Authors: Darlly Rojas Lesmes, Jorge Sanchez Lozano, Jim Nelson – Brigham Young University
Title: Improving the Accuracy of GEOGloWS ECMWF Global Hydrologic Model using a Bias Correction Methodology based on Quantile-Matching Approach
Abstract: Global hydrological models are essential for water resources management, climate change impact assessment, and flood and drought forecasting. However, these models face significant challenges when being used on local scales such as managing Big Data, communication, adoption, and validation. The biggest challenge is validation, which requires extensive observed data that can be difficult to obtain. In this study, we evaluated the performance of GEOGloWS ECMWF Global Hydrologic Model and found it has systematic biases limiting its reliability and accuracy. To correct these biases, we propose a bias correction methodology based on a quantile-matching approach. Our approach was effective in correcting the magnitude and seasonality of simulated streamflow values in different regions, climate groups, watershed areas, and observed time series lengths. The results show that the bias correction method can significantly improve the accuracy of GEOGloWS Model ERA5 streamflow reanalysis and be useful in hydrological studies and water resources management.