Project aims

The surface of the Earth is made of many rigid pieces known as plates. The margins between these plates are either constructive (adding material to the plate), destructive (removing material from the plate), or conservative (material is neither gained nor lost) (figure 1). In the Arctic region, the boundary between the North American plate and the Eurasian plate is a constructive margin – these two plates are spreading apart from one another and new oceanic crust is being created and added to the outer shell of the planet along the Gakkel Ridge. The Gakkel Ridge has been active for the last 55 million years or so (Jackson and Gunnarsson, 1990). Before this, a different constructive margin orthogonal to the Gakkel Ridge existed and this ‘proto-Arctic’ Ocean was much smaller than the Arctic Ocean of today. Prior to about 130 million years ago, the Arctic Ocean did not even exist.

A widely held theory for the development of the proto-Arctic Ocean, known as the ‘rotation’ hypothesis, suggests that when North America and Eurasia began spreading apart (rifting), Eurasia rotated away from North America (Lawver et al., 2002). This rotation was anchored around a point (the ‘pole of rotation’) in the MacKenzie delta of Alaska and rifting was asymmetric, like the opening of a zipper (figure 1). According to the rotation hypothesis Wrangel Island, Chukotka and possibly the New Siberian Islands would all have been derived from the North American Arctic margin.

If the rotation hypothesis is correct then signatures of distinctive tectonic processes such as collision, rifting and subsidence should be recorded in the geologic record as orogenic belts, ophiolites, ocean basin sediments, etc. Using these signatures to evaluate the geologic evolution of the greater circum-Arctic region, we are testing the rotation hypothesis and the correlations suggested by it (e.g. Pease 2006). This work is part of international research efforts to understand the regional tectonic evolution of circum-Arctic palaeogeography, i.e. CASP (Circum-Arctic Sediment Provenance), PLATES & GATES (International Polar Year project no. 77: Plate Tectonics and Polar Gateways). This work also has important implications for global economic resources since significant proven and potential reservoir rocks also occur in Arctic shelf regions, as well as for the development of the modern global climate system because cold water flowing from the Arctic and Antarctic Oceans drives global ocean circulation.

Field season 2006

The main target for our fieldwork in 2006 was Wrangel Island in the Russian Arctic, but we also took advantage of unexpected and exciting opportunities to work on Axel Heiberg Island in arctic Canada and in North Greenland. Wrangel Island would have been near Axel Heiberg Island prior to its counterclockwise rotation from the North American margin. If the geology and age relationships on Wrangel Island are similar to those on Axel Heiberg Island, this will provide direct confirmation of the rotation hypothesis. Unfortunately the expedition to Axel Heiberg Island was grounded in Resolute; winter never left the island and float planes were unable to land. Work in North Greenland provided an opportunity to continue developing our reference database: North Greenland has a fantastic, continuous and in situ (non-rotated) sedimentary succession to which our other circum-Arctic locales can later be compared.

Wrangel Island

Fieldwork

Our international group (American, Russian, and Swedish participants) met in Moscow and on 10^th July flew northeast across Siberia to Pevek. In Pevek, supplies were organized and local geology was investigated while awaiting our ‘weather window’ to fly by helicopter via Cape Schmidt to Wrangel Island, 120 km north of mainland Russia. We reached Wrangel Island on 27^th July and on 29^th July arrived at our first base camp in the Tsentralny Mountains. After a few days evaluating the stratigraphy we divided into separate working groups, each operating from several base camps (figure 2). We typically hiked 10–15 km per day, either directly from camp or, if the objective was far away, from a vehicle drop-off point. The expedition experienced generally fine weather and motorized vehicle transportation ensured that our original scientific objectives were exceeded!

After two weeks in the field our working groups re-combined along the Mamontovaya River and on 12^th August headed to Somnitelnaya to await our return flight to the Russian mainland. Some days were used to work locally and on 21^st August the helicopter flew direct from Wrangel Island to Pevek, where we made our connections to Moscow and beyond.

Preliminary results

Wrangel Island represents a north-vergent fold and thrust belt. All rock units, from Precambrian to Triassic in age, are folded at the meso- and macro-scale and thrust faulted. Folding varies from close to isoclinal. Bedding and cleavage generally dip 20–40° to the south, though steeper angles are seen locally. Variable bedding and cleavage relationships have resulted from folding (e.g. upright versus overturned limbs, steeper cleavage-tobedding intersections in hinges). As a result of this folding and faulting, the stratigraphic section is both duplicated and/or missing! Consequently, though outcrop exposure is good and a general sequence stratigraphy is identifiable (figures 3 and 4), determining the original thickness of strata and the detailed relationships between units is difficult.

The geology of Wrangel Island is dominated by Palaeozoic and Mesozoic clastic and shallow water marine sequences, that is erosional sediments and shelf carbonates, deposited unconformably on Precambrian basement (figures 3 and 4). Though folding and faulting across Wrangel Island has resulted in the widespread out-of-sequence juxtaposition of strata, gradational contacts between Palaeozoic and Mesozoic units are seen locally. These sediments document a change in depositional environment, from a subsiding passive margin to an actively uplifting margin; predominantly terrigenous clastics (Devonian) overlain by shallow marine carbonates (Carboniferous) and deeper marine (basinal) black shale (Permian) record the former, while the immature, organic-rich turbiditic sandstone and shale (Triassic) overlying Permian shale records the latter (figures 3 and 4). In the coming months, provenance analysis of Triassic sediment will be important for constraining plate tectonic reconstructions associated with the development of the Arctic Ocean (for example Miller et al., 2005).

The nature of the Precambrian basement is poorly understood. Numerous age data exist for these rocks (predominantly K-Ar whole rock analyses), but because they are metamorphosed (greenschist facies) and variably deformed (figures 3 and 4), these ages are interpreted to reflect metamorphic resetting. A few conventional U-Pb zircon dates range from 630–700 Ma, with inherited grains suggesting the possible presence of older source material at depth (see Kosko et al. 1993, for a summary of existing work). The geochemistry of these rocks is essentially unknown. Thus a major focus of our work on these rocks will be to define the age and magmatic evolution of this basement using zircon U-Pb, O-, and Hf-isotopic analyses on single crystals at NORDSIM (the Nordic ion-microprobe facility in Stockholm). This information will be used to correlate circum-Arctic Precambrian terranes.

Notable results from work to date include:

1. recognition of highly strained Devonian conglomerates and sandstones, with possible mylonitic deformational fabrics – the verification of mylonitic fabrics is important, as it would suggest thrust faulting occurred at deeper levels than previously thought;
2. identification of Carboniferous pepperite, mixtures of lava and wet sediment (previously reported as basaltic breccia);
3. a sampling cross-section from the central mountains to the southern coast (Precambrian through Triassic) for microfauna, fissiontrack and provenance analyses. Further work on these topics requires sample processing and laboratory access, which will also begin in the coming months.
   

   Field photographs – A) Polydeformed precambrian orthogneiss. B) Polymict Devonian conglomerate with granitic clast indicated by arrow. C) Permian black shale. D) Carboniferous limestone forming ledge against the skyline. E) Thick Triassic sandstone bed. Photos: Victoria Pease, Vladimir Verzhbitsky, Elizabeth Miller.

North Greenland

Fieldwork

Our field party of Danish and Swedish scientists assembled in Copenhagen on 17^th July and travelled by air to Svalbard. From Svalbard equipment and personnel were loaded into a rugged aircraft (Twin Otter) and transferred to Station Nord, North Greenland. On 18^th July we flew to our first base camp at the western end of Ovre Midsommersø (figure 5), where we spent ten days mapping, logging and collecting samples. On 27^th July we transferred to our second base camp and spent three days near Brønlundhus (figure 5) before returning to station Nord. The weather was excellent, permitting a large dataset of field observations and samples to be collected.

Preliminary results

Base camps in North Greenland were selected for access to the late Precambrian–Cambrian stratigraphy, i.e. rocks 1 000–545 million years old. These rocks include an important glacial unit known as tillite, which represents icedeposited material of inferred Neoproterozoic age (Collinson et al., 1989). Such deposits are important for understanding the timescales of climate change and variability on Earth. For example, Hoffman et al. (1998) postulated a global glaciation in the Neoproterozoic with global mean temperatures about -50°C! Constraining the absolute age of the tillite from the Morænesø Formation will be an important part of our research in the coming months.

The Morænesø Formation fills a series of ancient valleys cut into the underlying Independence Fjord group (figure 5). The Morænesø formation is made of river and aeolian sediments and its base has proximal rock fall deposits. More widespread conglomerates overlie valley-floor fluvial sands and contain evidence for ice-push. These conglomerates, unlike the rock fall deposits (figure 6A), contain fartravelled clasts, many of which are granitic. Some clasts have striated surfaces suggesting a phase of glacial transport prior to final emplacement (figure 6B). The granitic clasts will help define the older basement of the region, which is presently covered by the Greenland ice cap. Further evidence of peri-glacial process operating at the time of deposition includes fossil polygonal ground and dropstones (pebbles deposited from floating ice). Towards the top of the sequence, carbonate rocks are preserved as spectacular stromatolite domes (figure 6C). Such sequences in locations around the world have been interpreted to reflect deglaciation and associated flooding of continental shelves while ice sheets melted. The lack of an erosion surface between the conglomerates and the dolomite indicates a relatively rapid transition between siliciclastic and carbonate deposition. However in order to understand Neoproterozoic climatic processes we need absolute time constraints.

The study area is essentially undeformed and stratigraphic relationships show that the oldest unit in the region is the Independence Fjord group, a sequence of sedimentary rocks over 2 km thick. This group comprises deltaic cross bedded, arkosic and abundantly ripple-marked sandstones (figure 6D). It is interpreted to overlie ‘crystalline basement’ which, however, is not exposed in the region. Age constraints on the Independence Fjord group are sparse, with only a lower age limit provided by the approx. 1 380 Ma intrusive Zig-Zag Dal basalt suite (Kalsbeek and Jepsen, 1984). Studies of detrital minerals in this group will provide further age constraints and also test the hypothesis that the Morænesø Formation was locally derived.

The youngest unit investigated was the Portfjeld Formation consisting of 500–700 m of dominantly carbonate rocks. Sitting unconformably above the Portfjeld is the Early Cambrian Buen formation (figure 5). We sampled sandstone intervals within the Portfjeld formation, as well as a possible ash bed which may provide an absolute time marker for the sequence. The Portfjeld formation was considered to be Early Cambrian in age and correlated to the Ella Bay Formation of Ellesmere Island (Peel and Christie, 1982). The Ella Bay formation is one of the oldest stratigraphic units exposed in the Franklinian margin sedimentary sequence in the Canadian Arctic Islands, but is now considered to have a probable late Neoproterozoic age (Dewing et al., 2004). Constraint from detrital zircon crystals will help clarify the depositional timing of the Portfjeld and its correlative formations.

Concluding remarks

Circum-Arctic projects cover a very large region and because we generally target a single location per year, it will take many years to visit each important location necessary to understand the growth and development of the Arctic Ocean. The International Polar Year 2007–2008 (IPY) is designed to increase awareness of the importance of the Earth’s polar regions (the Arctic and Antarctica), which are significant not only for societal resources (petroleum, fishing, shipping lanes, etc.), but also for the planet’s biodiversity, environmental monitoring, etc. The circum-Arctic region will continue to be a focus of our research during the IPY, when we hope to visit Axel Heiberg Island, as well as the Franz Josef Land Islands of the Russian Arctic.