Authors: David Baude, Rebecca Diehl, Ijaz Ul Haq, Kristen Underwood, Alexander Prescott, and Beverley Wemple – University of Vermont; Belize Lane and Colin Phillips – Utah State University
Title: Assessing Where HAND-Based Approaches Are Suitable for Flood Inundation Mapping
Presentation Type: Poster Presentation
Abstract: While flood inundation mapping based on low-complexity (i.e., HAND) approaches oversimplifies flooding dynamics in some settings leading to errors in mapped extents, it has been shown to be adequate for parts of the river network. Differences in applicability and accuracy across the landscape may be a function of the variability of watershed and hydraulic characteristics of the river and floodplain. To characterize the relationship between river settings and the suitability of HAND-based inundation mapping, we identified 82 benchmark sites distributed across CONUS. We describe the range of HAND stages defined by the benchmark inundation maps and identify the HAND stages that optimize the agreement between benchmark maps and those developed using the HAND-SRC (synthetic rating curve) approach.
Departures between optimized and synthetic rating curves from NOAA’s National Water Model are described by their offset and understood relative to attributes of the setting. We find that HAND-based inundation mapping performance may be described, in part, by landscape characteristics that can be derived from broadly available terrain datasets. This work suggests that topographic features may help facilitate spatially variable model selection and bias correction, supporting more computationally efficient approaches to large-scale flood inundation forecasts.