National Bureau of Soil Survey and Land Use Planning, Amravati Road, Nagpur, Maharashtra, 440010.
Present address:1 Regional Centre, NESS & LUP, Bangalore, 560024.
Field trials on influence of soil-site characteristics vs management for rainfed soybean were conducted for three years each on varying soil units in Khapri (1990–92), Saongi (1990–92) and Gondkhairi (1993–95) watersheds near Nagpur. Tho soil parameters, viz. depth, plant available water holding capacity (pAWC), length of growing period (LGP), cation exchange capacity (CEC) and clay content were the main yield contributing soil parameters. For achieving maximum yield, the optimum values from the quadratic curves, were worked out to be 140 cm (at optimum crop management) of soil depth, to 200 cm (at low crop management), 200 mm of PAWC, > 50 cmol (p+) kg−1 of CEC and 50 ± 10 per cent of clay. As per the nature of relationships of soil parameters with seed yield and based on the optimum values for obtaining maximum yield, the soil-site requirements have been defined for rainfed soybean. At low crop management level, among the 15 multiple linear regression (MLR) models established by least square technique, model No. V (Yield =-208.10 + 1.852 slope -7.223 erosion -0.142 depth +0.861 coarse fragments -0.767 sand +9.803 organic carbon -0.595 CEC +0.813 CaCO3 +25.37 pH +0.310 LGP) was found to be the best in forecasting the yield. The 10 soil-site parameters which are all significant in the above model explain 96 per cent (R2 = 0.961) variability in yield in swell-shrink soils. Similarly at optimum level of crop management, out of the five MLR models fitted, model No. V(Yield =-221.76+2.616 slope-10.116 erosion -0.151 depth+0.921 coarse fragments -0.860 sand +11.140 organic carbon -0.749 CEC +0.869 CaCO3 +28.290 pH +0.334 LGP) appears to be most accurate (R2 = 0.980).
Soybean, soil suitability criteria, swell-shrink soils, Vertisols, modelling