AI-based unmixing of medium and source signatures from seismograms: ground freezing patterns

Abstract

Seismograms always result from mixing many sources and medium changes that are complex to disentangle, witnessing many physical phenomena within the Earth. With artificial intelligence (AI), we isolate the signature of surface freezing and thawing in continuous seismograms recorded in a noisy urban environment. We perform a hierarchical clustering of the seismograms and identify a pattern that correlates with ground frost periods. We further investigate the fingerprint of this pattern and use it to track the continuous medium change with high accuracy and resolution in time. Our method isolates the effect of the ground frost and describes how it affects the horizontal wavefield. Our findings show how AI-based strategies can help to identify and understand hidden patterns within seismic data caused either by medium or source changes.Plain Language Summary Seismic waves, emitted by a seismic source and then traveling through the Earth, contain crucial information about the sources and the medium. However, often multiple sources emit simultaneously, while the elastic properties of the medium can change over time. Unmixing and identifying the different processes in the seismograms is a complex task, which we try to solve with methods of artificial intelligence. In a completely data-driven fashion, we are able to mute the variation in the seismograms due to anthropogenic seismic sources and reveal a continuous medium change due to freezing and thawing. This approach could reveal hidden information in complex environments such as volcanoes, where many different source and medium processes occur.

Deciphering the Whisper of Volcanoes: Monitoring Velocity Changes at Kamchatka's Klyuchevskoy Group With Fluctuating Noise Fields

“…Thereby, they succeeded to discriminate time windows with different dominant sources. In another approach, they analyzed the output features of the ICA and found a feature that closely resembles the velocity change of the medium (Steinmann, Seydoux, & Campillo, 2022). While this approach seems to retrieve very accurate and highly resolving dv / v estimates for their dataset, one must have successfully retrieved a low‐resolution baseline estimate of dv / v using a conventional algorithm.…”

Recent advances in earthquake seismology using machine learning

“…Recently, more effective clustering has been attempted by automatically extracting features directly from unprocessed datasets, such as seismic waveforms or their spectrograms (Seydoux et al 2020;Yin et al 2022b). This approach is used for various purposes, including the detection/classification of earthquakes (Seydoux et al 2020;Steinmann et al 2022a), quarry blasts and rockfalls (Hammer et al 2013), volcanic earthquakes/tremors (Esposito et al 2008;Unglert and Jellinek 2017;Soubestre et al 2018;Ren et al 2020;, surface freezing and thawing (Steinmann et al 2022b), reflection wave (Ali et al 2022, and seismic waves radiated in glaciological processes (Jenkins et al 2021).…”