An integrated approach for risk assessment of land subsidence in Xi'an, China using optical and radar satellite images

Abstract

Land subsidence, induced by the over-exploitation of groundwater resources, is becoming a huge hazard in many large cities around the world. Risk assessment of the infrastructure under the threat of land subsidence plays a vital role in urban planning and design. In this study, an integrated approach that combines the land subsidence, the ground fissures, and the elements at risk is proposed. The ground surface deformation is studied with time-series Interferometric Synthetic Aperture Radar (InSAR), and the differential settlement is assessed by an index of angular distortion. Then, the elements that are potentially affected by the land subsidence are recognized through a land use classification analysis, which is based on a synergetic use of optical and radar images using an object-based approach. Finally, the risk due to land subsidence is assessed using a risk matrix that integrates the differential settlement, the ground fissures, and the land use classification results. To demonstrate the effectiveness of the proposed method, a risk assessment of the land subsidence in the City of Xi'an, China is conducted. Our results also illustrate the advantages of the synergetic land use classification over the pixel-based classification.