High‐Resolution InSAR Reveals Localized Pre‐Eruptive Deformation Inside the Crater of Agung Volcano, Indonesia

Abstract

Localized deformation at volcanoes is caused by shallow processes including magma movement, pressurization of a hydrothermal system, or instability of the edifice, and is often observed shortly before or during an eruption (Cassidy et al., 2019). Thus, observing localized deformation is critical for understanding the final stages of magma ascent and eruption forecasting. However, sufficiently frequent and dense spatial observations are rarely available from ground-based instrument networks. High-resolution Interferometric Synthetic Aperture Radar (InSAR) can achieve <1 m spatial resolution and sub-weekly observations (e.g., COSMO-Skymed (CSK) (Italiana, 2007), TerraSAR-X (TSX) (Mittermayer et al., 2014), and ICEYE (Ignatenko et al., 2020)), making it an excellent tool for the detection, monitoring, and understanding of localized volcano deformation. For example,

Multi‐Temporal InSAR, GNSS and Seismic Measurements Reveal the Origin of the 2021 Vulcano Island (Italy) Unrest

“…Localized deformation at volcanoes, such as those measured at La Fossa Caldera between early August and late October 2021, may be very relevant because they are often indicators of the hydrothermal systems pressurization and can be precursors of phreatic explosions (Bemelmans et al., 2023; Kobayashi et al., 2018). Accordingly, the combined use of MT‐InSAR and GNSS data is valuable because it allows examining surface deformation and the time evolution of localized hydrothermal systems, along with the analysis of seismic signals.…”

The utility of TerraSAR-X, TanDEM-X, and PAZ for studying global volcanic activity: Successes, challenges, and future prospects

“…To measure ground deformation, orbital information provided by the space agencies and Digital Elevation Models (DEMs) are used to remove orbital and topographic effects. InSAR ground deformation measurements can be affected by changes in atmospheric water vapor that can be estimated using weather models; however, such models typically do not resolve features <10 km in scale, which are the focus of the high-spatialresolution data studied here [Bemelmans et al 2023]. Further, comparing interferograms with and without weather model corrections did not show significant differences at Masaya volcano, Nicaragua [Stephens et al 2020] and San Miguel volcano, El Salvador (Supplementary Figures page 136; Supplementary Figures file available in the data repository * ).…”

“…For all these types of measurements, high-resolution DEMs are necessary to correct topographic distortions and accurately georeference products without significant loss of resolution [e.g. Bemelmans et al 2023;Grémion et al 2023]. However, in most of the cases we did not have access to high-resolution DEM; therefore, we used open source 30-m resolution DEMs: the NASA Shuttle Radar Topography Mission (SRTM) DEM, acquired in 2000, and the Copernicus DEM [Farr et al 2007;European Space Agency].…”

“…Based on the limited literature available, dozens of volcanoes annually might exhibit changes in the high-resolution data that might be missed using other tech-niques [e.g. Bemelmans et al 2023]. For example, in a reconnaissance study, noted that >35 % of the potentially active volcanoes in Latin America had topographic changes resolvable from space.…”

“…An exciting advance in satellite observations of volcanoes is the development of radar and optical datasets with spatial resolutions of 3 m pixel −1 or lower with repeat intervals of a week or less (e.g. Planet, WorldView, Pléiades, Radarsat-2, TerraSAR-X, COSMO-SkyMed, Capella, ICEYE) that have allowed investigations of critical, smaller-scale features and their changes over time [Richter et al 2013;Eugenio et al 2014;Aldeghi et al 2019;Waythomas et al 2020;Dai et al 2022;Bemelmans et al 2023]. For example, high spatial resolution Synthetic Aperture Radar (SAR) images from RADARSAT-2 and TerraSAR-X satellites showed changes before an eruption at Merapi, Indonesia and were used along with other data to inform evacuation orders that saved thousands of lives in 2010 [Pallister et al 2013].…”