Please login first
Evaluation of Soil Fertility, Geospatial Mapping and Quality Index using Kriging operation in Agricultural Land of Karaikal District, Puducherry, India
* 1 , 1 , 1 , 2 , 1
1  Department of Soil Science and Agricultural Chemistry, Pandit Jawaharlal Nehru College of Agriculture and Research Institute, Nedungadu Post, Karaikal, Puducherry, India, 609603
2  Department of Agricultural Economics and Extension, Division of Agricultural Statistics, Pandit Jawaharlal Nehru College of Agriculture and Research Institute, Nedungadu Post, Karaikal, Puducherry, India, 609603
Academic Editor: Sanzidur Rahman

Abstract:

Proper soil thematic maps are essential for developing effective soil nutrient strategies. This study aimed to demarcate soil nutrient properties and spatial variability across the Thirunallar region of Karaikal District through a digital survey, adopting a toposheet and base map from Sentinel 2 satellite data for sample collection in each grid (320-meter interval) at 0-15cm depth. The soil data were fitted into the model and layer maps were generated using ArcGIS (v 10.8.2) considering the Kriging function and the geostatistical method. The results revealed that soils were in the acidic to alkaline range (5.18 - 8.93), exhibited less saline (0.035 - 3.502 dS m-1), and were low to high in SOC (0.24 - 1.41%), respectively. The available N ranged from low to high (142.80 - 739.20 kg ha-1), while the range was medium to high for available P (15.33 - 98.44 kg ha-1) and low to high for available K (90.18 - 493.42 kg ha-1). Sulphur was reported to be in the medium to high range (9.94 - 99.67 mg kg-1). Exchangeable properties were sufficient, as were micro-nutrient (Fe, Mn and Cu) levels, except for Zn. The coefficient of variation was reported to be high in soil EC (103.36%) and low for pH (10.87%). The efficiency of quantification with respect to the R2 value and Root Mean Square Error provided explained variance and residuals in the derived model, and the majority of soil properties were best fitted for the spherical model. Semivariogram modelling indicates a strong spatial dependency (SpD) level. The anticipated dataset for each parameter showed the lowest RMSE, which accounted for soil EC and SOC (0.524; 0.154), and the R2 values corresponded to good model fitting for SOC (0.984) and Cu (0.954). The SQI derived from the PCA underscores the fact that calcium and sulfur had a greater contribution towards soil quality. Thus, integrating spatial analytical data and the SQI provides better regional soil management practices, optimum fertilizer use and future sustainable practices.

Keywords: fertility; geostatistics; kriging; remote sensing; variability

 
 
Top