Structural and lithological controls on landfill site suitability in Tenth of Ramadan City, Egypt using remote sensing and machine learning

TL;DR

This study identifies suitable landfill sites in Tenth of Ramadan City, Egypt, using remote sensing and machine learning to analyze geology and urban expansion.

Contribution

The novel integration of SVM and PCA for lithological and structural analysis improves landfill site suitability assessment in arid urban environments.

Findings

SVM classification achieved 82.49% accuracy in lithological mapping.

PCA increased detected lineament lengths by 25% compared to legacy data.

16.2% of the study area was identified as highly suitable for landfill siting.

Abstract

The study area, located east of Tenth of Ramadan City, represents a vital urban center on the desert fringes of the eastern Nile Delta. In accordance with governmental regulations for landfill construction in new developments, this study aims to identify optimal landfill sites by integrating environmental and geotechnical considerations. Remote sensing and machine learning techniques (K-means and Support Vector Machine (SVM)) were applied to Landsat 5 TM imagery for lithological mapping. The SVM classification achieved an overall accuracy of 82.49% and a Kappa coefficient of 0.7235, providing a significant spatial refinement over the regional geological map by identifying localized lithological variations and recent urban expansion. Additionally, structural analysis using PCA yielded a 25% increase in detected lineament lengths 19.3 km compared to legacy data 15.4 km. Furthermore, ASTER GDEM data were used to generate a digital elevation model to visualize topographic variations and support structural analysis. The results revealed clear delineation of lithological units and highlighted zones of high lineament density, with dominant NE–SW (Syrian Arc) and NW–SE (Clysmic) structural trends influencing hydrogeological and geotechnical stability. Multi-criteria decision analysis (MCDA) was employed to map suitability, indicating that approximately 16.2% of the study area is highly suitable for landfill siting. These findings provide a practical framework for urban planners, providing a reliable decision-support framework for landfill selection based on integrated lithological and structural evidence.

Figures