Flood Modeling in the Coastal Plains and Mountains: Analysis of Terrain Resolution

The number of flood disasters has increased worldwide in recent decades. Identifying the optimal resolution or scale at which to represent digital terrain models (DTMs) is critical in order to improve our ability to accurately and efficiently model floods. Few studies have attempted to compare flood modeling results using different resolutions of DTMs in divergent environments. In this study flooding on the Tar River in the coastal plains and the Watauga River in the mountains of North Carolina were modeled using hydrologic information obtained following Hurricanes Floyd and Ivan. The effectiveness of DTMs derived from light detection and ranging and United States Geological Survey elevation data at commonly available resolutions in North Carolina were assessed. A quantitative diagnostic method based on measuring the distance flooded along transects was applied for evaluating the horizontal extent and internal pattern of flooding. The use of additional diagnostic metrics (area, volume, and shape) along with a visual graphic assessment enhanced the evaluation of flood modeling results. The extent and internal pattern of flooding in the low-relief coastal plains was found to be especially sensitive to the representation of terrain, and in the mountains 30 x 30-m data regardless of source were found to be dramatically unsuitable for flood modeling.