My research interests include seismic data processing, imaging, and modelling techniques, applied to volcanic structures and fault zones. Another topic is the analysis of induced seismicity in geothermal reservoirs. Software development is an important part of my work.
2010–2012: Contract Researcher at AMRA Scarl, Naples, Italy (RISSC Lab) 2006–2010: Contract Researcher at University of Naples Federico II, Department of Physics, Naples, Italy (RISSC Lab) 2004–2005: Post-doc Research Fellow at NORSAR, Kjeller, Norway 2000–2003: Scientist (part-time) at GeoForschungsZentrum (GFZ), Potsdam, Germany
Abstract: High-resolution imaging with microseismic events requires the use of large and consistent data sets of seismic phase arrival times. In particular the S-phase is important to derive physical parameters of the subsurface. Typically this phase is identified on one of the horizontal seismogram components by a change of signal amplitude and frequency as compared to the previous P-phase. However, reliable S-phase identification can be difficult for local events because of a signal overlap with the P coda, the presence of converted phases, and possible S-wave splitting due to anisotropy. In this study we propose a new data processing technique aiming at uniquely identifying the S phase arrival using all available records from a seismic network. The technique combines polarization analysis of single three components recordings of an event with analysis of lateral waveform coherence across the network. This makes it possible to construct seismic sections in which the first arrival is the S-phase. This graphical representation can support an operator in both the analysis of single events and in semi-automatic analyses of large datasets. In addition, an automated stacking velocity analysis provides S-wave velocities from these sections. We demonstrate the applicability of this technique using synthetic seismograms, and we evaluate the efficacy on a dataset of three-component velocimeter records from local earthquakes of the Campania-Lucania Apennines (southern Italy) recorded by the Irpinia Seismic Network (ISNet).
Abstract: The Mw 6.3 L'Aquila earthquake, central Italy, on April 6, 2009 has been recorded by the Irpinia Seismic Network (ISNet) about 250 km south-east of the epicenter. Up to 19 three-component accelerometer stations could be used to infer the main source parameters with different seismological methods. We obtained an approximate location of the event from arrival times and array-based backazimuth measurements and estimated the local magnitude (6.1) from an attenuation relation for southern Italy. Assuming an omega-square spectral model we inverted S-wave displacement spectra for moment magnitude (6.3), corner frequency (0.33 Hz), stress drop (2.5 MPa) and apparent stress (1.6 MPa). Waveform modeling using a point source and an extended source model provided consistent moment tensors with a centroid depth around 6 km and a prevalently normal fault plane solution with a dominant directivity toward south-east. The relatively high corner frequency and an overestimated moment magnitude of 6.4 from moment tensor inversions are attributed to the rupture directivity effect. To image the rupture geometry we implemented a beamforming technique that back-projects the recorded direct P-wave amplitudes into the earthquake source region. A NW-SE striking rupture of 17 km length is imaged, propagating with an average velocity up to 3 km/s. This value is significantly higher than our estimate of 2.2 km/s from S-wave spectra. Our case study demonstrates that the use of array techniques and a dense accelerometer network can provide quick and robust estimates of source parameters of moderate-size earthquakes located outside the network.
Abstract: Fault zones are the locations where motion of tectonic plates, often associated with earthquakes, is accommodated. Despite a rapid increase in the understanding of faults in the last decades, our knowledge of their geometry, petrophysical properties, and controlling processes remains incomplete. The central questions addressed here in our study of the Dead Sea Transform (DST) in the Middle East are as follows: (1) What are the structure and kinematics of a large fault zone? (2) What controls its structure and kinematics? (3) How does the DST compare to other plate boundary fault zones? The DST has accommodated a total of 105 km of left-lateral transform motion between the African and Arabian plates since early Miocene (~20 Ma). The DST segment between the Dead Sea and the Red Sea, called the Arava/Araba Fault (AF), is studied here using a multidisciplinary and multiscale approach from the micrometer to the plate tectonic scale. We observe that under the DST a narrow, subvertical zone cuts through crust and lithosphere. First, from west to east the crustal thickness increases smoothly from 26 to 39 km, and a subhorizontal lower crustal reflector is detected east of the AF. Second, several faults exist in the upper crust in a 40 km wide zone centered on the AF, but none have kilometer-size zones of decreased seismic velocities or zones of high electrical conductivities in the upper crust expected for large damage zones. Third, the AF is the main branch of the DST system, even though it has accommodated only a part (up to 60 km) of the overall 105 km of sinistral plate motion. Fourth, the AF acts as a barrier to fluids to a depth of 4 km, and the lithology changes abruptly across it. Fifth, in the top few hundred meters of the AF a locally transpressional regime is observed in a 100-300 m wide zone of deformed and displaced material, bordered by subparallel faults forming a positive flower structure. Other segments of the AF have a transtensional character with small pull-aparts along them. The damage zones of the individual faults are only 5-20 m wide at this depth range. Sixth, two areas on the AF show mesoscale to microscale faulting and veining in limestone sequences with faulting depths between 2 and 5 km. Seventh, fluids in the AF are carried downward into the fault zone. Only a minor fraction of fluids is derived from ascending hydrothermal fluids. However, we found that on the kilometer scale the AF does not act as an important fluid conduit. Most of these findings are corroborated using thermomechanical modeling where shear deformation in the upper crust is localized in one or two major faults; at larger depth, shear deformation occurs in a 20-40 km wide zone with a mechanically weak decoupling zone extending subvertically through the entire lithosphere.
Abstract: We present a method to estimate seismic velocity and density contrasts at a given interface in a 1-D layered model using PS-to-PP reflection amplitude ratios. The velocity structure above the reflector is constrained by travel time modelling, and the amplitude ratios are determined using using the same source-receiver pair for the measured PP and PS amplitudes (common-offset geometry). Thereby, source and receiver site effects are cancelled, and the remaining propagation effects are included in the ray-theoretical forward modelling of theoretical PS-to-PP ratios. A minimisation of the least-squares misfit between observed and modelled ratios provides the remaining elastic parameters below the reflector of interest (P-velocity, P-to-S velocity ratio, density). 1-D examples and a 2-D synthetic case study with a dipping reflector and a laterally varying overburden demonstrate the possibilities and limitations of the method. An application of the method to a 0.6 km deep reflector below the Campi Flegrei caldera, Italy, reveals a strong contrast with a P-velocity increase from less than 2 km/s to 3.5 km/s and a decrease of the P-to-S velocity ratio from 3.6 to 1.75. The proposed PS-to-PP amplitude ratio analysis is applicable for wide-angle seismic reflection data, especially when strong elastic parameter contrasts are expected and when source amplitudes or site effects are poorly known.
Abstract: Campi Flegrei is an active, resurgent caldera that is located a few kilometres west of the city of Naples, a densely populated urban settlement in southern Italy. Identifying, locating at depth and better defining the geometry of the magma feeding system of the caldera is highly relevant for assessing and monitoring its volcanic hazard. Based on a high resolution seismic reflection dataset, we investigated the deep structure of the volcano. Here we show that seismic wave amplitude variations with distance from the radiating source provide clear evidence for large amplitude seismic reflections from the top of an extended gas- and/or brine-bearing rock formation at about 3000 m and of an about 7500 m deep, low velocity layer, which is associated with a mid-crust, partial melting zone beneath the caldera. The modeling of magma properties based on measured seismic velocities indicates a relatively high melt percentage (in the range 80-90%). These new data suggest that a large magmatic sill is present well within the basement formations, which is possibly linked to the surface through a system of deep fractures bordering the caldera. The lateral extension and similar depth of the melt zone observed beneath the nearby Mt. Vesuvius support the hypothesis of a single continuous magma reservoir feeding both of these volcanoes.
Abstract: BARENTS50, a new 3-D geophysical model of the crust in the Barents Sea Region has been developed by the University of Oslo, NORSAR and the U.S. Geological Survey. The target region comprises northern Norway and Finland, parts of the Kola Peninsula and the East European lowlands. Novaya Zemlya, the Kara Sea and Franz-Josef Land terminate the region to the east, while the Norwegian-Greenland Sea marks the western boundary. In total, 680 1-D seismic velocity profiles were compiled, mostly by sampling 2-D seismic velocity transects, from seismic refraction profiles. Seismic reflection data in thewestern Barents Sea were further used for density modelling and subsequent density-to-velocity conversion. Velocities from these profiles were binned into two sedimentary and three crystalline crustal layers. The first step of the compilation comprised the layer-wise interpolation of the velocities and thicknesses. Within the different geological provinces of the study region, linear relationships between the thickness of the sedimentary rocks and the thickness of the remaining crystalline crust are observed. We therefore used the separately compiled (area-wide) sediment thickness data to adjust the total crystalline crustal thickness according to the total sedimentary thickness where no constraints from 1-D velocity profiles existed. The BARENTS50 model is based on an equidistant hexagonal grid with a node spacing of 50 km. The P-wave velocity model was used for gravity modelling to obtain 3-D density structure. A better fit to the observed gravity was achieved using a grid search algorithm which focussed on the density contrast of the sediment-basement interface. An improvement compared to older geophysical models is the high resolution of 50 km. Velocity transects through the 3-D model illustrate geological features of the European Arctic. The possible petrology of the crystalline basement in western and eastern Barents Sea is discussed on the basis of the observed seismic velocity structure.
Notes: BARENTS50 is available at http://www.norsar.no/c-90-Barents-Sea-3D-Model.aspx
Abstract: In a high-resolution small scale seismic experiment we investigated the shallow structure of the Wadi Araba Fault (WAF), the principal fault strand of the Dead Sea Transform System between the Gulf of Aqaba/Eilat and the Dead Sea. The experiment consisted of 8 sub-parallel 1 km long seismic lines crossing the WAF. The recording station spacing was 5 meters and the source point distance was 20 m. The first break tomography yields insight into the fault structure down to a depth of about 200 m. The velocity structure varies from one section to the other which were 1 to 2 km apart, but distinct velocity variations along the fault are visible between several profiles. The reflection seismic images show positive flower structures and indications for different sedimentary layers at the two sides of the main fault. Often the superficial sedimentary layers are bent upward close to the WAF. Our results indicate that this section of the fault (at shallow depths) is characterized by a transpressional regime. We detected a 100 to 300 m wide heterogeneous zone of deformed and displaced material which, however, is not characterized by low seismic velocities at a larger scale. At greater depth the geophysical images indicate a blocked cross-fault structure. The structure revealed, fault cores not wider than 10 m, are consistent with scaling from wear mechanics and with the low loading to healing ratio anticipated for the fault.
Abstract: Magnetotelluric and seismic methods provide complementary information about the resistivity and velocity structure of the subsurface on similar scales and resolutions. No global relation, however, exists between these parameters and correlations are often valid for only a limited target area. Independently derived inverse models from these methods can be combined, using a classification approach, to map geologic structure. The method employed is based solely on the statistical correlation of physical properties in a joint parameter space and is independent of theoretical or empirical relations linking electrical and seismic parameters. Regions of high correlation (classes) between resistivity and velocity can in turn be mapped back and reexamined in depth section. The spatial distribution of these classes, and the boundaries between them, provide structural information not always evident in the individual models. This method is applied to a 10 km long profile crossing the Dead Sea Transform in Jordan. Several prominent classes are identified with specific lithologies in accordance with local geology. An abrupt change in lithology across the fault, together with vertical uplift of the basement suggest the fault is sub-vertical within the upper crust.
Abstract: Seismic tomography, imaging of seismic scatterers, and magnetotelluric soundings reveal a sharp lithologic contrast along a 10 km long segment of the Arava Fault (AF), a prominent fault of the southern Dead Sea Transform (DST) in the Middle East. Low seismic velocities and resistivities occur on its western side and higher values east of it, and the boundary between the two units coincides partly with a seismic scattering image. At 1-4 km depth the boundary is offset to the east of the AF surface trace, suggesting that at least two fault strands exist, and that slip occurred on multiple strands throughout the margin's history. A westward fault jump, possibly associated with straightening of a fault bend, explains both our observations and the narrow fault zone observed by others.
Abstract: The Barents Sea and its surroundings is an epicontinental region which previously has been difficult to access, partly because of its remote Arctic location and partly because the region has been politically sensitive. Now, however, this region, and in particular its western parts, has been very well surveyed with a variety of geophysical studies, motivated in part by exploration for hydrocarbon resources. Since this region is interesting geophysically as well as for seismic verification, a major study was initiated in 2003 to develop a three-dimensional (3-D) seismic velocity model for the crust and upper mantle, using a grid density of 50 km.
This study, in cooperation between NORSAR, the University of Oslo (UiO), and the United States Geological Survey (USGS), has led to the construction of a higher-resolution, regional lithospheric model based on a comprehensive compilation of available seismological and geophysical data. Following the methodology employed in making the global crustal model CRUST5.1, the new model consists of five crustal layers: soft and hard sediments, and crystalline upper, middle, and lower crust. Both P- and S-wave velocities and densities are specified in each layer. In addition, the density and seismic velocity structure of the uppermost mantle, essential for Pn and Sn travel time modeling, are included.
Notes: The model is available at http://www.norsar.no/c-90-Barents-Sea-3D-Model.aspx
Abstract: To address one of the central questions of plate tectonics – How do large transform systems work and what are their typical features? – seismic investigations across the Dead Sea Transform (DST), the boundary between the African and Arabian plates in the Middle East, were conducted for the first time. A major component of these investigations was a combined reflection/refraction survey across the territories of Palestine, Israel and Jordan. The main results of this study are: (1) The seismic basement is offset by 3-5 km under the DST, (2) The DST cuts through the entire crust, broadening in the lower crust, (3) Strong lower crustal reflectors are imaged only on one side of the DST, (4) The seismic velocity sections show a steady increase in the depth of the crust-mantle transition (Moho) from ~26 km at the Mediterranean to ~39 km under the Jordan highlands, with only a small but visible, asymmetric topography of the Moho under the DST. These observations can be linked to the left-lateral movement of 105 km of the two plates in the last 17 Myr, accompanied by strong deformation within a narrow zone cutting through the entire crust. Comparing the DST and the San Andreas Fault (SAF) system, a strong asymmetry in subhorizontal lower crustal reflectors and a deep reaching deformation zone both occur around the DST and the SAF. The fact that such lower crustal reflectors and deep deformation zones are observed in such different transform systems suggests that these structures are possibly fundamental features of large transform plate boundaries.
Abstract: With controlled seismic sources and specifically designed receiver arrays, we image a subvertical boundary between two lithological blocks at the Arava Fault (AF) in the Middle East. The AF is the main strike-slip fault of the Dead Sea Transform (DST) in the segment between the Dead Sea and the Red Sea. Our imaging (migration) method is based on array beamforming and coherency analysis of P-to-P scattered seismic phases. We use a 1-D background velocity model and the direct P arrival as a reference phase. Careful resolution testing is necessary, because the target volume is irregularly sampled by rays. A spread function describing energy dispersion at localised point scatterers and synthetic calculations for large planar structures provides resolution estimates of the images. We resolve a 7 km long steeply dipping reflector offset roughly 1 km from the surface trace of the AF. The reflector can be imaged from about 1 km down to 4 km depth. Previous and ongoing studies in this region have shown a strong contrast across the fault: low seismic velocities and electrical resistivities west and high velocities and resistivities east of it. We therefore suggest that the imaged reflector marks the contrast between young sedimentary fill in the west and Precambrian rocks in the east. If correct, the boundary between the two blocks is offset about 1 km east of the current surface trace of the AF.
Abstract: On several recordings of linear seismometer arrays crossing the Arava Fault (AF) in the Middle East, we see prominent wave trains emerging from in-fault explosions which we interpret as waves being guided by a fault zone related low-velocity layer. The AF is located in the Arava Valley and is considered the principal active fault of the mainly N-S striking Dead Sea Transform System in this section. Observations of these wave trains are confined to certain segments of the receiver lines and occur only for particular shot locations. They exhibit large amplitudes and are almost monochromatic. We model them by a two-dimensional (2-D) analytical solution for the scalar wave field in models with a vertical waveguide embedded in two quarter spaces. A hybrid search scheme combining genetic algorithm and a local random search is employed to explore the multimodal parameter space. Resolution is investigated by synthetic tests. The observations are adequately fit by models with a narrow, only 3-12 m wide waveguide with S wave velocity reduced by 10-60% of the surrounding rock. We relate this vertical low-velocity layer with the damage zone of the AF since the location of receivers observing and of shots generating the guided waves, respectively, match with the surface trace of the fault. The thickness of the damage zone of the AF, at least at shallow depths, seems to be much smaller than in other major fault zones. This could be due to less total slip on this fault.
Abstract: High-resolution seismic tomography and magnetotelluric (MT) soundings of the shallow crust show strong changes in material properties across the Dead Sea Transform Fault (DST) in the Arava valley in Jordan. 2D inversion results of the MT data indicate that the DST is associated with a strong lateral conductivity contrast of a highly conductive layer at a depth of approximately 1.5 km cut-off at a position coinciding with the surface trace of the DST. At the same location, we observe a sharp increase of P wave velocities from <4 km/s west of the fault to >5 km/s to the east. The high velocities in the east probably reflect Precambrian rocks while the high electrical conductivity west of the DST is attributed to saline fluids within the sedimentary filling. In this sense, the DST appears to act as an impermeable barrier between two different rock formations. Such a localized fluid barrier is consistent with models of fault zone evolution but has so far not been imaged by geophysical methods. The situation at the DST is remarkably different from active segments of the San Andreas Fault which typically show a conductive fault core acting as a fluid conduit.
Abstract: The Dead Sea Rift Transect (DESERT 2000) is a multinational and interdisciplinary study of the Dead Sea Rift. The project began field work in February 2000 and the first experiments were successfully completed in May. The seismic, seismological, and magnetotelluric experiments presented here, along with the future electromagnetic, gravity, magnetic, geodynamic, and geological studies, will provide the basic geophysical frame for further geoscientific research. DESERT 2000 should also help to address a fundamental question of plate tectonics: How do shear zones work and what controls them?
Abstract: Here, we propose a method for the determination of Vp/Vs ratios in a horizontally layered propagation media using maximization of a coherency function along theoretical travel-times of PS reflected phases. The theoretical travel-times are computed using the information about the propagation media that is extracted by velocity analysis or by topographic analysis performed on the first arrivals. The method is also a valid tool for the identification of the PS phases associated with a fixed seismic reflector, and it is particularly suitable for data that is stored in common mid-point and common conversion point binning; for this kind of data the hypothesis of horizontally and layered media can usually be verified.
We applied the method to both simulated and real datasets. The use of the real data that was acquired in the Campi Flegrei caldera (southern Italy) allowed us to estimate a relatively high Vp/Vs ratio (3.5±0.6) for a very shallow layer (maximum depth 600 m). This hypothesis has been tested by theoretical rock physical modeling of the Vp/Vs ratios as a function of porosity, suggesting that the shallow layer appears to be formed of unconsolidated, water-saturated, volcanic and marine sediments that filled Pozzuoli Bay during the post-caldera activity.
Abstract: Elastic parameters derived from seismic reflection data provide information on the lithological contrast at an interface and support the geological interpretation. We present a method to estimate elastic parameter contrasts at a given interface in a 1-D layered medium from PS-to-PP amplitude ratios. The method is applied to synthetic data to demonstrate its possibilities and limitations. First results for real data acquired in the Campi Flegrei caldera (southern Italy) reveal a gas-bearing layer at around 3 km depth and indicate a strong negative velocity contrast at 7.5 km depth, possibly related to the presence of partial melt.
Abstract: The shallow crustal structure of the Campi Flegrei caldera, southern Italy, is imaged with P-to-P scattered seismic waves. The imaging (migration) method is based on array beamforming and coherence analysis of the scattered phase. Airgun shots from a controlled-source seismic experiment are grouped into arrays, and a 1-D background velocity model is used to calculate travel times for potential scatterer locations on a regular grid in the image volume. A high coherence measured on aligned traces indicates the presence of a scatterer at the corresponding location. A stack of the individual images from several pairs of a shot array and a single receiver provides the final image. First results show a prominent region of seismic scattering that coincides with the buried caldera rim.
Abstract: A new dataset of surface-wave observations from more than 150 local and regional events with travel paths through the greater Barents Sea region was compiled and group-velocity dispersion curves were measured for Love and Rayleigh waves in the period range 14-160 s (c.f. Part I: Levshin et al., 2005). This large amount of new group-velocity measurements was used to enlarge the already existing data base compiled at the University of Colorado and to increase the path density in the region under investigation (Levshin et al., 2001). This combined data set of group-velocity observations was inverted for 2D group-velocity maps (Barmin et al., 2001; Ritzwoller et al., 2002; Levshin et al., 2005). Pasyanos (2005) recently published another set of group-velocity maps for Eurasia and the European Artic, which shows very similar large scale features in the greater Barents Sea region, but which have far less resolution for smaller scale anomalies in that area.
Abstract: This short contribution is a description of data now available in NORSAR's data archives from a temporary deployment during 2002-2004 of six seismic stations in northern Norway and Finland.
Explosions in underground as well as open-pit mines in the Khibiny massif of the Kola Peninsula of northwestern Russia are conducted on a frequent and relatively regular basis. It was decided to supplement the network of permanent stations in northern Fennoscandia and northwest Russia with temporarily deployed stations, in order to record these explosions, as well as other mining explosions and natural events occuring in this general area. The six temporary stations were deployed along two profile lines, extending westwards from the Khibini massif. The rationale for this deployment was to collect data to examine distance as well as azimuthal dependence of seismic discriminants. The southernmost of the two profile lines runs through the permanent seismic array ARCES in northern Norway.
Abstract: It goes without saying that planet Earth is dynamic – the actual shape of the Earth's outer shell is continuously being formed and changed through processes of plate tectonics. Three key elements in the deformation and movement of these massive plates are plate rifting, plate subduction in the Earth's mantle, and horizontal transform faulting.
Despite numerous investigations at different fault zones such as the San Andreas Fault Zone in California (SAF), essential processes controlling such large transforms could not yet be fully understood. The Dead Sea Transform is, alongside the SAF, a key location for the study of transform faulting. The apparently simple structure of the DST system and the relatively low movement rate of approx. 0.5 cm per year distinguishes the DST significantly from the SAF which moves at a rate of 3.5 cm per year due to the complicated interaction of oceanic plate movement. Apart from the necessity for basic research, a further important reason for our study is that the investigation of historical earthquakes, paleoseismic studies and instrumental seismology over the past 100 years have shown that a number of destructive earthquakes have occurred along the DST. The DST, thus, represents a significant seismic risk for the inhabitants of Palestine, Israel and Jordan.
Due to the political situation in the Middle East in recent years, an investigation of the DST has not been possible up to now. The Dead Sea Rift Transect (DESERT) project is thus the first geoscientific project to cross the DST. This research group aims to study the structure and dynamics of the crust and outer Earth's mantle, the fine structures and dynamics of the DST, the thermal conditions and the geodynamic evolution.
Abstract: The Dead Sea Rift / Dead Sea Transform acts as a hinge between the Alpine-Himalayan-Mountain Belt, stretching East-West from the Mediterranean to Indonesia, and the largest active continental rift system, the Afro-Arabian Rift System, which runs South-North from East Africa to the Dead Sea. Except for a mild compressional deformation starting about 180 Ma ago, the Dead Sea region has remained a stable platform almost since its formation in the late Proterozoic. This tectonic stability was only recently (ca. 18 Ma ago) interrupted by the formation of a transform with a left-lateral motion of about 105 km as of today. The simplicity of this system, especially in the Arava Valley, the valley between the Dead Sea and the Red Sea, puts it in strong contrast to other large transform systems like the North Anatolian Fault System, which is in the middle of an orogenic belt, and the San Andreas Fault System, which suffered repeated accretional episodes and the interaction with a triple junction. The simplicity of the Dead Sea Rift / the nearly linear Dead Sea Transform provides a natural laboratory to study and understand transfrom faults, one of the key elements of plate tectonics together with subduction and rifting. Despite the central role of this world geological site, up to now no geophysical profile has crossed the Dead Sea Rift / the Dead Sea Transfrom (DST).
The DEad SEa Rift Transect (DESERT) is a multinational and interdisciplinary study of the Dead Sea Rift, and the main goal of the DESERT project is to help address a fundamental question of plate tectonics: How do shear zones work and what controls them, on different scales?
The project began with field work in February 2000, and first experiments were completed by the DESERT Team in May 2000. The seismic, seismological, and magnetotelluric experiments presented here, along with the future electromagnetic, gravity, magnetic, petrological, geothermal, geodynamic, and geological studies, will provide the geophysical and geodynamic frame for further geoscientific research.
Within the DESERT project, scientists from Germany, Israel, Jordan, and the Palestine Territories joined together for the first time to study the crust and upper mantle, the main shear zones, and the geodynamics of the DST. Over 30 scientists of the GFZ, the universities of Potsdam, Kiel, Köln, and Göttingen; the universities of Tel-Aviv and Jerusalem, the national Ministry of Infrastructure and the Geophysical Institute of Israel; the Natural Resources Authority, Jordan; and the An-Najah University in Nablus and the Palestine Water Authority, Palestine Territories; work together in this project. The 260 km long transect across the DST traverses Israel, Jordan, and the Palestine Territories.
First results of the geophysical experiments show – contrary to the expectations – basically no up-doming of the Moho under the rift, suggesting that the mantle has played a rather insignificant role in the extension process associated with the Dead Sea Rift. The role of an 8 km thick structure, imaged as a band of strong reflectors in the lower crust under the elevated rift shoulder in Jordan, the dynamics of the rift / transform requires additional geophysical studies planned for the next year. On the kilometer scale, a 3-D tomographic image of the fault regions shows significant changes in seismic P-velocities across the fault, and the analysis of fault guided waves indicates a low-velocity zone with a velocity reduction of 15 to 25% and a thickness of only 10 to 20 meters at the surface location of the Arava Fault. This fault is also the location where a roughly 1.5 km thick body with very good conductivity in 2 km depth stops abruptly, possibly indicating the contact of sediments with saline brines to the west of the fault with a high velocity body with bad conductivity to the east.
Abstract: At Merapi volcano, central Java, an active seismic experiment was carried out in the years 1997 and 1998 as a contribution to the German-Indonesian joint project MERAPI. Using deterministic wide-angle refraction and stochastic methods, the seismic structure was investigated along three profiles until approximately 300 m depth. The seismic source was a mudgun operated in three artificial water basins located on the flanks of the volcano at altitudes of about 1000 m. Shots were recorded along seismometer lines arranged radially with respect to the summit and consisting of up to 30 three-component seismometers each (100 m spacing).
The standard data processing included trace editing, stacking of up to 100 single shots, and in some cases the application of f-k filters to separate back and forth travelling waves. To analyse vector properties of the wave field, the covariance matrix of the three-component data is calculated and various measures of rectilinearity and particle motion are determined. These analyses lead to the design of a polarisation filter that enhanced the first break and later onsets with a mostly vertical particle motion in the otherwise structureless wave field behind the first break.
Arrival times of first and later onsets were modelled by two-dimensional ray tracing in a layered 1-D model with P-velocities of some hundred m/s near the surface and more than 3000 m/s at a maximum depth of about 300 m. Later onsets in the seismograms are interpreted as refracted waves reflected at steeply-dipping fracture zones. The most prominent fracture zones observed on the southern and western volcano flanks are located at the same distances (~2 km, ~3.5 km) to the summit of Mt. Merapi and may belong to a larger, approximately circular weakness zone.
Abstract: Within the MERAPI project, the seismological aspects of Merapi volcano are investigated by two complementary approaches, the observation of seismicity using a network of permanent recording stations, and seismic measurements with artificial sources. The active seismic experiment has the following objectives: - to determine the seismic velocity structure of the volcano in order to identify anomalies and to improve the localization of natural seismic sources, - to investigate properties affecting seismic signal propagation such as reflectivity, scattering, and absorption in order to support the analysis of the mechanisms of natural seismic sources, and - to provide a basis for active seismic repeat measurements to monitor temporal changes in the velocity structure of the volcano caused by magma or fluid movement or deformation. The concept of the measurements is to use a seismic source with well defined spectral characteristics and high repetition accuracy, and to sound the volcano in 3D along geophone lines arranged radially with respect to the summit. In summer 1997 some measurements serving as a pilot study were performed along lines BEB and TRO. They are discussed below.
Abstract: The Geysers is a vapor-dominated geothermal field located about 120 km north of San Francisco, California. The field is actively exploited since the 1960s, and it is now perhaps the most important and most productive geothermal field in the USA. The continuous injection of fluids and the stress perturbations of this area has resulted in induced seismicity which is clearly felt in the surrounding villages. Thus, based on these considerations, in the present work Ground Motion Prediction Equations (GMPEs) are derived, as they play key role in seismic hazard analysis control and for monitoring the effects of the seismicity rate levels. The GMPEs are derived through the mixed non-linear regression technique for both Peak Ground Velocity (PGV) and Peak Ground Acceleration (PGA). This technique includes both fixed effects and random effects and allows to account for both inter-event and intra-event dependencies in the data. In order to account for site/station effects, a two steps approach has been used. In the first step, regression analysis is performed without station corrections and thus providing a reference model. In the second step, based on the residual distribution at each station and the results of a Z-test, station correction coefficients are introduced to get final correct model. The data from earthquakes recorded at 29 stations for the period September 2007 through November 2010 have been used. The magnitude range is (1.0 < Mw < 3.5) while the hypocentral distances range between 0.5 km and 20 km. The final models are compared with standard models obtained using data collected in different tectonic environments and magnitude ranges to understand the compatibility of the model obtained from data collected in geothermal fields with respect to those obtained from natural seismic events. The residual analysis is performed at individual stations to check the reliability of the station corrections and for evaluating the fitting reliability of the retrieved model. The best model has been chosen on the basis of inter-event standard error and R-square test. After the introduction of the site/station correction factor, an improvement in the fit is observed, which resulted in total standard error reduction and increased R-square values.
Notes: Poster presentation in session SM5.2/ERE5.6 at the EGU General Assembly, Vienna, Austria, April 22-27, 2012
Abstract: The effect of induced seismicity of geothermal systems during stimulation and fluid circulation can cover a wide range of values from light and unfelt to severe and damaging. If the design of a modern geothermal system requires the largest efficiency to be obtained from the social point of view it is required that the system could be managed in order to reduce possible impact in advance. In this framework, automatic control of the seismic response of the stimulated reservoir is nowadays mandatory, particularly in proximity of densely populated areas. Recently, techniques have been proposed for this purpose mainly based on the concept of the traffic light. This system provides a tool to decide the level of stimulation rate based on the real-time analysis of the induced seismicity and the ongoing ground motion values. However, in some cases the induced effect can be delayed with respect to the time when the reservoir is stimulated. Thus, a controlling system technique able to estimate the ground motion levels for different time scales can help to better control the geothermal system. Here we present an adaptation of the classical probabilistic seismic hazard analysis to the case where the seismicity rate as well as the propagation medium properties are not constant with time. We use a non-homogeneous seismicity model for modeling purposes, in which the seismicity rate and b-value of the recurrence relationship change with time. Additionally, as a further controlling procedure, we propose a moving time window analysis of the recorded peak ground-motion values aimed at monitoring the changes in the propagation medium. In fact, for the same set of magnitude values recorded at the same stations, we expect that on average peak ground motion values attenuate in same way. As a consequence, the residual differences can be reasonably ascribed to changes in medium properties. These changes can be modeled and directly introduced in the hazard integral.
We applied the proposed technique to a training dataset of induced earthquakes recorded by Berkeley-Geysers network, which is installed in The Geysers geothermal area in Northern California. The reliability of the techniques is then tested by using a different dataset performing seismic hazard analysis in a time-evolving approach, which provides with ground-motion values having fixed probabilities of exceedence. Those values can be finally compared with the observations by using appropriate statistical tests.
Abstract: The increasing number of seismic networks with high density of stations offers an ever larger amount of three-component recordings of earthquakes in a wide range of magnitudes. The analysis of these data can provide detailed information on both the propagation medium and the seismic source. In particular, the S-wave velocity is a key parameter for the understanding of the compositional and physical state of the lithosphere. On the other hand this requires a tool for identifying the seismic phase.
The S-phase can be identified by a change in amplitude and frequency content of the signal with respect to the P-phase. The precise identification of S-phase is generally made difficult by the interference of P-coda waves, the arrival of converted phases generated beneath the recording site or the S-wave splitting. These factors can lead the operator to misidentify the phase or, very often, to abandon reading itself.
In this study, we propose a data processing technique aimed at univocally identifying the arrival-time of the S-phase by using three component recordings available at all stations of a seismic network. The proposed technique provides an additional support to the operators to be used for both the analysis of a single event or for the massive, quasi-automatic analysis of huge datasets.
The technique is based on the combination of a polarization detector mainly used in passive seismology and the move-out and stack analysis of trace gathers as for the velocity analysis in exploration seismics. The processing consists of four main steps. The first consists in P-phase picking and event location. The second step is the setting-up polarization detector: we rotate the three-component seismograms into the ray-coordinate system (L,Q,T), using theoretical backazimuths and incidence angles from P-phase polarizations. In the new system we calculate the directivity D, which is defined as the normalized angle between the P-phase polarization L and the actual polarization direction, the rectilinearity R and the ratio between transverse and total energy H. The product of the squares of the three filter operators yields the characteristic function (CF) for S-wave detection, with which we weight the transverse component traces. The third step deals with the seismic section analyses. Once the CF has been defined, the waveforms are displayed in common receiver gathers as a function of hypocentral distance. On each section we evaluate the lateral coherence of S-phase through a linear velocity analysis. The resulting S-wave velocity can be used to compute a reference pick for the S-phase at each station. In the fourth step an automatic picker is used around the reference value.
In the present study, we apply our S-wave detection-picking approach to a dataset of 5675 three component, ground velocity recordings of 626 local earthquakes with magnitude ML (0.1, 3.2), which occurred in southern Italy and were recorded by the Irpinia Seismic Network in the period December 2007 to March 2010. To assess the performance of the proposed methodology, we compare the residuals of the automatic and the theoretical arrivals with the residuals between the manual readings and the theoretical arrivals. The dispersion of the residual distributions obtained from the refined picking is consistent with the dispersion obtained from the manual picks, while the total number of available readings is increased.
Abstract: The Mw 9.0 megathrust earthquake off northeastern Honshu, Japan, in March 2011 had an unexpected size for a region which experienced only few events with magnitude larger than 8.0 in the past millennium. The event originated at crustal depths along a segment of the Pacific slab of the Japanese subduction zone. Large slip deficit and strong interplate coupling have been previously detected there by inland deformation measurements. The pattern of seismicity occurrence and the mechanical coupling between the different sectors of the Japan slab suggest that its morphology and segmentation may be strongly influenced by the presence of landward oceanic fracture zones. The aim of this study is to image the locations of strongly radiating sources and the rupture development during the faulting process. We used strong-motion records from the dense Japanese accelerometer arrays, integrated twice to obtain ground displacement, and filtered in different frequency bands between 0.04 Hz and 2.0 Hz. We applied a move-out and stacking technique to back-project the S-wave displacement amplitudes onto the subducting plate boundary, including the proper correction for geometrical spreading and source radiation pattern. Thus, the resulting images are consistently mapped into the slip distribution during the rupture development. Image resolution and sensitivity to processing parameters is assessed by synthetic tests. Our results show that the great Tohoku earthquake started as a smaller size rupture, slowly propagating upward along the slab segment and triggering the break of a larger size asperity at shallower depths near the trench. In that region also the largest slip has been observed in various studies. For a large amount of its duration, the rupture remained confined in a 100-150 km wide slab stripe, delimited by two Northwest-Southeast trending oceanic fractures. After about a minute, the rupture propagated at relatively high speed toward Southwest, parallel to the trench. The occurrence of large slip amplitudes at shallow depths likely favored the rupture to propagate across contiguous slab segments and contributed to build up a giant size earthquake. The lateral variations in the slab surface geometry may act as geometrical and/or mechanical barriers finally controlling the earthquake rupture nucleation, evolution and arrest.
Notes: Video: The entire session U42A <I>The Great March 11, 2011 Tohoku Earthquake IV</I> can be viewed at http://vimeo.com/33388479.
Abstract: The growing installation of industrial facilities for subsurface explorations worldwide, particularly in proximity of urbanized areas, requires refinements in understanding both the mechanisms for triggering the induced seismicity and their effects in terms of hazard. In fact, particularly in proximity of densely populated areas, induced low-to-moderate, high-frequency seismicity can be clearly felt by population and, in some cases, can produce damages to non-structural elements of buildings or even structural damages when rural buildings are involved. As a consequence, it is nowadays definitely important to be able to estimate time-dependent seismic hazard for providing a guide during the field operations and for monitoring their direct effects in the surrounding areas.
In the present work a time-dependent probabilistic seismic hazard analysis is presented. The technique is aimed at integrating the models of earthquakes occurrence which best correlate with field operations and ground-motion prediction techniques in a Bayesian framework to estimate in a time-evolving approach the probability of exceedance of selected ground motion parameters, that are of engineering interest.
Using data from different geothermal areas, e.g. The Geysers in Northern California, seismic hazard analysis is performed through: 1. Identification of the earthquake occurrence model which, on the basis of statistical tests, best correlates with the observed seismicity; 2. Time and space analysis of the recurrence relationship, mainly the b-value of the Gutenberg-Richter relationship; 3. Estimation of the maximum expected magnitude earthquake; 4. Selection of the best ground-motion parameters and predictive equation; 5. The selection of the most appropriate exposure times that cannot be classic ones for example 475 years.
Finally, the availability of high-quality catalogues covering long periods offers a unique opportunity for testing the proposed technique. In this respect, using different portions of the available catalogues a first setting of the technique is performed, and the prediction capability is evaluated through a statistical analysis.
Abstract: The Campi Flegrei located west of the city of Naples, is one of the most active calderas in the world. With several hundred thousand people living within its borders, this area is considered at high risk for eruptive scenarios. With the aim of investigating and reconstructing the volcanic structure of the Campi Flegrei caldera, the extensive marine investigation SERAPIS was carried out in the area in September 2001. The large amount of data provided by the experiment is still able to provide new insights into the velocity structure underneath the Campi Flegrei area. This study discusses a 3D velocity model with seismic interfaces for the Campi Flegrei caldera from the joint inversion of first and secondary seismic phases based on the SERAPIS dataset. The traveltimes used during the inversion are obtained by a refined manual picking of first arrivals and of main reflected/converted (PP and PS) phases. The dataset of first arrivals is the same as used during the previous tomographic studies of the area, but the travetimes have been manually re-picked. Concerning the reflected dataset, we performed a refined picking of the main PP and PS reflections on vertical and radial seismic sections composed with traces arranged in 3D Common Mid Point gathers. Three main reflection events have been inferred through graphic display of the seismic sections and analysis of the lateral coherency of reflection events by visual comparisons of different gathers along EW and NS profiles. Preliminary information on the morphology of reflectors has been obtained using the residual traveltimes between observed reflected picks and reflected traveltimes computed for 1D interfaces and using an average 1D velocity model with the aim of indicating large wavelength anomalies of the reflector depths. A final joint inversion has then been applied for the determination of the morphology of the reflector using CAT3D software which models both the velocity in the bulk and the shapes of the layers. Moreover, the joint use of first arrivals and reflected/converted data investigates the extension of a new 3D velocity model at greater depths as compared to tomographic models based on the inversion of only first arrival traveltimes.
Abstract: Campi Flegrei (Phlegraean Fields) is an active, resurgent volcanic caldera that is located a few kilometres west of the city of Naples, a densely populated urban settlement in southern Italy. To image the subsurface structure of the caldera, an extensive marine seismic survey was carried out in the area in 2001 (SERAPIS experiment). Previous results from this survey include smooth 3-D velocity models for the upper 4-5 km from first-arrival tomography, and a 1-D layered model from PP and PS reflection travel times and amplitudes. The layered model shows three dominant reflectors, interpreted as the base of marine unconsolidated sediments, the top of a gas-bearing rock formation around 3 km depth, and as the top of a magma layer with high melt fraction at about 7.5 km depth. Here we present results on the 3-D morphology of these reflectors, obtained by tomographic inversion of reflection travel times. The two shallowest reflectors are well-constrained by data and show maximum depth variations of 150 m and 300 m, respectively. The reflector around 3 km depth has a basin-like shape with a morphological high coinciding with the buried caldera rim, as previously imaged as a high-velocity anomaly. The deepest reflector is less well resolved, and it appears rather smooth with a small maximum in the western part of the submarine caldera boundary. Rock physics modelling helps to relate the seismic velocities to lithological properties and supports our previous interpretation of the main structural discontinuities. The dominant lithological units listed above essentially extend through the entire imaged volume beneath the caldera. We discuss lateral velocity variations both in terms of their significance based on model resolution, and regarding their implications on lithology. While petrological data support the presence of melt below 7.5 km, there are no indications for larger melt reservoirs at shallower depths within the resolved crustal volume. Kinematic seismic imaging leading to the currently available velocity models is not able to resolve structures less than 1 km in diameter, such as small magma patches, which have been suggested by petrologists.
Abstract: S5 project was aimed at supporting and integrating the ongoing research on three selected Italian test sites (the Alto Tiberina Fault (ATF), the Messina Strait and the Irpinia fault system) where advanced multi-parametric monitoring infrastructures are available. A new site (L’Aquila) has been added to the S5 project in September 2009 after the occurrence of the Mw 6.3 earthquake on April 6th 2009. The main objective of the project was to improve the understanding of earthquake generation processes in Italy and to characterize the earthquake source and medium properties in the sites by developing and applying innovative methodologies aimed at the massive processing, analysis and modelling of multi-parametric geophysical data available in real-time and off-line data banks. The work breakdown structure was organized according to scheme with four main Tasks each of them dedicated to the research activities developed in the selected active fault test sites: Alto-Tiberina fault, Irpinia fault, the Messina Strait and 2009 L’Aquila fault.
Abstract: This work concerns the high-resolution study of Irpinia region through the refined estimates of micro-earthquake source parameters in the magnitude range 1<M<3 and the determination of appropriate velocity models. Using accurate manual picks we first located about 900 events using a non-linear global approach and after we refined locations considering a double-difference technique. Moreover we relocated subsets of similar event using also high-precision differential arrivals from cross-correlation differential arrival times of P- and S-waves. Finally we propose an innovative technique to identify S-waves on seismograms since the complexity of the area produces phases that overlap the first S-wave arrival.
Abstract: The Mw 6.3 L'Aquila earthquake on April 6, 2009 has been recorded by the Irpinia Seismic Network (ISNet) about 250 km SE of the epicenter. The ISNet array has an aperture of about 80 km and consists of 25 stations, each equipped with a 3C CMG-5T accelerometer. Waveforms from 19 stations could be used to estimate rupture geometry, event magnitude, and moment tensor.
Standard array methods applied to low-pass filtered waveforms provided backazimuth and slowness of the incoming waves for subsequent event location. To image the rupture geometry we implemented a modified beamforming technique that back-projects the recorded direct P-wave amplitudes into the earthquake source region. A NW-SE striking rupture of 17 km length is imaged, propagating with an average velocity up to 3 km/s.
We inverted P- and S-wave displacement spectra to determine the main source parameters, assuming an omega-square spectral model and a constant-Q attenuation function. The estimated seismic moment (2.1 x 10^18 Nm) is consistent for P- and S-waves and yields a maximum Mw of 6.1. Relatively high P- and S-corner frequencies (0.6 and 0.4 Hz), due to directivity effects, lead to an underestimation of the rupture length and a high static stress drop value. However, an apparent stress of about 1.5 MPa is measured from the radiated seismic energy, which is less sensitive to directivity effects.
We determine the moment tensor solution for the earthquake by waveform modelling. One approach uses a point source approximation and a grid search over a set of trial source positions to identify the optimal centroid position, time, and moment tensor. In another approach the rupture is represented by a finite 1D source model, approximated by a summation over point sources along the fault strike. The focal mechanism and the linear seismic moment distribution along the strike are inverted simultaneously by a grid search combined with simulated annealing. We find a centroid depth of about 5 km and a prevalently normal fault plane solution with a dominant directivity toward SE.
Our studies demonstrate that the use of array techniques and a dense accelerometer network can provide quick and robust estimations of source parameters of moderate-size earthquakes located outside the network.
Abstract: The research activity of RU6 is targeted at the Irpinia test site, one of the most active seismic zones of the Southern Apennines. Large destructive earthquakes occurred both in historical and recent times in this region, which was struck in 1980 by the strongest event (Mw = 6.9) of the past century in the Southern Apennines. The main research facility of the area is the Irpinia Seismic Network (ISNet), managed by AMRA (Analysis and Monitoring of Environmental Risks). ISNet is a high-density, high-dynamics seismic and accelerometric network, built on innovative technological and methodological concepts, and focused on real-time data acquisition, processing and modeling. ISNet complements the INGV network in the region, making the Irpinia test site one of the highest instrumented seismic areas in Italy and an ideal site for experimenting new approaches for seismic monitoring and imaging of active fault systems.
In the framework of S5 project, the research carried by RU6 is based on the real-time and off-line analysis of noise and microearthquake data collected by the ISNet and INGV networks. In particular, we are actively developing and experimenting new methodologies in the following research fields:
(a) Seismic noise analysis and green functions: use of the random wavefield to retrieve images of the sub-soil through cross-correlation and stacking of continuous recorded signal; study of the ambient noise level and the seismic detection threshold for ISNet.
(b) Refined estimates of micro-earthquake source parameters: retrieval of high resolution images of the fault system through the accurate determination of location, size and fault mechanisms in the magnitude range 1<M<3; study of the scaling relationships for seismic moment, radiated energy, corner-frequency, stress drop and source radius.
(c) Reflection seismology applied to earthquake data: use of reflection tomography to infer depths and geometries of subsurface reflectors, and to constrain the velocity structure below the seismogenic zone.
The central access point for the real-time and off-line analysis results is the ISNet bulletin (http://lxserver.ov.ingv.it/cgi-bin/isnet-events/isnet.cgi) where full waveforms and source parameters for automatically detected and manually revised events are available.
Abstract: The Mw 6.3 Central Italy (L'Aquila) earthquake on April 6, 2009 has been recorded by the Irpinia Seismic Network (ISNet) about 250 km SE of the epicenter. The ISNet array has an aperture of about 80 km and consists of 25 stations, each equipped with a 3C CMG-5T accelerometer. Waveforms from 19 stations could be used to estimate rupture geometry, event magnitude, and moment tensor.
The recorded, low-pass filtered waveforms (f<2 Hz) are very coherent and permit the application of array methods to measure backazimuth and slowness of the incoming waves for subsequent event location. To image the rupture geometry we implemented a modified beamforming and stacking technique that back-projects the recorded direct P-wave amplitudes into the earthquake source region using travel times from a 1D velocity model. Amplitudes were measured in a 0.5 s moving time window until the arrival of secondary phases. A NW-SE striking rupture of 17 km length is imaged, propagating with an average velocity up to 3 km/s.
The P- and S-wave displacement spectra recorded by the 3C accelerometric and some broad-band sensors have been inverted to determine the main source parameters. We assumed an omega-square spectral model and a constant-Q attenuation function, with a parameter t* (T/Q) directly estimated from the high-frequency spectral decay. The estimated seismic moment (2.1 x 10^18 Nm) is consistent for P- and S-waves within a factor 2, providing a maximum Mw 6.1. This value is also consistent with the local magnitude inferred from Wood-Anderson synthetic records. Relatively high P- and S-corner frequencies (0.6 and 0.4 Hz), due to directivity effects, are the cause for an underestimation of the rupture length and an extremely high static stress drop value. On the other hand, an apparent stress of about 1.5 Mpa is measured from the radiated seismic energy, which is less sensitive to source directivity effects.
We determine the moment tensor solution for the earthquake by modeling the strong-motion waveforms using two different approaches. The first one uses a point source approximation and a grid search over a set of trial source positions and time shifts to identify the optimal centroid position, time, and moment tensor. In the second method the rupture is represented by a finite 1D source model. Source finiteness is approximated by a summation over point sources aligned along the fault strike. The focal mechanism and the linear seismic moment distribution along the strike are inverted simultaneously by an optimized grid search combined with a simulated annealing algorithm. Moreover the technique provides some insight on the modality of the rupture. We find a centroid depth of about 5 km and a prevalently normal fault plane solution with a dominant directivity effect toward SE.
Our studies demonstrate that the use of array techniques and a dense accelerometer network can provide quick and robust estimations of source parameters of moderate-size earthquakes located outside the network.
Abstract: Campi Flegrei (Phlegraean Fields) is an active, resurgent volcanic caldera that is located a few kilometres west of the city of Naples, a densely populated urban settlement in southern Italy. Identifying, locating at depth and better defining the geometry of the magma feeding system of the caldera is highly relevant for assessing and monitoring its volcanic hazard. Based on a high resolution seismic reflection dataset, we investigated the deep structure of the volcano. Here we show that seismic wave amplitude variations with distance from the source provide clear evidence for large amplitude reflections from the top of an extended, supercritical fluid-bearing rock formation at about 3000 m depth and of an about 7500 m deep low-velocity layer, which is associated with a mid-crust, partial melting zone beneath the caldera (around 70 percent of molten rock). These new data suggest that a large magmatic sill is present well within the basement formations, which is possibly linked to the surface through a system of deep fractures bordering the caldera. The lateral extensions and similar depths of melt layers observed beneath Campi Flegrei and beneath the nearby Vesuvius volcano support the hypothesis of a single continuous magma reservoir feeding both of these volcanoes.
Abstract: DESERT and DESIRE, two multi-national, interdisciplinary research efforts by teams from Germany, Israel, Jordan and Palestine focused on the Dead Sea Transform (DST) and the Dead Sea Basin (DSB), respectively. The DST has accommodated left-lateral transform motion of 105 km between the African and Arabian plates since early Miocene (ca. 20 Ma), creating during this process also the prime example of a pull-apart basin, the DSB.
Within DESERT the DST segment between the Dead Sea and the Red Sea called Arava/Araba Fault (AF) was studied with the following results. On plate tectonic scale the AF is a narrow, sub-vertical zone cutting through crust and lithosphere to more than 50 km depth, while the Moho depth increases smoothly from 26 km to 39 km from W to E under the DST. Several faults exist in the upper crust in a ca. 40 km wide zone around the AF, but none has kilometer-size zones of decreased seismic velocities/zones of high electrical conductivities typical for damage zones. Across the sub-vertical AF abrupt changes in lithology can be identified to a depth of 4 kilometers. The AF also acts as a barrier to fluids. The AF is the main active fault of the DST system but it has only accommodated a limited part (up to 60 km) of the overall 105 km of sinistral plate motion. Now inactive fault strands in the vicinity of the present day AF took up lateral motion until about 5 Ma ago, when the main, active fault trace shifted ca. 1 km westward to its present position. In the top few hundred meters of the AF a locally transpressional regime occurs in a 100 to 300 m wide zone of deformed and displaced material, bordered by sub-parallel faults forming positive flower structures. The damage zones of the individual faults are only 5 to 20 m wide. This narrow width is significantly smaller than at other major strike-slip faults of similar magnitude. Most of these findings are corroborated by thermo-mechanical modeling that show shear deformation in the lithosphere under the DST/AF first localizes in a 20 to 40 km wide zone with a mechanically weak decoupling zone extending sub-vertically through the entire lithosphere. As time progressed upper crustal deformation became quickly focused in a few faults.
Within DESIRE the DSB, the largest basin along the DST, is studied using again a multi-disciplinary and multi-scale approach. Some of the open questions presently being addressed in the DESIRE project which started in 2006, are: (1) What is the fault pattern of the DSB at depth? and (2) What is the deep structure of the DSB and the depth and configuration of the major crustal interfaces, e.g. the Moho beneath the basin? We will also report results of DESIRE, with an emphasis on the findings from geophysical studies and modeling.
Abstract: Elastic parameters derived from seismic reflection data provide information on the lithological contrast at an interface and support the geological interpretation. Standard amplitude variation with offset methods analyse reflection coefficients and thus require the knowledge of source, receiver, and propagation effects. These effects are mostly reduced when looking at amplitude ratios of PP reflected and PS converted phases from the same interface. We present a technique to model and invert PS-to-PP amplitude ratios to estimate arbitrary strong elastic contrasts at a given interface in a layered medium. Our method is first applied to synthetic data to assess its possibilities and limitations. Second, we analyse amplitude ratios from real data acquired in the Campi Flegrei caldera (Phlegrean Fields, southern Italy), and we show first results for main discontinuities identified in this volcanic region.
Abstract: A new 3D model for the seismic velocity and density structure of the crust in the greater Barents Sea region has been developed by a team consisting of NORSAR, the University of Oslo and the U.S. Geological Survey. The crustal model is based on a large amount of 1D seismic velocity profiles that were compiled. Deep-seismic reflection data were used to define the geometry of sedimentary basins and, where possible, estimate crustal thickness. The 3D velocity model is based on an equidistant grid with a node spacing of 50 km, and for each node point two sedimentary and three crystalline layers have been defined in terms of their depths and physical properties. A complementary upper mantle model has also been developed in cooperation with the University of Colorado. This model is based on a large dataset of new surface-wave observations from more than 150 local and regional events with travel paths through the greater Barents Sea region. This has resulted in group-velocity observations that were inverted for 2D group-velocity maps, and for a new 3D S- and P-wave upper mantle model with a nominal resolution of 1 by 1 degree. In the final step of model construction, we combined the new crustal model with the new surface-wave inversion mantle structure sampled laterally at 50 km, resulting in a complete 3D velocity model for the crust and upper mantle. The new model reveals strong horizontal irregularities and inhomogeneities and can now be used, among other interesting scientific purposes, as a basis for estimation of more precise seismic travel times and thereby also for improved locations of seismic events in this part of the European Arctic.
Abstract: A new 3D geophysical model for the greater Barents Sea Region was developed by the University of Oslo, NORSAR and the U.S. Geological Survey. A considerable amount of continuous seismic velocity transects were compiled and deep-seismic multi-channel data in the SW was further used for density modeling and subsequent density-to-velocity conversion. The final velocity model consists of five crustal layers. The model compilation introduced in this study is based on geological provinces of different sizes, such as sedimentary basins, structural highs or volcanic provinces with individual sedimentary, tectonic and magmatic histories. Firstly, the layer velocities and thicknesses were laterally inter- and extrapolated. Within the provinces, linear relationships between the thickness of the sedimentary rocks and the thickness of the remaining crystalline crust are observed. We used therefore the additionally compiled (area-wide) depth-to-basement data to adjust the crystalline crustal thickness according to the sedimentary thickness where no data constraints are given. After the compilation, the P-wave velocity model was converted into a density model, in order to test the structure against independent gravity data and to model densities for each layer. Our model is completed by an upper mantle surface wave inversion model. The resultant model was subsequently verified in terms of traveltimes. In order to do so, we compiled a list of well-located reference events and arrival times not used to construct the model.Within the uncertainty limits, modeled traveltimes fit the available observations and qualitatively validate the model. An outstanding improvement compared to other (older) geophysical models is the high resolution of 50 km throughout the target region. Velocity transects through the 3D model and a new depth-to-Moho map exemplifies geological features of the European Arctic and the formerly unknown diversity of the crustal structure in the greater Barents Sea region.
Abstract: Existing global and regional tomographic models have limited resolution in the European Arctic due to the small number of seismic stations, relatively low regional seismicity, and poor knowledge of the crustal structure. During the last decades, new seismic stations were permanently ortemporarily installed in and around this region. To improve the surface wave data set in the region, we extensively searched for broadband data from stations in the area from the beginning of the 1970s until 2005 and were able to retrieve surface wave observations from the data archives at NORSAR, the Universities of Bergen and Helsinki, the Kola Science Center in Apatity, the Geological Survey of Denmark and Greenland, and from IRIS and GEOFON. Recently, a new crustal model of the Barents Sea and surrounding areas had been derived in a joint Norwegian-American project. This detailed information on crustal thickness and sedimentary basins helps to constrain the tomographic inversion of the upper mantle velocity structure based on surface wave data. Rayleigh and Love wave group velocity measurements from 10 to 150 sec period obtained on the newly retrieved data were combined with existing data provided by the University of Colorado. Using these data, we constructed a new 3D shear velocity model of crust and upper mantle beneath the European Arctic, which provides higher resolution and accuracy than previous models. The new model reveals substantial variations in shear velocities in the upper mantle across the region. Of particular note are clari ed images of the mantle expression of the continent-ocean transition in the Norwegian Sea and a deep high wave speed lithospheric root beneath the Eastern Barents Sea. Currently, the data set is being extended southwards to cover mainland Scandinavia and leading to a 3D model of the crust and upper mantle for Norway and adjacent areas.
Abstract: In this three-year project we have been addressing the problem of energy partitioning at distances ranging from very local to regional for various kinds of seismic sources. On the local and regional scale (20-220 km) we have targeted events from the region offshore of Western Norway, where we have both natural earthquake activity and frequent underwater explosions carried out by the Norwegian Navy. On the small scale we have focused on analysis of observations from an in-mine network of 16-18 sensors in the Pyhäsalmi mine in central Finland. This analysis has been supplemented with 3-D finite difference wave propagation simulations in a realistic mine model to investigate the physical mechanisms that partition seismic energy in the near source region in and around the underground mine.
The results from modeling and analysis of local and regional data show that mean S/P amplitude ratios for explosions and natural events differ at individual stations and are in general higher for natural events and frequency bands above 3 Hz. However, the distributions of S/P ratios for explosions and natural events overlap in all analyzed frequency bands. Thus, for individual events in our study area, S/P amplitude ratios can only assist the discrimination between an explosion and a natural event. This observation is supported by synthetic seismograms calculated for simple 1-D models, which demonstrate that explosions generate shear-wave energy if they are fired close to an interface with a strong material contrast (as is the case for most explosions), e.g., free surface or the ocean bottom. The larger difference in S/P ratios between earthquakes and explosions for higher frequencies can be explained by the fact that at low frequencies (larger wavelengths), discontinuities and structural heterogeneities in the explosion source region are stronger generators of converted S energy. The S* phase, for example, is most efficiently generated whenever an explosion source is located close to (within one wavelength) a strong discontinuity.
High-frequency (50-400 Hz) S/P ratios for mine blasts (explosions) and rockbursts recorded at the Pyhäsalmi in-mine network do not show any significant dependency on the distance to the events, which ranges between 40 and 400 m. The Pyhäsalmi explosions have generally lower S/P ratios than the rockbursts for all frequencies, but the difference is far too small to be significant for classification purposes. S/P ratios for explosions and rockbursts located in the same small area of the mine show results very similar to those for the full data set. This indicates that the observed differences in S/P ratios between explosions and rockbursts are due to differences in the source characteristics, and not to propagation effects along paths in the mine.
Three-dimensional finite-difference simulations were used to model seismic events within the Pyhäsalmi mine. In particular, a January 26, 2003, rockburst was modeled at frequencies of 50 Hz (4 m grid) and 100 Hz (2 m grid). We were able to match the characteristics of the observed data at 50 Hz particularly well, and the characteristics of the 100 Hz data reasonably well. The simulations showed that significant shear-energy can be produced due to the geologic and structural heterogeneities within the mine. In fact, mode-converted shear-energy generated from mine heterogeneity can dominate the compressional energy from an explosive source. A strong correlation is observed between the distance of a source from a mine heterogeneity and the magnitude of generated shear-energy. The ratio of shear to compressional energy is about a factor of two larger when the source is located within one wavelength from a mine heterogeneity. The simulations also suggest that excavated mine volumes are significantly stronger contributors to shear-energy generation than are geologic heterogeneities.
Abstract: Existing global and regional tomographic models have limited resolution in the European Arctic due to the small number of seismic stations, relatively low regional seismicity, and poor knowledge of the crustal structure. During the last decades, new seismic stations were permanently or temporarily installed in and around this region. However, many of the data from these stations are not easily accessible via the international data centers but only by direct request to the different data operators. Recently, a new crustal model of the Barents Sea and surrounding areas had been derived in a joint project between the University of Oslo, NORSAR and the USGS (Bungum et al., 2005). This model with its detailed information on crustal thickness and sedimentary basins in the area helps to constrain the tomographic inversion of the upper mantle velocity structure based on surface wave data. To improve the surface wave data set in the region, we extensively searched for broadband data from stations in the area from the beginning of the 1970s until 2005 and were able to retrieve surface wave observations from the data archives at NORSAR, University of Bergen, University of Helsinki, the Kola Science Center in Apatity, and the Geological Service of Denmark in addition to data from the data centers of IRIS and GEOFON. Rayleigh and Love wave group velocity measurements from 10 sec to 150 sec period obtained on these seismograms were combined with the existing data set provided by the University of Colorado (e.g. Levshin et al., 2001). Using these data, we constructed a new 3-D shear velocity model of the crust and upper mantle beneath the European Arctic which provides higher resolution and accuracy than previous models.
The new model reveals substantial variations in shear wave speeds in the upper mantle across the region. Of particular note are clarified images of the mantle expression of the continental-ocean transition in the Norwegian Sea and a deep high wave speed lithospheric root beneath Novaya Zemlya.
Abstract: We here present BARENTS50, a new 3D geophysical model of the crust in the Barents Sea region. The target region of interest comprises northern Norway and Finland, parts of the Kola Peninsula and the East European lowlands. Novaya Zemlya, the Kara Sea and Franz-Josef Land terminate the region to the east, while the Norwegian-Greenland Sea marks the western boundary. In total, 680 one-dimensional seismic velocity profiles were compiled, mostly by sampling 2D seismic velocity transects, from seismic refraction profiles, every 25 km. Seismic reflection data in the western Barents Sea were further used for density modeling and subsequent density-to-velocity conversion. Velocities from these profiles were binned into two sedimentary and three crystalline crustal layers. The first step of the compilation comprised the layer-wise interpolation of the velocities and thicknesses. Within the different geological provinces of the study region, linear relationships between the thickness of the sedimentary rocks and the thickness of the remaining crystalline crust are observed. We therefore used the separately compiled (area-wide) sediment thickness data to adjust the crystalline crustal thickness according to the sedimentary thickness where no constraints from 1D velocity profiles existed. The BARENTS50 model is based on an equidistant hexagonal grid with a node spacing of 50 km. The P-wave velocity model was used for gravity modeling in order to obtain 3D density structure in the study region. A better fit to the observed gravity was achieved using a grid search algorithm which focused on the density contrast of the sediment-basement interface. The high resolution of 50 km is an improvement compared to older geophysical models.
Notes: LA-UR-06-5471. BARENTS50 is available at http://www.norsar.no/c-90-Barents-Sea-3D-Model.aspx
Abstract: A new 3D geophysical model for the greater Barents Sea Region was developed by the University of Oslo, NORSAR and the U.S. Geological Survey. A considerable amount of continuous seismic velocity transects were compiled and deep-seismic multi-channel data in the SW was further used for density modeling and subsequent density-to-velocity conversion. The final velocity model consists of five crustal layers. The model compilation introduced in this study is based on geological provinces of different sizes, such as sedimentary basins, structural highs or volcanic provinces with individual sedimentary, tectonic and magmatic histories. Firstly, the layer velocities and thicknesses were laterally inter- and extrapolated. Within the provinces, linear relationships between the thickness of the sedimentary rocks and the thickness of the remaining crystalline crust are observed. We used therefore the additionally compiled (area-wide) depth-to-basement data to adjust the crystalline crustal thickness according to the sedimentary thickness where no data constraints are given. After the compilation, the P-wave velocity model was converted into a density model, in order to test the structure against independent gravity data and to model densities for each layer. Our model is completed by an upper mantle surface wave inversion model. The resultant model was subsequently verified in terms of traveltimes. In order to do so, we compiled a list of well-located reference events and arrival times not used to construct the model. Within the uncertainty limits, modeled traveltimes fit the available observations and qualitatively validate the model. An outstanding improvement compared to other (older) geophysical models is the high resolution of 50 km throughout the target region. Velocity transects through the 3D model and a new depth-to-Moho map exemplifies geological features of the European Arctic and the formerly unknown diversity of the crustal structure in the greater Barents Sea region.
Abstract: Existing global and regional tomographic models have limited resolution in the European Arctic due to the small number of seismic stations, relatively low regional seismicity, and poor knowledge of the crustal structure. During the last decades, new seismic stations were permanently or temporarily installed in and around this region. However, many of the data from these stations are not easily accessible via the international data centers but only by direct request to various data operators.
Recently, a new crustal model of the Barents Sea and surrounding areas had been derived in a joint project between the University of Oslo, NORSAR and the USGS (Bungum et al., 2005). This model with its detailed information on crustal thickness and sedimentary basins in the area helps to constrain the tomographic inversion of the upper mantle velocity structure based on surface wave data.
To improve the surface wave data set in the region, we extensively searched for broadband data from stations in the area from the beginning of the 1970s until 2005 and were able to retrieve surface wave observations from the data archives at NORSAR, University of Bergen, University of Helsinki, the Kola Science Center in Apatity, and the Geological Service of Denmark in addition to data from the data centers of IRIS and GEOFON.
Rayleigh and Love wave group velocity measurements from 10 sec to 150 sec period obtained on these seismograms were combined with the existing data set provided by the University of Colorado (e.g. Levshin et al., 2001). Using these data, we constructed a new 3-D shear velocity model of the crust and upper mantle beneath the European Arctic which provides higher resolution and accuracy than previous models.
The new model reveals substantial variations in shear wave speeds in the upper mantle across the region. Of particular note are clarified images of the mantle expression of the continent-ocean transition in the Norwegian Sea and a deep high wave speed lithospheric root beneath the Eastern Barents Sea.
Currently, the data set is being extended southwards to cover mainland Scandinavia and leading to a 3-D model of the crust and upper mantle for Norway and adjacent areas.
Abstract: Existing global and regional tomographic models have limited resolution in the European Arctic due to the small number of seismic stations, relatively low regional seismicity, and poor knowledge of the crustal structure. During the last decades, new seismic stations were permanently or temporarily installed in and around this region. However, many of the data from these stations are not easily accessible via the international data centers but only by direct request to the different data operators.
Recently, a new crustal model of the Barents Sea and surrounding areas had been derived in a joint project of the University of Oslo, NORSAR and the USGS (Bungum et al., 2005). This model with its detailed information on crustal thickness and sedimentary basins in the area perfectly constrains the tomographic inversion of the upper mantle velocity structure based on surface wave data.
In this project, we have extensively searched for broadband data from stations in the area from the beginning of the 1970s until 2005 and were able to retrieve surface wave observations from the data archives at NORSAR, University of Bergen, University of Helsinki, the Kola Science Center in Apatity, and the Geological Service of Denmark in addition to data from the data centers IRIS and GEOFON. The group velocity measurements of Rayleigh and Love waves in the range of period 10-150 s from these seismograms were combined with the existing data set provided by the Center for Imaging the Earth's Interior at the University of Colorado (Levshin et al., 2001). From this updated data set we have constructed a new crustal and upper mantle model of the European Arctic of higher resolution and accuracy than existing models.
Abstract: We constructed a 3D seismic P velocity model for the extended Barents Sea region including Svalbard, Novaya Zemlya, the Kara Sea, and the Kola-Karelia regions. The crustal model is based on a large number of existing 1D and 2D velocity profiles, constrained by geological observations, and the nominal resolution is 50 km. Each grid node is filled with a five-layer crustal model (plus water and ice), and the continuous upper mantle velocity structure is taken from published regional models. Seismic S velocities and the density structure shall be included in the near future. The final model also aims to improve seismic monitoring and verification in this region, including improved event locations and event size estimation, and a better understanding of regional seismic wave phases.
Validation of the velocity model includes forward modeling of observed travel times and relocation of seismic events. For this purpose we compiled a set of reference events with known or well-located epicenters, referred to as Ground Truth (GT) events. The GT events comprise quarry blasts and announced chemical explosions located mainly in Scandinavia and the Kola Peninsula, nuclear explosions in northwestern Russia and on Novaya Zemlya, and natural earthquakes. With these events we obtain good Pn and Sn ray coverage in the main target region. Phase arrival times of multiple events at some sites provides estimates on timing errors at some stations.
Here we present the model in terms of regional contour maps and seismic velocity transects. North-south trending transects in the western Barents Sea show Moho depths between 10 and 45 km, with average values around 35 km. Thicknesses of sedimentary layers vary considerably and reach locally more than 10 km. Strongest variations are observed in the west between northern Norway and Svalbard, whereas transects further east exhibit a more simple, layered crustal structure. As the crustal velocity structure in the source or receiver region strongly influences phase arrival times, the detailed 3D model is therefore also important for the monitoring of events at regional distances, where corresponding ray paths run mainly in the upper mantle. Furthermore, there are evidences for significant velocity variations in the mantle of our study region.
Abstract: We have addressed the question of how seismic energy is partitioned between P and S waves at regional distances between 20 and 220 km. The database for this study consists of recordings at the stations of the National Norwegian Seismic Network (NNSN) of events from the region offshore Western Norway. For this region we have both natural earthquake activity as well as frequent occurrence of underwater explosions carried out by the Norwegian Navy. The data were provided by the University of Bergen.
For a set of 49 presumed earthquakes and 24 presumed underwater explosions we measured the S/P ratios in different frequency bands at seven stations of the NNSN (KMY, BLS, ODD, ASK, SUE, and FOO). For both event populations we investigated the effects of epicentral distance and frequency band, as well as individual station effects.
A result from this study is that the mean S/P amplitude ratios for explosions and natural events differ at individual stations and are in general higher for natural events and frequency bands above 3 Hz. However, the distributions of S/P ratios for explosions and natural events overlap in all analyzed frequency bands. Thus, for individual events in our study area, S/P amplitude ratios can assist the discrimination between an explosion or a natural event, but other measures such as spectral analysis should be included in the interpretation.
Abstract: During the last decades various efforts have been made to unveil the lithospheric structure of the Scandinavian mountain chain. However, a careful review shows that a need exists for new, detailed seismic and seismological investigations of the deep crust and upper mantle.
The Norwegian mainland and offshore areas are covered by high-quality potential field data sets. The offshore continental margin of Norway is in addition well investigated by scientific and commercial seismic studies. However, the onshore-offshore correlation of lithospheric structures is not well known. Onshore-offshore studies are often well constrained in the offshore part by a combination of potential field data and seismics, while the investigation of the onshore part bases mainly on the ambiguous potential field data, due to the sparse distribution and resolution of the available deep seismic lines. (e.g. Olesen et al., 2003).
Using existing deep seismic lines Kinck et al. (1993) compiled a map of the crust-mantle boundary below Fennoscandia. Since then a variety of passive and active experiments have been carried out in the central Baltic Shield, which increased the quality of the available lithosphere information (e.g. Kozlovskaya et al., 2004). However, in Norway few significant improvements to Kinck et al. (1993) have been achieved. A recent study of crustal structure of Norway using teleseismic results (Ottemöller and Midzi, 2003) shows for most stations differences in the depth estimate of the crust-mantle boundary compared to the compilation by Kinck et al. (1993). However, the general shape of the crust-mantle boundary of Kinck et al. (1993) was confirmed.
The analysis of gravity data to provide an estimate on the crust-mantle boundary in Fennoscandia is very limited. Comparison between maps of the seismic Moho and a simple isostatic Moho show nearly no correlation. Density models based on seismic data in most cases do not provide sufficient constraints to allow secure interpretations of lithospheric structures. This is especially true in areas where the old seismic experiments had not the quality to resolve crustal structures.
Even where the geometry of the deep lithosphere is known, considerable uncertainties exist which justify a new, detailed study of the lithosphere structure in Norway. New, cooperative efforts should be made to accomplish an integrated study composed of both passive and active seismic experiments. Such a study is necessary to close the gap between the well-known offshore areas and the central Fennoscandian shield and provide the critical "missing link" between the onshore and offshore domains.
Abstract: We concluded comprehensive ground truth collection at the Khibiny, Olenegorsk, Kovdor and Zapolyarnyi mines, and have basic information on 2052 explosions. In the past two years we used this ground truth information to extract waveform data from the ARCES array and a number of regional stations (KEV, LVZ, APA) as well as from six stations that we deployed along two lines stretching between the Khibiny Massif mines and the region around the ARCES array. We calculated P/S ratios using the ARCES array data for many of these events comprising several source types (compact underground explosions, underground ripple-fired explosions, surface ripple-fired explosions). We found that the P/S ratios of small compact underground explosions in mines of the Khibiny Massif are systematically lower than the P/S ratios of large ripple-fired surface explosions. We had anticipated that smaller underground shots would appear more like single well-coupled explosions, thus having higher P/S ratios than large ripple-fired explosions. A possible explanation for this phenomenon is that the compact underground explosions in these mines are designed to fracture and drop a large quantity of ore from the ceiling of a horizontal shaft. The potential energy released by the falling ore may express as shear wave energy, which may be considerably greater than the (P wave) energy released directly by the explosive.
We concluded the deployment of the six stations along the Khibiny-ARCES lines this past summer; this year we are examining the data from these stations to see how P/S ratios vary with range from the source. We expect to have an update on the P/S ratio analysis contrasting different source types for this years SRR meeting, with the addition of an analysis of range dependence using data from the temporary stations.
The portable stations were redeployed in the fall of 2004 to the Kiruna and Malmberget underground mines in northern Sweden. The stations deployed in Malmberget also record events from the surface mining operations at the Aitik mine, located some 15 km from Malmberget mine. The data from these stations will allow comparisons of seismic waveforms resulting from different types of shooting practices at different locations within the mines. These stations will provide ground truth on a large number of explosions at these mines allowing future analyses of the dependence of discriminants on source type, possibly assessing the portability of results obtained with the Khibiny explosion observations.
Abstract: We address the problem of energy partitioning at distances ranging from very local to regional for various kinds of seismic sources, and are now in the last year of this three-year effort. On the small scale we have focused on analysis of observations from an in-mine network of 16-18 sensors in the Pyhäsalmi mine in central Finland. This analysis has been supplemented with 3D finite difference wave propagation simulations to investigate the physical mechanisms that partition seismic energy in the near source region in and around the underground mine. On the local and regional scale (20-220 km) we have targeted events from the region offshore Western Norway where we have both natural earthquake activity as well as frequent occurrence of underwater explosions carried out by the Norwegian Navy.
Since the previous reporting of this project at the 2004 Seismic Research Review (Bungum et al., 2004), we have extended the finite difference simulations in the 3D geological model of the Pyhäsalmi mine. This model, which encompasses a geologic volume 500 meters in each direction, includes 3-D representations of the ore bodies, excavated regions, tunnels, and voids. The model is discretized on both 2 and 4 meter grids making it possible to simulate seismic energy up to 100-200 Hz. We perform a variety of sensitivity tests to determine the mechanisms that produce shear energy in an underground mine environment. For example, we conduct a suite of 15000 (2-D) explosive source simulations to quantify the influence of source location on the amplitude of generated shear energy. We find that shear energy generation is particularly prevalent when the source is located near a geologic or structural boundary of the mine. In fact, most of the shear energy appears to be generated within 10-20 meters from the source (at frequencies of 50 Hz). Examination of waveforms reveals that both geologic heterogeneity and the structural influences of the mine are contributors to the near-source generation of shear energy. There is some suggestion that the geologic inhomogeneity is significant early in the wavetrain, whereas the mine structure is likely to produce scatter and be more significant later in the waveforms. As a validation measure, the synthetic waveforms are compared with observed data from single and multi-component instruments located in the mine. The simulated data match the amplitude and character of the observed waveforms particularly well, especially at frequencies at and below 50 Hz. This suggests that we can reliably infer energy partitioning phenomena based on these simulations.
A database of underwater explosions and earthquakes from the region offshore Western Norway, recorded at seven selected stations of the National Norwegian Seismic Network (NNSN), were analyzed for differences in the S/P amplitude ratios. In order to separate the path and source effects for the two event populations, we have investigated the station, distance and frequency dependencies of the recorded data in detail. The results indicate that the mean S/P amplitude ratios for both underwater explosions and natural events vary from station to station but are, in general, higher for natural events. For frequencies above 3 Hz, the difference in S/P ratios between explosions and natural events is higher than for lower frequencies. However, the distributions of S/P ratios for explosions and natural events overlap in all analyzed frequency bands. Thus, for individual events in our study area, S/P amplitude ratios can assist the discrimination between an explosion or a natural event, but other measures such as spectral analysis should be included in the interpretation.
Abstract: We have compiled a 3D seismic velocity model for the crust and upper mantle in the greater Barents Sea region including northern Scandinavia, Svalbard, Novaya Zemlya, the Kara Sea, and the Kola-Karelia regions. While the general motivation for developing this model is basic geophysical research, a more specific goal is to create a model for research on the identification and location of small seismic events in the study region, and for operational use in locating and characterizing seismic events in the study region.
The observational basis for the velocity model are previous, crustal-scale 2D seismic reflection and refraction profiles, and passive seismological recordings, supplemented by potential field data to provide additional constraints on the crustal structure. The model is defined at grid tiles spaced every 50 km, and each tile is represented by up to two sedimentary and three crystalline crustal layers (plus water and ice). For crustal regions not constrained by primary velocity data, we developed an interpolation scheme based on several defined geological provinces that are characterized by individual tectono-sedimentary histories. The interpolation utilizes the observed strong correlation between sediment and crystalline crustal thickness within continental provinces. For comparison, an alternative interpolation approach applies a continuous curvature gridding algorithm within each of the provinces.
To provide a complete lithospheric model, we complemented the crustal model with an upper mantle velocity model based on surface wave inversion, thereby covering depths essential for Pn and Sn travel time modeling. As an extension to the previously existing data set, we recently retrieved a large amount of surface wave data recorded or excited in the European Arctic during the last three decades. The merged surface wave data set will enable us to refine the upper mantle velocity structure in the study region significantly. Preliminary group velocity maps for Rayleigh and Love waves reflect large-scale geological structures and demonstrate lateral velocity variations in the mantle.
Validation of our velocity model includes travel time modeling and relocation of seismic events. For this purpose we compiled a set of Ground Truth (GT) events comprising chemical and nuclear explosions, and natural earthquakes. Phase arrival times of multiple events at some sites provide timing error estimates at some stations. With the GT events we obtain a rather good Pn and Sn ray coverage in the main target region. Besides the comparison of observed and modeled travel times along selected transects, we have computed source-specific station corrections (SSSCs) from our 3D model.
The crustal velocity models are also evaluated by comparison of predicted gravity fields with the observed free-air gravity. To model the gravity field, we used standard velocity-density relationships for crustal rock types and the density structure of the upper mantle from previous studies. The inferred gravity fields both reflect and exaggerate the basic geological features. Accomplishments so far have been concerned with implementation of a forward modeling procedure and software development needed to support the complex 3D model structure. The forward modelling is done in order to reduce the misfit between observed and modelled gravity and finally to supplement our crustal velocity model with a density distribution.
Abstract: We constructed a 3D seismic P velocity model for the extended Barents Sea region including Svalbard (Spitsbergen), Novaya Zemlya, the Kara Sea, and the Kola-Karelia regions. The crustal model is based on a large number of existing 1D and 2D velocity profiles, constrained by geological observations, and the nominal resolution is 50 km. Each grid node is filled with a five-layer crustal model (plus water and ice), and the continuous upper mantle velocity structure is taken from published regional models. Seismic S velocities and the density structure shall be included in the near future. The final model also aims to improve seismic monitoring and verification in this region, including improved event locations and event size estimation, and a better understanding of regional seismic wave phases.
Validation of our velocity model includes forward modeling of observed travel times and relocation of seismic events. For this purpose we compiled a set of reference events with known or well-located epicenters, referred to as Ground Truth (GT) events. The GT events comprise quarry blasts and announced chemical explosions located mainly in Scandinavia and the Kola Peninsula, nuclear explosions in northwestern Russia and on Novaya Zemlya, and natural earthquakes. With these events we obtain good Pn and Sn ray coverage in the main target region. Phase arrival times of multiple events at some sites provides estimates on timing errors at some stations.
Here we present our model in terms of regional contour maps and seismic velocity transects. North-south trending transects in the western Barents Sea show Moho depths between 10 and 45 km, with average values around 35 km. Thicknesses of sedimentary layers vary considerably and reach locally more than 10 km. Strongest variations are observed in the west between northern Norway and Svalbard, whereas transects further east exhibit a more simple, layered crustal structure. As the crustal velocity structure in the source or receiver region strongly influences phase arrival times, the detailed 3D model is therefore also important for the monitoring of events at regional distances, where corresponding ray paths run mainly in the upper mantle. Furthermore, there are evidences for significant velocity variations in the mantle of our study region. We show comparisons of observed travel times and GT data along selected transects and initial time correction surfaces calculated from our model.
Abstract: xisting global and regional tomographic models have limited resolution in the European Arctic due to the small number of seismic stations, relatively low regional seismicity, and poor knowledge of the crustal structure. During the last decades, new seismic stations were permanently or temporarily installed in and around this region. However, many of the data from these stations are not easily accessible via the international data centers but only by direct request to various data operators. Recently, a new crustal model of the Barents Sea and surrounding areas had been derived in a joint project between the University of Oslo, NORSAR and the USGS (Bungum et al., 2005). This model with its detailed information on crustal thickness and sedimentary basins in the area helps to constrain the tomographic inversion of the upper mantle velocity structure based on surface wave data. To improve the surface wave data set in the region, we extensively searched for broadband data from stations in the area from the beginning of the 1970s until 2005 and were able to retrieve surface wave observations from the data archives at NORSAR, University of Bergen, University of Helsinki, the Kola Science Center in Apatity, and the Geological Service of Denmark in addition to data from the data centers of IRIS and GEOFON. Rayleigh and Love wave group velocity measurements from 10 sec to 150 sec period obtained on these seismograms were combined with the existing data set provided by the University of Colorado (e.g. Levshin et al., 2001). Using these data, we constructed a new 3-D shear velocity model of the crust and upper mantle beneath the European Arctic which provides higher resolution and accuracy than previous models. The new model reveals substantial variations in shear wave speeds in the upper mantle across the region. Of particular note are clarified images of the mantle expression of the continental-ocean transition in the Norwegian Sea and a deep high wave speed lithospheric root beneath Novaya Zemlya.
Notes: EOS Vol. 86, No. 52, Fall Meeting Supplement, abstract S51E-1053
Abstract: The principal objective of the present study is to compile a three-dimensional (3D) seismic velocity model of the crust and upper mantle for the larger Barents Sea region, at a spatial resolution of nominally 50×50 km. The main accomplishments so far have been concerned with compilation, collation and review of primary existing geophysical data, including first of all deep seismic wide-angle profiles (OBS and ESP – two-ship expanded spread profiles), deep multichannel seismic reflection (MCS) profiles, and shallower 1D velocity profiles. The main source of data has been a data base compiled over many years at the University of Oslo (UiO), supplemented by data compiled at the United States Geological Survey (USGS) and from collaboration partners in Norway and in Russia. Subsequently, detailed comparisons of data and models between the UiO and USGS have been performed, and unified criteria for quality assessment have been developed. The result is a full integration of the underlying data, and a common unified model.
The 50×50 km grid tiles in the target region have been defined in an optimum way such that the tiles form a fully equidistant grid. The filling of the grid tiles so far shows a very good coverage in the western Barents Sea, a reasonable coverage also in the Novaya Zemlya region, but is less-constrained in the northern and northeastern parts of the target region. The results show that the depths to Moho vary from about 10 km in the oceanic crustal domain to more than 40 km in coastal regions of Norway and Russia and in the Kara Sea, while sediment thicknesses are 15-20 km in the south-western and eastern parts of the Barents Sea. Following the USGS methodology each grid tile is represented with layers for ice, water, soft sediments, hard sediments, and crystalline upper, middle and lower crust. Finally there is a layer describing the seismic velocity and density of the uppermost mantle, which is controlling Pn and Sn travel times.
We have also acquired an upper mantle model (Shapiro and Ritzwoller, 2002) that eventually may be integrated with our new crustal model and tested for a Ground Truth (GT) data base of about 50 events that also has been established as a part of this study. Since some large regional distances also will be used when comparing observed GT travel times with computed travel times through the established model, a mantle velocity model down to about 400 km is needed. To facilitate this travel time testing a number of 2D profiles have been established through the regions that are well covered with initially sampled 1D velocity functions, followed by different smoothing and interpolating techniques, dependent on which modelling method that will be applied. So far 2D ray tracing and finite difference methods have been used here, with preliminary testing also of 3D methods.
Since the grid tiles filled with primary data are unevenly distributed it has been necessary also to develop and to test methodologies for interpolation and extrapolation, in order to have all tiles filled. In regions where the data coverage is not dense, geological provinces are defined which are supposed to hold similar tectonic histories. In this case a method is being tested in which the velocity profiles have been used to calculate the crystalline crustal thickness as a function of sediment thickness. Since the crustal rock velocity distribution for a particular geological province is known, a regional depth-to-basement map can used to calculate representative 1D velocity-depth functions for the entire crust. This technique looks promising as a means for providing an equally sampled crustal model, as is required for seismological purposes.
Abstract: Our study region includes the entire Barents Sea from northern Fennoscandia to the region north of Svalbard, and it extends from the Norwegian Sea and the Fram Strait to the western Kara Sea, surrounding the declared Russian nuclear test sites on Novaya Zemlya. This region consists of Precambrian shields, oceanic and extended continental crust, fold and thrust belts, and deep sedimentary basins.
Our main objective is to develop a 3D seismic velocity model of the crust and upper mantle in the study region. This is a two-year project funded by the U.S. DoE and scheduled until September 2005. To compile our 3D model with a nominal resolution of 50×50 km, we use results from previous, crustal-scale 2D seismic wide-angle and normal incidence profiles. Passive seismological recordings, previous receiver function studies, and potential field data provide additional constraints for our model. The upper mantle of our model is based on published regional models. In order to verify our a priori model, we model observed arrival times of several seismic phases with known origins (ground truth events). The final model aims to improve seismic monitoring in this region, which includes improved event locations and event size estimation, and a better understanding of regional seismic wave phases. The model will also serve as a basis for estimating and calibrating travel times of regional phases in this region, and it will complement, with a better resolution, global crustal models such as CRUST5.1.
Here we focus on the crustal structure in the study region and present seismic velocity structures in terms of preliminary regional contour maps, and seismic velocity transects. N-S trending transects in the western Barents Sea show Moho depths between 10 and 45 km, with average values around 35 km. Thicknesses of sedimentary layers vary considerably and reach locally more than 10 km. Strongest variations are observed in the west between northern Norway and Svalbard, whereas transects further east exhibit a more simple, layered crustal structure. The crustal velocity structure causes variations of first arrival travel times of more than 1 s in distances more than about 300 km from the sources. The detailed crustal velocity model is therefore also important for the monitoring of events at regional distances, where corresponding ray paths run mainly in the upper mantle. Furthermore, delay times of a few seconds for events at Novaya Zemlya, recorded at the NORSAR array in Norway, indicate significant velocity variations in the mantle also for our study region.
Abstract: We present a 3D seismic velocity model for the extended Barents Sea region, including Svalbard, Novaya Zemlya, the Kara Sea and the Kola-Karelia Regions. The purpose of developing a higher-resolution velocity model is to improve generally the seismic event localization in the target region. The model should improve the future monitoring facilities and the accompanied travel-time modeling. Initial testing of the model will base on the modeling of a series of seismic ground-truth events recorded by the surrounding stations. The model has a spatial resolution of 50×50 km and includes 1490 nodes. Each node is filled with a 5-layer crustal model (plus water/ice- and additional mantle layers): Nodes within the oceanic and continental domains bear two sedimentary layers (low/high Vp) and three "crystalline" crustal layers (low/intermediate/high Vp). Basis of this model is a recent compilation of seismic velocities taken from published wide-angle profiles, unpublished ESP profiles and additional gravity modeling along deep MCS-profiles. Over 700 1D velocity profiles are collected. In order to interpolate the velocity/depth-information from the randomly distributed 1D profiles on the equal-spaced grid, the following technique was applied: Analyzing the database, we found a strong linear trend between the total thickness of the sediment layers and the remaining crystalline crust within pre-defined continental provinces (e.g. distinct sedimentary basins, plateaus, basement highs, etc.). Area-wide depth-to-basement information, based on the integrated analysis of seismic, gravity and magnetic data is used to calculate the crystalline and total crustal thicknesses as functions of sediment thickness. The mean seismic velocities and thickness-rates for each of the 5 crustal layers are calculated from the compiled database. Analysis of the regressions show that about 75-90% of the data input is fitted by the calculated functions with a maximum of 20% deviation relative to its total thickness. The compiled database provides further excellent statistical background for composition of crystalline crustal rocks in the target region. The overall distribution of seismic velocities within crystalline crust shows a clear bimodal structure with velocity peaks at 6.4 and 6.8 km/s. First modeling tests along four selected transects were carried out to evaluate the constructed 3D seismic model. According to the tests travel-time deviations can exceed 2 s at distances of 300-800 km (by comparison to a standard 1D model).
Abstract: Magnetotelluric (MT) and seismic methods provide information about the conductivity and velocity structure of the subsurface on similar scales and resolutions. The independent electrical and seismic tomograms can be combined, using a classification approach, to map lithologic, tectonic, and hydrologic boundaries. The method employed is independent of theoretical/empirical relations linking electrical and seismic parameters, and based solely on the statistical correlation of physical property models in parameter space. Regions of high correlation (classes) can in turn be examined in the spatial domain. The spatial distribution of these clusters, and the boundaries between them, provide structural information not always evident from the individual models. The method is applied to coincident seismic velocity and electrical resistivity models from two active transform margins. Along the San Andreas Fault, classification studies reveal the strong lithological contrast across the fault, suggesting it is sub-vertical in the upper crust throughout central California. A possible hydrologic boundary is further identified to the northeast of the fault. Classification studies along the Dead Sea Transform reflect the dominant lithologies surrounding the fault, and suggest the fault is again vertical in the upper crust, but offset to the east of the surface trace. There are indications that the basement is uplifted by 2 km east of the fault. These results suggest a quantitative, joint interpretation of MT and seismic data can greatly improve our ability to delineate lithologic, tectonic, and hydrologic boundaries, thus overcoming some of the resolution limitations inherent to the MT and seismic methods.
Abstract: A combined statistical analysis has been applied to MT and seismic models. A high degree of correlation between resistivity and seismic velocity is observed, with several pronounced clusters in resistivity/velocity space. These clusters are further used to map out regions of uniform physical properties, and the borders between these regions are often sharply defined. A comparison of seismic scattering and resistivity models suggest that pronounced scatterers are coincident with pronounced resistivity gradients. Taken together, these studies suggest the Dead Sea Transform forms a sharp vertical boundary within the upper 5 km of the crust and is offset to the east relative to the surface trace.
Abstract: Coincident seismic and magnetotelluric data, collected as part of the international DESERT (DEad SEa Rift Transect) project, have provided high-resolution velocity and resistivity images of the upper-crustal structure of the Dead Sea Transform. While changes in velocity or resistivity may signal geologic, tectonic, or hydrologic boundaries, looking at these properties in tandem allows for more precise resolution of such boundaries. A combined statistical analysis of the DESERT resistivity and velocity models will be presented. Data from both local and regional profiles exhibit a strong correlation between velocity and resistivity. Cluster analysis can be further used to map out regions of uniform physical properties, and the borders between these regions are often sharply defined. Preliminary results suggest the Dead Sea Transform forms a sharp vertical boundary within the upper 5 km of the crust.
Abstract: With controlled seismic sources and specifically designed receiver arrays, we imaged a subvertical boundary between two lithological blocks at the Arava Fault (AF) in the Middle East. The AF is the main strike-slip fault of the Dead Sea Transform (DST) between the Dead Sea and the Red Sea. Our imaging (migration) method is based on array beamforming and coherency analysis of P-to-P scattered seismic phases. We use a 1-D background velocity model and the direct P arrival as a reference phase. A spread function describing energy dispersion at localised point scatterers and synthetic calculations for large planar structures provided resolution estimates of the images. We resolve a 7 km long steeply dipping reflector offset roughly 1 km from the surface trace of the AF. The reflector can be imaged down to about 4 km depth. Our results suggest that the AF consists of one dominant fault strand in the uppermost crust. Previous and ongoing studies in this region have shown a strong contrast across the fault: low seismic velocities and electrical resistivities west and high velocities and resistivities east of it. We therefore suggest that the imaged reflector marks the contrast between sedimentary fill in the west and precambrian rocks in the east. This implies that the boundary between the two blocks, i.e. the actual fault location, is about 1 km east of the surface trace of the AF.
Abstract: Within the DESERT project, the structure of crust and upper mantle in the southern part of the Dead Sea Transform (DST) was studied by a series of geophysical experiments. This transform, stretching from the Red Sea to the Tauros-Zagros collision zone, is one of the world's major active continental shear zones, exhibiting a total slip of about 100 km within the last 20 Myr. In the southern part, the Arava fault (AF) is considered to be the main fault strand.
Latest seismic investigation was a small-scale high-resolution experiment (receiver distance: 5 m, source distance: 20 m) which provided detailed P wave velocity models (first-break tomography) and reflection images of the shallow subsurface structure (< 1000 m) along eight 1 km long profiles crossing the AF at a 10 km long segment. These images directly complement previous studies at larger scale and the analysis of explosion generated guided waves in the same area.
We observe a strong cross-fault velocity contrast at depths greater than 1 km, with higher velocities east and lower velocities west of the fault (which we relate to the sedimentary basin fill). In the uppermost layers (< 100 m) the velocity images appear in part patchy, on some profiles the AF seems to distinguish domains with different velocities. Even in the high-resolution tomographic pictures we see no indication for a fault-zone related low-velocity zone, but CMP stacks show cross-fault changes in the reflectivity pattern. The observations of guided waves suggest that at some segments the fault shows a very narrow sub-vertical low-velocity layer (< 20 m wide).
Our results suggest that the AF is characterized by the juxtaposition of different blocks separated by a very narrow damage zone. This can be explained by the fact that the total slip within the DST system is/was distributed in space and time over several fault strands, resulting in a reduced slip on the currently active strand of the AF. Furthermore, the shallow velocity structure probably reflects the interaction of movement along the fault and the deposition of sediments.
Abstract: Which information can we expect by using seismic methods as a monitoring tool for geothermal activities in sedimentary rocks? To answer this question, it is necessary to evaluate the resolving capability of seismic methods at target depths. Therefore a case study in the northeastern part of Germany was carried out to obtain information for an optimal configuration of such a seismic experiment. The study area is located near Groß Schönebeck, approximately 40 km north of Berlin. 630 vibrator sweeps at the same location were recorded by a fixed geophone spread. Different frequencies and sweep lengths were used to investigate stability of amplitude and waveform information in the seismic traces.
The analysis provides an estimate of the possibilities and limitations of seismic timelapse experiments for monitoring property changes in deep reservoirs. The geometry was designed to focus on the stability of a strong reflector (Permian/Triassic boundary) in 2.4 km depth, overlain by a partly unconsolidated sedimentary cover. We present results of the estimation of factors, such as source and receiver stability and influence of seismic noise, will have an impact on future experiments.
Abstract: Mit einer Länge von mehr als 1100 km gehört die Dead Sea Rift Transform (DST) zu den größten Scherzonen der Erde. Sie verläuft vom Golf von Aqaba am Roten Meer über das Tote Meer bis in die Süd-Türkei und hat in den vergangenen 18 Ma eine linkslaterale Verschiebung von ca. 105 km erfahren. Eine der wichtigsten aktiven Störungen der DST ist die Arava-Störung (Wadi Arava Fault, WAF) südlich des Toten Meeres. Ergänzend zu großräumigen seismischen 2-D Weit- und Steilwinkelmessungen in dieser Region sind im Rahmen des Projekts DESERT 2000 im näheren Umfeld um die Arava-Störung (WAF) mehrere kleinskalige, flächenhafte seismische Experimente durchgeführt worden (Controlled Source Array, CSA). Diese Experimente sollen die seismische Geschwindigkeitsstruktur in diesem Gebiet und die dreidimensionale Geometrie der Störung liefern. Dazu werden verschiedene Wellentypen betrachtet, wie beispielsweise direkte, refraktierte und im Bereich der Störung reflektierte Wellen, sowie geführte Wellen (fault zone guided waves).
Im April 2000 wurden 53 Schüsse mit Ladungen zwischen 45 und 60 kg auf drei jeweils etwa 10 km langen Profilen mit einem Geophonabstand von 100 m registriert. Die Profile waren ungefähr senkrecht zur WAF angeordnet und hatten einen Abstand von 4-5 km. Zur detaillierten Untersuchung späterer Phasen dienten zusätzlich 9 Arrays und 2 Kurzprofile bestehend aus kurzperiodischen und breitbandigen Dreikomponenten-Seismometern. Aus P-Ersteinsätzen konnte mittels tomographischer Inversion ein dreidimensionales Geschwindigkeitsmodell bestimmt werden. Der gut aufgelöste Bereich erreicht etwa 12 mal 12 km und eine Tiefe von rund 3 km im Zentrum des Modells. Auffällig ist ein starker Geschwindigkeitskontrast an der WAF mit höheren Geschwindigkeiten östlich der Störung. Im Norden zeigt sich ein komplizierteres Bild, das mit teilweise schon bekannten geologischen Strukturen korrelliert.
Für ein detaillieres Abbild der WAF in Oberflächennähe sind im Oktober 2001 jeweils rund 50 Schüsse (300 g) auf acht 1 km langen Profilen senkrecht zur WAF mit einem Geophonabstand von 5 m aufgenommen worden. Die gewonnenen Daten werden reflexionsseismisch und mittels zweidimensionaler Tomographie ausgewertet. Von diesem Teil des Experiments werden erste CMP-Stapelsektionen und Ergebnisse der Inversion für P-Geschwindigkeiten gezeigt.
Abstract: The Dead Sea Transform (DST), stretching from the Red Sea to the Tauros-Zagros collision zone, is one of the world's major active continental shear zones, exhibiting a total slip of about 100 km within the last 20 Myr. In the southern part, the Arava Fault (AF) is considered to be the main fault strand.
Within the DESERT project, the structure of crust and upper mantle in the southern part of the transform was studied by a series of geophysical experiments. Latest seismic investigation was a small-scale high-resolution experiment (receiver distance: 5 m, source distance: 20 m) which provided detailed P wave velocity models (first-beak tomography) and reflection images of the shallow subsurface structure (< 1000 m) along 8 one kilometre long profiles crossing the AF at a 10 km long segment. These images directly complement previous studies at larger scale and the analysis of explosion generated guided waves in the same area.
We observe a strong cross-fault velocity contrast at depths greater than 1 km, with higher velocities east and lower velocities west of the fault (which we relate to the sedimantary basin fill). In the uppermost layers (< 100 m) the velocity images appear in part patchy, on some profiles the AF seems to distinguish domaines with different velocies. Even in the high-resolution tomographic pictures we see no indication for a fault-zone related low-velocity zone, but guided waves suggest that at some segments the fault shows of a very narrow subvertical low-velocity layer (< 20 m wide). Our results suggest that the uppermost part of the AF is characterized by a very narrow damage zone. This can be explained by the fact that the total slip within the DST system is/was distributed in space and time over several fault strands, resulting in a reduced slip at the AF. Furthermore, the shallow velocity structure probably reflects the interaction of movement along the fault and the deposition of sediments.
Abstract: As part of the Dead Sea Rift Transect Project (DESERT 2000) we conducted an active seismic experiment to study the small-scale structure of the Wadi Arava fault zone (WAF). This fault is considered the principal active fault in the southern part of the Dead Sea Transform system, which extends over a length of about 1000 km and which is characterized by a sinistral movement of 105 km within the last 18 Ma. One of the aims of the project was to generate and observe seismic guided waves in the fault zone. Guided waves are multiple-reflected waves propagating in narrow low-velocity channels. They provide information on properties and geometry of the fault zones itself which are often not obtained by conventional seismic experiments. In April 2000 we placed 12 detonations within or very close to the surface trace of the WAF. The charges consisted of 45 kg of chemical explosives placed in 20 m boreholes. Seismic signals were recorded at 5 densely-spaced linear geophone arrays crossing the fault. The recordings show prominent wave trains emerging from 2 in-fault explosions. We interpret these phases as waves being guided by a fault-zone related low-velocity layer. Observations of these wave trains are confined to certain segments of the receiver lines and occur only for certain shot locations, matching the surface trace of the WAF. They show high energy and monochromatic behaviour. We model the guided waves by using an analytical solution for the wavefield (Ben-Zion and Aki, 1990). The model is characterized by a vertical low-velocity layer embedded in two quarter spaces. Results of the analysis will be presented. Although strong trade-offs between the free parameters are present, preliminary calculations suggest that the observations are adequately fit by models with a narrow (10 to 30 m) vertical layer with a reduced S wave velocity (15 to 25%). We relate the vertical low-velocity layer to the damage zone of the WAF. Compared to other major continental shear zones, the damage zone of the shallow (!) part of the WAF at this location seems to be rather narrow.
Abstract: With a length of more than 1100 km the Dead Sea Rift Transform (DST) comprises one of the largest transform systems on Earth. During the last 18 Ma it experienced a left-lateral motion of about 100 km, and todays activity is indicated by recent seismicity.
As part of the project DESERT 2000 we conducted an active seismic experiment in and around one of the major faults of the DST, the Arava Fault, complementing data from wide-angle refraction, near-vertical reflection, and passive seismological experiments. Main goal of this experiment is the investigation of properties and 3-D geometry of the fault by the observation of fault reflections, converted waves, and fault zone guided waves.
We deployed 270 geophone channels on three 10 km lines crossing the Arava Fault roughly W-E and 9 mini arrays of 10 stations equipped with three-component seismometers. 53 chemical explosions within shot clusters around the study area were used as seismic sources.
We see clear secondary arrivals which can be interpreted as reflections of the steeply dipping Arava Fault, and the seismic data reveal a rather complex velocity structure west and east of the fault. The mini arrays running continuously for a period of 7 days recorded various regional and teleseismic events and two small-magnitude local earthquakes in the direct vicinity of the fault. We present preliminary results of our studies.
Abstract: As part of the Dead Sea Rift Transect Project (DESERT 2000) we conducted an active seismic experiment to study the small-scale structure of the Wadi Arava fault zone (WAF). This fault is considered the principal active fault in the southern part of the Dead Sea Transform system, which extends over a length of about 1000 km and is characterized by a sinistral movement of 105 km within the last 18 Ma. One of the aims of the project was to generate and observe seismic guided waves in the fault zone. Guided waves are multiple-reflected waves propagating in narrow low-velocity channels. They provide information on properties and geometry of the fault zones itself which are often not obtained by conventional seismic experiments. In April 2000 we placed 12 detonations within or very close to the surface trace of the WAF. The charges consisted of 45 kg of chemical explosives placed in 20 m boreholes. Seismic signals were recorded at 5 densely-spaced linear geophone arrays crossing the fault. The recordings exhibit prominent wave trains emerging from 2 in-fault explosions. We interpret these phases as waves being guided by a fault-zone related low-velocity layer. Observations of these wave trains are confined to certain segments of the receiver lines and occur only for certain shot locations, matching the surface trace of the WAF. They show high energy and monochromatic or weak dispersive behaviour. We model the guided waves by using an analytical solution for the wavefield (Ben-Zion and Aki, 1990). The model is characterized by a vertical low-velocity layer embedded in two quarter spaces. Although strong trade-offs between the free parameters are present, preliminary calculations suggest that the observations are adequately fit by models with a 10 to 30 m thick vertical layer where the S wave velocity is reduced by approximately 15 to 25% relative to the surrounding rock. We relate the vertical low-velocity layer to the damage zone of the WAF. Compared to other major continental shear zones, the damage zone of the shallow part of the WAF at this location seems to be rather narrow. The WAF shows 100 km of cumulative offset, about a fifth of the San Andreas Fault. The damage zone is proportionately thinner, suggesting a scaling between damage zone thickness and offset.
Abstract: With a length of more than 1100 km the Dead Sea Rift Transform (DST) comprises one of the largest transform systems on Earth running from the Gulf of Aqaba to southern Turkey. During the last 18 Ma it experienced a left-lateral motion of about 100 km, and todays activity is indicated by recent seismicity. Under the roof of the multinational and interdisciplinary research project "Dead Sea Rift Transect (DESERT 2000)" we conducted an active seismic experiment (Controlled Source Array, CSA) in and around one of the major faults of the DST, the Arava Fault, complementing data from wide-angle refraction, near-vertical reflection, and passive seismological experiments. Main goal of this experiment is the investigation of properties and 3-D geometry of the fault(s) by the observation of fault reflections, converted waves, and fault zone guided waves. In April 2000 we deployed 270 geophone channels (spacing 100 m) on three 10 km lines crossing the Arava Fault roughly W-E. Additionally 9 mini arrays of 10 to 13 stations (aperture ~1 km) equipped with three-component seismometers (some of them broadband) were installed in this area of 10 by 10 km. 41 chemical explosions within 7 shot clusters around the study area and at the ends of the lines were used as seismic sources. Furthermore, with two additional densely-spaced receiver lines crossing the fault (spacing ~15 m) we tried to record fault guided waves generated by 12 in-fault shots. On several recordings we see clear secondary arrivals which can be interpreted as reflections of the steeply dipping Arava Fault. Furthermore, the seismic data reveal a rather complex velocity structure west and east of the fault. The mini arrays running continuously for a period of 7 days recorded various regional and teleseismic events and two small-magnitude local earthquakes in the direct vicinity of the fault. We present preliminary results of our studies.
Abstract: Within the Indonesian German MERAPI project the seismic structure of the Merapi Volcano is investigated with an active seismic experiment. First seismic profiling was performed along 4 profiles arranged radially with respect to the summit. Each profile consists of 30 3-component seismometers with a station spacing of 100 m. As a seismic source, a 2.5 l mudgun was placed in three different water basins at a distance of 5 km from the summit. This special seismic source has a well defined spectral characteristic and high repetition accuracy.
Along the 3 km long near-source profiles we observed refracted waves. The corresponding 1D depth profiles show low-velocity surface layers of some hundred m/s P-wave velocity. A greater depth the P-wave velocity increases to approximately 3000 m/s.
The volcanic strata leads to strong scattering and absorption effects. A significant decrease of the main signal frequency from approximately 10 Hz to 5 Hz between 0 km and 3 km offset can be observed. This frequency shift coincides with the formation of a long coda with maximum amplitudes several seconds after the first break. The coda wavefield is mainly incoherent, which suggests some sort of random character of signal formation. We use randomly distributed scatterers to model the seismogram envelopes and to determine scattering parameters as well as Coda-Q-values.
An analysis in the frequency-wavenumber domain shows, however, that besides the incoherent energy, the interference of coherent waves travelling back and forth along the profile also contibutes to the coda.
We conclude that to understand the mechanisms of the natural volcanic seismic sources of Merapi, scattering and absorption effects along the propagation path must be considered.
Abstract: Within the Indonesian-German research project MERAPI the subsurface structure of the high risk volcano Mt. Merapi, Java/Indonesia, is investigated with an Active Seismic Experiment (ASE). This experiment uses artificial sources with a high repitition accuracy and seismometer lines arranged radially with respect to the summit. Only three component seismometers are included in the spread.
The objective of this thesis is to analyze the vector properties of the recorded wave field, to make use of selected polarisation parameters in a multicomponent digital filter and to interpret the resulting travel time curves. The thesis mainly focusses on the data of one seismic profile in the south of Mt. Merapi. The processing of the vertical component of that profile includes the stacking of single shots, the frequency analysis with a comparison of artificial and volcanic signals, and the calculation of the frequency-wavenumber spectrum (f-k spectrum). A f-k filter can partly separate waves propagating forward and backwards. The polarisation analysis yields a high degree of linear polarisation for the first arrival, while the strong coda waves show diffuse results. The direction of polarisation is found out to be a parameter distinguishing later, reflected onsets from the remaining wave field. The polarisation filter combines a measure of rectilinearity and the direction of polarisation, and it emphasizes later onsets of reflected waves at profiles with an offset up to 4 km. The application of the filter to data of one far offset profile (up to about 7.5 km) amplifies the first onset and enables an improved evaluation of the travel times. Also at this profile reflected waves become visible. The reflections can be explained with a simple two dimensional model based on the ray theory, and it is shown that they are caused by open fracture zones.
Notes: In German, English title: Polarisation analysis of refraction seismic data from Merapi volcano, Indonesia.
Related software is published as part of CWP/SU SeismicUnix at http://www.cwp.mines.edu/cwpcodes/.
Abstract: The Dead Sea Transform (DST) is a prominent shear zone in the Middle East. It separates the Arabian plate from the Sinai microplate and stretches from the Red Sea rift in the south via the Dead Sea to the Taurus-Zagros collision zone in the north. Formed in the Miocene 17 Ma ago and related to the breakup of the Afro-Arabian continent, the DST accommodates the left-lateral movement between the two plates. The study area is located in the Arava Valley between the Dead Sea and the Red Sea, centered across the Arava Fault (AF), which constitutes the major branch of the transform in this region.
A set of seismic experiments comprising controlled sources, linear profiles across the fault, and specifically designed receiver arrays reveals the subsurface structure in the vicinity of the AF and of the fault zone itself down to about 3-4 km depth. A tomographically determined seismic P velocity model shows a pronounced velocity contrast near the fault with lower velocities on the western side than east of it. Additionally, S waves from local earthquakes provide an average P-to-S velocity ratio in the study area, and there are indications for a variations across the fault. High-resolution tomographic velocity sections and seismic reflection profiles confirm the surface trace of the AF, and observed features correlate well with fault-related geological observations.
Coincident electrical resistivity sections from magnetotelluric measurements across the AF show a conductive layer west of the fault, resistive regions east of it, and a marked contrast near the trace of the AF, which seems to act as an impermeable barrier for fluid flow. The correlation of seismic velocities and electrical resistivities lead to a characterisation of subsurface lithologies from their physical properties. Whereas the western side of the fault is characterised by a layered structure, the eastern side is rather uniform. The vertical boundary between the western and the eastern units seems to be offset to the east of the AF surface trace.
A modelling of fault-zone reflected waves indicates that the boundary between low and high velocities is possibly rather sharp but exhibits a rough surface on the length scale a few hundreds of metres. This gives rise to scattering of seismic waves at this boundary. The imaging (migration) method used is based on array beamforming and coherency analysis of P-to-P scattered seismic phases. Careful assessment of the resolution ensures reliable imaging results.
The western low velocities correspond to the young sedimentary fill in the Arava Valley, and the high velocities in the east reflect mainly Precambrian igneous rocks. A 7 km long subvertical scattering zone (reflector) is offset about 1 km east of the AF surface trace and can be imaged from 1 km to about 4 km depth. The reflector marks the boundary between two lithological blocks juxtaposed most probably by displacement along the DST. This interpretation as a lithological boundary is supported by the combined seismic and magnetotelluric analysis. The boundary may be a strand of the AF, which is offset from the current, recently active surface trace. The total slip of the DST may be distributed spatially and in time over these two strands and possibly other faults in the area.
Abstract: Since the advent of plate-tectonics the Dead Sea Transform (DST) has been considered a prime site to examine geodynamic processes. It has accommodated a total of 105 km of left-lateral transform motion between the African and Arabian plates since early Miocene (~20 My). Large historical earthquakes on the DST with magnitudes up to 7 and the 1995 Nueiba M7.2 event, as well as ongoing micro-seismic activity show that the DST is a seismically active plate boundary. The DST therefore poses a considerable seismic hazard to Palestine, Israel, and Jordan. The DST segment between the Dead Sea and the Red Sea, called Arava/Araba Fault (AF), is studied in DESERT in detail, using a multi-disciplinary and multi-scale approach from the micrometer to the plate-tectonic scale. This volume contains the results of the DESERT project running from 2000 to 2006. It opens with a review paper (DESERT Group, 2009) followed by 33 special papers.
Abstract: In this project we have addressed the problem of energy partitioning at distances ranging from very local to regional for various kinds of seismic sources. On the local and regional scale (20-220 km) we have targeted events from the region offshore Western Norway where we have both natural earthquake activity as well as frequent occurrence of underwater explosions carried out by the Norwegian Navy. On the small scale we have focused on analysis of observations from an in-mine network of 16-18 sensors in the Pyhäsalmi mine in central Finland. This analysis has been supplemented with 3-D finite difference wave propagation simulations in a realistic mine model to investigate the physical mechanisms that partition seismic energy in the near source region in and around the underground mine.
The results from modeling and analysis of local and regional data show that mean S/P amplitude ratios for explosions and natural events differ at individual stations and are in general higher for natural events and frequency bands above 3 Hz. However, the distributions of S/P ratios for explosions and natural events overlap in all analyzed frequency bands. Thus, for individual events in our study area, S/P amplitude ratios can only assist the discrimination between an explosion or a natural event. This observation is supported by synthetic seismograms calculated for simple 1-D models which demonstrate that explosions also generate shear-wave energy if they are fired close to an interface with a strong material contrast (as is the case for most explosions), e.g., free surface or the ocean bottom. The larger difference in S/P ratios between earthquakes and explosions for higher frequencies can be explained by the fact that at low frequencies (larger wavelengths), discontinuities and structural heterogeneities in the explosion source region are stronger generators of converted S energy. The S*-phase, for example, is most efficiently generated whenever an explosion source is located close (within one wavelength) to a strong discontinuity.
The Pyhäsalmi explosions have generally lower S/P ratios than the rockbursts for all frequencies, but the difference is far too small to be significant for classification purposes. The maxima for the explosion distributions are all below 2, whereas they are all above 2 for the rockbursts. The rockbursts also have a wider distribution of S/P ratios, which can be explained by the variability of the radiation patterns from the rockburst sources. S/P ratios for explosions and rockbursts located in the same small area of the mine show results very similar to those for the full data set. This indicates that the observed differences in S/P ratios between explosions and rockbursts are due to differences in the source characteristics, and not due to propagation effects along paths in the mine. 3-D finite-difference simulations were used to model seismic events within the Pyhäsalmi mine. In particular, a January 26, 2003 rockburst was modeled at frequencies of 50 Hz (4 meter grid) and 100 Hz (2 meter grid). We were able to match the characteristics of the observed data at 50 Hz particularly well, and the characteristics of the 100 Hz data reasonably well. These results help validate the 3-D geologic mine model and the reliability of our simulations. The simulations showed that significant shear-energy can be produced due to the geologic and structural heterogeneities within the mine. In fact, mode-converted shear-energy generated from mine heterogeneity can dominate the compressional energy from an explosive source. A strong correlation is observed between the distance of a source from a mine heterogeneity and the magnitude of generated shear-energy. The ratio of shear to compressional energy is about a factor of two larger when the source is located within one wavelength from a mine heterogeneity. The simulations also suggest that excavated mine volumes are significantly stronger contributors to shear-energy generation than geologic heterogeneities. However, the simulations reveal that the magnitude of shear-energy generated as part of a shear-producing source mechanism (e.g., rockburst, mine collapse) can be as large or larger than that caused by heterogeneity within the mine.
Abstract: This short contribution is a description of data now available in NORSAR's data archives from a temporary deployment during 2002-2004 of six seismic stations in northern Norway and Finland.
Explosions in underground as well as open-pit mines in the Khibiny massif of the Kola Peninsula of northwestern Russia are conducted on a frequent and relatively regular basis. It was decided to supplement the network of permanent stations in northern Fennoscandia and northwest Russia with temporarily deployed stations, in order to record these explosions, as well as other mining explosions and natural events occuring in this general area. The six temporary stations were deployed along two profile lines, extending westwards from the Khibini massif. The rationale for this deployment was to collect data to examine distance as well as azimuthal dependence of seismic discriminants. The southernmost of the two profile lines runs through the permanent seismic array ARCES in northern Norway.
Abstract: The Dead Sea Transform (DST) is a prominent shear zone in the Middle East. It separates the Arabian plate from the Sinai microplate and stretches from the Red Sea rift in the south via the Dead Sea to the Taurus-Zagros collision zone in the north. Formed in the Miocene 17 Ma ago and related to the breakup of the Afro-Arabian continent, the DST accommodates the left-lateral movement between the two plates. The study area is located in the Arava Valley between the Dead Sea and the Red Sea, centered across the Arava Fault (AF), which constitutes the major branch of the transform in this region.
A set of seismic experiments comprising controlled sources, linear profiles across the fault, and specifically designed receiver arrays reveals the subsurface structure in the vicinity of the AF and of the fault zone itself down to about 3-4 km depth. A tomographically determined seismic P velocity model shows a pronounced velocity contrast near the fault with lower velocities on the western side than east of it. Additionally, S waves from local earthquakes provide an average P-to-S velocity ratio in the study area, and there are indications for a variations across the fault. High-resolution tomographic velocity sections and seismic reflection profiles confirm the surface trace of the AF, and observed features correlate well with fault-related geological observations.
Coincident electrical resistivity sections from magnetotelluric measurements across the AF show a conductive layer west of the fault, resistive regions east of it, and a marked contrast near the trace of the AF, which seems to act as an impermeable barrier for fluid flow. The correlation of seismic velocities and electrical resistivities lead to a characterisation of subsurface lithologies from their physical properties. Whereas the western side of the fault is characterised by a layered structure, the eastern side is rather uniform. The vertical boundary between the western and the eastern units seems to be offset to the east of the AF surface trace.
A modelling of fault-zone reflected waves indicates that the boundary between low and high velocities is possibly rather sharp but exhibits a rough surface on the length scale a few hundreds of metres. This gives rise to scattering of seismic waves at this boundary. The imaging (migration) method used is based on array beamforming and coherency analysis of P-to-P scattered seismic phases. Careful assessment of the resolution ensures reliable imaging results.
The western low velocities correspond to the young sedimentary fill in the Arava Valley, and the high velocities in the east reflect mainly Precambrian igneous rocks. A 7 km long subvertical scattering zone (reflector) is offset about 1 km east of the AF surface trace and can be imaged from 1 km to about 4 km depth. The reflector marks the boundary between two lithological blocks juxtaposed most probably by displacement along the DST. This interpretation as a lithological boundary is supported by the combined seismic and magnetotelluric analysis. The boundary may be a strand of the AF, which is offset from the current, recently active surface trace. The total slip of the DST may be distributed spatially and in time over these two strands and possibly other faults in the area.
Researcher-ID: D-1816-2009 Updated: May 2012
