aethetics only

Updated: Oct 13, 2010

9th International Symposium on Deep Seismic Reflection Probing of the Continents and Their Margins

Elsevier logo

Ulvik, Norway, 18-23 June 2000

from TECTONOPHYSICS Volume 355, Issues 1-4 (30 September 2002) Edited by H. Thybo

Abstracts of papers presented for publication in a special issue of Tectonophysics

(Full copies of these papers may be purchased for about US$30 each from Elsevier through their web site and follow the "journals" link to Tectonophysics)

Deep seismic probing of the continents and their margins

Pages 1-5
H. Thybo

Elsevier online abstract

Lithospheric growth at margins of cratons

Pages 7-22
D. B. Snyder

Deep seismic reflection profiles collected across Proterozoic–Archean margins are now sufficiently numerous to formulate a consistent hypothesis of how continental nuclei grow laterally to form cratonic shields. This picture is made possible both because the length of these regional profiles spans all the tectonic elements of an orogen on a particular cratonic margin and because of their great depth range. Key transects studied include the LITHOPROBE SNORCLE 1 transect and the BABEL survey, crossing the Slave and Baltic craton margins, respectively. In most cases, the older (Archean) block appears to form a wedge of uppermost mantle rock embedded into the more juvenile (Proterozoic) block by as much as 100–200 km at uppermost mantle depths and Archean lithosphere is therefore more laterally extensive at depth than at the surface. Particularly bright reflections along the Moho are cited as evidence of shear strain within a weak, low-viscosity lower crustal channel that lies along the irregular top of the indenting wedge. The bottom of the wedge is an underthrust/subduction zone, and associated late reversal in subduction polarity beneath the craton margin emerges as a common characteristic of these margins although related arc magmatism may be minor.

Elsevier online abstract

Basement control on oblique thrust sheet evolution: seismic imaging of the active deformation front of the Central Andes in Bolivia

Pages 23-39
Ralph Hinsch, Charlotte M. Krawczyk, Christoph Gaedicke, Raul Giraudo and Daniel Demuro

In the area of the Bolivian Orocline, we examine the deformation pattern associated with the active development of a new thrust sheet. A dense grid of reprocessed 2–D seismic lines from hydrocarbon exploration industry is interpreted and a 3–D simplified structural and kinematic model is deduced. In the Boomerang Hills, onlapping Paleozoic and foredeep sediments are detached from the underlying S-dipping basement. They are thrust northeastwards by less than 2 km. Two zones can be differentiated along the Andean deformation front: (1) a W–E to NW–SE striking frontal segment of predominantly orthogonal shortening, comprising a thrust and anticline system; (2) a WSW–ENE striking lateral zone of oblique shortening within a complex system of thin-skinned strike–slip faults and minor folds. The deformation front always follows a pronounced edge in the topography of the top basement surface close to the boundary of the Paleozoic basin. The observed deformation pattern indicates intensified strain partitioning caused by the interaction of contraction direction and basement topography, which provides a near oblique ramp for the onlapping wedge of sediments. The SW–NE thrusting direction is divided into orthogonal and tangential components. These are accommodated by convergent and strike–slip structures, respectively, which sole into a common detachment horizon. The structural evolution of the new thrust sheet in the Bolivian Orocline is primarily controlled by the paleorelief of the Brazilian Shield because: (1) the shape of the basement affects the taper of the thrust wedge and localizes the deformation front and (2) small asperities in/close to the top of the basement promote fault localization. The coincidence of a relatively high basement position and a structural high of the Eastern Cordillera leads to the conclusion that the shape of the Brazilian Shield also controls the structural evolution of the pronounced eastern border of the Bolivian Orocline.

Elsevier online abstract

Deep seismic reflection profiling across the Ou Backbone range, northern Honshu Island, Japan

Pages 41-52
Hiroshi Sato, Naoshi Hirata, Takaya Iwasaki, Makoto Matsubara and Takeshi Ikawa

Knowledge of the crustal structure, especially the geometry of seismogenic faults, is key to understanding active tectonic processes and assessing the size and frequency of future earthquakes. To reveal the relationship between crustal structure and earthquake activity in northern Honshu Island, common midpoint (CMP) deep reflection profiling and earthquake observations by densely deployed seismic stations were carried out across the active reverse faults that bound the Ou Backbone range. The 40-km-long CMP profiles portray a relatively simple fault geometry within the seismogenic layer. The reverse faults merge at a midcrustal detachment just below the base of the seismogenic layer, producing a pop-up structure that forms the Ou Backbone range. The top of the reflective middle to lower crust (4.5 s in travel time (TWT)) nearly coincides with the bottom of seismogenic layer. The P-wave velocity structure and surface geology suggest that the bounding faults are Miocene normal faults that have been reactivated as reverse faults.

Elsevier online abstract

Seismological features of island arc crust as inferred from recent seismic expeditions in Japan

Pages 53-66
Takaya Iwasaki, Toshikatsu Yoshii, Tanio Ito, Hiroshi Sato and Naoshi Hirata

Crustal studies within the Japanese islands have provided important constraints on the physical properties and deformation styles of the island arc crust. The upper crust in the Japanese islands has a significant heterogeneity characterized by large velocity variation (5.5–6.1 km/s) and high seismic attenuation (Qp=100–400 for 5–15 Hz). The lateral velocity change sometimes occurs at major tectonic lines. In many cases of recent refraction/wide–angle reflection profiles, a "middle crust" with a velocity of 6.2–6.5 km/s is found in a depth range of 5–15 km. Most shallow microearthquakes are concentrated in the upper/middle crust. The velocity in the lower crust is estimated to be 6.6–7.0 km/s. The lower crust often involves a highly reflective zone with less seismicity, indicating its ductile rheology. The uppermost mantle is characterized by a low Pn velocity of 7.5–7.9 km/s. Several observations on PmP phase indicate that the Moho is not a sharp boundary with a distinct velocity contrast, but forms a transition zone from the upper mantle to the lower crust. Recent seismic reflection experiments revealed ongoing crustal deformations within the Japanese islands. A clear image of crustal delamination obtained for an arc–arc collision zone in central Hokkaido provides an important key for the evolution process from island arc to more felsic continental crust. In northern Honshu, a major fault system with listric geometry, which was formed by Miocene back arc spreading, was successfully mapped down to 12–15 km.

Elsevier online abstract

A possible Caledonide arm through the Barents Sea imaged by OBS data

Pages 67-97
Asbjørn Johan Breivik, Rolf Mjelde, Paul Grogan, Hideki Shimamura, Yoshio Murai, Yuichi Nishimura and Asako Kuwano

The assembly of the crystalline basement of the western Barents Sea is related to the Caledonian orogeny during the Silurian. However, the development southeast of Svalbard is not well understood, as conventional seismic reflection data does not provide reliable mapping below the Permian sequence. A wide-angle seismic survey from 1998, conducted with ocean bottom seismometers in the northwestern Barents Sea, provides data that enables the identification and mapping of the depths to crystalline basement and Moho by ray tracing and inversion. The four profiles modeled show pre-Permian basins and highs with a configuration distinct from later Mesozoic structural elements. Several strong reflections from within the crystalline crust indicate an inhomogeneous basement terrain. Refractions from the top of the basement together with reflections from the Moho constrain the basement velocity to increase from 6.3 km s-1 at the top to 6.6 km s-1 at the base of the crust. On two profiles, the Moho deepens locally into root structures, which are associated with high top mantle velocities of 8.5 km s-1. Combined P- and S-wave data indicate a mixed sand/clay/carbonate lithology for the sedimentary section, and a predominantly felsic to intermediate crystalline crust. In general, the top basement and Moho surfaces exhibit poor correlation with the observed gravity field, and the gravity models required high-density bodies in the basement and upper mantle to account for the positive gravity anomalies in the area. Comparisons with the Ural suture zone suggest that the Barents Sea data may be interpreted in terms of a proto-Caledonian subduction zone dipping to the southeast, with a crustal root representing remnant of the continental collision, and high mantle velocities and densities representing eclogitized oceanic crust. High-density bodies within the crystalline crust may be accreted island arc or oceanic terrain. The mapped trend of the suture resembles a previously published model of the Caledonian orogeny. This model postulates a separate branch extending into central parts of the Barents Sea coupled with the northerly trending Svalbard Caledonides, and a microcontinent consisting of Svalbard and northern parts of the Barents Sea independent of Laurentia and Baltica at the time. Later, compressional faulting within the suture zone apparently formed the Sentralbanken High.

Elsevier online abstract

Crustal structure of the southern part of the Vøring Basin, mid-Norway margin, from wide-angle seismic and gravity data

Pages 99-126
T. Raum, R. Mjelde, P. Digranes, H. Shimamura, H. Shiobara, S. Kodaira, G. Haatvedt, N. Sørenes and T. Thorbjørnsen

During the summer of 1996, five profiles with a total length of 892.5 km and 30 recovered Ocean Bottom Seismographs (OBSs) were acquired along the southern part of the Vøring Margin, mid-Norway. These profiles are oriented in both strike and dip direction, mapping the crustal structure along the western-most part of this area. By performing both forward and inverse velocity travel-time modelling of the vertical components of the OBSs (P-waves) regional models have been established. The thickness of the Cenozoic sediments ranges from 0.5 km at the structural highs (Møre Marginal High) to more than 3 km in the Vigrid Syncline. The Cretaceous layers vary considerably in thickness, from 1.7 km close to the Jan Mayen Fracture Zone, increasing to approximately 8 km below the Helland Hansen Arch. Several intrusions are found in the Cretaceous layers and they individually range in thickness from 0.1 to 0.6 km. These intrusions can be followed through crossing profiles, in both in the models presented in this paper, and in the models inferred from earlier OBS surveys along this margin. The pre-Cretaceous sediments are modelled as a layer with P-wave velocity of more than 5 km/s. The thickness of this layer is generally less than 5 km, but locally a thickness of 6.5 km is modelled. The crystalline basement is modelled as a 6+ km/s refractor with a general trend shallowing towards south- and northwest. Along a strike-profile, the transition from continental to oceanic crust is modelled by increased P-wave velocity from 6.2 to 7.3 km/s, as the profile crosses the Jan Mayen Lineament towards the Møre Marginal High. The thickness of the continental crystalline basement is largest landward towards the Helland Hansen Arch where it exceeds 10 km. Towards the SW and NW, where the crystalline basement is elevated, the thickness decreases to less than 5 km. In the lower crust, a high velocity 7+ km/s body, interpreted as magmatic underplating, is found along three of the profiles in the present work. The SE boundary of this magmatic underplated layer is located beneath the Helland Hansen Arch and towards the Møre Marginal High. The southwestward termination is less certain, but the Jan Mayen Fracture Zone seems to control the emplacement of intrusions in the southern Vøring Basin. The depth to Moho varies from 12 to 28 km, deepening landwards.

Elsevier online abstract

Origin of the northern Indus Fan and Murray Ridge, Northern Arabian Sea: interpretation from seismic and magnetic imaging

Pages 127-143
Christoph Gaedicke, Hans-Ulrich Schlüter, Hans Albert Roeser, Alexander Prexl, Bernd Schreckenberger, Heinrich Meyer, Christian Reichert, Peter Clift and Shahid Amjad

The nature and origin of the sediments and crust of the Murray Ridge System and northern Indus Fan are discussed. The uppermost unit consists of Middle Miocene to recent channel–levee complexes typical of submarine fans. This unit is underlain by a second unit composed of hemipelagic to pelagic sediments deposited during the drift phase after the break-up of India–Seychelles–Africa. A predrift sequence of assumed Mesozoic age occurring only as observed above basement ridges is composed of highly consolidated rocks. Different types of the acoustic basement were detected, which reflection seismic pattern, magnetic anomalies and gravity field modeling indicate to be of continental character. The continental crust is extremely thinned in the northern Indus Fan, lacking a typical block–faulted structure. The Indian continent–ocean transition is marked on single MCS profiles by sequences of seaward–dipping reflectors (SDR). In the northwestern Arabian Sea, the Indian plate margin is characterized by several phases of volcanism and deformation revealed from interpretation of multichannel seismic profiles and magnetic anomalies. From this study, thinned continental crust spreads between the northern Murray Ridge System and India underneath the northern Indus Fan.

Elsevier online abstract

The effect of crustal anisotropy on reflector depth and velocity determination from wide-angle seismic data: a synthetic example based on South Island, New Zealand

Pages 145-161
N. J. Godfrey, N. I. Christensen and D. A. Okaya

When deriving velocity models by forward modelling or inverting travel time arrivals from seismic refraction data, a heterogeneous but isotropic earth is usually assumed. In regions where the earth is not isotropic at the scale at which it is being sampled, the assumption of isotropy can lead to significant errors in the velocities determined for the crust and the depths calculated to reflecting boundaries. Laboratory velocity measurements on rocks collected from the Haast Schist terrane of South Island, New Zealand, show significant (up to 20%) compressional (P) wave velocity anisotropy. Field data collected parallel and perpendicular to the foliation of the Haast Schist exhibit as much as 11% P-wave velocity anisotropy. We demonstrate, using finite-difference full-wavefield modelling, the types of errors and problems that might be encountered if isotropic methods are used to create velocity models from data collected in anisotropic regions. These reflector depth errors could be as much as 10–15% for a 10-km thick layer with significant (20%) P-wave velocity anisotropy. The implications for South Island, New Zealand, where the problem is compounded by extreme orientations of highly anisotropic rocks (foliation which varies from horizontal to near vertical), are considered. Finally, we discuss how the presence of a significant subsurface anisotropic body might manifest itself in wide-angle reflection/refraction and passive seismic datasets, and suggest ways in which such datasets may be used to determine the presence and extent of such anisotropic bodies.

Elsevier online abstract

Continental mantle seismic anisotropy: a new look at the Twin Sisters massif

Pages 163-170
Nikolas I. Christensen

Recent interpretations of upper continental mantle seismic anisotropy observations have often relied on fabric measurements and calculated anisotropies of upper mantle xenoliths. Seismic ray paths of P and S waves, which provide information on azimuthal compressional wave anisotropy and shear wave splitting, are tens to hundreds of kilometers, whereas, xenoliths are usually only a few centimeters in diameter. To place better constraints on field-based anisotropy observations and to evaluate anisotropy information provided by xenoliths, it is important to examine anisotropy in large ultramafic massifs which have originated in the upper mantle. One such massif is the Twin Sisters Range located in the western portion of the North Cascades of Washington State, USA. The Twin Sisters massif, a slab of unaltered dunite, is 16 km in length, 6 km in width and 3 km thick. Exposed along its south and west sides are mafic granulite facies rocks, which likely represent lower continental crustal fragments. The ultramafic rocks are porphyroclastic in texture, consisting of strained, flattened porphyroclasts of olivine and enstatite and strain-free olivine mosaics. Olivine fabrics are typical of those formed at high temperatures and low strain rates. Petrofabrics and calculated anisotropies of individual samples vary throughout the massif, however, overall anisotropy of the body is significant, with maximum P and S waves anisotropies of 5.4% and 3.9%, respectively. The maximum delay time for split shear waves traveling through a 100-km-thick slab is 0.8 s and two directions of shear wave singularity are observed. The directions of maximum shear wave splitting and shear wave singularities do not coincide with the directions of maximum and minimum compressional wave velocity. In general, individual hand samples show significantly higher anisotropy than the overall anisotropy of the massif. It is concluded that simple averages of xenolith anisotropies are unreliable for use in the interpretation of field anisotropy observations.

Elsevier online abstract

2D model of the crust and uppermost mantle along rift profile, Siberian craton

Pages 171-186
G. A. Pavlenkova, K. Priestley and J. Cipar

A two–dimensional model of the crust and uppermost mantle for the western Siberian craton and the adjoining areas of the Pur-Gedan basin to the north and Baikal Rift zone to the south is determined from travel time data from recordings of 30 chemical explosions and three nuclear explosions along the RIFT deep seismic sounding profile. This velocity model shows strong lateral variations in the crust and sub-Moho structure both within the craton and between the craton and the surrounding region. The Pur-Gedan basin has a 15-km thick, low-velocity sediment layer overlying a 25-km thick, high-velocity crystalline crustal layer. A paleo-rift zone with a graben-like structure in the basement and a high-velocity crustal intrusion or mantle upward exists beneath the southern part of the Pur-Gedan basin. The sedimentary layer is thin or non-existent and there is a velocity reversal in the upper crust beneath the Yenisey Zone. The Siberian craton has nearly uniform crustal thickness of 40–43 km but the average velocity in the lower crust in the north is higher (6.8–6.9 km/s) than in the south (6.6 km/s). The crust beneath the Baikal Rift zone is 35 km thick and has an average crustal velocity similar to that observed beneath the southern part of craton. The uppermost mantle velocity varies from 8.0 to 8.1 km/s beneath the young West Siberian platform and Baikal Rift zone to 8.1–8.5 km/s beneath the Siberian craton. Anomalous high Pn velocities (8.4–8.5 km/s) are observed beneath the western Tunguss basin in the northern part of the craton and beneath the southern part of the Siberian craton, but lower Pn velocities (8.1 km/s) are observed beneath the Low Angara basin in the central part of the craton. At about 100 km depth beneath the craton, there is a velocity inversion with a strong reflecting interface at its base. Some reflectors are also distinguished within the upper mantle at depth between 230 and 350 km.

Elsevier online abstract

CDP and DSS data along the Uchta–Kem profile (the Baltic Shield)

Pages 187-200
R. G. Berzin, Yu. G. Yurov and N. I. Pavlenkova

Detailed Common Depth Point (CDP) and Deep Seismic Sounding (DSS) observations were carried out along the W–E trending Kem–Uchta profile between the White Sea and the Russian–Finnish border. The cross-sections obtained from wide-angle and near–vertical reflections show different features. In the CDP cross-section, several inclined boundaries are traced from the surface down to 25–30 km depth. They correlate with the well-known fault zone between the Belomorian Mobile Belt and the Korelian Protocraton. The DSS data also show these inclined reflectors as well as a near horizontal boundary, K1 (velocity of 6.3 versus 6.4 km/s, depth of 10–15 km), under a low velocity zone. The boundary generates high amplitude wide-angle reflections, which cross the CDP inclined reflections. A lower crustal boundary, K2 (velocity of 6.7 versus 6.8 km/s, depth of 30 km), has no clear expression in the CDP reflectivity pattern as well. Strong reflections from Moho (PmP) coincide with the boundary between reflective lower crust and transparent upper mantle on the CDP section (depth 40 km). This suggests that the near-horizontal crustal boundaries and the Moho are transition zones with high velocity gradients and not sharp discontinuities. The data analysis suggests the crustal layering to be a result of rock transformation (metamorphism, change in mechanical properties) after consolidation of the crust. Tectonic features of orogenic periods were overprinted by the metamorphic processes in the lower crust. In the middle crust, a destruction (low velocity) zone and the K1 boundary were created, but the initial tectonic structure of the upper crust was preserved. The Moho is a young feature that was formed during the platform stage of the crustal evolution.

Elsevier online abstract

High-resolution reflection seismic imaging of the upper crust at Laxemar, southeastern Sweden

Pages 201-213
B. Bergman, C. Juhlin and H. Palm

A major cost in exploring the upper 1–2 km of crystalline crust with reflection seismics is the drilling required for explosive sources. By reducing the charge size to a minimum, shallow inexpensive shotholes can be drilled with handheld equipment. Here, we present results from a full-scale test using small charges for high-resolution seismic surveying over a nuclear waste disposal study site (not an actual site). Two 2–2.5–km–long crossing profiles were acquired in December 1999 with 10–m shot and geophone spacing in the Laxemar area, near Oskarshamn in southeastern Sweden. After standard processing, including dip moveout (DMO), several subhorizontal to moderately dipping reflections are imaged. Many of the dipping ones can be correlated to fracture zones observed in a ca. 1700–m–deep borehole where the profiles cross and/or to fracture zones mapped on the surface. The imaged fracture zones form a complex 3D pattern illustrating the necessity of having 3D control before interpreting seismic reflection data. Analyses of sonic and density logs from the borehole show that greenstones have significantly higher impedances than the more dominant granite found in the borehole (granite/greenstone reflection coefficient is +0.065). These greenstones may contribute to the reflectivity when associated with fracture zones. In some cases, where they are present as larger subhorizontal lenses, they may be the dominant source of reflectivity. A set of north-dipping (10°) reflectors at 3–3.5–km depth can be correlated to a similar set observed below the island of Ävrö about 3 km to the east.

Elsevier online abstract

Testing the resolution of a 3D velocity tomogram across the Chicxulub crater

Pages 215-226
J. V. Morgan, G. L. Christeson and C. A. Zelt

An integrated offshore/onshore reflection and refraction experiment was shot across the Chicxulub impact crater in 1996. The refraction data were previously inverted in 3D using first-arrival travel-time tomography. A regularized inversion, in which both data misfit and model roughness are minimized simultaneously, was used to determine a smooth velocity tomogram across the inner crater region. However, the experimental geometry for the refraction data was irregular, causing concern that this velocity model might not be well resolved. In this paper, we present a suite of checkerboard tests to investigate the lateral resolution of our velocity model. The Chicxulub crater is located partly onshore and partly offshore, with its centre close to the Yucatan coastline in Mexico. The shallow water limited acquisition of marine reflection data to distances of not closer than 25 km from the crater centre, and the centre of the structure is imaged with refraction data only. Intriguing velocity anomalies were observed across the central crater region, providing constraints on the lithological and structural form of Chicxulub. A high-velocity body within the central crater is most likely to represent lower-crustal rocks that were stratigraphically uplifted during the formation of this complex crater. The concave shape of this stratigraphic uplift suggests clues to the mechanics of large-crater collapse—the rocks appear to have moved upward and outward. An inward-dipping zone of lowered velocity has been interpreted as delimiting the outer edge of a central zone of melt-rich rocks. The resolution tests presented here indicate that these observed velocity anomalies are likely to be real.

Elsevier online abstract

Multimode migration of scattered and converted waves for the structure of the Hikurangi slab interface, New Zealand

Pages 227-246
John N. Louie, Sergio Chávez-Pérez, Stuart Henrys and Stephen Bannister

Reflectivity imaging of local earthquake seismograms has revealed the structure of the Hikurangi subduction interface at the location of two strong earthquakes that occurred in 1990. The earthquakes originated within the continental plate of the North Island of New Zealand and below in the subducting Pacific slab. We used seismograms from 500 well-located events in two earthquake sequences recorded by a small temporary seismograph deployment to directly image the structure and multiphase reflectivity of the plate interface. Synthetic tests of the imaging method show the effects of the poor 3-d geometric coverage afforded by the seismometer array. Kirchhoff summation image sections computed from synthetics show accurate depth imaging of backscattering interfaces. Phase-converting interfaces imaged with forward-scattered waves are smeared by poor ray coverage to 5-km depth inaccuracy and are only imaged over a small range of their horizontal extent. From the data, we computed image sections for P–P, P–S, S–P and S–S scattering. We mitigated imaging artifacts due to poor ray coverage with an obliquity factor, an antialiasing criterion and enhancement by resampling statistics. Imaging used a sharply layered velocity model. We tested for the effects of imaging with first-arriving headwaves by imaging through smoothly varying velocity models. For our ray geometry, early-arrival headwaves contribute little to the images.

The plate interface appears as a 3–5-km thick P–P and possibly S–S backscatterer with 5° NW dip, offset 5 km down-to-the-NW above a normal fault in the slab. When illuminated from below, a wedge of the interface on the downdip side of the slab fault forms a very prominent P–P forward scatterer. The edges of the wedge forward-scatter some S–P and S–S energy, but an order of magnitude less than the P–P forward scattering. The imbalances between forward scattering of P and S energy suggest a wedge of subducted sediment retaining significant porosity but with rigidity close to that of surrounding rocks.

Elsevier online abstract

Double–sided onshore–offshore seismic imaging of a plate boundary: "super-gathers" across South Island, New Zealand

Pages 247-263
David Okaya, Stuart Henrys and Tim Stern

Onshore–offshore seismic refraction profiling allows for the determination of crustal and mantle structures in the transition between continental and oceanic environments. Islands and narrow landmasses have the unique geometry of allowing for double–sided onshore–offshore experiments that favor the construction of composite "super-gathers" using the acquisition of onshore–offshore and ocean–bottom seismometer receiver gathers, land explosion shot gathers, and near-vertical incidence multichannel seismic (MCS) profiling. A number of sites at plate boundaries are amenable to the application of double–sided onshore–offshore imaging, including the Indo–Australian/Pacific transform boundary on South Island, New Zealand. By comparing the ratio of island width to mantle refraction (Pn) "maximum" crossover distance, using nondimensional distances, we provide an indicator of raypath "coverage" for crustal illumination. Islands or narrow land masses whose widths are less than twice their maximum crossover distance are candidates for double–sided onshore–offshore experiments. The SIGHT (South Island GeopHysical invesTigation) experiment in New Zealand is located where the width of South Island is sufficiently narrow with respect to its crustal thickness that a double–sided onshore–offshore experiment allows for complete crustal imaging of the associated plate.

Elsevier online abstract