Agent-Based Modeling for Managed Retreat and Flood Risk

Abstract

Managed retreat is an increasingly important climate adaptation strategy for communities exposed to recurrent flooding. Agent-based modeling helps represent the household-level decisions and policy feedbacks that determine whether relocation programs succeed or fail. This topic focuses on managed retreat under flood risk, emphasizing heterogeneity in household resources, risk tolerance, social attachment, and relocation incentives. Hyperspectral weak-signal detection is included as a parallel literature stream because both managed retreat and contamination detection involve early identification of latent risk. In managed retreat, the weak signals may include repeated minor flooding, insurance stress, declining property values, or rising adaptation costs. The literature supports modeling approaches that connect individual decision-making, policy design, and system-level risk reduction.