Interpolation for Environmental Application Using Kriging Spatial Technique

This research deals with the use of a technique of spatial interpolation, which is known as the prediction of the values of the spatial variable at unmeasured points. The usual Kriging method was used to obtain the best mathematical model. The experimental variogram function was applied to the spatial data of soil compounds data for Mosul/Iraq. The data consists of (121) real values for each of the soil properties (PH, SO4=, NO3-, PO4-, EC). The objectives of this research are to explore the properties of the variogram function under the real data in all compass directions. By applying this data to the prediction formulas, accurate results were obtained that prove the validity of the uncertainty. What supports these results is the minimum error variance and the results of the weights that give the sum equal to one in the prediction process as well as the results of the prediction accuracy verification procedures. Through the prediction values that were reached, the best mathematical model was reconciled through the basic variance functions. In conclusion, this improvement in the prediction process does not depend on the most widely used statistical methods, but higher quality and a large number of environmental variables should be used to improve the predictions, and sometimes the distance and weights are the closest results to the Normal model and cross-validation shows better performance of kriging techniques with the closest interpolation model(exponential, spherical and Gaussian).All programs used in this research were implemented in MATLAB language