Search

We present a technique to automatically classify the wave type of seismic phases that are recorded on a single six-component recording station (measuring both three components of translational and rotational ground motion) at the Earth's surface. We make use of the fact that each wave type leaves a unique 'fingerprint' in the six-component motion of the sensor (i.e. the motion is unique for each wave type). This fingerprint can be extracted by performing an eigenanalysis of the data covariance matrix, similar to conventional three-component polarization analysis. To assign a wave type to the fingerprint extracted from the data, we compare it to analytically derived six-component polarization models that are valid for pure-state plane wave arrivals. For efficient classification, we make use of the supervised machine learning method of support vector machines that is trained using data-independent, analytically derived six-component polarization models. This enables the rapid classification of seismic phases in a fully automated fashion, even for large data volumes, such as encountered in land-seismic exploration or ambient noise seismology. Once the wave-type is known, additional wave parameters (velocity, directionality and ellipticity) can be directly extracted from the six-component polarization states without the need to resort to expensive optimization algorithms. We illustrate the benefits of our approach on various real and synthetic data examples for applications such as automated phase picking, aliased ground-roll suppression in land-seismic exploration and the rapid close-to real-time extraction of surface wave dispersion curves from single-station recordings of ambient noise. Additionally, we argue that an initial step of wave type classification is necessary in order to successfully apply the common technique of extracting phase velocities from combined measurements of rotational and translational motion. [ABSTRACT FROM AUTHOR]

Interior exploration using Seismic Investigations, Geodesy and Heat Transport's (InSight) seismometer package Seismic Experiment for Interior Structure (SEIS) was placed on the surface of Mars at about 1.2 m distance from the thermal properties instrument Heat flow and Physical Properties Package (HP3) that includes a self‐hammering probe. Recording the hammering noise with SEIS provided a unique opportunity to estimate the seismic wave velocities of the shallow regolith at the landing site. However, the value of studying the seismic signals of the hammering was only realized after critical hardware decisions were already taken. Furthermore, the design and nominal operation of both SEIS and HP3 are nonideal for such high‐resolution seismic measurements. Therefore, a series of adaptations had to be implemented to operate the self‐hammering probe as a controlled seismic source and SEIS as a high‐frequency seismic receiver including the design of a high‐precision timing and an innovative high‐frequency sampling workflow. By interpreting the first‐arriving seismic waves as a P‐wave and identifying first‐arriving S‐waves by polarization analysis, we determined effective P‐ and S‐wave velocities of vP=119−21+45 ${v}_{P}=11{9}_{-21}^{+45}$ m/s and vS=63−7+11 ${v}_{S}=6{3}_{-7}^{+11}$ m/s, respectively, from around 2,000 hammer stroke recordings. These velocities likely represent bulk estimates for the uppermost several 10s of cm of regolith. An analysis of the P‐wave incidence angles provided an independent vP/vS ratio estimate of 1.84−0.35+0.89 $1.8{4}_{-0.35}^{+0.89}$ that compares well with the traveltime based estimate of 1.86−0.25+0.42 $1.8{6}_{-0.25}^{+0.42}$. The low seismic velocities are consistent with those observed for low‐density unconsolidated sands and are in agreement with estimates obtained by other methods. Plain Language Summary: In the framework of the NASA Interior exploration using Seismic Investigations, Geodesy and Heat Transport (InSight) mission, two scientific instruments were placed on the surface of Mars: A seismometer to detect signals from marsquakes and other sources generating seismic (elastic) waves and a self‐hammering temperature sensor that was designed to penetrate the Martian subsurface. The hammering of the temperature sensor generated vibrations that were measured by the seismometer and could be used to determine the elastic parameters of the shallow subsurface of Mars. We found low seismic velocities for the shallowest several tens of cm that are typical for low‐density loose sands. This information is important to further study the local geological setting at the InSight landing site and the shallow Martian subsurface in general. Key Points: Seismic signals from the Heat flow and Physical Properties Package mole provide a unique opportunity to study the shallow regolithFirst‐arrival traveltimes and P‐wave incidence angles constrain elastic parameter estimatesLow seismic velocities are consistent with unconsolidated low‐density sand [ABSTRACT FROM AUTHOR]

In December 2018, the NASA InSight lander successfully placed a seismometer on the surface of Mars. Alongside, a hammering device was deployed at the landing site that penetrated into the ground to attempt the first measurements of the planetary heat flow of Mars. The hammering of the heat probe generated repeated seismic signals that were registered by the seismometer and can potentially be used to image the shallow subsurface just below the lander. However, the broad frequency content of the seismic signals generated by the hammering extends beyond the Nyquist frequency governed by the seismometer's sampling rate of 100 samples per second. Here, we propose an algorithm to reconstruct the seismic signals beyond the classical sampling limits. We exploit the structure in the data due to thousands of repeated, only gradually varying hammering signals as the heat probe slowly penetrates into the ground. In addition, we make use of the fact that repeated hammering signals are sub‐sampled differently due to the unsynchronized timing between the hammer strikes and the seismometer recordings. This allows us to reconstruct signals beyond the classical Nyquist frequency limit by enforcing a sparsity constraint on the signal in a modified Radon transform domain. In addition, the proposed method reduces uncorrelated noise in the recorded data. Using both synthetic data and actual data recorded on Mars, we show how the proposed algorithm can be used to reconstruct the high‐frequency hammering signal at very high resolution. Key Points: Hammering of the InSight heat probe generates high‐frequency seismic signals that exceed the Nyquist frequency of the seismometerWe developed a new data acquisition and reconstruction workflow that allows for the recovery of the full‐bandwidth hammering signalsDuring hammering, we deliberately turned off the seismometer's anti‐aliasing filters and reconstructed the aliased signal using a sparseness‐promoting algorithm [ABSTRACT FROM AUTHOR]

A 6 degrees-of-freedom (6DoF) sensor, measuring three components of translational acceleration and three components of rotation rate, provides the full history of motion it is exposed to. In Earth sciences 6DoF sensors have shown great potential in exploring the interior of our planet and its seismic sources. In space sciences, apart from navigation, 6DoF sensors are, up to now, only rarely used to answer scientific questions. As a first step of establishing 6DoF motion sensing deeper into space sciences, this article describes novel scientific approaches based on 6DoF motion sensing with substantial potential for constraining the interior structure of planetary objects and asteroids. Therefore we estimate 6DoF-signal levels that originate from lander–surface interactions during landing and touchdown, from a body's rotational dynamics as well as from seismic ground motions. We discuss these signals for an exemplary set of target bodies including Dimorphos, Phobos, Europa, the Earth's Moon and Mars and compare those to self-noise levels of state-of-the-art sensors. [ABSTRACT FROM AUTHOR]

Spatial wavefield gradient data can be estimated from dense receiver arrays using a finite-difference approximation. Array-derived spatial wavefield gradient data to estimate, for example, strain and rotation components of the wavefield usually suffer from receiver perturbations, such as varying sensor coupling and/or differences in the instrument channel responses. We present a novel methodology to compute spatial seismic wavefield gradient data in the presence of receiver specific perturbation effects. We simultaneously invert the seismic wavefield recorded with an array consisting of closely spaced sensors for the spatial wavefield derivatives and station specific frequency-dependent calibration filters that remove the local receiver perturbations. Using synthetic and field data, we demonstrate that our scheme has the potential to significantly improve gradient estimation results for arrays with a low number of stations suitable, for instance, for land seismic exploration. The extracted calibration filters do not only correct for receiver perturbations but also partly compensate for the truncation error of the Taylor series expansion about the central (master) station of the array. The application of the calibration filters results in gradient estimates that are more accurate compared to conventional approaches. Multiple source recordings are used for reducing the trade-off between the extracted receiver perturbation filters and the spatial wavefield derivatives in the proposed inversion scheme. [ABSTRACT FROM AUTHOR]

We provide a six-component (6-C) polarization model for P-, SV-, SH-, Rayleigh-, and Lovewaves both inside an elastic medium as well as at the free surface. It is shown that single-station 6-C data comprised of three components of rotational motion and three components of translational motion provide the opportunity to unambiguously identify the wave type, propagation direction, and local P- and S-wave velocities at the receiver location by use of polarization analysis. To extract such information by conventional processing of three-component (3-C) translational data would require large and dense receiver arrays. The additional rotational components allow the extension of the rank of the coherency matrix used for polarization analysis. This enables us to accurately determine the wave type and wave parameters (propagation direction and velocity) of seismic phases, even if more than one wave is present in the analysis time window. This is not possible with standard, pure-translational 3-C recordings. In order to identify modes of vibration and to extract the accompanying wave parameters, we adapt the multiple signal classification algorithm (MUSIC). Due to the strong nonlinearity of the MUSIC estimator function, it can be used to detect the presence of specific wave types within the analysis time window at very high resolution. We show how the extracted wavefield properties can be used, in a fully automated way, to separate the wavefield into its different wave modes using only a single 6-C recording station. As an example, we apply the method to remove surface wave energy while preserving the underlying reflection signal and to suppress energy originating from undesired directions, such as side-scattered waves. [ABSTRACT FROM AUTHOR]

Interest in measuring displacement gradients, such as rotation and strain, is growing in many areas of geophysical research. This results in an urgent demand for reliable and field-deployable instruments measuring these quantities. In order to further establish a high-quality standard for rotation and strain measurements in seismology, we organized a comparative sensor test experiment that took place in November 2019 at the Geophysical Observatory of the Ludwig-Maximilians University Munich in Fürstenfeldbruck, Germany. More than 24 different sensors, including three-component and single-component broadband rotational seismometers, six-component strong-motion sensors and Rotaphone systems, as well as the large ring laser gyroscopes ROMY and a Distributed Acoustic Sensing system, were involved in addition to 14 classical broadband seismometers and a 160 channel, 4.5 Hz geophone chain. The experiment consisted of two parts: during the first part, the sensors were co-located in a huddle test recording self-noise and signals from small, nearby explosions. In a second part, the sensors were distributed into the field in various array configurations recording seismic signals that were generated by small amounts of explosive and a Vibroseis truck. This paper presents details on the experimental setup and a first sensor performance comparison focusing on sensor self-noise, signal-to-noise ratios, and waveform similarities for the rotation rate sensors. Most of the sensors show a high level of coherency and waveform similarity within a narrow frequency range between 10 Hz and 20 Hz for recordings from a nearby explosion signal. Sensor as well as experiment design are critically accessed revealing the great need for reliable reference sensors. [ABSTRACT FROM AUTHOR]

Recent progress in rotational sensor technology has made it possible to directly measure rotational ground-motion induced by seismic waves. When combined with conventional inertial seismometer recordings, the new sensors allow one to locally observe six degrees of freedom (6DOF) of ground-motion, composed of three orthogonal components of translational motion and three orthogonal components of rotational motion. The applications of such 6DOF measurements are manifold—ranging from wavefield characterization, separation, and reconstruction to the reduction of non-uniqueness in seismic inverse problems—and have the potential to revolutionize the way seismic data are acquired and processed. However, the seismological community has yet to embrace rotational ground-motion as a new observable. The aim of this paper is to give a high-level introduction into the field of 6DOF seismology using illustrative examples and to summarize recent progress made in this relatively young field. It is intended for readers with a general background in seismology. In order to illustrate the seismological value of rotational ground-motion data, we provide the first-ever 6DOF processing example of a teleseismic earthquake recorded on a multicomponent ring laser observatory and demonstrate how wave parameters (phase velocity, propagation direction, and ellipticity angle) and wave types of multiple phases can be automatically estimated using single-station 6DOF processing tools. Python codes to reproduce this processing example are provided in an accompanying Jupyter notebook. [ABSTRACT FROM AUTHOR]