Agent-based Models in Flood Simulation - A Case Study for New Brunswick’s Flooding Events in 2018

Abstract

This thesis introduces an Agent-Based Modeling (ABM) framework for flood simulation and inversion modeling in flood-prone areas, aiming to improve our understanding of the complex dynamics of flooding and provide valuable insights for flood management. The developed model incorporates multi-source terrain datasets, and integrates water flow and meteorological conditions from remote sensing data sources. It takes into account factors such as precipitation patterns, geographical features, and Digital Elevation Model (DEM) resolution to accurately represent flood characteristics. The model's parameter settings are derived from extensive experimentation, allowing for effective control and meaningful results. By considering the impact of precipitation and the presence of rivers, the model demonstrates its ability to simulate flood inundation with a reasonable level of accuracy. Overall, this comprehensive model provides a valuable tool for flood simulation and offers insights into flood dynamics for effective flood management and mitigation strategies.