Spatial Interpolation Techniques of Some Soil Chemical Properties with Prediction

Abstract

This paper deals with spatial prediction of two interpolate methods. The purpose of this research is to obtain estimate parameters based on a sample in the spatial process. This work using kriging technique and inverse distance weighting method. The objectives of this research are to; explore properties of variogram function under real data in all directions; to provide and to estimate the parameters of covariance models based on regionalized variables. The data adopted of this work taken from (60) data for each soil chemical from Mosul Quadrangle of Mosul/Iraq. Data of (PH , Na , Ca , Ec , Hco3 , and So4), after applied kriging technique and inverse distance weighting method, we show the minimization of the estimation variance amounts to choosing the estimator, we get the smallest standard errors with the best covariance models also, the constraint of weights satisfied. In addition, kriging is the best linear unbiased estimator. We note that kriging technique is better than inverse distance weighting method because, the performance of each interpolation technique that is clear by comparing the deviation of the estimates from the measured data with cross-validation. In conclude, that improvement does not depend on more used statistical methods, but rather higher quality and a larger number of soil environmental variables should be used to improve predictions. The prediction of this work carried out by using matlab language.