Assessing The Spatial Variability of Soil Nutrients Prediction Using GIS-based Interpolation Techniques

GIS-based spatial interpolation techniques are utilized in soil sciences to analyze and predict the soil quality in precession agriculture. This research involved the analysis of seventy soil samples at depths of up to 40 centimeters from randomly selected farm plots surrounded by an area of 300ha from the Tarakeswar block in Hooghly region, West Bengal, India during the post-harvest period 2019-2020. Current work assigns the five spatial interpolation techniques namely IDW, RBF, LPI, OK, and EBK for the prediction of soil nutrients on a local scale with the site-specific soil management through GPS-aided Geographical Information System. The accuracy of different interpolation techniques examines by the coefficient of determination (R2) and root mean square error (RMSE) through the cross-validation method. Exponential models were appropriate to the experimental semivariograms for the soil OC, Zn, pH, EC, K, Sand, Silt, and Clay while P and N were best suited to the Gaussian model. The outcomes demonstrate that LPI and EBK is the most preferred method with the highest R2 and smallest RMSE value for interpolation of spatial variability of soil nutrients parameters distribution. It is widely believed that soil variability is an instrument that can help improve the management of land and reduce conflicts within the rural community.