The applicability of the global brook90-R physically-based hydrologic model in some river basin in Turkey: Flow estimation study

The use of hydrological models is very important in terms of understanding the hydrological process in the basins, estimating the hydraulic/hydrological variables, revealing the effects of climate change on water resources, creating future climate projections and determining water management strategies. In this study, the physically based Global BROOK90-R (GB90-R) hydrological model, which was used for the first time in Turkey, was established in the Çarşamba Stream, Karasu Stream and Körkün Stream Basins to predict monthly streamflow amounts. Meteorological (precipitation, temperature, wind speed and solar radiation) and physical data (digital elevation map, land cover and soil maps) were entered as input to the model. The model was run for the period 1977-2016 for the Çarşamba Stream Basin, 1979-2020 for the Karasu Stream Basin and 1992-2017 for the Körkün Stream Basin. The success of the model results obtained for each basin was evaluated using the coefficient of determination (R2), Nash-Sutcliffe Efficiency (NSE) coefficient, Kling-Gupta Efficiency (KGE) coefficient and percent bias (PBias) metrics. According to the results obtained, the highest success in monthly streamflow estimation with the GB90-R model was obtained for the Karasu Stream Basin with NSE=0.570 value. The success of the model in the monthly flow estimation in the Çarşamba Stream Basin (with a value of NSE=0.521) is very close to the success of the Karasu Stream. The lowest performance in monthly flow estimation with the GB90-R model was obtained in the Körkün Stream Basin with NSE=0.337 value. In the study, evaporation (EVPP), soil moisture (SWATT) and snow (SNOW), other outputs of the GB90-R model other than flow, were evaluated in terms of water budget for all three basins. The most important advantage of the GB90-R model is that it requests only the basin location as input, while providing detailed hydrological parameters as output.