Spatial variability of macro-nutrients of N. P. K in order to soil fertilization management

Chemical fertilizers are the main resources to supply plant nutrients and optimum use of them, have an important role in field of environmental sustainability and the agricultural products safety. In order to management of the fertilizers, there is necessary to know the soil nutrients amount that can supply the plants. Unrecognizing of the soil variability in different locations and the using unique amount of fertilizers may lead to over or under supplying of soil nutrients. This research was done to study of spatial variability of the soil macro-nutrients; N, P and K, comparing Geostatistics methods to estimate them and carrying out the related maps in order to fertilizer management in Uremia plain, W. Azarbaijan. To estimate fertilize elements; Organic Carbon, Phosphorus and Potassium on non-sampled points, Kriging, Co-Kriging and Weighted Moving Average methods were used by GS+ and GIS. To compare the applicability these methods the Cross Validation was used by MAE and MBE. Results showed that kriging method has the highest with %99 correlation coefficient and gausian semi-variorum, has high accordance to estimate the non-sampled points. There is also obtained the MAE=0.18 to 0.75 and MBE= -0.002 to 0.12 meq/100gr soil in this method.