The agri-environmental sciences offer endless possibilities for the application of Interpretable Machine Learning methods.These methods can aid in evaluating mechanistic assumptions or in revealing unforeseen connections between variables. Covering different spatial and temporal scales and linking between them, scientists have proposed and used a wide variety of IML models, each with their unique demands regarding data, statistical analysis and interpretation of results.
To gain an appreciation of this range, this session invites contributions involving interpretable Machine Learning and xAI applications in the agri-environmental and forestry sciences. The applied models may cover different a-priori assumptions, data requirements, spatial and temporal scales, and methods of statistical analysis. The rationale of their choice should be communicated clearly to give (early career) scientists methodological inspiration for their own pursuits.