Climate change projection and the effect of its possible scenarios on water resources: The example of Di̇cle-Fırat sub-basin

As the Earth's climate system undergoes unprecedented transformations, understanding the complex interplay between climate change and water resources emerges as an extraordinary scientific pursuit. In this study, in order to examine the effects of climate change on the Euphrates and Tigris sub-basins, flow, precipitation, temperature, relative humidity and evaporation variables covering different years obtained from 23 meteorological and 11 hydrological stations were firstly analyzed by Mann-Kendall (M-K), Spearman Rho (SR) and Innovative. Trends were examined using Trend Analysis (ITA) methods. In this way, detailed information about trend assets is obtained and an idea is obtained about how the stations will change from past to present and in the coming years. Then the stations in the work area. Sequential Mann-Kendall (SMK) and Cusum tests were applied to detect sudden changes. Secondly, monthly, 3-month and 12-month SP, SPEI and SDI meteorological and hydrological drought indices of the stations in the study area were determined. By making trend analyzes of the drought indices of these stations, detailed information about the drought of the stations was obtained. In addition, the correlations between these indices were examined and the relationships between them were determined. Thirdly, projection maps were produced for meteorological data for RCP 4.5 and RCP 8.5 scenarios of GFDL-ESM2M, HadGEM2-ES, MPI-ESM-MR global climate models. Since we worked with forecast data while producing these maps, station data from provinces outside the study area were also used in order to produce more accurate projection maps. Finally, by using machine learning methods such as Multi-Layer Perceptron (MLP), Random Forest (RF), Support Vector Machine (SVM), the model with the highest accuracy was determined by establishing models in which the meteorological data in the observed years were independent and the flow data was the dependent variable. Then, rainfall, temperature, evaporation and humidity projection data were used as input and flow data was predicted as output. In this way, the future value of the flow is revealed, contributing to the aim of more accurately planning the use of water assets in the future