Summary The costly power requirements of delivering seismic data back to Earth from planetary missions requires the development of algorithms for lander-side signal analysis for telemetry prioritization. This is difficult to explicitly program, especially if no prior seismic data are available from the planetary body. Deep learning computer vision has been used to generalize seismic signals on Earth for earthquake early warning problems but such techniques have not yet been expanded to planetary science. We demonstrate that Convolutional Neural Networks can be used to accurately catalog planetary seismicity without local training data by building binary noise/signal classifiers from a single Earth seismic station and applying the models to moonquakes from the Apollo Passive Seismic Experiment (PSE) and the Lunar Seismic Profiling Experiment (LSPE). In order to promote generality and reduce the amount of training data, the algorithms use spectral images instead of time-series. Two- to five-layer convolution models are tested against a subset of 200 Grade-A events from the PSE and obtained station accuracy averages of 89–96%. As the model was applied to an hour trace of data (30 minutes before and after the Grade-A event), additional detections besides the Grade-A event are unavoidable. In order to comprehensively address algorithm accuracy, additional seismic detections corresponding to valid signals such as other moonquakes or multiples within a particularly long event needed to be compared with those caused by algorithm error or instrument glitches. We developed an ”extra-arrival accuracy” metric to quantify how many of the additional detections were due to valid seismic events and used it to select the three-layer model as the best fit. The three-layer model was applied to the entire LSPE record and matched the lunar day-night cycle driving thermal moonquake generation with fewer false detections than a recent study using Hidden Markov Models. We anticipate that these methods for lander-side signal detection can be easily expanded to non-seismological data and may provide even stronger results when supplemented with synthetic training data.
The seismic noise recorded by the Interior Exploration using Seismic Investigations, Geodesy, and Heat Transport (InSight) seismometer (Seismic Experiment for Interior Structure [SEIS]) has a strong daily quasi-periodicity and numerous transient microevents, associated mostly with an active Martian environment with wind bursts, pressure drops, in addition to thermally induced lander and instrument cracks. That noise is far from the Earth’s microseismic noise. Quantifying the importance of nonstochasticity and identifying these microevents is mandatory for improving continuous data quality and noise analysis techniques, including autocorrelation. Cataloging these events has so far been made with specific algorithms and operator’s visual inspection. We investigate here the continuous data with an unsupervised deep-learning approach built on a deep scattering network. This leads to the successful detection and clustering of these microevents as well as better determination of daily cycles associated with changes in the intensity and color of the background noise. We first provide a description of our approach, and then present the learned clusters followed by a study of their origin and associated physical phenomena. We show that the clustering is robust over several Martian days, showing distinct types of glitches that repeat at a rate of several tens per sol with stable time differences. We show that the clustering and detection efficiency for pressure drops and glitches is comparable to or better than manual or targeted detection techniques proposed to date, noticeably with an unsupervised approach. Finally, we discuss the origin of other clusters found, especially glitch sequences with stable time offsets that might generate artifacts in autocorrelation analyses. We conclude with presenting the potential of unsupervised learning for long-term space mission operations, in particular, for geophysical and environmental observatories.
SUMMARY Fluid injection for geothermal production has the potential to produce subsidence and microseismicity that can incur heavy financial cost or hazard. Due to this, novel ways to monitor subsurface deformation to supplement existing methods are highly sought after. We use seismic ambient noise to obtain time-dependent measurements of shear velocity within the geothermal reservoirs of Rotokawa and Ngatamariki, two producing geothermal fields in the Taupō Volcanic Zone located in the central North Island of New Zealand and operated by Mercury Energy. We investigate the relationship between shear wave velocity changes and geothermal injection by selecting time periods at the fields when injection and production volumes were significantly altered: 2009–2010 at Rotokawa, when geothermal injection was quadrupled due to the start-up of a new power station, and 2012–2013 at Ngatamariki, the beginning of geothermal injection for electricity production at that field. Shear wave velocity changes are computed from the ambient noise cross-correlation coda using the Moving Window Cross-Spectral (MWCS) technique, with a reference stack encompassing all data prior to the change in injection rate and moving stacks of 10–50 d. Gradual positive and negative shear velocity changes with a periodicity of approximately 12 months were observed at both sites, with maximum amplitude of 0.06 ± 0.04 and –0.08 ± 0.03 per cent at Rotokawa and 0.07 ± 0.03 and –0.06 ± 0.02 per cent at Ngatamariki. We hypothesize that these changes are due to seasonal rainfall, as seismic velocities computed by ambient noise are sensitive to the filling and emptying of near-surface pore space. In addition to these gradual responses, we found several sharp negative changes in velocity that reach minimum values over a few days and then gradually equilibrate to prior values over a few weeks. The amplitude of these responses is between –0.03 and –0.07 per cent and coincides with regional and local earthquakes. We hypothesize that these responses are primarily produced by the creation of new fractures, the same mechanism that produces gradual groundwater level decreases at regional distances from earthquake epicentres. We analyse a periodic signal within the time-delay measurements and determine that it is at least in part caused by the MWCS window size smoothing the cross-coherence of the ambient seismic signal. We do not observe shear wave velocity changes coinciding with geothermal injection, which may suggest that the signal has lower amplitude compared to the seasonal and seismic responses. We use bandstop filters and polynomial curve fitting to remove the contribution of the seasonal signal, but see no evidence of a shear wave velocity response due to geothermal fluid injection.
We have determined subsurface structure using the refraction microtremor (ReMi) method at the Ngatamariki geothermal field, Central North Island, New Zealand. The local geology is such that refraction and reflection studies are hindered by energy scattering and attenuation in the near-surface layers. The ReMi method uses surface waves from ambient noise and active sources to determine S-wave velocities in the shallow subsurface. We have deployed two lines of 72-channel, 10 Hz vertical geophones with 10 m spacing, and we were able to model near-surface S-wave velocity to depths of 57–93 m for 2D profiles and as much as 165 m for 1D profiles. Shear-velocity anomalies were detected on one line that were spatially correlated with a fault. The location of the fault was previously inferred from stratigraphic offset in the geothermal wells, suggesting that the ReMi method can provide important constraints on near-surface geology in noisy geothermal settings.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.