Authors: Riley McDermott, Sagy Cohen, Reihaneh Zarrabi – University of Alabama
Title: Determination of Bankfull Channel Conditions Using Percentile Estimation and Regression
Abstract: The introduction of new continental-scale hydrological datasets has opened up the possibility of more accurate large-scale modeling of river channel geometry. Previous methods of modeling channel geometry were limited by dataset size and quality of channel dimension measurements. The availability of a new robust dataset containing Acoustic Doppler Current Profiler (ADCP) measurements, the gold standard of underwater channel attribute measurements, has fueled the development towards a new model of channel geometry. The dataset used in this study is limited in its ability to model channel geometry by its inclusion of measurements of river dimensions under flood conditions, which do not reflect the dimensions within the channel. This project aims to isolate measurements within the channel dimensions to generate the optimal dataset for bankfull channel geometry modeling. To create this dataset, observations measuring flood conditions are identified for removal via percentile estimation of bankfull conditions as well as identification of bankfull and flood conditions through fitted polynomial regression. The resulting dataset containing only within channel measurements and its subsequent models are all created in an effort to improve future flood inundation mapping techniques.