Nikolaj Dahmen

Last Name

Dahmen

First Name

Nikolaj

ORCID

0000-0002-9114-6747

Organisational unit

03476 - Giardini, Domenico (emeritus) / Giardini, Domenico (emeritus)

Show all metadata

Search Results

Publications 1 - 10 of 20

Near Surface Excitation of the Martian Ground as Measured by InSight
Pou, Laurent; Panning, Mark P.; Kedar, Sharon; et al. (2023)
AGU Fall Meeting Abstracts
Magnitude Scales for Marsquakes Calibrated from InSight Data
Böse, Maren; Stähler, Simon Christian; Deichmann, Nicholas; et al. (2021)
Bulletin of the Seismological Society of America
Revisiting Martian seismicity with deep learning-based denoising
Dahmen, Nikolaj; Clinton, John Francis; Stähler, Simon Christian; et al. (2024)
Geophysical Journal International
The analysis of seismic events recorded by NASA’s InSight seismometer remains challenging, given their commonly low magnitudes and large epicentral distances, and concurrently, strongly varying background noise. These factors collectively result in low signal-to-noise ratios (SNR) across most event recordings. We use a deep learning denoising approach to mitigate the noise contamination, aiming to enhance the data analysis and the seismic event catalogue. Our systematic tests demonstrate that denoising performs comparable to fine-tuned bandpass filtering at high SNRs, but clearly outperforms it at low SNRs with respect to accurate waveform and amplitude retrieval, as well as onset picking. We review the denoised waveform data of all 98 low-frequency events in the Marsquake Service catalogue version 14, and improve their location when possible through the identification of phase picks and backazimuths, while ensuring consistency with the raw data. We demonstrate that several event waveforms can be explained by marsquake doublets—two similarly strong quakes in spatio-temporal proximity that result in overlapping waveforms at InSight—and we locate them in Cerberus Fossae (CF). Additionally, we identify and investigate aftershocks and an event sequence consisting of numerous relatively high magnitude marsquakes occurring within hours at epicentral distances beyond CF. As a result of this review and interpretation, we extend the catalogue in event numbers (+8 per cent), in events with epicentral distances and magnitudes (+50 per cent), and events with backazimuths and a resulting full locations (+46 per cent), leading to a more comprehensive description of Martian seismicity.
MarsQuakeNet: A More Complete Marsquake Catalog Obtained by Deep Learning Techniques
Dahmen, Nikolaj; Clinton, John Francis; Meier, Men-Andrin; et al. (2022)
Journal of Geophysical Research: Planets
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.
The Marsquake catalogue from InSight, sols 0–478
Clinton, John Francis; Ceylan, Savas; van Driel, Martin; et al. (2021)
Physics of the Earth and Planetary Interiors
The InSight (Interior Exploration using Seismic Investigations, Geodesy and Heat Transport) mission began collecting high quality seismic data on Mars in February 2019. This manuscript documents the seismicity observed by SEIS, InSightâ€™s seismometer, from this time until the end of March 2020. Within the InSight project, the Marsquake Service (MQS) is responsible for prompt review of all seismic data collected by InSight, detection of events that are likely to be of seismic origin, and curation and release of seismic catalogues. In the first year of data collection, MQS have identified 465 seismic events that we interpret to be from regional and teleseismic marsquakes. Seismic events are grouped into 2 different event families: the low frequency family is dominated by energy at long period below 1 s, and the high frequency family primarily include energy at and above 2.4 Hz. Event magnitudes, from Mars-specific scales, range from 1.3 to 3.7. A third class of events with very short duration but high frequency bursts have been observed 712 times. These are likely associated with a local source driven by thermal stresses. This paper describes the data collected so far in the mission and the procedures under which MQS operates; summarises the content of the current MQS seismic catalogue; and presents the key features of the events we have observed so far, using the largest events as examples.
The Seismicity of Mars Recorded by InSight
Dahmen, Nikolaj (2024)
The InSight mission is the first Martian lander since the Viking missions in the late 1970s that has successfully deployed a seismometer in order to measure the seismicity of Mars. From early 2019 to late 2022, InSight collected a nearly-continuous high-quality seismic data set that demonstrates that Mars is seismically active. Over 1300 seismic event signals were cataloged by the Marsquake Service (MQS), which are associated both to tectonic quakes and meteorite impacts. Generally, recorded seismic events have low magnitudes and occur in large epicentral distances, which results in small ground motion at InSight. The very broadband (VBB) seismometer is capable of recording their small ground motions due to its extraordinarily high sensitivity, and the absence of microseismic noise on Mars. However, with seismicity with these low amplitude levels, the signals from the harsh local environment that are recorded on the VBB instrument become significant. In response to strong winds and large diurnal temperature changes, the seismic background noise varies strongly throughout the Martian day and contains various transient signals. Consequently, the seismic event recordings commonly have low signal-to-noise ratios (SNR), making their identification and any type of analysis challenging. This thesis is dedicated to advance our understanding of the InSight seismic data set and the recorded seismicity of Mars. To identify and interpret seismic event signals, it is essential to comprehend the seismic background noise. Specifically, we focus on various persistent signals that shape the frequency spectrum. While most of these spectral peaks are attributed to wind-excited mechanical resonances of the lander - lander modes - or artifacts produced in the measurement system, a broad resonance at 2.4 Hz stands out. Since this 2.4 Hz resonance is excited by seismic events, and might be used to infer subsurface properties through ambient noise analysis, it is critical to understand its origin. We investigate all spectral peaks in the continuous data streams, and in particular, evaluate their sensitivity to the local atmosphere and the seismic event energy. This analysis demonstrates the high correlation between lander modes and local winds, and the distinct nature of the 2.4 Hz resonance supporting its interpretation as a feature related to a ground structure. In addition to the classic seismic events located between 10’s to 1000’s of kilometers away from the InSight lander, the recordings also revealed a separate class of signals called super high frequency (SF) events. Their short duration without distinct phase arrivals, the high frequency content, and the absence of lander mode excitation suggests that SF events originate from InSight’s vicinity, but are not associated to the InSight lander itself. We detect these signals across the mission and show that SF events cluster in event families with similar waveforms and repetitive recurrence times. The events are mainly observed during the sunset period when the atmospheric temperatures are rapidly decreasing, indicating a thermal source reminiscent of thermal moonquakes. All other seismic events are identified and cataloged by MQS in manual review of the data, and classified by their frequency content into low and high frequency event families. Here we also show an automatic approach to detect signals of both event families across the mission. Due to the complex background noise, conventional signal detection methods do not perform well, and we follow a deep learning approach instead. Limited by the small labeled data set, we train the model on synthetic event recordings which are combined with recorded noise. The deep learning model is capable of effectively detecting event signals among large noise signals, enabling, for the first time, the automatic generation of a marsquake catalog. This catalog substantiates the MQS catalog and extends it by 60% with low SNR events. The extended catalog confirms seasonal trends of high frequency events, and demonstrates substantial interannual variations in their event rate. To address the limitations in event analysis arising from low SNRs, we also use the deep learning approach to mitigate all types of noise contamination. The model enables us to decompose the recorded signal into its event and noise components and produce denoised event waveforms. We systematically assess the denoising performance and demonstrate its advantage over commonly applied band pass filters, specifically in recovering waveform characteristics, peak amplitudes and signal onsets at low SNRs. We produce a denoised data set for all cataloged events and use it to revise the low frequency family catalog and improve event locations when possible. The denoised data set also allows for the identification and investigation of marsquake doublets, aftershocks, and a strong event sequence. Our revised catalog further extends the MQS catalog in event numbers (+8%), and contains substantially more events with epicentral distance (+50%), epicenter (+46%), and magnitude estimates (+50%). As an outlook, possible applications for the processing the high frequency family data set are outlined. In summary, this thesis offers a more comprehensive understanding of InSight’s background noise, and contributes to a more thorough seismicity catalog by detecting and investigating all types of events from low to super high frequency. As a result, more seismic events could be cataloged and located, providing a more complete and detailed perspective on the seismicity of Mars recorded by InSight.
S1222a – The Largest Marsquake Detected by InSight
Kawamura, Taichi; Clinton, John Francis; Zenhäusern, Géraldine; et al. (2023)
Geophysical Research Letters
NASA’s InSight has detected a large magnitude seismic event, labelled S1222a. The event has a moment magnitude of M(Ma)(W)4.7, with 5 times more seismic moment compared to the second largest even. The event is so large that features are clearly observed that were not seen in any previously detected events. In addition to body phases and Rayleigh waves, we also see Love waves, minor arc surface wave overtones, and multi-orbit surface waves. At long periods, the coda event exceeds 10 hours. The event locates close to the North-South dichotomy and outside the tectonically active Cerberus Fossae region. S1222a does not show any evident geological or tectonic features. The event is extremely rich in frequency content, extending from below 1/30 Hz up to 35 Hz. The event was classified as a broadband type event; we also observe coda decay and polarization similar to that of very high frequency type events.
Machine learning and marsquakes: a tool to predict atmospheric-seismic noise for the NASA InSight mission
Stott, Alexander E.; Garcia, Raphaël F.; Chédozeau, Armand; et al. (2023)
Geophysical Journal International
The SEIS (seismic experiment for the interior structure of Mars) experiment on the NASA InSight mission has catalogued hundreds of marsquakes so far. However, the detectability of these events is controlled by the weather which generates noise on the seismometer. This affects the catalogue on both diurnal and seasonal scales. We propose to use machine learning methods to fit the wind, pressure and temperature data to the seismic energy recorded in the 0.4-1 and 2.2-2.6 Hz bandwidths to examine low- (LF) and high-frequency (HF) seismic event categories respectively. We implement Gaussian process regression and neural network models for this task. This approach provides the relationship between the atmospheric state and seismic energy. The obtained seismic energy estimate is used to calculate signal-to-noise ratios (SNR) of marsquakes for multiple bandwidths. We can then demonstrate the presence of LF energy above the noise level during several events predominantly categorized as HF, suggesting a continuum in event spectra distribution across the marsquake types. We introduce an algorithm to detect marsquakes based on the subtraction of the predicted noise from the observed data. This algorithm finds 39 previously undetected marsquakes, with another 40 possible candidates. Furthermore, an analysis of the detection algorithm's variable threshold provides an empirical estimate of marsquake detectivity. This suggests that events producing the largest signal on the seismometer would be seen almost all the time, the median size signal event 45-50 per cent of the time and smallest signal events 5-20 per cent of the time.
Potential Pitfalls in the Analysis and Structural Interpretation of Seismic Data from the Mars InSight Mission
Kim, Doyeon; Davis, Paul; Lekic, Ved; et al. (2021)
Bulletin of the Seismological Society of America
The Seismic Experiment for Interior Structure (SEIS) of the InSight mission to Mars has been providing direct information on Martian interior structure and dynamics of that planet since it landed. Compared with seismic recordings on the Earth, ground-motion measurements acquired by SEIS on Mars are not only made under dramatically different ambient noise conditions, but also include idiosyncratic signals that arise from coupling between different InSight sensors and spacecraft components. This work is to synthesize what is known about these signal types, illustrate how they can manifest in waveforms and noise correlations, and present pitfalls in structural interpretations based on standard seismic analysis methods. We show that glitches (a type of prominent transient signal) can produce artifacts in ambient noise correlations. Sustained signals that vary in frequency, such as lander modes that are affected by variations in temperature and wind conditions over the course of the Martian sol, can also contaminate ambient noise results. Therefore, both types of signals have the potential to bias interpretation in terms of subsurface layering. We illustrate that signal processing in the presence of identified nonseismic signals must be informed by an understanding of the underlying physical processes in order for high-fidelity waveforms of ground motion to be extracted. Whereas the origins of the most idiosyncratic signals are well understood, the 2.4 Hz resonance remains debated, and the literature does not contain an explanation of its fine spectral structure. Even though the selection of idiosyncratic signal types discussed in this article may not be exhaustive, we provide guidance on the best practices for enhancing the robustness of structural interpretations.
Seasonal variations of subsurface seismic velocities monitored by the SEIS-InSight seismometer on Mars
Compaire, Nicolas; Margerin, Ludovic; Monnereau, Marc; et al. (2022)
Geophysical Journal International
The SEIS (Seismic Experiment for Interior Structure) seismometer deployed at the surface of Mars in the framework of the NASA-InSight (Interior Exploration using Seismic Investigations, Geodesy and Heat Transport) mission has been continuously recording the ground motion at Elysium Planitia for more than one martian year. In this work, we investigate the seasonal variation of the near-surface properties using both background vibrations and a particular class of high-frequency seismic events. We present measurements of relative velocity changes over one martian year and show that they can be modelled by a thermoelastic response of the Martian regolith. Several families of high-frequency seismic multiplets have been observed at various periods of the martian year. These events exhibit complex, repeatable waveforms with an emergent character and a coda that is likely composed of scattered waves. Taking advantage of these properties, we use coda wave interferometry (CWI) to measure relative traveltime changes as a function of the date of occurrence of the quakes. While in some families a stretching of the coda waveform is clearly observed, in other families we observe either no variation or a clear contraction of the waveform. These various behaviors correspond to different conditions of illumination at the InSight landing site, depending on the season. Measurements of velocity changes from the analysis of background vibrations above 5 Hz are consistent with the results from CWI. We identify a frequency band structure in the power spectral density (PSD) that can be tracked over hundreds of days. This band structure is the equivalent in the frequency domain of an autocorrelogram and can be efficiently used to measure relative traveltime changes as a function of frequency. We explain how the PSD analysis allows us to circumvent the contamination of the measurements by the Lander mode excitation which is inevitable in the time domain. The observed velocity changes can be adequately modelled by the thermoelastic response of the regolith to the time-dependent incident solar flux at the seasonal scale. In particular, the model captures the time delay between the surface temperature variations and the velocity changes in the subsurface. Our observations could serve as a basis for a joint inversion of the seismic and thermal properties in the first 20 m below InSight.

Publications 1 - 10 of 20

Nikolaj Dahmen

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results