Flood Modeling and the Influence of Digital Terrain Models: A Case Study of the Swannanoa River in North Carolina

An increase in flood disasters nationally and internationally has highlighted the need for accurate flood modeling regarding flood insurance and emergency response. Topographic data is the most important variable in determining flood modeling accuracy according to the National Research Council. Increasing availability of airborne light detection and ranging (LiDAR) data warrants the investigation of the optimal resolution or range of resolutions needed to represent digital terrain models for accurate operational flood modeling. Few studies have focused on flood modeling in mountain environments. The Swannanoa River, located within the Appalachian Mountains of western North Carolina, was selected for this study based on unique physical characteristics, a substantial built environment within the 100 year floodplain, and significant recorded levels of historical flooding.Flood modeling accuracy was evaluated using LiDAR elevation data represented at a series of equivalent resolutions (3.77m, 6m, 8m, 10m, 12m, 15m, 20m, 25m, and 30m) and United Stated Geological Survey Level 2 digital elevation model data represented at 10m and 30m resolutions combined with a series of flood recurrence intervals (10yr, 25yr, 50yr, 100yr, and 500yr). A variety of descriptive and inferential statistics were used to evaluate generated water surface profiles and depth grids.