Agent-based modelling of crop management

This study aims to explore the benefits of integrating Agent-Based Models (ABMs) of farmer behaviour with biophysical models to describe and understand the complex agroecological systems that influence decision-making in arid and semi-arid regions. Through a mixed-methods approach combining surveys, interviews, and ABM, the research provides insights into the complex dynamics shaping farmer behaviour and evaluates the potential impacts of various management strategies on agricultural sustainability. Initial online surveys across diverse agro-climatic zones in Morocco revealed that farmer decisions are influenced by environmental pressures, crop characteristics, and water availability. Follow-up in-depth interviews in the Al Haouz Basin highlighted institutional barriers like land tenure insecurity and bureaucratic processes as key constraints to adopting sustainable practices. The study integrates empirical data with Structural Equation Modelling and the Theory of Planned Behaviour to parameterize an ABM. This coupled behavioural-biophysical simulation captures feedback loops between environmental conditions and human decisions. Model simulations revealed potential unintended consequences of policies aimed at increasing productivity, such as increased soil salinization and land abandonment resulting from expanded groundwater access. Key contributions include advancing the understanding of temporal adaptation dynamics in agricultural systems under climate change and developing a novel methodological framework integrating qualitative and quantitative approaches for studying complex socio- ecological systems. By bridging social and natural sciences, this research establishes a comprehensive framework for addressing agricultural sustainability challenges in water-scarce regions.