Abstract
Despite termite-induced soil mixing, summarizing termite-affected soil horizons is difficult, while the lack of accurate information on the pedogenic processes featured by termite bioturbation, topography, and land use limits an effort to address land degradation. A study was therefore carried out to quantitatively classify the soils and describe them based on rangeland uses. Based on cluster analysis, five representative soil profiles were studied at different topographical positions. Soil samples were collected from mounds and adjacent soils under enclosure, cultivated, and open-grazing land at the summit and foot slope positions. Agglomerative clustering showed low Ca(2+), CEC, pH, and Mg(2+) that described cambic horizons formed Cambisols at the summit and back slope. Eluviation-illuviation processes formed Luvisols on the toe slope and foot slope, whereas clay and high CEC described argic horizons. High Ca(2+), CEC, pH, and Mg(2+) described calcic horizons that formed Calcisols on the bottom slope. Divisive clustering showed that soil properties varied slightly between Cambisols and Luvisols at different topographies. However, the Luvisols on the toe slope were differentiated from the soil on the foot slope by predominant pedogenetic clay formation and a distinctly increased CEC. Calcisols are placed in other clusters due to their distinct properties. Agglomerative clustering reflected pedogenic processes and differentiated diagnostic horizons, while divisive clustering matched WRB classification. The results of this study also showed that termite-mediated soil properties were dictated by rangeland use, and pedogenesis was more noticeable on open-grazing land than on enclosure or cultivated land.