GIS-Based Interpolation Methods for Estimating Spatial Distribution of Nitrogen Content in the Soil

The spatial statistical method has been found favorable among agricultural scientists who seek to map yield estimates or soil properties of an area from a small number of samples of known location scattered throughout the area. Spatial statistics is concerned with the analysis of spatially referenced data and the study of associated spatial statistical models and processes. The agricultural benefits of accurate interpolation of spatial distribution patterns of nitrogen content (N) are well recognized. For the study different interpolation techniques in a geographical information system (GIS) were analyzed and compared for estimating the spatial variation of N at four different blocks of Raipur. Number of sample was selected by Stratified random samples with PPS (probability proportional to sample) sampling method. Different interpolation methods such as inverse distance weighting (IDW), ordinary kriging (OK) and lognormal kriging were used to generate spatial distribution of N. The cross validation is applied to evaluate the accuracy of interpolation methods through coefficient of determination (R²) and root mean square error (RMSE). After comparison it has been observed that inverse distance weighting (IDW) did not perform well due to its higher values of RMSE in comparison to ordinary kriging and lognormal kriging Interpolation techniques. The values of R² show the strength of relationship among the nitrogen content at different locations. Results showed that after transforming the data for skewed distribution gave the précised result; it means that lognormal kriging provides best result as compared to ordinary kriging interpolation technique.