Prediction of Ground Surface Deformation Induced by Earthquake on Urban Area Using Machine Learning

Earthquakes can inflict significant damage to structures and infrastructures. This paper presents a machine learning model predict ground surface deformation (GDS) induced by earthquake events. The data on historical GSD is extracted from radar product of Synthetic Aperture Radar (SAR) one-year over five magnitude earthquakes that occurred within 200 kilometers the Kota Padang Regency, West Sumatra. Building topology its footprint area, distance shoreline, elevation, coordinate were incorporated as main features in dataset. parameters taken USGS catalog. Four algorithms Neural Network (NN), Random Forest (RF), k-Nearest Neighbors (kNN), Gradient Boosting (GB) are applied. trained models predicted compared with measured SAR’s product. performances proposed evaluated terms statistical index. A new dataset event March 2022 used further test performance models. Overall, four have outstanding performance, coefficient determinant more than 0.9. kNN algorithm outperforms others delineating GSD. gave deficient prediction correlation 0.228 RF algorithm. Additional datasets unique will improve algorithms.