Spatial and Ecological Assessment of The

Environmental Issues Series no. 11

Spatial and Ecological Assessment of the TEN: Demonstration of Indicators and GIS Methods Progress Report of the DGVII-DGXI-Eurostat-EEA Working Group of the SEA of the TEN April 1998 Spatial and Ecological Assessment of the TEN

NOTE

The contents of this report do not necessarily reflect the official opinion of the European Commission or other European Communities institutions. Neither the European Environ- ment Agency nor any person or company acting on the behalf of the Agency is responsible for the use that may be made of the information contained in this report.

This is a progress report of the DGVII-DGXI-Eurostat-EEA working group, prepared by the European Environment Agency. Main contributors to this report are: Chris Steenmans, EEA, Ivone Pereira Martins, EEA, Wim Devos, GIM Geographic Informa- tion Management, Ann Dom, DGXI, Johan Geert Koier, DGVII, Technical assistance mapping: Klaas Scholte, University of Utrecht

A great del of information on the European Union is available on the Internet. It can be accessed through the Europa server (http://europa.eu.int)

Cataloguing data can be found at the end of this publication

Luxembourg: Office for Official Publications of the European Communities, 1998

Cover and layout: Folkmann Design

ISBN xxx

European Environment Agency Kongens Nytorv 6 DK-1050 Copenhagen K Denmark Tel. (+45) 33 36 71 00 Fax (+45) 33 36 71 99 E-mail: [email protected] Home page: http://www.eea.eu.int Table of Contents

Table of Contents

1. Introduction ...... 7

2. Executive Summary ...... 8

3. Context...... 10 3.1. Background ...... 10 3.2. The Commission Work Programme on a SEA of the TEN ...... 10

4. Methodology ...... 12 4.1. Purpose and Scope of a SEA of the TEN ...... 12 4.2. Environmental Objectives and Targets ...... 12 4.3. Impact Indicator Selection ...... 12

5. Data and Information Sources ...... 16 5.1. Major Data Sources and Limitations ...... 16 5.1.1. Transport Infrastructure Data...... 16 5.1.2. Environmental Data ...... 17

6. TEN alternatives ...... 18 6.1. Selection of Alternatives...... 18 6.2. The TEN in Perspective ...... 21

7. Nature and Biodiversity ...... 22 7.1. Main Objectives Concerning Nature and Biodiversity .. 22 7.2. Vicinity to Sites of Nature Conservation Interest ...... 22 7.2.1. Methodological Approach ...... 23 7.2.2. The Results ...... 24 7.3. Partitioning of Land ...... 26 7.3.1. Methodological Approach ...... 27 7.3.2. The Results ...... 27 4 Spatial and Ecological Assessment of the TEN

8. Water Resources ...... 29 8.1. Main Objectives Concerning the Protection of Water Resources...... 29 8.2. Crossing of Surface Waters ...... 29 8.2.1. Methodological Approach ...... 29 8.2.2. The Results ...... 31

9. Noise ...... 32 9.1. Main Objectives Concerning Noise ...... 32 9.2. People in the Vicinity of Infrastructure ...... 32 9.2.1. Methodological Approach ...... 32 9.2.2. The Results ...... 33 9.3. Crossing of Tranquil Zones ...... 33 9.3.1. Methodological Approach...... 33 9.3.2. The Results ...... 34

10. Land Resources ...... 36 10.1. Main Objectives Concerning Land Resources ...... 36 10.2. Land Take ...... 36 10.2.1. Methodological Approach ...... 36 10.3. Sensitive Zones Mapping ...... 39 10.3.1. Methodological Approach ...... 39 10.3.2. The Results ...... 43

11. Evaluation and Conclusions ...... 46 11.1. Indicator Overview ...... 46 11.2. Conclusions...... 49 Table of Figures 5

Table of Figures

Figure 4.1: Relevant spatial and ecological issues for a strategic environmental assessment of the TEN ... 13 Figure 5.1: Available digitalised data concerning European TEN infrastructure ...... 16 Figure 5.2: Available (digitalised) data for environmental assessment on European scale ...... 17 Figure 6.1: Decision No 1692/96/EC Annex III priority projects ...... 18 Figure 6.2: The three transport modes included in this study and a multi-modal overview including the 14 specific projects...... 19 Figure 6.3: Example of the multi-modal TEN network compared to the national and regional network... 20 Figure 6.4 Estimated length of the trans-European transport network compared to the overall European infrastructure (kilometres) ...... 20 Figure 7.1: Site description according to its legal statute...... 23 Figure 7.2: Designated sites potentially affected by the development of the trans-European transport network. Status in mid 1997 (overlaps between sites not taken into account) ...... 23 Figure 7.3: Indication of potential conflicts between complete TEN and nature sites. Nature site classification according to neighbouring links ...... 25 Figure 7.4: Indication of potential conflicts between complete TEN and nature sites. TEN infrastructure classification according to ...... neighbouring nature sites...... 26 Figure 7.5: Number of land units and their cumulative surface - existing and TEN planned situation ...... 26 Figure 7.6: Partitioning of Land effects of the complete TEN .... 28 Figure 8.1: Water crossings of the existing trans-European transport network ...... 29 Figure 8.2: Comparison of the number of water crossings under different TEN variants ...... 30 6 Spatial and Ecological Assessment of the TEN

Figure 8.3: Complete Multimodal TEN classification according to potential conflicts with surface water .. 30 Figure 9.1: Population susceptible to noise exposure due to TEN infrastructure ...... 32 Figure 9.2: Population exposed to noise impacts per TEN variant...... 33 Figure 9.3: Population density map and multi-modal TEN network ...... 33 Figure 9.4: Intersection with tranquil zones per TEN variant ..... 34 Figure 9.5: Intersection with tranquil zones according to TEN variants ...... 34 Figure 9.6: Noise disturbance zones around the Complete Multimodal TEN network ...... 35 Figure 10.1: Estimated surface requirements per infrastructure type ...... 36 Figure 10.2: Land take by TEN variant ...... 37 Figure 10.3: Distribution of the major land cover types in Europe ...... 38 Figure 10.4: Land take by land cover type for TEN variants ..... 39 Figure 10.5: Sensitivity thresholds attributed to baseline data sets ...... 40 Figure 10.6: Assignment rules for determining sensitivity per group ...... 40 Figure 10.7: Sensitivity map based on Group I criteria (physical-environmental criteria) ...... 41 Figure 10.8: Sensitivity map based on Group II criteria (environmental legal status criteria) ...... 42 Figure 10.9: Sensitivity map based on Group III criteria (demographic criteria) ...... 43 Figure 10.10: Identification of high sensitivity cells crossed by infrastructure links ...... 44 Figure 10.11: Sensitivity index per impact group and TEN variant ...... 44 Figure 10.12: Sensitivity index graphs per impact group and TEN variant ...... 45 Figure 11.1: Summary of indicator characteristics...... 47 Figure 11.2: Assessment of used indicators according to OECD criteria ...... 48 Introduction 7

1. Introduction

The present report is written in view of the Commission’s aim to develop appropriate methods for the strategic evaluation of the environmental impacts of the trans-European transport network (TEN). In order to study the possibilities of evaluating a particular type of environmental impact – namely spatial and ecological impacts – the Directorate-General for Trans- port (DGVII) and the Directorate-General for Environment, Nuclear Safety and Civil Protec- tion (DGXI) in March 1997 initiated a co-operative effort with the European Environment Agency (EEA) and Eurostat.

The Commission’s request led to the creation of a working group consisting of DGVII, DGXI, Eurostat and EEA representatives, and the present study was steered and developed by this working group. The EEA European Topic Centre on Land Cover carried out most of the technical work, including the GIS development. The EEA European Topic Centre on Nature Conservation provided data on nature conservation and gave methodological feedback.

The objectives of the present methodological study were:

I. to develop and evaluate a number of indicators and GIS assessment methods and to demonstrate their feasibility - taking account of the present data available - for an assessment of the overall trans-European transport network, and consequently provide a common discussion platform between and within Commission services, with external experts, policy makers and other stakeholders on this issue; II. to identify issues for further research as a result of this broad and multi-disciplinary consultation.

In this context, the reader should bear in mind that all the results and figures that are presented in this methodological report serve only to illustrate the potential output of the various assessment methods. They are not the final results of the assessment. 8 Spatial and Ecological Assessment of the TEN

2. Executive Summary

In accordance with the provisions of the used. For the current study, it was decided to European Union (EU) guidelines for the define overall network variants that do not trans-European transport network (TEN), a represent true policy options, but that are European Commission working group has useful in raising awareness about the possi- investigated the feasibility of carrying out a ble environmental impacts of the trans- full spatial and ecological assessment of the European transport network. The variants TEN. The aim of the study was to develop are: and test a number of indicators, to consider the availability of data and to identify issues I. a “do-nothing” option: no further exten- for further research, if necessary. The study sions to the existing network; and the resulting report are intended to II. a “do-minimum” option: the existing promote further discussion between all network is extended by the execution of stakeholders on the issue of strategic envi- the fourteen specific projects agreed at ronmental assessment (SEA) in relation to the Essen European Council in 1994; TENs. III. the existing network is further developed through the carrying out of all road On the basis of EU environmental objectives, projects; mainly specified in the 5th Environmental IV. the existing network is further developed Action Plan, the working group identified a through the realisation of all rail number of indicators that could be taken projects; into account in a spatial and ecological V. the existing network is further developed assessment of the TEN. Five different envi- through the implementation of all ronmental themes were highlighted: inland waterway projects; biodiversity and nature conservation, water VI. the existing network is completed by the resources, coastal zones, noise and land development of all planned projects. resources. Two broad types of indicators were tested. The working group considered only quanti- The first type is based on identifying specific tative evaluation methods, using a Geo- sites in the vicinity of planned infrastructure graphical Information System (GIS) as an links. In principle this technique can be used analysis tool. The methodological approach to evaluate the potential impact on, for was mainly based on land-suitability or example, designated biodiversity or nature sensitivity analysis. This type of analysis sites and water catchment areas. focuses on the environmental characteristics of the land and relies on expert judgement The second category of indicators is based (based on scientific knowledge). The quanti- on the identification of different land cover tative approach involves indicator categories types according to their degree of vulnerabil- selection, impact estimation and ratings or ity to certain impacts. For example, certain priorities assignment. Map overlay tech- habitat types may be more vulnerable from a niques are the most suitable for carrying out biodiversity point of view than others. the analysis; therefore, GIS tools - given the Alternatively, one could define non-frag- scale of the analysis and the geographical mented areas that are more likely to have a differentiation of the impacts - have proven high value for nature and biodiversity than to be the most useful. With these types of zones that are already fragmented by existing methodologies, impacts are implied but not (TEN) infrastructure. This type of indicator explicitly reported. Qualitative results may be can also be used to combine different integrated subsequently but an evaluation environmental aspects into an overall vulner- without any degree of quantification was not ability/sensitivity index per zone. considered feasible, given the objectives of the study. It must be noted that the results of an assessment at EU-level must be interpreted Six different TEN network development with great care, especially when it concerns variants - considering only linear road, rail spatial and ecological impacts whose signifi- and water infrastructures - were used to cance very much depends on local circum- demonstrate how the indicators could be stances. The design and alignment of the Executive Summary 9

infrastructure, the vulnerability of the As a broad conclusion, the present methodo- species in the area concerned, the traffic logical study can be considered successful as volume and several other factors determine it realised most of its original objectives. It the extent of the ecological impacts. Moreo- demonstrated the technical feasibility, but ver, mitigation measures at project level can also the limitations, of a European-wide SEA reduce (sound barriers) or even eliminate of the TEN, and stressed the usefulness of (tunnelling) adverse impacts. Therefore, an multi-disciplinary data collection at an EU-level network assessment should prima- European scale. rily be seen as a warning system for potential risks. As a result, the present exercise helped suggest further research and methodological The feasibility study clearly underlined the improvements: practical difficulties that arise in an assessment of the overall network. The quality and 1. Refining the indicators and methods availability of data at EU-level turned out to through a wide and multi-disciplinary be rather poor. For example, digitalised consultation process; maps of the TENs are only available on a 2. Filling the major data gaps on the TEN very large scale (1:1.000.000), information and on the environment; concerning the possible alignment of new 3. Carrying out a full spatial and ecological links is often lacking, the actual boundaries assessment of the TEN, including the of designated areas are unknown or not integration into the GIS of the results of available in GIS-format, and overlaps be- the traffic-related impacts and a compre- tween different designation categories hensive multi-criteria analysis of the cannot be identified. predicted impacts; 4. Initiating longer term research on the The insufficiency of information limits the above outlined issues, in order to gradu- performance of the assessment methods. ally improve methods, data and tools for Based on the currently available data, the the ecological and spatial impact assess- measurability and analytical validity of the ment of transport plans; proposed indicators cannot be fully assured. 5. Performing “zoom analyses” or “sample Therefore, either additional data must be surveys”, enabling a testing of the inter- collected at EU-level or a different approach relation between different planning/ must be followed, namely a more detailed geographical levels and corresponding but still strategic, assessment per infrastruc- assessment techniques. ture scheme. An example of this method is the way in which strategic assessments are carried out for the German Bundesverkehrswegeplan. Furthermore, additional data collection would be particularly important for improving information on planned TEN infrastructure. Another priority action is the completion of the NATURA 2000 network, as this would provide a more consistent basis for assessment of impacts on habitats.

The models and parameters that were used in this study have been applied in a rather straightforward manner, and have been employed using expert judgement, which inevitably involves a certain degree of subjectivity. The key assumptions that have been made and their transferability to EU scale should be improved by a broad and multi- disciplinary consultation of experts. In particular the EU wide definition of sensitive zones, which inevitably relies to a certain extent on a subjective use of sensitivity values, should be further researched. 10 Spatial and Ecological Assessment of the TEN

3. Context

3.1. Background environmental assessments may become mandatory for a wide range of plans and In July 1996 the European Parliament and programmes if the Commission’s proposal the Council adopted Community guidelines for a Directive on SEAs is adopted². In April for the development of the multi-modal 1997, the Council of Ministers of the Euro- trans-European transport network (TEN)¹. pean Conference of Ministers of Transport By 2010, it will gradually be established by formally endorsed the principles of SEA at its integrating land, sea and air transport session in Berlin and called for SEA to be infrastructure networks throughout the put into practice. Community. The transport infrastructure will comprise road, rail and inland waterways networks, seaports and inland waterway 3.2. The Commission Work Programme ports, and other interconnection points. The on a SEA of the TEN guidelines provide the framework for the development of the TEN and specify objec- Following the adoption of the TEN guide- tives, principles and outline maps. One of lines, the Directorate-General for Transport the main objectives of the TEN is “...to (DGVII) and the Directorate-General for ensure the sustainable mobility of persons Environment, Nuclear Safety and Civil and goods within an area without internal Protection (DGXI) of the European Com- frontiers under the best possible social and mission started to co-operate closely . A state- safety conditions...” (article 2 of the guide- of-the-art review showed that a wide range of lines). The network should not only SEA methods and techniques already exist strengthen the economic and social cohe- and that SEA practice in the transport sector sion of the Union, but should also help to is growing. However, there is little experi- achieve the Union’s environmental objec- ence in applications on a scale comparable tives. Integration of environmental concerns to the trans-European transport network. into the design and development of the TEN Thus, a practical-oriented work programme is considered a key priority. was developed and adopted by both DGs. It included the following major tasks: The Commission has long recognised that the development of a process of strategic • carrying out a pilot SEA of the overall environmental assessment (SEA) is a prereq- trans-European transport network, in co- uisite in this regard. This view has been operation with the EEA and in consulta- 1) Decision N° 1692/96/ EC on Community confirmed in the guidelines. Article 8 §2 of tion with the Member States; Guidelines for the the TEN guidelines states that “the Commis- • supporting a number of pilot assessments development of the sion will develop appropriate methods of of TEN corridors conducted by the trans-European analysis for strategically evaluating the Member States in co-operation and with transport network, OJ L environmental impact of the whole network co-financing of the Commission; 228, Volume 39, 9 September 1996 [...] and will develop appropriate methods of • developing a methodological handbook corridor analysis covering all relevant trans- for SEA of infrastructure networks and 2) Proposal for a port modes without prejudice to the defini- corridors, featuring the findings of past Council Directive on the tion of the corridors themselves [...]”. The state-of-the-art studies, case studies and assessment of the article also stipulates that the result of the the above mentioned pilot SEAs. Practi- effects of certain plans and programmes on work will, as and when appropriate, be taken cal guidance is also given. the environment - into account by the Commission in its report COM(96)511 final on the revision of the guidelines, which is to The present work forms part of the pilot SEA be published by July 1999. of the overall trans-European network, which 3) European Commis- - for pragmatic reasons - has been split in sion - DGXI, Spatial and ecological assessment It should be noted that the concept of three main parts: of the trans-European strategic environmental assessment is not transport network: limited to transport policies or plans. In I the assessment of spatial and ecological scope, methods, data several Member States and in a number of impacts of the TEN. This part focuses on and research needs, countries outside Europe, SEAs are already the mere physical impacts of the infra- Workshop proceedings, Brussels, 24 - 25 April commonly carried out for land use and structure. The idea is to evaluate land 1997. development plans. Moreover, strategic use, disturbance and fragmentation of Context 11

nature areas, etc. by using geographical Land Cover and Nature Conservation, a information system tools; number of international NGOs, as well as II the assessment of traffic-generated consultants and scientists. impacts such as emissions of greenhouse gases, acidifying gases and pollutants, The work, steered and developed by a energy consumption, safety and - if DGVII-DGXI-EEA-Eurostat working group, possible - noise. This task, which requires consisted of the following tasks: the use of predictive traffic and environmental models, will be carried out by an • the development of an integrated data- international research team in the base and Geographical Information context of the Commission’s 4th Frame System (GIS) on the TEN, including work Programme on Transport RTD; thematic data maps on infrastructure, III the aggregation of the results of the land cover, demography, geography, previous two steps and the comparative environment and nature; evaluation of the predicted impacts in • the selection and review of indicators for the light of the Community’s environ- assessing the spatial and ecological mental objectives and targets. impacts of the TEN, taking into account the current data availability and limita- The present methodological study was tions; conducted as a preparatory phase for the • the development and testing of a first part of the pilot assessment, i.e. the number of GIS assessment techniques; spatial and ecological assessment of the TEN. • the compilation of the results in a GIS It builds on the findings of a technical demonstration package to allow an workshop on this issue organised by the interactive demonstration of indicators Commission in April 1997³. The workshop and methods; was attended by representatives of the • a broad consultation on the findings and European Commission (DGXI, DGVII, and a consensus-building process regarding Eurostat), the EEA and its Topic Centres on full assessment methodology. 12 Spatial and Ecological Assessment of the TEN

4. Methodology

4.1. Purpose and Scope of a SEA of the TEN enable more detailed evaluations and will have to take into account specific and more The SEA is a systematic process of evaluating detailed national and regional characteristics the environmental impacts of a strategic and objectives. action and its alternatives. The main aim of a SEA is to ensure that environmental considerations are addressed at the earliest appro- 4.2. Environmental Objectives and Targets priate stage of decision-making and on a level equal to that of economic and social Strategic assessment is generally considered considerations. as an objective-led process. The purpose of a SEA is to assess to what extent the strategic In developing methods for the SEA of the action and its alternatives contribute to the TEN, it is important to keep in mind the realisation of certain environmental objec- general aim of such an evaluation and the tives and targets. A framework of environ- information that it is expected to generate. mental objectives and targets is essential for The main objective of developing a SEA the identification of proper indicators and system for the TEN is to raise awareness for the evaluation of impact predictions. about the environmental consequences of developing the TEN, to improve planning The broad (environmental and other) and to assist investment decisions. The SEA objectives of the TEN are set out in Article 2 of the TEN should therefore provide the of the TEN guidelines. For a SEA at the TEN following information: level, the targets of the 5th Environmental Action Programme (5EAP)4 and its recent • an overall evaluation of the extent to review5 provide a minimum framework. The which the current TEN outline plan will White Paper on the Common Transport help to achieve the Community (and Policy (CTP) also identified a number of other) objectives and targets for environ- priority environmental objectives. As regards mental protection. This evaluation nature conservation, the designated areas should distinguish the effects per trans- covered by the Directives 79/409/EEC port mode, the potential modal shift and (Birds Directive) and 92/43/EEC (Habitats) the effect of induced traffic; and other (inter)nationally designated • a broadbrush impression of the geo- conservation areas should be incorporated. graphic distribution of positive and The present study focuses on Community- negative impacts of the TEN. In particu- level targets. At a later stage, an investigation lar, the assessment should allow the should be conducted as to whether and how identification of potential conflict areas. national or regional strategic options and The SEA will not address alignment objectives can be included within the frame- options per link or corridor, but can for work of the SEA of the TEN. 4) European example be useful in determining the Commission, DG XI size of the corridor that should subse- (1992): Towards quently be assessed in more detail by the 4.3. Impact Indicators Selection Sustainability: A Member States at a more local planning European Community Programme of Policy level (e.g. corridor level). An indicator is a variable that enables an and Action in Relation evaluation of the impact intensity for a to the Environment and The SEA should produce information that is particular environmental stake or objective. Sustainable Develop- of greatest relevance to the particular deci- Environmental indicators can be measured ment (The Fifth Action sion-making level it addresses and on a level in physical units (e.g. emissions), on an Programme). Brussels. of detail that is appropriate. Thus, the ordinal scale (e.g. landscape quality) or can 5) European Environ- assessment of the TEN will differ in scope simply be described in a qualitative manner ment Agency (1995): and detail with the assessment of the TEN (e.g. biotopes). Environment in the corridors: an assessment at TEN-level will European Union 1995. inevitably be more broad and will focus on One of the main aims of the working group Report for the Review of the Fifth Environmen- impacts of “Community-importance” (e.g. was to identify possible indicators of spatial tal Action Programme. greenhouse effect, acidification, loss of and ecological impacts of the TEN. In the Copenhagen. biodiversity). Corridor assessments will context of this study, ecological and spatial Methodology 13

impacts are broadly defined to include the • Impacts for which indicators with a following impact categories: spatial dimension can be used. An example of this is noise, which can be • Ecological impacts of infrastructure, i.e. evaluated among other ways by estimat- impacts that are directly related to ing the number of people living in the ecosystems, habitats and species and the vicinity of the infrastructure or by defin- degree of biodiversity. Examples are ing noise sensitive zones (“tranquil” habitat loss, habitat fragmentation and zones) on the basis of a number of disturbance or loss of species. Indicators spatial/ecological criteria. are mostly related to habitats. • Functional impacts, e.g. on agriculture The scope and choice of impacts and indica- and forestry: large-scale infrastructure tors should match the appropriate policy development can create barriers, divid- level (in this case the EU level). For the SEA ing functional land units such as parcels of the TEN simple indicators have to be of farmland. They can consume part of used, which are applicable to the whole of the agricultural area and may make the Union. For more local-level assessments further exploitation impossible substitu- (e.g. corridor assessments), other, more tion effect). detailed indicators can be used, which also

Relevant selected spatial and ecological issues for a strategic environment assessment of the TEN Figure 4.1

Theme Environmental objectives Environmental targets Potential indicator

Biodiversity and safeguard of biodiversity conservation of designated vicinity of sites to the nature conservation areas and the wider environment infrastructure (5th EAP, CTP, Birds Directive, Habitats Directive, Ramsar density of sites within buffers Convention, Biodiversity around the infrastructure Convention, etc.) consumption of natural habitats

Water resources sustainable use of water reduction of pollution, protection number of crossings of waterways resources of watersheds, prevention of environmental damage from number of water and groundwater maintenance and improvement shipping activities protection zones touched of groundwater quality

maintenance of the ecological quality of surface fresh water

eduction of discharges into marine water (5th EAP, CTP)

Coastal zones sustainable development of improve the balance of land use number of coastal zones touched coastal zones and their resources and conservation and the use of (5th EAP) natural resources size of coastal zone stretches taken by links performe integrated planning and management

improve co-ordination between relevant EC policies and between EC, national and regional policies

Noise avoid exposure to dangerous reduction of noise exposure area under influence along the links noise levels for health and quality (especially night-time exposure) of life (5th EAP) number of tranquil zones touched

Land resources sustainable maintenance of avoid disrupting functional units land take (agricultural land, forestry) economic activities

improve land-use planning

Figure 4.1: Relevant selected spatial and ecological issues for a strategic environmental assessment of the TEN 14 Spatial and Ecological Assessment of the TEN

account for more specific characteristics of In general, an impact assessment consists of the immediate environment. In short, the three different steps: initial selection of spatial and ecological indicators has been made based on the impact prediction; making an objective esti- following criteria: mate of the type and magnitude of the actions likely to have future impacts; • relevance in regard to Community environmental objectives and targets; impact evaluation; the subjective appraisal of • applicability on EU scale (+ Switzerland). the significance of the predicted impacts, taking into account the sensitivity or impor- As a second step, the working group assessed tance of the receiving environment, and by the feasibility of the impact indicators in comparing (e.g. using multi-criteria analysis) terms of the availability of suitable data and the impact levels with environmental objec- evaluated the uncertainties involved. tives and targets;

Figure 4.1 summarises the impacts and analysis of the uncertainties that underlie the potential indicators that were identified by impact prediction and evaluation. One the working group. The figure also indicates manner of dealing with uncertainties in a SEA the policies that relate to various impacts and is by basing the analysis on a relative compari- indicators. It distinguishes five different son of alternatives, rather than on absolute themes or issues: biodiversity and nature impact predictions. It is of course important conservation, water resources, coastal zones, that the uncertainties at strategic level are not noise and land resources. The environmen- so large as to make subsequent assessments at tal objectives and targets concerning these more local levels of decision-making signifi- issues primarily follow what is laid out in the cantly different (and perhaps contradictory). 5th Environmental Action Plan, the White Paper on the Common Transport Policy, the The working group focused on the first and Birds Directive, the Habitats Directive as well third steps of impact assessment. The current as a number of other international agree- report is an attempt to demonstrate how ments. In general, the aim is to protect and impacts could be predicted using the indica- conserve the areas concerned and in the tors suggested by the working group. The case of noise, to reduce the number of report does not include an evaluation of the people exposed to certain noise levels. impacts, since the predicted impacts should at this stage be seen as methodological tests. An According to its modal scope this study evaluation of the impacts would be prema- focuses on terrestrial, linear TEN infrastructure. However, the working group has consid- ture, i.e. motorways, conventional and high ered the degree of uncertainty linked to the speed rail links and inland waterways. different indicators, taking into account the quality of the currently available data. Airports, maritime ports and terminals are not included, mainly because the TEN In this demonstration phase, a range of guidelines do not identify specific planned assessment techniques was tested, including: projects for nodal points (with the excep- tion of priority project 10, Malpensa Air- • A proximity analysis of the planned TEN port). Naturally, an assessment requires infrastructures to the (legally and scien- proposed plans or actions. For airports the tifically) designated sites. This provides a guidelines do contain outline maps, but the very rudimentary indication of the maps only subdivide the main European potential risk that planned TEN links airports into four categories (international may constitute in terms of disturbance or connecting points, EU connecting point, damage to habitats.Calculations of single regional and accessibility point, and airport indicators for the whole network, such as as part of an airport system) without indicat- total land take (per habitat type). ing any planned projects. As to maritime • A vulnerability mapping analysis, i.e. a ports and inland terminals, the guidelines confrontation of the TEN alternatives provide a general reference, but provide no with sensitive zones. The zonesare outline maps nor an indication of planned defined on the basis of a combination of projects. The Commission has meanwhile various indicators and are evaluated prepared a proposal for the integration of using indices of significance. ports, inland ports and terminals in the trans-European network, in a similar way to All assessment methods described in the that of airports. present report were designed for consistency Methodology 15

with geographical information system (GIS) In order to test the different indicators techniques. A GIS is a computer-based mentioned above, several GIS techniques system used for entering, storing, manipulat- were applied: ing, visualising and analysing geographical and tabular data together. It is clear that • conversion algorithms were used to bring computer hardware and software play a the data into the appropriate formats. central role in a GIS; but data, models, Raster formats proved more suitable for expertise and documentation - orgware - are overlay processes, vector formats were equally, if not more, important. used for neighbourhood functions; • spatial analysis operations were carried GIS tools are particularly useful for environ- out including masking, proximity analy- mental assessments, because they can com- sis, overlay, and data set re-combination; bine different databases and make overlays • generation functions enabled the crea- of different maps. The ability to present data tion of specific support data such as and indicators in the form of maps greatly sampling grids and perimeters. facilitates the understanding and interpreta- tion of the evaluation. It should be noted Each individual indicator should be consid- that the maps included in the present ered a result of a series of data transforma- document are only very small-scale represen- tion and manipulation steps using several of tations of the actual output of the system and the above activities. do not do justice to the degree of detail that is in fact present in the databases. 16 Spatial and Ecological Assessment of the TEN

5. Data and Information Sources

A first important task in the demonstration tion is an elementary requirement for a work consisted of bringing together all proper assessment of the ecological impact necessary data in a single information system of a planned road or rail connection. and checking their quality and appropriate- ness for use in the assessment. Guidance on Also, comprehensive and digitalised informa- the range of assessment methods to be tested tion concerning, for example, the number of was obtained through a review of various motorway lanes or rail tracks is lacking. existing studies and SEA practices, which can be found on a regional, national and, to a The European Commission has initiated a lesser extent, international level. number of activities to enhance the existing database on the TEN projects. These activities involve co-operation with the Member 5.1. Major Data Sources and Limitations States and different international organisations (e.g. WERD and CCFE). Through this 5.1.1. Transport Infrastructure Data data collection exercise, the Commission’s The use of transport infrastructure in the GIS will gradually contain more information SEA requires information on the location, about the quality of the TEN links. However, the mode, the type and the state of the it is not expected that more detailed infor- European Networks. Eurostat-GISCO pro- mation concerning the location of new links vided a number of spatially-referenced data will be obtained. sets covering the necessary information. These databases were reorganised to en- It should also be noted that the scale of the hance their suitability for analysis. available maps for the whole trans-European transport network is much larger than the It is important to note that the infrastructure scale of the maps used in any SEA applica- data have important limitations. The outline tions used thus far. In fact, a recent study on plans that are included in the TEN guide- a SEA concluded that “any GIS mapping at lines only give very rough indications about larger than 1:200.000 scale would be inad- the (future) location of the new infrastruc- equate for SEA purposes”6. ture links. For some links, such information was completely lacking and their location has The working group that prepared the been entered in the Geographical Informa- present document fully acknowledges the 6) Steer Davies Gleave, “State of the art on tion System on the basis of a “best guess”. In limitations of the available infrastructure strategic environmental addition, there is very limited information at data and therefore stresses that the output assessment for transport European level currently on the status of the should not be seen as actual predictions of infrastructure”, prepared planned infrastructure links. In the case of the possible environmental impacts of the for the Directorate- roads, conventional rail and inland water- planned TEN links. The aim of the current General for Transport of the European ways, the guidelines do not distinguish study was primarily to investigate how indica- Commission, Final report upgrades of existing infrastructure from tors and GIS could best be used and to test July 1996, p. 113. completely new links. Of course, this distinc- whether meaningful results can be pro-

Figure 5.1 Available digitalised data concerning European TEN infrastructure

Theme Content Source Year

trans-European rail Network as published in the Official Journal Eurostat 1996

trans-European road Network as published in the Official Journal Eurostat 1996

trans-European water Network as published in the Official Journal Eurostat 1996

European Roads 1/1.000.000 road network Eurostat updated 1996

European Railways 1/1.000.000 rail network Eurostat updated 1997

Specific TEN projects 13 projects as published by the CEC EEA 1997 Data and Information Sources 17

duced, taking into account the quality of the lack of data. The EEA and its Topic Centres available data. are making continuous efforts to solve the problems encountered in the environmental 5.1.2. Environmental Data data sets. The term “environmental data” refers to a large variety of spatially referenced data sets. The main problems are the following: It is the combination of the different data that enables the creation of a model of the • coastal zones have only been defined in a European terrain. From the basic material, linear way; there is no EU definition that derived data sets are produced for assess- enables the identification of the real or ment purposes. total area that should be considered; • the database on population density (to The figure below describes the core raw data be used for noise impact assessment) is used for the present study. All databases have heterogeneous in the sense that data a nominal scale of 1/1.000.000, unless come from different years; Eurostat- specified otherwise. GISCO is working on an improved database; As Figure 5.2 illustrates, some of the databases • the land cover data is a compilation of are rather heterogeneous as far as age is various (national) databases. The EU concerned (land cover and population database is not yet fully harmonised, but density in particular). The stability column the European Topic Centre on Land provides an indication of the need to update Cover is working on this issue; the data sets. “Stable” indicates that the • the databases on designated sites and database concerned does not require fre- nature inventories only contain informa- quent updating (e.g. topography, coastal tion on the size and the centre co- zones and water pattern) whereas “dynamic” ordinates of the sites; information on the databases need regular updating over time actual site boundaries is not (yet) avail- (administrative units and nature inventories). able. DGXI has started initiatives to obtain information on site boundaries; The available environmental spatially refer- • the database on water patterns is hetero enced data on the EU level also suffer geneous and coverage is incomplete. certain limitations, either in scale, coverage The European Topic Centre on Inland or content. Certain impacts (e.g. impacts on Waters is currently involved in updating soil quality and groundwater) had to be the database. excluded from the assessment because of a

Available (digitalised) data for environmental assessment on European scale Figure 5.2

theme content source year stability

Coastal zone derived from administrative regions Eurostat 1996 stable

Topography 1 km resolution digital elevation model Eurostat 1994 stable

Administrative units mask defining the European Territory Eurostat 1996 dynamic

Population population density by smallest Eurostat 1981/93 average administrative units SFSO(CH)

Land cover 1/100.000 CORINE land cover classes ETC/LC 1986/97 average of the EU

Ecological regions 1/3.000.000 digitised map of the ETC/NC 1997 stable European ecological regions

Designated sites sites designated under international ETC/NC 1993 average conventions version

Nature inventories sites complying to scientific criteria of ETC/NC, 1989/95 dynamic importance to nature conservation Birdlife Int.

Water pattern surface water pattern CEC DGXI, 1985 stable ETC/IW 18 Spatial and Ecological Assessment of the TEN

6. TEN Alternatives

6.1. Selection of Alternatives 1. the existing TEN network (existing primary and secondary road, rail and The comparison of alternatives forms the water infrastructure is not included) core of each SEA. In the case of the TEN, 2. the existing TEN network plus the the choice of alternatives must focus on fourteen priority projects that were modal choices and on varying degrees of agreed at the 1994 Essen European network extensions. Council (except project N° 10, the Malpensa Airport) In line with the concept of “tiering” and for 3. the existing TEN network plus all practical reasons, it seemed appropriate that planned rail projects an assessment of an European network does 4. the existing TEN network plus all not consider alternatives for each individual planned inland waterway projects link, but rather illustrates the (potential) 5. the existing TEN network plus all environmental impact of different overall planned road projects network settings (or scenarios). The selec- 6. the complete multi-modal TEN tion of the alternatives to consider is an issue in itself that requires further development The first alternative represents the “do- and also a wide consultation process. It nothing” option (or present situation) that demonstrates that the European network can be considered as a reference situation assessment should primarily be seen as a tool against which other alternatives can be to raise awareness about the possible envi- evaluated. Variant 2 could be considered a ronmental consequences of the TEN, which “do-minimum” alternative. The other vari- may compare different general orientations ants represent differing degrees of imple- for the TEN. mentation of the TEN plans. Variants 3 to 6 are based on the outline maps 2 to 4 in For this pilot assessment, six alternative Annex 1 of the TEN guidelines; planned network developments have been selected. projects are those projects indicated as They must be considered only as examples of “planned” in the outline maps. The projects possible options. The six network alterna- of specific interest are those projects that tives are: were adopted by the European Council at

Figure 6.1 Decision No 1692/96/EC Annex III priority projects

Project Mode Member States Location

1 High speed rail DE, AU, IT Berlin-Nürenberg, Brenner axis

2 HSR UK, FR, BE, NL, DE Paris-Brussels-Köln-Amsterdam-London

3 HSR FR, ES Madrid-Montpellier, Madrid-Dax

4 HSR FR, LU, DE Paris-Karlsruhe, Mannheim, Luxembourg

5 Rail NL, DE Betuwe Line

6 Combined FR, IT France-Italy

7 Road GR Greek Motorways

8 Road PT, ES Motorway Lisboa-Valladolid

9 Rail IR L link Cork-Dublin-Belfast-Larne-Stranraer

11 Rail/road DK, SE Øresund link (Denmark-Sweden)

12 Rail/road SE, FI Nordic Triangle

13 Road IRL, UK, NL, BE Ireland/UK/Benelux road link

14 Rail UK West Coast main line TEN Alternatives 19

The three transport modes included in this study and a multi-modal overview including the 14 specific projects Figure 6.2

Essen on 9 and 10 December 1994 and are evaluation of the modal liability for environ- contained in Annex III of the TEN guide- mental impacts. lines. Project n° 10, the Malpensa Airport, is not considered in this report, as this study’s As mentioned before, plans for non-linear scope is limited to linear infrastructure infrastructure (maritime ports and airports) plans. have not been taken into account.

Hence, the alternatives should not be inter- The working group acknowledges that by preted as true policy options. They reflect defining the variants in the above mentioned (hypothetical) extreme situations, but make way, the “added value” for decision-making is possible the identification of the range of limited, though important in raising aware- possible outcomes concerning the environ- ness about the environmental implications mental impact of the TEN and allow an and modal liability of the TEN development. 20 Spatial and Ecological Assessment of the TEN

Figure 6.3 Example of the multi-modal TEN network compared to the national and regional network

It would be preferable to define alternatives and extensive consultation and discussion that represent true strategic options and with all parties involved, in particular the possibly different investment levels. One Member States. The working group consid- could for example think of a development ered the definition of such TEN strategy variant that focuses on key corridors, a options outside the scope of its primarily development variant that primarily aims at methodological work, but recommends this strengthening cohesion, a variant that as an important issue for further research. requires limited investment and a variant that requires high investment levels. How- Figure 6.2 consists of four maps showing the ever, definition of such variants requires existing and planned infrastructure that is selection or prioritisation of individual links considered in the current study. and therefore requires careful preparation

Estimated length of the trans-European transport network Figure 6.4 compared to the overall European infrastrukture (kilometres)

mode existing TEN (1996) planned TEN (2010)* existing overall European infrastructure**

upgrade new

road 49,598 n.a. +26,875 49,024***

conventional rail 48,477 n.a. +1,372 155,836

high speed rail 4,901 +14,408 +10,088 2,406

inland waterways 12,239 n.a +1,412 30,191

* All planned roads, conventional rail links and waterways are assumed to be new ** Source: European Commission, “EU transport in figures - statistical pocketbook”, 2nd issue 1997 (figures refer to 1995) *** Motorways only TEN Alternatives 21

6.2. The TEN in Perspective the planned links. The estimates were performed at a cartographic scale of 1/ In order to facilitate the understanding of 1.000.000 and provided the following cumu- the scope of the trans-European transport lative lengths of the network alternatives. network, Figure 6.3 sketches the density of The calculations are based on the assump- the TEN compared to the national and tions mentioned earlier, i.e. all planned regional road and rail networks. roads, conventional rail and inland waterways are considered to be new infrastructure. Figure 6.4 shows a different way of putting the A comparison of these measured values is trans-European network into perspective. It made with the reference values made by provides graphical estimates of the length of DGVII. 22 Spatial and Ecological Assessment of the TEN

7. Nature and Biodiversity

7.1. Main Objectives Concerning Nature broader sense. The latter would include and Biodiversity other sites, like habitats and species, as well as buffer-zones and corridors. This leads to The importance of safeguarding nature and two different types of analyses. biodiversity is undisputed. It is also one of the key priorities of the EU’s 5th Environ- The first analysis focuses on the vicinity of mental Action Plan. the existing and planned TEN infrastructures to the (legally and scientifically) The construction and use of roads, railways designated sites of nature conservation and canals produces several negative effects interest. If a link is located in or near a on nature and biodiversity. The infrastruc- designated site, it is assumed to constitute a ture in itself may cause a loss of nature sites potential risk to the site concerned. or lead to partitioning and isolating ecosystems and species populations. Secondly, the The second evaluation method specifically traffic that is using the infrastructure can considers the partitioning effect of the TEN disturb nature through noise, vibrations and infrastructure. The partitioning, or fragmen- accidents. tation effect constitutes an important element of the adverse environmental impacts Different conventions and agreements have that infrastructure may have. The evaluation been established to protect nature sites and uses information concerning designated biodiversity, ranging from international areas, but also uses other information. legislation to regional programmes. At international level, the United Nations Convention on Biological Diversity (CBD) 7.2. Vicinity to Sites of plays an important role. The signatories to Nature Conservation Interest the Convention (including all EU Member States) have committed themselves to the A first way of estimating the impact of the conservation of biological diversity and the TEN infrastructure on nature and sustainable use of its components.7 In 1998, biodiversity is through identifying the vicinity the European Commission adopted a Com- of planned and existing TEN links to desig- munity Biodiversity Strategy in view of the nated nature sites. If a link runs through or fulfilment of the European Community’s is located near a designated site, it may be obligations under the CBD. assumed that there is a potential risk of adverse impacts to the site concerned. At European level, the 5th Environmental Action Plan obviously constitutes a central This type of analysis has been suggested and element, together with the Directive on the tested before by the Royal Society for the conservation of natural habitats and of wild Protection of Birds (RSPB) in its report “The fauna and flora (commonly referred to as impact of trans-European networks on the Habitats Directive) and the Directive on nature conservation: a pilot project” (1995). the conservation of wild birds (Birds Direc- 7) The CBD defines tive)8. The EU policy aims to create an It should be noted that the strategic network “biological diversity” as ecological network of protected areas across assessment can only identify potential risks; the variability among living organisms from all the Union, the so-called Natura 2000 net- the actual impacts naturally depend on the sources including, inter work. The network will consist of areas exact location of the link (and site), the type alia, terrestrial, marine designated under the Habitats and Birds of construction and the mitigation measures and other aquatic Directive. that are taken. However at the strategic level, ecosystems and the it is neither possible nor appropriate to take ecological complexes of which they are part; this When it comes to assessing a potential account of detailed information per link (in includes diversity within conflict the TEN might create in respect to accordance with the principle of “tiering”, species, between nature conservation interests and biodiver- this should be left to corridor or project species and of eco- sity maintenance, the question arises assessment). The aim of the network assess- systems. whether only impacts on formally designated ment is simply to present an overall picture 8) Council Directives 92/ sites should be examined or whether nature of the possible environmental impact of the 43/EC and 79/409/EC. and biodiversity should be interpreted in a TEN. Of course, by presenting potential risks Nature and Biodiversity 23

rather than actual impacts, the assessment known. Instead, the databases provide centre paints a “worst-case scenario”: not all poten- co-ordinates and an indication of the surface tial risks will become real impacts. of the site. Secondly, certain databases do not cover all countries and/or contain data 7.2.1. Methodological Approach of varying age (this is particularly true for Using a methodological approach, the first the CORINE Biotopes ). Thirdly, areas may question to be answered is what sites are be double counted, as the same site may be taken into account. Several thousand sites designated under various conventions. A are defined throughout Europe. The major- typical example is the Camargue area in the ity, however, can only be mapped on a local south of France. The problem of double scale. Digitalised information on the Euro- counting may be solved in the future once pean scale is currently only available for sites more detailed site descriptions become designated on a formal international or available under Natura 2000. European basis. The nationally designated sites have not Several limitations of the data must be noted. been considered in this study either, due to Firstly, the existing site lists contain only the current lack of consistency and heteroge- extremely limited information. For example, neity of the data (definitions of the sites the actual site boundaries are in general not depend on national legislation that vary

Site description according to its legal statute Figure 7.1

Sites Description

Designated sites protected under community legislation

Bird Directive (SPA) Directive 79/409/EEC provides a general system of protection for all species of wild birds. Its purpose is to regulate the hunting, capturing, killing and sale of such species. Its main objective is the conservation and restoration of natural habitats. Data is not complete for all countries (only a sample of sites has been tested).

Designations under international conventions

European Diploma Network of the CoE to promote better management of important natural sites, scientific research and awareness raising.

Barcelona Convention (16 Feb. 1976). The Convention for the Protection of the Mediterranean Sea constitutes the main legal instrument for the protection of the Mediterranean environment. It is a framework convention that has been progressively developed through the adoption of further instruments and protocols.

Biosphere Reserve The UNESCO “Man and the Biosphere” programme aims to encourage interdisciplinary research, training and demonstration activities with a view to improve the understanding and management of the resources of the various ecosystems. A BR aims to reconcile the conservation of biodiversity with its sustainable use by involving local communities.

Biogenetic Reserve A network of the CoE to include sites with special emphasis on endemic species and biodiversity.

World Heritage The UNESCO World Cultural and Natural Heritage Convention preserves outstanding ecosystems and landscapes.

Ramsar Convention (2 Feb. 1971) The Convention on Wetlands of International importance, with emphasis on waterfowl habitat, aims to combine long-term national policies with co-ordinated international action.

Scientific inventories

CORINE Biotopes A scientific inventory of sites important for nature conservation, regardless of the designation given. Sites were selected according to the presence of species, habitats or landscapes. The data varies considerably in age.

Important Bird Area (IBA) Birdlife International used its global network of national organisations to create a database to monitor the conservation of bird species and habitats, identify and secure conservation of the most important sites, and build a greater understanding of the relationships required for sustainable development. 24 Spatial and Ecological Assessment of the TEN

enormously). The EEA European Topic noise and vibrations - stretches further than Centre for Nature Conservation is compiling the actual borders of the infrastructure itself. a list of all national sites (without however An additional complication arises from the harmonising the definition of the sites) in fact that many sites have an oblong shape, the common database on designated areas. thus are not compact. The working group acknowledges that by making random As already indicated, the information con- sample surveys, an indication of the “average cerning the location of new TEN links, as uncertainty” of the geographical information well as the boundaries of the designated might be obtained. Such sample surveys have sites, is rather imprecise. The databases on not been carried out, but could be a subject the TEN provide only rough indications of for further research. the locations of new links (sometimes “best guesses”) and the nature sites can often only An indication for choosing the width of the be represented symbolically by a circle with a buffer-zones can be obtained from other radius that is proportional to the site surface. studies. The RSPB used for example corridors of 2 and 10 kilometre wide9. The In order to take account of these uncertain- French Ministry of Transport generally ties, a buffer-zone around the infrastructure applies 10 kilometre buffer-zones in the centre lines can be used. Of course, the assessments that are carried out on corridor width of the buffer-zones is a significant level10. Consequently it was decided for the determining factor for the outcome of the current analysis, a 10 kilometre buffer-zone study. On the one hand, the buffer-zone should apply. should be wide enough to reflect all potential impacts in a reasonable way. On the 7.2.2. The Results other hand, the buffer-zone should be There are different ways in which the results narrow enough to avoid unnecessary overes- can be presented. A first option is to identify timation of the possible risks. Ideally the per TEN development variant, the number buffer-zone should be a fair reflection of the of sites that is potentially at risk. Figure 7.2 “average uncertainty” concerning the loca- shows the proportion of potentially threat- tion of the nature site and the planned ened sites for the six network alternatives, infrastructure link. Simultaneously, it should according to its statute. It suggests that the take account of the fact that the impact of development of the planned road network the infrastructure - through for example will increase the number of potentially

Designated sites potentially affected by the development of the trans-European transport network. Figure 7.2 Status in mid 1997 (overlaps between sites not taken into account)

variant (% of sites within 10 km)

site type and inventories 1 2 3 4 5 6 number of sites

S.P.A. 53 53 55 53 60 61 743 (EU12)

Barcelona C. site 43 44 44 44 47 49 94

Eurodiploma site 44 53 47 44 56 59 34

9) RSPB, idem. Biosphere Reserve 66 68 66 68 72 75 47 Biogenetic Reserve 52 53 54 52 57 59 253 10) See Ministère de l’Equipement, des World Heritage 20 20 20 20 20 20 5 Transport et du Tourisme - Direction Ramsar C. site 71 72 72 71 79 80 246 des Routes, “Flux est- ouest au sud du bassin CORINE Biotopes inventory 49 50 51 49 56 58 7740 (EU12) parisien, dossier pour un débat”, Décembre Important bird area 51 52 53 51 58 59 1644 1994. It must be noted Variants: 1) existing TEN network that the scope and 2) existing TEN network + 13 specific projects nature of this study is 3) existing TEN network + all planned rail projects slightly different from 4) existing TEN network + all planned inland waterway projects the network assessment 5) existing TEN network + all planned road projects at European level. ) complete multi-modal TEN Nature and Biodiversity 25

Indication of potential conflicts between complete TEN and nature sites. Figure 7.3 Nature sites classification according to neighbouring links

threatened sites by 14 percent. The develop- (distinction between new links and upgrad- ment of the rail projects has a considerable ing) and for the environmental data (double smaller impact. This is barely surprising counting of sites). considering that the extension of the road network is assumed for this exercise to Figure 7.3 gives a geographical representation consist of around 27,000 kilometres of new of the sites that are potentially at risk. The infrastructure, whereas the rail network is map distinguishes four different risk catego- only extended by around 11,000 kilometres. ries. The sites indicated with a green dot are not within the 10 kilometre buffer-zone of The presented results should only be consid- any of the existing or new TEN infrastructure ered as hypothetical, considering the uncer- links. The sites represented by a blue dot are tainties both for the infrastructure data “threatened” by one or two links, the orange 26 Spatial and Ecological Assessment of the TEN

ones by 3 or 4 links and the red sites by 5 or high probability to exposure. It could be more links. considered that special caution is required when dealing with projects on these loca- Instead of showing the results per site, the tions (red flag approach). outcome can also be presented per link. Figure 7.4 shows the average number of designated sites at risk for each link. The 7.3. Partitioning of Land map is based on a density calculation per link because a longer link naturally has a A particular effect that linear infrastructure higher chance of neighbouring sites than a has on nature - and landscape - is that it shorter link. partitions land or, more specifically, fragments habitats.11 Habitat fragmentation can Combining approaches allows the identifica- be defined as the splitting of natural ecosys- tion of problematic links and sites with a tems into smaller and more isolated units

Indication of potential conflicts between complete TEN and nature sites. Figure 7.4 TEN infrastructure classification according to neighbouring nature sites.

11) More detailed research on the problem of habitat fragmentation is planned in the framework of COST Action 341 (European Co-operation in the field of Scientific and Technical Research). Nature and Biodiversity 27

Number of land units and their cumulative surface - existing and TEN planned situation Figure 7.5

thus endangering the survival of animal and acknowledges that further testing is required plant species and communities. The decline to determine how and to what extent influ- in biodiversity witnessed after a land unit has ences other than the TEN, can be dealt with been split into smaller fragments depends on properly. the condition of the original unit, the size and distribution of the newly formed parts Another simplifying assumption has been and the barrier effect of the infrastructure made-to ignore original land units (deline- delineating the unit. Moreover, partitioning ated by the existing TEN network) that were can constitute a form of visual intrusion to smaller than 1000 km². This was done purely valuable landscapes. for scale reasons. It does not imply that smaller land units of biological interest are 7.3.1. Methodological Approach considered to be unimportant. However, For the above stated reasons, a good indica- because of the rough scale of environmental tor for measuring the partitioning effect data on EU level and the uncertainty of the should take both size and ecological quality location of TEN links, it is not possible, at into account. The size of a unit can be this level, to evaluate effects on smaller land determined graphically; yet to assess its units. This is a task for national, regional or ecological quality is more difficult. Two local level assessment. approaches to the quality problem have been explored, i.e. quality evaluation based on: A similar type of analysis has been tested by the French consultancy BCEOM for the I. the proportion of the land unit area that assessment of national infrastructure plans is designated by international conven- (“Schémas Directeurs”).12 The French study tions considered all main railway lines, all roads II. the proportion of the land unit area that and motorways with an average daily traffic is covered by forest or semi-natural of more than 4000 vehicles per day, all canals habitats. and all natural fragmentations (such as valleys). Even though large-scale maps were If a land unit scores above the European used for other parts of the assessment, average on both observations, it is consid- BCEOM used relatively detailed maps for the ered of high quality; if it fails on both ac- evaluation of habitat fragmentation. The counts, it is considered of lower quality. methodology proved very feasible in a test application at a regional level. 12) BCEOM, “Etude stratégique d’impact For the present exercise, only the TEN links sur l’environnement - were included. The other existing primary 7.3.2. The Results essai méthodologique”, and secondary road and rail infrastructure Again there are different ways in which the Ministère de l’Environ- was not taken into account. It follows that results can be presented. Figure 7.5 shows a nement - Direction de this is a simplistic assumption and that the graph with the number of land units com- la Nature et du Paysage - Sous-direction de derived results can only be considered as pared to their cumulative surface. The green l’Aménagement et du “dummy” results that merely serve to illus- line represents the existing situation; the red Paysage, Décembre trate the methodology. The working group line indicates the situation after implementa- 1994. 28 Spatial and Ecological Assessment of the TEN

tion of the trans-European transport net- existing situation and map 7.6-B shows the work. The red line runs clearly below the situation after the development of the TEN. green line. This illustrates that the same Map 7.6-C illustrates the differences between cumulative area (for example 2,000,000 ha) map 7.6-A and 7.6-B. The green areas on is formed by more individual land units after map 7.6-C have not changed due to the the realisation of the TEN than before. The execution of the TEN; whereas the brown disadvantage of this presentation is that it areas have deteriorated from “high quality” does not take account of the quality of the land (with an above average proportion of fragmented land. designated land and an above average proportion of land covered by forest or semi- Figure 7.6 below illustrates the partitioning natural habitats) to “poor quality” land (a effect taking into account the characteristics below average proportion of designated land of the land units. Map 7.6-A shows the and forest/natural habitats).

Figure 7.6 Partitioning of Land effect on the complete TEN

A. Quality of land units according to size and ecological criteria. Existing situation.

B. Quality of land units according to size and ecological criteria. Complete TEN planned situation.

C. Induced land quality changes by land partitioning effect of the complete TEN (Drop in indicator value set at 20%). Water Resources 29

8. Water Resources

8.1. Main Objectives Concerning the framework for the water policy. The Water Protection of Water Resources Framework Directive would require Member States to prepare a programme of measures Infrastructure building and use can affect to attain “good” surface water and water quality through pollution from exhaust groundwater status by the end of 2010. emissions, de-icing salts, weed control along roads and railways, fuel spills, sewage from There are some regional and international train discharges onto tracks, and pollutants agreements that concur to the same broad associated with transport of hazardous objectives and targets. Regional agreements materials. Transport infrastructure building include the Rhine Action Plan, the Elbe can furthermore affect the hydrological Action Programmes, the Helsinki Conven- functioning of ground and surface water, e.g. tion - Baltic Sea Joint Comprehensive Envi- can disturb the groundwater table and affect ronmental Action Programme - 1993 to the course and run-off of surface water. 2012, the Oslo and Paris Commission [OSPAR] - North Sea Ministerial Confer- Surface and groundwater quality are key ences, the Hague Conference of 1990, the elements in a healthy environment as it Mediterranean Action Plan, and the Arctic forms the very basis of life itself. National Monitoring and Assessment Programme. and international networks monitor water International agreements include the Con- quality , and various kinds of protection vention on the Protection and Use of measures are implemented throughout the Transboundary Watercourses and Interna- Union accordingly. tional Lakes.

The European Union Fifth Environmental Action Programme identified the most 8.2. Crossing of Surface Waters important problems and set priorities. With respect to groundwater and surface freshwa- 8.2.1. Methodological Approach ter, it stressed the need to integrate resource In principle, a similar type of indicator as for conservation and sustainable use criteria into biodiversity and designated areas could be other policies including agriculture, industry used to identify the surface water areas or and land-use planning. Concerning the water catchment areas at risk by determining qualitative aspects, and taking into account the vicinity to new infrastructure links. surface freshwater, it called for the safe- Similar to the former chapter, the actual guarding of existing high quality and the impact on water quality and hydrology are improvement of the ecological quality of extremely difficult to predict as it would water. require accurate knowledge and data, for example, on the complex relation between The proposal for an EU Action Programme the emissions of pollutants and their disper- for Integrated Groundwater Protection and sion in the water. Again, the strategic assess- Management (COM(96) 315 final), adopted ment at European level can only aim to by the Commission in August 1996, together identify potential impact risks on the basis of with the proposal for a Water Framework a very simplified risk indicator. Directive (COM(97) 49 final), issued in February 1997, aim to protect groundwater, It did not appear feasible to consider water inland surface waters, estuaries, coastal catchment areas simply because EU-wide waters and groundwater, which form the spatially referenced information on the

Water crossings of the existing trans-European transport network Figure 8.1

observed crossings road rail

number of crossings of the existing TEN network 3373 6050

density of crossings of the existing TEN network 6.8/100km 8.9/100km 30 Spatial and Ecological Assessment of the TEN

Figure 8.2 Comparison of the number of water crossings under different TEN variants

Alternative road rail

existing TEN network 100 100

existing TEN network + 14 projects of specific interest 108 106

existing TEN network + all planned rail projects 100 116

existing TEN network + all planned road projects 166 100

complete multi-modal TEN (rail and road links only) 166 116

Figure 8.3 Complete Multimodal TEN classification according to potential conflicts with surface water Water Resources 31

quality of surface water and groundwater Figure 8.2 presents the increase in the resources is lacking. Therefore, only a very number of crossings compared to the exist- simplistic, morphological indicator was ing network. The figures are given as indices tested. It is based on the crossing of surface (existing network = 100). The table suggests water. In other words, the density of cross- an extremely large increase of water crossings, measuring the potential proximity ings due to the construction of new roads. between links and surface waters, was consid- However, similar to all results on the previ- ered as an indicator for environmental risk ous chapter, the outcomes should not be to surface waters. seen as estimates of the actual impact, but rather as “dummies” to illustrate the output For self-evident reasons, the TEN inland of a methodology. waterways were not included in this exercise. Results can also be presented per link. 8.2.2. The Results “Critical” links can be identified by measur- Figure 8.1 summarises the number of water ing the density of surface water crossings crossings and the density of crossings (per they induce. Figure 8.3 includes a map drawn 100 kilometre of infrastructure) on the for the complete TEN variant and provides existing situation. It suggests that on average an illustration of this type of presentation. the rail network crosses more water patterns than the road network. 32 Spatial and Ecological Assessment of the TEN

9. Noise

9.1. Main Objectives Concerning Noise Nevertheless, two approaches have been tested. The first approach considered the Noise pollution is one of the most tangible population within buffer-zones around the and direct effects related to the use of existing and new infrastructure. The second transport infrastructure. The importance of approach consisted of the identification of this issue on urban areas was pointed out in so-called “tranquil zones” that could be at the 5EAP - where it was shown that more risk due to the TEN infrastructure develop- than 16% of the Community population ment. In fact, the two approaches illustrate suffers because of noise levels over Leq 65 typical issues when assessing noise. On the dB(A) during the night, mainly resulting one hand, the aim is to avoid noise exposure from road and air traffic. Noise pollution for people in their dwellings, offices and has also recently been highlighted in the work places, and on the other hand, the aim Commission’s green paper on a future noise is to maintain “tranquil areas” both for policy13. reasons of nature conservation and recrea- tion. Noise is an impact that can be effectively mitigated at project level (e.g. through the use of tunnels, cuttings, noise screens). Still, 9.2. People in the Vicinity of Infrastructure it is also important to consider the impact at earlier planning stages, in particular at the 9.2.1. Methodological Approach level of route determination. The increase in the number of people susceptible to noise disturbance could be Noise emissions not only depend on traffic used as an indicator. Ideally, the indicator for characteristics (e.g. traffic intensity, speeds) noise impact should take into account the and vehicle technology but also on the traffic intensity for the infrastructure con- infrastructure design and materials. Further- cerned and identify critical noise contours more, the degree of nuisance that noise (e.g. 65 dBA). However, this would require emissions cause is very much determined by traffic forecasts and modelling, which is local elements, as noise propagation de- outside the scope of this report. pends largely on the typology of the local environment (i.e. topography, existing In order to measure the increase in the obstacles and screens such as buildings or number of people susceptible to noise vegetation) and meteorological conditions. disturbance, a database describing the This means that although noise levels can be detailed position of dwellings, offices and effectively measured and modelled at a local other working places (e.g. special ones such scale, it is much less straightforward to as schools and hospitals) for the entire model noise impact quantitatively at Euro- European Union would be necessary. Such a pean scale. database is, however, not available. The most

Figure 9.1 Population susceptible to noise exposure due to TEN infrastructure

Influence zones Alternative 250 m 1 km 5 km

Existing TEN network 100 100 100

Existing TEN network + 14 projects of specific interest 104 103 101

Existing TEN network + all planned rail projects 107 105 102 13 ) European Existing TEN network + all planned water projects 101 100 100 Commission (1996), Future Noise Policy, Existing TEN network + all planned road projects 109 107 105 European Commission Green Paper, Complete multi-modal TEN 116 113 107 COM(96)540 final Noise 33

Population exposed to noise impacts per TEN variant Figure 9.2

250 m 1 km 5 km 120 120 120 115 115 115 110 110 110 105 105 105 increase increase increase 100 100 100 95 95 95 Var. 4 Var. 5 Var. 6 Var. 4 Var. 5 Var. 6 Var. 4 Var. 5 Var. 6 Var. Var. 1 Var. 2 Var. 1 Var. 2 Var. 1 Var. 2 Var. 3 Var. 3 Var. 3

detailed information is on population and is The results can also be presented in dia- available at local administrative level (NUTs grams (Figure 9.2) or on maps. Figure 9.3 level 5, e.g. communes). By assuming that shows an overlap of the population density this population is evenly distributed over the and the multi-modal network. territory of the considered administrative level, one obtains a fairly accurate population density map at European scale. 9.3. Crossing of Tranquil Zones

An additional problem in assessing noise 9.3.1. Methodological Approach impact is that its harmful effects are rather The French consultancy BCEOM suggested local. Risks will generally only occur within a in its 1997 report for the Ministry of Environ- 250 metre buffer-zone around the infrastruc- ment to identify “tranquil zones.” This is to a ture. Inhabitants living at greater distances certain extent comparable to the definition would only perceive disturbance from the of non-fragmented areas in Chapter 7. constant background noise from, for exam- Tranquil zones have been defined as (semi-) ple, a motorway. On the other hand, the natural areas - forests, semi-natural land and uncertainty around the exact location of wetlands - that are at least 7.5 kilometres future TEN links is quite large. Partly to take remote from existing TEN infrastructure14. account of this uncertainty, the use of a 10 kilometre wide buffer-zone was suggested in Chapter 7 (analysis of nature sites). Taking account of the different kind of impacts of Population density map Figure 9.3 each subject, namely its spatial behaviour and multi-modal TEN network using a similar buffer-zone for estimating possible noise impacts, would lead to a disproportionate overestimation of the effect. Therefore, a 1 kilometre and a 5 kilometre buffer exercise was performed.

9.2.2. The Results Figure 9.1 presents the proportional increase of the population in the vicinity of the TEN infrastructure. The table sketches outcomes for the different TEN variants using a 250 14) Ministère de metre, a 1 kilometre and a 5 kilometre l’environnement (1997), buffer-zone. The results are presented in the Evaluation environne- form of index figures (existing network = mentale des schémas 100). d’infrastructures de transport - Essai méthodologique The As in the case of all other indicators, one can French report argues argue that for a proper evaluation of the that, on average, noise results, the effects of the different TEN generated by transport variants should not be compared to the infrastructure can be detected as far as 7.5 existing trans-European network, but to the kilometres from the effect of the general existing infrastructure. source track. 34 Spatial and Ecological Assessment of the TEN

Figure 9.4 Intersection with tranquil zones per TEN variant

Intersection with Alternative tranquil zones (km2)%

Existing TEN network 411 413 100

Existing TEN network + 14 projects of specific interest 427 208 104

Existing TEN network + all planned rail projects 435 615 106

Existing TEN network + all planned water projects 413 277 101

Existing TEN network + all planned road projects 490 902 119

Complete multi-modal TEN 512 556 125

This excludes areas where noise producing A possible indicator for disturbance of human activities are intrinsically present tranquil zones is the intersection between (urbanised areas, areas with predominant those zones and new TEN infrastructure industrial and agricultural activities). (including a 7.5 km influence zone). This can be estimated within a geographical Although noise generated by transport information system by overlaying the net- infrastructure can be detected as far as 7.5 works with the tranquil areas. The surface of km from the source track, at this distance its the tranquil area that falls within the 7.5 level will not be high enough to have actual kilometre buffer-zone around the infrastruc- health effects, even if it can still render a ture is calculated relatively easily. In fact, by certain disturbance. “Silence” is a quality that calculating overlapping surfaces, this ap- becomes more and more difficult to find in proach goes a step further than a simple our society. On the one hand, the possibility vicinity analysis that was used in Chapter 7. for people to temporarily escape urban This can be justified for two reasons: firstly, pressures and to relax in remote tranquil the actual boundaries of the tranquil zones areas contributes to the general quality of are known, whereas the designated areas life. Tranquillity may also be an essential were represented only by a circle around the prerequisite for nature conservation. centre co-ordinates. Secondly, the number and size of the tranquil zones (under the current definition) is so much larger that it seems less appropriate to speak of individual tranquil zone, but rather consider it as a specific land cover type.

Intersection with tranquil zones 9.3.2 The Results Figure 9.5 according to TEN variants The results can be presented in a table such as the one in Figure 9.4, and by Figure 9.5, which summarise the overall increase in the 130 intersection between tranquil zones and TEN infrastructure per variant, compared to 125 the intersection with the existing network. The effects of developing the rail projects 120 appear to be less than those of the development of the planned road network, which triggers both a partitioning and a deteriora- 115 tion process. Both options seem to have a cumulative effect on the amount of tranquil 110

% increase zones threatened.

105 Again, this indicator does not account for the loss of smaller tranquil areas, as the data 100 available for an EU level assessment do not include such smaller-scale effects. These 95 should be addressed at regional or corridor Var. 1 Var. 2Var. 3 Var. 4 Var. 5 Var. 6 level. Noise 35

A critique, which also applies to this indica- Noise disturbance zones around Figure 9.6 tor, is the fact that other infrastructures, not the Complete Multimodal TEN network included in the TEN, were not considered. Possible improvement in the use of this indicator could be achieved by also including non-TEN infrastructure in the definition criteria for tranquil areas. In this respect, an investigation should be made as to which infrastructure level should be taken into account (i.e. should all roads included in the GISCO secondary road network be included, or only roads of a certain importance).

The results can also be shown on a map as the one displayed on Figure 9.6, useful merely as an illustration of the indicator analysed. 36 Spatial and Ecological Assessment of the TEN

10. Land Resources

10.1. Main Objectives Concerning simply look at the land take of existing and Land Resources new infrastructure. It was not considered appropriate to merely examine designated In Europe, land resources are relatively nature sites, populated areas, tranquil zones, scarce, and reaching a sustainable balance water catchment areas, and so on separately. between competing land uses is a key issue Instead, it was thought important to consider for all development policies. Land use any land take as potentially indicating planning and land resources are not dealt adverse environmental impacts. For exam- with specifically as strategic themes in the ple, land take in semi-natural areas may lead 5th Environmental Action Programme. to a decrease of biodiversity. Land take in However, their importance is fully refer- urban areas could represent the risk of enced as a requisite for some strategic issues impacts on humans (safety and noise) and such as biodiversity and nature conservation, the use of agricultural or forestry land could management of water resources and coastal be considered harmful for nature in general, zones, as well as for the sustainable develop- for landscape (visual intrusion) or even for ment of several selected target sectors, such certain economic activities. as transport, agriculture and tourism. Naturally, looking at land take tout court is Thus, identifying land take according to the an ultimate simplification and a “worst-case” various major habitat types of land cover can estimate, but given the flaws in the methods constitute an indicator for a wide variety of and data (in particular) available for more impacts. For example, the calculation of land sophisticated analyses, this crude indicator take in semi-natural areas can be considered has its merits. One could go one step further an indicator for a potential decrease of and, as a minimum, take account of the land biodiversity. Similarly, land take in urban cover types, but the report will show that this areas may indicate the risk of impacts on method has the same limitations. humans, while the taking of agricultural or forestry land may indicate how the TEN may 10.2.1. Methodological Approach affect certain economic activities. Roads, rail and waterways consume land directly or indirectly. Direct land take refers to the land needed for the actual infrastruc- 10.2. Land Take ture and is a function of the number of traffic lanes or tracks, the number of safety The broadest way of assessing the spatial and lanes, embankments, dug-outs, etc. Indirect ecological impact of the TEN would be to land take is caused by land reallocation due

Figure 10.1 Estimated surface requirements per infrastructure type

Source: EUROPEAN COMMISSION (1993), Traffic mode Total cross section (m) Surface (ha/km) The European HSR Network: Rail Conventional 25 2.5 Environmental Impact HSR upgrade 32 3.2 Assessment HSR new 35 3.5

Road 2x1 lanes 60 6

2x2 lanes 90 9

2x3 lanes 100 10

2x4 lanes 120 12 Land Resources 37

Land take by TEN variant Figure 10.2

Land take Alternative Area (km2) Score

Existing TEN network 8,598 100

Existing TEN network + 14 specific projects 9,408 109

Existing TEN network + all planned rail projects 8,964 104

Existing TEN network + all planned inland waterway projects 8,767 102

Existing TEN network + all planned road projects 11,285 131

Complete TEN 12,632 137

to the infrastructure construction, such as to widths of 32 up to 120 meters. By combin- parking places, rail stations, etc. ing these factors with the estimated length of the existing and new network, total land take In the present study, only direct land take has can be obtained. Results are summed up in been considered. The proposed indicator is Figure 10.2, which presents total land take for direct land take (derived from the lengths of the existing network and the 5 network the infrastructure development) differenti- variants. The comparison of the figures is ated by major land cover types that underlie facilitated by using index scores (the existing the existing and planned links. Of course, the network is 100). The smaller width of rail fact that the location of the routes is still not infrastructure explains the “favourable” precisely defined introduces a factor of score of the planned rail network. uncertainty into the estimates. It goes without saying that land take in itself The surface required by transport infrastruc- is not necessarily directly environmentally ture is directly related to its construction harmful. The potential impact depends on characteristics. Because these are increas- the type of land that is affected. A refine- ingly set to international standards, the ment of this indicator can be achieved taking variations between different countries and into account land cover types. Land cover data sources are marginal. Figure 10.1 illus- types were obtained from the European trates the surface requirements of different CORINE land cover inventory supplemented infrastructures. The figures are based on a by national land cover inventories. study carried out for DGVII of the European Commission in 1993. The land type or habitat information that was used for this study is available at a resolution To get an indication of the land use effi- of 250 metres, which seems compatible with ciency of each mode, the direct land take the infrastructure widths mentioned. must furthermore be put into the perspective of the related traffic capacity that the The CORINE database that was used for this infrastructure offers. For example, taking analysis distinguishes seven major land cover into account the capacities of a motorway or “habitat” types (in order of increasing (1500 person car units per hour per traffic environmental sensitivity): lane) and of a HSR (15 trains per hour per direction), the capacity of a HSR line equals • vegetated, strongly artificial land (which that of a motorway with 2x4 lanes, but includes artificial non-agricultural requires much less land resources.15 vegetated areas, arable land, permanent crops and annual crops associated with Data on the number of traffic lanes for TEN permanent crops); roads and the number of rail tracks per TEN- • vegetated, less artificial land (which in- link are as yet not available. For the purpose cludes pasture, complex cultivation of the present demonstration work, the patterns, agriculture with natural vegeta- amount of land take by linear infrastructure tion and agro-forestry areas); 15) European Commis- is estimated to take on average 3.2 ha/km for • semi-natural land (which includes shrub sion (1993), The Euro- pean HST Network: rail lines, 12 ha/km for inland waterways and and/or herbaceous vegetation and open Environmental Impact 10 ha/km for motorways. This corresponds space with little or no vegetation); Assessment 38 Spatial and Ecological Assessment of the TEN

Figure 10.3 Distribution of the major land cover types in Europe Land Resources 39

• artificial land (which includes urban Based on a quantitative approach, this areas, industrial areas and mines); method uses categories or impact group • forests; selection, ratings or assignment of priorities • water bodies (which includes inland and can also yield a large index calculation. waters, coastal lagoons and estuaries); Map overlay techniques are the most suitable • wetlands (which includes inland and to perform this methodology, which was a coastal wetlands). precursor for early GIS developments.

If the different major habitat types are 10.3.1. Methodological Approach considered, it is clear that vegetated land will The present identification of sensitive zones suffer the most. However, data also shows the is based on a methodological approach impact of the planned road network on which, generally combines the use of GIS forests and semi-natural habitats. and the use of indices, and which integrates existing environmental and non-environmental information into general potential 10.3. Sensitive Zones Mapping impact maps. The aim is to give an indication of the sensitivity of an environmental A second option for refining an indicator system (e.g. local community, ecosystems) based on land use would be by defining according to the potential impacts of a sensitive zones on the basis of more criteria certain activity (such as transport). The than land cover type only. Various countries method consists of the following basic steps: commonly use “sensitive zone mapping” as a tool in strategic environmental assessment. • The selection and mapping of areas that However, given the scale of the TEN and the are considered sensitive for specific data limitations at EU scale, only a very related impacts, in this case sensitive to simplified version of these national or transport infrastructure. This is done on regional approaches can be applied. the basis of an analysis of the geographic, demographic and environmental charac- Sensitive zone mapping derives from the so- teristics of the area and by aggregating a called land-suitability analysis, an approach limited number of spatial/ecological commonly used in the USA since the early indicators. This is carried out from the 1970s. Its main characteristics are that: i) it perspective of different “impact groups”; focuses principally on natural characteristics • An appraisal of the expected sensitivity of the land, that is, it can be spatially refer- of the area using simple qualitative enced and reported; ii) expert judgements indices such as “not sensitive”, “moder- (based on scientific knowledge) play a ately sensitive” or “highly sensitive”. central role in the method; and iii) impacts are implied but not explicitly reported.

Land take by land cover type for TEN variants Figure 10.4

140.0 Variants : 1) existing TEN network; 120.0 2) existing TEN network + 14 projects of specific interest; 100.0 3) existing TEN network + all planned rail projects; 4) existing TEN network + 80.0 vegetated-strongly artificial all planned inland vegetated-less artificial waterway projects; 60.0 semi-natural 5) existing TEN network + artificial all planned road projects; amount of land 40.0 forests 6) complete multi-modal water bodies TEN. The sum of the wetland contributions for variant 1 20.0 adds up to 100.

0.0 1 2 3 4 5 6 variant 40 Spatial and Ecological Assessment of the TEN

Figure 10.5 Sensitivity thresholds attributed to baseline data sets

* class 1: lowland; Sensitivity group Baseline data Low sensitivity Moderate sensitivity High sensitivity class 2: hilly land; class 3 : mountainous. I water patterns <10 km /cell 10-20 km/cell >20km/cell ** class 1: within 1 km semi-natural <25% SN area 25 -75% SN area >75% SN area from the coast; class 2: between 1 and mountainous* class 1 dominant class 2 dominant class 3 dominant 5 km from the coast; class 3: between 5 and II coastal zone** class 1 present class 2 present class 3 present 10 km from the coast. designated area <2% DG area 2-10% DG area >10% DG area

III population <50 inh/km2 50 - 250 inh/km2 >250 inh/km2

5 - 20 % BU area

built up area <5 % BU area >20% BU area

An overlay of the infrastructure outline plans patterns, occurrence of semi-natural with the sensitivity maps enables the identifi- land, and mainly mountainous character- cation of areas where conflicts can arise. The istics (possible environmental impacts objective of infrastructure planning should include habitat fragmentation, loss of be to avoid, where possible, crossing through biodiversity, and visual intrusion of highly sensitive zones, or, when this is una- landscapes); voidable, to ensure that proper mitigation • Group II measures are taken. Sensitivity based on environmental legal status (biodiversity, habitat frag- The mapping of sensitive areas is done mentation, coastal zones); according to “impact groups” or “sensitivity • Group III groups” using existing environmental, Sensitivity based on territorial demo- physical and social baseline data. graphic characteristics, such as population density or share of built up area For the present report, three different (possible environmental impacts are sensitivity groups were distinguished: noise and accidents).

• Group I The present overall assessment scheme uses Sensitivity based on the general environ- water patterns, semi-natural land, and mental physical nature of the area: water mountainous areas (derived from DEM

Figure 10.6 Assignment rules for determining sensitivity per group

Combination of two risk data set (group II and III) Combination of three risk data set (group I)

Original celle score Resulting Original celle score Resulting sensitivity sensitivity

high + high high high + high + high high

high + moderate high high + high + moderate high

moderate + moderate moderate high + moderate + moderate high

high + low moderate high + moderate + low moderate

moderate + low low high + low + low moderate

low + low low moderate + moderate + moderate moderate

moderate + moderate + low moderate

moderate + low + low low

low + low + low low Land Resources 41

Sensitivity map based on Group I criteria (physical-environmental criteria) Figure 10.7

Note: surface water data for Finland and northern Sweden is missing

altitude and slope data-base) as GROUP I Using expert judgement, simple scores for base-line data; coastal zone boundaries and sensitivity indices are attributed to the areas. designated areas as GROUP II base-line data; A three-class criteria or sensitivity threshold and population and built-up area as GROUP was attributed to the 7 baseline data sets. As a III base-line data. GROUP II impact group result of this operation, each baseline data includes coastal zones, as they constitute an set will provide a sensitivity distribution map irreplaceable environmental heritage with containing three impact classes. The thresh- unique ecological, cultural and economic olds are applied to square territorial units or resources subject to increasing pressures. As cells measuring 10 km by 10 km. Data an example, approximately a third of the analysis was conducted on a raster format. Union’s wetlands are located on the coast, as well as more than 30% of the SPAs. To combine the above indices into an impact group, assignment rules were used. Sensitiv- 42 Spatial and Ecological Assessment of the TEN

Figure 10.8 Sensitivity map based on Group II criteria (environmental legal status criteria)

ity maps were then built, based on these data Figure 10.5 illustrates the criteria or sensitivity sets, and according to defined assignment thresholds for the 7 baseline data sets. As a rules. Semi-natural areas, mountainous areas result of this operation, each baseline data and the density of surface water patterns set will provide a sensitivity distribution map were combined and overlaid to estimate the containing three impact classes (low sensitiv- environmental/physical sensitivity of the ity, moderate sensitivity and high sensitivity). area. Secondly, legally defined nature sites The thresholds are applied to square territo- and coastal zones combined to form an rial units or cells (“grids”) measuring 10 by environmental/legal sensitivity map. Lastly, 10 kilometres. The thresholds can only be built up areas derived from the major land considered, in this exercise, as rules of cover types database and population density, thumb. They are commonly, methodologi- gave rise to a territorial sensitivity map. cally based on expert judgement, which is Land Resources 43

Sensitivity map based on Group III criteria (demographic criteria) Figure 10.9

generally derived from the application of according to each impact group - that are some expert judgement assessment rule, crossed by infrastructure links. which was in fact not the case. 10.3.2. The Results The next step was to combine the above Applying the described methodology, sensi- indices to identify a general measure of tivity maps were processed for each impact sensitivity per group based on agreed assign- or sensitivity group. These maps are pre- ment rules. (Figure 10.6) sented in Figures 10.7, 10.8 and 10.9.

Overlaying the multi-modal TEN network Network assessment was then performed by with the three sensitivity maps enables the identifying those links that cross the sensitive identification of high risk cells (“red flags”) areas, using spatial overlay techniques. Each 44 Spatial and Ecological Assessment of the TEN

Overlay of the complete multimodal TEN network with sensitivity maps. Figure 10.10 Identification of high sensitivity cells crossed by infrastructure links.

Figure 10.11 Sensitivity index per impact group and TEN variant

Impact

Alternative Group I Group II Group III

Existing TEN network 100 100 100

Existing TEN network + 14 specific projects 111 112 111

Existing TEN network + all planned rail projects 110 108 108

Existing TEN network + all planned inland waterway projects 101 101 101

Existing TEN network + all planned road projects 125 125 118 Land Resources 45

Sensitivity index graphs per impact group and TEN variant Figure 10.12

Group I Group II Group III 130 130 130 125 125 125 120 120 120 115 115 115

increase 110 increase 110 increase 110 105 105 105 100 100 100 95 95 95 Var. 1 Var. 2 Var. 3 Var. 4 Var. 5 Var. 1 Var. 2 Var. 3 Var. 4 Var. 5 Var. 1 Var. 2 Var. 3 Var. 4 Var. 5 Var.

link is subsequently classified according to forward the potential risks vis-à-vis all the highest sensitivity category it crosses three sensitivity categories. (Figure 10.10), thus forming an average 2. The incomplete water pattern baseline sensitivity index for each variant (Figure data set illustrates the importance of 10.11 and Figure 10.12). high quality source material. Data gaps and data heterogeneity complicate the Figures 10.11 and 10.12 show that the specific risk assessment process and are clearly projects are primarily located in built up and distinguished in the final results. populated areas and that they, compared to 3. In this exercise, all baseline data were overall planned road and rail networks, have considered equivalent. If some kind of less impact on the more “natural” areas. weighting were applied to accentuate or attenuate some vulnerability factors, fine- An investigation of Figures 10.7 to 10.9 tuning could be within reach. However, demonstrates the following: irresponsible or biased use of the weight factors could easily undermine the goal 1. As it is not always feasible to avoid both of this procedure, which is to obtain an areas of nature conservation interest and objective and general vulnerability areas of dense population (as these are assessment of the territory. to a certain extent complementary), choices must be made with respect to the Finally, the present method could be signifi- prioritisation of sensitive zones. However, cantly improved once estimates become the present method does not attempt available on the predicted magnitude of this “weighting”, as such decisions should traffic impacts on the links (i.e. traffic be taken on a broader policy level (e.g. volume, emissions of pollutants and noise). in the context of existing national or This new information could take into ac- regional policies and circumstances), count the expected environmental pressures, and in consultation between decision- which can differ amongst modes and makers and stakeholders. The only aim amongst links. of the present method is to clearly bring 46 Spatial and Ecological Assessment of the TEN

11. Evaluation and Conclusions

11.1. Indicator Overview The use of these indicators showed that they are quite different in concept, scope, com- As stated before, one of the main aims of the prehensiveness, usefulness for decision- working group was to identify possible making and communication to a broader indicators for explaining spatial and ecologi- public, as well as in the degree of uncertainty cal impacts of the TEN, considering both that is linked to them, taking into account Community-level environmental objectives the quality of the currently available data. and targets, as well as data availability. The scope of the ecological and spatial impacts The Working Group felt that some assess- suggested the following impact categories: ment procedure should be applied to the chosen and tested set of indicators, keeping • Ecological impacts of infrastructure, i.e. in mind further development of the exercise, impacts that are directly related to namely additional research and consultation, ecosystems, habitats and species, as well as well as impact evaluation on a strategic as the degree of biodiversity; level of TEN alternatives. • Functional impacts like those derived from barriers, as a division of functional In 1993, the OECD defined the following land units and the reduction of the issues as criteria for indicator selection: viability of land units; policy relevance and utility for users, analyti- • Impacts with a spatial dimension like cal soundness, and measurability (OECD, noise, which can be evaluated by estimat- 1993, OECD - Core Set of Indicators for ing for instance the number of people Environmental Performance Reviews - A living in the vicinity of the infrastructure synthesis report by the group on the state of or by defining noise sensitive zones. the environment, Environment Monographs No. 83, Paris). The chosen indicators are based on the level of planning (in this case the EU level), as well As stated in the report, these criteria, which as data availability, as selected indicators have will not all be met in practice, describe an to be used Europe-wide. For more local-level “ideal indicator” . Nevertheless, according assessments (like corridor or project assess- to these criteria, the assessment of indica- ments), other, more detailed (and data tors can highlight their characteristics, and constrained) indicators can be used, which show their usefulness for certain specific account for more specific characteristics of objectives. The same report specified the the immediate environment. In short, the issues within these criteria that should apply initial selection of spatial and ecological to each indicator. Concerning Policy Rel- indicators was based on the following criteria: evance and Utility for Users, each indicator should: • relevance in regard to the Community environmental objectives and targets; • provide a representative picture of environ- • applicability on EU scale; mental conditions, pressures on the • feasibility in terms of the available data. environment or society’s responses; • be simple, easy to interpret and able to A group of seven indicators was chosen, show trends over time; namely: •be responsive to changes in the environment and to related human activities; • vicinity to sites of nature conservation • provide a basis for international compari- interest; sons; • partitioning of land; • be either national in scope or applicable to • water crossings; regional environmental issues of national • population in surrounding of infrastruc- significance; ture; • have a threshold or reference value against • crossing of tranquil zones; which to compare, so that users are able • land take by land cover type; to assess the significance of the values • sensitive zones analysis associated with it. Evaluation and Conclusions 47

Summary of indicator characteristics Figure 11.1

Policy relevance and Analytical soundness Measurability utility for users

Vicinity to sites of Relates to defined Buffer-zones around Site boundaries lacking nature conservation environmental objectives network should depend interest and targets on habitat/species type, Databases do not for which information is cover all countries Identifies potential risks lacking as problematic links and Overlap of designated sites (“red-flag” approach) Allows (and requires) areas “sample surveys” at Easy to understand regional/local level Definition of sites depend on national legislation

Characteristics and location of new links missing

Partitioning of land Relates to defined environ- Based on several assumptions, Ignores land units less mental objectives and difficult to agree upon than 1000km2 targets (quality evaluation, thresholds) Infrastructure other than Easy to understand Most applicable at regional level TEN not considered (relatively detailed maps required) Difficult to interpret Characteristics and location of new links missing

Crossing of surface Relates to defined environmental Not only water proximity may Characteristics and location waters (density) objectives and targets lead to environmental impacts of new links, information on catchment areas and Easy to understand and interpret vulnerability of surface and groundwater is missing Too simplistic

Allows for “red-flag” approach

People in vicinity of Not directly related to defined Gives only an indication of the Characteristics and location of infrastructure environmental objectives and potential population affected, new links, traffic forecasts and targets but doesn’t apply to impact noise modelling missing. magnitude Easy to understand Dwellings, working places, special infrastructure (hospitals, Little added value, as it is an schools) not available “implied” indicator

Crossing of tranquil Not directly related to defined Characteristics and location zones environmental objectives and of new links missing targets

Easy to understand and interpret

More useful for issues other than noise (e.g. biodiversity)

Land take (by land Not directly related to defined Characteristics and location cover type) environmental objectives and of new links missing targets

Easy to understand

Little added value

Sensitive zones Not directly related to defined Data weighting not Characteristics and location mapping environmental objectives and introduced of new links missing targets Suited to raise awareness Easy to understand and at strategic level to interpret Requires wide expert Based on an extremely consultation simplified model

Useful as a general indicator 48 Spatial and Ecological Assessment of the TEN

Concerning Analytical soundness, each indica- policy relevance and utility for users, they are tor should : all considered as good or very good, except the crossing of surface waters and land take • be theoretically well founded in technical by land cover type, which is mainly due to and scientific terms; the “plain” characteristics of these direct • be based on international standards and indicators. international consensus about its validity; • lend itself to being linked to economic Concerning analytical soundness, it generally models, forecasting and information scores under the pointed relevance or utility, systems. which means that further research on their theoretical validity as well as technical and Concerning Measurability, each indicator scientific consensus building must be con- should: ducted. Although this is a broad conclusion over the general scope of the indicators, • be readily available or made available at a considering its importance for the SEA of reasonable cost/benefit ratio; the TEN, it points to further consultation • be adequately documented and of known between Commission services, Member quality; States, the EEA and the scientific community • be updated at regular intervals in accord- on the most feasible and appropriate meth- ance with reliable procedures. ods and indicators.

Figure 11.1 summarises the weaknesses and Some problems should be raised concerning strengths of the indicators identified in the the measurability criteria, which relates to present study and Figure 11.2 presents an data availability, quality and updating. Direct evaluation of the indicators according to the indicators such as the crossing of surface OECD criteria. The assessment was con- waters and land take, as well as the vicinity to ducted combining the results of an “experts sites of nature conservation were considered panel”, where each expert was asked to do as the most reliable, considering the data the ranking according to presented scores. driven issues. The importance of accurate The evaluation was conducted in conjunc- and up-to-date data on land cover must be tion with the presented criteria, indicator stressed, as it serves as baseline data for most concept, required environmental data and of the indicators used. information display. Lack of data on infrastructure was also taken into account, as it Considering biodiversity indicators, vicinity to seems more important for some indicators sites of nature conservation interest seems (population affected by noise) than for more meaningful and appropriate to be others (e.g. crossing of surface waters). carried out at the corridor level. A necessary indicator refinement would be the improve- Some conclusions can be derived from ment of the designated sites database and a assessing the indicators. In terms of their qualitative analysis establishing the relation

Figure 11.2 Assessment of used indicators according to OECD criteria

++ very good Policy relevance and Analytical soundness Measurability + good utility for users +/- fair - poor Vicinity to sites of nature + +/- + -- very poor conservation interest Partitioning of land + + -

Crossing of surface waters +/- +/- +

Population in surrounding + +/- +/- of infrastructure

Crossing of tranquil zones ++ +/- +/-

Land take (by land-cover type) +/- + +

Sensitive zones mapping ++ - - Evaluation and Conclusions 49

between sites/habitats/species and respec- Land resources indicators, like land take, is tive impact vulnerability, performed by straightforward, easy to understand, and specific “zoom analysis”. This work would be useful as a general indicator for raising best linked, at a first stage, to the corridor awareness at the strategic level. Sensitive analysis already under way, and through zone mapping seems to be very useful for consultation with the Member States. Special decision-makers at policy and strategic level. attention must be given to potential risk sites The specific characteristics of the method, identified by this exercise (“red-flag” analy- based on data combination through map sis), but also to other specific situations overlay, make the quality of data a significant which, because of an inconsistency in defini- issue. Again, the method could be signifi- tion, or information available (the case of cantly improved once estimates become nationally designated sites) have not been available on the predicted magnitude of considered. traffic impacts on the links (i.e. traffic volume, emissions of pollutants and noise). Concerning the partitioning of land indica- This would allow better accounting for the tors, in spite of being easy to understand as a expected environmental pressures, which concept, it is difficult to interpret. This is can differ amongst modes and amongst partly because it is based on many assump- links. tions, which are difficult to agree upon, particularly at a European level. It is best used at a regional level (where smaller units 11.2. Conclusions could be considered in the analysis), and it should be refined to consider infrastructure The importance of improving certain other than existing TEN. The impact evalua- databases has been clearly raised in the tion also requires information on the charac- previous chapters. In fact data shortage and teristics and location of new links. inconsistencies, as well as all the assumptions used to overcome them, limit the usefulness The Water resources indicator is simplistic and of the chosen indicators. Their measurability easy to understand, but its usefulness is quite and their soundness, in particular, cannot be limited. That is partly because environmental assured. impacts are not just a result of proximity between water surfaces and infrastructures . One of the main problems encountered in An analysis based on water catchment areas, the study is the poor information on the relating hydrology and quality of surface and planned TEN-infrastructure. Special atten- groundwater, lacked accurate data at an tion should be paid to collecting additional European level. Again, at a strategic level, data on the distinction between upgrade and this indicator can be useful for scoping more new constructions for the planned TEN links in-depth analyses at a regional or corridor and to the improvement of information on assessment. the location of the planned new infrastructure. Also estimates on the predicted magni- The indicators used for noise assessment are tude of traffic impacts on the links (i.e. an example of the need for more detailed traffic volume, emissions of pollutants and infrastructure data, as their relation to noise noise) would enable a better understanding is indirect and inferred. They do not con- of the expected environmental pressures, sider noise levels and their spatial distribu- which can differ amongst modes and tion (critical noise contours), which can only amongst links. be calculated based on traffic estimates and physical infrastructure characteristics (pave- Another priority action is the completion of ments). They are both relatively easy to the information on the NATURA 2000 understand, but their added value for noise network, as this would provide a more impact evaluation is very small. The first one consistent basis for the assessment of impacts is more a “social-indicator” as it considers the on habitats. Furthermore, this underlines potential number of affected people (by the importance of developing a SEA system noise but also by other effects such as “bar- for the TEN in an ‘iterative’ manner, since rier effect”). As a “social indicator,” it is this would allow the refining of the assess- important to note that detailed information ment techniques and scope as more accurate on location of special equipment and infra- data become available. structure (hospitals, schools, and residential areas) is missing. The second one seems, The models and parameters that were used also, to be more useful for issues other than in this study have been applied in a rather noise, e.g. for biodiversity and social impacts. straightforward manner, and using expert 50 Spatial and Ecological Assessment of the TEN

judgement which inevitably involves a types, species and their vulnerability to certain degree of subjectivity. Parameters impacts of infrastructure and traffic. At this and threshold values (such as the criteria scale, it is not feasible to evaluate the im- and indices used for identifying sensitive pacts on small valuable areas. It is therefore zones) have been entered in a pragmatic essential that the SEA of the TEN be comple- way, based on literatureexamples gathered mented by more local-level assessments, i.e. from various national, regional or local assessments of national/regional networks environmental assessments. The key assump- and of corridors, and, possibly, pilot project tions that have been made and their transfer- environmental impact assessments (EIA). ability to EU scale have to be improved by Using such a “tiered” assessment process, the broad and multi-disciplinary consultation of SEA of the TEN should be considered as a experts. In particular the EU-wide definition first broadbrush screening instrument for of sensitive zones, which inevitably relies to a identifying those corridors or series of certain extent on a subjective attribution of corridors that should be addressed as a sensitivity values, should be further re- priority. searched. During the April 1997 workshop on the In this study, it was also assumed that the Spatial and Ecological Assessment of the impact a transport infrastructure exerts when TEN, some issues were identified, which crossing a sensitive environment is similar have not yet been addressed in the present for all types of infrastructure (e.g. motor- study, as they require in-depth research, ways, rail or inland waterways). This may not including: necessarily be the case. Also, it was not possible to differentiate between various • development of indicators and methods types of motorways (e.g. number of lanes). for the spatial assessment of secondary In the future, attributing weights to various development around TEN-nodes; types of infrastructure could increase the • identification of ecological corridors and value of the assessment. This could be done the use of such corridors in infrastruc- on the basis of, for instance, infrastructure ture network and corridor assessments. characteristics (such as the number of lanes or tracks per link). Another solution would Finally, in order for the SEA to become an be to integrate the results of the prediction integrated decision-aiding tool within the of traffic-related impacts in the GIS (i.e. overall revision of the guidelines, it should traffic volume, emissions of pollutants and be conducted in parallel with the evaluation noise for each link). This would enable a of the socio-economic impacts of the TEN better comprehension of the various degrees and of its impact on cohesion and accessibil- of environmental pressures, and thus allow a ity. The studies that are carried out on all better weighting of the impacts according to those aspects should be carefully co- mode and to links. ordinated, and should be based on the same underlying assumptions and data sets. In order to ensure the compatibility of the outputs of both parts of the assessment, the The present methodological study can be RTD project that the Commission is cur- considered successful as it realised most of its rently launching on the assessment of global original objectives: impacts should therefore be closely co- ordinated with the current work on spatial • it provided a comprehensive compilation and ecological assessment. of a Union-wide spatially referenced database, holding information on the The study also made clear that an assessment state of the TEN, as well as of a selection at EU level is naturally limited in scope and of environmental and other data, which level of detail, due to data limitations. The is relevant for spatial and ecological methods that were tested allow the identifica- impact assessment; tion of potential conflicts with certain • the project initiated the development, environmental concerns (“red-flag” ap- testing and application of a series of proach), but do not permit a prediction of impact assessment methodologies such the magnitude of the actual impacts (in as proximity analysis, impact prediction terms of losses in species or damage to models and sensitivity mapping. These human health, for example). This would preliminary results will form a practical require extremely detailed data on species basis for refinements and further meth- and ecosystems and a qualitative investiga- odological developments; tion into the relationship between habitat • the provisional results are stored in a Evaluation and Conclusions 51

new database that can be made available 2. Filling of the major data gaps on the to all parties concerned in order to TEN and on the environment. launch the discussions and raise aware- 3. The carrying out of the full spatial and ness on the environmental impact of the ecological assessment of the TEN, TEN implementation; including the integration into the GIS of • the importance of this SEA of the TEN the results of the traffic-related impacts exercise in the screening for corridor (emissions, noise, safety) and a compre- assessment must also be highlighted. hensive multi-criteria analysis of the predicted impacts. For this, sufficient The study furthermore demonstrated the resources (in terms of budget and staff) technical feasibility, but also the limitations, have to be made available. of a Europe-wide SEA of the TEN, and re- 4. Initiation of longer term research on the affirms the usefulness of the multidiscipli- above issues, in order to gradually nary data collection at European scale. improve methods, data and tools for the ecological and spatial impact assessment The present exercise also pointed to issues of transport plans. needing further research and methodologi- 5. Perform some “zoom analyses” or “sam- cal improvements: ple surveys”, enabling the testing of the interrelation between different plan- 1. Optimisation of the indicators and ning/geographical levels and corre methods through a wide and multi- spondent assessment techniques (rela- disciplinary consultation of experts, tion between strategic assessment and stakeholders and Member States. In this corridor and project assessment). For respect, the following actions can be this purpose a follow-up and feedback of envisaged: the 5 corridor assessments currently • organisation of a technical workshop being carried out at national level with (involving the EEA, the Commission support from the Commission, should be services, external experts and NGOs), performed. which should allow consensus building 6. The selection of the alternatives, should on the methods and indicators. be a subject for further research, due to • presentation of the methods to the its strategic nature and assessment Member States in the next meeting of implications. the Ad Hoc Group on the Environmen- tal Evaluation of the TEN. 52 Spatial and Ecological Assessment of the TEN

List of Acronyms

CBD - United Nations Convention on Biological Diversity CCFE - Commission des Chemins de Fer Européens CoE - Council of Europe CORINE - Coordination of Information on the Environment DEM - Digital Elevation Model DGVII - European Comission Directorate-General for Transport DGIX - European Comission Directorate-General for Environment, Nuclear Safety and Civil Protection 5EAP - 5th Environmental Action Programme EEA - European Environment Agency ETC/LC - EEA European Topic Centre on Land Cover ETC/NC - EEA European Topic Centre on Nature Conservation GIS. - Geographical Information System GISCO - EUROSTAT Directorate on Social and Regional Statistics and Geographical Infor- mation Systems HSR - High Speed Rail NATURA 2000 Network - Process related to the Habitats and Birds Directives, on the conservation of european natural resources for the year 2000 and beyond NGO - Non Governmental Organization NUTS - Nomenclature of Territorial Units for Statistics OECD - Organisation for Economic Co-operation and Development RSPB - The Royal Society for the Protection of Birds SEA - Strategic Environmental Assessment SPA - Special Protected Areas under directive 79/409/EEC TEN - Trans-European Transport Network WERD - Western European Road Directorates