Pedotransfer functions for cation exchange capacity estimation in highly weathered soils of the tropical highlands of NW Cameroon

Abstract

Cation exchange capacity (CEC) is a very important soil property widely used in the assessment of soil quality. Unfortunately, its determination in the laboratory can be very tedious, especially when samples are many, necessitating high costs and time. To curb these challenges, pedotransfer functions (PTFs) can be used as indirect and reliable tools for CEC estimation using easy-to-measure soil properties such as pH, soil organic carbon (OC), and texture. Soil survey data from 98 soil profiles in the North West region of Cameroon, belonging to eleven Reference Soil Groups (RSGs) classified as Acrisols, Andosols, Cambisols, Ferralsols, Fluvisols, Histosols, Regosols, Nitisols, Alisols, Luvisols, and Chernozems were used to develop pedotransfer functions for estimating CEC in these soils. Pedotransfer functions were developed using linear and non-linear regressions, while their reliability was evaluated using the error ratio (ER), the geometric mean error ratio (GMER), and root mean square error (RMSE). Predicted and measured CECs were equally compared using non-parametric tests to assess the deviations observed. To enhance the quality of the PTFs, stratifications of the data were done at various levels; (i) pooled data for all soil samples, (ii) within each RSG and (iii) within pedogenic soil horizons. Results indicate that PTFs obtained within the various reference soil groups were better than those obtained from pooled data due to the reduced variability of soil properties. Best results were obtained when soils were stratified according to pedogenic soil horizons. The best PTFs were observed in the most highly weathered soils, the Ferralsols, probably associated with their existence in the near steady state with greater stability and equilibrium among various soil properties. In most of the cases, OC proved to be the dominant soil property contributing to CEC. Results of this study will enable estimation of CEC from available soil survey data in the study area and similar environments, as well as the need to consider stratifications in terms of soil types and pedogenic soil horizons as reported in the literature to improve the reliability of PTFs.