ANALYSIS AND COMPARISON OF SPATIAL RAINFALL DISTRIBUTION APPLYING DIFFERENT INTERPOLATION METHODS IN PORSUK RIVER BASIN, TURKEY

Abstract

Porsuk River Basin, located in the Central Anatolia, Turkey, has a drainage area of 10818.41 km2 and a total length of 460 km,\rwhich makes it a significant region for a variety of hydrological and hydropower studies. Therefore, it is necessary to determine\rthe rainfall distribution over the basin so that related projects and processes, such as dam planning works, can be properly\rperformed. To fulfil this aim, the meteorological data gauged between 1927-2015 by 15 different stations within and around\rthe study area were used to perform interpolation functions with three widely known spatial distribution methods, namely\rThiessen Polygons (TP), Spline (SP) and Inverse Distance Weighting (IDW), for the determination of the rainfall distribution.\rMoreover, the reliability of these methods was also evaluated and compared in terms of their Mean Absolute Error (MAE),\rMean Square Error (MSE), Root Mean Square Error (RMSE) and Correlation Coefficient (R2) values. The results revealed that\rIDW method, in general, was the most appropriate option for Porsuk River Basin in comparison with SP and TP methods, as\rMAE, MSE, RMSE and R2 values of this method was found 33.359, 1710.385, 41.357 and 0.7118 respectively. However, TP\rdisplayed smoother results at the points where the rain gauges were closer to each other or dense.