Municipal Decision-Making Processes and Learning From Flood Simulation Exercises Using Climate Projections: Insights From Two Municipalities in the Tokachi River Basin, Hokkaido, Japan

Climate change is increasing the frequency and intensity of extreme rainfall events, highlighting the limitations of structural flood control measures and the growing importance of evacuation and emergency response. Municipalities play a central role in issuing evacuation information, yet research on their pre-flood decision-making processes and learning remains limited. This study demonstrates that scenario-based simulation exercises grounded in climate projections can enhance municipal preparedness and adaptive evacuation decisions under climate change, providing empirically grounded evidence based on the experiential learning framework. To this end, we conducted blind-type exercises in two municipalities in the Tokachi River Basin, Hokkaido, Japan, using climate projection data from the d4PDF (Database for Policy Decision Making for Future Climate Change) to design unprecedented flood scenarios. Decision-making processes were systematically recorded and reviewed through post-exercise semi-structured interviews, and analyzed using Kolb’s experiential learning model, focusing on concrete experience, reflective observation, and abstract conceptualization. The results indicated that municipal decision-making followed a sequence of interrelated processes, beginning with precautionary monitoring of typhoon tracks and meteorological warnings, and extending to staff deployment, shelter preparation, and the issuance of evacuation information. While river water levels were decisive for issuing evacuation orders, early precautionary actions based on meteorological forecasts were essential, especially to avoid nighttime evacuations. These findings suggest that simulation exercises for sequential decision-making against unprecedented flood disasters should incorporate comprehensive scenarios covering both meteorological conditions and river levels from the early pre-flood stages. Post-exercise interviews facilitated reflection that transformed experience into learning, enabling officials to identify challenges, generate improvements, and formulate hypotheses for decision-making and preparedness. Key insights were the importance of early preparedness, bold decision-making under uncertainty, and the necessity of monitoring smaller rivers alongside major ones. Moreover, differences in learning were evident between the municipalities, reflecting differences in exercise scale: in Obihiro, where the simulation was conducted as a risk-management-unit–level exercise, reflections focused primarily on evacuation-information decision-making, whereas in Memuro, where the simulation was conducted as a full-scale municipal exercise, learning extended to broader organizational and field-level operations. Expanding the scale and participation of simulation exercises was found to enhance learning beyond evacuation decisions, extending to field operations and headquarters management.