Geology

doi: 10.1130/0091-7613(1987)15<1118:CTATWB>2.0.CO;2 1987;15;1118-1122Geology

 Olav Eldholm, Jan Inge Faleide and Annik M. Myhre marginContinent-ocean transition at the western Barents Sea/Svalbard continental  

Email alerting servicescite this article

to receive free e-mail alerts when new articleswww.gsapubs.org/cgi/alertsclick

Subscribe to subscribe to Geologywww.gsapubs.org/subscriptions/click

Permission request to contact GSAhttp://www.geosociety.org/pubs/copyrt.htm#gsaclick

viewpoint. Opinions presented in this publication do not reflect official positions of the Society.positions by scientists worldwide, regardless of their race, citizenship, gender, religion, or political article's full citation. GSA provides this and other forums for the presentation of diverse opinions andarticles on their own or their organization's Web site providing the posting includes a reference to the science. This file may not be posted to any Web site, but authors may post the abstracts only of theirunlimited copies of items in GSA's journals for noncommercial use in classrooms to further education and to use a single figure, a single table, and/or a brief paragraph of text in subsequent works and to makeemployment. Individual scientists are hereby granted permission, without fees or further requests to GSA, Copyright not claimed on content prepared wholly by U.S. government employees within scope of their

Notes

Geological Society of America

on August 18, 2014geology.gsapubs.orgDownloaded from on August 18, 2014geology.gsapubs.orgDownloaded from

Continent-ocean transition at the western Barents Sea/Svalbard continental marg;in

Olav Eidholm, Jan Inge Faleide, Annik M. Myhre Department of Geology, University of Oslo, P.O. Box 1047, 0316 Blindem, Oslo 3, Norway

ABSTRACT The change in crustal type at the western Barents Sea/Svalbard

margin takes place over a narrow zone related to primary rift and shear structures reflecting the stepwise opening of the Greenland Sea. Regionally, the margin is composed of two large shear zones and a central rifted-margin segment. Local transtension and transpression at the plate boundary caused the early Cenozoic tectonism in Svalbard and the western Barents Sea, and might explain the prominent mar-ginal gravity and velocity anomalies.

INTRODUCTION Plate tectonic models predict an early Tertiary regional shear system

initiating the formation of the continental margin off the Barents Sea and Svalbard (Fig. 1). The geology of this margin is still poorly known due to limited seismic penetration, absence of identifiable sea-floor spreading anomalies in the Greenland Sea, and lack of data from the conjugate margin. Most plate reconstructions (Talwani and Eldholm, 1977; Unter-nehr, 1982; Nunns, 1983) show a gap in the southern Greenland Sea and considerable overlap to the north. This was, however, partly accounted for by Myhre et al. (1982), who proposed a two-stage opening of the Green-land Sea and a continent-ocean boundary (COB) at the shelf north of lat 73°N.

An extensive grid of multichannel seismic (MCS) profiles is now available in the western Barents Sea (Riis et al., 1986), whereas regional lines exist north of lat 76°N (Myhre and Eldholm, 1987). Although only a small number of lines extend onto oceanic crust, the seismic data enable us to map the main regional structural elements at the margin and to suggest a first-order evolutionary model based on a stepwise opening ocean in the Paleogene.

MARGIN GEOLOGY The seismic character and structural style change from west to east

along the entire margin. In general, the top of oceanic basement continues beneath the outer continental slope, where it changes into a more complex zone characterized by structural relief and poorer seismic continuity (Fig. 2). The landward termination of this zone is abrupt and is interpreted to mark a first-order geologic boundary that appears to relate to the change in crustal type across the margin. South of lat 76°N, we associate this change with the COB (Fig. 1). However, because significant regional differences are associated with the transition from continental to oceanic crust, we divide the margin into three main segments: (1) the Senja Fracture Zone margin south of lat 73°N and (2) the Bjornaya-Spitsbergen margin north of lat 74°N both exhibit.a north-northwest structural trend, and are linked by (3) a complex intermediate region offsetting the margin toward the northeast (Figs. 1, 2).

The main structural elements of the western Barents Sea are the Troms«) and Bjerney basins and the Svalbard Platform (Fig. 1). These two basins were initiated by Late Jurassic-Early Cretaceous tension and un-derwent maximum subsidence in Albian-Aptian times; the subsidence gradually decreased during the Late Cretaceous. The pre-Cretaceous strata shallow toward the west; thus, the main depocenter was east of the Senja Fracture Zone. In the north, the Svalbard Platform comprises Paleozoic

and early Mesozoic sediments (Faleide et a l , 1984a). The Stappen High, which protrudes from the platform south of Bjarnaya, was, however, uplifted in the ;arly Tertiary. Most of the Barents Sea was emergent in the Tertiary, when a huge wedge of sediments prograded westward into the young ocean.

The positive elongate free-air anomaly over the Senja Fracture Zone (Fig. 1) lies in a region of lateral velocity discontinuity at depth (Houtz and Windisch, 1974) and marks a change from a linear magnetic pattern to a quiet zone or the landward side (Am, 1975). Talwani and Eldholm (1977) interpreted the anomaly as the COB reflecting a sheared-margin segment formed prior to anomaly 13, a time of a major change in relative plate motion between Greenland and Eurasia. The deep penetration of the MCS data reveals the existence of the fracture zone proper. As the margin is approached from the west, the Cenozoic sedimentary apron thickens toward a narrow fault zone, east of which the character of the seismic record changes significantly. Most profiles show local basement relief just west of the fault zone. On the other side, the sedimentary beds below the Cenozoic wedge are strongly faulted (Fig. 2). We interpret these rock units to predate the early Eocene opening of the Norwegian-Greenland Sea, forming the western flanks of the Troms« and Bjflrnoy basins. On the basis of geologic continuity, the Senja Fracture Zone is placed at the base of the westernmost fault, i.e., beneath the eastern flank of the gravity anomaly (Fig. 1). Hence, we move the crustal transition 20-25 km east of the position suggested by Talwani and Eldholm (1977), who placed it at the maximum fre 3-air gravity anomaly.

Northeast of the Senja Fracture Zone, the main feature is a marginal basement higli (relative to the adjacent oceanic crust). This high, which is fault bounded (COB, Fig. 1) toward the southeast, is covered by a distinct opaque acoustic basement reflector interpreted to be early Eocene volcan-ics. Locally, (he volcanics are observed a short distance east of the fault. The opaque horizon may mask structures below, and we note some weak deeper refleci.ors at the innermost high. Actually, this region might com-prise more small rifted and sheared-margin segments than shown in Figure 1. Except for the absence of seaward-dipping subbasement reflectors, the high is simih.r to other marginal highs in the Norwegian-Greenland Sea formed during the earliest opening of the ocean (Eldholm and Thiede, 1987). The intersection between the Senja Fracture Zone and the marginal high is structurally complex, and we have identified buried basement peaks at the flank of the high (Fig. 2C). These peaks and adjacent extrusives that postdate the deeper volcanic acoustic basement reflector are related to a rejuvenation of volcanism in the earliest Oligocene (Faleide et al., 1987).

North cf lat 74°N, the structural trend is again north-northwest. The Hornsund Fuult Zone at the central shelf, defined as a narrow zone of downfaulted blocks, is the dominant feature. The fault blocks are inter-preted to be fragments of the inner shelf complex, which is without clear seismic stratification and resembles the Svalbard Platform. South of Spits-bergen the oceanic basement can be followed almost all the way to an outer fault scarp (Fig. 2). The crustal transition is placed near the base of the scarp. It appears that the Bjarnaya-Sarkapp Fault Zone farther east (Fig. 1) acted as a hinge line for the Cenozoic downfaulting of the Sval-bard Platform. The present spreading axis, the Knipovich Ridge, progres-sively approaches the margin west of Spitsbergen, and the axial mountains

1118 GEOLOGY, v. 15, p. 1118-1122, December 1987

on August 18, 2014geology.gsapubs.orgDownloaded from

underlie the lower slope in the north. In general, we observe features similar to those farther south; however, we cannot trace oceanic basement beneath the thickest sediments adjacent to the fault zone. The limit of well-identified oceanic crust is shown in Figure 1; the COB lies between this line and the westernmost downfaulted block.

CONTINENT-OCEAN BOUNDARY We define the COB as the location of change from oceanic to conti-

nental basement, realizing that isostatic considerations require a more gradual transition deeper in the crust. In addition to the MCS profiles, our interpretation is based on gravity and magnetic data and new seismic techniques such as deep crustal profiling and two-ship velocity measurements.

Figure 1 demonstrates relations between the free-air gravity field and structural lineaments at the margin. First, we note the positive anomaly belts seaward of the COB, and second, two of the belts exhibit very prominent amplitudes. The southern belt is associated with the Senja Fracture Zone, and the Hornsund anomaly between Bjarneya and south-ern Svalbard is associated with the Hornsund Fault Zone. However, no satisfactory explanation for their origin has been offered. Between these belts, along the central margin segment, there is another well-defined but less prominent anomaly belt over the marginal high. An outer gravity high of similar magnitude has been observed over the oldest oceanic crust at other rifted-margin segments (Talwani and Eldholm, 1973). This marginal high exhibits the smallest gravity anomaly, which suggests a genetically different origin from the two main anomaly belts. Nevertheless, the loca-tions of all three anomaly belts strongly implies a relation with fundamen-tal properties of the oldest oceanic crust and with structures formed during the initial phase of opening.

There are few crustal velocity data, but continental crust is inferred east of the Hornsund Fault Zone at lat 77-79°N (Guterch et al., 1978). Moreover, a traverse of expanded spread profiles centered along profile B (Figs. 1, 2) reveals a major change in crustal structure just west of the Hornsund Fault Zone (Kitterad, 1986). Normal oceanic crust is present west of the gravity anomaly. Beneath the anomaly the crust maintains a normal oceanic thickness, but a 7.1 km/s velocity is measured just below the basement surface. East of the Hornsund Fault Zone, continental crust with a prominent Moho reflector at a depth of about 30 km is observed in the deep seismic profiles (Gudlaugsson et al., 1987). Simitar results are obtained along expanded spread profile traverses across the Senja Fracture Zone.

By integrating the seismic, gravity, and magnetic (south of lat 73°N) data, we feel confident that the location of the crustal transition can be confined to a narrow zone along the margin south of Svalbard. The COB in Figure 1 marks the most landward extent of oceanic crust. At the central rifted margin, however, volcanic extrusives might hide thinned continental crust extending a short distance on the seaward side, as proposed for the Varing marginal high (Skogseid and Eldholm, 1987). Off Svalbard, the seismic penetration only allows us to place the COB within a 10-30-km-wide zone (Fig. 1).

MARGIN EVOLUTION Abnormally high seismic velocities near the top of oceanic basement

suggest that isolated high-density bodies produce the large gravity anoma-lies west of the Senja and the southern Hornsund fault zones. However, oceanic crust emplaced at a normal accretionary plate boundary does not exhibit the observed velocity-depth relation (Ewing and Houtz, 1979).

Figure 1. Main geologic provinces, structural features, and selected geophysical data. Lines A - D refer to seismic sections in Figure 2. 1: COB and main structural lineaments; 2: bathymetry (m); 3: eastern limit of identified oceanic crust in seismic record; 4: magnetic linea-tions (Talwani and Eldholm, 1977); 5: Spitsbergen fold and thrust belt (Steel et al., 1985); 6: Tertiary Central Basin (Steel et al., 1985); 7: area of volcanic overprint; 8: marginal free-air gravity anomalies (Faleide et al., 1984b) (A: 50-100 mgal, B: >100 mgal). BB: Bjdrndy Basin; B-S: Bjirntfya-Stf rkapp; HB: Hammerfest Basin; LH: Loppa high; MR: Mol-loy Ridge; SH: Stappen high; SR: Senja Ridge; TB: Tromsu basin; TFP: Troms-Finnmark platform.

SVALBARD

PLATFORM

Q B J 0 R N 0 Y A

GEOLOGY, December 1987 1119

on August 18, 2014geology.gsapubs.orgDownloaded from

w E

K N I P O V I C H R IDGE 0

8

0

8

0

r^TJ O C E A N I C L - J C R U S T

2 V O L C A N I C F L O W S

4

[ 1 s t w t

6 1 0 k m

8

0

6

8

Moreover, the anomalies are found at the predominantly sheared margin segments, where large offsets between spreading axes have caused conti-nental crustal blocks to slide along each other for considerable time periods. We have also noticed a small but distinct angle between the COB and the flowlines describing the relative plate motion, suggesting nonideal transforms and deformation along the pilate boundary. This is a leaky transform (Menard and Atwater, 1968) that created a rhomboid-shaped depression in which spreading was very oblique (Fig. 3). Because these leaky transforms are surrounded by relatively thick continental crust, in-trusions of asthenospheric material from levels deeper than at a spreading ridge might occur (Schubert and Garfunkel, 1984). As the plates separate, the plate boundary adjusts to a normal transform. We consider this mode of crustal generation to be oceanic, although it might be called mantle diapirism rather than sea-floor spreading. Structurally, this type of passive margin could be classified as an oblique-shear margin (Fig. 3). A mecha-nism of this kind is compatible with the geophysical data and the pre-Oligocene plate reconstructions shown in Figure 4. These models are relatively schematic; a better fit between the gravity anomalies and the extent of the diapiric crust can be obtained by a more complex plate-boundary configuration.

Prior to opening in the earliest Eocene, crustal extension initiated a regime of wrench tectonics between the nascent Lofoten-Greenland and Eurasia basins. Shearing took place along north-northwest-trending faults between Svalbard and northeast Greenland (Harland, 1969). The main

H O R N S U N D F A U L T Z O N E

0 Figure 2. Line drawings of selected seismic sections (Fig. 1) crossing various geologic

2 margin provinces. Parts of lines A and C have been presented by Schluter and Hinz (1978)

4 and Hinz and Schluter (1978). BT: base Tertiary reflector.

wrench region was located between the Hornsund and Bjerneya-Sarkapp fault zones and the Trolleland Fault Zone (Fig. 4). To the north, transpres-sion initiated ( he fold and thrust belt of Spitsbergen (Steel et al., 1985). To the south, trar spressional motion caused local inversion at the Senja Ridge and the Stappen High. Conversely, transtension weakened the crust, and dense mantle material was intruded locally. These intrusions have been preserved as :he Senja Fracture Zone and Hornsund gravity anomalies. This tectnoic regime continued after the time of initial opening.

Figure 4 shows the evolution of the southern Greenland Sea prior to the plate reorganization at the time of anomaly 13. The principal feature of this first-order model is a stepwise northeastward propagation of the rift axis. Our timing may contain some error due to the lack of structural data from the Greenland margin and spreading anomalies in the Greenland Sea. In the model, no oceanic crust formed east of the Greenland-Senja Frac-ture Zone from initial opening to the time of anomaly 23. The plate boundary appears to have continued into the Trolleland Fault Zone in northeast Greenland (H&kansson and Stemmerik, 1984) as a large continent-continent transform. However, crustal translation also took place to the eiist as the crust was thinned, prior to opening along the central rifted-margin segment, at about the time of anomaly 23 to 22. At the eastern end of the new ocean basin, crustal accretion by oblique spreading began at the time of anomaly 22. Finally, the plate boundary moved a short distance to the northeast between anomaly 21 and 13, when the main Hornsuid fault became the regional transform to the Eurasia Basin.

R i f t - m a r g i n

Figure 3. Schematic diagram depicting formation of oblique shear margin.

R i f t - m a r g i n

Figure 3. Schematic diagram depicting formation of oblique shear margin.

1120 GEOLOGY, December 1987 1119

on August 18, 2014geology.gsapubs.orgDownloaded from

o tn O

A 2 3 ( 5 4 M a ) A 21 (49 .5 M a ) 5 10

A 1 3 ( 3 6 M a )

COB, Main structure v 1 * Oblique shear crust HR : Hovgaard Ridge

Limit identified Free-air gravity (mGal) GFZ : Greenland F2 oceanic crust

Figure 4. Greenland Sea Paleocene to pre-Oligocene plate reconstructions based on rotation poles of Talwani and Eidholm (1977).

on August 18, 2014geology.gsapubs.orgDownloaded from

During this entire period the plate configuration caused transpression between northeast Greenland and northern Spitsbergen. Our models indi-cate maximum crustal shortening prior to anomaly 21, gradually diminish-ing until anomaly 13 time. This event was responsible for the Spitsbergen orogeny, and the timing is compatible with data from the Central Basin in Spitsbergen that suggest major faulting with a possible strike-slip compo-nent from early Paleocene time. Observations in both the Central Basin and the fold and thrust belt also suggest a transpressive regime starting in the late Paleocene (Steel et al., 1985).

The spreading pattern in Figure 4 was maintained until the time of anomaly 13, when the relative direction of plate movement became northwest, first causing crustal thinning and, later, sea-floor spreading in the northern Greenland Sea. This change is reflected in the normal faults composing the Hornsund Fault Zone and in subparallel features on the continental shelf. Downfaulting probably followed the existing zones of weakness. A shift of the plate boundary also separated the Hovgaard ridge from the Svalbard margin (Myhre et al., 1982), and volcanism occurred locally. Subsequent to anomaly 13 time, the plate tectonic evolution might have been complex, gradually changing into the almost continuous plate boundary now present at the Knipovich Ridge.

As the ocean matured, a large volume of sediments from the sur-rounding landmasses was transported into the young basins. The shelf edge migrated west from hinge lines at the eastern Tromsa and Bjarnay basins and along the Bjerneya-Serkapp and northern Hornsund fault zones. The Cenozoic stratigraphy and sediment thickness suggest a significant increase in deposition and margin subsidence during the past 6 m.y. (Myhre and Eldholm, 1987). This might relate to glaciations, but a similar history applies to the entire Norwegian margin. No satisfactory mechanism has yet been proposed for this young pulse of vertical motion.

CONCLUSIONS The geology of the margin reflects a stepwise opening of the Green-

land Sea in the Paleogene. This evolution formed long shear, or oblique shear, margin segments at the Senja Fracture Zone and the southern Hornsund Fault Zone, linked by a rifted segment southwest of Bjcrndya. This segment comprises a marginal high characteristic of volcanic rifted margins. Late Paleocene and early-middle Eocene compression north of the evolving ocean basin created the Spitsbergen orogeny. The early Ter-tiary wrench regime was also accompanied by local transpression and transtension forming the present structural framework of the western Bar-ents Sea. This setting allowed local intrusions of mantle material along shear zones within a relatively thick continental crust. Seismic velocity data show that these elongate, high-density, intrusive bodies give rise to large gravity anomalies. Finally, the major plate-tectonic change in the North Atlantic realm in the early Oligocene created a predominantly tensional regime and opened the northern Greenland Sea. We conclude that the COB is located at or just seaward of the system of well-defined structural lineaments south of Svalbard, and within a narrow zone farther north. In our model, the intrusive bodies represent the oldest oceanic crust. The position we infer for the COB, partly located under the continental shelf, removes much of the overlap encountered in previous plate reconstructions.

REFERENCES CITED Am, K., 1975, Magnetic profiling over Svalbard and surrounding shelf areas: Norsk

Polarinstutt Arbok 1973, p. 87-100. Eldholm, O., and Thiede, J., 1987, Summary and preliminary conclusions, ODP

Leg 104: Proceedings, Ocean Drilling Program Initial Reports, Part A, p. 751-771.

Ewing, J., and Houtz, R., 1979, Acoustic stratigraphy and structure of the oceanic crust: American Geophysical Union, Mauri« Ewing Series, v. 2, p. 1-14.

Faleide, J.I., Gudlaugsson, S.T., and Jacquart, G., 1984a, Evolution of the western Barents Sea: Marine and Petroleum Geology, v. 1, p. 123-150.

Faleide, J.I., Gudlaugsson, S.T., Johansen, B., Myhre, A.M., and Eldholm, O., 1984b, Free-air gravity maps of the Greenland Sea and the Barents Sea: Norsk Polarinstit Jtt Skrifter no. 180, p. 63-67.

Faleide, J.I., Myhre, A.M., and Eldholm, O., 1987, Early Tertiary volcanism at the western Barents Sea margin: Geological Society of London (in press).

Gudlaugsson, S T., Faleide, J.I., Fanavoll, S., and Johansen, B., 1987, Deep seismic reflection profiles across the western Barents Sea margin: Royal Astronomical Society Gfophysical Journal, v. 89, p. 273-278.

Guterch, A., Pajchel, J., Perchuc, E., Kowalski, J., Duda, S., Komber, J., Bojdys, G., and Scllevoll, M.A., 1978, Seismic reconnaissance measurement on the crustal stri cture in the Spitsbergen region 1976: Bergen, University of Bergen Seismological Observatory, 61 p.

Hikansson, E., and Stemmerik, L., 1984, Wandel Sea Basin—The North Greenland equivalent to Svalbard and the Barents shelf, in Spencer, A.M., ed., Petroleum geology of the North European margin: London, Graham & Trotman, p. 97-108.

Harland, W.B., 1969, Contribution of Spitsbergen to understanding of tectonic evolution of the North Atlantic region, in Kay, M., ed., North Atlantic geology and continental drift: American Association of Petroleum Geologists Memoir 12, p. 817- 851.

Hinz, K., and Schlüter, H-U., 1978, The geological structure of the western Barents Sea: Marine Geology, v. 26, p. 199-230.

Houtz, R.E., and Windisch, C.C., 1974, Barents Sea continental margin sonobuoy data: Geological Society of America Bulletin, v. 88, p. 1030-1036.

Kittered, N.O., 1986, Velocity-depth changes. Interpretation procedures for ESP and sonobuoy data [thesis]: Oslo, University of Oslo, 193 p. (in Norwegian).

Menard, W.R., and Atwater, T., 1968, Changes in the direction of sea floor spread-ing: Nature, v. 222, p. 1037-1040.

Myhre, A.M., a ad Eldholm, O., 1987, The western Svalbard margin (74-80°N): Marine ancl Petroleum Geology (in press).

Myhre, A.M., Edholm, O., and Sundvor, E., 1982, The margin between Senja and Spitsberger fracture zones: Implications from plate tectonics: Tectonophysics, v. 89, p. 1-32.

Nunns, A.G., IS83, Plate tectonic evolution of the Greenland-Scotland Ridge and surrounding regions, in Bott, M.H.P., et al., eds., Structure and development of the Greenla nd-Scotland Ridge: New York, Plenum Press, p. 11-30.

Riis, F., Vollset, J., and Sand, M., 1986, Tectonic development of the western margins of :he Barents Sea and adjacent areas: American Association of Petro-leum Geologists Memoir 40, p. 661-676.

Schlüter, H-U., und Hinz, K., 1978, The continental margin of West Spitsbergen: Polarforschung, v. 48, p. 151-169.

Schubert, G., and Garfunkel, Z., 1984, Mantle upwelling in the Dead Sea and Salton Trot gh-Gulf of California leaky transforms: Annales Geophysicae, v. 2, p. 633-648.

Skogseid, J., ancl Eldholm, O., 1987, Early Cenozoic crust at the Norwegian con-tinental margin and the conjugate Jan Mayen Ridge: Journal of Geophysical Research (in press).

Steel, R., Gjeltx.rg, J., Nattvedt, A., Heiland-Hansen, W., Kleinspehn, K., and Rye-Larsen, M., 1985, The Tertiary strike-slip basins and orogenic belt of Spitsbergen: Society of Economic Paleontologists and Mineralogists Special Publication 37, p. 339-360.

Talwani, M., and Eldholm, O., 1973, The boundary between continental and oceanic cruiit at the margins of rifted continents: Nature, v. 241, p. 325-330. 1977, Evolution of the Norwegian-Greenland Sea: Geological Society of America Bulletin, v. 83, p. 3575-3608.

Unternehr, P., 1982, Structural and kinematic studies of the Norwegian-Greenland Sea and the evolution of the Jan Mayen Ridge [Ph.D. thesis]: France, Univer-sity of Bretagne, 227 p. (in French).

ACKNOWLEDGMENTS Partly supported by the Royal Norwegian Council for Science and the Human-

ities, Norwegian Petroleum Directorate, and Elf Aquitaine Norge A.S. We thank Karl Hinz, Eirik Sundvor, the Continental Shelf and Petroleum Technology Re-search Institute, and the Norwegian Petroleum Directorate for making seismic data available.

Manuscript received April 20, 1987 Revised manuscri pt received August 19,1987 Manuscript accep ted September 10, 1987

1122 Primed in U.S.A. GEOLOGY, December 1987

on August 18, 2014geology.gsapubs.orgDownloaded from