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

Abstract

An increase in volcanic thermal emissions can indicate subsurface and surface processes that precede, or coincide with, volcanic eruptions. Space-borne infrared sensors can detect hotspots—defined here as localized volcanic thermal emissions—in near-real-time. However, automatic hotspot detection systems are needed to efficiently analyze the large quantities of data produced. While hotspots have been automatically detected for over 20 years with simple thresholding algorithms, new computer vision technologies, such as convolutional neural networks (CNNs), can enable improved detection capabilities. Here we introduce HotLINK: the Hotspot Learning and Identification Network, a CNN trained to detect hotspots with a dataset of −3,800 satellite-based, Visible Infrared Imaging Radiometer Suite (VIIRS) images from Mount Veniaminof and Mount Cleveland volcanoes, Alaska. We find that our model achieves an accuracy of 96% (F1-score 0.92) when evaluated on −1,700 unseen images from the same volcanoes, and 95% (F1-score 0.67) when evaluated on −3,000 images from six additional Alaska volcanoes (Augustine Volcano, Bogoslof Island, Okmok Caldera, Pavlof Volcano, Redoubt Volcano, Shishaldin Volcano). In comparison with an existing threshold-based hotspot detection algorithm, MIROVA (Coppola et al., Geological Society, London, Special Publications, 2016, 426, 181–205), our model detects 22% more hotspots and produces 12% fewer false positives. Additional testing on −700 labeled Moderate Resolution Imaging Spectroradiometer (MODIS) images from Mount Veniaminof demonstrates that our model is applicable to this sensor’s data as well, achieving an accuracy of 98% (F1-score 0.95). We apply HotLINK to 10 years of VIIRS data and 22 years of MODIS data for the eight aforementioned Alaska volcanoes and calculate the radiative power of detected hotspots. From these time series we find that HotLINK accurately characterizes background and eruptive periods, similar to MIROVA, but also detects more subtle warming signals, potentially related to volcanic unrest. We identify three advantages to our model over its predecessors: 1) the ability to detect more subtle volcanic hotspots and produce fewer false positives, especially in daytime images; 2) probabilistic predictions provide a measure of detection confidence; and 3) its transferability, i.e., the successful application to multiple sensors and multiple volcanoes without the need for threshold tuning, suggesting the potential for global application.

Detection and Characterization of Seismic and Acoustic Signals at Pavlof Volcano, Alaska, Using Deep Learning

“…Figure 8 shows the multidisciplinary timeline spanning 1 January 2021-1 March 2023, which bounds the 5 August 2021-17 December 2022 eruption. It displays: our seismic and infrasonic VOISS-Net timelines (c, g), the 2-year spectrograms (b, f), the AVO Aviation Color Code (a, which is a qualitative proxy for the level of volcanic unrest), the median-filtered PS1A reduced displacement calculations (d), radiative power (e, in black) (Saunders-Shultz et al, 2024), SO 2 emissions determined by AVO (e, in red) (Lopez et al, 2021), and an explosion catalog comparison between VOISS-Net and one independently derived using the infrasound network and the Reverse Time Migration (RTM) method of Fee et al (2021) (h). Focusing on the period of background activity from January 2021 to May 2021, the VOISS-Net seismic model accurately returns infrequent seismic tremor classifications which agree well with AVO interpretations of intermittent tremor during this period.…”