EOMF-01. Using integrated DInSAR+GNSS time series and synthetic surface deformation models for volcanic early warning systems

Abstract
Differential Interferometric Synthetic Aperture Radar (DInSAR) and Global Navigation Satellite System (GNSS) document comprehensive ground motions or ruptures at or near the Earth’s surface. These datasets may be independently applied to detect and analyze natural hazard phenomena, such as ground motion leading to major volcanic eruptions. Additionally, when combined into a single dataset, assimilated DInSAR + GNSS deformation results have improved accuracy [Corsa et al., 2022]. Here, we attempt to generate a synthetic dataset that best matches our integrated DInSAR + GNSS time series results over Hawaii from November 2015 through January 2022 by comparing multiple volcanic deformation source models, including a Mogi, diking and distributed source models. We discuss the benefits and limitations of such models and the applicability to other volcanic systems with varying parameters. We then demonstrate how the synthetic- and real- time series will be streamed through machine learning algorithms to help identify precursor motion leading to major eruptions, contributing towards improved early warning systems.