Small baselines techniques of time series InSAR to monitor and predict land subsidence causing flood vulnerability in Sidoarjo, Indonesia

Abstract

Floods are the most common sort of natural disaster that occurs when a body of water overflows and submerges on dry terrain. Several regions in Sidoarjo District, East Java such as Kedungbanteng and Banjarasri Village, have experienced floods with varying heights since 2018 due to heavy rainfall in the areas. According to the Meteorological, Climatological, and Geophysical Agency (BMKG), rainfall intensity in these areas reach 100 mm/day. Furthermore, the physical environment in both areas shows land subsidence that causes a change in the flow pattern, which triggers its occurrence. This research incorporated different small baseline techniques of InSAR time series analysis and used GPS measurement to evaluate and monitor the land subsidence in both areas. The reseachers used Sentinel-1 SAR data both from ascending and descending for the period between 2017 and 2021. The results showed significant subsidence value by up to −200 mm/year both in Kedungbanteng and Banjarasri Village, and GPS observation also confirmed approximately −3.2 cm of subsidence value during 2 months of sampling observations. Using the neural network analysis, the reseachers predicted the time series displacement for one year and showed that a continuous trend of deformation still existed.