Site Specific Nutrient Management of an Intensively Cultivated Farm Using Geostatistical Approach

Abstract

Spatial variability and site specific fertility management of six soil parameters viz. soil pH, soil organic carbon, available nitrogen, available phosphorus, available potassium and available sulphur were undertaken using both classical statistics and geostatistical analysis for the intensively cultivated research farm of Indian Agricultural Research Institute, New Delhi that represents agro-ecological sub-region of north Punjab plain, Ganga Yamuna Doab and Rajasthan upland under hot semi-arid eco-subregion. The soil properties exhibited variability with highest variation observed in available sulphur (72.7 %) whereas the lowest variation was in pH (7.3 %). Bivariate correlation coefficient analysis showed highly significant (P < 0.01 %) relationship within six attributes of the soil, except between the available sulphur and pH. The exponential model for pH, available phosphorus and available potassium were the best fitted semi-variogram models while that of soil organic carbon, available nitrogen and available sulphur was spherical model. The nugget and sill ratio indicated moderate spatial dependency for all soil properties within the farm. Ordinary kriging was used for interpolation of soil attributes in non-sampled areas. It was observed that use of geostatistical method on a small scale could accurately generate the spatial variability map of soil nutrients in alluvial soils. Based on the study recommended level of fertilization except sodic soil patches where reclamation measures are needed for sustainable crop productivity are advocated.