Seismic ambient noise has three bands- cultural noise, microseism noise, and infra-gravity waves. In this study, we have considered two bands: cultural noise and microseism noise. Anthropogenic activities generate cultural noise(1-10Hz), whereas ocean, atmosphere, and solid earth interaction lead to the genesis of microseism noise (0.05-1Hz). Our study region is on the Indian subcontinent. We have taken continuous data from the National Center for Seismology, including stations from inland, island, and coastal parts of the subcontinent. A frequency-dependent polarization method is applied. From our observations, the seismic stations located in the busiest part of the city show a high amplitude of noise in the cultural noise band. Diurnal variations are also observed in this band. However, for stations located in the crowded part of the city, the noise level remains high even at night due to the active nightlife in the cities. Seasonal variations are not observed in the cultural noise band. In the microseism band, seasonal variations are observed in both the direction of the noise source and the amplitude of noise due to the significant contribution of the southwest monsoon to the Indian subcontinent. For island stations, the peak in the noise is observed in the microseism band, while in the land stations, cultural noise dominates. Sudden increases in the amplitude of microseism noise are detected in the coastal stations because of the disturbed sea-state conditions during the cyclone period. Overall, our study provides a comprehensive idea about the distribution of seismic ambient noise in the Indian sub-continent.
How to cite: Pradhan, G., Reddy, R., and Singha Roy, P. N.: Seismic Ambient Noise Analysis of the Indian Subcontinent, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-619, https://doi.org/10.5194/egusphere-egu25-619, 2025.
Ambient seismic noise has the potential to significantly reduce the detection capabilities of the planned next generation gravitational wave detector (so-called Einstein telescope). The noise field impacts the detector performance either directly by transferring seismic movements into the detector system or indirectly by changes to the gravitational attraction of the surrounding rockmass due to spatio-temporal density changes caused by seismic waves (so-called Newtonian noise). In order to find the best geometries for the placement of the subsurface detection chambers, allow for active suppresion of detector vibrations and estimate the influence of Newtonian noise, a precise knowledge of the local ambient seismic noise field at the surface and at placement depth is vital. In this context, the amplitudes, source locations and durations of distinct natural and anthropogenic noise sources of the ambient noise field are all of interest.
Here, we present first results from investigations in the EMR region (Euregio Meuse-Rhine) Einstein telescope candidate site in area between Maastricht, Liège and Aachen. Data was recorded by several temporary deployments of short period sensors running for several weeks each. Possible noise sources in this region include ocean microseism from the North Sea/North Atlantic for frequencies below 1 Hz and wind farms, transportation infrastructure like highways and railways and heavy machinery as anthropogenic sources in the frequency range above 1 Hz. Apart from spectral analysis to determine the frequency resolved spatio-temporal amplitude changes at the recording sites, we also investigate the signal coherence over the network to estimate frequency bands, time intervals and (sub-)networks that could be used for further coherence based analysis. We then apply classical array analysis and Matched Field Processing (MFP) to pinpoint possible source locations of the seismic wavefield outside and within the posssible placement area of the Einstein telescope candidate site.
How to cite: Becker, D., Hammer, C., and Hadziioannou, C.: Spatio-temporal characteristics of the ambient seismic noise field at the EMR gravitational wave telescope candidate site, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-9154, https://doi.org/10.5194/egusphere-egu25-9154, 2025.
Cyclones, typhoons or hurricanes over the ocean generate oceanic waves. The interactions between these waves produce pressure fluctuations close to the ocean surface, which are the source of secondary microseisms in the frequency range 0.1–1 Hz. This study investigates secondary microseisms generated by the cyclone Gillian, with a specific focus on the impact of Horizontal Polarization Anomaly (HPA) of surface waves.
Cyclone Gillian developed in March 2014 over the Indian Ocean near southern Indonesia. Initially, it moved westward across the Indonesian Islands, then it reached a minimum distance of ~200 km from the Indonesian shoreline on 21 March. The cyclone then shifted in a west-southwest direction, intensifying as it moved southward. By 23 March, Gillian reached its peak wind speed, with gusts of ~315 km/h, when it was located over 1000 km from Indonesia.
During Gillian's activity, a temporary seismic array consisting of 46 three-component seismometers, with interstation distances of ~2 km, was deployed around the Merapi volcano and its surrounding region in Indonesia. Analysis of this seismic dataset reveals that secondary microseism extended to higher frequencies (up to ~1 Hz) at most stations, coinciding with the cyclone’s closest approach to the array on 20–22 March. In addition, beamforming analysis shows that during periods of strong wind speeds (22–24 March), seismic waves at 0.11 Hz arrived at the network from multiple directions and with various slowness values.
To quantify the effect of the cyclone on secondary microseisms, the full seismic cross-correlation tensor was computed for the cyclone activity period and the subsequent seven months. The Optimal Rotation Algorithm (ORA) (Roux et al. 2009) was applied to estimate the horizontal polarization anomaly of surface waves (Saade et al. 2017). Results demonstrate a significant increase in HPA on 23 March, corresponding to the cyclone’s peak wind speed. The rapid release of the polarization anomaly following the end of the cyclone suggests that this pulse can be attributed to the effects of the cyclone, acting as a moving noise source.
References:
Roux, P. (2009). Passive seismic imaging with directive ambient noise: application to surface waves and the San Andreas Fault in Parkfield, CA. Geophysical Journal International, 179(1), 367-373.
Saade, M., Montagner, J. P., Roux, P., Shiomi, K., Enescu, B., Brenguier, F. (2017). Monitoring of seismic anisotropy at the time of the 2008 Iwate-Miyagi (Japan) earthquake. Geophysical Journal International, 211(1), 483-497.
How to cite: Montagner, J.-P., Riahi, A., Saade, M., Kazantsev, A., Stutzmann, É., and Métaxian, J.-P.: Seismic Ambient Noise and Horizontal Polarization Anomaly During Cyclone Gillian at the Merapi Volcanic Region, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16353, https://doi.org/10.5194/egusphere-egu25-16353, 2025.
Ocean generated low-frequency seismic noise signals called microseisms are linked to the ocean environment, the subsurface and the atmosphere. The energy associated with microseisms is closely related to ocean wave amplitude, globally and throughout the seasons, and shows great potential in the analysis of climate variability.
To comprehend the generation and propagation mechanisms of secondary microseisms offshore Sicily, a detailed analysis of Distributed Acoustic Sensing (DAS) data is being conducted on the MEOC fibre optic cable of the INFN-LNS submarine infrastructure offshore Catania. DAS technology exploits the backscattering properties of fiber optic cables which enables data acquisition over large distances of cables acting as a densely distributed array for recording strain rate at the seafloor. The DAS data presented in this study was collected over five days (October 10–15, 2020) and focuses on the first 20 km section of the MEOC cable.
The appearance of pronounced energy observed in both the DAS data and nearby land seismometer within the secondary microseism band (~3s), persisting for a prolonged duration and coincident with a “storm” event, confirms the impact of regional weather conditions on microseism generation. This specific time window is analysed in detail to explore the propagation effects of the secondary microseism wavefield arriving on the cable. Hindcast data from the WAVEWATCH III ocean wave model identifies a secondary microseism source location south of Sicily, generated in response to high winds from northwest to southeast. The 20 km long array with a channel spacing of 2 m enables, through effective FK (frequency-wavenumber) domain analysis, a detailed examination of the spatial variability of the wavefield arriving at the cable. The FK analysis plots show almost equal energy on both positive and negative wavenumbers with a slight dominance in seaward propagation directions. A simple forward seismo-acoustic simulation performed to see the effects of bathymetry corroborates the results obtained from the DAS observations. The simulations demonstrate that while the wavefield reaches the cable at a normal angle, a dominant seaward propagation is observed towards the cable's end due to the wavefront getting redirected from the continental shelf. Multiple simulations with different source locations are discussed in order to further understand the influence of the source position as well as regional bathymetry on the wavefield recorded with the DAS. These findings highlight the complex interplay between the role bathymetric features in microseism propagation and dominant microseism source location. Spatially dense DAS data recorded over long distances can play a key role in unravelling these dynamics.
How to cite: Vijayan, A., Le Pape, F., Bean, C. J., Murphy, S., Ker, S., Jousset, P., Riccobene, G., Viola, S., Gutscher, M., Currenti, G., Aurnia, S., and Pulvirenti, S.: Multiscale Secondary Microseism Propagation Analysis Using Full Waveform Modelling and DAS Observations offshore Catania, Sicily, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-12126, https://doi.org/10.5194/egusphere-egu25-12126, 2025.
Seismic interferometry yields a correlation wavefield that is closely related to the Green’s function of the medium under the condition of homogeneously distributed sources. In cases where this condition is not met, iteratively computing the correlations of correlation wavefields (“higher-order correlations”) has been argued to improve Green’s function retrieval and thus be a useful processing step for imaging applications. Higher-order correlations can also retrieve correlation wavefields between stations that were not installed simultaneously, which can help to homogenize the imaging conditions for asychronous deployments.
We show that higher-order correlations do not improve correlation wavefields when isolated noise sources are present, which is common for seismic field data. Instead, higher-order correlations enhance the travel time bias introduced by such source distributions. This impacts both far-field investigations, such as tomographic studies, and near-field investigations, such as spatial autocorrelations and focal spots. We simulate several source scenarios numerically to showcase this behaviour. Field data observations from a large-N nodal deployment in Eastern Austria confirm these considerations. This work exposes the need for a reliable strategy to assess the correlation wavefield properties before applying advanced processing, such as higher-order correlations.
How to cite: Schippkus, S., Hillers, G., and Hadziioannou, C.: On the undesired behaviour of higher-order correlations, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5594, https://doi.org/10.5194/egusphere-egu25-5594, 2025.
Accurate subsurface seismic velocities are crucial for drilling exploration wells, exploring geothermal resources, or locating seismic events. Due to their dispersive nature and prevalence in ambient seismic noise, surface wave velocities can be used to obtain shear velocities beneath seismic arrays. Localized shear velocity anomalies indicate the presence or absence of fluids; Temporal variations in shear velocities can indicate changes in fluid content or poisson ratio over time, i.e. during geothermal operations.
Three-component ambient noise beamforming (B3Am) detects waves passing over seismic arrays and retrieves their propagation direction, propagation speed, and polarization parameters. Since B3Am analyzes the wavefield at discrete frequencies, with the frequency band limited by the array size, and in short time windows, typically ten times the period, results can be stacked over short time periods and temporal analysis of dispersion curves, backazimuth, and polarization parameters becomes possible. However, before interpreting temporal variations caused by physical changes in the subsurface, the overall variability needs to be estimated to accurately estimate uncertainties. Changing noise source fields do not impact the B3Am results directly but could cause deviations in absolute parameters. Furthermore, seasonal changes in water content in sediments could introduce seasonal variations not related to geotechnical activities.
Using an existing dataset recorded with 23 broadband seismometers deployed over a range of ten months in an area of about 15 km in diameter in Germany, we analyse the seismic noise wavefield and estimate the temporal stability of surface wave dispersion curves. We calculate probabilistic power spectral densities, investigate their variability over time, and compare them to the wavefield composition, i.e. body to surface wave ratio, computed with B3Am. We plot backazimuths and velocities against frequency for all wave types for selected days. We compare different stack lengths in a bootstrapping-type analysis to see the minimum number of detections, i.e. recording length, needed for accurate results. We see that for frequencies below 1 Hz, five days of continuous noise recordings produce a stable dispersion curve. However, we see large seasonal variations between results in Autumn and Spring to results in Winter. These variations can be attributed to changes in the noise field, both of natural and anthropogenic origin. Finally, we derive expected uncertainties and provide insights about the impact for depth inversions.
How to cite: Finger, C., Neugebauer, S., and Löer, K.: Temporal stability of surface wave dispersion extracted from ambient seismic noise using three-component beamforming, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11944, https://doi.org/10.5194/egusphere-egu25-11944, 2025.
Interactions between oceanic waves and the seafloor generate seismic waves recorded globally and referred to as natural ambient “noise.” In particular, the 3-10s period band, known as the secondary microseismic band, corresponds to non-linear oceanic wave-wave interaction and represents the highest peak in a seismic station PSD. While surface waves are prominent in this period band, body waves, which sample deeper areas and are less scattered, can also be identified. These body waves are valuable for examining the properties of the deep Earth due to their sensitivity to the inner medium.
In the last decade, improvements in global oceanographic hindcast, such as the WAVEWATCHIII model, have allowed seismologists to track the spatiotemporal behavior of these ocean-generated seismic sources. Since these unconventional sources are non-impulsive, interferometric methods, by correlating signals between stations for a few hours, are necessary to highlight surface and body waves from local to global scale.
We introduce the WMSAN Python library for Wave Model Sources of Ambient Noise, which allows for the visualization of oceanic sources of ambient noise distribution and computation of proxy for seismic observables in a user-friendly fashion. This library provides functions and simple examples to map secondary microseismic source distributions for Rayleigh, P, and SV waves using oceanographic data. Seismic data counterparts are then inferred from these source distributions, such as synthetic spectrograms and cross- or auto-correlation functions. We will detail the benchmark examples of this library and its application to extract body wave interference (PP-P) differential travel times from a single secondary microseismic event occurring 8-11 December 2014 in the Northern Atlantic Ocean.
How to cite: Tomasetto, L., Boué, P., Ardhuin, F., Stutzmann, É., Xu, Z., De Plaen, R., and Stehly, L.: WMSAN: Wave Model Sources of Ambient Noise Python Library. From Modeling to Applications., EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5776, https://doi.org/10.5194/egusphere-egu25-5776, 2025.
The fine structure of the upper and lower limits of the mantle’s transition zone (so-called 410 and 660 discontinuities) enlightens us on the temperature and composition of the mantle with, consequently, major impact on our modelling of mantle dynamics. We demonstrated in 2012 that it is possible to extract subvertically reflected waves (Pv410P and Pv660P) on the discontinuities from the seismic noise. We review how our knowledge has progressed since then, for better imaging of the 410 and 660 discontinuities. Rather than recovering an approximate Green’s function, based on high levels of diffraction and/or an even source distribution, Pv410P and Pv660P result from cross terms of a limited geographical coverage of P-wave sources from distant storms. It is possible to identify, in the data or with models of seismic noise, the time windows which effectively contribute to Pv410P and Pv660P, but the methods carry inherent risk of source location dependent time and amplitude bias which can be quantified through numerical modeling. Together, these studies indicate practical ways forward for being able to extract mantle discontinuities reflections with higher signal to noise ratios than previously, and for adequately interpreting them through a deeper understanding of potential bias and uncertainties.
How to cite: Pedersen, H.: Seismic noise and mantle discontinuity reflections – insight gained and future directions, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3481, https://doi.org/10.5194/egusphere-egu25-3481, 2025.
The Tainan deformation front is an anticline located at the Chinese continental shelf-slope break in the transition from subduction to collision in southwestern Taiwan. Several NE-SW trending mud diapirs and mud volcanoes have been identified by marine geophysical survey offshore southwest of Tainan, where one submarine-diapirs is spatially closely linked to the onshore Tainan anticline. In contrast, distinct fold-and-thrust belts occur east of the Tainan anticline, where thrust faulting on the detachment folds is often considered the dominant mechanism for the Tainan anticline. Whether a frontal thrust or diapirism formed the Tainan anticline relates to the earthquake potential of the ~1.8 million populated Tainan city. We selected 2-3 months-long continuous seismic recordings of 33 seismic stations covering a 15 by 10 km2 area in the Tainan deformation front of southwestern Taiwan from February to June 2021 to conduct ambient noise tomography. With an inter-station distance of about 1 km, the seismic array spanned from west to east across four major tectonic regimes: Anping Plain, Tainan Tableland, Dawan Lowland, and Chungchou Tableland. Two to three months of ambient noise were cross-correlated between each station pair, and clear fundamental-mode Rayleigh waves were observed between 1 and 3 seconds. For the Eikonal tomography analysis, five BATS (Broadband Array in Taiwan for Seismology) stations and six CWASN (Central Weather Administration Seismographic Network) stations were chosen at distances ranging from 40 to 80 kilometers away from the Tainan array, with azimuths between 45° to 140° and 270° to 360°. We then calculated the cross-correlation function (CCF) between 33 seismometers and these stations, and we measured the relative surface wave arrival times using the beamforming method. To perform Eikonal tomography, we calculated the surface wave propagation of the ambient noise and the shallow velocity structure for each period between 4 and 9 seconds. The ambient noise tomography and the Eikonal tomography results were combined to construct the three-dimensional shear wave velocity model of the Tainan anticline. Our result shows that the velocity at 1-3 seconds under the study area is almost uniform, while the velocity at 4-9 seconds shows up to +18% velocity perturbations beneath the Tainan anticline. This suggests that there are no significant lateral variations in lithology at a shallow level (1-3 seconds) and a high-density body, the mud-diapir extruded upward beneath the Tainan anticline at a deeper level (4-9 seconds).
How to cite: Rau, R.-J., Yang, T.-C., and Chen, Y.-N.: Ambient noise tomography of the Tainan deformation front in southwestern Taiwan, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-2303, https://doi.org/10.5194/egusphere-egu25-2303, 2025.
The seismic Rayleigh wave focal spot imaging technique utilizes ambient seismic noise fields to estimate local Rayleigh wave phase velocities. Records of scattered and diffuse wavefields in dense seismic arrays are now routinely used to obtain virtual surface waves propagating between stations by cross-correlation, which supports ambient noise tomography. Rayleigh waves that refocus on the virtual source form the spatial autocorrelation field or focal spot. The shape of the narrow-band focal spot is used to obtain the local Rayleigh wave phase velocity at each virtual source or sensor in an array. It has been demonstrated that the lateral resolution of focal spot imaging depends on the data range on the order of one wavelength that is used to constrain the Bessel function model from the focal spot data. This can be observed as lateral spreading or averaging of velocities in inhomogeneous velocity distributions. Here we conjecture that the spreading effect is similar to the blurring effect observed in optical images, where the blurring is quantified by the point spread function that is the operator describing how the imaging device affects the image. Undoing the blurring in conventional images caused by the imaging device point spread function can be achieved by deconvolution methods. In seismic imaging, however, the exact properties of the focal spot imaging point spread function remain unknown. Determining the focal spot imaging point spread function properties has the potential to yield better resolved focal spot images. Experimental determination of the microscope point spread function is a routine task in microscopy, allowing for sharper images of near-diffraction limit scale objects through deconvolution. In microscopy, the empirical point spread function is determined by imaging sub-diffraction limit scale fluorescent beads acting as point sources. We adopt a similar approach to determine the empirical seismic focal spot imaging point spread function by imaging known velocity structures in synthetic focal spot imaging configurations using two-dimensional acoustics simulations in a reverberating cavity. Point-like velocity distributions are imaged to obtain empirical point spread functions. The empirical point spread functions are validated by deconvolving blurred synthetic images where the original velocity structure is known. As image deconvolution is an ill-posed problem, regularization methods are used to stabilize the solution. We utilize traditional spectral filtering methods such as truncated singular value decomposition and Tikhonov regularization, and total variation regularization to reconstruct the original velocity distribution using the empirical point spread function. Updated results of the empirical seismic focal spot imaging point spread function are presented.
How to cite: Kolehmainen, K., Hillers, G., Giammarinaro, B., Juvonen, M., Meaney, A., and Siltanen, S.: Estimation of the seismic focal spot imaging point spread function, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6164, https://doi.org/10.5194/egusphere-egu25-6164, 2025.
Seismic monitoring is essential for the successful and sustainable exploration and operation of underground reservoirs and storage systems. Distributed Acoustic Sensing (DAS) has emerged as a solution for the acquisition of seismic data with high spatial density and extensive coverage, benefiting seismic monitoring efforts. Making use of unused telecommunication fibers, or dark fibers, is a particularly attractive opportunity due to the widespread availability of this infrastructure. It can help address the challenges associated with deploying and maintaining extensive seismic networks, particularly in urban areas targeted for geothermal energy development. This study uses a 3 km section of the telecommunication network at the Karlsruhe Institute of Technology (KIT) to conduct seismic monitoring near the planned DeepStor geothermal research infrastructure. The research includes a preliminary verification of the fiber's location and reports observations from local seismic events, harnessing the high spatial density of sensing points for beamforming analysis. Additionally, a signal classification framework is designed to detect and categorize frequent vehicle passages. The analysis of the associated signals makes it possible to extract virtual shot gathers. These gathers facilitate the analysis of dispersion curves at relatively high frequencies, which are subsequently used to invert shear-wave velocity profiles. This complements lower-frequency analyses derived from microseism signals during periods of minimal anthropogenic activity. Continuous seismic wavefield recordings were collected over an eight-month period, providing access to a significant time series for analysis of temporal variations and signal stacking. Our results provide a basis for future seismic monitoring of the upcoming DeepStor research infrastructure on the KIT Campus North. They also demonstrate the potential of ambient seismic wavefield analysis for advanced subsurface characterization and contribute to the broader application of DAS technology in urban seismic monitoring.
How to cite: Proenca, T., Gaucher, E., Rietbrock, A., and Azzola, J.: Seismic Imaging and Monitoring with Distributed Acoustic Sensing on Dark Fibers at the KIT Campus, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-9091, https://doi.org/10.5194/egusphere-egu25-9091, 2025.
Landslides are difficult to predict and can therefore be a serious threat to populations and infrastructures. Understanding landslide processes and their precursor parameters is an important challenge for scientists and for public managers. Landslide monitoring is essential for determining the hazard associated with the unstable slopes. Computing seismic velocity changes from ambient seismic noise has been proven an effective tool for landslide monitoring. Before failure, a drop in rigidity in the landslide leads to a decrease in shear wave velocity, creating a potential precursory signal that can be retrieved by seismic interferometry (1). An important step is distinguishing these precursory signals from environmental influences on the seismic velocities that are often observed.
The Séchilienne landslide, located 25 km southeast of Grenoble, has been instrumented with broadband seismic stations since 2013 (2). We use this extensive dataset to compute relative seismic velocity changes (dV/V) from ambient seismic noise over a period of 10 years. We observe seasonal cycles that could be associated with the thermomechanical state of the slope and/or the water table fluctuations. These seasonal cycles are most prominent in low frequency ranges (2–8 Hz). The dV/V fluctuations remain in a range of a few percents. In particular, no significant velocity drop was observed, which is coherent with no observed sudden acceleration within these 10 years.
This work was partially funded by the European Research Council (ERC) under grant No. 101142154 - Crack The Rock project.
(1) Mainsant, G., E. Larose, C. Brönnimann, D. Jongmans, C. Michoud, and M. Jaboyedoff (2012), Ambient seismic noise monitoring of a clay landslide: Toward failure prediction, J. Geophys. Res., 117, F01030, doi:10.1029/2011JF002159.
(2) Seismic data have been acquired by the French National Landslide Observatory (OMIV), and are available at doi.org/10.15778/RESIF.FR and doi.org/10.15778/RESIF.MT
How to cite: Wienk, I., Guillemot, A., Radiguet, M., Helmstetter, A., and Larose, E.: Decennial Monitoring of the Séchilienne Landslide with Seismic Noise, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8256, https://doi.org/10.5194/egusphere-egu25-8256, 2025.
Ambient seismic noise interferometry is a powerful tool to monitor changes in seismic velocities within the upper crust induced by various forcings. Several studies have shown that the dynamics of aquifers can generate significant seismic velocity variations, concealing more subtle variations linked to other phenomena. Here, we present a temporal and spatial analysis of subsurface seismic velocity variations over a portion of the Upper Rhine Graben in north-eastern France, hosting one of the largest watertable in Europe. We analyze 4 years of continuous seismic records between 2019 and 2023 from 144 permanent and temporary seismological stations, together with data from 195 piezometers from the APRONA observatory. Ambient seismic noise cross-correlations were calculated using horizontal and vertical components records, and we performed velocity variations in different frequency bands (ranging from 0.1 to 4 Hz) and lapse times. We systematically compared temporal variations in seismic velocities with ground water level variations. Overall, our results indicate a strong seasonality of seismic velocity changes above 1 Hz mainly in the ballistic surface waves time window and the beginning of the coda of correlations. This signature persists at lower frequencies, around 0.5 Hz, for longer times in the coda of correlations only. This suggests a possible influence of aquifer dynamics at greater depths. We spatially localised velocity changes above 1 Hz using coda waves sensitivity kernels and found patterns consistent with piezometric observations and the known limits of the water table.
How to cite: Mattern, F., Zigone, D., Vergne, J., and Schmittbuhl, J.: Monitoring groundwater dynamics in the shallow crust over the Upper Rhine Graben (France) using ambient seismic noise interferometry, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-9523, https://doi.org/10.5194/egusphere-egu25-9523, 2025.
Climate change significantly impacts groundwater resources by altering recharge rates and thus availability, making it crucial to manage these vital reserves sustainably to ensure long-term water security. In this study we seismically monitor a series of groundwater pumping tests in the municipality of Nickelsdorf (Burgenland, Austria). Due to expected increasing demand for water due to population development, wells were installed to ensure a sustainable drinking water supply in the long term. Traditionally monitored through point-wise hydrological wells, our approach combines nodal seismic sensors and ambient noise to broaden insights into subsurface processes affected by pumping activity. Seismic ambient noise was continuously recorded over three months in early 2023, including periods before, during, and after pumping. Our study evaluates various ambient noise sources and seismic signals, especially those generated by passing trains. To gain broader understanding of the subsurface processes we perform a time-lapse tomography to identify the location and strength of the velocity variations. Based on our analysis, we resolve increases/decreases in seismic velocity of around 10 % in the uppermost meters of the subsurface during pumping operations related to local reduction in the water table. This holistic approach aims at unveiling the behavior of the subsurface during and post-pumping, potentially offering a comprehensive understanding beyond individual hydrological wells.
How to cite: Kramer, R., Lu, Y., Estève, C., Gosselin, J., Jochum, B., and Bokelmann, G.: Time-Lapse Tomography of a Groundwater Pumping Experiment, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16250, https://doi.org/10.5194/egusphere-egu25-16250, 2025.
Natural oscillatory stress sources can be exploited as ‘pump’ to ‘probe’ the time-dependent response of active fault zones to stress variations at various temporal and spatial scales and investigate time-dependent variations of their elastic properties (Delorey et al., 2021). Recently, we carried out a multidisciplinary study along the Irpinia Fault System (IFS, Southern Apennines) to investigate the response of the crust to hydrological forcing associated with phases of recharge/discharge of karst aquifers in terms of time-dependent variations of its elastic and hydraulic properties.
Charge/discharge phases of the karst aquifers in the Apennines cause significant seasonal and multi-annual strain transients (Silverii et al., 2016), that modulate the secular, tectonic deformation (~3mm/yr extension across the Apennines). These seasonal and multi-annual transients correlate with the seismicity rate (D’Agostino et al., 2018) and seismic velocity variations (De Landro et al., 2022; Poli et al., 2020). Previous studies (D’Agostino et al., 2018; Silverii et al., 2016) showed the high sensitivity of the IFS volume to hydrological stresses reflected in a complex, time-dependent response of deformation and seismicity.
Within this framework, we performed a natural analogue to a quasi-static laboratory ‘pump-probe’ experiment to assess the non-linear behaviour of the seismogenic volumes in response to non-tectonic deformations. We used the seasonal horizontal strains associated with discharge/recharge of karst aquifers as the ‘pump’. We computed continuous in-time seismic velocity variations δv/v using empirical Green's functions (the ‘probe’) reconstructed by autocorrelation on continuous 14-year-long time series of ambient-noise (Shapiro & Campillo, 2004). We initially analyzed two different sites (co-located GPS and seismic stations), near and afar the IFS. We found that δv/v are significant (∼0.2%) nearby IFS (shallow carbonate rocks), rather than far away from it.
We compared for the site near IFS with the time series of Caposele spring discharge, strain and seismicity-rate. Our observations are coherent at seasonal and multi-annual scales and can be explained with the same mechanism. During periods of maximum hydraulic head within the aquifer, hydrologically related extensions are correlated with a decrease in the seismic wave velocity. During these episodes, the occurrence of microearthquakes is favored within the extensionally deforming belt along the Apennines thanks to the contribution of hydrological forcing. The non-linear elasticity suggested the presence of a multi-fractured and damaged crust subject to periodic seasonal phases of weakening/healing, potentially affecting earthquake nucleation processes.
Then, we extended our seismological analysis computing δv/v within the Campania-Lucania region. The other sites around IFS behave similarly, with a decrease in the seismic wave velocity commonly related to hydrologically extensions within the shallow carbonate rocks. Our observations confirm the significance of the hydrological forcing as a source of changes in elastic properties at a regional scale, a characteristic likely shared by the volumes surrounding the largest karst aquifers in the Apennines. For faults in a critical state, cyclical softening, g.e. caused by external forcing, may lead to failure and seasonal seismicity.
How to cite: Tarantino, S., Poli, P., D'Agostino, N., Vassallo, M., Festa, G., Ventafridda, G., and Zollo, A.: Non-linear elasticity, earthquake triggering and seasonal hydrological forcing along the Irpinia fault, Southern Italy, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13417, https://doi.org/10.5194/egusphere-egu25-13417, 2025.
The eruption of the Hunga Tonga-Hunga Ha’apai (hereafter referred to as Hunga-Tonga) on January 15, 2022, generated atmospheric pressure waves (Lamb waves) that traveled around the globe (Matoza et al., 2022). These waves were associated with ground deformation in the solid Earth and analyzed as a pressure-loading problem in a previous study by Anthony et al. (2022). Strictly speaking, the deformation of the solid Earth is an integral component of Lamb waves in the coupled Earth system. In this study, we develop an analysis method for Lamb waves within a coupled Earth model.
Our focus is on analyzing the ratio between vertical displacement and surface pressure, referred to as the compliance ratio, which provides critical insights into the elasticity of the upper crust. We demonstrate an inversion method to utilize this ratio for determining shallow crustal elasticity. This approach is analogous to the compliance method used for seismic noise to constrain the elasticity of sedimentary layers on the ocean floor (Crawford et al., 1991) and the elasticity of shallow structures at seismic stations equipped with co-located pressure sensors (e.g., Tanimoto and Wang, 2018, 2019).
Depth sensitivity kernels for the compliance ratio can be obtained through numerical differentiation. These compliance data primarily exhibit sensitivity to the near-surface shear modulus, with additional sensitivity to the shallow bulk modulus at hard rock sites. Given the Lamb wave phase speed of approximately 310 m/s and the high-coherence range limit of around 0.01 Hz, the depth range of reliable resolution is confined to the upper crust (approximately 5-15 km deep).
This method, initially developed for stations with co-located pressure and seismic sensors, can also be extended to stations equipped only with seismic sensors. This extension is feasible because Lamb wave waveforms exhibit minimal variation. By analyzing the coherence between seismic and pressure data from nearby locations, we can select suitable pairs of seismic and pressure data and apply our compliance method.
We demonstrate that this method can not only derive new information about shallow structures but also serve as a valuable tool for testing shallow structures in existing seismic velocity models. Improvements in our understanding of shallow elasticity structures are crucial for accurate ground motion predictions in seismically active regions worldwide.
How to cite: Tanimoto, T. and Alvarez, M.: Shallow Earth Elasticity from Sweeping Atmospheric Pressure Waves in the Coupled Earth , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3851, https://doi.org/10.5194/egusphere-egu25-3851, 2025.
Passive seismic methods have gained significant attention in recent decades due to their cost-effective and non-invasive nature, making them ideal for continuous monitoring of subsurface dynamics. Among these methods, ambient noise coda-wave interferometry is widely used for detecting time-lapse changes in seismic velocities and has been successfully applied in diverse geological settings. However, its effectiveness might be limited by the complexity of coda wave composition, which complicates the estimation of the depth sensitivity of velocity changes, and by the instability of noise sources, which can introduce artificial velocity changes. These limitations highlight the need for alternative methodologies.
In this study, we evaluate the potential of Rayleigh wave ellipticity as a non-interferometric tool for detecting near-surface variations. Using the degree of polarization method, combined with time-frequency analysis to isolate Rayleigh waves, we analyze time-lapse variations in ellipticity from multiple field case studies. To quantify these variations, we employ normalized cross-correlation and cross-covariance metrics. Unlike other widely used noise-based methods, such as spectral analysis or horizontal-to-vertical spectral ratios, our approach examines the full ellipticity function. This allows for the detection of velocity changes over a broader depth range without being limited to specific features of the curve.
Our results demonstrate the robustness of Rayleigh wave ellipticity in detecting shallow subsurface changes, which is source-unbiased. This approach addresses key limitations of existing geophysical methods and expands the toolbox for seismic monitoring, enabling a more comprehensive analysis of subsurface dynamics. Potential applications include monitoring groundwater variations, assessing infrastructure stability, and contributing to the understanding of subsurface dynamics in complex environments, making it a versatile and powerful tool for environmental and geological studies. Rayleigh wave ellipticity emerges as a robust, independent alternative, offering a refined approach for monitoring subsurface changes and addressing the challenges faced by other noise-based methods.
This work has received funding from the AGEMERA project, financed by the European Union’s Horizon Europe research and innovation programme under grant agreement N° 101058178. As well, this work has benefited from partial support of the STONE project (CPP2021-0090072), financed with funds from the Ministry of Science and Innovation through the State Agency for Innovation (MCIN/AEI/10.13039/501100011033) and the European Union-Next Generation through the Recovery, Transformation and Resilience Plan (PRTR).
How to cite: Seivane, H., Schimmel, M., Martí, D., and Sánchez-Pastor, P.: Near-Surface Monitoring with Rayleigh Wave Ellipticity, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11471, https://doi.org/10.5194/egusphere-egu25-11471, 2025.
Microseisms are predominantly generated through two mechanisms: the interaction between ocean waves and the shore (primary microseisms, PM) and the interaction of ocean swells traveling in opposite directions with similar frequencies (secondary microseisms, SM). In some regions, the SM spectrum exhibits a splitting into long-period and short-period components (LPSM and SPSM, respectively), typically associated with local wave-wave interactions.
As part of the EPYSIM project and with the aim to perform an ambient noise imaging study, we deployed an array of 19 broad-band stations in two different profiles in Catalonia (NE of the Iberian peninsula). The profiles are surrounded by coastline at different azimuths and the distance to the coast along each profile increases progressively. This particular setting provides an ideal environment to investigate the characteristics of SPSM. Using the EPYSIM stations and other nearby permament stations, we conduct a detailed analysis of microseisms, including spectral time evolution, polarization analysis, and attenuation patterns.
Our findings are consistent across all analyzed years and reveal distinct back-azimuths for the microseisms: PM and LPSM predominantly originate from the North Atlantic, while SPSM exhibits markedly different back-azimuths, pointing towards the Mediterranean. Additionally, the spectral intensity of SPSM differs significantly from that of PM and LPSM, suggesting a local source for SPSM, consistent with observations in other regions.
This work has benefited from support of the EPYSIM Project, funded by the Spanish Ministry of Science and Innovation (Ref.: PID2022-136981NB-I00).
How to cite: Jorde, S., Schimmel, M., Sánchez-Pastor, P., Seivane, H., and Díaz, J.: Seismic noise characterization in NE Iberia, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10742, https://doi.org/10.5194/egusphere-egu25-10742, 2025.
The Arena of Verona is a Roman amphitheater built in the first century AD. It is a remarkable example of the historical architecture and cultural heritage of Italy. With an elliptical structure measuring approximately 150 meters in length and 120 meters in width, it remains one of the largest and best-preserved amphitheaters of its kind. Located in the center of Verona, the Arena continues to host cultural events that attract thousands of people. To ensure the long-term safety of the structure and decrease risks to visitors, it is important to evaluate its structural integrity and to understand the interaction with the soil. The Arena is surrounded by a densely populated urban environment that generates significant seismic (vibrational) noise. While this poses some challenges for geophysical studies and structural monitoring, it allows for non-invasive characterization using mechanical waves.
In this study, seismic noise was recorded continuously for 7 days by a square 2D seismic array, with side of 18 m and 3 m sensor spacing. The network consisted of 40 1-component and 10 3-component SmartSolo seismic nodes, deployed on the stone floor at the center of the Arena.
Data were processed using passive seismic techniques. First, the frequency content and amplitude distribution of the measured seismic noise over time were investigated. Second, the spatial distribution of the noise sources was derived using one-component beamforming. Third, the shear wave velocity (Vs) profile of the underlying medium was derived using passive seismic interferometry. The Vs profile will be used for seismic risk assessment studies.
The present study is being carried out within the framework of the USES2 project, which receives funding from the EUROPEAN RESEARCH EXECUTIVE AGENCY (REA) under the Marie Skłodowska-Curie grant agreement No 101072599.
How to cite: Nesterova, O., Boaga, J., Cassiani, G., Brovelli, A., Pavoni, M., Peruzzo, L., and Barone, I.: Urban seismic noise characterization at the Verona Arena, Italy, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-12002, https://doi.org/10.5194/egusphere-egu25-12002, 2025.
At regional scale, P waves reflected on mantle discontinuities extracted by stacking seismic noise analysis are observed in cross-correlation functions (Poli et al., 2012). Under specific noise field condition, Pedersen et al. (2022), significantly improved the SNR of Pv410p and Pv660p. Initially it was assumed that the waves extracted reflected, albeit in a non perfect wave, the so-called Green's function. However, as the noise field is not perfectly distributed, the full Green's function is not retrieved, and details on these reflected wave properties (constituents in the CC, amplitudes, time bias, ...) remain unknown. We here show which major coherent phases interact to reconstruct Pv410p and Pv660p, and give a way to analyze bias between the observed cross-correlation and the Green's function, both in terms of time delays and relative amplitudes. These waves stem from long-range mantle phases that interact in the correlation function to enhance short-range reflections. We also simulated cross-correlation functions using very distant sources, ranging from a single point to realistic ocean models. The complexity of the wave field, which introduces cross terms into the correlation function, converges better when the source is extended. Our results demonstrate how to correct the simple modeling of these waves in order to obtain a detailed characterization of the observables in the noise cross-correlation. This opens up the possibility of using body-wave retrieval to obtain an accurate picture of mantle discontinuities and to constrain their compositions and temperatures using forward modeling and thermophysics.
How to cite: Pawlowski, P., Pedersen, H., Boué, P., and Tauzin, B.: Modeling teleseismic P wave interference from distant oceanic sources for upper mantle reflection imaging., EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3682, https://doi.org/10.5194/egusphere-egu25-3682, 2025.
Seismic imaging has traditionally been applied to far-field signals using earthquake tomography and, in more recent decades, ambient noise tomography. However, with modern large-N seismic arrays it has been demonstrated that structural observations at sub-wavelength distances are possible utilizing the surface wave focal spot imaging method. The focal spot is the zero-lag correlation amplitude field derived from dense array noise correlations. One advantage compared to noise tomography is the ability to analyze waves at length scales that are large compared to the array size, which enhances the depth resolution. The lateral resolution of focal spot imaging also improves at short distances when the station density is high. The focal spot is the time domain representation of the spatial autocorrelation field (SPAC). We can thus use established analytical SPAC methods to constrain local Rayleigh wave phase velocity estimates from the focal spot shape. Here, we investigate the effectiveness of the method in the context of critical raw material exploration. We apply focal spot imaging in the Kylylahti polymetallic mining area, hosting sulphide ore deposit in ophiolite-derived rock assemblage in the Finnish Outokumpu belt, located in eastern Finland. Our study utilizes a dense array dataset acquired in August-September 2016 within the ERA-MIN COGITO-MIN project. Our passive focal spot imaging extends the original objective of the project to produce structural imaging of the Kylylahti area using 2D seismic reflection profiles, 3D body-wave reflection seismic interferometry, and sparse 3D active-source survey. The array consisted of 994 stations that covered an area of 10.5 km² and featured lines spaced 200 m apart with 50 m receiver spacing. Each of the 994 stations consisted of six co-located 10 Hz vertical-component geophones and a data logger. Seismic noise was recorded at 500 Hz for 20 hours per day over a 30-day period, generating approximately 600 hours of passive seismic data. As part of the project, numerous active shots were conducted, which we found to disturb the noise records. To enhance the quality of the noise correlations by ensuring better compatibility with the diffusivity assumption of the ambient wavefield we remove the project-related shot data from the noise records using cataloged shot information. We decimate the daily records to 125 Hz and apply standard noise tomography pre-processing steps. We calculate the cross-correlations of the 1-hour time windows and stack them linearly to create 30 s maximum lag cross-correlations. The noise correlations are filtered using a Gaussian filter around the center frequency and the focal spot is parameterized with the SPAC Bessel function model to estimate the Rayleigh wave phase velocity. We present phase velocity maps produced at different periods, which allow us to characterize the geology of the Kylylahti mine area in the top kilometer. The effectiveness and accuracy of this method is demonstrated by comparing our results with previous active and passive imaging results.
How to cite: Hopiavuori, V., Tsarsitalidou, C., Kolehmainen, K., Hillers, G., Giammarinaro, B., Boué, P., Stehly, L., Malinowski, M., and Heinonen, S.: Rayleigh wave focal spot imaging of the Kylylahti ore deposit, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6190, https://doi.org/10.5194/egusphere-egu25-6190, 2025.
We assembled a large dataset of seismic waveforms with a non-uniform geographical distribution of permanent and temporary stations from regional-scale passive seismic experiments in central Europe, including the recent large-scale pan-European experiments AlpArray, PACASE and AdriaArray. This integrated dataset provides a viable source of observables for high-resolution travel-time surface wave tomography. However, the resolution and clarity of images resulting from various tomographic methods using different seismic waves are often compromised by the need of regularisation (i.e., damping and smoothing) to balance images from irregular input data distribution.
In this study, we developed a weighted regularisation scheme (WRS) for surface-wave travel-time tomography to reduce the bias caused by non-uniformly distributed data. The WRS is based on the implementation of ray-path coverage weights in localised azimuth-distance geographical segments in the inversion.
We present calibration of spatial weighting function on synthetic inter-station travel-times with regular ray-path distribution. On the integrated dataset with real ray-path distribution, we compare results of surface-wave travel-time tomography of synthetic travel times (calculated over the “spike” and “checkerboard” velocity fields) and observed travel times (derived from the cross-correlation delay times of the ambient noise) by applying:
- conventional regularisation approach (i.e., smoothing and damping)
- station de-selection to regular geographical grid, followed by conventional approach
- weighted regularisation scheme(WRS)
We show that the conventional surface-wave travel-time inversion is biased in favour of solutions in areas with the dense ray-path coverage and has poor resolution and reliability in areas with sparse ray-path coverage. The proposed WRS balances the inversion bias from the non-uniform distribution of travel-time measurements, allows the use of milder regularisation parameters (i.e., lower smoothing and damping), and resolves better broader areas with higher reliability.
How to cite: Kvapil, J. and Plomerová, J.: Weighted regularisation scheme for surface-wave tomography to mitigate bias caused by non-uniformly distributed ray-path coverage., EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16815, https://doi.org/10.5194/egusphere-egu25-16815, 2025.
Recent advances in tomographic methods, such as Helmholtz/Eikonal tomography, leverage seismic wavefields recorded by dense regional seismic networks to produce finely resolved images of the subsurface. This approach directly uses phase and amplitude measurements of coherent wavefronts generated by large earthquakes, bypassing the need to solve large tomographic inverse problem or compute synthetic seismograms. However, since these earthquakes are mainly located at plate boundaries, the azimuthal distribution of the sources is often limited. In addition, obtaining robust phase velocity maps typically requires years of continuous seismic data. In contrast, oceanic microseisms - responsible for ambient seismic noise - are more evenly distributed and generate coherent wavefronts continuously. In this study, we apply a matched filtering technique to iteratively extract these coherent surface wavefronts. Within a 4-hour time window, we typically extract 10 to 20 coherent wavefronts, from which we can measure arrival time and amplitude at each station. After removing outliers, phase measurements are interpolated using smoothing splines on either a Cartesian or a spherical grid, depending on the size of the domain under study. The gradient of the interpolated phase velocity surfaces is then used in the eikonal equation to generate phase velocity maps. These maps are stacked to produce average isotropic phase velocity maps for periods ranging from 6 to 25 s. We will present applications of this method in California and Western Europe.
How to cite: Delage, K. and Chevrot, S.: Eikonal tomography using Coherent surface Wavefronts extracted from Ambient seismic Noise with a Matched-Filtering Approach, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10052, https://doi.org/10.5194/egusphere-egu25-10052, 2025.
The detection of volcanic unrest is a critical component of volcanic monitoring and risk mitigation, especially for volcanoes with persistent hydrothermal activity and no recent eruptions. Identifying early signs of reactivation in such systems is particularly challenging due to the complex interplay between magmatic and hydrothermal processes.
Vulcano, one of the seven volcanic islands in the Aeolian archipelago (Southern Italy), is characterized by fumarolic activity and well-documented historical eruptions. The last eruptive event, occurring in 1888–1890 AD, featured episodic explosive activity of varying intensity, with the most violent explosions ejecting bombs and blocks over 1 km from the crater. Due to its small size and the proximity of active volcanic features to densely populated and tourist areas, Vulcano represents a critical site for volcanic risk management. In mid-September 2021, the island experienced significant degassing episodes at La Fossa cone, marking a period of unrest without eruptive activity. This unique scenario makes Vulcano an ideal natural laboratory for studying volcanic unrest in the absence of eruptions, providing valuable insights into the underlying magmatic-hydrothermal system.
Using continuous seismic records from Vulcano, we analyze relative seismic velocity changes (dv/v) through the cross-correlation of ambient noise, employing the MSNoise package \citep{lecocq2014msnoise}. Our analysis covers the pre-unrest, unrest, and post-unrest periods from 2016 to 2024, offering a long-term perspective on the temporal evolution of the system. Preliminary results show significant changes in dv/v that appear to be related to the 2021 episode of unrest, which corresponds to increased seismic activity, variations in gas emissions, and ground deformation. Long-term monitoring of Vulcano is crucial for identifying early signs of reactivation, which can significantly improve eruption forecasting and risk mitigation strategies. These findings highlight the potential of seismic noise analysis for real-time monitoring and to advance our understanding of the dynamics of volcanic unrest.
References
Lecocq, T., Caudron, C., & Brenguier, F. (2014). Msnoise, a python package for monitoring seismic velocity changes using ambient seismic noise. Seismological Research Letters, 85 (3), 715–726.
How to cite: Lo Bue, R., Cannavò, F., De Plaen, R., Lecocq, T., and Cannata, A.: Ambient Noise Analysis Reveals Seismic Velocity Changes during the 2021 Unrest at Vulcano Island, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8916, https://doi.org/10.5194/egusphere-egu25-8916, 2025.
Understanding the nonlinear elastic behavior of rocks has primarily been based on laboratory observations or numerical simulations. However, due to the inherent complexities and uncontrollable nature of real-world systems, field measurements of nonlinear elasticity remain a significant challenge. In-situ “Pump-Probe” type experiments, analogous to those conducted in laboratories, provide a valuable approach for characterizing the nonlinear mechanical properties of Earth materials. Environmental factors, such as tidal forces, hydrological loading, and thermal elasticity, serve as potential “pump” sources for these experiments. Seismic wave relative velocity changes (dv/v) are crucial proxies for investigating nonlinear elastic variations within the Earth's crust.
During the 36th Chinese Antarctic Expedition, we conducted a “Pump-Probe” type experiment near Dalk Glacier in East Antarctica. Over the course of one month, we collected ambient seismic noise data from the region. By reconstructing the noise cross-correlation functions (NCFs) from this seismic data, we applied coda wave interferometry to calculate dv/v.
This study further explores the hysteresis characteristics of dv/v in relation to strain, further analyzing the temporal delay effects of dv/v in response to various environmental factors, including tidal forces, temperature, humidity. Through time-domain analysis, we quantified the lag of the dv/v time series relative to the environmental parameters. In the frequency domain, we examined the diurnal and semi-diurnal variations in dv/v and their correlation with environmental factors, shedding light on the underlying mechanisms driving the observed fluctuations.
Additionally, we applied degree-day model and energy balance model to assess the melting dynamics of the ice sheet, allowing us to examine the response of dv/v to ice sheet melting. These findings contribute to a deeper understanding of the complex interactions between environmental factors and nonlinear elasticity, with potential implications for monitoring subsurface disturbances in polar regions.
How to cite: Bai, H., Feng, X., Fu, L., Zhou, H., Fehler, M., Brown, S., Zhang, M., and Nilot, E.: Monitoring Nonlinear Elasticity Near Dalk Glacier in East Antarctica, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-2265, https://doi.org/10.5194/egusphere-egu25-2265, 2025.
Seismic interferometry (SI) retrieves seismic responses by, for example, cross-correlating observations at different receiver locations, offering a new data for subsurface imaging. Monitoring soil conditions along railway embankments is essential for ensuring the stability and safety of rail infrastructure. This study explores the use of active and passive seismic sources, combined with SI technique, to enhance subsurface imaging.
Active seismic sources produce high-resolution reflections and surface-wave data, critical for identifying and monitoring key soil properties. These controlled sources provide superior signal-to-noise ratios and establish a reliable baseline for subsurface imaging. Passive seismic sources, such as train-induced vibrations, complement active-source data by providing continuous and natural excitation of the subsurface. Both active and passive sources are processed using SI technique, with adaptive subtraction applied to suppress dominant surface waves and improve imaging clarity.
The integration of active- and passive-source data might help achieve better interpretation of the subsurface and monitoring for possible changes. Although this study is still under development, it demonstrates the potential to deliver a scalable, non-invasive solution for monitoring railway embankments and surrounding soils. By advancing our understanding of subsurface conditions, this approach could contribute substantially to the predictive maintenance and safety of rail infrastructure, paving the way for future innovations in geophysical monitoring.
How to cite: Harajchi, S. and Draganov, D.: Application of Seismic Interferometry for Railway Embankment and Soil Monitoring, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3525, https://doi.org/10.5194/egusphere-egu25-3525, 2025.
Botswana is situated in central Southern Africa and is characterized by diverse geology, including prominent cratons such as the Congo and Kalahari cratons, as well as two sedimentary basins. Previous studies of the crustal structure beneath Botswana have primarily relied on traditional regional and teleseismic earthquake tomography. In this study, we use ambient seismic noise tomography to image the crustal structure of Botswana and its surrounding region. Using two years of seismic data (2019–2020) from 40 broadband stations including stations from the Botswana Seismological Network (BSN) and neighbouring regions, cross-correlation functions (CCFs) are computed and used to reconstruct surface waves propagating between station pairs. Dispersions of the surface waves are extracted and used to produce Rayleigh wave group and phase velocity maps of the region. Here we present the first results of Rayleigh wave group velocities and discuss these findings. This study aims to enhance our ability to image crustal structures in this low-seismicity region. The resulting velocity maps will contribute to the development of detailed 3D velocity models of Botswana’s crustal structure, providing new insights into the region’s subsurface structure and geodynamics. Future work will extend these results by integrating Love wave data and investigating crustal anisotropy.
How to cite: Ntibinyane, O., Qorbani Chegeni, E., and Bokelmann, G.: Mapping Botswana's Crustal Structure from Ambient Seismic Noise, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10849, https://doi.org/10.5194/egusphere-egu25-10849, 2025.
We investigate the internal properties and characteristics of the Mantle Transition Zone (MTZ) beneath the contiguous U.S. by detecting the presence of the 520-km discontinuity (d520) and determining its depth variation. Observations of d520 are performed using short-period reflected waves extracted from noise correlations (3-10 s period) employing a data selection strategy based on quantitative noise phase composition analysis. Detection of d520 is supported by analysis of the relative energy values between MTZ reflection phases. Our results reveal significant lateral variations in d520 depth, with deeper than average depths observed in the Western U.S. and shallower depths in the Eastern U.S. The Central U.S. exhibits transitional behavior. Analysis of relative phase energies indicates a strong d520 reflection phase across the central U.S., consistent with high seismic velocity contrasts and a likely higher olivine content in this region. In contrast, the Eastern U.S. shows a weak d520 reflection phase, potentially due to a gradual transition and/or potentially lower water content. The Western U.S., characterized by depressed d520 and d410 depths, likely reflects a warmer upper MTZ. The MTZ composition likely varies across the U.S., with potential basalt accumulation in the southwest due to past subduction.
How to cite: Aiman, Y. A., Lu, Y., Esteve, C., and Bokelmann, G.: Detailed structure within the Mantle Transition Zone beneath the contiguous U.S.: Insight from the 520-km discontinuity revealed by ambient noise correlations, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-9511, https://doi.org/10.5194/egusphere-egu25-9511, 2025.
The 2024 MW 7.5 Noto earthquake in Japan was preceded by an intense seismic swarm, likely driven by upward fluid migration. Crustal seismic velocities are influenced by external perturbations from earthquakes, with the presence of pressurized fluids in the crust amplifying these induced velocity changes. Consequently, we characterize subsurface fluids by examining the coseismic velocity changes associated with the 2024 mainshock. We combined data from permanent Hi-net and temporary seismic stations. The temporary stations were deployed by Tohoku University and Earthquake Research Institute, the University of Tokyo (1, 2, 3).
Our analysis reveals significant coseismic velocity drops, averaging ~0.5% in the near field (i.e., the Noto Peninsula) and reaching 0.6–0.8% near the coseismic slip peaks. These observed velocity drops correlate strongly with modeled velocity drops by coseismic static stress changes. Peak Ground Velocity (PGV) and Peak Ground Acceleration (PGA), which are proxies for dynamic stress changes, are also strongly correlated with the observed velocity drops. However, disentangling the contributions of static and dynamic stress changes to the observed velocity drops remains challenging due to their similar spatial patterns. In the far field (i.e., outside the Noto Peninsula), the coseismic velocity drops are on average ~0.1%, predominantly attributed to dynamic stress changes, as static stress changes are negligible at greater distances.
While the addition of temporary stations significantly enhances resolution in the pre-mainshock swarm zone, no significant coseismic velocity drop anomalies were detected in the shallow crust down to ~2.5 km depths. This suggests that the volume of pressurized fluids in the shallow crust is not anomalously large, implying that the fluid migration preceding the mainshock is likely confined to greater depths.
1: Sakai et al., https://doi.org/10.5281/zenodo.6767362, 2022
2: Okada et al., https://doi.org/10.1186/s40623-024-01974-0, 2024a
3: Okada et al., https://doi.org/10.5281/zenodo.10939231, 2024b
How to cite: Paris, N., Itoh, Y., Brenguier, F., Wang, Q.-Y., Sheng, Y., Okada, T., Uchida, N., Higueret, Q., Takagi, R., Sakai, S., Hirahara, S., and Kimura, S.: Coseismic crustal seismic velocity changes associated with the 2024 MW 7.5 Noto earthquake, Japan, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11265, https://doi.org/10.5194/egusphere-egu25-11265, 2025.
Passive seismic interferometry based on noise correlations has become an efficient way to detect tiny temporal changes of properties of Earth’s subsurface and crust. In particular, this method has successfully been used for environmental seismology issues, in a view of investigating the response of shallow subsurface to environmental changes, in particular thermal and hydrogeological forcings (1). An accurate spatial localization of elastic variations is thus a key objective to understand the processes involved in the whole volume probed.
Nevertheless, estimating the spatial distribution of these changes using coda waves is not a straightforward problem, regarding the complexity of scattered waves. Several assumptions are often used to simplify the coda waves and to estimate their sensitivity kernels to elastic changes (2). Here, our work focuses on the vertical localization of seismic variations observed by coda wave interferometry, reducing the problem to an estimation of the depth of shear stiffness variations over time in a 1D layered medium.
A time-lapse coda wave inversion scheme is commonly used, solving a linear least-square inverse problem (3) (4). Several other methods can also be tested, from Marko Chain Monte-Carlo Bayesian inversion to gradient descent algorithms. The choice of regularization parameters and checkerboard tests are discussed here, and help to discriminate the relevant method regarding the accuracy of the results and the vertical resolution obtained. Additionally, we applied these time-lapse inversion procedures to real seismic noise datasets recorded on slope instabilities such as rock glaciers and landslides.
This work illustrates how estimating the depth of seismic velocity changes contributes to characterizing the shallow subsurface and monitoring its sensitivity to various forcings.
References
- Richter, T., Sens‐Schönfelder, C., Kind, R., & Asch, G. (2014). Comprehensive observation and modeling of earthquake and temperature‐related seismic velocity changes in northern Chile with passive image interferometry. Journal of Geophysical Research: Solid Earth, 119(6), 4747-4765
- Obermann, A., & Hillers, G. (2019). Seismic time-lapse interferometry across scales. In Advances in geophysics(Vol. 60, pp. 65-143). Elsevier.
- Mordret, A., Courbis, R., Brenguier, F., Chmiel, M., Garambois, S., Mao, S., ... & Hollis, D. (2020). Noise-based ballistic wave passive seismic monitoring–Part 2: surface waves. Geophysical Journal International, 221(1), 692-705.
- Fokker, E., Ruigrok, E., Hawkins, R., & Trampert, J. (2023). 4D physics‐based pore pressure monitoring using passive image interferometry. Geophysical Research Letters, 50(5), e2022GL101254.
How to cite: Guillemot, A., Larose, E., and Baillet, L.: Assessing the vertical localization of seismic variations retrieved by passive surface wave interferometry, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15116, https://doi.org/10.5194/egusphere-egu25-15116, 2025.
Ambient noise interferometry has emerged as a powerful technique for tracking changes in volcanic systems by providing continuous data on seismic velocity variations (dv/v). Due to its sensitivity to structural variations (e.g., crack opening or closing), magma intrusions, and external factors like precipitation, this method offers valuable insights into dynamic volcanic processes. Here, we use ambient noise interferometry to study the September 2024 East Rift Zone (ERZ) eruption of Kīlauea volcano. This eruption and the preceding intrusions were captured by a dense nodal seismic network of 116 stations distributed over an area of 30×60 km in the ERZ. With this new dataset we investigate magma-tectonic interactions by monitoring seismic velocity variations (dv/v) across space and time. These variations, driven by both internal factors (magmatic and tectonic activity) and external forces (e.g. precipitation), were contextualized using complementary geophysical observations, including data from tiltmeters, GPS stations, and precipitation sensors maintained by the Hawaiian Volcano Observatory. Initial results revealed a velocity decrease of up to 1% during the July 2024 intrusion event, correlating with increased seismicity and deformation observed at the summit and in the Middle East Rift Zone. This study highlights how the integration of various geophysical observations can improve the understanding of volcanic precursors and magma dynamics and demonstrates the potential of dense seismic networks for volcano monitoring.
How to cite: Rohnacher, A. and Lanza, F.: Ambient noise interferometry as a tool for volcanic system monitoring: insights from the 2024 Kīlauea East Rift Zone eruption , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15322, https://doi.org/10.5194/egusphere-egu25-15322, 2025.
