Skip to content

A framework for systematic cataloguing of volcano deformation source parameters from Sentinel-1 InSAR data

Ben Ireland 1, Juliet Biggs1, Nantheera Anantrasirichai2, Edna Warsame Dualeh1, Fabien Albino3

  • Affiliations: 1School of Earth Sciences, Wills Memorial Building, University of Bristol, UK 2Visual Information Laboratory, University of Bristol, UK 3ISTerre, Université Grenoble-Alpes, Grenoble, France 

  • Presentation type: Poster

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

  • Poster Board Number: 38

  • Programme No: 2.3.24

  • Theme 2 > Session 3

Abstract

Understanding global patterns of monitoring signals at volcanoes is important for inferring magmatic processes, forecasting of eruptions and unrest, and identifying similarly behaving systems. Satellite datasets provide global volcano deformation measurements, but existing global compilations rely on metadata parameters. These are not systematic and suffer from biases and uncertainties, due to methodological differences in data and modelling. Growing archives of Sentinel-1 InSAR data, systematically measuring volcano deformation globally, now provide an opportunity to overcome these limitations. Here, we present a framework to create systematic, data-driven volcano deformation catalogues from Sentinel-1 InSAR datacubes, tested using a dataset of 16 deforming volcanoes along the East African Rift System (EARS). First, we extract temporal parameters by fitting functions to deformation timeseries. Then, we develop methods to systematically determine source parameters building on the Bayesian non-linear inversion software, GBIS. Pre-processing steps, particularly downsampling and initial model geometries, were validated with synthetic examples. We model the signals using Mogi and Penny sources and compare fits against alternate source geometries using Akaike's Information Criterion. 14/16 EARS signals were automatically located correctly and 12/16 are fit with a Mogi source, with another two signals fit best with Okada sources. Variations in source parameters correspond to differences in inferred magmatic processes and tectonic setting. For simple sources, the output parameters are within ranges of previous modelling studies, but further work is needed to develop a systematic approach for datasets containing multiple signals or deformation episodes. Applied globally, such catalogues could aid eruption forecasting and analogue volcano identification.