Predicting Channel Conveyance in the Obion River Watershed Using SAMBLE Method

Abstract

Digital elevation models (DEMs) are widely used for hydraulic modeling, but one of the challenges in using a DEM for hydraulic modeling is its limitations in accurately representing the channel conveyance. This limitation is usually disregarded because the collection of bathymetry data is time and resource intensive, and it is often unavailable for well-gauged watersheds as well. Thus, a DEM-based hydraulic model may lead to less accurate conclusions even if detailed hydrologic data are available. The situation is more challenging in large and morphologically active river systems because DEMs only represent the river condition of their acquisition time. Therefore, the objectives of this research are to (1) develop a DEM correction method for predicting channel conveyance at the watershed scale and (2) demonstrate the significance of riverbed-level change for flood risk assessment in the Obion River. The proposed DEM correction method is called Slope Adjusted Mean Bed Level Elevation (SAMBLE) and utilizes the hydrologic data of a watershed to predict average riverbed level and thus channel conveyance at a watershed scale. It is tested on the Obion River watershed in northwest Tennessee. Analysis of the results showed that the SAMBLE method can predict average riverbed level with less than 1 m error that is more accurate than the best available Light Detection and Ranging (LiDAR) DEM for the Obion River watershed. Additionally, it can predict riverbed-level changes using time series of hydrologic data. The significance of riverbed-level change on flood risk is also demonstrated by estimating aggradation-degradation (via SAMBLE) in a morphologically active segment of the Obion River.