MULTIPARAMETER LAND SUBSIDENCE VULNERABILITY ASSESSMENT THROUGH SATELLITE IMAGERY, GIS, AND SPATIAL DATA INTEGRATION

Abstract

Land subsidence is a substantial issue today, particularly in some regions where it has the ability to interrupt future development, halt the process, and even modify the development plan. In general, this study focuses on updating current knowledge on land subsidence and performing risk assessments using a case study in Semarang City, Central Java, where land subsidence is a serious problem. Several satellite imageries, such as Sentinel-1 and Sentinel-2, are used as databases in this work, with various target analysis and Geographic Information Systems (GIS) methods used to handle and alter the data. Sentinel-1's radar data, along with the displacement analysis approach of Differential Interferometry Synthetic Aperture Radar (DInSAR), is primarily used to provide a snapshot of the present state of land subsidence under this area in 2022. The result demonstrates the variation of vertical displacement values ranging from -7.7 to 6.65 cm with subsidence spread mostly in the northern region, using two datasets with 60-day intervals between January and March. Using a combination of the Normalize Difference Vegetation Index (NDVI) and the Normalize Difference Built-in Index (NDBI) to create Built-up Index (BU) data, the Sentinel-2 data with optical-based images was then used to map the human-made feature and expansions, notably buildings. Each piece of information considered multiparameter was afterwards overlapped to build a vulnerability map, while also considering the geological dataset and validated by ground verification. This finding will be presented to the section of the city that can help support the city's future growth in terms of its vulnerability to land subsidence dangers.