Mapping soil hydropedological response groups to analyse patterns of observed landslide occurrence at catchment scale in the Eastern Cape province of South Africa
Year: 2026
Type: article
Abstract: Landslides are widespread natural hazards, often driven by complex interactions between terrain, hydrology and soil properties. This study investigates the hydropedological characteristics of soils in relation to mapped landslide sites near Port St Johns and Lusikisiki in the Eastern Cape province of South Africa. The Soil-Landscape Inference Model (SoLIM) was used to produce a rule-based digital soil map of Hydropedological Response Groups (HRGs) representing soil–water dynamics across landscape positions. Field-based HRG classification was informed by 186 soil points across dominant land types, complemented by terrain and hydrological covariates. The HRGs map was validated with an overall classification accuracy of 68% and Cohen’s Kappa coefficient of 0.60. Overlay analysis showed that landslide occurrences were predominantly associated with recharge (shallow) and recharge (slow) soil types found on mid- to upper-slope positions with moderate to steep slopes. The study demonstrates that rule-based HRG mapping, when grounded in local soil and terrain understanding, is effective in delineating zones of heightened landslide occurrence. Moreover, it provides a framework for integrating soil hydrology into regional landslide hazard assessments by classifying soils into HRGs. Limitations include potential misclassification due to limited subsoil information and spatial resolution constraints.
Authors: Jaco Kotzé, Jay Le Roux, Johan van Tol
Institution: University of the Free State
Language: English
Subfield: Management, Monitoring, Policy and Law
Field: Environmental Science
Domain: Physical Sciences
SDG: Life on land
Open Access status: hybrid
Awards: G2261091802, G2880324134, G9001368419
OpenAlex ID: W7226030664
