Basement structure of the north-western Yermak Plateau

Transcription

1 GEOPHYSICAL RESEARCH LETTERS, VOL. 35, L05309, doi: /2007gl032892, 2008 Basement structure of the north-western Yermak Plateau W. Jokat, 1 W. Geissler, 1 and M. Voss 1 Received 13 December 2007; revised 28 January 2008; accepted 6 February 2008; published 12 March [1] The separation of Northeast Greenland and Svalbard was achieved by large strike slip movements in Cenozoic times. Evidence for these movements can be found onshore both on North Greenland and Svalbard. However, the role of the Yermak Plateau in this process is quite speculative. New multichannel seismic (10 km spacing) and aeromagnetic data (7.5 km spacing) across the northwestern part of the plateau show that the acoustic basement has a similar strike direction to that of the geological units onshore Svalbard. A prominent fault zone separates these most likely continental structures in the west from a more N-S extended transitional crustal block in the eastern part of the plateau. This part of the plateau is characterized by strong magnetic anomalies at least indicating highly intruded and stretched continental or even oceanic crust. However, the seismic data show that the plateau-like bathymetry is quite young. During most of its Cenozoic history the Yermak Plateau had a rough topography, similar to the topography onshore Svalbard. Thus, the paleobathymetry might have played an important role for the water exchange between the Arctic Ocean and the North Atlantic prior to the opening of the Fram Strait, which is today the main pathway for the deep-water masses. Citation: Jokat, W., W. Geissler, and M. Voss (2008), Basement structure of the north-western Yermak Plateau, Geophys. Res. Lett., 35, L05309, doi: /2007gl Introduction [2] The Fram Strait is the only deep-water connection between the Arctic Ocean and the global oceans. Its deepest part hosts an active mid-ocean ridge, the Lena Trough. The extension in the Lena Trough is the latest stage of the separation of Greenland and Svalbard, which started with large strike-slip movements about 50 Ma. According to published geodynamic models, a deep-water connection was established about 14 Ma [Kristoffersen, 1990; W. Jokat et al., Timing and geometry of the Fram Strait opening, submitted to Geophysical Journal International, 2008], but the Arctic Ocean was already ventilated through a shallow to intermediate water connection earliest 17.5 Ma [Jakobsson et al., 2007]. [3] Two prominent plateaus bound the mid-ocean ridge in the Lena Trough to the east and west: the Yermak Plateau and Morris Jesup Rise, respectively (Figure 1). Currently, these rises are thought to have formed 35 Ma [Feden et al., 1979], when relative movements between Svalbard and Greenland became significant and, as a consequence, volcanic material erupted onto these rises and/or formed them 1 Alfred Wegener Institute for Polar Research, Bremerhaven, Germany. Copyright 2008 by the American Geophysical Union /08/2007GL completely. There is no information available to constrain the amount of magmatic material that erupted during this time period. According to some speculations, the northeastern Yermak Plateau is oceanic in origin [Feden et al., 1979; Jackson et al., 1984], while its southern part might consist of stretched continental crust [Ritzmann and Jokat, 2003]. The latter interpretation is based on an on/offshore seismic refraction experiment off Northern Svalbard. Speculations on the oceanic origin of the Yermak Plateau [Feden et al., 1979] rely on aeromagnetic investigations, which show strong (up to 1000 nt) magnetic anomalies over both rises, which might be related to magmatism during continental break-up. However, no systematic seismic data existed to constrain the interpretations by defining the nature of the basement and deeper crust of the northern and northeastern part of the plateau. In the 2004 summer season, a systematic seismic network and a dense grid of aeromagnetic data was acquired across the north-western Yermak Plateau to map its sedimentary and basement structures. In this contribution, we present new information on the magnetic anomaly distribution and the structure of the acoustic basement of the Yermak Plateau, and discuss the consequences of the new results for its tectonic evolution. 2. Experiment Set-Up and Data Processing [4] Before the expedition ARK XX/3 with RV Polarstern in 2004 [Stein, 2005], available ice maps showed that the ice edge in the Fram Strait had retreated as far north as 82 N. This observation was more or less confirmed upon arrival, although the presence of compact ice fields drifting in the previously open water area made seismic investigations difficult. The seismic data were acquired using an 800 m long streamer (96 channels, 6.25 m group spacing), and an airgun array with a total volume of 24 l. The Yermak Plateau east of 10 E was not accessible with this experimental set-up, because thick multi-year ice floes did not allow any single ship seismic investigations. North of 82 N only two profiles were acquired until the ship got stuck in thick ice floes. [5] Contemporaneously with seismic profiling across the Yermak Plateau an aeromagnetic survey was performed by helicopter. The aim was to complete profiles in a previous magnetic survey from 1999 (Figure 1), and to densify the existing magnetic data set in order to resolve more details in the magnetic field of the Yermak Plateau, especially in the transition area from low/moderate to strong positive anomalies (ca nt) in the east. The line spacing was 7.5 km. The flight altitude was mainly 100 m, and the data were acquired at a speed of 150 km/h (40 m/s). Within nine days and 20 flights, more than 7250 kilometres of new aeromagnetic data were acquired. Standard data reduction has been applied to the measurements including regional L of6

2 Figure 1. Location of the seismic profiles on the Yermak Plateau underlain by the new aeromagnetic data. The thin lines indicate the flight tracks. The bathymetry [Jakobsson et al., 2000] is contoured in 500 m intervals. The red lines indicate seismic profiles acquired in The black bold line is shown in this study, while the E-W trending grey lines show the seismic network, which is used to map basement topography shown in Figure 4. Abbreviations: 005, AWI ; GR, Gakkel Ridge; LR, Lomonosov Ridge; LT, Lena Trough; MJR, Morris Jesup Rise; NG, North Greenland; SV, Svalbard; YP, Yermak Plateau. IGRF (international geomagnetic reference field) and diurnal corrections. Diurnal variations were taken from recordings of a base station located in Ny Ålesund (Spitsbergen) permanently operated by the University of Tromsø ( geo.phys.uit.no/viewasc/). The local total magnetic field was recorded in intervals of 10 s. Prior to the reduction, an 1800 s time filter was applied in order to suppress high frequency local variations, and to remove the regional trend Figure 2. Misties of the magnetic data presented in this study prior and after levelling. 2of6

3 Figure 3. Unmigrated seismic section of profile AWI with line drawing of acoustic basement. The distance between the CDPs is 25 m. The gravity (dashed line) ( and magnetic data (solid line) along the profile are drawn on top. 3of6

4 Figure 4. Compilation of depth to basement in TWT for the seismic network across the north-western Yermak Plateau. The thin grey lines show the seismic lines in this area, which were accessible for this study. In the right lower corner part of the Svalbard mainland is displayed. of the variations. An upward continuation of all ground and airborne magnetic recordings to 1000 m allowed merging of the two separate surveys (1999 and 2004) and combined processing and level adjustment at line intersections. Levelling, using LCT software (Fugro Ltd.), and appropriate adjustments revealed a reduction of an average mistie at cross points from a mean of 78 nt to less than 11 nt (Figure 2). 3. Results [6] The profile AWI is the easternmost line in our 2004 seismic survey of the plateau (Figure 1). It crosses the Yermak Plateau in a region, where strong magnetic anomalies start to appear. The present sea floor is almost flat, but disturbed by iceberg scours. At deeper levels, the basement topography shows surprising variations. Starting in the NNW, a basement high terminates the bathymetric plateau (Figure 3, CDP ). The slope is covered by more than 2 km of sediments. The magnetic anomaly across this feature is 750 nt. The central part of the line imaged another basement high (Figure 3, CDP ) that crops out at the sea floor. This high is associated with a magnetic anomaly of about 250 nt. Again the SSE termination of the plateau is underlain by a basement high (Figure 3, CDP ) buried by more than 1 s TWT (Two way travel time) of thick drift sediments. The grabens between the basement highs are up to 2.5 s TWT deep. Sedimentary units follow the shape of the grabens, indicating subsidence and probably gravity driven mass transport. At shallower levels (less than 1.7 s TWT) deposition seems to become increasingly influenced by currents with channels formed at the flanks of the central basement high. Some reflectors crop out at the sea floor. This may indicate that the present day bathymetry is a consequence of glacial erosion (Figure 3, CDPs 4800, 3800, 2900). Indications for free gas at 1.5 s TWT can be found between CDP 2800 and [7] A basement map in two-way travel time for the new seismic network shows a N-S trend in the acoustic basement (Figure 4). West of 002 E, a similar trend is also evident in the oceanic basement, which was formed during the evolution of the Lena Trough. East of 002 E the basement of the Yermak Plateau is much shallower, but still shows N-S basement structures that correlate with a negative free air gravity anomaly (Arctic gravity grid; [8] The high-resolution magnetic image reveals a clear segmentation of the large positive anomaly (Figure 1). A clear NE-SW trend of the major anomaly appears separated from the positive anomaly west of it. South of the ridge-like structure the plateau seems almost magnetically unaffected and an extent of the anomaly to the SE, compared to the regional magnetic map after Verhoef et al. [1996] cannot be supported. 4. Interpretation and Discussion [9] The general trend of the Yermak Plateau basement is of special interest for the tectonic evolution and origin of the plateau. Although the spacing of the seismic profiles is still 4of6

5 about 10 km, we are confident that the data capture the major structural information. The seismic data show that the buried topography of the Yermak Plateau is quite rough. Ridges and mountains with a relief of up to 3000 m are present. The present plateau-like shape is a consequence of mainly current driven sedimentation resulting in infilling the deep depressions between the topographic highs. The strike of the horst and grabens north of 81 N is similar to that of geological structures onshore/offshore Svalbard. Here, the shelf off western Svalbard, as well as the onshore geology is highly deformed due to strike slip movements between Svalbard and North Greenland. [10] The region between Svalbard and North Greenland was strongly modified, when the Gakkel Ridge started to propagate southwards some 35 Ma [Vogt et al., 1979]. The resulting separation of North Greenland and Svalbard was accompanied by strong volcanism, which partly formed the two conjugate plateaus, the Morris Jesup Rise and the Yermak Plateau. Since no rock samples from this volcanic phase are recovered so far from both plateaus, their exact age and rock composition is unknown. Thus, we can only speculate to what extent e.g. the Yermak Plateau was modified during the initial break-up phase. The volcanism is inferred from the strong magnetic anomalies with amplitudes up to 1000 nt (Figure 1). The shape of the magnetic anomalies has fostered speculations that the north-eastern part of the Yermak Plateau is of oceanic origin [Feden et al., 1979; Jackson et al., 1984]. Such interpretations are difficult to support from the newly acquired data. [11] Having mapped the shape of the acoustic basement, we propose that a) the acoustic basement west of 10 E was deformed during strike-slip movements between North Greenland and Svalbard, b) the composition of the Yermak Plateau acoustic basement is, in general, identical to that of mainland Svalbard, though the crust of the Yermak Plateau is highly stretched, and c) the general strike direction suggests that the Hornsund Fault Zone (Figure 4), which marks the continent-ocean transition in the south, continues northwards, where it may broaden. [12] Basement outcrops were identified during earlier investigations and named Sverdrup Bank (Figure 4) [Sundvor and Austegard, 1990; Eiken, 1994]. The new data indicate that Sverdrup Bank is part of a plateau-wide fault zone. In our interpretation the eastern flank of the Sverdrup Bank not only dissects the plateau, but also points to differences in the geological processes that modified its deeper structure. [13] We speculate that the geology west of this fault zone is mainly influenced by transpressional/transtensional movements between North Greenland and Svalbard. Here, the magnetic field is rather smooth and, thus, provide little evidences for extensive or large magmatic intrusions. Our study confirms earlier interpretations based on deep seismic sounding data that the Yermak Plateau may consist of extended continental crust [Ritzmann and Jokat, 2003]. [14] The situation changes east of the fault zone. Here, we assume that the NNW-SSE basement trend is mainly caused by rifting of the Lomonosov Ridge and the propagation of the Gakkel Ridge. Our seismic line seems to be located on the transition zone between the two regimes. Figure 1 shows that first the strong magnetic anomalies occur only at the northernmost termination of the Yermak Plateau (Figure 3), and then jump to the centre of the NE-Yermak Plateau. We interpret the northernmost basement block (Figure 3, CDP ) to be of volcanic origin, and to have formed during the final separation of the Morris Jesup Rise/Yermak Plateau. This interpretation is supported by Jackson et al. [1984], who found high seismic velocities at shallow depths in the investigated area. We relate the high seismic velocity to intrusions causing the strong magnetic anomaly along line The central basement high (Figure 3, CDP ) has only a small magnetic anomaly of 250 nt. Further south the magnetic field is flat and almost constant. Just northeast of line , the plateau exhibits magnetic anomalies with amplitudes of up to 1000 nt. However, the deeper structure and the bathymetry of this area are unknown. Only one seismic transect crosses the plateau in the east at N15 E (Figure 1, red line), and shows a much simpler basement structure [Jokat et al., 1995]. Thus, it seems likely that the central basement high on line is simply the southern termination of the basement structure that is associated with the strong magnetic anomalies in the northeast. [15] Currently it is unknown, if older rift grabens are filled a) by mainly volcanic material, or b) marine or terrestrial sediments deposited since the formation of the Yermak Plateau. Neither the acoustic nor the potential field data support an interpretation that the fault zone in the central part of the Yermak Plateau is a continent-ocean transition. Thus, we favour a model of stretched and intruded continental crust for the north-eastern part of the Yermak Plateau. Some of the NNW-SSE extension of the NE Yermak Plateau might already been triggered during the separation of the Lomonosov Ridge from the Barents Sea margin around 55 Ma. In this case, the SE-NW trending fault zone would have separated two stress regimes, and accommodated differential movements between them. 5. Conclusions [16] From geodynamic models it can be assumed that the basement highs in the seismic network formed during the final split of the Yermak and Morris Jesup plateaus and/or during the early rifting period (35 55 Ma). Thus, the oldest sediments in the grabens of the Yermak Plateau might be around 35 Ma old. The top of the acoustic basement strikes almost N-S in the investigated area, and it might consist of rocks as old as Devonian. Mapping the acoustic basement within the new seismic network indicates that the Yermak Plateau contains a) structures that were created during strike-slip movements between North Greenland and Svalbard, and b) structures in the NE that might at least partly have formed during the Early Cenozoic rifting of the Lomonosov Ridge, and might have been overprinted by the final split of the Morris Jesup Rise and Yermak Plateau. Thus, our study does not support models, which suggest oceanic crust for the Yermak Plateau. [17] Future geodynamic reconstruction should include the older topography of the Yermak Plateau. The deep basins in-between the basement highs of the north-western Yermak Plateau might have allowed already a limited water exchange between the Arctic and the young North Atlantic 5of6

6 Ocean, already well before the opening of the Fram Strait. To what extent a shallow water connection existed, is related to the subsidence history of the northern North Atlantic, which is subject of future investigations. [18] Acknowledgments. We thank the captain and crew of RV Polarstern as well as the helicopter crew and the chief scientist for their excellent support during the expedition. Finally we thank Volker Leinweber for supporting the final compilation of the magnetic data, and two anonymous reviewers for helpful comments on the manuscript. References Eiken, O. (1994), Seismic atlas of western Svalbard: A selection of regional seismic transects, Nor. Polarinst. Medd., 130, Feden, R. H., P. R. Vogt, and H. S. Fleming (1979), Magnetic and bathymetric evidence for the Yermak Hot Spot northwest of Svalbard in the Arctic Basin, Earth Planet. Sci. Lett., 44, Jackson, H. R., G. L. Johnson, E. Sundvor, and A. M. Myhre (1984), The Yermak Plateau: Formed at a triple junction, J. Geophys. Res., 89, Jakobsson, M., N. Z. Cherkis, J. Woodward, R. Macnab, and B. Coakley (2000), New grid of Arctic bathymetry aids scientists and mapmakers, Eos Trans. AGU, 81, 89. Jakobsson, M., et al. (2007), The early Miocene onset of a ventilated circulation regime in the Arctic Ocean, Nature, 447, , doi: /nature Jokat, W., E. Weigelt, Y. Kristoffersen, T. Rasmussen, and T. Schöne (1995), Geophysical and bathymetric results from the Morris Jesup Rise, Yermak Plateau and Gakkel Ridge, Geophys. J. Int., 123, Kristoffersen, Y. (1990), On the tectonic evolution and the paleoceanographic significance of the Fram Strait Gateway, in Geological History of the Polar Oceans: Arctic Versus Antarctic, edited by U. Bleil and J. Thiede, pp , Kluwer, Amsterdam. Ritzmann, O., and W. Jokat (2003), Crustal structure of northwestern Svalbard and the adjacent Yermak Plateau: Evidence for Oligocene detachment tectonics and non-volcanic breakup, Geophys. J. Int., 152, Stein, R. (2005), Scientific cruise report of the Arctic expedition ARK-XX/ 3ofRVPolarstern in 2004: Fram Strait, Yermak Plateau and East Greenland continental margin, Ber. Polarforsch. 517, 188 pp., Alfred Wegener Inst. for Polar and Mar. Res., Bremerhaven, Germany. Sundvor, E., and A. Austegard (1990), The evolution of the Svalbard Margins: Synthesis and new results, in Geological History of the Polar Oceans: Arctic Versus Antarctic, edited by U. Bleil and J. Thiede, pp , Kluwer, Amsterdam. Verhoef, J., R. Macnab, W. Roest, J. Arjani-Hamed, and the project team (1996), Magnetic anomalies of the Arctic and North Atlantic and adjacent land areas. GAMMAA5 (Gridded Aeromagnetic and Marine Magnetics of the North Atlantic and Arctic, 5 km) [CD-ROM], Open File 3125a, Geol. Surv. of Can., Ottawa. Vogt, P. R., P. T. Taylor, L. C. Kovacs, and G. L. Johnson (1979), Detailed aeromagnetic investigations of the Arctic Basin, J. Geophys. Res., 84, W. Geissler, W. Jokat, and M. Voss, Alfred Wegener Institute for Polar Research, Bremerhaven D-27568, Germany. (jokat@awi-bremerhaven.de) 6of6