Investigation of the effects of different spatial interpolation methods on soil mapping

Technological developments in our country and in the world are used intensively in agricultural activities as in every field. Considering that agricultural activity is extremely vital for countries and therefore people, it is an extremely important need to increase agricultural productivity day by day. One of the methods that should be done to increase agricultural productivity is to provide the most appropriate elements for the growth and life of plants by conducting soil surveys. In order to meet these needs of the plants in the most appropriate way, the soil surveys should be in an appropriate distribution and the most appropriate estimation should be made for other regions by making use of these soil surveys. Many spatial estimation methods have been developed for this estimation process. In this context, it has been estimated for 200 test points with the help of 600 model data, using the 5 most widely used spatial estimation methods (Nearest neighborhood, Inverse weighting, Surface trend analysis method, Point KRIGING and Artificial neural network method) and artificial neural networks. An android application was developed using Android Studio for 3 of these estimations (Nearest Neighborhood, Inverse of Distance Weighting and Surface trend analysis), Surfer program for Point KRIGING Method, which is another of the spatial estimation methods, and MATLAB program for ANN (Artificial Neural Networks). Thematic maps of the results of the predictions were created in the Arcgis program, the results were analyzed statistically and their usability was examined.