Fuzzy Relation Between The RDIst Index And The Water Table Of A Coastal Aquifer of Nestos Delta, Greece

Drought is a complicated natural extreme event associated with the decline of water availability below the normal conditions of a hydrological system, both from a temporal and from a spatial point of view. This work aims to assess the drought impacts on groundwater fluctuation in a shallow coastal unconfined aquifer via fuzzy approach. Meteorological drought intensity is estimated by the Standardized Reconnaissance Drought Index (RDIst) based on precipitation (P) and potential evapotranspiration (PET). In addition, groundwater modeling is carried out using MODFLOW and then, the simulated values of water table (WT) of a coastal unconfined aquifer are utilized. In order to relate the RDIst with WT, a fuzzy linear regression (FLR) is applied. FLR based on Tanaka model produces a fuzzy band, where all the data must be included within, incorporating the system uncertainty. The suitability of the achieved fuzzy regression model is tested by using appropriate measures. The propounded methodology is applied in the eastern area of Nestos River Delta, Prefecture of Xanthi, Greece.