Volcanic Hot-Spot Detection Using SENTINEL-2: A Comparison with MODIS–MIROVA Thermal Data Series

Abstract

In the satellite thermal remote sensing, the new generation of sensors with high-spatial resolution SWIR data open the door to an improved constraining of thermal phenomena related to volcanic processes, with strong implications for monitoring applications. In this paper, we describe a new hot-spot detection algorithm developed for SENTINEL-2/MSI data that combines spectral indices on the SWIR bands 8a-11-12 (with a 20-meter resolution) with a spatial and statistical analysis on clusters of alerted pixels. The algorithm is able to detect hot-spot-contaminated pixels (S2Pix) in a wide range of environments and for several types of volcanic activities, showing high accuracy performances of about 1% and 94% in averaged omission and commission rates, respectively, underlining a strong reliability on a global scale. The S2-derived thermal trends, retrieved at eight key-case volcanoes, are then compared with the Volcanic Radiative Power (VRP) derived from MODIS (Moderate Resolution Imaging Spectroradiometer) and processed by the MIROVA (Middle InfraRed Observation of Volcanic Activity) system during an almost four-year-long period, January 2016 to October 2019. The presented data indicate an overall excellent correlation between the two thermal signals, enhancing the higher sensitivity of SENTINEL-2 to detect subtle, low-temperature thermal signals. Moreover, for each case we explore the specific relationship between S2Pix and VRP showing how different volcanic processes (i.e., lava flows, domes, lakes and open-vent activity) produce a distinct pattern in terms of size and intensity of the thermal anomaly. These promising results indicate how the algorithm here presented could be applicable for volcanic monitoring purposes and integrated into operational systems. Moreover, the combination of high-resolution (S2/MSI) and moderate-resolution (MODIS) thermal timeseries constitutes a breakthrough for future multi-sensor hot-spot detection systems, with increased monitoring capabilities that are useful for communities which interact with active volcanoes.

The Impact of Complex Volcanic Plumbing on the Nature of Seismicity in the Developing Magmatic Natron Rift, Tanzania

“…This seismic gap may represent the region of persistent partial melt (i.e., Hudson et al, 2017) that connects the volcano to its deeper plumbing system, and/or has heated nearby country rock resulting in ductile rock behavior for extended time periods in otherwise brittle crust (Castaldo et al, 2019;Parisio et al, 2019). Indeed, renewed and persistent volcanic activity at Oldoinyo Lengai since late 2018, as inferred from satellite data (Massimetti et al, 2020), require open conduits transporting volatiles and melt to the surface. From tomography and complementary receiver function studies (Plasman et al, 2017), the volcano is underlain by a low S-wave anomaly and high Vp/Vs ratios at 5 km depth, which continue well under the rift shoulder (Roecker et al, 2017) and may represent zones of partial melt below the volcano.…”

Constructive and Destructive Processes During the 2018–2019 Eruption Episode at Shiveluch Volcano, Kamchatka, Studied From Satellite and Aerial Data

“…MODIS infrared data have a 1 km spatial resolution and a high revisit frequency (up to four times daily) for volcanic monitoring purposes. Highspatial resolution thermal satellite datasets, including S2 and L8, have been investigated by applying a novel hotspot detection algorithm based on fixed ratios in the shortwave infrared (SWIR) regions with a contextual threshold derived from a statistical distribution of hotspot pixel clusters (Massimetti et al, 2020), as SWIR signals, in particular, record almost purely thermal emissions produced by hot emitting bodies (Blackett, 2017). Images were analyzed considering the SWIR top-of-atmosphere (TOA) reflectances in the ρ12 (2.19 µm), ρ11 (1.61 µm), and ρ8A (0.86 µm) bands for the S2 MSI and the ρ7 (2.11-2.29 µm), ρ6 (1.57-1.65 µm), and ρ5 (0.85-0.88 µm) bands for the L8 OLI.…”

“…Remote sensing is an efficient and safe method for monitoring the evolution of a volcano; effective observation techniques include thermal remote sensing (Coppola et al, 2020), synthetic aperture radar (SAR) analysis from spaceborne or airborne platforms (Schaefer et al, 2016), and multisensor approaches (Corradini et al, 2016). Freely available satellite data and smart data science allow the monitoring of deformation fields, thermal anomalies, and gas emissions in near real time (e.g., Valade et al, 2019;Massimetti et al, 2020). In this study, we use a multisensor remote sensing approach to investigate Shiveluch, one of the most active volcanoes in Kamchatka (Figures 1A,B).…”

Exploratory Analysis of Driving Force of Wildfires in Australia: An Application of Machine Learning within Google Earth Engine

“…All these locations were a part of the sample training dataset, comprising of 10,800 training points across the Australian mainland. The fire locations were visually verified from active fire alerts calculated from the Sentinel-2 data, as this can successfully recognize thermal trends [37]. The highlighted hot spots were calculated using short-wave infrared (SWIR-Band 12) with near infrared wavelengths (NIR-Band 5) and stress burned areas and a wider range of short-wave infrared wavelengths providing more purified hotspot detection [38].…”