Mapping of Cation Exchange Capacity Using Geostatistics and Particle Component Analysis

Soil Cation Exchange Capacity (CEC) is an important vital indicator of soil fertility and as well, of pollutant sequestration capacity. Throughout the present study, spatial variability of soil CEC was investigated Through Kriging and corking with the principal components derived from soil physico-chemical properties including texture, (clay, sand, and silt content), organic carbon, electrical conductivity as well as pH. To follow the purpose, 247 soil samples were collected from central areas of Guilan province. Seventy five percent of the soil samples were used for training and 25% for testing. The first two Principal Components (PC1 and PC2) together explained 68.54% of the total variance of soil physico-chemical properties. PC1 explained the highest significantly positive correlation with CEC (r=0.81, P<0.01), whereas there was no significant correlation observed between CEC and PC2 (r=-0.19). PC1 was then used as an auxiliary variable in cokriging method for the prediction of soil CEC. Root mean square error of kriging for the test dataset was found 0.159 and that of cokriging for the dataset amounted to 0.118. The cross-validation determination coefficient (R2) for the test dataset was recorded 0.49 for kriging while 0.71 for cokring at a 0.01 level. The results show that interpolation through cokriging, with an auxiliary variable PC1 derived from soil physico-chemical properties, proves more reliable than through kriging. In addition, the principal components that bear the highest positive significant correlation with the dependent variable are of the most potential for prediction through cokriging.