Periodic Reporting for period 3 - BESTMAP (Behavioural, Ecological and Socio-economic Tools for Modelling Agricultural Policy)
Reporting period: 2022-09-01 to 2024-02-29
Around 40% of EU land is agricultural. Processes like land-use intensification and climate change threaten the ecosystem services provided by these agro-ecosystems. Policy makers at all levels face the challenge of enhancing agricultural sustainability whilst maintaining farmers’ livelihoods. Current policy impact assessment models (PIAMs) used by the Commission largely ignore the complexity of farmers’ decision making, and existing models focus on economics, mostly ignoring policy impacts on rural natural, social and cultural assets.
BESTMAP (Behavioural, Ecological & Socio-Economic Tools for Modelling Agricultural Policy) developed a new modelling framework utilising behavioural theory based individual-farm Agent-Based Models (ABM), linked to spatially explicit ecosystem services (ESS) models. These new modular and customizable tools allowed BESTMAP to quantitatively model, monitor and map policy change impacts on the environment, biodiversity, ecosystem services, and socio-economic conditions. BESTMAP provided a direction to improve and contribute to existing tools used by the Commission, national and regional decision-makers and expert personnel. To promote those goals, BESTMAP used a range of external communication and dissemination activities to build capacity for researchers, national and EU Directorate-General staff and parliamentarians, to model policy impacts and improve policy design and monitoring.
BESTMAP (Behavioural, Ecological & Socio-Economic Tools for Modelling Agricultural Policy) developed a new modelling framework utilising behavioural theory based individual-farm Agent-Based Models (ABM), linked to spatially explicit ecosystem services (ESS) models. These new modular and customizable tools allowed BESTMAP to quantitatively model, monitor and map policy change impacts on the environment, biodiversity, ecosystem services, and socio-economic conditions. BESTMAP provided a direction to improve and contribute to existing tools used by the Commission, national and regional decision-makers and expert personnel. To promote those goals, BESTMAP used a range of external communication and dissemination activities to build capacity for researchers, national and EU Directorate-General staff and parliamentarians, to model policy impacts and improve policy design and monitoring.
Work Package 1, ‘Project management’, focussed on monitoring the progress of the project and managing risks and mitigation. The data management plan was regularly updated and model codes/outputs and results from modelling exercises and analysis have been stored on Zenodo, GitLab (https://git.ufz.de/) GeoNetwork (https://geonetwork.ufz.de) and the project dashboard. Guidelines and protocols were created and developed to harmonise activities across case studies (CSs). WP1 also ensured exploitation activities were completed, including agreeing a Memorandum of Understanding for future exploitation of the expertise and modelling tools developed during the project.
Work Package 2, ‘Co-design and co-development’, was responsible for developing a co-design approach with stakeholders to identify policy scenarios relevant for the CSs to model its impacts, and 5 agri-environmental schemes and their application or non-application scenario were selected. The sessions were also key to co-design the interactive dashboard (https://www.ogc.grumets.cat/bestmap/) with future users, by performing a user requirement of functionalities that stakeholders would need. Additionally, an economic baseline scenario development using the CGE model DART-Bio model was done, which focussed on the synergies between the EU's Renewable Energy Directive (RED 2) policy, global biofuel quotas, and international climate policies under the Paris Agreement. The study revealed that implementing biofuel policies in non-EU regions leads to a global shift towards more cropland used for biofuel feedstock, affecting pastureland and crops not used for biofuel production. This task also highlights the fact that climate policies influence land-use changes differently based on whether they target CO2 emissions or all greenhouse gas emissions.
Work Package 3, ‘Farming System Archetypes’ (FSAs), was responsible for compiling all geospatial data required to create the models. A set of customizable, open-source and spatially-explicit biophysical models were developed and/or adapted to estimate the effects of selected Agri-Environment Practices (AEP) on ESS. The models were used to map ESS provision, biodiversity and socio-economic outputs under different AEP adoption scenarios in the 5 CSs of the project. WP3 also developed FSAs, i.e. a categorization of farms based on their specialisation and economic size, which were mapped for each CS and can be viewed in the dashboard.
Work Package 4, ‘Agent-based modelling and analysis’, focussed on development and implementation of ABMs in each CS, with the main structure of the ABMs being consistent across CS but with adaptations to local specificities due to differences in the availability of data and existing policies and regulations. The ABMs were applied to different agri-environmental scheme (AES) scenarios to investigate the effects of AES parameters on AES adoption rates as compared to the status-quo adoption rates. WP4 also examined how the adoption of AES drives trade-offs and synergies between ESS, biodiversity and socio-economic outputs, focusing on two different scenarios, with and without AES. In addition, WP4 reviewed policy indicators from different sources linked to agricultural practices and the most relevant indicators selected and analysed to determine the level of their association with BESTMAP model results. Finally, WP4 synthesised and compared the main findings of the regional CSs, which was used as a basis for BESTMAP policy briefs.
Work Package 5, ‘Upscaling’, was responsible for using the CS-level ESS results to predict average NUTS3-level per-farm ESS results across the other NUTS3 regions in the EU, and other European countries. The upscaling was undertaken using European-level environmental and economic predictors that were used in meta-models to predict Europe-wide results, which were determined by inter-CS region predictions. The European-wide ABM predicted the uptake of AES within the EU using a generalised linear model approach, which showed how the AES uptake varied across countries. WP5 was also responsible for trialling a pilot to investigate remote sensing methods of crop type mapping, crop yield mapping and field boundary mapping – the results in brief show the potential of remote sensing techniques in generating supplemental datasets for mapping FSA dimensions, but lacked precision for metrics, like crop yield or composition. A roadmap was produced for the expansion of BESTMAP towards an operational pan-European modelling platform, as well as exploring via pilot analyses several areas for improvement and future research.
Work Package 6, ‘Capacity building and dissemination’, was responsible for clustering with other projects by developing an engagement plan to establish a strategic channel of relationship-building processes with identified related parties, including joint participation in the AGRIMODELS cluster at different events. Project information and results were disseminated through a wide variety of outlets, and policy notes were also developed covering proposals for the new EU statistical regulation and the Common Agricultural Policy, along with case-specific scenarios.
Work Package 7, ‘Ethics’, ensured the ethics requirements of the project were delivered.
Work Package 2, ‘Co-design and co-development’, was responsible for developing a co-design approach with stakeholders to identify policy scenarios relevant for the CSs to model its impacts, and 5 agri-environmental schemes and their application or non-application scenario were selected. The sessions were also key to co-design the interactive dashboard (https://www.ogc.grumets.cat/bestmap/) with future users, by performing a user requirement of functionalities that stakeholders would need. Additionally, an economic baseline scenario development using the CGE model DART-Bio model was done, which focussed on the synergies between the EU's Renewable Energy Directive (RED 2) policy, global biofuel quotas, and international climate policies under the Paris Agreement. The study revealed that implementing biofuel policies in non-EU regions leads to a global shift towards more cropland used for biofuel feedstock, affecting pastureland and crops not used for biofuel production. This task also highlights the fact that climate policies influence land-use changes differently based on whether they target CO2 emissions or all greenhouse gas emissions.
Work Package 3, ‘Farming System Archetypes’ (FSAs), was responsible for compiling all geospatial data required to create the models. A set of customizable, open-source and spatially-explicit biophysical models were developed and/or adapted to estimate the effects of selected Agri-Environment Practices (AEP) on ESS. The models were used to map ESS provision, biodiversity and socio-economic outputs under different AEP adoption scenarios in the 5 CSs of the project. WP3 also developed FSAs, i.e. a categorization of farms based on their specialisation and economic size, which were mapped for each CS and can be viewed in the dashboard.
Work Package 4, ‘Agent-based modelling and analysis’, focussed on development and implementation of ABMs in each CS, with the main structure of the ABMs being consistent across CS but with adaptations to local specificities due to differences in the availability of data and existing policies and regulations. The ABMs were applied to different agri-environmental scheme (AES) scenarios to investigate the effects of AES parameters on AES adoption rates as compared to the status-quo adoption rates. WP4 also examined how the adoption of AES drives trade-offs and synergies between ESS, biodiversity and socio-economic outputs, focusing on two different scenarios, with and without AES. In addition, WP4 reviewed policy indicators from different sources linked to agricultural practices and the most relevant indicators selected and analysed to determine the level of their association with BESTMAP model results. Finally, WP4 synthesised and compared the main findings of the regional CSs, which was used as a basis for BESTMAP policy briefs.
Work Package 5, ‘Upscaling’, was responsible for using the CS-level ESS results to predict average NUTS3-level per-farm ESS results across the other NUTS3 regions in the EU, and other European countries. The upscaling was undertaken using European-level environmental and economic predictors that were used in meta-models to predict Europe-wide results, which were determined by inter-CS region predictions. The European-wide ABM predicted the uptake of AES within the EU using a generalised linear model approach, which showed how the AES uptake varied across countries. WP5 was also responsible for trialling a pilot to investigate remote sensing methods of crop type mapping, crop yield mapping and field boundary mapping – the results in brief show the potential of remote sensing techniques in generating supplemental datasets for mapping FSA dimensions, but lacked precision for metrics, like crop yield or composition. A roadmap was produced for the expansion of BESTMAP towards an operational pan-European modelling platform, as well as exploring via pilot analyses several areas for improvement and future research.
Work Package 6, ‘Capacity building and dissemination’, was responsible for clustering with other projects by developing an engagement plan to establish a strategic channel of relationship-building processes with identified related parties, including joint participation in the AGRIMODELS cluster at different events. Project information and results were disseminated through a wide variety of outlets, and policy notes were also developed covering proposals for the new EU statistical regulation and the Common Agricultural Policy, along with case-specific scenarios.
Work Package 7, ‘Ethics’, ensured the ethics requirements of the project were delivered.
ESS, biodiversity and socio-economic models for each case study were conducted looking at two scenarios, current status quo and zero adoption of agri-environmental schemes. Their results contributed to the dashboard and formed the basis of policy recommendations created and presented in Brussels at the final dissemination event.
An online survey, in the form of a discrete choice experiment was conducted in Spring 2022 in all 5 case studies. Analysis of the data took place and the results formed an input to the ABM.
Co-design sessions were held with stakeholders comprising policy makers at national level in all 5 case studies. Their feedback informed the development of the dashboard.
An online survey, in the form of a discrete choice experiment was conducted in Spring 2022 in all 5 case studies. Analysis of the data took place and the results formed an input to the ABM.
Co-design sessions were held with stakeholders comprising policy makers at national level in all 5 case studies. Their feedback informed the development of the dashboard.