̇ Considers 90 % of customers Restoration process after the physical damage restored within 3-days following an of the network earthquake having a 10% chance of exceedance in 50-years ̇ A typical water utility will be able Performance 4 to isolate most of the leaking and ALA (2005) level criterion broken pipes within 1 day or so Propose a diagram between equivalent damage ratio (km) and Percentage (%) of customers with water

D3.9 - Fragility functions for water and waste-water system elements

3.9.2 Waste-Water System/ component performance indicators

For waste-water system, ALA (2004) propose as performance indicators, capacity measures (e.g. flow of waste-water at selected points); measures of reliability (such as frequency and magnitude of sanitary or combined sewer overflows (SSOs, CSOs), and the frequency and magnitude of discharge of inadequately treated sewage, percentage treated, etc.); measures of safety and health (similar to reliability examples as they impact water quality); and financial measures. The Environmental Protection Agency National Pollution Discharge Elimination System (EPA NPDES) permit requirements incorporate relevant performance measures such as discharge volume and water quality. Potential metrics recommended for the performance of waste-water system according to ALA (2004), are: 1) Public health/backup of raw sewage: This accounts for the probability of achieving performance objective (e.g. – 90% probability of achieving), the probabilities of occurrence (e.g. 50% in 50 years) and different criteria as a function of method of contact (backup into buildings, overflow onto city streets). 2) Discharge of raw/inadequately treated sewage: Metrics commonly used quantify the impact on public health and the environment (e.g. flow associated with biochemical oxygen demand, dissolved oxygen of the receiving water). 3) Direct damage/financial impact: Direct damage to waste-water system components can include cleanup and repair costs associated with flood inundation of a treatment plant or repair cost of the collection system (pipelines, tunnels etc) while secondary damage (economical cost) can be occurred to commercial or industrial facilities (e.g., factories shut down) due to loss of waste-water service. 4) Security system performance: The performance objective is stated in terms of probability of limiting raw sewage discharge when subjected to a design basis threat. Moreover, performance indexes for waste-water system can account “Societal Factors” (ALA, 2004): o Fines and/or jail time - resulting from illegal discharges. o Loss of public confidence – resulting from release of raw sewage, back-up of raw sewage into households, or discharging partially treated sewage into the receiving body. o Political – resulting from peer pressure from other regional waste-water organizations, or local politicians concerned about discharge of raw or partially treated sewage in their area. o Public health and safety – injury or death to utility staff or the public due to exposure to raw or partially treated sewage, chemical release, or building collapse In addition, several other factors (economic factors) can describe waste-water performance (ALA, 2004) such as: o Substantial fines levied by regulating authorities. o Direct loss - repair costs of facilities damaged in hazard events. o Capital improvement plan – identify and prioritize projects to optimize a capital improvement plan.

D3.5 -Fragility functions for water and waste-water system elements

o Project design – define capacity, reliability or other parameters to optimize a new project. o Level of service (outage time) – define expected service outage times associated with various events with associated probabilities of occurrence.

Simpler, waste-water system performance indicators can be described by waste-water flow [m3/h], discharge / number of people supplied / km2 treated (or ratio per zones [%]) / ratio of critical facilities supplied [%].

D3.9 - Fragility functions for water and waste-water system elements

Table 3-9 Summary of Waste-Water Component Performance Indicators (PPIs).

A/A Approach Component Description Reference 1 Operability Collection* and treatment systems Achieving performance objective (% probability of achieving) ALA (2005) 2 Functionality Collection and treatment systems Estimation of violation maximum duration e.g. 7 days, 30 days

*The collection and conveyance system is the system of pipes that collects the sewage from the sources and conveys it to a central point for treatment and/or disposal.

Table 3-10 Summary of Waste-Water System Performance Indicators (WWSPIs) – ALA (2004)

Performance Objective Category 100-Year Return Event 500-Year Return Reference (40% in 50 years) Event (10% in 50 years) Public Health Not acceptable (less than 1% Not acceptable (less than 5% probability Backup of any raw sewage into buildings probability of occurrence) of occurrence) Acceptable in localized areas; Acceptable (treatment plant is Overflow of raw sewage into streets less than 24 hrs inundated) less than 72 hrs Environmental Discharge of raw sewage to stormwater Acceptable in localized areas; Acceptable system, ditch or stream less than 72 hrs less than 7 days Acceptable in accordance with Acceptable ALA (2004) Discharge of raw sewage to lake or river CSO/NPDES less than 30 days Acceptable in accordance with Acceptable Discharge of raw sewage to salt water CSO/NPDES less than 90 days Acceptable Acceptable Discharge of disinfected primary effluent less than 30 days less than 180 days Discharge of disinfected secondary effluent Acceptable Acceptable (meet NPDES permit requirements

D3.9 - Fragility functions for water and waste-water system elements

4 Fragility functions for water and waste-water system elements

4.1 STATE-OF-THE-ART FRAGILITY CURVES PER COMPONENT OF WATER SYSTEM

Table 4-1 presents a brief review of existing fragility functions for water source, water treatment plant, pumping station, storage tanks, pipes, tunnels, canals and conduits

D3.9 - Fragility functions for water and waste-water system elements

Table 4-1 Review of existing fragility functions for potable water elements * Anchored equipment in general refers to equipment designed with special seismic tiedowns or tiebacks, while unanchored equipment refers to equipment designed with no special considerations other than the manufacturer's normal requirements.

Earthquake Component Reference Methodology Classification Damage States and Index descriptor Water NIBS, HAZUS – empirical fragility Complex component. Peak Ground Five damage states:

Source 2004 functions. A distinction is made Acceleration None (ds1), slight/minor (ds2), Two parameters (median according to: (PGA) moderate (ds3), extensive (ds4) and and standard deviation く) - Subcomponents complete (ds5). log-normal cumulative (anchored or unanchored) Index: distributions. Description of the type and extent (level) of structural damage and serviceability state.

Water NIBS, HAZUS – empirical fragility Complex component. Peak Ground Five damage states:

Treatment 2004 functions. A distinction is made Acceleration None (ds1), slight/minor (ds2), Plant Two parameters (median according to: (PGA) moderate (ds3), extensive (ds4) and and standard deviation く) - Subcomponents complete (ds5). log-normal cumulative (anchored or Index: distributions. unanchored)* Description of the type and extent - Size (small, medium or (level) of structural damage and large) serviceability state. Water SRM-LIFE HAZUS – empirical fragility Complex component. Peak Ground Five damage states: Treatment 2003- 2007 functions. - anchored Acceleration None (ds1), slight/minor (ds2), Plant Two parameters (median subcomponents (PGA) moderate (ds3), extensive (ds4) and and standard deviation く) independently from the complete (ds5). log-normal cumulative size Index: distributions. Description of the type and extent (level) of structural damage and serviceability state.

D3.5 -Fragility functions for water and waste-water system elements

Earthquake Component Reference Methodology Classification Damage States and Index descriptor Water Ballantyne TCLEE 2009 Complex component. Peak Ground Three damage states: Treatment et al., 2009 Each of the WTP’s Acceleration Light, Moderate, Severe Plant There are no fragility system components were (PGA) and According to: Permanent curves given evaluated using: Functionality and Restoration Cost Ground ASCE Seismic Evaluation (% of replacement cost) of Existing Buildings Deformation (ASCE 31.03) (PGD) American Concrete Institute Code Requirements for Environmental Engineering Concrete Structures (ACI- 350- 06). Pumping NIBS, HAZUS – empirical fragility Anchored or unanchored Peak Ground Five damage states:

Station 2004 functions. subcomponents Acceleration None (ds1), slight/minor (ds2), Two parameters (median (PGA) moderate (ds3), extensive (ds4) and and standard deviation く) complete (ds5). log-normal cumulative Index: distributions. Description of the type and extent (level) of structural damage and serviceability state. Pumping SRM-LIFE Empirical fragility Anchored or unanchored Peak Ground Five damage states: Station 2003- 2007 functions. subcomponents Acceleration None (ds1), slight/minor (ds2), Two parameters (median (PGA) moderate (ds3), extensive (ds4) and and standard deviation く) complete (ds5). log-normal cumulative Index: distributions. Adapted to Description of the type and extent SRM-LIFE BTM (Kappos (level) of structural damage and et al., 2006) serviceability state.

D3.9 - Fragility functions for water and waste-water system elements

Earthquake Component Reference Methodology Classification Damage States and Index descriptor Pumping Scawthorn, No fragility functions There are no fragility - Reliability index: Station 1996 Pumping station fault tree curves given for Low, Moderate, High diagram. subcomponents. HAZUS – empirical fragility Above ground RC tanks Peak Ground None, Slight, Moderate, Extensive, functions. Acceleration Complete Storage NIBS, Two parameters (median (PGA) Description of the type and extent tanks 2004 and standard deviation く) (level) of structural damage and loss log-normal cumulative of context distributions. Storage O’Rourke Empirical fragility On-grade steel tanks Peak Ground Four damage states: tanks and So, functions. Height to diameter ratio, Acceleration None (ds1), slight/minor (ds2), 1999 amount of stored content (PGA) extensive (ds3) and complete (ds4). Description of the type and extent (level) of structural damage in the roof and loss of context

D3.5 -Fragility functions for water and waste-water system elements

Earthquake Component Reference Methodology Classification Damage States and Index descriptor Storage ALA, Empirical A distinction is made according to: Peak Ground Four damage states tanks 2001a, b fragility - Anchorage Acceleration according to: functions - Material (redwood, steel, post-tensioned (PGA) & - Roof damage circular concrete tank, R/C) Permanent - Anchor bolts - Size (according to gallons) Ground damage Deformation - Seismic design (no, nominal) - Overflow pipe (PGD) - Roof (integral shell roof, wood roof, over open damage cut reservoir) - Elephant foot buckle Types: - Inlet pipe leak - Unanchored redwood tank (50,000 - 500,000 - Wall uplift gall) - Elephant foot buckle - Unanchored post-tensioned circular concrete - Hoop Overstress tank (1,000,000+ gallons)

- Unanchored steel tank with integral shell roof

(100,000 - 2,000,000 gallons)

- Unanchored steel tank with wood roof (100,000 - 2,000,000 gallons) - Anchored steel tank with integral steel roof (100,000 - 2,000,000 gallons) - Unanchored steel tank with integral steel roof (2,000,000+ gallons) - Anchored steel tank with wood roof (2,000,000+ gallons) - Anchored reinforced (or prestressed) concrete tank (50,000 - 1,000,000 gallons) - Elevated steel tank with no seismic design - Elevated steel tank with nominal seismic design - Roof over open cut reservoir

D3.9 - Fragility functions for water and waste-water system elements

Component Reference Methodology Classification Earthquake Damage States and Index descriptor Katayama Empirical relation According to the soil Peak Ground Repair rate per km et al., 1975 log(R.R/km)= A+6,39*log(PGA) conditions and pipeline age Acceleration (A- coefficient) (PGA) (g)

Eguchi, Empirical numbers According to material - Y: break/ 1000 feet 1983 Y= 1.5 ( Asbestos Cement (AC) Y= 1.0 (Cast- iron (CI) Y= 0.8 ( Welded steel with Caulked joints (WSCJ) Y= 0.7 ( Welded steel with Gas- welded joints (WSGWJ) Pipe Y= 0.1 ( Welded steel with Arc- welded joints (WSAWJ) ATC- - Buried pipelines - None, Slight, Light, 13,1985 Moderate, Heavy, Major, Destroyed (based on RR/km) Isoyama Empirical relation (RR/km)= For Cast iron pipes Peak Ground Repair rate per km and 1.698*10-16*PGA6.06 Acceleration Katayama, (PGA) 1982

Memphis, Empirical relation According to soil conditions 』mm: Mercalli N: Repair rate per km Ã Ô Tennessee gÄ MMI / く Õ and diameter Intensity n ? C C 10 Å Ö , 1985 d g

D3.5 -Fragility functions for water and waste-water system elements

Component Reference Methodology Classification Earthquake Damage States and Index descriptor

Wang et Empirical relation According to soil conditions 』mm: MSK intensity Y: Breaks/ km al., 1991 Poor soil conditions: Log Y= 1.837*(I) -14.065 Average soil conditions: Log Y=1.717*(I)-14.221 Good soil conditions: Log Y=1.522*(I)-14.045 O’ Rourke Empirical relation RR/km= According to pipe material Peak Ground Repair rate per km and Ayala, K*(0.0001*PGV2.25) (flexible, rigid) Velocity (PGV) 1993 (cm/sec)

Eidinger et Empirical relation (RR/km) According to pipe material Peak Ground Repair rate per km al., 1995; 0.0012*PGV0.7677 (asbestos-cement, cast – Velocity (PGV) Eidinger, 0.0006*PGV1.5542 iron, steel) (cm/sec) 1998 6*10-5 * PGV2.2949 Isoyama, Empirical relation According to pipe material Peak Ground Repair rate per km 1998 RR/km = and diameter Velocity (PGV) -3 1.3 (cm/sec) Cp*Cd*3.11*10 * (PGV-15)

O’Rourke Empirical relation Peak Ground Repair rate per km et al., 1998 RR/km = Acceleration 2 101.25log10(PGA-0.63) (PGA) (cm/sec ) O’Rourke Empirical relation According to diameter Peak Ground Repair rate per km and Leon, RR/km = Velocity (PGV) 1999 0.865 (cm/sec) 0.050*(vscaled) , 1.138 vscaled = PGV/ Do

D3.9 - Fragility functions for water and waste-water system elements

Component Reference Methodology Classification Earthquake Damage States and Index descriptor Eidinger Empirical relation According to pipe material, Peak Ground Repair rate per km and Avila, RR/km = diameter, joint type and soil Velocity (PGV) 1999 K1*1.512*(PGV1.98) (m/sec)

Isoyama et Empirical relation For Cast – iron pipes Peak Ground Repair rate per km al., 2000 R.R(ゅ) = 2.88*10-6*(PGA-100)1.97 Acceleration 2 R.R(ゅ) = 3.11*10-3*(PGV-15)1.3 (PGA) (cm/sec ) ]LA, 2001 Empirical relation According to pipe material Peak Ground Repair rate per km R.R/km =K1* 0.241*PGV Velocity (PGV) (m/sec) Pipe Toprak, Empirical relation For all buried pipes Peak Ground Repair rate per km 1998 Log(RR)=1.36*log(PGA)-0.61 Acceleration (PGA) Hung, 2001 Empirical relation For all buried pipes Peak Ground Repair rate per km 4.29 RR/km=26.58*PGA Acceleration (PGA) (cm/sec2)

O’Rourke Empirical relation Buried pipelines Peak Ground Repair rate per km and Deyoe, (rigid pipes) Brittle pipes, R or S waves Velocity (PGV), 2004 R.R./km =k1*513* i0.89 strain (wave propagation) R.R./km =k1*724* i0.89 (wave propagation & permanent deformation) Porter et Empirical relation According to pipe material Permanent al., 1991 Ground Deformation (PGD) (inches)

D3.5 -Fragility functions for water and waste-water system elements

Component Reference Methodology Classification Earthquake Damage States and Index descriptor Honegger Empirical relation According to pipe material Permanent Repair rate per km and R.R/km =【*(7.821*PGD0.56) Ground Eguchi, Deformation 1992 (PGD) Empirical relation According to pipe material Permanent Repair rate per km 100*[1-exp[(0.283*PGD)1.33]] and joint type Ground 100*[1-exp[(0.899*PGD)1.11]] Deformation (PGD) (m) 100*[1-exp[(0.578*PGD)1.55]]

100*[1-exp[(1.120*PGD)1.69]] Heubach, 100*[1-exp[(0.743*PGD)0.71]] 1995 100*[1-exp[-(1.120*PGD)0.761]] 100*[1-exp[-(0.644*PGD)1.37]] 100*[1-exp[-(1.530*PGD)1.62]] 100*[1-exp[-(0.961*PGD)1.64]] 100*[1-exp[-(1.830*PGD)1.83]] Eidinger et Empirical relation According to pipe material Permanent Repair rate per km al.,1999 R.R./km =K2*23.674*(PGD)0.53 and joint type Ground Deformation (PGD) (m) ]LA, According to pipe material Permanent Repair rate per km 2001a,b Empirical relation and joint type Ground R.R./km = K2*11.223*PGD0.319 Deformation (PGD) (m) Pipe Yeh et al., Empirical relation Ji – Ji earthquake Peak Ground Repair rate per km 2006 RR = 1.028ゅ10-3* PGA0.9735 (R2= Acceleration 0.9388) (PGA)

D3.9 - Fragility functions for water and waste-water system elements

Component Reference Methodology Classification Earthquake Damage States and Index descriptor Ballantyne Empirical figure According to material Permanent Five Vulnerability Class and (welded steel, old steel and Ground (High, Moderate- High, Heubach, cast iron, locked converse, Displacement Moderate, Low- Moderate, 1996 asbestos cement, cast iron (PGD) Low) according to Damage post 1955) Rate Tunnel As in Roadline System

Canal ALA, Empirical Minor damage: Peak Ground Four Vulnerability Class 2001a,b 0.1 repairs/ km Velocity (PGV) (No, Minor, Moderate, (PGV = 20 - 35 inches/sec) and Permanent Major) Ground 0.01 repairs/ km Index: according to Deformation hydraulic performance of a (PGV = 5 - 15 inches/sec) (PGD) canal 0 below PGV < 5 inches/sec Moderate damage: for PGD= 1-5 inches Major damage: for PGDs > 6 inches

D3.9 - Fragility functions for water and waste-water system elements

4.2 STATE-OF-THE-ART FRAGILITY CURVES PER COMPONENT OF WASTE-WATER SYSTEM

Table 4-2 presents a brief review of existing fragility functions for waste landfill, waste-water treatment plant, lift station, pipes, tunnels and conduits Table 4-2 Review of existing fragility functions for waste-water system elements

Damage States and Component Reference Methodology Classification Earthquake descriptor Index Solid Waste Matasovic et According to the real - - Five damage categories: Landfill al., 1998 damage observations V: Major damage, IV: Significant damage, III: Moderate damage, II: Minor damage, I: Little or No damage Index: According to restoration process (need time to repair) Waste- NIBS, 2004 HAZUS – empirical Complex component. Peak Ground Acceleration Five damage states: water fragility functions. A distinction is made (PGA) None (ds1), slight/minor Treatment Two parameters according to: (ds2), moderate (ds3), Plant (median and standard - Subcomponents extensive (ds4) and deviation く) log-normal (anchored or complete (ds5). cumulative distributions. unanchored)* Index: - Size Description of the type (small, medium or large) and extent (level) of structural damage and serviceability state.

D3.5 -Fragility functions for water and waste-water system elements

Damage States and Component Reference Methodology Classification Earthquake descriptor Index Waste- SRM-LIFE, SRM-LIFE based on Complex component with Peak Ground Acceleration Five damage states: water 2003- 2007 HAZUS empirical anchored subcomponents (PGA) None (ds1), slight/minor Treatment fragility functions. independently from the (ds2), moderate (ds3), Plant Two parameters size but according to the extensive (ds4) and (median and standard building type complete (ds5). deviation く) log-normal Index: cumulative distributions. Description of the type and extent (level) of structural damage and serviceability state. Conduits Mataki et al., Design Code According to “Earthquake Strain 1996 Compression strain: Resistant Design code for Gas Pipeline (High- ic = 35*te/ Dm (%) Pressure)” and Tensile strain: it = 3% “Earthquake Resistant Design Code for Gas Pipeline (Medium & Low Pressure) Lift Station As in potable water system Pipes As in potable water system Tunnel / As in roadline system

Buildings See Task 3.1 See also Task 3.2

D3.5 -Fragility functions for water and waste-water system elements

4.3 VALIDATION / ADAPTATION / IMPROVEMENT

Recent destructive earthquakes (Kocaeli, Ms=7.8, 17-08-1999 & Düzce, Ms= 7.3, 12-11- 1999 in Turkey and Lefkas, Ms=6.4, 14/8/2003 in Greece) provoked important damages to lifelines due to ground shaking or/and permanent ground deformations.

Fig. 4-1 Location of Düzce and Lefkas island

Few hundreds of damages to buried pipelines of the water supply systems and waste-water network were reported in Düzce while in Lefkas (Fig. 4-1) the reported damages were much lower but equally important. The aim of this section is to compare the estimated damages with the observed and reported ones in the two cities, in order to validate existing fragility curves. This comparative study is one of the first well-documented cases in the whole Mediterranean region, where we have an important lack of data regarding lifeline damages during earthquakes. The methodology applied is based on a detailed inventory of the observed damages and a site-specific ground response analysis to simulate the spatial variability of ground motions during the two severe earthquakes occurred in Düzce and Lefkas. Several studies were performed in Düzce (Alexoudi, 2005; Pitilakis et al., 2005; Alexoudi et al., 2007, 2008, 2010) for water and waste-water system aiming to record the observed damages and to compare with the computed ones obtained when several commonly used fragility curves are applied.

D3.5 -Fragility functions for water and waste-water system elements

4.3.1 Validation of vulnerability models for pipes

4.3.1.1 Düzce

Düzce is situated between Ankara and Istanbul and is located nearby the North Anatolian Fault (NAF) and next to Düzce fault that is small branch of NAF. Two major earthquakes Kocaeli (17/8/1999, 40.702 N, 29.987 E, Mw= 7.4) and Düzce (12/11/1999, 31.15E, 40.77N, Mw= 7.2, h=10km) earthquakes occurred in the area provoking important damages in Düzce.

DÜZCE POTABLE WATER SYSTEM The water supply system in Düzce dates back to 1940’s. The pre-existing network is thought to be about 500 km in length, although no maps exist to confirm this (Tadday and Sahin, 2001). This old network was still in use at the time of Kocaeli and Düzce earthquakes. The new network is connected to the old one with a series of bypasses. The old network is mainly CI (cast iron), with some AC (Asbestos cement) pipes. The whole distribution network is therefore made up of pipes normally classified as brittle. A 600mm diameter AC pipe conveys raw water from the main source, the River Ugur, to the water treatment plant which lies to the south of the town. A 1m diameter steel pipe then carries the treated water to the distribution network, joining the town in the Azmimilli District. Twin CI pipes, of diameter 125mm, transport water from a well-field and reservoir to supplement the main river water supply; these pipes join the town in the north-east. The digitized network is a mixed system as is comprised by some old water branches and the new network. The total length is 298km and the average depth of the water pipes of Düzce water system is 1.50m. A site-response study was conducted for the city of Düzce using as input the deconvoluted time history of the 17/8/1999 Kocaeli and 12/11/1999 Düzce main-shock that was recorded in the Meteorological Station. The geotechnical map for Düzce derived from the existing geological and geotechnical data, numerous very shallow, 10-20m, boreholes and few (10) well documented deep (40-100m) boreholes which were collected in the framework of a research project (SRM-DGC, 2006-2008). Bedrock’s depth (B120m) was defined using both geological and seismic data (compilation of a large number of aftershocks at the Meteorological Station and estimation of the H/V spectral ratio). The above result was also validated with data from topographic maps of the area as well as microtremor measurements (Kudo et al., 2000; Rosenblad et al., 2001). Using the aforementioned geotechnical and geological data, numerous 2D cross-sections were constructed along the city of Düzce. Based on the 2D cross-sections, approximately thirty typical soil profiles were proposed in specific sites along the city in order to perform a set of 1D equivalent linear analysis. The spatial distribution of the computed mean values of the peak ground acceleration and peak ground velocities combined with the digitized water system is presented in Fig. 4-2 for both Kocaeli and Düzce earthquakes respectively.

D3.5 -Fragility functions for water and waste-water system elements

a(1) a(2)

b(1) b(2)

Fig. 4-2 Düzce. Analyzed method: 1D linear equivalent, Local Soil Condition: Based on Soil Profiles, a) Earthquake: Kocaeli, 1999, PGA (g) [a(1)], PGV (m/sec) [a(2)], b) Earthquake: Düzce, 1999, PGA (g) [b(1)], PGV (m/sec) [b(2)]

In order to validate the available fragility curves for water pipes, different vulnerability functions were selected in order to compare the estimated damages in Düzce (Turkey) water pipeline network with the observed ones after Düzce and Kocaeli earthquakes. Table 4-3 gives the computed water pipe failures due to ground shaking for four different fragility expressions and the two input motion for the digitized network of 298.81km. For 2 months after Kocaeli and Düzce earthquakes, about 298 and 238 potable water pipe failures respectively were recording by Tromans (2004) in a water network of 433.60km in 29 mahallas. After the available transforming of the two lengths (298.81/433.60) in order to compare the results, the recorded water pipe damages are 200 and 164 for Kocaeli and

D3.5 -Fragility functions for water and waste-water system elements

Düzce earthquake respectively. The average monthly repairs before the earthquakes were 95 and the real water losses were calculated to 80% of the initial supply.

Table 4-3 Computed water pipe failures in the water network of Düzce due to ground shaking for different fragility expressions, and input motions (Alexoudi et al., 2010)

Fragility curves/ Earthquake DÜZCE KOCAELI

O’ Rourke and Ayala (1993) 147 116

Isoyama et al. (1998) 80 66

Eidinger and Avila (1999) 104 84

]LA (2001) 28 25

Recorded 164 200

Comparing the computed (Table 4-3) and the recorded damages, after excluding the average pre-earthquake monthly repairs, it derives that the O’ Rourke and Ayala (1993) relation describes better the real event given the inherent uncertainties in the pipes individual characteristics and the recorded damages from Kocaeli earthquake. ALA (2001) fragility curve, underestimates the failures induced by wave propagation compared with other relations, while the failures that Eidinger and Avila (1999) predicts is 20 - 30 % lower compared to the recorded ones from the two earthquakes. The estimated failures by Isoyama et al. (1998) relation, are about the half of the ones that are obtained when the O’ Rourke and Ayala (1993) relation is applied. It is noticed that the recorded failures from Kocaeli earthquake is unjustified larger compared with the ones from Düzce earthquake, although the parameters of input motion and Aries Intensity connected with Düzce earthquake is 2 times larger than the Kocaeli earthquake. Also, Düzce earthquake had larger duration compared with Kocaeli earthquake. Moreover, for Eidinger and Avila (1999) and O’Rourke and Ayala (1993) fragility relations, a spatial distribution of the computed damages in each mahalla is presented in Fig. 4-3 and Fig. 4-4 for the two earthquakes. Analyzing the results, it is shown that the spatial distribution of damages of the O’Rourke and Ayala relation is generally well correlated with the Tromans (2004) and Alexoudi (2005) recorded data (Fig. 4-5).

D3.5 -Fragility functions for water and waste-water system elements

pipe_failure ¯ pipe_failure ¯ REHAZKOC REEIDKOC Low Low Moderate Moderate

High High

1 1

3 5 3 5

2 2 4 4

7 8 7 8

6 10 6 10

12 12 9 11 9 11 13 13

18 14 18 14

19 20 19 20 16 17 21 16 17 21

24 22 15 24 22 15 27 23 27 23 25 25

28 29 28 29 26 26

0550 1,100 2,200 3,300 4,400 0550 1,100 2,200 3,300 4,400 a) km b) km

Fig. 4-3 Mahallas that present low, moderate and extensive failures as result of Kocaeli earthquake and O’Rourke and Ayala (1993) (a) and Eidinger and Avila (1999) (b) relationships. The points represent the well documented damages shown earlier. Earthquake: Kocaeli 1999, Microzonation study (Alexoudi et al. , 2007)

Legend Legend waterfailure ¯ waterfailure ¯ REHAZDUZ 1 REEIDDUZ 1 Low Low Moderate Moderate High 3 5 High 3 5

2 2 4 4

7 8 7 8

6 10 6 10

12 12 9 11 9 11 13 13 18 14 18 14 [ 19 20 21 [ 19 20 16 17 16 17 21

24 22 15 24 22 27 23 27 23 15 25 25

28 29 28 29 26 26

0600 1,200 2,400 3,600 4,800 0600 1,200 2,400 3,600 4,800 a) km b) km

Fig. 4-4 Mahallas that present low, moderate and extensive failures as result of Düzce earthquake and O’Rourke and Ayala (1993) (a) and Eidinger and Avila (1999) (b) relationships. The failures collected are illustrated with points. For each mahalla, ID is corresponded. Earthquake: Düzce 1999, Microzonation study. (Alexoudi et al., 2007)

D3.5 -Fragility functions for water and waste-water system elements

Legend Legend waterfailure ¯ waterfailure ¯ REPRESEISM 1 REKOCAEL 1 Low Low Moderate Moderate High 3 5 High 3 5 2 4 2 4

7 8 7 8

6 10 6 10 12 9 11 13 12 9 11 18 14 13 [ 19 20 18 14 16 17 21 [ 19 20 21 24 22 16 17 27 23 15 25 24 22 27 23 15 28 29 25 26 28 29 26

0600 1,200 2,400 3,600 4,800 km 0600 1,200 2,400 3,600 4,800 a) b) km

Legend Legend waterfailure ¯ waterfailure ¯ REDUZCE 1 REEBRU 1 Low Low Moderate Moderate High 3 5 High 3 5

2 2 4 4

7 8 7 8

6 10 6 10

12 12 9 11 9 11 13 13 18 18 14 14 [ 19 20 [ 19 20 16 17 21 16 17 21

24 22 24 22 15 27 23 15 27 23 25 25

28 29 28 29 26 26

0600 1,200 2,400 3,600 4,800 0600 1,200 2,400 3,600 4,800 c) km d) km

Fig. 4-5 Mahallas that presents low, moderate and extensive failures (a) before the two earthquakes, (b) after Kocaeli earthquake, (c) after Düzce earthquake (d) present research as result of both earthquakes. Points illustrate the failures collected while the ID corresponds to each mahalla.

DÜZCE WASTE-WATER SYSTEM The waste-water supply system in Düzce is a gravity network that dates back to the 1940’s although several parts of the system are dating back to the early 1900’s. The pre-existing network is estimated to be about 300 km in length, although no maps exist to confirm this. Both old and new networks were in use at the time of Kocaeli and Düzce earthquakes. The parts of the network that was digitized consist of 50.60km pipes-conduits with circular shape while the rest (3.44km) has different shapes (rectangular, oval, and orthogonal. The material

D3.5 -Fragility functions for water and waste-water system elements of waste-water pipes is concrete and the distribution of their diameters is illustrated in Fig. 4-6. Information about the dimension, the shapes and the material for the rest network is not available. Taking into account the 93% of the material type of waste-water pipes, whole network can be characterized as a brittle network (Alexoudi, 2005).

Waste-Water pipes/ tunnel (diameter)

4000mm 4000mm 1200mm 1000mm 1000mm 800mm 900mm 800mm 400mm 600mm

Diameter (mm) 400mm 200mm 300mm 0,00 5,00 10,00 15,00 20,00 200mm Length (km)

Fig. 4-6 Digitized Waste- Water network (left) in Düzce and distribution of waste-water pipe/ conduits diameters (up)

Applying, O’ Rourke and Ayala (1993) fragility function we estimate a total number of 52 damages (10 breaks, 42 leaks) and 44 damages (9 breaks, 35 leaks) as a result of ground shaking for Düzce and Kocaeli earthquake respectively (Fig. 4-7).

4% 5% 10% 11% Break Leak Break No- damage Leak No- damage

84% a) 86% b)

Fig. 4-7 Estimated damages of waste-water network as percentage of the total length of the network for Kocaeli (a) and Düzce (b) earthquake (Alexoudi et al., 2008)

The spatial distribution of the damages of waste-water network as result of Düzce and Kocaeli earthquake is illustrated in Fig. 4-8. Tromans (2004) database for water pipes was used for the validation of the estimated damages of waste-water system resulted from the conducted vulnerability assessment of Kocaeli and Düzce earthquakes. It is assumed that the failures of water system of Düzce are quite similar to the damages of waste-water system, an estimation that is made by the

D3.5 -Fragility functions for water and waste-water system elements

Waste - Water Company of Düzce, although, some individual characteristics of the two networks can enlarge the different seismic response of the two networks. In particular, the material, the oldness of the network and the construction practice can alter greatly the response of a pipe.

a) b)

Fig. 4-8 Spatial distribution of waste-water pipe damages in Düzce network for Kocaeli (a) and Düzce (b) earthquake (Alexoudi et al., 2008)

A comparison between the recorded water pipe damages derived from Tromans (2004) database and the estimated damages of waste-water system as result of Kocaeli earthquake are illustrated in Fig. 4-9 a,b. It can be noticed that the expected damages from the two earthquakes are located in the southern part of the city in almost the same mahallas that important damages in potable water system were observed and high PGV values were calculated. For the Düzce earthquake the corresponding damages (Fig. 4-9 c,d) have some minor differences, mainly due to the limited time for recovery between the two earthquakes. Moreover, the damages in waste-water system are very hard to recognize as they are not related with the reduction of pressure or even flow and there were no available records before and after the earthquakes.

D3.5 -Fragility functions for water and waste-water system elements

a) b)

c) d)

Fig. 4-9 Estimated waste-water pipe damages per mahalla for Kocaeli earthquake (a), for Düzce earthquake (c) and recorded water pipe damages per mahalla after Kocaeli earthquake (b) and Düzce earthquake (d) - (Alexoudi et al., 2008)

4.3.1.2 LEFKAS

LEFKAS POTABLE WATER SYSTEM Lefkas water supply system was constructed in 1978 for daily design consumption of 5400m3. It was designed to provide water for drinking and fire protection. Nowadays, in winter time, it serves about 9.000 people (old city and new parts of the city) and more than 12.000 people in touristic period (May-October). 29.114 km of pipes are in the old city and more than 20 km in the new city. The main water source is river Louros at the main land, but in touristic period the city uses also ground water supply from 2 shallow wells (150m3/ day/ well). Moreover, an R/C tank with a capacity of 1000m3 serves distribution network, as an external reservoir in order to cover the increased summer daily demands. Lefkas potable water distribution network is composed by 86% PVC pipes (internal pressure: 10atm) with special couplings in the joints and about 14% asbestos-cement pipes (older than 1978). The

D3.5 -Fragility functions for water and waste-water system elements system was in very good condition with a very small number of pre-earthquake reported leaks. After the August 2003 seismic event (14/08/2003, Ms=6.4) the main water network of the city of Lefkas suffered 10 failures in water mains (old city), 5 in the marina area and more than 80 damages in service connections in both parts of the city. The location of the failures is illustrated in Fig. 4-10.

Fig. 4-10 Water distribution network of old city of Lefkas and the location of main water system failures and secondary connections (p-primary network, sec-secondary network-connections with customers.

In general, the damages observed along the coastline (3 damages- 1 double damage) and in “Gyra” (3 damages), resulted from permanent ground deformation due to soil liquefaction. The rest 4 damages can be attributed to wave propagation and material failures. The failure modes that were observed for PVC and asbestos cement pipes in Lefkas earthquake were direct failures of the pipe body and a slip-out of joints. The failures in Marina and in coastline are attributed to the large vertical and horizontal displacements due to liquefaction. The basic geotechnical-geological formations in the city of Lefkas are constituted from recent deposits (present at depths varying from 10.6 to 16.0m) overlying to a stiff to hard marl layer extended to the bedrock surface. Top deposits include an upper layer of soft to medium cohesive soils (shear wave velocities Vs=180-250m/sec2) with locally situated layers of loose saturated sandy-silty soils, quite susceptible to liquefaction, mostly present at the coastal parts of the examined region, underlying a layer of debris 1.0-5.7m deep. The lower layer of the deposits, are medium clays and silts in the central part of the city, while in the coastal region medium to dense layers of silty sands prevail. The soil classification and simplified geotechnical characterization of the area was based on several cross sections along with the information from laboratory and in-situ tests (mostly NSPT). Shear wave velocities were estimated using both existing cross-hole data and empirical correlations with NSPT, which seemed to be in reasonable agreement with the available cross-hole data. The geotechnical information is based on 17 geotechnical boreholes with SPT and in few cases with cross-hole Vs measurements. The dynamic properties (G-け-D curves) are rather well known from RC tests. The available record of the main shock (PGA-0.45g) is recorded

D3.5 -Fragility functions for water and waste-water system elements

in a site where the soil profile of 60m is very well known with all the necessary data. This was particularly important to conduct the deconvolution analyses. In order to account for the effect of liquefaction phenomena on the ground motion characteristics, several elastoplastic analyses (using the 1D-Cyclic program, 2001) were performed for selected profiles along the coastal part of the city and the marina area, where liquefaction induced phenomena were observed after the earthquake. The latter were conducted using the same input motions with the equivalent linear elastic analyses for wave propagation. The recorded PGV is 39.6cm/sec (EW component) while the computed PGV values vary from 30cm/sec to 46.60cm/sec. The estimated number of repairs based on different fragility curves is presented in Table 4-4 both for wave propagation and permanent ground deformation. A comparison between the number of repairs, the repair rate/km and the observed damages for the potable water network of Lefkas is given in Table 4-5 and Table 4-6 (Alexoudi, 2005; Pitilakis et al., 2005).

Table 4-4 Estimated number of repairs for Lefkas earthquake using different fragility curves

Vulnerability Wave propagation Permanent deformation Combination relations PGVew=30- PGD= 1.0- 40.42cm 46.60cm/sec

O’Rourke RR/kmPGV RR/kmPGD RR/kmPGVPGD and Ayala 4 repairs 4 repairs 6 repairs (1993) & (3 leaks, (1 leak, (2 leaks, Honegger RR/km= 1 break) RR/km= 3 breaks) RR/km= 4 breaks) and Eguchi 0.137 0.137 0.206 (1992) (NIBS, 2004) 3 repairs 26 repairs 25 repairs Eidinger and RR/km= (2 leaks, RR/km= (5 leaks, RR/km= (3 leaks, Avila (1999) 0.103 1 break) 0.893 21 breaks) 0.859 22 breaks)

1 repairs 22 repairs 21 repairs RR/km= (1 leaks, RR/km= (4 leaks, RR/km= (3 leaks, ALA (2001) 0.034 0 break) 0.756 18 breaks) 0.721 18 breaks)

Isoyama et 3 repairs RR/km= al. (1998) (2 leaks, 0.103 1 break) Heubach 9 repairs (1995) RR/km= (2 leaks, 0.309 7 breaks)

D3.5 -Fragility functions for water and waste-water system elements

Table 4-5 Comparison of Repair Rate/km (wave propagation) with the recorded damages of water network of Lefkas

Eidinger O’ Rourke and Isoyama LA Recorded RR/ km and Ayala ] Avila (1998) (2001a.b) damages (1993) (1999)

RRPGV/km 0.137 0.103 0.103 0.034 0.137

Table 4-6 Comparison of the number of failures (wave propagation) for water system of Lefkas

Vulnerability relations No. of failures Recorded damages O’ Rourke and Ayala (1993) 4 Eidinger and Avila (1999) 3 4 Isoyama et al. (1998) 3 ]LA (2001) 1

Applying O’ Rourke and Ayala (1993) fragility relation four damages were estimated for water system in Lefkas for the seismic scenario of 2003 Lefkas earthquake. ALA (2001) underestimates the damages for wave propagation as it predicts only one. For the case of permanent deformation, Honegger and Eguchi (1992) relation estimates 4 damages, while ALA (2001) 22 damages. In general, NIBS (2004) gives very close to the observed failures comparing to ALA (2001) which overestimates the damages for permanent ground deformation. The spatial distribution of estimated damages (lines with red- breaks, with orange- leaks) of potable water system via the recorded ones (points) is given for 4 different fragility curves in Fig. 4-11 to Fig. 4-14 (for the case of wave propagation).

D3.5 -Fragility functions for water and waste-water system elements

! !

! Legend ! waterfsecond

!! ! waterfailures

Waterpipes(PGVHAZREP)

break

leak

full-function

PGV_EW (cm/sec)

High : 46.60cm/sec !

Medium : 38.30cm/sec !

26002130 60m Low : 30.00cm/sec !

Fig. 4-11 Vulnerability assessment of potable water system (Fragility curve: O’ Rourke and Ayala, 1993, Earthquake: Lefkas 2003)

! !

! Legend !

waterfsecond

!! ! waterfailures

Waterpipes(PGVEIDREP)

break

leak

full-function

PGV_EW (cm/sec)

! High : 46.60cm/sec

! Medium : 38.30cm/sec

26002130 60m Low : 30.00cm/sec !

Fig. 4-12 Vulnerability assessment of potable water system (Fragility curve: Eidinger and Avila, 1999, Earthquake: Lefkas 2003)

D3.5 -Fragility functions for water and waste-water system elements

! !

! Legend !

waterfsecond

!! ! waterfailures

Waterpipes(PGVISOY)

break

leak

full-function

PGV_EW (cm/sec)

! High : 46.60cm/sec

! Medium : 38.30cm/sec

260130 0 260 m Low : 30.00cm/sec !

Fig. 4-13 Vulnerability assessment of potable water system (Fragility curve: Isoyama et al., 1998, Earthquake: Lefkas 2003)

! !

Legend !

waterfsecond !!

! waterfailures

Waterpipes(PGVALA)

leak

full-function

PGV_EW (cm/sec)

High : 46.60cm/sec !

Medium : 38.30cm/sec !

260130 0 260 m Low : 30.00cm/sec

Fig. 4-14 Vulnerability assessment of potable water system (Fragility curve: ]LA, 2001, Earthquake: Lefkas 2003)

D3.5 -Fragility functions for water and waste-water system elements

4.4 FINAL PROPOSAL

4.4.1 WATER SYSTEM ELEMENTS

4.4.1.1 Water Source

Wells are complex components that include several subcomponents. HAZUS (NIBS, 2004) gives fragility curves for anchored and for unanchored subcomponents. Although, there are no specific guidelines in Europe, all subcomponent are anchored. In order to account the uncertainty in their final response, a semi- anchorage of subcomponents can be defined. The description of damage states for water source is provided in Table 4-7 while the corresponding fragility curves due to peak ground acceleration are given in Table 4-8.

Table 4-7 Description of damage states for water source subject to ground shaking

Damage Restoration cost Description Serviceability state (%) Malfunction of well pump and motor for a short time Normal flow Operational (less than three days) due Minor 10-30 and water after limited to loss of electric power and pressure repairs backup power if any, or light damage to buildings Malfunction of well pump and motor for about a week due to loss of electric power and backup power if any, Operational Moderate 30-50 considerable damage to after repairs mechanical and electrical Reduce flow equipment, or moderate and water damage to buildings pressure The building being Partially extensively damaged or the operational Extensive well pump and vertical shaft 50-75 after extensive being badly distorted and repairs non-functional Building collapsing Not water Complete 75-100 Not repairable available

D3.5 -Fragility functions for water and waste-water system elements

Table 4-8 Parameters of fragility curves for water source (wells)

Peak Ground Acceleration (PGA) Description Damage state く Median (g) (log-standard deviation) Anchored Minor 0.16 0.70 components (low- Moderate 0.18 0.65 rise R/C building with low seismic Extensive 0.30 0.65 code design) Complete 0.40 0.75 Anchored Minor 0.25 0.55 components (low Moderate 0.45 0.50 height R/C building with advanced Extensive 0.85 0.55 seismic code 2.10 0.70 Complete design)

Wells (anchored components) Low-rise building with low seismic code design 1,00

0,80

0,60

0,40

0,20 [Probability Ds> ds / PGA] ds Ds> [Probability

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g)

Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-15 Fragility curves for wells (Anchored components, low – rise R/C building with low seismic code design) subjected to ground shaking

D3.5 -Fragility functions for water and waste-water system elements

Wells (anchored components) Low-rise building with advange seismic code design 1,00

0,80

0,60

0,40

0,20 [Probability Ds> / PGA] ds

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g)

Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-16 Fragility curves for wells (Anchored components, low – rise R/C building with advanced seismic code design) subjected to ground shaking

Table 4-9 Subcomponent Damage Algorithms for Wells with Anchored Components (SRM-LIFE, 2003-2007)

Peak Ground Acceleration

Damage Median Subcomponents State (g) minor 0.50 0.60 Electric Power (Backup) moderate 0.70 0.80 minor 0.15 0.40 Loss of commercial Power moderate 0.30 0.40 Well Pump extensive 1.00 0.60 Electric Equipment moderate 0.80 0.60 minor 0.18 0.73 Building (low-rise R/C moderate 0.23 0.73 building with low seismic code design) extensive 0.30 0.73 complete 0.41 0.73 minor 0.28 0.73 Building (low height R/C moderate 0.72 0.73 building with advanced seismic code design) extensive 1.66 0.73 complete 2.17 0.73

D3.5 -Fragility functions for water and waste-water system elements

Comment: For the buildings sub-component, the typology and fragility curves proposed in SRM-LIFE (2003-2007) were used. The upgrade of fragility curves will be made after the finalization of D3.1 “Fragility functions for common RC building types in Europe” and the proposal of buildings’ typologies and fragility functions for SYNER-G.

4.4.1.2 Water Treatment Plant

Water Treatment Plants are complex components that include several subcomponents. HAZUS (NIBS, 2004) gives fragility curves for anchored and for unanchored subcomponents for different sizes of Water Treatment Plants. There are no specific guidelines in the anchorage of the subcomponents in Europe for Water Treatment Plants. In order to account for the uncertainty in their final response as a result of the different European practices used for Water Treatment Plants of different sizes and the semi- anchorage of subcomponents, only one fragility curve for Water Treatment Plant is proposed independently of the size. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario). The description of damage states for Water Treatment Plant is provided in Table 4-10 while the corresponding fragility curves are given in Table 4-11.

Table 4-10 Description of damage states for Water Treatment Plant subjected to ground shaking

Damage Restoration Description Serviceability state cost (%) Malfunction of plant for a short time (<3 days) due to loss of electric power, considerable damage to various Normal flow Operational equipment, light damage to Minor 10-30 and water after limited sedimentation basins, light damage to pressure repairs chlorination tanks, or light damage to chemical tanks. Loss of water quality may occur. Malfunction of plant for about a week due to loss of electric power and backup power if any, extensive damage to various equipments, considerable Operational Moderate damage to sedimentation basins, 30-50 after repairs considerable damage to chlorination tanks with no loss of contents, or Reduce flow considerable damage to chemical tanks. and water Loss of water quality is imminent pressure Partially The pipes connecting the different basins operational and chemical units being extensively Extensive 50-75 after damaged. This type of damage will likely extensive result in the shutdown of the plant. repairs The complete failure of all pipings or Not water Not Complete 75-100 extensive damage to the filter gallery available repairable

D3.5 -Fragility functions for water and waste-water system elements

Table 4-11 Parameters of fragility curves for Water Treatment Plant

Peak Ground Acceleration (PGA) Description Damage state く Median (g) (log-standard deviation) Minor 0.15 0.30 Water Treatment Plants with Moderate 0.30 0.25 anchored Extensive 0.55 0.60 components Complete 0.90 0.55

Table 4-12 Subcomponent Damage Algorithms for Water Treatment Plants with Anchored Components

Peak Ground Acceleration Subcomponents Damage State Median (g) Loss of commercial Power minor 0.15 0.40 moderate 0.30 0.40 Chlorination minor 0.50 0.60 Equipment moderate 0.85 0.70 Sediment Flocculation minor 0.36 0.50 moderate 0.60 0.50 Chemical minor 0.35 0.70 Tanks moderate 0.55 0.70 Electric Equipment moderate 0.80 0.60 Elevated Pipe extensive 0.53 0.60 complete 1.00 0.60 Filter Gallery complete 2.00 1.00

D3.5 -Fragility functions for water and waste-water system elements

Water Treatment Plant (anchored components, without back-up power) 1,000

0,800

0,600

0,400

[Probability Ds> ds / PGA] ds Ds> [Probability 0,200

0,000 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-17 Fragility curves for Water Treatment Plant (Anchored components) subjected to ground shaking

4.4.1.3 Pumping Station

Pumping Stations are complex components that include several subcomponents. HAZUS (NIBS, 2004) gives fragility curves for anchored and for unanchored subcomponents for different sizes of Pumping Stations. There are no specific guidelines in the anchorage of the subcomponents in Europe for pumping stations. In order to account for the uncertainty in their final response as a result of the different European practices used for Pumping Stations of different sizes and the semi- anchorage of subcomponents, only one fragility curve for Pumping Stations is proposed independently of the size for different building categories. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario).The description of damage states for pumping station is provided in Table 4-13 while the corresponding fragility curves are given in Table 4-14.

D3.5 -Fragility functions for water and waste-water system elements

Table 4-13 Description of damage states for Pumping Station subjected to ground shaking

Damage Restoration cost Description Serviceability state (%) Malfunction of plant for a Normal flow Operational short time (< 3 days) due to Minor 10-30 and water after limited loss of electric power or pressure repairs slight damage to buildings The loss of electric power for about a week, considerable damage to Operational Moderate 30-50 mechanical and electrical after repairs equipment, or moderate Reduce flow damage to buildings. and water pressure The building being Partially extensively damaged or the operational Extensive 50-75 pumps being badly after extensive damaged beyond repair repairs Not water Complete The building collapsing. 75-100 Not repairable available

Table 4-14 Parameters of fragility curves for pumping station

Peak Ground Acceleration (PGA) Description Damage state く Median (g) (log-standard deviation) Anchored Minor 0.10 0.55 components (low- Moderate 0.15 0.55 rise R/C building with low seismic Extensive 0.30 0.70 code design) Complete 0.40 0.75 Anchored Minor 0.15 0.30 components (low- Moderate 0.30 0.35 rise R/C building with advanced Extensive 1.1 0.55 seismic code Complete 2.1 0.70 design)

D3.5 -Fragility functions for water and waste-water system elements

Pumping station (anchored components, low-rise building with low seismic code design, without back-up power) 1,00

0,75

0,50

0,25 [Probability Ds> ds / PGA] / ds Ds> [Probability

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g)

Minor damages Moder ate damages Extensive damages Complete damages

Fig. 4-18 Fragility curves for pumping station (Anchored components, low-rise R/C building with low seismic code design) subjected to ground shaking

D3.5 -Fragility functions for water and waste-water system elements

Pumping station (anchored components, low-rise building with advance seismic code design, without back-up power) 1,00

0,75

0,50

0,25 [Probability Ds> ds PGA] /

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-19 Fragility curves for pumping station (Anchored components, low -rise R/C building with advanced seismic code design) subjected to ground shaking

Table 4-15 Subcomponent Damage Algorithms for Water Treatment Plants with Anchored Components

Peak Ground Acceleration Subcomponents Damage State Median (g) Loss of commercial Power minor 0.15 0.40 moderate 0.30 0.40 Electric and Mechanical 0.80 0.60 moderate Equipment Vertical/ Horizontal Pump* extensive 1.25/1.60 0.60 Low-rise R/C building with minor 0.18 0.73 low seismic code design moderate 0.23 0.73 extensive 0.30 0.73 complete 0.41 0.73 Low height R/C building minor 0.28 0.73 with advanced seismic moderate 0.72 0.73 code design extensive 1.66 0.73 complete 2.17 0.73

D3.5 -Fragility functions for water and waste-water system elements

Comment: For the buildings sub-components, the typology and fragility curves proposed in SRM-LIFE (2003-2007) were used (Kappos et al., 2006). The upgrade of fragility curves will be made after the finalization of D3.1 “Fragility functions for common RC building types in Europe” and the proposal of buildings’ typologies and fragility functions for SYNER-G.

4.4.1.4 Storage tanks

Different fragility curves are illustrated (Table 4-16 - Table 4-19) by ALA (2001a,b) and HAZUS (NIBS, 2004) for wave propagation (PGA) and for permanent ground deformation (PGD)- (Table 4-20, Table 4-21). In Europe, the more common typology is R/C tanks without anchorage.

Table 4-16 Fragility curves for anchorage R/C at grade tanks (wave propagation)- ALA (2001a,b)

Failure Type Serviceability Median PGA (g) Uplift of wall– 1.30 0.50 Crush concrete Cracking or No operational shearing of tank 1.60 0.50 wall Sliding 1.10 0.50 Hoop overstress Operational 4.10 0.50

Table 4-17 Fragility curves for unanchorage R/C at grade tanks (wave propagation)- ALA (2001a,b)

Median PGA Failure Type Serviceability (g) Cracking or shearing Loss of No operational 1.05 0.45 of tank wall context No loss of Roof damage 2.60 0.45 context Uplift of wall– Crush Operational Small leak 2.00 0.45 concrete Sliding Small leak 0.25 0.45 Loss of No operational 0.75 0.45 Hoop overstress context Small leak Operational 0.45 0.45

D3.5 -Fragility functions for water and waste-water system elements

Table 4-18 Fragility curves for Open reservoirs with or without seismic design code (wave propagation) ALA (2001a,b)

Failure Type Serviceability Median PGA (g) Roof Extensive 1.00 0.55 Operational damage Minor 0.60 0.55

Table 4-19 Fragility curves for unanchorage R/C at grade tanks (permanent deformations)- ALA (2001a,b)

Typology Serviceability Median PGD (m) Anchored R/C 0.06 0.50 Un-anchored No operational At columns 0.06 Steel 0.50 At grade 0.09 Wooden No operational 0.09 0.50 Without roof Operational 0.20 0.50

Table 4-20 Fragility curves for at-grade R/C tanks (wave propagation)- (HAZUS; NIBS, 2004)

Typology Damage states Median PGA (g) Anchored at-grade minor 0.25 0.55 R/C tank moderate 0.52 0.70 extensive 0.95 0.60 complete 1.64 0.70 Unanchored minor 0.18 0.60 at-grade R/C tank moderate 0.42 0.70 extensive 0.70 0.55 complete 1.04 0.60

Table 4-21 Fragility curves for buried R/C tanks (permanent ground deformation)- (HAZUS; NIBS, 2004)

Typology Damage states Median PGA (g) Buried R/C minor 0.05 0.50 tanks moderate 0.10 0.50 extensive 0.20 0.50 complete 0.30 0.50

A comparison between the different R/C tanks based on ALA (2001) and HAZUS (NIBS, 2004) is illustrated in Fig. 4-20 and in Fig. 4-21. In Europe, there is no available studies, as far as we know, that evaluate the two different fragility curves. In SYNER-G, ALA (2001) fragility curves are proposed for the estimation of the vulnerability through the operability of the tanks.

D3.5 -Fragility functions for water and waste-water system elements

Above ground R/C tanks (wave propagation) 1,00

0,80

0,60

0,40 [Probability Ds> ds / PGA]

0,20

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) Minor damages Moderate damages Extensive damages Complete damages ALA_Operative ALA_No- operative

Fig. 4-20 Fragility curves for above ground R/C tanks (wave propagation)

Above ground R/C tanks (permanent deformation)

1,00

0,80

0,60

0,40 [Probability Ds> ds / PGA] 0,20

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGD (m) Minor damages Moderate damages Extensive damages Complete damages ALA_PGD_no_operate

Fig. 4-21 Fragility curves for above ground R/C tanks (permanent ground deformations)

D3.5 -Fragility functions for water and waste-water system elements

4.4.1.5 Canal

Failures in canals can be produced by landslides and by the damage of other infrastructures that can influence the flow.

Table 4-22 Description of damage states for Canals (ALA, 2001a,b)

Damage Description Damage Rate state The canal has the same hydraulic Minor damage to unreinforced liners No damage performance after the earthquake or unlined embankments may be Some increase in the leak rate of the canal expected at Repair Rate/km 0.1 for has occurred. Damage to the canal liner ground shaking velocities of PGV = may occur, causing increased friction 20 to 35 inches/ sec. The minor between the water and the liner and damage rate drops to 0.01 repairs per lowering hydraulic capacity. The liner kilometer for ground shaking damage may be due to PGDs in the form of velocities of PGV = 5 to 15 inches/ settlements or lateral spreads due to sec and 0 below that. Damage to liquefaction, movement due to landslide, reinforced liners is one quarter of Minor offset movement due to fault offset, or these rates. Bounds on the damage excessive ground shaking. Landslide debris estimate can be estimated assuming may have entered into the canal causing plus 100% to minus 50% at the plus higher sediment transport, which could or minus one standard deviation level, cause scour of the liner or earthen respectively. embankments. Overall, the canal can be operated at up to 90% of capacity without having to be shut down for make repairs. Some increase in the leak rate of the canal has occurred. Damage to the canal liner has occurred, causing increased friction between water and the liner, lowering hydraulic capacity. The liner damage may be due to PGDs in the form of settlements or lateral spreads due to liquefaction, movement due to landslide, offset Moderate damage is expected if movement due to fault offset, or excessive lateral or vertical movements of the ground shaking. Landslide debris may have embankments due to liquefaction or entered into the canal causes higher landslide are in the range of 1 to 5 sediment transport, which could cause Moderate inches. Moderate damage occurs due scour of the liner or earthen embankments. to fault offset across the canal of 1 to Overall, the canal can be operated in the 5 inches. Moderate damage is short term at up to 50% to 90% of capacity; expected if small debris flows into the however, a shutdown of the canal soon canal from adjacent landslides after the earthquake will be required to make repairs. Damage to canal overcrossings may have occurred, and temporary shutdown of the canal is needed to make repairs. Damage to bridge abutments could cause constriction of the canal’s cross-section to such an extent that it causes a significant flow restriction.

D3.5 -Fragility functions for water and waste-water system elements

Damage Description Damage Rate state The canal is damaged to such an extent that immediate shutdown is required. The damage may be due to PGDs in the form of settlements Major damage is expected if PGDs of or lateral spreads due to liquefaction, movement the embankments are predicted to be due to landslide, offset movement due to fault six inches or greater. Major damage offset, or excessive ground shaking. Landslide occurs due to fault offset across the debris may have entered the canal and caused canal of six inches or more. Major excessive sediment transport, or may block the damage is expected if a significant canal’s cross-section to such a degree that the amount of debris is predicted to flow Major flow of water is disrupted, overflowing over the into the canal from adjacent damage canal’s banks and causing subsequent flooding. landslides. The differentiation of Damage to overcrossings may have occurred, moderate or major damage states for requiring immediate shutdown of the canal. debris flows into the canal should Overcrossing damage could include the collapse factor in hydraulic constraints caused of highway bridges and leakage of non-potable by the size of the debris flow, the material pipelines such as oil, gas, etc. Damage potential for scour due to the type of to bridge abutments could cause constriction of debris and water quality the canal's cross-section to such an extent that requirement a significant flow restriction which warrants immediate shutdown and repair.

Table 4-23 Vulnerability of canals (wave propagation, ALA, 2001a, b)

Typology PGV 0.5 m/s PGV>0.5 m/s (R.R=0.1 repair/km) Unreinforced liners or unlined No Minor Reinforced liners No No

Table 4-24 Vulnerability of canals (permanent deformations, ALA, 2001a, b)

Typology PGD 0.025 m PGD 0.025 m PGD 0.15 m Unreinforced liners or unlined No/minor Moderate Major damages Reinforced liners

4.4.1.6 Pipes

The proposed vulnerability curves for pipes, based on the validation provided before (§4.3) are the empirical fragility curves of O’Rourke and Ayala (1993) for the case of wave propagation and Honneger and Eguchi (1992) for the case of permanent ground deformation.

4.4.1.7 Tunnels

As proposed in D3.7 “Fragility functions for roadway system elements”

D3.5 -Fragility functions for water and waste-water system elements

4.4.2 WASTE-WATER SYSTEM ELEMENTS

4.4.2.1 Waste-Water Treatment Plant

Waste-Water Treatment Plants are complex components that include several subcomponents. HAZUS (NIBS, 2004) gives fragility curves for anchored and for unanchored subcomponents for different size of Waste-Water Treatment Plants. There are no specific guidelines referring to the anchorage of the subcomponents in Europe for Waste- Water Treatment Plants. In order to account for the uncertainty in their final response as a result of the different European practices used for Waste-Water Treatment Plants of different sizes and the semi- anchorage of subcomponents, only one fragility curve for Waste-Water Treatment Plant is proposed independently of the size. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario). The description of damage states for Waste-Water Treatment Plant is provided in Table 4-25 while the corresponding vulnerability curves are given in Table 4-26.

Table 4-25 Description of damage states for Waste-Water Treatment Plant subjected to ground shaking

Damage Restoration Description Serviceability state cost (%) Malfunction of plant for a short time (< 3 days) due to loss of electric power, considerable Operational damage to various equipment, Normal flow Minor 10-30 after limited light damage to sedimentation and pressure repairs basins, light damage to chlorination tanks, or light damage to chemical tanks. Malfunction of plant for about a week due to loss of electric power, extensive damage to various equipment, considerable damage to Operational Moderate 30-50 sedimentation basins, after repairs considerable damage to chlorination tanks with no loss Reduce flow of contents, or considerable and pressure damage to chemical tanks. Partially The pipes connecting the operational different basins and chemical Extensive 50-75 after units being extensively extensive damaged. repairs The complete failure of all pipings or extensive damages Not Complete 75-100 No available of the buildings that with repairable various equipment.

D3.5 -Fragility functions for water and waste-water system elements

Table 4-26 Parameters of fragility curves for Water Treatment Plant

Peak Ground Acceleration (PGA) Description Damage state く Median (g) (log-standard deviation) Minor 0.15 0.35 Waste-Water Treatment Plants with anchored components (low-rise R/C Moderate 0.30 0.20 building with low seismic code Extensive 0.45 0.50 design) Complete 0.50 0.50 Minor 0.15 0.35 Waste-Water Treatment Plants with anchored components (low-rise R/C Moderate 0.30 0.20 building with advanced seismic code Extensive 0.45 0.50 design) Complete 1.00 0.50

Waste- Water Treatment Plants with anchored components (low-rise R/C building with low seismic code design, without back-up power) 1,00

0,80

0,60

0,40

0,20 [Probability Ds> dsPGA] /

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g)

Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-22 Fragility curves for Waste- Water Treatment Plant (Anchored components) subjected to ground shaking (low-rise R/C building with low seismic code design)

D3.5 -Fragility functions for water and waste-water system elements

Waste- Water Treatment Plants with anchored components (low-rise R/C building with advance seismic code design, without back-up power) 1,00

0,80

0,60

0,40

0,20 [Probability Ds> ds / PGA]

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00

PGA (g)

Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-23 Fragility curves for Waste- Water Treatment Plant (Anchored components) subjected to ground shaking (low-rise R/C building with advanced seismic code design)

Table 4-27 Subcomponent Damage Algorithms for Waste- Water Treatment Plants with Anchored Components

Peak Ground Acceleration Subcomponents Damage State Median (g) Loss of commercial Power minor 0.15 0.40 moderate 0.30 0.40 Chlorination minor 0.65 0.60 Equipment moderate 1.00 0.70 minor 0.36 0.50 Sediment Flocculation moderate 0.60 0.50 extensive 1.20 0.60 Chemical minor 0.40 0.70 Tanks moderate 0.65 0.70 Electrical/ Mechanical Equipment moderate 1.00 0.60 extensive 0.53 0.60 Elevated Pipe complete 1.00 0.60 Building (low-rise R/C building with complete 2.17 0.73 low seismic code design) Building (low height R/C building with complete 0.41 0.73 advanced seismic code design)

D3.5 -Fragility functions for water and waste-water system elements

4.4.2.2 Lift station

Lift Stations are complex components that include several subcomponents. HAZUS (NIBS, 2004) gives fragility curves for anchored and for unanchored subcomponents for different sizes of lift stations. There are no specific guidelines referring the anchorage of the subcomponents in Europe for lift station. In order to account for the uncertainty in their final response as a result of the different European practices used for lift stations of different sizes and the semi- anchorage of subcomponents, only one fragility curve for Pumping Station is proposed independently of the size for different building types. It is also assumed that there is no back-up power in case of loss of electric power (worst case scenario). The description of damage states for lift station is provided in Table 4-28 while the corresponding vulnerability curves are given in Table 4 29.

D3.5 -Fragility functions for water and waste-water system elements

Table 4-28 Description of damage states for Lift Station subjected to ground shaking

Damage Restoration cost Description Serviceability state (%) Malfunction of lift station for Operational a short time (< 3 days) due Minor 10-30 Normal flow after limited to loss of electric power or repairs slight damage to buildings The loss of electric power for about a week, considerable damage to Operational Moderate 30-50 mechanical and electrical after repairs equipment, or moderate damage to buildings. Reduce flow The building being Partially extensively damaged, or the operational Extensive 50-75 pumps being badly after extensive damaged beyond repair repairs Complete The building collapsing. 75-100 Not water Not repairable

D3.5 -Fragility functions for water and waste-water system elements

Table 4-29 Parameters of fragility curves for lift station

Lift station (anchored components, low-rise building with low seismic code design, without back-up power)

1,00

0,75

0,50

0,25 [Probability Ds> ds / PGA]

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-24 Fragility curves for lift station (Anchored components, low-rise R/C building with low seismic code design) subjected to ground shaking

D3.5 -Fragility functions for water and waste-water system elements

Lift station (anchored components, low-rise building with advance seismic code design, without back-up power) 1,00

0,75

0,50

0,25 [Probability Ds> ds / PGA] ds / Ds> [Probability

0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g)

Minor damages Moderate damages Extensive damages Complete damages

Fig. 4-25 Fragility curves for lift station (Anchored components, low-rise R/C building with advanced seismic code design) subjected to ground shaking

Table 4-30 Subcomponent Damage Algorithms for Lift Station with Anchored Components

Peak Ground Acceleration Subcomponents Damage State Median (g) Loss of commercial minor 0.15 0.40 Power moderate 0.30 0.40 Electric and 0.80 0.60 Mechanical moderate Equipment Vertical/ Horizontal extensive 1.25/1.60 0.60 Pump* minor 0.18 0.73 Building (low-rise R/C moderate 0.23 0.73 building with low seismic code design) extensive 0.30 0.73 complete 0.41 0.73 minor 0.28 0.73 Building (low-rise R/C moderate 0.72 0.73 building with advance seismic code design) extensive 1.66 0.73 complete 2.17 0.73

D3.5 -Fragility functions for water and waste-water system elements

Comment: For the buildings sub-components, the typology and fragility curves proposed in SRM-LIFE (2003-2007) were used. The upgrade of fragility curves will be made after the finalization of D3.1 “Fragility functions for common RC building types in Europe” and the proposal of buildings’ typologies and fragility functions for SYNER-G.

4.4.2.3 Conduits

For tunnels as proposed in D3.7 “Fragility functions for roadway system elements” For pipes as proposed for potable water system: O’Rourke and Ayala (1993) for the case of wave propagation and Honneger and Eguchi (1992) for the case of permanent ground deformation.

D3.5 -Fragility functions for water and waste-water system elements

5 Coding and digital description of fragility functions

System Water System Element at risk Well Code PWSW Reference NIBS, 2004 Method Empirical Function Lognormal Typology Component anchorage, according to building typology Damage states None Minor Moderate Extensive Complete Malfunction of Malfunction of well The building Building well pump and pump and motor for being collapsing. - motor for a about a week due to extensively short time (less loss of electric damaged or than three power and backup the well days) due to power if any, pump and loss of electric considerable vertical power and damage to shaft being backup power mechanical and badly if any, or light electrical equipment, distorted damage to or moderate damage and non- buildings to buildings functional Functionality Usable Operational Operational after Partially Not states after limited repairs operational repairable repairs after extensive repairs Seismic intensity Peak Ground Acceleration PGA (g) parameter

Figures Wells (anchored components) Wells (anchored components) Low-rise building with low seismic code design Low-rise building with advange seismic code design 1,00 1,00 0,80 0,80

0,60 0,60

0,40 0,40

0,20 0,20 [Probability Ds> ds / PGA] [Probability / Ds> ds PGA] 0,00 0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) PGA (g) Minor damages Moderate damages Extensive damages Complete damages Minor damages Moderate damages Extensive damages Complete damages Parameters (median values, く values)

Comments Distinction according to building typology.

D3.5 -Fragility functions for water and waste-water system elements

System Water System Element at risk Tunnels Code Comments See D3.7 “Fragility functions for roadway system elements”

System Water System Element at risk Pipes Code PWSPIPES Reference NIBS, 2004 Method Empirical Function O’Rourke and Ayala (1993) – wave propagation Honneger and Eguchi (1992) - permanent ground deformation. Typology Pipe material (flexible, rigid) Damage states No damage Leak Break Functionality - Reduced supply No water supply is states and pressure available Seismic intensity Peak Ground Acceleration PGA (g) – wave propagation parameter Permanent Ground Deformation PGD (m) Parameters RR/km= K*(0.0001*PGV2.25) å Wave Propagation RR/km =【*(7.821*PGD0.56) å Permanent Ground Deformation Comments -

100

D3.5 -Fragility functions for water and waste-water system elements

System Water System Element at risk Water Treatment Plant Code PWSWTP Reference SRM-LIFE, 2003-2007 Method Empirical Function Lognormal Typology Independently of the size (anchored components, no back-up power) Damage states None Minor Moderate Extensive Complete Malfunction of Malfunction of plant The pipes The plant for a short for about a week due connecting complete - time (<3 days) to loss of electric the different failure of due to loss of power and backup basins and all pipings electric power, power if any, chemical or considerable extensive damage to units being extensive damage to various equipments, extensively damage various considerable damaged. to the equipment, damage to This type of filter light damage to sedimentation damage will gallery sedimentation basins, considerable likely result in basins, light damage to the shutdown damage to chlorination tanks of the plant. chlorination with no loss of tanks, or light contents, or damage to considerable chemical tanks. damage to chemical Loss of water tanks. Loss of water quality may quality is imminent occur. Functionality Usable Operational Operational after Partially Not states after limited repairs operational repairable repairs after extensive repairs Seismic intensity Peak Ground Acceleration PGA (g) parameter Figures Water Treatment Plant (anchored components, without back-up power) 1,000

0,800

0,600

0,400

[Probability Ds> ds [Probability ds / PGA] Ds> 0,200

0,000 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) Minor damages Moderate damages Extensive damages Complete damages Parameters (median values, く values)

Comments -

101

D3.5 -Fragility functions for water and waste-water system elements

System Water System Element at risk Pumping Station Code PWSP Reference SRM-LIFE, 2003-2007 Method Empirical Function Lognormal Typology Independently of the size (anchored components, no back-up power) according to building typology Damage states None Minor Moderate Extensive Complete Malfunction of The loss of electric The building The building plant for a short power for about a being collapsing time (< 3 days) week, considerable - extensively due to loss of damage to electric power mechanical and damaged or or slight electrical equipment the pumps damage to or moderate being badly buildings damage to damaged buildings. beyond repair

Functionality Usable Operational Operational after Partially Not states after limited repairs operational repairable repairs after extensive repairs Seismic intensity Peak Ground Acceleration PGA (g) parameter

Pumping station Pumping station Figures (anchored components, low-rise building with low seismic code (anchored components, low-rise building with advance seismic design, without back-up power) code design, without back-up power) 1,00 1,00

0,75 0,75

0,50 0,50

0,25 0,25 [Probability Ds> ds PGA] / [Probability Ds> ds / PGA] ds / Ds> [Probability

0,00 0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) PGA (g) Minor damages Moderate damages Extensive damages Complete damages Minor damages Moderate damages Extensive damages Complete damages Parameters (median values, く values)

Comments Distinction according to building typology.

102

D3.5 -Fragility functions for water and waste-water system elements

System Water System Element at risk Canals Code PWSC Reference ALA, 2001 Method Empirical Function - Typology - Damage states None Minor Moderate Major damage Functionality states The canal has Some increase in Some increase in the leak The canal is damaged to the same the leak rate of the rate of the canal has such an extent that hydraulic canal has occurred. occurred. Damage to the immediate shutdown is performance Damage to the canal liner has occurred, required. The damage after the canal liner may causing increased friction may be due to PGDs in earthquake occur, causing between water and the the form of settlements increased friction liner, lowering hydraulic or lateral spreads due to between the water capacity. The liner damage liquefaction, movement and the liner and may be due to PGDs in the due to landslide, offset lowering hydraulic form of settlements or movement due to fault capacity. The liner lateral spreads due to offset, or excessive damage may be liquefaction, movement ground shaking. due to PGDs in the due to landslide, offset Landslide debris may form of settlements movement due to fault have entered the canal or lateral spreads offset, or excessive ground and caused excessive due to liquefaction, shaking. Landslide debris sediment transport, or movement due to may have entered into the may block the canal’s landslide, offset canal causes higher cross-section to such a movement due to sediment transport, which degree that the flow of fault offset, or could cause scour of the water is disrupted, excessive ground liner or earthen overflowing over the shaking. Landslide embankments. Overall, the canal’s banks and debris may have canal can be operated in causing subsequent entered into the the short term at up to 50% flooding. Damage to canal causing to 90% of capacity; overcrossings may have higher sediment however, a shutdown of occurred, requiring transport, which the canal soon after the immediate shutdown of could cause scour earthquake will be required the canal. Overcrossing of the liner or to make repairs. Damage damage could include earthen to canal overcrossings may the collapse of highway embankments. have occurred, and bridges and leakage of Overall, the canal temporary shutdown of the non-potable material can be operated at canal is needed to make pipelines such as oil, up to 90% of repairs. Damage to bridge gas, etc. Damage to capacity without abutments could cause bridge abutments could having to be shut constriction of the canal’s cause constriction of the down for make cross-section to such an canal's cross-section to repairs. extent that it causes a such an extent that a significant flow restriction. significant flow restriction which warrants immediate shutdown and repair Seismic intensity Peak Ground Velocity PGV (g) – wave propagation parameter Permanent Ground Deformation PGD (m) Parameters

Comments -

103

D3.5 -Fragility functions for water and waste-water system elements

System Water System Element at risk Storage Tank Code PWSST Reference ALA (2001a,b) Method Empirical Function Lognormal Typology According to material, anchorage, Damage states According to different material and type the damage states alters - Uplift of wall– Crush concrete, Cracking or shearing of tank wall, Sliding, Hoop overstress, Roof damage - Minor, moderate, extensive, complete Functionality - No loss of context, Small leak, Loss of context states - No operational, Operational Seismic intensity Peak Ground Acceleration PGA (g) parameter Permanent Ground Deformation PGD (m)

Figures Above ground R/C tanks (wave propagation) Above ground R/C tanks (permanent deformation) 1,00 1,00

0,80 0,80

0,60 0,60

0,40 0,40 [Probability Ds> ds / PGA] [Probability Ds> ds / PGA] / ds Ds> [Probability 0,20 0,20

0,00 0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 PGA (g) PGD (m) Minor damages Moderate damages Extensive damages Complete damages Minor damages Moderate damages Extensive damages ALA_Operative ALA_No- operative Complete damages ALA_PGD_no_operate Parameters Anchorage R/C at grade Unanchorage R/C at (median values, tanks (wave grade tanks (wave Open reservoirs with or く values) propagation) propagation) without seismic design code (wave propagation)

Unanchorage R/C at grade tanks (permanent deformations)

Comments -

104

D3.5 -Fragility functions for water and waste-water system elements

System Waste-Water System Element at risk Waste-Water Treatment Plant Code WWSWWTP Reference SRM-LIFE, 2003-2007 Method Empirical Function Lognormal Typology Independently of the size (anchored components, no back-up power) based on building typology Damage states None Minor Moderate Extensive Complete Malfunction of Malfunction of plant The pipes The plant for a short for about a week due connecting complete time (< 3 days) - to loss of electric the failure of all due to loss of electric power, power, extensive different pipings or considerable damage to various basins and extensive damage to equipment, chemical damages of various considerable damage units being the equipment, to sedimentation extensively buildings light damage to basins, considerable damaged. that with sedimentation damage to various basins, light chlorination tanks equipment. damage to with no loss of chlorination contents, or tanks, or light damage to considerable damage chemical tanks. to chemical tanks Functionality - Operational Operational after Partially Not states after limited repairs operational repairable repairs after extensive repairs Seismic intensity Peak Ground Acceleration PGA (g) parameter

Waste- Water Treatment Plants with anchored components (low-rise R/C Figures Waste- Water Treatment Plants with anchored components (low-rise building with advance seismic code design, without back-up power) R/C building with low seismic code design, without back-up power) 1,00 1,00

0,80 0,80

0,60 0,60

0,40 0,40

0,20 0,20 [Probability Ds> ds PGA] / [Probability Ds> dsPGA] /

0,00 0,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00

PGA (g) PGA (g) Minor damages Moderate damages Extensive damages Complete damages Minor damages Moderate damages Extensive damages Complete damages Parameters (median values, く values)

Comments -

105

D3.5 -Fragility functions for water and waste-water system elements

System Waste-Water System Element at risk Pipes Code WWSPIPES Reference NIBS, 2004 Method Empirical Function O’Rourke and Ayala (1993) – wave propagation Honneger and Eguchi (1992) - permanent ground deformation. Typology Pipe material (flexible, rigid) Damage states No damage Leak Break Functionality - Reduced supply No water supply is states and pressure available Seismic intensity Peak Ground Acceleration PGA (g) – wave propagation parameter Permanent Ground Deformation PGD (m) Parameters RR/km= K*(0.0001*PGV2.25) å Wave Propagation RR/km =【*(7.821*PGD0.56) å Permanent Ground Deformation Comments The same vulnerability functions as in potable water system

System Waste-Water System Element at risk Tunnels Code Comments See D3.7 “Fragility functions for roadway system elements”

106

D3.5 -Fragility functions for water and waste-water system elements

System Waste-Water System Element at risk Lift Station Code WWSLS Reference SRM-LIFE, 2003-2007 Method Empirical Function Lognormal Typology Independently of the size (anchored components, no back-up power) according to building typology Damage states None Minor Moderate Extensive Complete Malfunction of The loss of electric The building The building lift station for a power for about a being collapsing short time week, considerable - extensively (< 3 days) due damage to to loss of mechanical and damaged, or electric power electrical the pumps or slight equipment, or being badly damage to moderate damage damaged buildings to buildings. beyond repair

Lift station (anchored components, low-rise building with low Lift station (anchored components, low-rise building with Figures seismic code design, without back-up power) advance seismic code design, without back-up power) 1,00 1,00

0,75 0,75

0,50 0,50

0,25 0,25 [Probability Ds> ds / PGA] ds / Ds> [Probability [Probability Ds> ds / PGA]

Comments Distinction according to building typology.

107

D3.5 -Fragility functions for water and waste-water system elements

References

Adachi, T. and B.R. Ellingwood (2008). “Serviceability of earthquake-damaged water systems: Effects of electrical power availability and power backup systems on system vulnerability”. Reliability Engineering and System Safety, 93:78-88. Alexoudi M, (2005) “Contribution to seismic assessment of lifelines in urban areas. Development of holistic methodology for seismic risk” PhD Thesis, Civil Engineer Department, Aristotle University of Thessaloniki, Greece (in Greek) Alexoudi M, Manou D, Pitilakis K. (2008) “Seismic vulnerability analysis of waste-water system. Methodology and application for Düzce and Kocaeli earthquakes” Proceedings of the 14th WCEE, Beijing, China, ｠て.06-0119 Alexoudi M.N, Manou D.K, Hatzigogos Th.N, (2007) “The influence of probabilistic loading and site-effects in the earthquake risk assessment of water system. The case of Düzce” Proceeding of 10th International Conference on Applications of Statistics and Probability in Civil Engineering, Tokyo, Japan, July 31- 3 August, Paper Number: 176 Alexoudi M, Manou D, Pitilakis K(2010) “Influence Of Local Site Conditions In The Seismic Risk Assessment Of The Water System In Düzce (Turkey)” 6th National Conference of Geotechnical and Environmental Engineering, ｠て. 230200, Volos, Greece (in Greek) American Lifelines Alliance (2001a). “Seismic Fragility Formulations for Water Systems. Part 1 – Guideline”, ASCE-FEMA, 104 p. American Lifelines Alliance (2001b). “Seismic Fragility Formulations for Water Systems. Part 2 – Appendices”, ASCE-FEMA, 239 p. American Lifelines Alliance (2002) “Development of Guidelines to Define Natural Hazards Performance Objectives for Water Systems, Volume I” prepared by ASCE, FEMA and NIBS American Lifelines Alliance (2004), “Wastewater System Performance Assessment Guideline” Part 1 Guideline, prepared by ASCE, FEMA and NIBS American Lifelines Alliance (2005) “Seismic Guidelines for Water Pipelines” prepared by ASCE, FEMA and NIBS Aldea A., Arion C., Petrescu F. (2002). “March 4, 1977 - Damage to city supply facilities” UTCB internal report in RISK-UE (2001- 2004) research project. Applied Technology Council (ATC) (1985). “ATC-13-Earthquake Damage Evaluation Data for California”, Redwood City, California. ASCE (1987) “Planning, Operation, Rehabilitation and Automation of Irrigation Water Delivery Systems” Proceedings of a Symposium sponsored by the Irrigation and Drainage Division, July 28-30, Editor Darell D. Zimbelman, 391pp ASCE (1997) “Guidelines for Seismic Evaluation and Design of Petrochemical Facilities” prepared by Task Committee on seismic evaluation and design of petrochemical facilities of the petrochemical committee of the energy division of the American Society of Civil Engineers, New York, 375pp. Ballantyne D, Heubach W (1996) “Use of GIS to evaluate earthquake hazard effects and mitigation on pipeline systems- Case studies” Proceedings from the Sixth Japan- U.S

109

D3.5 -Fragility functions for water and waste-water system elements

Workshop on Earthquake Resistant Design of Lifeline Facilities and Countermeasures Against Soil Liquefaction, Technical Report NCEER -96-0012. Ballantyne D., Perimon T., Martins P. (2009) “Water Treatment Plant Seismic Risk Assessment for the Joint Water Commission, Hillsboro, Oregon” Proceedings of TCLEE 2009, Lifeline Earthquake Engineering in a Multihazard Environment, ASCE, June 28- July 1, Oakland, California Dueñas-Osorio, L., J.I. Craig and B.J. Goodno (2007a). ”Seismic response of critical interdependent networks” Earthquake Engineering and Structural Dynamics, 36:285-306. Dueñas-Osorio, L., J.I. Craig, B.J. Goodno and A. Bostrom (2007b) “Interdependent response of networked systems” Journal of Infrastructure Systems, 13:185-194. Dueñas-Osorio, L., and S. M. Venuru (2009) “Cascading failures in complex infrastructure Systems” Structural Safety, 31:157-167. EERI (1990) “Loma Prieta earthquake reconnaissance report” Earthquake Spectra, Supplement to Vol.6, Earthquake Engineering Research Institute, Oakland, CA EERI (1995) “Northridge earthquake of January 17, 1994 reconnaissance report” Earthquake Spectra, Supplement C to Vol.11, Earthquake Engineering Research Institute, Oakland, CA. Eguchi R.T (1983) “Seismic Component Vulnerability Models for Lifeline risk Analysis” J. Wiggins Co, Technical Report No. 82-1396-22 Eidinger J.M, Maison B, Lee D, Lau B (1995) “East Bay Municipality Utility District Water Distribution Damage in 1989 Loma Prieta Earthquake” Proceedings of the Fourth U.S Conference on Lifeline Earthquake Engineering, ASCE, TCLEE, Monograph 6, ASCE, pp. 240-247. Eidinger J (1998) “Lifelines, Water Distribution System” in the Loma Prieta, California, Earthquake of October 17, 1989, Performance of the Build Environment –Lifelines, US Geological Survey Professional Paper 1552-A, pp. A63-A80, A. Schiff Ed. Eidinger J, Avila E (1999) “Guidelines for the seismic upgrade of Water Transmission Facilities” ASCE, TCLEE, Monograph No. 15. Heubach W (1995) “Seismic Damage Estimation for Buried Pipeline Systems” Proceedings of the Fourth US Conference on Lifeline Earthquake Engineering, TCLEE, Monograph No.6, ASCE, pp312-319. Honegger, D. G., R.T. Eguchi (1992). “Determination of the Relative Vulnerabilities to Seismic Damage for Dan Diego Country Water Authority (SDCWA) Water Transmission Pipelines”. Hung J-H (2001) “The analysis of water pipeline Damages of Wufeng Shiang in the 921 Ji-Ji Earthquake, Master’s Thesis, National Taipei University of Technology, Taipei, Taiwan (in Chinese) Isoyama, R., E. Ishida, K. Yune, T. Shirozu (1998). “Seismic Damage Estimation Procedure for Water Supply Pipelines”. Proceedings, Water & Earthquake ’98 Tokyo, IWSA International Workshop, Anti-Seismic Measures on Water Supply, International Water Services Association and Japan Water Works Association, Tokyo Japan.

110

D3.5 -Fragility functions for water and waste-water system elements

Isoyama, R., Ishida, E., Yune, K. & Shirozu, T. (2000). “Seismic damage estimation procedure for water supply pipelines,” Proceedings of the Twelfth World Conference on Earthquake Engineering, CD-ROM Paper No. 1762, 8pp. Isoyama & Katayama, (1982) “Reliability Evaluation of Water Supply Systems during Earthquakes” Report of the Institute of Industrial Science, University at Tokyo, February, vol.30, no.1 Imai T., Koike T. (2010) “Seismic risk management for an existing lifeline system” Safety, Reliability and Risk of Structures, Infrastructures and Engineering Systems – Furuta, Frangopol & Shinozuka (eds), 2010 Taylor & Francis Group, London, ISBN 978-0-415- 47557-0 JWWA (1997) “Seismic Design and Construction Guidelines for Water Supply Facilities”, Japan Water Works Association. Katayama T, Kubo K, Sato N (1975) “Earthquake Damage to Water and Gas Distribution Systems” Proceedings of the U.S National Conference on Earthquake Engineering, Oakland, California, EERI, pp. 396-405. Kappos, A., G. Panagopoulos, Ch. Panagiotopoulos and Gr. Penelis (2006). A hybrid method for the vulnerability assessment of RC and URM buildings. Bulletin of Earthquake Engineering 4(4): 391-413. Kawakami H (1990) “Earthquake physical damage and functional serviceability of lifeline network models” Earthquake Engineering and Structural Dynamics, Vol. 19, 1153-1165 Kudo K., Kanno T., Okada H., Ozel O., Erdik M., Takahashi M., Sasatani T., Higashi S., Yoshida K, 2000) “Site specific issues on strong ground motion during the Kocaeli, Turkey Earthquake of August 17,1999” as inferred from array observations of microtremors and aftershocks. BSSA Special Issue. LESSLOSS, (2004-2007). Risk Mitigation for Earthquakes and Landslides. Research Project, European Commission, Sixth Framework Programme, Priority 1.1.6.3, Global Change and Ecosystems, Contract Number: GOCE-CT-2003-505448. Mataki Sh., Deguchi T., Honda K., Suzuki T. (1996)”Investigation of aseismic design methods for highly- reliable telecommunication conduit facilities in liquefied ground” Proceedings from the Sixth Japan- U.S Workshop on Earthquake Resistant Design of Lifeline Facilities and Countermeasures Against Soil Liquefaction, Technical Report NCEER -96-0012. Matasovic N, Kavazanjian Ed, Anderson R. (1998) “Performance of Solid Waste Landfills in Earthquakes” Earthquake Spectra, Vol. 14, (2), pp.319-334 Moghtaderi-Zadeh, M., K. Wood, A. Der Kiureghian and R.E. Barlow (1982) “Seismic reliability of lifeline networks” Journal of Technical Councils, ASCE, 108:60-78. National Institute of Building Sciences (NIBS) (2004). “Earthquake loss estimation methodology. HAZUS ’04”, Technical manual, Vol.1. Federal Emergency Management Agency (FEMA), Washington, D.C. NCEER (1994) “The Northridge, California Earthquake of January 17, 1994: Performance of Gas Transmission Pipelines” T.D O’Rourke, M.C Palmer, TR NCEER-94-0011, National Center for Earthquake Engineering Research, Buffalo, NY

111

D3.5 -Fragility functions for water and waste-water system elements

NCEER (1995) “The Hanshin- Awaji Earthquake of January 17, 1995 Performance of Lifelines” M. Shinozuka, TR NCEER-95-0015, National Center for Earthquake Engineering Research, Buffalo, NY. NIST (1994) “The January 17, 1994, Northridge Earthquake (California)”, H.S Lew, J. Cooper, S. Hacopian, W.Hays, M. Mahokey, Special Publication 871, Editor Noel, J. Ranfaste, Gaithersburg, Maryland NIST (1996) “The January 17, 1995 Hyogoken-Nanbu (Kobe) Earthquake, Performance of Structures, Lifelines and Fire Protection System”, Special Publication 902 (ICSSC TR18), Gaithersburg, Maryland. NISTIR (1994) “Performance of Structures, Lifelines and Fire Protection Systems” D. Todd, N. Carino, R. Chung, H.S Lew, A.W Taylor, W.D Walton, Publication 5396. NRC (1994) “Practical lessons learned from Loma Prieta earthquake” National Research Council, National Academy Press, Washington DC. O’ Rourke & Ayala (1993). "Pipeline damage due to wave propagation." Journal of Geotechnical Engineering. ASCE. 119 (9), 1490-1498. O' Rourke, T.D., Toprak, S, & Sano, Y. (1998). "Factors affecting water supply damage caused by the Northridge earthquake." Proceedings of the Sixth US National Conference on Earthquake Engineering. [CD-ROM]. O’Rourke, T.D. & Jeon, S.-S. (1999). “Factors affecting the earthquake damage of water distribution systems,” Proceedings of the Fifth US Conference on Lifeline Earthquake Engineering, Seattle, WA, ASCE, Reston, VA, 379-388. O’Rourke M.J., Liu X. (1999). “Response of buried pipelines subject to earthquake effects” MCEER. Monograph series n°3. 249 p. O’Rourke T.D and Palmer M.C (1996) “Earthquake performance of gas transmission pipelines” Earthquake Spectra, 12 (3), pp. 493- 527 O’Rourke T.D, Toprak S, Sano Y (1998) “Factors affecting water supply damage caused by the Northridge Earthquake”, Proceedings of the 6th US National Conference on Earthquake Engineering, Seattle, WA, pp. 1-12. O’Rourke, M. J., and P. So. 2000. Seismic fragility curves for on-grade steel tanks. Earthquake Spectra 16(4). ぁ’Rourke & Deyoe E (2004) “Seismic Damage to Segmented Buried Pipe” Earthquake Spectra, Vol.20, No.4, pp.1167-1183. Pitilakis K, Alexoudi M, Kakderi K, D. Manou, E. Batum, D. Raptakis, (2005) “Vulnerability analysis of water systems in strong earthquakes. The case of Lefkas (Greece) and Düzce (Turkey)” in International Symposium on the Geodynamics of Eastern Mediterranean: Active Tectonics of the Aegean, June 15-18, Istanbul, Turkey Rosenblad, B.L., Rathje, E.M., and Stokoe, K.H. (2001). “Shear Wave Velocity Profiling by SASW Method at Selected Strong-Motion Stations from the 1999 Turkey Earthquakes,” Final Report to Pacific Earthquake Engineering Research Center, September. Scawthorn Ch (1996) “Reliability – based Design of Water Supply Systems” Proceedings from the Sixth Japan- U.S Workshop on Earthquake Resistant Design of Lifeline Facilities and Countermeasures Against Soil Liquefaction, Technical Report NCEER -96- 0012.

112

D3.5 -Fragility functions for water and waste-water system elements

Shrestha B. K (2001) “Disaster Reduction and Response Preparedness in Japan: A Hyogo Approach” Proceedings of the 2nd Conference on Disaster Communications (CDC 2001), Tampere, Finland 28-30 May, 2001 Singhal A.C (1984) “Nonlinear behavior of ductile iron pipeline joints” Journal of Technical Topics in Civil Engineering, 110, No1, ASCE. SRMLIFE (2003-2007). “Development of a global methodology for the vulnerability assessment and risk management of lifelines, infrastructures and critical facilities. Application to the metropolitan area of Thessaloniki”, Research Project, General Secretariat for Research and Technology, Greece. Porter K.A, Scawthorn C, Honegger D.G, O’rourke T.D, Blackburn F (1991) “Performance of Water Supply Pipelines in Liquefied Soil” Proceedings 4th US – Japan Workshop on Earthquake Disaster Prevention for Lifeline Systems, NIST Special Publication 840. RISK-UE, (2001-2004). An Advanced Approach to Earthquake Risk Scenarios with Applications to Different European Towns. Research Project, European Commission, DG ゅ』I2001-2004, CEC Contract Number: EVK4-CT-2000-00014. SRM-DGC (2006-2008). Development and proposition for implementation of an efficient methodology and appropriate local instruments for the management, prevention and reduction of seismic risk in Düzce -Turkey, Grevena - Greece and Catania – Italy. Research Project, Interreg III B Archimed, Code Number. A.1.010. Tadday, K. & Sahin, M, (2001). “Situation analysis and recommendations related to improvement measures for drinking water supply system in Düzce,” Proceedings of the Symposium on Water Supply Situation in Earthquake Affected Region, Düzce Municipality, Koru Hotel, Düzce, Turkey, 4 – 5 April, 1-54. TCLEE (1995) “Northridge earthquake lifeline performance and post-earthquake response” Monograph No.8, Edited by Schiff, published by ASCE, New York, 339pp. Terzi V, Alexoudi M, Hatzigogos Th, (2007) Numerical Assessment of damage state of Segmented Pipelines due to Permanent Ground Deformation” Proceeding of 10th International Conference on Applications of Statistics and Probability in Civil Engineering, Tokyo, Japan, July 31- 3 August, Paper Number: 202 Tromans J, (2004) “Behaviour of buried water supply pipelines in earthquake zones” PhD thesis, Department of Civil and Environmental Engineering, Imperial College of Science, Technology and Medicine, London. Toprak S (1998) “Earthquake Effects on Buried Lifeline Systems” PhD Dissertation, Cornell University Yeh, C.-H., Shih, B.-J., Chang, C.-H., Chen, W., Liu, G.-Y.and Hung., H.-Y. 2006. Seismic Damage Assessment of Potable Water Pipelines, Proc. 4th Int. Confer. Earthquake Eng., Paper No. 247, Oct. 12-13, Taipei, Taiwan. Wang LRL, Wang JCC, Ishibashi I (1991) “GIS applications in seismic loss estimation model for Portland, Oregon water and sewer systems” US Geological Survey, Open File Report.

113