A Clustering Approach for the Analysis of InSAR Time Series: Application to the Bandung Basin (Indonesia)

Abstract

Interferometric Synthetic Aperture (InSAR) time series measurements are widely used to monitor a variety of processes including subsidence, landslides, and volcanic activity. However, interpreting large InSAR datasets can be difficult due to the volume of data generated, requiring sophisticated signal-processing techniques to extract meaningful information. We propose a novel framework for interpreting the large number of ground displacement measurements derived from InSAR time series techniques using a three-step process: (1) dimensionality reduction of the displacement time series from an InSAR data stack; (2) clustering of the reduced dataset; and (3) detecting and quantifying accelerations and decelerations of deforming areas using a change detection method. The displacement rates, spatial variation, and the spatio-temporal nature of displacement accelerations and decelerations are used to investigate the physical behaviour of the deforming ground by linking the timing and location of changes in displacement rates to potential causal and triggering factors. We tested the method over the Bandung Basin in Indonesia using Sentinel-1 data processed with the small baseline subset InSAR time series technique. The results showed widespread subsidence in the central basin with rates up to 18.7 cm/yr. We identified 12 main clusters of subsidence, of which three covering a total area of 22 km2 show accelerating subsidence, four clusters over 52 km2 show a linear trend, and five show decelerating subsidence over an area of 22 km2. This approach provides an objective way to monitor and interpret ground movements, and is a valuable tool for understanding the physical behaviour of large deforming areas.

Quantitative Assessment and Impact Analysis of Land Surface Deformation in Wuxi Based on PS-InSAR and GARCH Model

“…In Indonesia's Bandung Basin, subsidence impacted several towns and industrial areas within the central basin, with rates as high as 187 mm/year. This led to extensive damage to buildings and infrastructure, increasing safety risks [6]. This ongoing geological movement continues to afflict neighboring regions with natural disasters, including earthquakes and landslides [7].…”

Trend Classification of InSAR Displacement Time Series Using SAE–CNN

“…A method for monitoring surface deformation was proposed by van de Kerkhof et al [18] using t-distributed stochastic neighbor embedding (t-SNE) dimension reduction and density-based clustering non-parametric algorithm (DBSCAN) clustering. Festa et al [19] employed principal component analysis (PCA) to reduce data dimensionality, while Rygus et al [20] utilized the uniform manifold approximation and projection (UNWAP) method for the same purpose. Subsequently, both researchers applied clustering algorithms such as hierarchical density-based spatial clustering of applications with noise (HDBSCAN) and k-means to conduct cluster analysis on InSAR time-series data.…”