scholarly journals Improving cross-correlations of ambient noise using an rms-ratio selection stacking method

2020 ◽  
Vol 222 (2) ◽  
pp. 989-1002
Author(s):  
Jinyun Xie ◽  
Yingjie Yang ◽  
Yinhe Luo
Keyword(s):  
Short Duration ◽  
Ambient Noise ◽  
Signal To Noise Ratio ◽  
Ambient Noise Tomography ◽  
Noise Sources ◽  
Time Duration ◽  
Crust And Upper Mantle ◽  
Phase Velocities ◽  
Noise Data ◽  
Cross Correlations

SUMMARY Stacking of ambient noise correlations is a crucial step to extract empirical Green's functions (EGFs) between station pairs. The traditional method is to linearly stack all short-duration cross-correlation functions (CCFs) over a long period of time to obtain final stacks. It requires at least several months of ambient noise data to obtain reliable phase velocities at periods of several to tens of seconds from CCFs. In this study, we develop a new stacking method named root-mean-square ratio selection stacking (RMSR_SS) to reduce the time duration required for the recovery of EGFs from ambient noise. In our RMSR_SS method, rather than stacking all short-duration CCFs, we first judge if each of the short-duration CCF constructively contributes to the recovery of EGFs or not. Then, we only stack those CCFs which constructively contribute to the convergence of EGFs. By applying our method to synthetic noise data, we demonstrate how our method works in enhancing the signal-to-noise ratio of CCFs by rejecting noise sources which do not positively contribute to the recovery of EGFs. Then, we apply our method to real noise data recorded in western USA. We show that reliable and accurate phase velocities can be measured from 15-d long ambient noise data using our RMSR_SS method. By applying our method to ambient noise tomography (ANT), we can reduce the deployment duration of seismic stations from several months or years to a few tens of days, significantly improving the efficiency of ANT in imaging crust and upper-mantle structures.

Evaluating Uncertainties of Phase Velocity Measurements from Cross-Correlations of Ambient Seismic Noise

2020 ◽  
Vol 91 (3) ◽  
pp. 1717-1729
Author(s):  
Yinhe Luo ◽  
Yingjie Yang ◽  
Jinyun Xie ◽  
Xiaozhou Yang ◽  
Fengru Ren ◽  
...  
Keyword(s):  
Phase Velocity ◽  
Ambient Noise ◽  
Signal To Noise Ratio ◽  
The United States ◽  
Velocity Measurements ◽  
Phase Velocities ◽  
Phase Velocity Dispersion ◽  
Noise Data ◽  
Cross Correlations ◽  
Fitting In

Abstract Ambient-noise tomography (ANT) has become a well-established method to image the crust and uppermost mantle structures in the past 15 yr. Having a good estimate of uncertainties of phase velocity dispersion measurements in ANT is critical as they can guide the level of data fitting in tomography. However, to date, there are still no systemic studies to evaluate these uncertainties. In this study, we obtain cross correlations with different stacking durations from 17 yr of ambient-noise data recorded at 120 stations in the United States. We analyze the variations of signal-to-noise ratio (SNR) and phase velocities of cross correlations. We find that the uncertainties of phase velocities are affected by SNRs, interstation distances, and stacking durations. However, none of those three variables can be solely used as a proxy to estimate the uncertainties of phase velocity measurements. Based on our analysis, we graphically present empirical relations of uncertainties of phase velocity measurements as a function of SNR, interstation distance, and stacking duration. These relations can be employed as a guide to estimate phase velocity uncertainties in applications of ANT, assisting in evaluating the reliability of resulting models from ANT.

scholarly journals Upper-plate structure in Ecuador coincident with the subduction of the Carnegie Ridge and the southern extent of large mega-thrust earthquakes

2019 ◽  
Vol 220 (3) ◽  
pp. 1965-1977 ◽  
Author(s):  
Colton Lynner ◽  
Clinton Koch ◽  
Susan L Beck ◽  
Anne Meltzer ◽  
Lillian Soto-Cordero ◽  
...  
Keyword(s):  
Ambient Noise ◽  
Velocity Structure ◽  
Sedimentary Basins ◽  
Past Century ◽  
Ambient Noise Tomography ◽  
Plate Interface ◽  
Crust And Upper Mantle ◽  
Southern Boundary ◽  
Plate Structure ◽  
Velocity Anomaly

SUMMARY The Ecuadorian convergent margin has experienced many large mega-thrust earthquakes in the past century, beginning with a 1906 event that propagated along as much as 500 km of the plate interface. Many subsections of the 1906 rupture area have subsequently produced Mw ≥ 7.7 events, culminating in the 16 April 2016, Mw 7.8 Pedernales earthquake. Interestingly, no large historic events Mw ≥ 7.7 appear to have propagated southward of ∼1°S, which coincides with the subduction of the Carnegie Ridge. We combine data from temporary seismic stations deployed following the Pedernales earthquake with data recorded by the permanent stations of the Ecuadorian national seismic network to discern the velocity structure of the Ecuadorian forearc and Cordillera using ambient noise tomography. Ambient noise tomography extracts Vsv information from the ambient noise wavefield and provides detailed constraints on velocity structures in the crust and upper mantle. In the upper 10 km of the Ecuadorian forearc, we see evidence of the deepest portions of the sedimentary basins in the region, the Progreso and Manabí basins. At depths below 30 km, we observe a sharp delineation between accreted fast forearc terranes and the thick crust of the Ecuadorian Andes. At depths ∼20 km, we see a strong fast velocity anomaly that coincides with the subducting Carnegie Ridge as well as the southern boundary of large mega-thrust earthquakes. Our observations raise the possibility that upper-plate structure, in addition to the subducting Carnegie Ridge, plays a role in the large event segmentation seen along the Ecuadorian margin.

scholarly journals Crustal structures beneath the Eastern and Southern Alps from ambient noise tomography

2020 ◽  
Author(s):  
Ehsan Qorbani ◽  
Dimitri Zigone ◽  
Mark R. Handy ◽  
Götz Bokelmann ◽  
Keyword(s):  
High Resolution ◽  
Crustal Structure ◽  
Ambient Noise ◽  
Eastern Alps ◽  
Southern Alps ◽  
Ambient Noise Tomography ◽  
Tauern Window ◽  
Velocity Models ◽  
Cross Correlations ◽  
Velocity Anomalies

Abstract. We study the crustal structure under the Eastern and Southern Alps using ambient noise tomography. We use cross-correlations of ambient seismic noise between pairs of 71 permanent stations and 19 stations of the EASI profile to derive new high-resolution 3-D shear-velocity models for the crust. Continuous records from 2014 and 2015 are cross-correlated to estimate Green's functions of Rayleigh and Love waves propagating between the station pairs. Group velocities extracted from the cross-correlations are inverted to obtain isotropic 3-D Rayleigh and Love-wave shear-wave velocity models. Our high resolution models image several velocity anomalies and contrasts and reveal details of the crustal structure. Velocity variations at short periods correlate very closely with the lithologies of tectonic units at the surface and projected to depth. Low-velocity zones, associated with the Po and Molasse sedimentary basins, are imaged well to the south and north of the Alps, respectively. We find large high-velocity zones associated with the crystalline basement that forms the core of the Tauern Window. Small-scale velocity anomalies are also aligned with geological units such as the Ötztal and the Gurktal nappes of the Austroalpine nappes. Clear velocity contrasts in the Tauern Window along vertical cross-sections of the velocity model show the depth extent of the tectonic units and their bounding faults. A mid-crustal velocity contrast is interpreted as a manifestation of intracrustal decoupling in the Eastern Alps and decoupling between the Southern and Eastern Alps.

An Adjoint Technique for Estimation of Interstation Phase and Group Dispersion from Ambient Noise Cross Correlations

2019 ◽  
Vol 109 (5) ◽  
pp. 1716-1728
Author(s):  
Rhys Hawkins ◽  
Malcolm Sambridge
Keyword(s):  
Ambient Noise ◽  
Signal To Noise Ratio ◽  
Spectral Element ◽  
Second Step ◽  
Earth Model ◽  
Rayleigh Surface Waves ◽  
Radial Anisotropy ◽  
Cross Correlations ◽  
Noise Correlation Function ◽  
Tomographic Inversions

Abstract A method of extracting group and phase velocity dispersions jointly for Love‐ and Rayleigh‐wave observations is presented. This method uses a spectral element representation of a path average Earth model parameterized with density, shear‐wave velocity, radial anisotropy, and VP/VS ratio. An initial dispersion curve is automatically estimated using a heuristic approach to prevent misidentification of the phase. A second step then more accurately fits the observed noise correlation function (NCF) between interstation pairs in the frequency domain. For good quality cross correlations with reasonable signal‐to‐noise ratio, we are able to very accurately fit the spectrum of NCFs and hence obtain reliable estimates of both phase and group velocity jointly for Love and Rayleigh surface waves. In addition, we also show how uncertainties can be estimated with linearized approximations from the Jacobians and subsequently used in tomographic inversions.

Upper crustal structure at the KTB drilling site from ambient noise tomography

2020 ◽  
Author(s):  
Ehsan Qorbani ◽  
Irene Bianchi ◽  
Petr Kolínský ◽  
Dimitri Zigone ◽  
Götz Bokelmann
Keyword(s):  
Ambient Noise ◽  
Velocity Model ◽  
Ambient Noise Tomography ◽  
Data Set ◽  
Crustal Rocks ◽  
Velocity Models ◽  
Path Coverage ◽  
Passive Seismic ◽  
Cross Correlations ◽  
Drilling Site

<p>In this study, we show results from ambient noise tomography at the KTB drilling site, Germany. The Continental Deep Drilling Project, or ‘Kontinentales Tiefbohrprogramm der Bundesrepublik Deutschland’ (KTB) is at the northwestern edge of the Bohemian Massif and is located on the Variscan belt of Europe. During the KTB project crustal rocks have been drilled down to 9 km depth and several active seismic studies have been performed in the surrounding. The KTB area therefore presents an ideal test area for testing and verifying the potential resolution of passive seismic techniques. The aim of this study is to present a new shear-wave velocity model of the area while comparing the results to the previous velocity models and hints for anisotropy depicted by former passive and active seismological studies. We use a unique data set composed of two years of continuous data recorded at nine 3-component temporary stations installed from July 2012 to July 2014 located on top and vicinity of the drilling site. Moreover, we included a number of permanent stations in the region in order to improve the path coverage and density. Cross correlations of ambient noise are computed between the station pairs using all possible combination of three-component data. Dispersion curves of surface waves are extracted and are then inverted to obtain group velocity maps. We present here a new velocity model of the upper crust of the area, which shows velocity variations at short scales that correlate well with geology in the region.</p>

Beamforming Reliability of DAS Ambient Noise Data and Wave Modes Identification

2021 ◽  
Author(s):  
Yumin Zhao ◽  
Yunyue Elita Li
Keyword(s):  
Ambient Noise ◽  
Amplitude Ratio ◽  
Propagation Direction ◽  
Love Waves ◽  
Motor Vehicles ◽  
Noise Sources ◽  
Rayleigh And Love Waves ◽  
Noise Data ◽  
Campus Drive ◽  
Wave Modes

<p>Ambient noise generated by the anthropological activities in the urban environments may contain both Rayleigh and Love waves. Due to the differences in the physics of Rayleigh and Love waves, a pre-knowledge of the wave modes in the cross-correlogram is essential for an accurate inversion of the subsurface velocity model. Several studies (Martin and Biondi, 2017; Martin et al., 2017; Luo et al., 2020) demonstrated that only Rayleigh waves can be extracted by cross-correlation if the virtual source is colinear with the DAS array based on the assumption that the ambient noise sources are random and uniformly distributed. However, in realistic cases, ambient noise sources may come from a certain direction (e.g., Dou et al., 2017; Zhang et al., 2019). Moreover, the source propagation direction should be resolved and used to correct the apparent dispersion curves. Zhao et al. (2020) and van den Ende et al. (2020) proposed that beamforming results are not always reliable due to the measurements of DAS.</p><p>Based on the synthetic DAS ambient noise data recorded by a near “L” shape array (Source-West corner of the Stanford DAS-1 array), we prove that beamforming can resolve the source direction when the ambient sources are mainly coming from one direction. Two important processing procedures are that: check the polarity in the data and apply polarity flip on one part of the data; apply amplitude normalization on the data if strong amplitude difference exits in the data. Based on the source direction, the coordinate of the DAS array, and amplitude ratio of the data recorded by the two segments of the DAS array, we propose an inversion method to calculate the amplitude ratio of the Rayleigh and Love waves generated by the ambient sources.</p><p>We apply the method to two 100-second DAS ambient noise data recorded by the Stanford DAS-1 array. We first resolve the source propagation direction from the two data. The results indicate that the ambient noise in the data were mainly generated by the motor vehicles running on the Campus Drive in the northwest of the array. Then we invert for the Rayleigh and Love waves amplitude ratio using the proposed method. The ratios for the two data are 0.2 and 0.13, respectively. The results suggest that the ambient noise generated by motor vehicles running on the northwest corner of the Campus Drive mainly contain Love waves.</p>

scholarly journals Crust and upper mantle structure of the Ligurian Sea revealed by ambient noise tomography using ocean bottom seismometer data

2020 ◽  
Author(s):  
Felix Noah Wolf ◽  
Dietrich Lange ◽  
Heidrun Kopp ◽  
Anke Dannowski ◽  
Ingo Grevemeyer ◽  
...  
Keyword(s):  
Phase Velocity ◽  
Ambient Noise ◽  
Cross Correlation ◽  
Velocity Structure ◽  
Signal To Noise Ratio ◽  
Time Windows ◽  
Ocean Bottom ◽  
Ocean Bottom Seismometer ◽  
Ligurian Sea ◽  
Crust And Upper Mantle

<p>The Liguro-Provencal-basin was formed as a back-arc basin of the retreating Calabrian-Apennines subduction zone during the Oligocene and Miocene. The resulting rotation of the Corsica-Sardinia block at roughly 32–24 Ma is associated with rifting, shaping the Ligurian Sea. It is highly debated though, whether oceanic or atypical oceanic crust was formed or if the crust is continental and experienced extreme thinning during the opening of the basin.</p><p>To investigate the velocity structure of the Ligurian Sea a network (LOBSTER) of 29 broadband Ocean Bottom Seismometer (OBS) was installed jointly by GEOMAR (Germany) and ISTerre (France). The LOBSTER array measured continuously for eight months between June 2017 and February 2018 and is part of the AlpArray seismic network. AlpArray is a European initiative to further reveal the geophysical and geological properties of the greater Alpine area.</p><p>We contribute to the debate by surveying the type of crust and lithosphere flooring the Ligurian Sea.<br>Because of additional noise sources in the ocean, causing instrument tilt or seafloor compliance, OBS data are rarely used for ambient noise studies. However, we extensively pre-process the data to improve the signal-to-noise ratio. Then, we calculate daily cross-correlation functions for the LOBSTER array and surrounding land stations. Additionally, we correlate short time windows that include strong events. The cross-correlations of these are dominated by earthquake signals and allow us to derive surface wave group velocities for longer periods than using ambient noise only. Finally, phase velocity maps are obtained by inverting Green’s functions derived from cross-correlation of ambient noise and teleseismic events, respectively. The phase velocity maps show strong heterogeneities for short periods (5-15 s, corresponding to shallow depths). Causes for these include varying sediment thickness, fault zones and magmatism. For longer periods (20-80 s) the velocity structure smoothens and reveals mantle velocities north-northwest of the basin centre. This might hint on an asymmetric opening of the basin. We do not see strong indications for an oceanic spreading centre in the Ligurian basin.</p>

Velocity and attenuation models of Tenerife and La Palma (Canary Islands, Spain) through Ambient Noise Tomography.

2020 ◽  
Author(s):  
Iván Cabrera ◽  
Jean Soubestre ◽  
Luca D'Auria ◽  
Edoardo Del Pezzo ◽  
José Barrancos ◽  
...  
Keyword(s):  
Group Velocity ◽  
Canary Islands ◽  
Ambient Noise ◽  
Dispersion Curves ◽  
Ambient Noise Tomography ◽  
Geothermal Exploration ◽  
La Palma ◽  
Cross Correlations ◽  
Attenuation Models ◽  
Volcanic Islands

<p>Tenerife and La Palma are active volcanic islands belonging to the Canarian archipelago. The island of La Palma is the most occidental and volcanically active island of the archipelago. The youngest volcanic rocks are located in the Cumbre Vieja volcanic complex, a fast-growing North-South ridge in the southern half part of the island. On the other hand, the central part of Tenerife island hosts the Teide composite volcano, the third tallest volcano on Earth measured from the ocean floor. The volcanic system of the island extends along three radial dorsals, where most of the historical eruptions occurred. Those two volcanic islands have potential geothermal resources that could be exploited to increase the percentage of renewable energy in the Canary Islands.</p><p> </p><p>The main objective of this work is the use of Ambient Noise Tomography (ANT) to determine high-resolution seismic velocity and attenuation models of the first few kilometres of the crust, in order to detect anomalies potentially related to active geothermal reservoirs. In the case of Tenerife, previous tomographic studies were performed on the island using active seismic data. They allowed to image the structure of the first 8 km depth. However, for the purpose of geothermal exploration, a higher spatial resolution is needed for the first few kilometres and the determination of the shear wave velocity has a particular importance when searching for fluid reservoirs. In the case of La Palma, no seismic tomography was performed yet.</p><p> </p><p>To realize the ANT, we deployed temporary broadband seismic networks in the two islands. In total, we deployed seismic stations on 41 measurements points in Tenerife and 23 points in La Palma. The campaigns lasted at least 1 month, using jointly the permanent seismic network Red Sísmica Canaria (C7) operated by INVOLCAN. After performing standard data processing to retrieve Green’s functions from cross-correlations of ambient noise, we retrieved the dispersion curves using the FTAN (Frequency Time ANalysis) technique. The inversion of dispersion curves to obtain group velocity maps was performed using a novel non-linear multiscale tomographic approach. The forward modelling of surface waves traveltimes was implemented using a shortest-path algorithm which takes the topography into account. The method consists of progressive non-linear inversion steps at increasing resolution. This technique allows retrieving 2D group velocity models in presence of strong velocity contrasts with up to 100% of relative variation.</p><p> </p><p>In parallel with velocity model, we retrieved maps of seismic attenuation (i.e. quality factor Q) retrieved from the coda envelope decay of noise cross-correlations (Q-coda). For each source-receiver pair, a Q-coda value was calculated, and mapped to the target area by using 2D empirical sensitivity kernels for diffusion (Del Pezzo and Ibañez, 2019). We compared 2D velocity and attenuation images at different dominant periods, evidencing structural features for Tenerife and La Palma islands which seem to be relevant for the purpose of geothermal exploration.</p>

Rapid Global Finite-Frequency Ambient Noise Source Inversion

2020 ◽  
Author(s):  
Jonas Igel ◽  
Laura Ermert ◽  
Andreas Fichtner
Keyword(s):  
Ambient Noise ◽  
Noise Source ◽  
Computational Cost ◽  
User Interaction ◽  
Real Data ◽  
Source Distribution ◽  
Source Inversion ◽  
Finite Frequency ◽  
Noise Sources ◽  
Cross Correlations

<p>Common assumptions in ambient noise seismology such as Green’s function retrieval and equipartitioned wavefields are often not met in the Earth. Full waveform ambient noise tomography methods are free of such assumptions, as they implement knowledge of the time- and space-dependent ambient noise source distribution, whilst also taking finite-frequency effects into account. Such methods would greatly simplify near real-time monitoring of the sub-surface. Additionally, the distribution of the secondary microseisms could act as a new observable of the ocean state since its mechanism is well understood (e.g. Ardhuin et al., 2011).</p><p>To efficiently forward-model global noise cross-correlations we implement (1) pre-computed high-frequency wavefields obtained using, for example, AxiSEM (Nissen-Meyer et al., 2014), and (2) spatially variable grids, both of which greatly reduce the computational cost. Global cross-correlations for any source distribution can be computed within a few seconds in the microseismic frequency range (up to 0.2 Hz). Similarly, we can compute the finite-frequency sensitivity kernels which are then used to perform a gradient-based iterative inversion of the power-spectral density of the noise source distribution. We take a windowed logarithmic energy ratio of the causal and acausal branches of the cross-correlations as measurement, which is largely insensitive to unknown 3D Earth structures.</p><p>Due to its parallelisation on a cluster, our inversion tool is able to rapidly invert for the global microseismic noise source distribution with minimal required user interaction. Synthetic and real data inversions show promising results for noise sources in the North Atlantic with the structure and spatial distribution resolved at scales of a few hundred kilometres. Finally, daily noise sources maps could be computed by combining our inversion tool with a daily data download and processing toolkit.</p>

Sign in / Sign up

Export Citation Format

Share Document