Examining the uncertainties of the factors in meteorological drought analysis

In order to predict the future effects of global warming, mathematical analysis of drought and hydrological calculation methods are essential to estimate water needs accurately. Therefore, in this study, Potential Evapotranspiration (PET), General Circulation Model(GCM), Representative Concentration Pathway(RCP) are questioned to provide mathematical approach to all uncertainties they cause. Otherwise, it would be wrong to estimate future drought without knowing how much offset the estimate could be. In this study, the results of every model and method are taken into the variance analysis test (ANOVA) and examined the results to provide a mathematical approach to uncertainties these models create. As a result, uncertainties of meteorological droughts are provided not just in historical data, but the future projections as well. In this study the temperature (T) and precipitation (P) data are taken from various measurement stations and optimized for modelling the basin which is one of the Midwestern basins of Turkey, Gordes. In order to calculate PET uncertainty, 12 PET models are created using MATLAB program. Conjuring the PET and P data Standardized Precipitation and Evapotranspiration Index (SPEI) is created for comparing results for different models. After the drought assessments of Gordes basin, downscaling for GCM data is provided. Then, 1980-2005 calendar data are changed into 1981-2005 water year data to ensure every data is used respectively at the same time. Afterwards, projection of GCM and RCP data are created using MATLAB program for water years of 2020-2050. This projection included 12 PET models, 12 GCM data, and 2 RCP data. The results of these 288 combination are tested using ANOVA. The results from variance analysis provided uncertainty factors for meteorological drought. After examining meteorological drought, for comparison, uncertainties of agricultural droughts are also calculated. The varieties of PET results were thought to cause the uncertainty in the beginning. However, in result GCM data is found to be the main reason for uncertainty in both Agricultural and Meteorological drought projections. At the end, preventing miscalculations caused by uncertainty is discussed and made suggestions for making better future meteorological drought analysis.