Assessment of the Interdependence of Soil Properties and Nutrients of District Mirzapur (U.P), India
Baishali Mishra *
Department of Mathematics and Statistics, C.C.S. Haryana Agricultural University, Hisar, India.
Vinod Kumar
Department of Mathematics, Statistics and Computer Science, G. B. Pant University of Agriculture and Technology, Pantnagar, India.
*Author to whom correspondence should be addressed.
Abstract
In present paper assessment of the interdependence of the soil characteristics and nutrients through Multiple Regression analysis and Path coefficient analysis was done. For this soil characteristic data related to the farmers of sampled villages of different blocks of Mirzapur district of Uttar Pradesh was used. The study reveals that pH has significant positive correlation with Sulphur and Boron and significant negative correlation with Nitrogen, Manganese and Iron. EC has shown significant positive correlation with Potassium, Sulphur and Boron and significant negative correlation with Iron, Manganese and Zinc. Moreover, OC is found to have positive and significant correlation with all the nutrients except Nitrogen. Path coefficient analysis reveals that EC, OC, Sulphur, Copper and Boron have positive direct effects on soil pH and remaining have negative direct effects on soil pH. Multiple Linear Regression Models are developed for estimating soil properties (pH, EC, OC) on the basis of given values of soil nutrients. These models are found to be quite efficient for the said purpose on the basis of large value of R2 and small value of Root Mean Square Error.
Keywords: Path coefficient, correlation coefficient, multiple linear regression, coefficient of determination, root mean square error