This project was aimed at adapting decision-theoretic tools to agri-environmental management. The process of decision-making within an agricultural context is complex because it spans multiple interdependent stages and involves many risks along the way. Decisions such as when to apply pesticides, how much to apply, when to prune, when to water, and even when to harvest can affect the quantity and quality of produce, and hence the short-term success of the enterprise.
Decisions will also determine the extent of environmental harm, which has been challenging to define, as is the value of ‘services’ provided by the ecosystem. To facilitate their inclusion in decision-making, the JDec project team developed models that are more flexible and holistic than common frameworks in operational research and that emphasise the inclusion of uncertainties and inter-dependencies.
Joint decision models
There are three key aspects underpinning the models developed by the project team. First, outcomes need to be valued by utility functions that comprehensively reflect costs and benefits. It is vital that these are evaluated along decision trajectories, including appropriate levels of memory and foresight, and interdependencies. For example, a herbicide treatment may look effective only as long as its indirect effect is ignored on the wild pollinators that had visited the weeds, and whose loss will need to be compensated with new costs.
Second, it is important to recognise that in an agricultural-environment context, decisions are not taken by humans alone. A modelling approach looking at decisions being taken jointly by the farmer, the crop, and the environment provides the flexibility needed to deal with interactions. We further allow for a higher level of uncertainty, given that the influence each of these agents has may itself depend on random events.
Third, the project team’s models acknowledge the temporal dimension and potential resource allocation constraints. In a large, interconnected, multi-stage system of land and resource management, past actions influence future decisions.
If we also consider the rapidly changing environment, with extreme weather events increasing in frequency, shifting pest and pollinator ranges, and resource depletion, we need to take account of the need for robust approximate solutions in model development. In other words, the challenge of having to make decisions in the ‘real world in real time’ requires a paradigm for ‘good enough’ decision-making, and a conceptualisation of the gap between these decisions and optimal solutions.
The aim of the project was ultimately to build a mathematical and statistical framework for decision modelling that covers the three aspects outlined above. Two case studies were used to achieve this.
The first was a system of wild pollinators in apple orchards, a particularly suitable testing ground for understanding indirect effects at the frontier between managed land and its surrounding landscape. The second case study explored the use of decision modelling in a large farm scale experiment with four crops and multiple intervention methods. This provided a rich data set for comparing decision strategies. The work undertaken during the project has the potential to directly benefit many citizens. This is not only true of crop scientists and land managers, but also ecologists, conservationists, local authorities, charities, and policymakers. The project team produced a number of models and applications that can be accessed via the UKRI Gateway to Research and which might inspire further work on decision models for agri-environmental settings.
Julia Alexandra Brettschneider (Principal Investigator)
Rosemary Collier (Co-Investigator)
1. Data management challenges for artificial intelligence in plant and agricultural research, F1000Research, April 2021.
|Link to the project page on the UKRI Gateway to Research: JDec: Joint decision models for citizens, crops, and environment|