Smart farm models in sustainable agriculture
The provision of healthy food with minimal impact on the environment requires efficient use of natural resources while adapting to climate change. The EU’s common agricultural policy(opens in new window) must evolve to address environmental sustainability and climate action by understanding the patterns guiding farmers’ decision-making criteria.
Farm-level modelling
Fully aligned with the farm to fork strategy, the EU-funded MIND STEP(opens in new window) project set out to develop advanced tools to integrate farm-level data into policy analysis. The consortium introduced detailed bio-economic, farm-level mathematical programming and econometric optimisation models. These models leverage individual farm data from the EU farm accountancy data network(opens in new window), assigning cost components to specific agricultural activities. By combining this data with biophysical information, the project has enhanced the spatial allocation of representative farms and developed yield response curves for grassland using remote sensing and statistical data. To capture the socio-psychological aspects influencing farmers’ decisions, MIND STEP conducted surveys incorporating behavioural factors such as risk preferences. An innovative risk module was implemented in the FarmDyn model(opens in new window), allowing integrated assessments of farm management measures under uncertainty. “Our models now better reflect the diversity in farmers’ behaviours and responses to policy measures, providing more accurate predictions of policy impacts at the farm level,” states project leader John Helming.
Comprehensive policy analysis
Given the complexity of agricultural systems, MIND STEP combined bio-economic farm models with agent-based models, agricultural sector models and economy-wide models. This integration allows for a bottom-up approach, starting from individual farm decisions and scaling up to regional and national analyses. “By linking detailed farm data with broader economic models, we can assess how policy changes affect not just individual farms but the entire agricultural sector and economy,” explained Helming.
Agricultural policy recommendations
MIND STEP’s research has led to several important insights that could shape future agricultural policy design. One key recommendation is to incorporate farm-level decision-making models and farmer-specific variables such as age, income and education level into the modelling used by the EU. This helps identify policy measures that are effective on paper and that farmers are more likely to adopt in practice. Another important finding is that due to the inherent heterogeneity (or differences) among farms, particularly in terms of risk attitudes, efficiency levels, and marginal abatement costs, uniform command-and-control policies aimed at reducing emissions may not be cost-effective. Instead, market-based policies must be tailored to reflect this diversity to achieve better environmental and economic outcomes. The project also found that a gradual introduction of emissions taxes would be the most effective way to reduce mineral nitrogen use and greenhouse gas emissions. This gives farmers time to adjust, plan investments and adopt mitigation strategies. Furthermore, reinvesting the revenues from these taxes into mitigation technologies has the potential to reduce emissions across the sector. “Collectively, our recommendations emphasise the importance of integrating individual farm data and models into the design of effective and targeted agricultural policies,” highlights Helming. Building on the success of MIND STEP, the development of bio-economic farm-level models will continue in other EU Horizon projects, such as AgEnRes, BrightSpace, LAMASSUS and Act4CAP. A significant next step involves further developing a farm model framework using FarmDyn as the core.
Keywords
MIND STEP, agricultural policy, sustainable agriculture, smart farm, farm-level modelling
