Improving SWAT model performance in the upper Blue Nile Basin using meteorological data integration and subcatchment discretization
- 1Chair of Hydrology and River Basin Management, Faculty of Civil, Geo and Environmental Engineering, Technische Universität München, Arcisstrasse 21, 80333 Munich, Germany
- 2Irrigation Engineering and Hydraulics Department, Faculty of Engineering, Alexandria University, El-Horia St., 21544 Alexandria, Egypt
- 3Chair of Geography and Geographic Remote Sensing, Faculty of Geography, Ludwig-Maximilians-Universität München, Luisenstrasse 37, 80333 Munich, Germany
Abstract. The Blue Nile Basin is confronted by land degradation problems, insufficient agricultural production, and a limited number of developed energy sources. Hydrological models provide useful tools to better understand such complex systems and improve water resources and land management practices. In this study, SWAT was used to model the hydrological processes in the upper Blue Nile Basin. Comparisons between a Climate Forecast System Reanalysis (CFSR) and a conventional ground weather dataset were done under two sub-basin discretization levels (30 and 87 sub-basins) to create an integrated dataset to improve the spatial and temporal limitations of both datasets. A SWAT error index (SEI) was also proposed to compare the reliability of the models under different discretization levels and weather datasets. This index offers an assessment of the model quality based on precipitation and evapotranspiration. SEI demonstrates to be a reliable additional and useful method to measure the level of error of SWAT. The results showed the discrepancies of using different weather datasets with different sub-basin discretization levels. Datasets under 30 sub-basins achieved Nash–Sutcliffe coefficient (NS) values of −0.51, 0.74, and 0.84; p factors of 0.53, 0.66, and 0.70; and r factors of 1.11, 0.83, and 0.67 for the CFSR, ground, and integrated datasets, respectively. Meanwhile, models under 87 sub-basins achieved NS values of −1.54, 0.43, and 0.80; p factors of 0.36, 0.67, and 0.77; r factors of 0.93, 0.68, and 0.54 for the CFSR, ground, and integrated datasets, respectively. Based on the obtained statistical results, the integrated dataset provides a better model of the upper Blue Nile Basin.