Detection of Subtle Thermal Anomalies: Deep Learning Applied to the ASTER Global Volcano Dataset

Abstract

Twenty-one years of ASTER global thermal infrared (TIR) acquisitions provide a large amount of data for volcano monitoring. These data, with high spatial and spectral resolution, enable routine investigations of volcanoes in remote and inaccessible regions, including those with no ground-based monitoring. However, the dataset is too large to be manually analyzed on a global basis. Here, we systematically process the data over several volcanoes using a deep learning algorithm to automatically extract volcanic thermal anomalies. We explore the application of a Convolutional Neural Network (CNN), specifically UNET, to detect subtle to intense anomalies exploiting the spatial relationships of the volcanic features. We employ a supervised UNET network trained with the largest (1500) labeled dataset of ASTER TIR images from five different volcanoes, namely Etna (Italy), Popocatépetl (Mexico), Lascar (Chile), Fuego (Guatemala), and Kliuchevskoi (Russia). We show that our approach achieves high accuracy (93%) with excellent generalization capabilities. The effectiveness of our model for detecting the full range of thermal emission is shown for volcanoes with very different styles of activity and tested at Vulcano (Italy). The results demonstrate the potential applicability of the proposed approach to the development of automated thermal analysis systems at the globalscale using future TIR data such as the planned NASA SBG mission.

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

“…The overall results presented here indicate that the investigated unrest phase (September-December 2021), that is, inflation and seismicity, together with anomalies in gasses emissions (Aiuppa et al, 2022;Federico et al, 2023;Inguaggiato et al, 2022Inguaggiato et al, , 2023, is compatible with the expansion of the hydrothermal system located within the La Fossa Caldera. Thermal anomalies also show a similar trend, with a peak again in October 2021 (Coppola et al, 2022;Corradino et al, 2023;Pailot-Bonnétat et al, 2023). While the thermal anomalies have been interpreted as an increase in temperatures of the fumarolic field present in the crater area of the La Fossa cone, the geochemical anomalies are interpreted as an increase in the parameters of the hydrothermal system which in turn is enriched in a strong and deep input of fluids released from an underlying magma batch (Aiuppa et al, 2022;Inguaggiato et al, 2022).…”

“…Since September 2021, Vulcano has undergone a new phase of unrest. Fumarolic activity increased in the latter half of September, accompanied by intensified seismic activity (Coppola et al., 2022; Corradino et al., 2023; Federico et al., 2023; Inguaggiato et al., 2023). Notably, Very Long Period (VLP; frequency peak ∼ 0.3 Hz) seismic events were recorded for the first time, with the source area located at 950 ± 270 m below sea level, North of La Fossa cone (Federico et al., 2023).…”

Automatic identification and quantification of volcanic hotspots in Alaska using HotLINK: the hotspot learning and identification network

“…For especially hot surfaces (>950 K), the peak radiance emission is in the shortwave infrared (SWIR, 1.4-3 µm) part of the spectrum. The distinct features produced by hotspots in MIR and TIR bands have been exploited to automate their detection by different algorithms (Higgins and Harris, 1997;Pergola et al, 2004;Wright et al, 2004;Ganci et al, 2011;Coppola et al, 2016;Gouhier et al, 2016;Lombardo, 2016;Valade et al, 2019;Castaño et al, 2020;Genzano et al, 2020;Layana et al, 2020;Massimetti et al, 2020;Corradino et al, 2023;Ramsey et al, 2023).…”

Cascading Machine Learning to Monitor Volcanic Thermal Activity Using Orbital Infrared Data: From Detection to Quantitative Evaluation

“…DL is widely used in computer vision to train a model to automatically catch spatial and spectral attributes from images and recognize its content. We successfully applied DL algorithms to automatically recognize subtle to intense thermal anomalies exploiting the spatial relationships of the volcanic features [18].…”