Mapping and assessing ecosystem services in the EU - Lessons learned from the ESMERALDA approach of integration

The European Union (EU) Horizon 2020 Coordination and Support Action ESMERALDA aimed at developing guidance and a flexible methodology for Mapping and Assessment of Ecosystems and their Services (MAES) to support the EU member states in the implementation of the EU Biodiversity Strategy’s Target 2 Action 5. ESMERALDA’s key tasks included network creation, stakeholder engagement, enhancing ecosystem services mapping and assessment methods across various spatial scales and value domains, work in case studies and support of EU member states in MAES implementation. Thus ESMERALDA aimed at integrating various project outcomes around four major strands: i) Networking, ii) Policy, iii) Research and iv) Application. The objective was to provide guidance for integrated ecosystem service mapping and assessment that can be used for sustainable decision-making in policy, business, society, practice and science at EU, national and regional levels. This article presents the overall ESMERALDA approach of integrating the above-mentioned project components and outcomes and provides an overview of how the enhanced methods were applied and how they can be used to support MAES implementation in the EU member states. Experiences with implementing such a large pan-European Coordination and Support Action in the context of EU policy are discussed and recommendations for future actions are given.


Background
The ESMERALDA*1 (Enhancing ecosystem services mapping for policy and decisionmaking) project has assembled a team of leading researchers and practitioners from ecosystem services (ES) science and application in a Coordination and Support Action (CSA) funded for 42 months from 2015-2018 under the European Union's (EU) Horizon 2020 Research and Innovation Programme. This CSA has been dedicated to enhancing mapping and assessment of ecosystems and their services in Europe in the context of the European Union's (EU) Biodiversity Strategy*2 to 2020. The Biodiversity Strategy foresees in its Action 5 that all EU Member States shall: 1) Map and assess the state of ecosystems and their services in their national territories by 2014 (European Commission 2011); 2) Assess the economic value of such services; and 3) Promote the integration of these values into accounting and reporting systems at EU and national level by 2020.
The consortium represented a deep and shared commitment amongst all participants (researchers, decision-makers and various other stakeholders) from all 28 EU member states, Switzerland, Norway, Israel as well as the EU outermost regions, to contribute to the effective accomplishment of the challenges of the Biodiversity Strategy and to address the underpinning, cutting edge research issues in this important and timely topic area. Thus, appropriate methods, information and data are needed on where and how, for example, food, water, clean air, other materials and recreation are provided and how climate, nutrients, natural disasters, pests and diseases are regulated (Bagstad et al. 2013, Harrison et al. 2018Burkhard and Maes 2017. Information and data on actual ES demands, beneficiaries and potential mismatches with their supply location as well as ES quality and quantity, are mandatory for appropriate management of natural resources (Bagstad et al. 2013). The protection of biodiversity, habitats and ecosystem functionality are keys for the sustainable supply of goods and services to human societies. Therefore, mapping and assessment of ecosystems and their services are core to the Biodiversity Strategy; they are essential if we are to take informed decisions (European Commission 2011).
Action 5 sets the requirement for an EU-wide knowledge-base, designed to be a primary resource for developing Europe's green infrastructure. This is to identify areas for ecosystem restoration and to set a baseline against which the goal of 'no net loss of biodiversity and ES' can be evaluated. A dedicated working group on Mapping and Assessment of Ecosystems and their Services (MAES*3) has been established by the European Commission and meets twice per year in order to inform and update each other on progress and new developments within EU member states. ESMERALDA has been closely cooperating with the MAES Working Group and provided regular updates on its progress during the whole lifetime of the project. One finding of the MAES process was that ES mapping has already been taking place in the majority of the EU member states, but that it is not uniformly developed. Thus, EU-based guidance (e.g. via a common approach) was needed (Maes et al. 2012). Assistance was also required in order to harness the broad range of ES mapping and assessment approaches currently available or under development . Transdisciplinary assessments have to integrate the state and functionality of ecosystems and their biodiversity as the basis for understanding the supply of ES, on the one hand and, on the other hand, for identifying the socio-economic system components and ES-related supply and demand patterns (Burkhard et al. 2012).

ESMERALDA objectives
ESMERALDA has worked along four key project strands: (i) Networking, (ii) Policy, (iii) Research and (iv) Application in order to help ensure the delivery of all EU member states of Action 5. A 'flexible MAES methodology' was developed that can simultaneously provide innovative building blocks for pan-European, national and regional MAES studies as well as for local assessments required, for instance, for spatial planning, agriculture, land degradation, climate, water and nature policy. A key part of ESMERALDA was the mobilisation of relevant actors from science, policy, practice and society involved in ES science and application and to enable them to fulfil their commitments under Action 5. Fig.  1 provides an overview of the ESMERALDA strands and key objectives, which are all interlinked with each other.
ESMERALDA's main objective was to use the expertise of its pan-European consortium members and their networks to build on and mobilise relevant ES mapping and assessment actors of on-going European, international and national initiatives and to exploit existing projects, networks, methods and data. This has provided opportunities for improving ES mapping and assessment methodologies and knowledge sharing through the wider stakeholder interactions supported by the project. In order to achieve the aims of the ESMERALDA CSA, the key objectives of ESMERALDA were: Mapping and assessing ecosystem services in the EU -Lessons learned from ... In the following, an overview of the ESMERALDA Coordination and Support Action's implementation and its achievements will be given.

ESMERALDA implementation
The work of ESMERALDA was organised along the four strands of Networking, Policy, Research and Application, which are described in more detail below. ESMERALDA started with identifying gaps and solutions for MAES implementation in the EU member states by intensive networking and stakeholder involvement, transmitting experiences through active processes of dialogue and co-creation of knowledge. The aim was to empower the participants and relevant stakeholders to achieve the aims of Action 5 by 2020. Based on relevant and completed or ongoing activities in Europe and worldwide (such as the EU MAES*3 Working Group, EU integrated projects such as OpenNESS*4, OPERAs*5, MESEU, BEST*6 and other related initiatives such as MA*7 or TEEB*8; see Maes et al. 2012) and national ecosystem studies (such as TEEB-DE*9; UK NEA*10; Spanish NEA*21), methods for mapping and assessing ecosystems and their services have been enhanced. ESMERALDA has been developing a flexible mapping approach that integrates biophysical, socio-cultural and economic valuation techniques.  (Vačkář et al. 2018, Lotan et al. 2018) scale case studies and addressed diverse themes from policy-and decision-making, businesses and citizens . While the applications and testings took place, insights and results were already being integrated into European and global activities (see timeline in Fig. 2).

Strand Networking
The Strand Networking went on through the whole project duration as represented by the increasing number of participating EU member states over the project lifetime. At its start, the consortium consisted of 25 partners from 19 EU member states and Switzerland. At the end of the project, ESMERALDA included 37 partners from all 28 EU member states, Switzerland, Norway and Israel (see Fig. 3and Suppl. material 1). This enlargement of ESMERALDA has enormously facilitated a pan-European assessment, testing, mapping and implementation phase towards the objectives of Action 5 of the EU Biodiversity Strategy. Timeline showing the integration of ESMERALDA into related EU and global activities.
Mapping and assessing ecosystem services in the EU -Lessons learned from ...
The Networking Strand has been especially active in the first phase of ESMERALDA, in which relevant stakeholders from science, policy and practice were identified and the current state of MAES implementation was systematically assessed in each EU member state. This stocktaking resulted in a clustering of all 28 EU member states according to their prerequisites and needs to perform ES mapping and assessment in the year 2015. 'Front-runner countries' could be identfied as well as some member states which had not yet started to implement MAES. A 'MAES barometer' has been developed and presented at MAES Working Group meetings. The barometer indicates each EU member state's stage of MAES implementation. Furthermore, an overview of various issues for MAES implementation were identified. Key gaps included: 1. Lack of engagement of national authorities in the MAES process, 2.
Unsatisfactory stakeholder involvement in the MAES process and 3.
Gaps in capacity and resources to carry out MAES in the country.
Based on ESMERALDA works, the following recommendations to overcome these gaps were identified: A follow-up stocktaking in late 2017 revealed that many EU member states had made substantial progress in the implementation of MAES and that all EU member states were at least engaged in it. Capacity building in terms of knowledge sharing, ES mapping and assessment methods and data availability still seemed to be major gaps in many countries.
The points from the initial and the follow-up stocktaking were taken up in the ESMERALDA workshops, which were a key element of the project. ESMERALDA workshops have taken place two to three times per year and in different EU member states across Europe. The workshops have successfully been linking the project consortium members with relevant stakeholders and specific topics of ES mapping and assessment in Europe. The workshops organised by ESMERALDA have considered different European regions and biomes, thematic questions from policy, businesses and citizens and were related to various case studies, as well as policy themes so as to increase the potential impact of the project on mainstreaming of ES in policy-and decision-making (see Suppl. material 2 for an overview of the ESMERALDA workshops). The organisation of the altogether ten workshops and the final project conference during the project lifetime has been logistically challenging and resource-intensive. The results of a strong network and an efficient results creation have, however, proved worth the efforts, because most of the outcomes could actually be directly achieved at the workshops.

Strand Policy
The interest from the policy-side in research and practice on ES mapping is high as ES are a very politically relevant topic. Ecosystem assessments usually start with a set of questions from policy, society, business or science . ES mapping and assessment methods, selected to address these questions, need to be flexible enough to be applied in all EU member states (including outermost regions, marine areas and specific biomes) and related to various questions from policy, business and society in the context of the EU Biodiversity Strategy, as well as of the national Biodiversity Strategies, where present.
A first set of MAES-relevant policy questions was published in Maes et al. 2013. A second survey of policy questions was organised during the 13th Meeting of the MAES Working Group in 2017 and amongst ESMERALDA project partners, resulting in 82 policy questions  This classification was then used to link ES mapping and assessment to policy questions and to reveal which question can (or cannot) be answered by a combination or an integration of available methods  Knowledge requests: are questions asking for a conceptual clarification, describing information needs, usually at the start of a policy cycle. The reports by the MAES working group (available for download from the MAES website*3) are essential sources of information to trigger and support ecosystem assessments in Europe.
Policy support questions: are questions framing the use of ES as a concept to support a particular policy objective. These can include policies that have a positive or a negative impact on ES or are regulating the use of natural resources including agricultural policy, climate policy, biodiversity policy, spatial planning, impact assessment, disaster risk reduction and economic policy.
Questions about resources and responsibilities relate to ES governance: these questions ask, what could organisationally be done or which kind of institutional setting is favourable to implement an ES-based approach. Questions about human capacities and financial resources needed to carry out ecosystem assessments (or to ensure that ecosystems and their services are integrated into decision-making) are important for consideration during the assessment.
Application of ES mapping questions: are 'how to' questions focusing on the implementation of approaches and how to use ES mapping and assessment outputs to support policy implementation.
Technical and methodological questions: are questions asking for specific technical details of ES mapping; commonly addressed issues are spatial scale, uncertainty, the appropriate use of certain methodologies, priority setting and preferences.

Strand Research
The Strand Research of ESMERALDA has been clearly orientated on ES mapping and assessment methods and aimed at identifying, reviewing, enhancing, integrating and interlinking existing methods. The identified methods include biophysical, socio-cultural and economic ES quantification, valuation and mapping techniques . Such methods are commonly applied to improve the understanding of ES supply and demand patterns over space and time . At the end of ESMERALDA, 49 groups of methods were identified, comprising 15 categories of biophysical, 22 of economic and 10 of socio-cultural methods. A multi-tiered approach for ES mapping and assessment was developed. The approach considers different methods (biophysical, socio-cultural and economic) at different levels of detail and complexity and can be applied according to specific needs, data and resources availability (Weibel et al. 2018).
To gain a comprehensive and systematic overview of existing methods, a methods database was created. This database contains information on existing ES mapping and assessment methods used by consortium partners in case studies and populated by a comprehensive pan-European literature review (

ESMERALDA achievements
Experience gained during the work in the four ESMERALDA Strands was used to establish a functioning stakeholder network (including stakeholder support groups in each country consisting of representatives from science, policy and practice), to improve ES mapping and assessment, to identify and improve related methods and to apply and test methods related to distinct questions from policy-and decision-making.
One of the major outcomes of the project was the easy-to-access online open access interface ESMERALDA MAES Explorer*12, which allows people to browse through the diverse end products. Amongst those products are project Deliverable reports, scientific open access publications, a text book on mapping ecosystem serivces , a glossary for ES mapping and assessment terminology ) and country and case study fact sheets illustrating MAES implementation and applications. Additionally, ES methods application cards focusing on the selection of appropriate methods for the mapping and assessment of one or more ES to answer underlying questions were provided.
The ESMERALDA MAES Explorer also includes the ESMERALDA MAES Methods Explorer, allowing users to browse the ESMERALDA methods database and to collect further information on methods. The interface will help users of ESMERALDA products to find what they need for MAES implementation in their country, region, area or case study.
Included guidance documents deliver detailed descriptions of MAES implementation in seven-steps. The guidance starts with relevant questions to be answered and the identification of stakeholders. It continues with network creation and the activation of the relevant stakeholders. The next steps are related to the ES mapping and assessment processes per se, their related methods, their application in case studies and further background information for ES mapping and assessment. An appropriate and userorientated dissemination and communication of (often complex) scientific findings is key for successful implementation in decision-making -the last two steps of the ESMERALDA seven-step MAES/Action 5 implementation plan. Flexibility in ES mapping and assessment methods was achieved in ESMERALDA by analysing existing methods in a systematic review, testing them in various real-world case studies and by interlinking biophysical, socio-cultural and economic methods in transdisciplinary integrated ecosystem assessments. The design of a tiered mapping and assessment approach from simple (Tier 1) to complex (Tier 3) methods (Weibel et al. 2018) is helping to harness the variety of methods. The approach is combining expert-and land cover-based approaches (Burkhard et al. 2012, Campagne andRoche 2018) and the use of existing ES indicator data, with more complex and comprehensive ES modelling frameworks (see . Depending on data and resources available, the most suitable approach can then be chosen. The interest of the additional European countries to join the consortium, although only very little financial resources could be provided, has proved ESMERALDA's recognition and acceptance in Europe and can be seen as a great success of the Action's implementation. ESMERALDA is certainly a very rare case of an EU project including partners from all 28 EU member states, Switzerland, Norway and Israel. Moreover, contacts with the EU outermost regions and overseas countries and territories as well as to EU enlargement countries (non-EU countries in Europe which are intending to join the EU in the future) have been established. The second survey amongst EU member states (see 'Strand Networking' above) revealed that ESMERALDA has helped EU member states to: • Bring legitimacy for MAES from the outside; outside experts promoting a European agenda might seem to be more convincing to national authorities; • Approach policy-makers, authorities and governments in order to convince them of linkages of ES and other themes such as Natural Poland. ESMERALDA has been successful in mobilising and increasing scientific capacity in these countries (e.g. . It has enabled reseachers and policy-makers to extend their national networks to regional and European scales. In turn, there is evidence that this has increased the capacity of those countries to lead and coordinate their research on ecosystems and their services at regional levels. ESMERALDA supported the creation of new European collaborations and the establishment of new national e.g. Poland, Greece (Dimopoulos et al. 2017) and regional (e.g. South-eastern Europe) ESP chapters and networks. Furthermore, new cooperations and projects with the EU outermost regions and overseas countries and territories could be established. The first ESMERALDA spin-off EU project MOVE*20 (Facilitating MAES to support regional policy in Overseas Europe: mobilising stakeholders and pooling resources) started in April 2018. Several ESMERALDA partners are involved in the new CSA project MAIA (Mapping and Assessment for Integrated ecosystem Accounting), replying to the Call on "Valuing nature: developing and implementing natural capital and ecosystem accounts in EU Member States and Associated Countries", which marks the next step in the MAES process. MAIA will start in autumn 2018 and can build directly on ESMERALDA outcomes. These initiatives help, with support of the European Commission, the ESP, common ES-related platforms such as OPPLA and BISE and the EU member states, to maintain ESMERALDA products after the project finishes and to make them available for future activities.

Conclusions and recommendations
The large pan-European ESMERALDA consortium created an extensive and detailed cross-disciplinary knowledge based on ES mapping and assessment, its application in the context of MAES in EU countries and has completed a comprehensive European case study and methods review. The case studies proved to be a very constructive and successful way to engage stakeholders from the member states, to test methods and to deliver relevant outcomes. Contacts with relevant projects, initiatives, knowledge-bases and stakeholders across Europe have been mandatory to achieve the ESMERALDA objectives. The regular ESMERALDA workshops in different European regions were used to bring together scientists and stakeholders from EU member states in order to discuss and develop knowledge, methods and cooperation. The integration of stakeholders into project activities, especially in the case studies, the regular project workshops and the project mid-term and final conferences were one key to the success of ESMERALDA.
Workshops and face-to-face meetings proved to be an excellent means for creating and maintaining a strong and functioning network. Intensive knowledge exchange and capacity building were able to occur during the ESMERALDA workshops. However, the organisation of all thematic workshops, the mid-term and the final project conferences have been logistically challenging and resource-intensive. Therefore, various project partners were selected as hosts for the workshops. Nevertheless, the workload on the ESMERALDA Executive Board (EB) members and work package leaders has been comparatively high, considering alone the 27 EB meetings until the final project conference in June 2018 was reached. Based on the available funding for CSAs and the large size of the consortium with, in the end, 37 partners, the budget available for personal costs has, in most cases, not been in accordance with the workload. ESMERALDA could harness its pan-European network and build on existing ES case studies provided by the consortium partners. Against this background, ESMERALDA, as a CSA, profited substantially from contributions made by the project partners and their institutions/existing networks to the Action which have not directly been financed by the project. The same goes for the eleven new ESMERALDA project partners who entered the project in this phase and who were willing to contribute to ESMERALDA, based on provided travel budget only. This dedication to an Action is exceptional and proves the strong interest of EU member state representatives in ESMERALDA. This can, however, not be taken for granted for all activities related to the implementation of the Biodiversity Strategy and the Horizon 2020 funding programme should be adapted in order to provide sufficient resources for pending tasks.
Open access to all ESMERALDA products and the early publication of the open access textbook on 'Mapping Ecosystem Services'  with the inputs of many ESMERALDA partners, providing guidance and background on how to map ES, have been additional keys for success of this CSA. The ESMERALDA dissemination team has been very active in promoting and distributing the project outcomes by different means of communication and by using relevant information channels. Promotional materials (ESMERALDA flyers, posters, stickers, folders, cookies, chocolate (Fig. 4)) have been produced at the early stage of the project.  ESMERALDA Logo, bringing together Europe in a generalised map.