Evaluation of observed and satellite-based climate products for hydrological simulation in data-scarce Baro -Akob River Basin, Ethiopia

Abstract

Hydrological modeling in the data-scarce regions is challenging. Recent advances in remote sensing make producing climate data easier and provided alternatives to the conventional climate data sources, particularly for hydro-climatological studies in the data-scarce regions. However, rigorous evaluations are warranted to evaluate the reliability and effectiveness of gridded climate products. This study evaluated the CFSR and CHIRPS products against the observed climate for hydrologic prediction at two streamflow gauging stations. The Soil and Water Assessment Tool (SWAT) is used to simulate hydrological processes in the watersheds. Model sensitivity analysis and calibration were conducted using the Sequential Uncertainty Fitting-2 (SUFI-2) algorithm in the SWAT-Calibration Uncertainty and Prediction (SWAT-CUP). Streamflow simulation performance was evaluated using several statistical, hydrological performances and prediction uncertainty of p-factor and r-factor within the SUFI-2 framework. Simulation of the model using the observed and gridded climate products provided satisfactory model performance in terms of the goodness of fit criteria and uncertainty prediction. Results show that CHIRPS rainfall combined with CFSR and observed temperatures demonstrates slightly better statistics and hydrological performance than observed, but shows substantial difference than CFSR. The simulations with CHIRPS and observed yielded minor differences in the water balance component estimates where the CFSR simulation gave lower average annual rainfall, resulting in lower water balance components. Quantifying the various water balance component in the watershed helps to tackle the water management issues. Overall, the CHIRPS rainfall combined with CFSR temperature can serve as a better alternative climate input for hydrological applications in the study region