Skip to content

Characterisation of activity at Semeru volcano using high resolution radar and optical imagery

Pierre Bouygues, Fabien Albino, Virginie Pinel, Diego Cusicanqui

  • Affiliations: Univ. Grenoble Alpes, Univ. Savoie Mont Blanc, CNRS, IRD, Univ. Gustave Eiffel, ISTerre, 38000 Grenoble, France

  • Presentation type: Poster

  • Presentation time: Friday 16:30 - 18:00, Room Poster Hall

  • Poster Board Number: 28

  • Programme No: 2.3.14

  • Theme 2 > Session 3

Abstract

The Semeru volcano, located in eastern Java, Indonesia, reactivated in December 2021 following by the collapse of a dome that had been growing since 2009. Monitoring its summit surface evolution is essential, but can hardly rely on ground-based measurements only, because of the edifice's inaccessibility and high sensors costs. Synthetic Aperture Radar interferometry (InSAR) overcomes these challenges by enabling high-resolution mapping of ground deformation and topographic changes. Studying volcano deformation at Semeru using InSAR is challenging. The steepness of the edifice, the tropical climate and its evolution through volcanic processes create significant noise in the InSAR phase. These effects can mask low-amplitude deformation signals, requiring corrections. To improve detection of low-amplitude displacement signals, we combine high-resolution SAR acquisitions from the TerraSAR-X, TanDEM-X, and PAZ missions, with data for the period 2022-2024. First, high-resolution DEMs are produced from eight TanDEM-X bistatic acquisitions and a Pleiades image. It enables to characterize dome evolution from 2015 to July 2021. During this period, the dome evolves heterogeneously, increasing 50 meters in height and reaching 1.35 million cubic meters in volume. Lava flows and pyroclastic deposits accumulate with a thickness up to 75 meters, filling trenches and creating new deposition channels eastward. Then, for TSX and PAZ repeat pass interferometry, we use the produced DEM to correct topographic fringes and we mitigate atmospheric delays using the ERA-5 weather model and GNSS dataset. By combining those three datasets, we expect to reduce noise in InSAR time series and therefore improve the detection of low amplitude signals.