Authors: John Helming (MIND STEP project coordinator) and Diego Rodriguez (Geonardo)
Editor: Blanca Casares (AEIDL)
The EU project MIND STEP, scheduled for completion in December this year, aims to support public decision-making in agricultural, rural, environmental, and climate policies related to agriculture such as the Common Agricultural Policy (CAP) of the EU.
The past four years MIND STEP has developed tools and models at different scales to do exactly this: monitor and assess policies related to agriculture. In describing and modelling the complexities and ever-evolving landscape of agriculture, the MIND STEP project has been a beacon of innovation and progress.
Let’s explore the transformative capabilities of the MIND STEP toolbox, which develops and integrates various tools and includes a wide range of sustainability indicators (environmental and socio-economic).
What can we do now thanks to the toolbox that we could not do before?
1. Better representation of the diversity of farms heterogeneity in modeling
The MIND STEP project has achieved groundbreaking milestones in modelling behaviour of individual decision makers in the agricultural sectors. MIND STEP has developed detailed bio-economic, farm level, mathematical programming and econometric optimisation and simulation models. These models are firstly based on individual farm data from the EU Farm Accountancy Data Network (FADN). Econometric activity-based cost accounting tools have been developed to assign different cost components in the EU FADN from the farm to the agricultural activity level. In a second stage, MIND STEP has developed tools to combine farm level data (FADN) with biophysical data. Using biophysical data, MIND STEP developed probabilities regarding spatial allocation of representative farms in the EU FADN. In addition, MIND STEP developed grassland yield response curves, combining remote sensing data (regarding number of cuts of grassland at parcel level), statistical information from census data at farm level and data from agronomic literature. Additional surveys have been conducted to combine statistical data with socio-psychological data helping to understand individual farmers preferences, behavior and adoption of risks.
The specific deliverable (D3.2) ‘An overarching IDM model structure Interfaces within the MIND STEP model toolbox’ has been published, describing the realisation within MIND STEP of a modular approach to model integration. Deliverable (D3.1) presents the ‘Specification of model requirements: Protocols for code and data’. In this respect, MIND STEP concludes that:
- Modular design of farm models increases complexity.
- There are no standardised test cases to test model behaviour.
- Training of users and developers is required. It is core to establish a network of model developers and users for the continued development of a generic and modular overarching farm level model structure.
2. Interactions between farms
MIND STEP developed innovative models and tools focusing on interaction between individual farms:
- The estimation of farm exit rates is intended to be the basis for integrating farm structural change into representative farm-level or equilibrium models. Innovative approaches include the use of neighbourhood effects in farm exit estimates.
- The new models and approaches include factors beyond economic incentives and combining agent-based modelling (ABM) and behavioral experiments.
- The estimate capturing market power along the supply chain and farmers’ power arising from contractual arrangements or the formation of producers.
- The training of machine-learning-based surrogate models. This allows efficient and consistent integration of detailed farm models in agent-based model capturing structural change implications of policies for the farm population.
3. Improved interfaces between data and models at different scales (farm, regional, national, EU)
MIND STEP made important contributions to improved micro-economic underpinnings of models at various scales, frequently used by the European Commission for assessments of policies related to agriculture: the individual farm model IFM-CAP, the world-wide bio-economic, agricultural sector model GLOBIOM and the world wide and economy-wide model MAGNET.
4. Transparency of methods, sustainable software development and model validation
As a roadmap, MIND STEP gained important experiences in the field of a) validation and proof of concept, b) importance of stakeholder workshops and c) policy evaluation.
To conclude, MIND STEP toolbox, enables better policy evaluation, identification of policy options with special emphasis on the CAP, scenario development and assessment of their impact on European agricultural production systems.