Upcoming Virtual Tomography Sessions

2 March Presentations, 10 a.m.–12:40 Pacific

10–10:20 a.m. Pacific

“An Indian Ocean Cluster of Mantle Plumes Imaged by Multifrequency P-Wave Tomography.” Maria Tsekhmistrenko, DIAS

10:20–10:40 a.m. Pacific

“Defining a New A-Priori Velocity Model for the Seismic Tomography Analysis of South Italy.” Cristina Totaro, University of Messina

10:40–11 a.m. Pacific

“Fault Damage and Pore Fluid Distribution Regionally around Kaikoura, New Zealand from Body-Wave Tomography.” Ben Heath University of Wisconsin–Madison

11–11:10 a.m. Pacific


11:10–11:30 a.m. Pacific

“Constraints on the Geometry of the Subducted Gorda Plate From Converted Phases Generated by Local Earthquakes.” Jianhua Gong, MIT/WHOI Joint Program

11:30–11:50 a.m. Pacific

“Level-Set Imaging of the Los Angeles Basin using the Hierarchical Ensemble Kalman Sampling.” Jack Muir, California Institute of Technology

11:50 a.m.–12:10 p.m. Pacific

“Getting to the Target with Marchenko-based Virtual Sources and Receivers.” Joeri Brackenhoff, Delft University of Technology

12:10–12:40 p.m. Pacific

Breakout Rooms

Thank you to SmartSolo for sponsoring the Virtual
Tomography series!



“Getting to the Target with Marchenko-based Virtual Sources and Receivers.” Joeri Brackenhoff, Delft University of Technology (j.a.brackenhoff@tudelft.nl); C.P.A. Wapenaar, Delft University of Technology (c.p.a.wapenaar@tudelft.nl)

In many applications of seismic methods, a frequent issue is the presence of events that have scattered more than once in the subsurface, commonly referred to as internal multiples. The problem is that many methods cannot distinguish the internal multiples from the events that have scattered only once. This can cause artifacts in the final result of a method, especially for deep targets.

In recent years, the Marchenko method has been employed to deal with this kind of internal multiples. The method employs reflection data, which are data that are measured using active sources and receivers located at the surface of the Earth, to simulate sources and receivers in the subsurface. These simulated sources and receivers are referred to as virtual sources and receivers. In order to achieve this, the only information that is required is the first arriving event from the virtual source or receiver position in the subsurface to the Earth’s surface. Such information can be obtained from a background velocity model. The main advantages of the method are that a virtual source or receiver can be created at any point in the subsurface without the need to resolve the overburden above the target. Other methods can achieve similar results for primaries, however the Marchenko method can handle the internal multiples properly.

We will present applications of seismic methods using the Marchenko method, namely the creation of virtual sources and receivers for the purpose of imaging the subsurface and monitoring wavefields in the subsurface. We will first demonstrate this on numerical data, using a model that contains an overburden with strong scatterers above a target with weak scattering energy. We will show the difference between the results achieved with the Marchenko method and achieved by a method that does not handle the internal multiples. We will show the application of these methods to field data, which have been pre-processed in order to apply the Marchenko method.

“Constraints on the Geometry of the Subducted Gorda Plate From Converted Phases Generated by Local Earthquakes.” Jianhua Gong, MIT/WHOI Joint Program

The largest slip in great megathrust earthquakes often occurs in the 10–30 km depth range, yet seismic imaging of the material properties in this region has proven difficult. We utilize a dense onshore-offshore passive seismic dataset from the southernmost Cascadia subduction zone where seismicity in the mantle of the subducted Gorda Plate produces S-to-P and P-to-S conversions generated within a few km of the plate interface. These conversions typically occur in the 10–20 km depth range at either the top or bottom of a ~5 km thick layer with a high Vp/Vs that we infer to be primarily the subducted crust. We use their arrival times and amplitudes to infer the location of the top and bottom of the subducted crust as well as the velocity contrasts across these discontinuities. Comparing with both the Slab1.0 and the updated Slab2 interface models, the Slab2 model is generally consistent with the converted phases, while the Slab1.0 model is 1–2 km deeper in the 2–20 km depth range and ~6–8 km too deep in the 10–20 km depth range between 40.25°N and 40.4°N. Comparing the amplitudes of the converted phases to synthetics for simplified velocity structures, the amplitude of the converted phases requires models containing a ∼5 km thick zone with at least a ~10%–20% reduction in S wave velocity. Thus, the plate boundary is likely contained within or at the top of this low velocity zone, which potentially indicates a significant porosity and fluid content within the seismogenic zone.

“Fault Damage and Pore Fluid Distribution Regionally around Kaikoura, New Zealand from Body-wave Tomography.” Ben Heath University of Wisconsin – Madison (baheath@wisc.edu); Donna Eberhart-Phillips, University of California, Davis

Individual earthquake ruptures are usually assumed to occur on individual faults and are often associated with narrow regions (< 5 km) of altered physical properties, such as areas of increased fracturing and/or increased pore fluids. Recently, earthquakes such as the 2016 Kaikoura, New Zealand earthquake have ruptured multiple faults with different orientations over regions with widths spanning > 25 km. We test whether such regions hosting these earthquakes are associated with anomalous physical properties. We use seismic arrival-time tomography in the Kaikoura region to investigate lateral variations in Vp and Vp/Vs, using these parameters to infer variation in crustal faulting/fracturing. By modeling the effect of fluid-filled fractures on lateral variations in Vp and Vp/Vs, we are able to attribute the lateral variation in seismic velocities (over scales of > 50 km) to fault damage and pore fluid distribution. We find that the immature fault zones ruptured during the Kaikoura earthquake are on average characterized by decreased Vp and elevated Vp/Vs, features that decay (over distances of 50 km) towards background levels with increased distance from Kaikoura earthquake faults (and increased proximity to more mature fault zones). Drops in Vp in the Kaikoura rupture region are found to linearly relate to increases in Vp/Vs at a rate that is consistent with elevated 0.01 aspect ratio fractures, with highest fracturing within 10 km of the ruptured faults. The broad regional fracture distribution is likely the result of distributed long-term deformation, with increased deformation in the Kaikoura region. In contrast to more mature fault zones, which have localized strain accommodation and limited regional fracture distribution, immature fault zones are characterized by broadened, extensive fractures which contribute to complicated rupture dynamics.

“Level-set Imaging of the Los Angeles Basin using the Hierarchical Ensemble Kalman Sampling.” Jack Muir, California Institute of Technology (jmuir@caltech.edu); Robert W Clayton, Caltech (clay@gps.caltech.edu); Victor C Tsai, Brown University (victor_tsai@brown.edu)

A key challenge facing modern tomographic analyses is the interpretability of the resulting images, both in terms of the geological features obtained from imaging, and in how the results of uncertainty quantification modulate our confidence in the results. An innovative choice of model parametrization that improves both the interpretability and robustness of seismic tomography is therefore highly desirable. We have developed a framework employing level-set parametrizations that seeks to define geological units, which we invert using a recently developed Hierarchical Tikhonov Ensemble Kalman Sampling scheme, a highly efficient derivative-free optimizer which includes uncertainty estimates and optimization of regularization parameters. We have applied this scheme in an effort to better map earthquake hazard and improve velocity models within the Los Angeles Basin by generating a local update to the SCEC CVMS-4.26 model. We utilize high resolution data from the Community Seismic Network, a 400-station permanent urban deployment, to invert Love-wave dispersion, derived from eikonal tomography of two-station cross-correlation travel-time delays, and relative amplification data from the Mw 7.1 July 5 2019 Ridgecrest Earthquake. We find that the data is best explained by a deepening of the LA Basin (compared to the CVMS-4.26 reference model) along its Northwest-Southeast axis relative to its deepest point, just south of downtown LA. Additionally, the deeper basin edge extends further to the East of downtown LA towards East Los Angeles. This result offers new progress towards the parsimonious incorporation of detailed local basin models within regional reference models utilizing an objective inverse-problem framework with uncertainty quantification, and highlights the importance of accurate basin geometry models when accounting for the potentially significant amplification of surface waves from regional earthquakes in the high-rise building frequency band.

“Defining a New A-priori Velocity Model for the Seismic Tomography Analysis of South Italy.” Cristina Totaro, University of Messina (ctotaro@unime.it); Giancarlo Neri, University of Messina (geoforum@unime.it); Barbara Orecchio, University of Messina (orecchio@unime.it);  Debora Presti, University of Messina (dpresti@unime.it); Silvia Scolaro, University of Messina, (silscolaro@unime.it)

By integrating data and constraints available in the literature we estimated a new “a-priori” 3D seismic velocity model depicting the lithospheric structure of South Italy, a highly complex area of the Mediterranean region characterized by the coexistence of Africa-Europe NNW-trending plate convergence and SE-ward residual rollback of the Ionian lithospheric slab subducting underneath the Tyrrhenian. The integration of data like velocity values from seismic profiles and/or tomographies, moho depth, and subduction interface geometry, is performed with a procedure derived to that already successfully applied in the area about a decade ago and it is aimed to furnish the simplest 3D velocity structure consistent with all the available data. Taking benefit from studies and analyses of the last decade we enlarged and improved the previous estimated model by adding further data and useful constraints. The so obtained “a-priori” velocity model has then been employed as starting model for the local earthquake tomography of the region. For the tomographic study a set of ca. 10000 earthquakes has been used by selecting from the Italian seismic database (www.ingv.it) all the M≥2 earthquakes that occurred in the time period 2000-2020 at depth less than 60 km and with at least 10 station readings. The obtained 3D velocity structure together with the related hypocenter locations have been interpreted in the frame of the geodynamic models proposed for the region.

“An Indian Ocean Cluster of Mantle Plumes Imaged by Multifrequency P-wave Tomography.” Maria Tsekhmistrenko, DIAS (mariat@cp.dias.ie); Karin Sigloch, University of Oxford (karin.sigloch@earth.ox.ac.uk); Kasra Hosseini, The Alan Turing Institute (khosseini@turing.ac.uk); Guilhem Barruol, IPGP CNRS (barruol@ipgp.fr)

Mantle plumes are commonly envisioned as thin, buoyant conduits rising vertically from the core mantle boundary (CMB) to the earth’s surface, where they produce volcanic hot spots. Most hotspots are located in the sparsely instrumented oceans, creating poor prospects for the seismic resolution of thin conduits in the deep mantle.

The RHUM-RUM experiment remedied this issue around the hotspot island of La Réunion by instrumenting 2000km x 2000km of seaoor for 13 months with 57 broadband ocean bottom seismometers (OBS). We present a 3-D P-wave tomography model that was computed from the RHUM-RUM waveform data, supplemented by a global data set of P-diffracted measurements, and by a selection of ISC picks. Multifrequency traveltimes were measured on the waveforms and inverted in a finite-frequency framework. We achieve high image resolution beneath the Indian Ocean hemisphere, and especially beneath La Réunion, from upper mantle to CMB.
We observe the Large Low Velocity Province (LLVP) rising 800 km above the CMB, forming a cusp beneath South Africa. A low-velocity branch undulates obliquely from this cusp region towards the uppermost mantle beneath La Réunion. Hence La Réunion’s connection to the lower mantle is more complex than previously envisioned, being neither a thin vertical conduit nor projecting down to an edge of the LLVP. The deep-mantle connections of the Afar and Kerguelen hotspots emerge from the same LLVP cusp beneath South Africa and extend towards the surface through tilted low- velocity branches.

Our results provide the first high-resolution image of a western Indian Ocean plume cluster, from the surface to the CMB. This represents a key advance for linking geophysical, geodynamic and geochemical observations.”