Small-scale adaptation reduces regional drought risk: an agent-based modelling study of farmers’ adaptive behaviour in the Netherlands

Increasing drought frequency and severity pose growing risks to agricultural water systems worldwide. Farmers can reduce drought impacts through on-farm adaptation measures, such as limiting drainage or enhancing groundwater retention, yet the feedbacks between adaptive behaviour and groundwater dynamics remain largely understudied. Therefore, we developed an agent-based model to assess the influence of individual adaptive behaviour of farmers on the local and regional groundwater. The model links farmer decision-making, based on the protection motivation theory, to the hydrological environment of the eastern Netherlands, simulated with the WALRUS hydrological model. We developed eleven simulation scenarios with the current climate and two future climate conditions. Model validation showed that the model realistically reproduces observed groundwater levels and adaptive behaviour in the study area. Our findings show that the adaptation measures can reduce the drought damage for farmers with 65% during moderate droughts, but only 13% during extreme droughts. Moreover, farmers who installed adaptation measures also experience slightly more damage from wet periods (0.27 d per year instead of 0.03), leading to higher risk for waterlogging. The impact of the adaptation measures reaches beyond the farm scale, resulting in less drought damage for both adapting and non-adapting farmers. Finally, in all scenarios the drought risk remained substantial, even with maximum adaptation rates. We conclude that small-scale adaptation lowers drought risk for farmers, but adapting to more severe and prolonged droughts requires more transformative change. Strengthening collaboration among farmers and implementing measures at the regional scale can contribute to build climate robust landscapes.