Mapping Subsidence Susceptibility and Risks in the Rhineland Coalfields: Leveraging EGMS Data and Machine Learning

The Rhineland coalfields, a major lignite mining hub in Germany, are vital for national energy production and economic stability. However, the region faces persistent challenges from subsidence driven by natural and anthropogenic factors, resulting in structural damage such as cracks in walls and differential settlement. Historical leveling data since the 1990s reveal vertical deformations of up to 4 meters in mining-impacted areas, highlighting the interplay of mining activities, geological features, fault lines, and groundwater dynamics that influence ground stability.

This study investigates subsidence susceptibility and its potential risks to infrastructure by integrating ground motion data from the European Ground Motion Service (EGMS) with historical leveling datasets. Machine learning techniques, including Random Forest and Light Gradient Boosting Machine (LightGBM), were employed to develop a robust model for identifying areas at high risk of subsidence. Geological, lithological, groundwater, and elevation data were utilized to create susceptibility maps, pinpointing regions of significant concern.

High-risk areas identified in the mapping were further analyzed for their impact on infrastructure. Using EGMS data, angular distortion and horizontal strain were evaluated to understand structural vulnerabilities. Results indicated angular distortion (β) of 1/150 and horizontal strain (ε) reaching 0.01% along fault zones, presenting critical threats to structural integrity.

The findings underscore the value of susceptibility mapping and risk analysis for managing subsidence in mining regions. By offering insights into deformation patterns and classifying risk zones, the study provides policymakers with essential tools to implement mitigation strategies and promote sustainable development. These approaches are critical for balancing energy production with environmental and infrastructure protection in regions facing geological instability.