Efficient Generation of Annual Exceedance Probability Maps

Abstract

The United States Federal Emergency Management Agency (FEMA) flood insurance rate maps (FIRMs) define a single line delineating the floodplain boundary at the 100-year recurrence interval. The U.S. Army Corps of Engineers (USACE), at the suggestion of the National Research Council, now promotes the concept of Annual Exceedance Probability (AEP) as the method used to determine the probability of flooding caused by the failure of a levee or other flood control structure at a particular study point. By using the HEC-1 (HMS) and HEC-RAS models for hydrologic and hydraulic calculations in conjunction with the Watershed Modeling System (WMS) for floodplain delineation, an AEP map can be created to determine and visualize the spatial distribution of the probability of flooding. An AEP map represents a continuous solution of the probability of flooding, rather than a single boundary line between areas considered at risk to a 100-year event and those not at risk. A single line definition often leads to unreasonable litigation over extents because, within limits of probability, engineers can use modeling input parameters to develop separate, yet reasonable results according to a client’s self interests. An AEP map is generated using Monte Carlo techniques to run hundreds or thousands of hydrologic/hydraulic/floodplain mapping simulations in order to capture the natural variability and model uncertainty inherent in the flood modeling/mapping process. The result is a spatial distribution of flood probability that, together with a map of infrastructure, can be integrated to determine actual flood damage (or flood protection) in terms of dollars. This paper/presentation will focus on the work of multiple AEP map case studies with the intent of better understanding the process of creating an AEP map in addition to learning more about the number of simulations required for convergence to a consistent AEP solution. Copyright ASCE 2005 EWRI 2005