Developing a Probabilistic Flood Plain Boundary Using HEC-1 and HEC-RAS

Despite the advances in computer software and modeling tools for hydrologic and hydraulic analysis, the science of developing flood plain extents and depths still involves a great deal of uncertainty. While the physics of rainfall runoff modeling and water surface profile computations is well known, complexities and uncertainties in deriving accurate input parameters still exist. The result is that for any project, such as the development of a flood insurance rate map (FIRM), a number of similar solutions can be generated that can be defended using sound engineering principles. A better approach to developing a flood plain boundary is to incorporate uncertainty in the modeling parameters during model development. Using a series of stochastic simulations a number of probable results from hydrologic and hydraulic models can be computed. Computer automated flood plain delineation tools can then be used to determine flood plain boundaries and inundation depths for the series of model solutions. Finally, a resulting probabilistic flood plain map can be generated from all of the solutions. The probabilistic flood plain shows contours of percent probability that a location is flooded. For example the 100% probability flood zone is the area that is flooded as a result of each of the model runs, whereas the 50% probability flood zone is only flooded by half of the resulting model runs.