Modelling flood, droughts and humans: A systematic review of hydrological hazard management in agent-based models

Agent-based modelling (ABM) is becoming a widely explored method to investigate human-water systems given their ability to represent heterogeneous actors and their decisions. ABM is also a bottom-up approach that can simulate how individuals interact and co-adapt with the environment; this is beneficial for understanding the effects of humans' decisions when facing the risk of hazards and climate change. However, individuals also adapt to organisational measures or lack thereof. This, for example, has been shown with the safe-development paradox, where governments’ hazard management strategies can impact individual risk preparedness. Therefore, ABMs can assist as tools for testing policies for improving flood and drought management.

The implementation of hydrological hazard management in ABMs has not yet been systematically evaluated. In this work, we aim to synthesize current knowledge on how hydrological hazard management and non-individuals are implemented in ABMs by performing a systematic review using the ROSES Protocol. A total of 377 unique articles were screened, and 78 articles were included in a full-text analysis. Our findings show that hydrological hazard management strategies in ABMs vary; both structural measures, such as levee and reservoir construction, and non-structural measures, such as water quota and insurance strategies, are implemented. Yet, there is a focus on individual agents taking measures against hazards in ABMs. Non-individual hazard management is often included as static scenarios or agents with ad-hoc or rational decision-making.

Our study demonstrates that the simplicity of hazard management in these models could limit the ability of the ABMs as a policy tool since the implemented hazard management does not adapt to the dynamics of human-water systems. Involving stakeholders or implementing bounded-rational decision-making could be an important shift to further improve the explanatory power of ABMs for challenges in hydrological hazard management.