Detection of slow-moving landslide using time-series SBAS and PS InSAR: case study of Palu, Indonesia

Abstract

The September 2018 Palu Earthquake triggered catastrophic flow-slides in areas such as Petobo, Jono Oge, and Balaroa, resulting in significant destruction and loss of life. While research has since focused on understanding the mechanisms behind these flow-slides, the post-failure activity of the landslides remains poorly understood, which is crucial for enhancing disaster management in Palu.

Effective landslide monitoring is essential to gather information on the dynamics of these events. Therefore, this study aims to detect slow-moving displacements in Petobo, Jono Oge, and Balaroa using two time-series InSAR approaches: Persistent Scatterer (PS) and Small Baseline Subset (SBAS). Additionally, the study evaluates the implementation of these methods. Sentinel-1 data were utilized to create time-series InSAR using the PS and SBAS approaches to measure velocity displacement over the landslide areas. The findings indicate that the PS approach is unsuitable due to pixel decorrelation, leading to information loss on the landslide body. In contrast, the SBAS approach performs better, providing valuable information that indicates slow-moving displacement in the Petobo landslide of up to 20mm/year. These findings are crucial for developing effective monitoring and disaster management strategies.

Furthermore, investigations into vertical slow-moving displacement were also discussed.