Bridging traditional and conventional knowledge for soil classification in landslide-prone areas using exploratory factor analysis

Introduction Understanding soil movement and landslide susceptibility requires accurate soil classification. Traditional knowledge (TK) from ancient Indian texts describes soil color, texture, smell, and taste as key indicators, yet such qualitative insights remain underutilized in modern geotechnical assessments. Methods A questionnaire comprising 32 TK-based soil characteristics (color, texture, smell, taste) was developed and applied by 12 experts across 55 landslide-prone sites in Himachal Pradesh, India. Soil samples were also collected for laboratory-based geotechnical testing. Exploratory Factor Analyses (EFA) were conducted on TK-only, geotechnical-only, and combined datasets to identify latent soil categories and compare their explanatory power. Results Three principal soil categories earthy, coarse-grained, and fine-grained were identified. Retained factors explained 65.17%, 68.32%, and 71.46% of total variance for TK, geotechnical, and combined datasets, respectively. Expert discipline and educational background significantly influenced soil categorization, particularly in the fine-grained and coarse-grained groups. Discussion Integrating traditional and modern indicators enhances soil classification accuracy and provides a cost-effective framework for assessing landslide susceptibility in Himalayan regions. TK-based soil descriptors, when combined with quantitative geotechnical data, offer a holistic approach to slope stability analysis and risk mitigation.