GIS Based Spatial Prediction of Major Nutrient Status of Soils at Farm Scale for Site Specific Crop Management

Abstract

ABSTRACT In the current research, Global Positioning System (GPS) based 121 surface soil samples were collected and analyzed in the laboratory for different soil fertility parameters. The results showed that the bulk density (BD) of soils of 1.39 Mg m−3, dark gray and grayish brown, sandy clay loam, neutral to slightly alkaline soil reaction (pH), safe in electrical conductivity (EC), low to medium in organic carbon (OC) and non-calcareous in nature. The nitrogen (N), phosphorus (P), potassium (K) and sulfur (S) were analyzed as low, medium, high and low with NI of 1.26, 2.19, 2.69 and 1.31, respectively. In the Farm, field (F) wise, OC, N, P, K and S content in soil varied from 4.89 (F20) to 7.33 g kg−1(F15), 196.71 (F20) to 264.25 kg ha−1 (F15), 4.10 (F19) to 27.04 kg ha−1 (F16), 292.72 (F2) to 674.80 kg ha−1 (F13) and 3.99 (F18) to 13.75 mg kg−1 (F13), respectively. Post hoc results also revealed that fields and land use types were significantly different. Ordinary Kriging in Geo-statistical tools revealed that the exponential model was best fitted for, N, P, K, OC, √CaCO3 and √K, which exhibited moderate spatial dependence (>25 but < 75%), whereas Gaussian model was used for BD, pH and CaCO3. Correlation ranges(m) for P was 717.57 m and exhibited strong spatial dependency (<25%). The J-Bessel model was used for EC and√EC. The spherical model was used for S which showed 57.95% spatial dependence. The spatial maps could be used as a basis for precision farming.