Effect of grouping soils based on their organic carbon and clay content on performance of hierarchical pedotransfer functions of soil cation exchange capacity

Cation exchange capacity (CEC) is one of the main factors in monitoring and management of soil quality in order to achieve sustainable production. Although CEC can be measured directly, these measurements are difficult, time-consuming and costly, especially for aridisols. One alternative method is the use of pedotransfer functions (PTF), in which CEC of soils is estimated through available soil information. A dataset including 1141 data points was used for development and accuracy test of functions and another independent data set with 232 data points was used for validation of the PTFs. The results showed that, grouping the soils based on clay and organic carbon contents are generally reduced the coefficient of variation of clay, organic carbon and CEC. In general, the grouping of soils declined the correlation of clay and organic carbon with CEC. In soils with organic carbon contents of less than 0.5%, CEC was predicted with reasonable accuracy using only the clay. Generally, grouping the soils enhanced the accuracy of the PTFs, while in most cases decreased their reliability. However grouping the soils based on their clay content, except for soils with clay content ≥ 35%, was more effective in improving the accuracy and reliability of the functions. Clay (except for soil with clay content of ≥ 35%) was the most influential factor in predicting the CEC of the soils, so that including the other variables were not significantly improve the accuracy and reliability of the functions.