Images of the Earth's Crust & Upper Mantle

Updated: Jun 02, 2016

3^rd International Symposium on Deep Seismic Profiling of the Continents and Their Margins

Canberra, Australia, 1-8 July 1988

from TECTONOPHYSICS Volume 173, Issues 1-4, Pages 1-641 (20 February 1990)
Edited by J. H. Leven, D. M. Finlayson, C. Wright, J. C. Dooley and B. L. N. Kennett

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 http://www.elsevier.com and follow the "journals" link to Tectonophysics)

• Pre- and post-stack migration of GLIMPCE reflection data
• Absolute reflection coefficients from deep seismic reflections
• Effective Q determination using frequency methods on BIRPS data
• Pseudo-3D study using crooked line processing from the Swiss Alpine western profile-Line 2 (Val d'Anniviers-Valais)
• Experimental three-dimensional imaging of crystal structure in the northwestern Canadian Arctic
• Seismic wavefield separation by multicomponent tau-p polarisation filtering
• Tomographic reconstruction of upper crustal velocity variations in the Arunta Block, central Australia
• The interpretation of expanding spread reflection profiles: examples from central and eastern Australia
• Wide-angle vibroseis data from the western Rhenish Massif
• Near-vertical vibroseis versus large-offset dynamite seismic reflection profiling: example of the ECORS northern France profile
• Fault-plane reflection processing and 3D display: the Darling Fault, Western Australia
• 3D isochronal modelling of reflections from the deep crust: application to reflection profiling in central Australia
• Design and operation of a deep seismic survey in the Tasman Sea Basin
• Imaging the lower crust in deep water
• The Cambridge digital seismic recorder for land and marine use
• An efficient method for sending a large seismic section from a mainframe computer to a dot matrix printer located at a remote site
• Basalts, water, or shear zones in the lower continental crust?
• Shear in the lower crust during extension: not so pure and simple
• Probing into extensional sedimentary basins: comparison of recent data and derivation of tentative models
• Comparative investigations of continental reflectivity
• Measured and calculated elastic wave velocities for xenoliths from the lower crust and upper mantle
• Seismic imaging of major tectonic features in the crust of Phanerozoic eastern Australia
• Mid-crustal detachments controlling basin deformation: ramp synforms in southwestern Queensland
• Deep seismic profiling in central Australia
• Seismic reflection profiling in the Proterozoic Arunta Block, central Australia: processing for testing models of tectonic evolution
• Australia's southern margin: a product of oblique extension
• Crystal seismic observations across the convergent plate boundary, North Island, New Zealand
• Deep seismic probing of continental crust in the lower Yangtze region, eastern China
• Deep seismic sounding in the Godavari Graben and Godavari (coastal) Basin, India
• The deep structure of the east Oman continental margin: preliminary results and interpretation
• Deep seismic profiling in the Nosop Basin, Botswana: cratons, mobile belts and sedimentary basins
• Interpretation of the reflections in the Siljan Ring area based on results from the Gravberg-1 borehole
• Reflectivity patterns in the Variscan mountain belts and adjacent areas: an attempt for a pattern recognition and correlation to tectonic units
• The identification of mantle reflections below Hungary from deep seismic profiling
• Progress in BIRPS deep seismic reflection profiling around the British Isles
• Major geodynamic processes studied from the ECORS deep seismic profiles in France and adjacent areas
• Main results of the ECORS Pyrenees profile
• Lateral variations in crystal reflectivity beneath the Paris Basin
• Deep structure of the Celtic Sea: a discussion on the formation of basins
• Dating the source of lower crystal reflectivity using BIRPS deep seismic profiles across the lapetus suture
• Geological inheritance and crystal dynamics of the northwest Scottish continental shelf
• Three-dimensional mapping of seismic reflections from the crust and upper mantle, northwest of Scotland
• Shear-wave evidence for an anisotropic lower crust beneath the Black Forest, southwest Germany
• Lower crustal petrology from wide-angle P- and S-wave measurements in the Black Forest
• Results of deep-seismic reflection investigations in the Rhenish Massif
• A geophysical study of the Hatton Bank volcanic margin: a summary of the results from a combined seismic, gravity and magnetic experiment
• Crustal anatomy of a transform continental margin
• Deep seismic and geochemical constraints on the nature of rift-induced magmatism during breakup of the North Atlantic
• Fault reactivation during Mesozoic extension in eastern offshore Canada
• Dipping shear zones and the base of the crust in the Appalachians, offshore Canada
• GLIMPCE seismic reflection evidence of deep-crustal and upper-mantle intrusions and magmatic underplating associated with the Midcontinent Rift system of North America
• Pg shingles: preliminary results from the onshore GLIMPCE refraction experiment
• Origin of deep crystal reflections: seismic profiling across high-grade metamorphic terranes in Canada
• LITHOPROBE: a scientific update-new images of the continental crust
• The structure and tectonic history of the western Canada subduction zone

Pre- and post-stack migration of GLIMPCE reflection data

Pages 1-13
Bernd Milkereit, Alan G. Green, Myung W. Lee, Warren F. Agena and Carl Spencer

GLIMPCE deep seismic reflection profiles across the Midcontinent Rift System beneath Lake Superior reveal a central asymmetric rift with an enormous thickness of volcanic and sedimentary rocks. True amplitude cmp-processing, poststack and prestack migration and forward modelling are used to improve images of steeply dipping faults, unconformities and other discontinuities in the deep seismic data. With prestack migration important steeply dipping structural features of the Lake Superior data set are revealed for the first time. Improved images of high-angle normal faults, later reactivated as reverse faults, provide key structural information for understanding the evolution of the rift system. Results illustrate that structural interpretations of complex deep reflection records, such as those recorded by GLIMPCE, should always be based on migrated data. Furthermore, depth-migrated sections provide useful starting models for forward velocity modelling of seismic data.

Elsevier online abstract

Absolute reflection coefficients from deep seismic reflections

Pages 15-23
Mike Warner

Many deep seismic reflection profiles show bright sub-horizontal layered reflections in the lower continental crust. Accurate estimates are obtained for normal incidence reflection coefficients for the brightest of these reflections by using a marine seismic acquisition system that has been calibrated using deep-water sea-bottom multiples. Minimum bounds are placed upon these reflection coefficients by assuming minimum losses due to scattering, anelastic attenuation and geometric spreading. Reflection coefficients of around 0.1 are obtained from the layered lower crust and around 0.15 from the Moho. It is argued that constructive interference between successive reflections from a layered sequence of reflectors is often unimportant, consequently these coefficients can be used to place useful limits on the absolute acoustic impedance contrasts present at depth.

Elsevier online abstract

Effective Q determination using frequency methods on BIRPS data

Pages 25-30
R. W. Hobbs

An investigation to determine Q for compressional waves from marine reflection seismic records is presented. The spectral ratio method is used to determine values for the lower crust. Results from spectral modelling from a known synthetic source wavelet are used to give minimum estimates for the whole crust. The effective Q for the reflective lower crust around Britain is found to be 500 ± 200 which is consistent with laboratory results for dry rocks and is probably too large to allow significant amounts of fluid to be present.

Elsevier online abstract

Pseudo-3D study using crooked line processing from the Swiss Alpine western profile-Line 2 (Val d'Anniviers-Valais)

Pages 31-42
L. Du Bois, L. Levato, J. Besnard, A. Escher, R. Marchant, R. Olivier, M. Ouwehand, S. Sellami, A Steck and J. J. Wagner

First results are presented of the processing of one of the seismic lines of the Swiss National Program to study the deep geological structure of the Swiss Alps, located in the Anniviers valley (Val d'Anniviers-Valais).

The rough topography of the study area gives rise to a crooked line geometry allowing analysis of lateral dips across the common mid-point (CMP) swath corresponding to a pseudo-3D study. The coherency of the seismic reflections has been improved by accounting for the lateral dips and applying a correction before performing the CMP stack. The main E-W dip found by the seismic pseudo-3D investigation confirms the surface geological survey.

Preliminary geological interpretation suggests large-scale recumbent folding at depths between 10 and 20 km.

Elsevier online abstract

Experimental three-dimensional imaging of crystal structure in the northwestern Canadian Arctic

Pages 43-52
Frederick A. Cook and Kevin C. Coflin

A three-dimensional crustal seismic reflection experiment was conducted in the northwestern Canadian Arctic where two regional linear reflection profiles intersected. VibroSeis (trademark CONOCO, Inc.) units shot broadside into one line as they were moved to varying offsets along the perpendicular line. The data were gathered into CMP bins of two-, four- and eight-fold data, providing a test of data quality versus fold. The improvement in data quality from two- to eight-fold did not warrant the higher fold. Preliminary analysis of the two-fold, 100 m × 100 m binned data allows delineation of three primary reflecting surfaces at about 1.5, 3.0 and 5.5-7.5 s. These surfaces show different orientations and a somewhat different geometry than that determined from two crossing 2D lines alone.

Elsevier online abstract

Seismic wavefield separation by multicomponent tau-p polarisation filtering

Pages 53-61
S. A. Greenhalgh, I. M. Mason, C. C. Mosher and E. Lucas

This paper presents a modified form of controlled direction reception (CDR) filter, to separate P and S waves in a multi-component seismic profile. The aim of CDR is to enhance the clarity of the section. The essence of the method (in seeking S-wave extinction) is to perform a dot product between the signal vector and the polarisation (slowness) vector during projection of the section into tau-p space, using the P-wave velocity along the array. Alternatively, S waves can be enhanced and all other wave types (including background noise) suppressed by using a conjugate rotation operator and the S velocity profile, in forming the slant stack.

A two-dimensional gain function, based on the rectilinearity of particle motion, is then applied in tau-p space as a non-linear boost function in order to enhance strongly polarised arrivals.

Synthetic and physical model examples are used to demonstrate the noise rejection properties of the CDR process. and to show how P- and S-wave seismic sections can be separately obtained.

Elsevier online abstract

Tomographic reconstruction of upper crustal velocity variations in the Arunta Block, central Australia

Pages 63-72
S. A. Greenhalgh, S. Sugiharto, C. Wright and B. R. Goleby

A seismic refraction/reflection programme of crustal/upper-mantle investigations was recently undertaken by the Australian Bureau of Mineral Resources within the Proterozoic Arunta Block, central Australia. One experiment involved the successive deployment of sixteen portable seismic recorders on two approximately orthogonal, intersecting traverses. Shots detonated on one traverse were recorded on the other traverse to give three-dimensional coverage. Shot-receiver distances varied from 1 to 39 km.

P-wave first-arrival times were processed using an OCSIRT (overlapping-cell simultaneous iterative reconstruction tomography) technique to produce a map of upper crustal velocity inhomogeneity. Synthetic data, using the same shooting geometry, were also processed as an aid to interpretation.

The final tomograms exhibit a velocity variation from 5.5 to 6.4 km s-1. The higher velocities (> 5.9 km s-1) are associated with mafic and felsic granulites of the Central Arunta Province that have been thrust over amphibolite facies gneisses and migmatites of the Southern Arunta Province. The lower velocities (< 5.9 km s-1) are generally associated with wave paths through the mylonites of the Redbank Thrust Zone and migmatitic gneisses of the Southern Arunta Province.

Elsevier online abstract

The interpretation of expanding spread reflection profiles: examples from central and eastern Australia

Pages 73-82
C. Wright, T. Barton, B. R. Goleby, A. G. Spence and D. Pfister

The Australian Bureau of Mineral Resources has recorded expanding reflection spreads coincident with regional deep seismic reflection profiles on land, both in sedimentary basins and in areas of exposed basement rocks. Arrival times of refracted, reflected and diffracted signals have been matched with synthetic times computed by ray tracing to give one- or two-dimensional velocity models for the regions immediately below the expanding spreads. Significant lateral variations in seismic velocity in both sedimentary and basement rocks over distances of about 20 km are widespread, making it difficult to fit refraction, near-vertical incidence and moderate- to wide-angle reflection data with reasonably simple velocity models. One expanding spread recorded across the Nebine Ridge in southern Queensland shows P-to-SV and SV-to-P converted reflections from a bright spot in the coincident near-vertical incidence reflection data at a depth of about 17 km. Another expanding spread recorded in the Amadeus Basin has provided a detailed velocity model for the 9 km thick sedimentary section. The resolution of deep basement velocities is best below the Surat Basin, although the absence of persistent, continuous reflections and the presence of reflections from outside the vertical plane through sources and receivers limits the reliability of estimates of basement velocities.

Elsevier online abstract

Wide-angle vibroseis data from the western Rhenish Massif

Pages 83-93
Dekorp Research Group, Ernst R. Flueh, Dirk Klaeschen and Rolf Meissner

A deep seismic reflection profile collected by DEKORP and BELCORP in the western Rhenish Massif was supplemented by wide-angle measurements. Signals from a vibrator source were successfully recorded to a distance of 60 km. A passive recording array was operated that recorded all shots along the profile. The wide-angle and near-vertical data were used to construct a velocity model for the profile. Most of the wide-angle reflections coincide with strong near-vertical reflections or bands of high reflectivity. The North Variscan Deformation Front, seen as a prominent shallow reflection on many profiles in this region, separates an upper crust with rather nigh velocities from a layer with lower velocities underneath. At a depth of 20-22 km a thin (2-3 km thick) layer of high velocities is found. The Moho is not reflective either in the near-vertical or in the wide-angle data, suggesting the presence of a thick crust-mantle transition zone.

Elsevier online abstract

Near-vertical vibroseis versus large-offset dynamite seismic reflection profiling: example of the ECORS northern France profile

Pages 95-106
B. Damotte and C. Bois

In the Paris basin (France), the ECORS project has recorded a deep seismic profile using two different sources i.e. vibrators and explosive charges, both using the same in line receiver spread. The explosive shots implemented for refraction surveying have been recorded with offsets up to 45 km to obtain deep reflections.

The results show these two techniques to be complementary. The VibroSeis 96-fold coverage shows a layered lower crust with a high seismic quality for two thirds of the profile. The explosive seismic survey, with a theoretical 3-fold coverage, has been displayed in three single-fold sections. More reflections are visible on these explosive source sections, especially where the quality of the VibroSeis section is poor.

We can explain this difference, in this last part, by a distorted overthrust unit that exists near the surface. When crossing this unit, the seismic rays are distorted and the conventional CMP stack is not optimal for imaging the lower crust.

Another difference between the interpretations made from the two sections is the extension through the lower crust of the Bray fault that is well mapped at the surface. The VibroSeis image suggests that this fault extends throughout the crust, whereas the image of the single explosive section does not show the fault.

Long-offset recording of powerful explosive shots using the same spread of geophones as those for the usual vertical seismic reflection survey can provide invaluable complementary data, especially in tectonically disturbed areas.

Elsevier online abstract

Fault-plane reflection processing and 3D display: the Darling Fault, Western Australia

Pages 107-117
R. A. Young, S. C. Stewart, M. R. Seman and B. J. Evans

A steeply dipping event observed on a 2D profile across the Darling Fault, Western Australia, is identified as a fault-plane reflection based on an analysis of shot records and traveltime modelling. A 3D image of the same event is formed from a lowfold 3D survey. Preliminary interpretation of this unmigrated time surface illustrates the usefulness of 3D graphics display in showing strike, dip and morphology of a fault plane.

Elsevier online abstract

3D isochronal modelling of reflections from the deep crust: application to reflection profiling in central Australia

Pages 119-128
Shunhua Cao, Brian L. N. Kennett and Bruce R. Goleby

Reflection profiling of the whole crust across ancient terranes encounters a range of interpretational problems and ambiguities, especially with regard to the geometry of faults. Such problems can be addressed by seismic modelling with full inclusion of diffractions and a suitable modelling procedure for such three-dimensional problems is the isochronal technique described by Cao and Kennett (1989)

Such modelling has been used to investigate the validity of some of the interpretations of a deep crustal refection profile in central Australia involving planar faults of moderate dip penetrating most of the crust and considerable Moho topography. Even with a relatively short (4 km) field spread it would be possible to detect the reflected energy from faults with dips of about 40°. The character of the reflections also suggests considerable variability in piysical properties within the fault zones. The largest fault, the Redbank Zone, has significant displacement of the Moho. But most of crustal faults appear to sole out into the crust-mantle interface. This gives rise to an undulating Mcho for which seismic modelling corresponds closely to the observed data.

Elsevier online abstract

Design and operation of a deep seismic survey in the Tasman Sea Basin

Pages 129-140
S. P. Kravis, M. F. Coffin and R. Whitworth

Deep crustal reflections, appearing at two-way times of 6-20 s, are characterised by a low signal-to-noise ratio and a low-frequency content (~ 5-30 Hz). Optimising the visibility of these reflections requires that:

1. the source energy be concentrated in this band;
2. the receiver be as long as practically possible;
3. the shot spacing be as snail as possible.

An airgun array simulation program was used to investigate the effect of operating a 10-gun array at different depths. It predicted that the source energy could be enhanced at low frequencies by running the gun array at depths of up to 25 m, which would improve the signal-to-noise ratio for deep reflectors by a factor of 2 relative to the same array at 7.5 m depth at the expense of some loss of temporal resolution.

The receiver length was limited to 4.4 km by the number of cable sections available and the maximum tension allowable in the cable. It was configured as 88 channels of 50 m length. To reduce the shot spacing, minimise the tow noise and reduce tension in the cable, the ship speed was reduced to 4 knots. A cable depth of 20 m was chosen to reduce sea state noise and keep the receiver ghost notch frequency outside the frequency band of interest.

A variety of operational problems was encountered. Fish bites holed 50% of the total cable inventory and cable sections were damaged when they sank below their maximum rated depth after being holed. Shot-generated noise was the limiting factor in deep water. Despite these problems, useful data were obtained over the Lord Howe Rise, Dampier Ridge, Gippsland Basin margin and the deep abyssal plain and the survey demonstrated the use of an airgun array tuned for maximum energy.

Elsevier online abstract

Imaging the lower crust in deep water

Pages 141-143
Robert J. J. Hardy, Michael R. Warner and The Birps Group

(No abstract available)

Elsevier online abstract

The Cambridge digital seismic recorder for land and marine use

Pages 145-154
T. R. E. Owen and P. J. Barton

The Cambridge digital seismic recorder is a simple, cheap and versatile recording system for portable equipment. The recorder stores data in digital form onto high-quality "Walkman" cassette recorders, with a capacity of over 5 megabytes per cassette. Most of the recorder functions are implemented in software, so that the system can be adapted for a wide variety of uses. We describe its use as the recording system with ocean bottom seismometers for controlled-source Seismology and as a triggered recorder for micro-earthquake recording.

Elsevier online abstract

An efficient method for sending a large seismic section from a mainframe computer to a dot matrix printer located at a remote site

Pages 155-161
R. F. Mereu, B. Dunn and J. Brunet

The processing of seismic reflection data sets is usually done on large computers which have both vector processing and large memory capabilities. Many researchers in university, government and industry laboratories have access to these machines via remote terminals and small work stations and therefore can run large jobs conveniently. However, a major problem occurs when one would like to see the results within a few minutes so that appropriate changes may be made. Although seismic centres have professional electrostatic plotters, it may not always be convenient to wait for the plots via courier services. Many university computing centres are equipped with only pen plotters. These are not practical to use for large seismic sections with shaded traces.

An efficient method of sending the data set from the mainframe computer to a small PC computer with a modem and a dot matrix printer is described in this paper. The method works primarily with two FORTRAN programs. The first is resident on the mainframe computer. This program strips the data set of all its non-essential bits, partially rasterizes the set and converts it into a simple ASCII character set sequence. The second program resides on the PC computer. It obtains the compressed data file from the mainframe and completes the rasterizing process by creating the bit map for the printer. Any reasonable size data set can be sent down the line as the method does not require the use of a hard disk on the PC. Screen dumps from a graphics terminal are not carried out. Instead there is a steady flow of data from the mainframe disk, through the PC and then to the printer.

In a test example, a 500-trace data set containing 8.4 megabytes of data on a CYBER 180/835 mainframe was compressed down to 1.2 megabytes and sent down a 9600-baud serial line to an IBM AT PC computer. The PC generated a bit map consisting of 32 million bits. This was plotted by a 180 dot per inch printer on a 6-foot by 14-inch (192.9 × 35.6 cm) plot. The total time required to transmit and plot the data was 25 min. This represents a rate of approximately 3 s for each 2250-point trace.

Elsevier online abstract

Basalts, water, or shear zones in the lower continental crust?

Pages 163-174
Mike Warner

Bright, sub-horizontal, laterally extensive, multi-cyclic layered reflections have now been recorded in the lower continental crust from many areas. In contrast, this reflection signature is rare in the continental upper crystalline crust and upper mantle and in the oceanic lithosphere. With the present state of knowledge it is sensible to seek a single quantitative explanation for all such layering.

Qualitative explanations for lower crustal layering abound and include igneous, metamorphic, structural and fluid related models. Measurements of reflection coefficients from the lower crust place strong quantitative constraints on possible models. Structurally induced anisotropy, metamorphic layering, or free fluids cannot realistically produce sufficiently bright reflections. Igneous layering or the juxtaposition of pre-existing heterogeneities by shear zones can produce strong reflections.

Results from crustal xenoliths, helium-isotope ratios, palaeo-geothermal gradients and chemistry of flood basalts,together with numerical modelling of lithospheric extension and asthenospheric melting, suggest that underplating and intrusion of the lower crust with mantle-derived mafic igneous rocks is common. The apparent spatial relationship between lower crustal layering and continental extension and the strength and continuity of the reflections, suggest that underplating, related to extension and/or elevated mantle temperatures, is a good explanation for lower crustal layering.

Elsevier online abstract

Shear in the lower crust during extension: not so pure and simple

Pages 175-183
T. J. Reston

As lower crustal reflectivity appears to be related to extension, it is sensible to consider the role of the lower crust in the extension of the lithosphere and, in particular, the way the lower crust deforms. The occurrence of transparent zones wrapped around by reflective bands can be interpreted as low-strain regions enclosed within anastomosing, shear zones, suggesting that deformation in the lower crust is by the relative movements of low-strain lozenges along acoustically reflective, high-strain zones. Although this deformation can easily accommodate bulk pure shear, it can also develop during bulk simple shear. A component of subhorizontal simple shear within the lower crust is necessary if deformation in the mantle is localised into shear zones with the geometry and distribution imaged on BIRPS profiles. The lower crust may be the critical link in the lithosphere, transferring deformation between the relatively strong upper crust and upper mantle and being strongly sheared in the process. The form of this shearing is a direct consequence of the way the entire lithosphere extends.

Elsevier online abstract

Probing into extensional sedimentary basins: comparison of recent data and derivation of tentative models

Pages 185-197
B. Pinet and B. Colletta

More and more deep seismic profiling is revealing the complex evolution of the continental crust, which bears the imprints of former plate movements. Previous subduction zones and mountain belts affected the crust before the formation of intra-cratonic sedimentary basins or passive margins. Therefore the former geological setting must always be taken into account in deep reflection studies of such areas. Recent deep seismic profiles in Europe and other parts of the world have demonstrated the largely heterogeneous nature of the continental crust, with dipping reflectors within the crust. While bedding heterogeneities or intrusive sill complexes may be invoked, these reflectors are most often associated with inherited thrust planes related to crustal shortening during previous orogenies. Some of these inherited faults have been imaged beneath sedimentary basins with contrasting styles and their reactivation depends on their strike in relation to the regional stress field. These pre-existing zones of weakness induce a heterogeneity in the upper crust, which may control the geometry of future extensional tectonics. When the strike of the inherited faults is perpendicular to the extensional stress field, they may be reactivated as a detachment surface controlling the sedimentary infill. More often, the obliquity of these structures induces local changes and segmentation of the rif: when they are crossed. In such a case, they correspond to transfer faults, acting as relays between major normal faults, or to "twist zones" where changes occur in the throw of the normal faults, sometimes evolving to an inversion of the polarity in the tilting of the major blocks.

During extension, the top of the lower crust may act as a mechanical decoupling boundary separating asymmetrical brittle failure within the upper crust and symmetrical ductile creep within the lower crust. Extensional modes may explain the great values of crustal thinning observed in both upper and lower crusts beneath rifts and passive margins related to the opening of an oceanic domain. It is more difficult to explain the comparatively small amount of horizontal extension generally observed.

Elsevier online abstract

Comparative investigations of continental reflectivity

Pages 199-206
Dekorp Research Group, Th. Wever, R. Meissner and P. Sadowiak

Based on observations of reflectivity and lengths of reflecting boundaries in various geological units a consistent picture for explaining the different reflectivity patterns is derived. Whereas in the upper, brittle part of old crusts reflectivity concentrates on fault zones of former or present compressional zones it is absent in warm extensional areas, where in addition, granitic plutons may disturb any upper crustal "layering". The lower crust of these young and warm areas with its reflecting lamellae is dominated by strain and shear processes i.e. by an ordering process in a low-viscosity environment.

Any change of sialic or low-grade to more mafic or high-grade layers at mid-crustal levels (Conrad discontinuity) is not related to the beginning of the lamellae but may be responsible for a finger-type appearance of lower crustal reflectivity.

Elsevier online abstract

Measured and calculated elastic wave velocities for xenoliths from the lower crust and upper mantle

Pages 207-210
Ian Jackson, Roberta L. Rudnick, S. Y. O'Reilly and C. Bezant

(No abstract available)

Elsevier online abstract

Seismic imaging of major tectonic features in the crust of Phanerozoic eastern Australia

Pages 211-216
D. M. Finlayson, K. D. Wake-Dyster, J. H. Leven, D. W. Johnstone, C. G. Murray, H. J. Harrington, R. J. Korsch and P. Wellman

Crustal architecture can be recognised from seismic fabric characteristics and velocity information. Across Phanerozoic eastern Australia, three major crustal sub-divisions are identified from data in southern Queensland. They occur under (west to east) the Nebine Ridge, Taroom Trough and New England Fold Belt, with each crustal sub-division having its own internal architecture but separated by major bounding structures (geosutures). These features are assessed against evolutionary models for the region. The significant conclusions are as follows:

1. The Nebine Ridge forms the southeastern part of the Thomson Fold Belt but has a crustal architecture distinctly different from that under the central part of the fold belt, tentatively supporting an evolutionary model involving fragmentation and eastward stretching and thinning of a Precambrian Australian craton. The ridge crust has possibly been considerably altered by Carboniferous plutonism and compressional events.
2. Between the Nebine Ridge and the crust under the Taroom Trough of the Bowen Basin there is a major low-angle (5-10°) geosuture (the Foyleview geosuture), which extends from upper to lower crustal levels along a series of prominent horizons. It was probably active during the major Carboniferous tectonism which severely structured Devonian basins west of the Nebine Ridge. The geosuture forms the southeastern boundary of the Thomson Fold Belt.
3. The crustal architecture under the Taroom Trough has little obvious structure above an outstanding Moho transition zone. However, the prominent Meandarra Gravity Ridge striking along the trough axis highlights the possibility of high-density mafic volcanics within the crust and the probability of a rift model being appropriate early in the trough's history. The sedimentary basin architecture supports the concept of an early extensional history for the crust within a poorly defined, N-S trending, strike-slip regime.
4. Between the Taroom Trough and the New England Fold Belt we interpret a high-angle geosuture extending through the crust (the Burunga-Mooki geosuture). Reactivation during late Palaeozoic-early Mesozoic times produced a series of reverse faults along the uplifted eastern margin of the trough. Some of these structures are tentatively considered to be positive flower structures above deeper faults with strike-slip movement. The geosuture separates regions with distinctly different seismic fabrics and played an important role in the formation of the Bowen-Gunne-dah-Sydney Basin system.
5. Within the New England Fold Belt, crustal domains can be recognized associated with late Palaeozoic oroclinal bending and early Mesozoic trans-tensional basins. Major oroclinal bending of the upper crust in the west of the fold belt is interpreted above a mid-crustal detachment (the Texas detachment). In the central part of the fold belt, the seismic data support a trans-tensional mechanism for the formation of Triassic basins with steeply dipping bounding faults. Under the east of the fold belt, an imbricate thrust stack/accretionary wedge is clearly imaged above a mid-crustal horizon (the Brisbane mid-crustal detachment). Upper crustal deformation appears to have occurred above this detachment.
6. From the Nebine Ridge to the coast, the Moho is clearly defined below a 3-km thick Moho transition zone. Hast of the Thomson Fold Belt, the Moho level is identified as a gently undulating feature at 36-38 km depth and probably re-established after the major late Palaeozoic tectonic events which formed the crust under the Taroom Trough and New England Fold Belt. This contrasts with the middle Palaeozoic lower crustal/Moho features which appear to be preserved under the Thomson Fold Belt.
7. The thickest crust under the Nebine Ridge (about 44 km) appears to be associated with lower crustal wedging from the east. There is an apparent thinning of the non-sedimentary crust under the deepest basin (Taroom Trough) of about 30% compared with the crust to the west and east. No crustal thickening is evident in the region of oroclinal bending within the New England Fold Belt.

Elsevier online abstract

Mid-crustal detachments controlling basin deformation: ramp synforms in southwestern Queensland

Pages 231-234
J. H. Leven, D. M. Finlayson and K. Wake-Dyster

Deep seismic data from the Adavale Basin region of southwestern Queensland suggest that westward Carboniferous movement of the upper crust over mid-crustal ramps was responsible for the present-day geometry of two synforms containing Devonian sediments. We interpret the troughs as hanging-wall synforms formed by sub-vertical simple shear within the upper crust during down-dip movement over the ramp.

One of these structures, the Quilpie Trough in the southern Adavale Basin, is an asymmetric trough interpreted to be an extensional ramp syncline. It overlies a west-dipping ramp in the boundary between the seismically non-reflective upper crust and the seismically reflective lower crust. Formation of this basin by the relative movement of the upper crustal Thomson Fold Belt over the mid-crustal ramp is postulated. The significant asymmetry of the trough suggests that lower crustal extension by pure shear may have occurred during upper crustal translation.

The other structure, the Westgate Trough, contains two Palaeozoic sequences. We propose that the Devonian lower sequence is in a ramp syncline formed by folding of the sediments in a conveyor belt style mechanism during a translation of the upper crust. The upper sequence is interpreted as a ramp basin in which sediments were deposited in the synform as it developed.

Deep seismic reflection data from the central Eromanga Basin region indicate that relative horizontal translation can occur between the upper and lower crust on mid-crustal detachments. The geometry of the synforms suggests that above these detachments, crustal deformation has a sub-vertical simple-shear style, whereas the lower crust below the detachments seems to deform in a pure-shear style. Recognition of relative movement of tens of kilometres on mid-crustal detachments is important for our understanding of basin geometry and development in the central Eromanga Basin.

Elsevier online abstract

Deep seismic profiling in central Australia

Pages 247-256
C. Wright, B. R. Goleby, C. D. N. Collins, R. J. Korsch, T. Barton, S. A. Greenhalgh and S. Sugiharto

Deep seismic profiling undertaken in central Australia comprised expanding spread reflection profiling, long-range refraction work and a small-scale three-dimensional refraction survey, as well as the more usual near-vertical incidence reflection profiling. Scismic reflection sections within the Arunta Block in central Australia show abundant northerly dipping events that are interpreted as reflections from dipping faults, many of which are evident in surface geological mapping. The Redbank Deformed Zone, a major thrust feature, has been imaged to depths of at least 30 km and defines a marked change in reflection character. South of the Redbank Zone, steep, northerly dipping reflections are absent; sub-horizontal basement reflections are prevalent below the Southern Arunta Province and Amadeus Basin. Seismic refraction profiles indicate that the crust is more than 50 km thick below both the boundary between the Southern and Central Arunta Provinces and the northern part of the Amadeus Basin. Expanding reflection spreads and a three-dimensional refraction survey have added fine details to the results in regions of special geological interest and include the resolution of complicated velocity variations in the sedimentary rocks of the northern part of the Amadeus Basin and the measurement of anisotropy in the granulites of the Central Arunta Province adjacent to the Redbank Zone.

Elsevier online abstract

Seismic reflection profiling in the Proterozoic Arunta Block, central Australia: processing for testing models of tectonic evolution

Pages 257-268
B. R. Goleby, B. L. N. Kennett, C. Wright, R. D. Shaw and K. Lambeck

The region extending from the northern Amadeus Basin to the Northern Arunta Province is characterised by a large change in Bouguer gravity (1400µ s-2), abrupt changes in teleseismic travel-time residuals and large thrust structures in which rocks from the lower crust are now exposed at the surface. Deep seismic reflection data, other geophysical data and geological mapping within this region have contributed to the development of a crustal model that would result from "thick-skinned" tectonic processes, in which deformation occurred along moderately dipping thrusts or shear zones that extended from the surface down to at least the crust-mantle boundary. Two important features of this model have emerged with the help of unconventional stacking techniques. The first is a series of reflections from the Redbank Deformed Zone that can be traced from surface outcrop to depths of more than 30 km. The second is evidence of a fault extending through the crust and displacing the Moho.

Elsevier online abstract

Australia's southern margin: a product of oblique extension

Pages 269-281
J. B. Willcox and H. M. J. Stagg

Recently developed detachment models of continental margin formation interpret the southern margin of Australia to have formed when the lower-plate Australian margin was pulled out from beneath the upper-plate Antarctic margin. Data now available and summarised in this paper, point very strongly to a generally NW-SE direction of initial continental extension for the southern margin, in contrast to the widely held picture of simple N-S rifting. The evidence for this extension direction comes from the analysis of deep seismic data acquired by the Bureau of Mineral Resources in 1986 in the central Great Australian Bight (GAB), the gravity field of the GAB, seismic and magnetic basement structures in the Eyre Sub-basin, Polda Trough, Ceduna Depocentre and Duntroon Basin and from the analysis of the magnetic seafloor spreading anomalies produced during the slow first phase of drifting between Australia and Antarctica.

Further, it is now believed that the formation of the southern margin of Australia can be described in terms of three phases of continental extension (El to E3) and two phases of drifting (D1 and D2). In summary, these phases were as follows.

1. E1: approximately 300 km of Late Jurassic (?or older) to Early Cretaceous NW-SE-oriented extension in the GAB, with strike-slip motion in the nascent Otway Basin and along the Tasmanian margin.
2. E2: 120 km of Early Cretaceous NNE-SSW-oriented extension which formed the basins of southeastern Australia (Otway, Bass, Gippsland) and which probably produced a structural overprinting in the GAB Basin.
3. E3/D1: minor continental extension and the first 500 km of slow drift between Australia and Antarctica on an azimuth of 165° wrenching on the Tasmanian margin.
4. D2: 2600 km of fast drifting between Australia and Antarctica on a N-S azimuth.

Elsevier online abstract

Crystal seismic observations across the convergent plate boundary, North Island, New Zealand

Pages 283-296
F. J. Davey and T. A. Stern

Seismic reflection profiles recorded across southern North Island, New Zealand, define the structural details of a convergent plate boundary. The eastern profile delineates the thrust wedges and back-tilted basins of the accretionary prism which overlies a detachment zone marking the top of the subducted Pacific plate. Most of the younger part of this wedge is isolated from the landward part of the sequence by an oceanward dipping detachment zone. The western profile defines a broad crustal downwarp in the "back-arc" region of the plate boundary overlying the 20-50 km deep subduction zone. The crustal section farther to the west comprises a generally transparent middle crustal layer and a reflective lower crust. Dipping reflective zones in the middle crust may delineate low-angle normal faulting associated with a Cretaceous extensional episode. Detailed Seismological data demonstrate the close association of the top of a zone of high seismicity, inferred to mark the subducted Pacific plate, with the base of the reflective sequence identified as the base of the overriding Australian plate.

Elsevier online abstract

Deep seismic probing of continental crust in the lower Yangtze region, eastern China

Pages 297-305
Weng Shijie, Chen Hushen, Zhou Xueqing and Cui Zhichen

In China, seismic probing of continental crust started several years ago, associated with the recording of regional seismic lines for petroleum exploration. During a survey for Palaeozoic oil prospects in the Lower Yangtze region, we extended the seismic record length to 16 seconds to obtain deep crustal data. Several Cenozoic basins with different subsidence histories have been transected by this deep seismic profile. Beneath these basins, the Palaeozoic basement has been deformed by a system of thrust faults which splay from a common detachment surface in a style similar to foreland basins.

A restite reflection signature below the Suzhou granite batholith suggests it was formed by crustal remelting. The structural history of the Tangchen-Lujiang fault involved thrust and subsequent strike-slip movement. The Yangtze Block was underthrust beneath the North China or Dabei Block. The structure of the Moho suggests it reformed recently and was influenced by the Cenozoic movements.

Elsevier online abstract

Deep seismic sounding in the Godavari Graben and Godavari (coastal) Basin, India

Pages 307-317
K. L. Kaila, P. R. K. Murty, V. K. Rao and N. Venkateswarlu

Deep seismic sounding investigations along two profiles, one along and the other across the Godavari Graben delineate the northeastern extension of the Bapatla Ridge which separates the Godavari Graben from the Godavari (coastal) Basin. The Godavari Graben is filled with a maximum of 2.8 km thick Lower Gondwana (Upper Permian to Lower Triassic) sediments. It is inferred on the basis of P-wave velocities that the Upper Precambrian Pakhal Formation does not continue south of Paloncha. The Upper Gondwana (Upper Jurassic to Lower Cretaceous rocks with marine incursions appear to continue towards the southeast for about 15 km from the exposures, below the Rajahmundry sandstone (Mio-Pliocene).

Two-dimensional velocity modelling of the upper crust indicates a major interface at 3.5 km depth in the upper crust where the velocity increases from 5.4-5.5 km s-1 to 6.2-6.4 km s-1. This interface shows a domal upwarp across the Godavari Graben which may be an indication of NE-SW crustal extension in this region. The deeper part of the crust consists of a large number of almost horizontal small reflector segments. Two-dimensional crustal velocity modelling for some correlatable wide-angle reflections yielded a three-layered structure below 10 km depth, with the Moho at about 41 to 43 km.

Elsevier online abstract

The deep structure of the east Oman continental margin: preliminary results and interpretation

Pages 319-331
P. J. Barton, T. R. E. Owen and R. S. White

We describe preliminary results of a coincident normal incidence and wide-angle seismic experiment across the east Oman continental margin just north of the Masirah Island ophiolite. This margin is affected by tectonic deformation and may be the site of margin-parallel shear in a zone parallel to the Owen Fracture Zone. The reflection profile is used to define upper crustal structure and shows a deep offshore basin dammed oceanwards by a ridge, interpreted here as a rotated fault block. Wide-angle data were collected using ten digital ocean-bottom Seismometers and 110 explosive shots. Preliminary raytracing of a crustal model based on the gravity model of Whitmarsh (1979) shows rapid changes in crustal thickness across the margin. A steep landward dipping reflector, probably the Moho, lies beneath the continental slope. The orientation of this reflector is exactly opposite to the direction of tectonic fabric predicted by a simple overthrust model of ophiolite emplacement from the ocean basin to the east.

Elsevier online abstract

Deep seismic profiling in the Nosop Basin, Botswana: cratons, mobile belts and sedimentary basins

Pages 333-343
James A. Wright and Jeremy Hall

Deep seismic reflection data from the Kalahari of western Botswana has been interpreted in terms of the tectonics of southern Africa. The 600 km of 12 fold data exhibit good quality reflections in the range 5-15 s TWT. The data outline the Nosop Basin, a depositional centre with greater than 10 km of Palaeozoic and younger sediments. Five prominent reflecting horizons are traced in the basement underlying the sediments. On the basis of the seismic data, the Nosop is now interpreted as a single basin. The previous multi-basin interpretation of the aeromagnetic data -esults from features within the magnetic basement. The uppermost two basement units are interpreted as Proterozoic sediments that rest on Archean crust. The seismic data show that the crust appears to thicken to the southeast as the Kaapvaal Craton is approached. At the same time, the sedimentary cover progressively thins suggesting that the western edge of the Kaapvaal is an extended margin. Interpreted deep crustal shear zones indicate a major terrane boundary separating the Proterozoic from the Archean. Depths to basement inferred from aeromagnetic data are consistent with the depth to the seismic boundary associated with the top of the Archean. Some of the aeromagnetic expression may be due to Karoo volcanic intrusives in the basement. A thrust belt in the northwest of the area now buried in the mid crust is correlated with the nearby Ghanzi-Chobe belt and allows the mid-crustal layering to be dated as late Proterozoic.

Elsevier online abstract

Interpretation of the reflections in the Siljan Ring area based on results from the Gravberg-1 borehole

Pages 345-360
C. Juhlin

During 1985 about 80 km of surface multichannel seismic reflection data were collected across the meteorite impact in the Siljan Ring area in central Sweden. The area consists mainly of granitic and gneissic rock ranging in age from 1400 to 2000 million years with remnants of Palaeozoic sedimentary rocks preserved by downfaulting after a meteorite impact at 360 Ma, thereby forming a circular ring about 40 km in diameter. Dolerite intrusions ranging in age from 850 to 1700 million years are also present. The seismic data revealed several high-amplitude, laterally continuous, sub-horizontal reflections in the northern part of the structure. The high-amplitude reflections and a possible intermediate low-velocity zone were contributing factors in choosing the site for the Gravberg-1 Deep Earth Gas test well. Drilling and vertical seismic profiling (VSP) found that the reflectors were associated with dolerite sills which had intruded into the granite and which range in thickness from a few metres up to 60 m and with a pre-impact area extent of at least 800 km2. Studies of amplitude and frequency versus offset (AFVO) show the observations are compatible with a model of simple granite/dolerite/granite layering.

Elsevier online abstract

Reflectivity patterns in the Variscan mountain belts and adjacent areas: an attempt for a pattern recognition and correlation to tectonic units

Pages 361-372
Dekorp Research Group, R. Meissner, Th. Wever and P. Sadowiak

The seismic reflection profiles of DEKORP (DEutsches KOntinentales ReflexionsSeismisches Programm) in the Federal Republic of Germany to date have been limited to areas of the Variscan orogeny. Nevertheless, the character of their reflections differs considerably and may be correlated to certain Variscan and post-Variscan developments. Lower crust lamellae develop in areas of high heat flow, mostly associated with post-Variscan extensional processes; "crocodile" and nappe tectonics are best preserved in the cores and at the flanks of older massifs which were incorporated into the Variscan orogeny. So far poor reflectivity has been observed only in the area of the London-Brabant Massif which was not involved in any of the Phanerozoic orogenies.

Elsevier online abstract

The identification of mantle reflections below Hungary from deep seismic profiling

Pages 379-385
K. Posgay, E. Hegedus and Z. Timar

In the 1970's the "Eötvös Lorand" Geophysical Institute performed seismic reflection deep sounding experiments with multiple coverage in several parts of the Pannonian Basin (Posgay. 1975). One of the most interesting results was that arrivals were received in the time range corresponding to the depth of the asthenosphere. They were interpreted as reflections (Posgay et al., 1981). Conspicuous arrivals from the time range of the asthenosphere (15-20 s) were also received along the Pannonian Geotraverse in 1987.

In the majority of sections published in other countries, the upper mantle seems to be transparent. The present paper discusses investigations and preliminary results, which lead to the conclusion that this apparent transparency is due to the frequency range below 10 Hz being neglected in both signal generation and recording in most deep continental reflection surveys.

The seismogeological conditions of the Pannonian Basin are certainly favourable for reflection studies of the upper mantle. The correctness of the assumption that arrivals from the time range of the asthenosphere are primary reflections is subjected to further investigation and conclusions are made concerning the structure of the asthenosphere.

Elsevier online abstract

Progress in BIRPS deep seismic reflection profiling around the British Isles

Pages 387-396
D. H. Matthews, The Birps Group (C.A. Flack, R.W. Hobbs, S.L. Klemperer, and D. B. Snyder, M.R. Warner, N.J. White)

The results of 4000 km of new BIRPS profiles are briefly outlined as an introduction to other BIRPS papers in this volume. Highlights of these profiles include the tracing for 900 km of a reflection pattern in the lower crust associated with the lapetus suture and the observation of more than 20 separate reflections from the mantle, while progress in solving the problem of deep-water multiples has enabled the joining of the continental and oceanic Moho reflections off SW England. Developments in old interpretational problems are reviewed and the new problem of dating reflections is mentioned.

Elsevier online abstract

Major geodynamic processes studied from the ECORS deep seismic profiles in France and adjacent areas

Pages 397-410
C. Bois and Ecors Scientific Party

For more than five years, the French ECORS project has obtained deep seismic profiles across major structural features that are related to the Variscan, Alpine and Pyrenean belts as well as the Permo-Mesozoic subsidence in the Celtic Sea, the English Channel and the Bay of Biscay. Profiles across the orogenic belts show striking similarities such as the flexure of the down-going plate and the geometry of the frontal detachment and related thrusts and nappes whatever the age of the belt. However, straight dipping reflections cutting through large sections of the crust seem to be common in the Variscan belt and are interpreted as late features in its tectonic history. The Tertiary Alpine and Pyrenean belts show extensive crustal thickening and crust-mantle imbrication, while the layered lower crust and its bottom are flat beneath the late Palaeozoic Variscan belt. This is interpreted as a late equilibration of the crust-mantle boundary, which involved deep metamorphism in the crust and occurred between the Permian and the middle Jurassic. The Permo-Jurassic basins reveal an unattenuated lower crust, while both the upper and lower crusts are thinned beneath the Cretaceous-Tertiary ones. Observation does not support the current models of basin formation and nowhere does extension appear to have played a major role in subsidence. The lower crust of the Permo-Jurassic basins may have been reworked and equilibrated after their rifting.

Elsevier online abstract

Main results of the ECORS Pyrenees profile

Pages 411-418
P. Choukroune, F. Roure, B. Pinet and Ecors Pyrenees Team

The 250 km long, deep seismic reflection survey across the Pyrenees from one foreland (Ebro Basin) to the other (Aquitaine Basin) was designed to study the structure of the deformed Mesozoic boundary between the Iberian and European Plates. The results are:

1. the entire crust shows well-defined reflectors with a general fan-shape geometry;
2. a deep layered domain is evident above the Moho discontinuity;
3. near the surface in the external domains, reflectors accurately define the geometry of major thrusts and structures affecting the Mesozoic and Cenozoic cover of the Pyrenees;
4. the middle crust shows two major sets of seismic events interpreted as Hercynian structures.

Using these results, a balanced structural section of the Pyrenean belt is proposed. The mode of thickening of the crust during the Alpine orogeny is briefly discussed; only the Iberian crust was significantly thickened by a southward stacking of slabs.

Elsevier online abstract

Lateral variations in crystal reflectivity beneath the Paris Basin

Pages 425-434
Jean-Michel Marthelot and Maksim Bano

The application of a new method of automatic extraction of reflections from a stacked seismic section to the ECORS North of France profile highlights the lateral variations in crustal reflectivity observed beneath the Paris Basin. Five characteristic zones can be distinguished along the 228 km long profile. A complex pattern of spatially interfering dipping events occurs in the lower crust of the transition region between the zones of reflective and transparent lower crust. Modelling reveals that these events are diffractions from discontinuous subhorizontal scatterers concentrated near 8 s. Hyperbolic events at Moho level (12 s) may indicate that the 1 s increase in travel time to the Moho observed across the transition region is accommodated by a discontinuous step. There is no spatial correlation between the lateral variations in reflectivity observed within the upper and the lower crust. Neither is there evidence that any structure within the upper crust continues into the lower crust. We interpret these two observations as an indication of a younger origin of the lower crustal reflectivity and discuss the possibility that the early extensional phase of evolution of the Paris Basin has contributed to the lateral variations of reflectivity observed within the lower crust.

Elsevier online abstract

Deep structure of the Celtic Sea: a discussion on the formation of basins

Pages 435-444
Jerome Dyment, Jean-Claude Sibuet and Bertrand Pinet

The SWAT (South West Approaches Traverse) deep seismic reflection profiles provide complete sections of the crust in the Celtic Sea area. Deep Mesozoic-Cenozoic basins are underlain by a non-reflective upper crust cut by Variscan thrusts and a reflective lower crust.

One of these thrusts, the Variscan Front, has been considered to be a detachment fault, which could have initiated the development of the North Celtic Sea Basin during Mesozoic times. This hypothesis is inconsistent with the obliquity of the front with respect to the basin, the migration of the basin depocentres and the existence of normal faults that controlled the subsidence of the basin and intersect the front.

In the absence of refraction data, we used two-dimensional gravimetric modelling to determine the depth of the Moho. This modelling shows that the gravimetric Moho is coincident (±3 km) with the bottom of the reflective lower crust.

The resulting depth sections show a nearly flat Moho and a lower crust with a relatively constant thickness. The thinning coefficients of the whole continental crust and the stretching coefficients, obtained by comparison of theoretical and computed (backstripping method) subsidence curves, are significantly different. We discuss three hypotheses to explain this discrepancy:

1. syn-rifting processes not taken into account in the stretching model;
2. previously thickened (Variscan) crust, departing from the initial state assumed by the model;
3. post-extensional modification of the crustal configuration resulting from rifting.

Each of these hypotheses or a combination of them may correctly account for the data. However, the flat nature of the Moho suggests that at least the third one occurred.

Elsevier online abstract

Dating the source of lower crystal reflectivity using BIRPS deep seismic profiles across the lapetus suture

Pages 445-454
Simon L. Klemperer, Richard W. Hobbs and Brett Freeman

Six BIRPS deep seismic reflection profiles around the British Isles that cross the trend of the lapetus suture allow a geophysical definition of the suture zone for 900 km along its strike and demonstrate that specific reflective structures have survived since the mid-Palaeozoic Caledonian orogen. In an area northeast of England, spanning 70 km along Caledonian strike and 200 km across strike, three profiles show very considerable similarities in the reflectivity patterns of the lower crust. When these similarities are mapped, the boundaries separating regions of lower crust with the same reflective character run parallel to Caledonian strike. This strongly suggests that the distribution of reflectivity in the lower crust in this area is directly controlled by Caledonian structures. Geologic models for the area imply that many of the reflectors themselves may also be of Caledonian age. However, though an age for the imposition of patterns of reflectivity can be inferred, the age of an individual reflector cannot be determined unless it can be traced to outcrop or unless there is independent evidence as to the physical nature of the reflector.

Elsevier online abstract

Geological inheritance and crystal dynamics of the northwest Scottish continental shelf

Pages 455-467
M. Stein and D. J. Blundell

Deep seismic reflection profiles have been combined with conventional seismic surveys across the northwest Scottish continental shelf to interpret basin development in the region. Post-Caledonian basins are built upon partially reactivated older structures which have strongly influenced their internal structure and stratigraphy. We show that the Minch Basin, along with several others, developed in the hangingwall of the partially reactivated Outer Isles Fault, a planar fault dipping at about 30° ESE that reaches nearly to the Moho. The internal structure of these hangingwall basins is compared with appropriate scaled analogue models to show that they have developed by passive hangingwall collapse above a footwall block that moved towards the WNW. This geometry is incorporated into a lithospheric dynamic model with northwest extension towards the Rockall Trough from a fixed lithosphere beneath the Scottish Highlands. This model requires non-linear lithosphere extension which is accommodated by movements on faults and block rotation in the upper crust, extension and thinning of the lower crust along a network of anastomosing ductile shears, extensional movement of the Flannan Fault in the upper mantle and horizontal displacement along the Moho beneath the Outer Hebrides. This dynamic model has wider application for other extensional regions.

Elsevier online abstract

Three-dimensional mapping of seismic reflections from the crust and upper mantle, northwest of Scotland

Pages 469-481
Catherine Flack and Mike Warner

A survey area 190 km by 110 km to the northwest of Scotland on the UK continental shelf, has been the site of intense exploration using the deep seismic reflection technique. A closely spaced grid of deep reflection data totalling 1600 line km has been acquired by the BIRPS group between 1981 and 1987. The majority of the data were recorded to 15 s two-way travel time (50 km depth); profiles with recording times up to 60 s (230 km depth) were also acquired. The survey includes a two-ship synthetic aperture profile, the SLAVE line, which synthesizes a 0-16 km offset CDP profile. The main features revealed and consistently imaged across the survey area by this combined dataset are: an upper crust containing half-graben style basins, bounded to the west by planar faults; a reflective lower crust; and dipping and sub-horizontal, high-amplitude reflections within the mantle. Three-dimensional mapping of the dipping reflectors shows a complex package of reflections which updomes into the lower crust in the central area of the survey, disrupting the Moho and continuing into the uppermost mantle, where it thins to a couplet of reflections. In the north and south of the survey area, this structure is observed only in the mantle, to a depth of 80 km. This dataset is unique in deep seismic reflection profiling, in terms of the quality of the image returned, the great depth to which reflections are imaged and the intensity of surveying of the lower crust and upper mantle. The continuity and strength of the reflectors seen on these records show that even at the great depths at which these structures exist the seismic reflection technique is able to image them as high-amplitude, coherent and highly continuous reflections. The reflections are interpreted as shear zones or faults within the lower crust and upper mantle. This indicates that in this part of the world at least, both the lower crust and upper mantle are able to sustain discrete zones of deformation.

Elsevier online abstract

Shear-wave evidence for an anisotropic lower crust beneath the Black Forest, southwest Germany

Pages 483-493
E. Lüschen, B. Nolte and K. Fuchs

Controlled shear-wave sources (a horizontal vibrator and dynamite) were used in near-vertical reflection studies to explore the nature of the reflective laminated lower crust in southwest Germany. The data reveal strong S-wave reflections from the lower crust. These lower crustal S-reflections were not observed by previous wide-angle surveys in the same region. A laminated crustal model with isotropic layers (and changing Poisson's ratio) cannot explain the differing near- and far-offset S-response. By introducing alternating anisotropic and isotropic lamellae, this discrepancy may be resolved. This seismic model is consistent with a petrological model of deformed amphibolites containing 10-30% of preferentially orientated hornblende. Observed amplitude differences in P- and S-wave reflections from the same depth in the lower crust suggest regions with a changing Poisson's ratio.

Elsevier online abstract

Lower crustal petrology from wide-angle P- and S-wave measurements in the Black Forest

Pages 495-505
K. -J. Sandmeier and F. Wenzel

The lower crust of the Black Forest in southwest Germany is strongly reflective for compressional waves but almost non-reflective for shear waves. If the reflectivity for P-waves is caused by alternating high- and low-velocity rocks, the S-wave observations require that the low-velocity rocks have a low Poisson's ratio and vice versa. A purely compositional, petrological model for lower crustal rocks can fit the observed P- and S-wave velocity data, provided a rather high quartz content (50%) is acceptable for the lower crust.

Elsevier online abstract

Results of deep-seismic reflection investigations in the Rhenish Massif

Pages 507-515
Dekorp Research Group

Within the "Deutsches Kontinentales ReflexionsSeismisches Programm" (= DEKORP) two deep-seismic reflection lines in the Rhenish Massif were surveyed in 1986 and 1987. The Rhenish Massif is part of the Mid-European Variscides, an orogenic belt which developed in the Upper Devonian-Carboniferous. Devonian rocks crop out at surface with only few exceptions. The Rhenish Massif is bordered by two sedimentary troughs: the Subvariscan Foredeep in the north and the Hessian depression in the south. In east-west direction it is subdivided by the River Rhine and another depression, the so-called Eifel North-South-Zone.

Both lines consist of two parts, respectively and cross the northern border of the Rhenish Massif. DEKORP 1A/B is located west of the River Rhine and was extended through the BELCORP group by 35 km in Belgium. The total length of the line is 145 km. DEKORP 2-North/2-North-Q is located east of the River Rhine and has a length of 220 km/60 km.

The aim of these surveys was to investigate the crustal structure of the Rhenish Massif and its relation to the adjacent areas in the north and south. The results indicate the presence of NW-vergent tectonics the effects of which can be traced down to the deep parts of the crust. Horizontal compression must play a dominant role. Differences between the western and eastern parts of the Rhenish Massif, the "Faille du Midi" and a prominent fault at its northern border are evident.

Elsevier online abstract

A geophysical study of the Hatton Bank volcanic margin: a summary of the results from a combined seismic, gravity and magnetic experiment

Pages 517-526
J. V. Morgan and P. J. Barton

This paper summarizes the results of a geophysical study of the Hatton Bank volcanic margin in the North Atlantic. Normal incidence and synthetic aperture profiles (SAPs) show seaward-dipping reflectors, interpreted as subaerial basalt flows, on the continental slope and in the adjacent ocean basin. Velocity-depth profiles, derived from expanding spread profiles (ESPs) shot along strike, show elevated velocities in the lower crust on the continental slope; the high velocities are interpreted as evidence for igneous underplating or intrusion. Gravity modelling shows high densities in the lower crust, supporting a model of igneous accretion. An ocean bottom Seismometer (OBS) and multichannel refraction line, shot perpendicularly to strike through the midpoints of the ESPs, further constrained the velocity structure across the margin. The velocity model was refined as each dataset was processed, clearly showing that multiple datasets were required in order to constrain the velocity structure tightly. In particular, ESPs were found to produce misleading results in places where the assumption that the sub-surface structure is one-dimensional, inherent in the ESP technique, was not valid. The final velocity model is used to infer the thickness of igneous material accreted across the Hatton margin, which is estimated to increase westwards from zero at the top of the bank to 18 km at the bottom of the continental slope and then thin rapidly to 11 km in the adjacent Iceland Basin. As the oceanic crust thins its seismic velocity decreases, suggesting a decrease in percentage partial melting. These observations strongly support a model in which volcanic margins are produced by elevated asthenospheric temperatures.

Elsevier online abstract

Crustal anatomy of a transform continental margin

Pages 527-529
C. E. Keen, W. A. Kay and W. R. Roest

This paper describes new deep seismic reflection results across the major transform margin southwest of the Grand Banks, off eastern Canada. An interpretation of these results is presented, supplemented with previous seismic refraction, gravity and magnetic measurements.

The results show a fairly sharp transition between oceanic and continental regions at the transform margin. A narrow zone, about 40 km wide, separates the two regions and is interpreted to be a zone in which shearing has destroyed the original fabric of the continental crust. Faults in this region extend throughout the crust and there is some evidence that a major strike-slip fault may extend some tens of kilometres into the upper mantle. Magmatism may have further modified the crust in this shear zone, which is bounded on the oceanic side by a small seamount. The continental crust is thinned by a factor of 3 or more in this zone, which may be due to erosion or to flow in the lower crust during shearing. The oceanic crust does not appear to thin significantly as the transform margin is approached. Crustal-scale dipping reflectors over a confined region below the shelf are interpreted to represent the contact between two Appalachian terranes: the Meguma and Avalon terranes. This supports earlier interpretations of a large magnetic lineament in the region as marking this boundary.

These crustal seismic data crossing a large transform margin are the first of their kind and will help us to constrain evolutionary models of transform margins. The picture they provide is different in many respects from that across rifted margins and similar in others. In some cases it might be difficult to distinguish between these two classes of Atlantic-type margins on the basis of crustal structure.

Elsevier online abstract

Deep seismic and geochemical constraints on the nature of rift-induced magmatism during breakup of the North Atlantic

Pages 545-547
Carolyn M. Zehnder, John C. Mutter and Peter Buhl

The deep crustal structure of the Norwegian and East Greenland conjugate passive margins has been investigated with two-ship multichannel reflection and refraction techniques in the region from the Iceland-Færøe Ridge to the Greenland Fracture Zone. The data demonstrate the presence of thick (15-20 km) igneous crust that was emplaced along the conjugate margins during the initiation of seafloor spreading. Seaward dipping units comprise the uppermost 3-6 km of the thick crust. The thickness of igneous crust diminishes seaward, with oceanic layer 2 approaching normal thickness nearer to the margin than does oceanic layer 3. Wide-angle reflection techniques, used in conjunction with expanded spread profiling, reveal deep Moho reflections beneath 20 km thick oceanic crust that continue seaward into the ocean basin.

Margins that have experienced unusually prolific magmatic activity during breakup have been termed "volcanic" to distinguish them from "non-volcanic" margins that have evolved into accreting plate boundaries without excessive volcanism. The origin of volcanic margins cannot be determined from seismically derived crustal thicknesses alone. However, thermal and chemical properties of the mantle source may be revealed in the compositions of magmatic products emplaced within volcanic passive margins. A combined investigation of crustal structure and associated basalt composition could better resolve processes that result in volcanic passive margins.

On the Vøring Plateau of the Norwegian margin, seismic lines coincide with DSDP Leg 38 drill sites. Whole rock major element abundances in the freshest basalts are compatible with moderate extents of partial melting of mantle much like that believed to underlie modern mid-ocean ridge systems and would be expected to result in the emplacement of oceanic crust about 9 (±2) km thick. This is about the thickness of oceanic crust located seaward of the dipping basaltic units off the Hatton, Møre and Vøring Plateau margins and 10 km south of Iceland on the Reykjanes Ridge. These results, together with regional heat flow, depth, gravity and geoid anomalies, suggest that the mantle beneath this region has experienced a moderate, long-lived temperature increase compared to mantle beneath normal ocean basins.

The much greater thicknesses of igneous crust adjacent to the margins do not appear to reflect an unusually hot flux associated with their construction, since this would lead to much greater extents of partial melting than are inferred from the basalt geochemistry. Evidently, the difference between volcanic and non-volcanic margins is not simply proximity to a hot spot. Instead, we suggest that the magmatism is closely related to the continental rifting, phase. Tectonism resulting in an abrupt transition from continental to oceanic lithosphere may induce small-scale convection that results in the circulation of a greater amount of mantle material into the pressure-temperature field where partial melting begins than would occur through passive upwelling alone. Initial crustal thicknesses are enhanced as a result of the convective partial melting. Crustal thickness is expected to diminish as this secondary convection slows, but the degree of partial melting of the source should not change dramatically.

Elsevier online abstract

Fault reactivation during Mesozoic extension in eastern offshore Canada

Pages 567-580
Beatrice de Voogd, Charlotte E. Keen and William A. Kay

New seismic data are presented that document reactivation of a thrust fault as a normal fault. Four deep seismic reflection profiles now straddle the Hibernia oil field region of the Grand Banks off eastern Canada. The profiles intersect each other near the western edge of the 22 km deep Jeanne d'Arc Basin, thereby giving a three-dimensional picture of basement structures. The basin-bounding fault which marks the eastern edge of the Bonavista platform appears to follow a pre-existing fault system. More precisely, the fault that bounds the Jeanne d'Arc Basin just north of Hibernia can be traced on the seismic data along strike to the south, beneath the Bonavista platform, where it has not been reactivated. The seismic data show that this feature was probably a Palaeozoic thrust fault. A laterally extensive Palaeozoic(?) basin occupies most of the Bonavista platform. The deeper part of the crust is quite reflective and is characterized by numerous, dipping reflective packages.

This detailed study of the Hibernia region suggests that the narrow elongated basins of the Grand Banks may have been formed within the reactivated hanging walls of older thrust sheets. Similar observations have been made for the Mesozoic extensional basins of the U.K. continental shelf. The pre-Mesozoic compressional structures documemed on the central Grand Banks lie along the Appalachian Orogen and may be of Variscan age.

Elsevier online abstract

Dipping shear zones and the base of the crust in the Appalachians, offshore Canada

Pages 581-593
Jeremy Hall, Garry Quinlan, François Marillier and Charlotte Keen

The Seismological character of the Appalachian mobile belt offshore from eastern Canada includes an unusual concentration of northwest-dipping reflectors which appear to dip through sub-horizontal reflectors at the base of the crust. The geometry of these structures has been characterised by ray-trace migration. They are not point diffractors. They appear to dip at between 25 and 45°, but flatten upwards towards a mid-crustal "bright" band and disappear downwards into the mantle. We interpret the reflectors as shear zones which ramp through the lower crust from a mid-crustal detachment to a diffuse and less reflective set of shears in the mantle. It is suggested that the reflectors were caused by the collision of continental blocks at the final closure of the Iapetus Ocean, but may have been subject to later reactivation, especially during late Carboniferous strike-slip.

Elsevier online abstract

GLIMPCE seismic reflection evidence of deep-crustal and upper-mantle intrusions and magmatic underplating associated with the Midcontinent Rift system of North America

Pages 595-599
J. C. Behrendt, D. R. Hutchinson, M. Lee, C. R. Thornber, A. Tréhu, W. Cannon and A. Green

Deep-crustal and Moho reflections, recorded on vertical incidence and wide angle ocean bottom Seismometer (OBS) data in the 1986 GLIMPCE (Great Lakes International Multidisciplinary Program on Crustal Evolution) experiment, provide evidence for magmatic underplating and intrusions within the lower crust and upper mantle contemporaneous with crustal extension in the Midcontinent Rift system at 1100 Ma. The rift fill consists of 20-30 km (7-10 s) of basalt flows, secondary syn-rift volcaniclastic and post-basalt sedimentary rock. Moho reflections recorded in Lake Superior over the Midcontinent Rift system have times from 14-18 s (about 46 km to as great as 58 km) in contrast to times of about 11-13 s (about 36-42 km crustal thickness) beneath the surrounding Great Lakes. The seismically complex deep-crust to mantle transition zone (30-60 km) in north-central Lake Superior, which is 100 km wider than the rift half-graben, reflects the complicated products of tectonic and magmatic interaction of lower-crustal and mantle components during evolution or shutdown of the aborted Midcontinent Rift. In effect, mantle was changed into crust by lowering seismic velocity (through intrusion of lower density magmatic rocks) and increasing Moho (about 8.1 km s-1 depth.

Elsevier online abstract

Pg shingles: preliminary results from the onshore GLIMPCE refraction experiment

Pages 617-626
R. F. Mereu, D. Epili and A. G. Green

In 1986, a series of near vertical seismic reflection lines and coincident seismic refraction lines were recorded across the Great Lakes in the GLIMPCE experiment. From a refraction point of view this experiment was unique as this was the first time in North America that long range refraction lines up to 350 km were recorded with a trace spacing of less than 100 metres. The source of energy was a large air gun. Some preliminary results of the nature of the onshore recordings made by the University of Western Ontario are presented in this paper.

Observations of many of the in-line profiles show that the character of the Pg crustal phase is not simple but has a "shingle-like" pattern which may be caused by wide-angle reflection effects from numerous long and short reflectors within the crust. Theoretical studies indicate that the amplitudes of the wide-angle reflected signals from thin high-velocity layers are in general much larger than those from thin low-velocity layers. Supercritical reflected waves are not possible from the top of a low-velocity layer. The closely spaced traces show in minute detail the manner in which one travel-time branch fades and the next one arrives. The results seem to indicate that the Pg phase in our data set has little direct wave component but, instead, is made up of a whole series of "shingled or fish-scaled" wide-angle reflection components, each with successively larger apparent velocities. These results are in agreement with the findings of the near-vertical reflection component of the experiment.

Elsevier online abstract

Origin of deep crystal reflections: seismic profiling across high-grade metamorphic terranes in Canada

Pages 627-638
Green, B. Milkereit, J. Percival, A. Davidson, R. Parrish, F. Cook, W. Geis, W. Cannon, D. Hutchinson, G. West and R. Clowes

In an attempt to better understand the origin of deep crustal reflections LITHOPROBE has sponsored or co-sponsored seismic reflection surveys across tracts of high-grade metamorphic rock in the Archean Superior craton, the Proterozoic Grenville orogen and the Phanerozoic Cordilleran orogen. Common to these three diverse terranes are near-surface zones of prominent seismic reflectivity that are typically associated with velocity discontinuities at highly strained contacts between gneissic rocks of varying lithology. At some locations the reflective layering resulted from transposition and rearrangement of previously layered rocks (stratified assemblages, sills, etc.), whereas in other regions it was generated by extreme attenuation, stretching and ductile flow of weakly layered or irregularly organized rocks. It seems likely that compositionally layered gneissic rock is a common source of reflections in the deep crust, with reflections originating at lithological boundaries and zones of mylonite.

Elsevier online abstract

LITHOPROBE: a scientific update-new images of the continental crust

Page 639
Ron M. Clowes

LITHOPROBE, Canada's national earth sciences research program which integrates geophysical, geological and geochemical investigations in a collaborative effort among scientists from universities, government and industry to extend and relate surface geology to structures at depth, is up and running. The Lithoprobe Secretariat is established at the University of British Columbia. The Lithoprobe Seismic Processing Facility (LSPF) for research reprocessing of Lithoprobe and related data is installed at the University of Calgary. Access by the Canadian research community will be available through remote site connections. All scientific aspects of the program are in operation and plans for the future are in hand. Some of the spectacular new images of continental crustal structures that have been obtained and their tectonic significance are highlighted in this summary. Individual contributions of specific studies by Lithoprobe scientists are included in this volume.

Along the eastern portion of the Southern Cordillera Transect, 270 km of crustal reflection data acquired in 1985 and subsequent magnetotelluric results define the westward extent of autochthonous North American basement (truncated by the crustal scale east-dipping Slocan Lake fault zone), the thickening of Purcell Proterozoic supracrustal rocks by folding and thrusting and a major compressional shear zone underlying the Valhalla gneiss complex. In 1988, seismic profiling was extended westward across the allochthonous terranes and plutonic complexes to the west coast. About 900 km of 60-fold, 18 s data were recorded by an industry contract crew; initial processing by contract is in progress.

In our Appalachian transect, Lithoprobe East, more than 3000 km of marine crustal reflection data have been acquired around Newfoundland and in the Gulf of St. Lawrence. Three deep crustal blocks, each of which exhibits a characteristic seismic signature, can be traced through the region. Intracrustal reflectors on the seismic line run across the continental margin northeast of Newfoundland help determine the nature of extensional tectonics and associated sedimentary basin evolution on the rifted margin.

The GLIMPCE Transect has enabled a glimpse of the mid-continent crust by recording marine reflection data in the Great Lakes. The sections from Lake Superior reveal - 10 km of sediments overlying a 20 km thick sequence of strongly reflective volcanics and inter-flow sediments, perhaps representing the greatest vertical extent of intra-continental rift deposits on Earth.The original Archean/Early Proterozoic crust is greatly thinned beneath the rift basin. In the Lake Huron/Georgian Bay profile, the Grenville front is imaged as the westernmost reflection of a spectacular 70 km wide band of southeast-dipping reflections that clearly truncates prominent sub-horizontal crustal reflections of the Superior granite-rhyolite terrane.

The Kapuskasing Structure Zone is a sequence of lower crustal rocks that have been thrust upward along the Ivanhoe Lake cataclastic zone and are now exposed at the surface. As part of the Kapuskasing Transect, 340 km of regional deep reflection data and 20 km of "high resolution" data (20 m station spacing, 20 m source interval and a higher sweep range) over the cataclastic zone were recorded in 1987-1988. The Abitibi sub-province of Ontario and Quebec contains mineralized belts for which it is famous. As a preliminary program for the Abitibi-Grenville Transect, 130 km of regional and 65 km of high-resolution seismic profiles also were recorded in 1987-1988. These were centred over major fault zones with which mineralization is related. Data quality is excellent and the data are in the process of being interpreted.

Elsevier online abstract

The structure and tectonic history of the western Canada subduction zone

Page 641
Ron M. Clowes, R. D. Hyndman, C. J. Yorath and E. E. Davis

As part of the Lithoprobe multidisciplinary earth sciences research program, multichannel seismic profiles from the deep sea across the continental shelf and on Vancouver Island, coupled with a wide range of other geophysical and geological studies, have permitted detailed delineation of the structure and tectonic history of the Juan de Fuca subduction zone in western Canada. The modern continental margin was built against a pre-Tertiary continent consisting of amalgamation of older terranes, the oldest and westernmost being Wrangellia. No pre-Eocene continental margin record exists and transform fault displacement of margin rocks north to Alaska is inferred. Eocene subduction was interpreted by a seaward jump in the trench axis in latest Eocene, trapping a section of marine volcanics (Crescent Terrane), together with sections of Mesozoic marine sedimentary rocks (Pacific Rim Terrane). These were placed against and under the margin on steeply dipping thrust faults that are well imaged in the seismic reflection data.

The terrane positions are also well delineated by magnetic and gravity data. Seaward and underlying the Crescent Terrane is the modern accretionary prism which is up to 100 km wide. Development of the deformation front at the toe of the continental slope is clearly shown by landward and seaward verging faults in the sediments overlying the basaltic oceanic crust. SeaMARC II acoustic images illustrate the surficial effects of the deformation front. Beneath the deep ocean, the continental shelf and the western part of Vancouver Island, the top of the downgoing oceanic plate is imaged well by a one or two cycle continuous reflection at depths which are consistent with those determined for the top of the plate from seismic refraction data and Wadati-Benioff zone earthquakes.

Above the subducting slab, two broad bands of high reflectivity dip beneath Vancouver Island at angles less than that of the slab. The deeper band coincides with a dipping layer of high conductivity interpreted from a magnetotelluric survey. On the basis of profile heat flow measurements which were converted to isotherms, this layer is isothermal at about 450°. One tectonic model for the origin of the reflective bands suggests that they are thick sections of underplated volcanic and sedimentary material, the lower one of which would be saturated with hot saline fluids. An alternative interpretation, supported particularly by the dipping isotherms, suggests that the lower band is caused by fluids that are driven off the downgoing oceanii: plate and are trapped below an impermeable horizon formed at a metamorphic front.

Elsevier online abstract