MarsQuakeNet: A More Complete Marsquake Catalog Obtained by Deep Learning Techniques. Issue 11 (24th November 2022)

• Record Type: Journal Article
• Authors: Dahmen, Nikolaj L.; Clinton, John F.; Meier, Men‐Andrin; Stähler, Simon C.; Ceylan, Savas; Kim, Doyeon; Stott, Alexander E.; Giardini, Domenico
• Is Part Of: Journal of geophysical research. Volume 127:Issue 11(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-11-24
• Subjects: marsquakes detection -- deep learning -- seismology -- denoising -- seasonality -- NASA InSight; Planets -- Periodicals; Geophysics -- Periodicals; 559.9
• DOI: 10.1029/2022JE007503
• Languages: English
• ISSNs: 2169-9097

Abstract

Abstract: NASA's Interior Exploration using Seismic Investigations, Geodesy and Heat Transport (InSight) seismometer has been recording Martian seismicity since early 2019, and to date, over 1, 300 marsquakes have been cataloged by the Marsquake Service (MQS). Due to typically low signal‐to‐noise ratios (SNR) of marsquakes, their detection and analysis remain challenging: while event amplitudes are relatively low, the background noise has large diurnal and seasonal variations and contains various signals originating from the interactions of the local atmosphere with the lander and seismometer system. Since noise can resemble marsquakes in a number of ways, the use of conventional detection methods for catalog curation is limited. Instead, MQS finds events through manual data inspection. Here, we present MarsQuakeNet (MQNet), a deep convolutional neural network for the detection of marsquakes and the removal of noise contamination. Based on three‐component seismic data, MQNet predicts segmentation masks that identify and separate event and noise energy in time‐frequency domain. As the number of cataloged MQS events is small, we combine synthetic event waveforms with recorded noise to generate a training data set. We apply MQNet to the entire continuous 20 samples‐per‐second waveform data set available to date (>1, 000 Martian days), for automatic event detection and for retrieving denoised amplitudes. The algorithm reproduces all high quality, as well as majority of low quality events in the manual, carefully curated MQS catalog. Furthermore, MQNet detects ∼60% additional events that were previously unknown with mostly low SNR, that are verified in manual review. Our analysis on the event rate confirms seasonal trends and shows a substantial increase in the second Martian year. Plain Language Summary: Interior Exploration using Seismic Investigations, Geodesy and Heat Transport's seismometer on Mars has recorded over 1, 300 marsquakes since its full deployment in early 2019. Marsquakes are often weak compared to the seismic background noise, which makes their detection and analysis challenging. For this reason, the current event catalog relies on identifying events in manual data review, which can result in an inconsistent event catalog with weak events being missed and quality standards changing over time. In this study, we use a type of artificial neural network for the automatic detection of marsquakes and to separate even signal and background noise. Since these artificial neural networks usually require many examples to learn from but the number of known marsquakes is low, we generate synthetic marsquake examples to train the network. We run the detection algorithm across the mission and compare its performance to the manually compiled event catalog: the algorithm can also detects the majority of identified marsquakes and additionally finds many weaker, previously missing events, thereby extending the number of known marsquakes by ∼60%, from 1, 297 to 2, 079. Further, our results show substantial variations in event numbers throughout the mission. Key Points: Marsquakes recorded by Interior Exploration using Seismic Investigations, Geodesy and Heat Transport's seismometer are challenging to detect and analyze due to typically low signal‐to‐noise‐ratio We present MarsQuakeNet—a convolutional neural network for marsquake detection and denoising—trained on synthetic data Our catalog is consistent with existing manual catalog, extends it by 60% and confirms significant changes in event rate across Martian years