Developing Multiple-Linear Regression Model to Predict Soil Cation Exchange Capacity for Nagaland Soils

Cation exchange capacity (CEC) is an important parameter to assess the soil health and fertility. The procedure for measuring CEC is complicated and time consuming. To overcome this issue, researchers have developed and tested models to estimate CEC, but no such model has been developed for North-eastern region (NER) of India. In the present study, a training dataset of 198 numbers of soil samples having data of soil texture, bulk density (BD), pH, soil organic carbon (SOC) and CEC was used to develop step-wise regression model for CEC. Correlation analysis was done to extract the influential parameters for predicting CEC. Results showed that basic soil parameters were able to predict CEC and can define 90 % of variability with SSE value of 2.76. The agreement between observed and predicted CEC in validation dataset with R2 value of 0.665 provided a strong basis to identify input variables for predicting CEC in the region.