Performance of Height Above Nearest Drainage (HAND) Model Compared to 2D Hydraulic Models for Flood Assessment

Riverine flooding is one of the main geohazards endangering human life and property. Delineating floods is critical to reducing recurring and future damage. For floodway delineation, floodplain managers have historically employed hydraulic models. The issues with such models are that they require many inputs and/or higher computational powers. More recently, with advances in light detection LiDAR) data acquisition, 2-D models need many inputs, higher processing power and computational time. Moving to this next dimension in technological advancement using LiDAR has improved the outputs obtained and perspective of the modeling drastically making flood-inundation mapping available across floodplains. However, these advancements still leave rural communities disadvantaged due to the high computational cost.

Here we are comparing traditional hydraulic models using SMS:SRH-2D (Surface-water Modeling System with GUI developed by Aquaveo for Sedimentation and River Hydraulics - Two-Dimension modeling system engine by the U.S. Bureau of Reclamation) and HEC-RAS 2D (Hydrologic Engineering Center's River Analysis System by the U.S. Army Corps of Engineers) with HAND to assess the usability of HAND in real-life scenarios for rural communities. Huron Creek, in Houghton, MI (approximate micropolitan population - 40,000), was selected as the study area, and simulations use a LiDAR DEM (digital elevation model) with 2 ft. resolution. The 100-year flood frequency was selected for the comparison. The hydraulic simulation results with SMS:SRH-2D and HEC-RAS 2D were similar. The inundation extent from the HAND was similar to results from 2D models and proved usable for enhancing flood resilience and adaptation in rural communities. Since the use of the HAND model for comparison is in its infancy, further refinement is recommended. Additionally, the HAND model can be integrated with the Google Earth engine; processing on the cloud, thus removing the need for higher processing power and making it more usable for rural communities to quickly determine flood-prone areas.