Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

journal homepage: www.elsevier.com/locate/palaeo

The Early Barents Sea Sill Complex: Distribution, 40Ar/39Ar geochronology, and implications for carbon gas formation

Stéphane Polteau a,⁎,BartW.H.Hendriksb,1, Sverre Planke a,c, Morgan Ganerød b, Fernando Corfu c,d, Jan Inge Faleide c,d,IvarMidtkandald,HenrikS.Svensenc,ReidunMyklebuste a Volcanic Basin Petroleum Research, Oslo Innovation Center, Gaustadalléen 21, 0349 Oslo, Norway b Geological Survey of Norway, Postboks 6315 Sluppen, 7491 Trondheim, Norway c The Center for Earth Evolution and Dynamics, University of Oslo, PO Box 1028 Blindern, 0315 Oslo, Norway d Department of Geosciences, University of Oslo, PO Box 1047, Blindern, 0316 Oslo, Norway e TGS, Lensmannslia 4, Asker, Norway article info abstract

Article history: Mafic igneous rocks of Cretaceous age (80–130 Ma) scattered around the are commonly referred to Received 3 March 2015 as the High Arctic Large Igneous Province (HALIP). We have mapped out the distribution of HALIP igneous rocks Received in revised form 3 July 2015 in the Barents Sea region over the past decade based on integrated seismic–gravity–magnetic interpretation, Accepted 8 July 2015 field work, review of publications, and analyses of new and vintage borehole and field samples. The mapping re- Available online 7 August 2015 veals abundant igneous rocks in the northern and eastern Barents Sea covering an area of ~900,000 km2 with a 3 Keywords: conservative volume estimate of 100,000 to 200,000 km of intrusions. The igneous province is dominated by HALIP sheet intrusions injected into Triassic and Permian sedimentary rocks. Hydrothermal vent complexes are rare, Large igneous provinces and only two potential vent complexes have been identified on seismic data in the eastern Barents Sea. We Cretaceous paleoclimate have further done extensive radiometric dating of the igneous samples in the Barents Sea region. New Barents Sea 40Ar/39Ar dating of thirteen samples from Svalbard reveal ages of crystallization and alteration. The large age 40Ar/39 Ar chronology span (60–140 Ma for the raw ages) is likely due to partial or complete overprint of the K/Ar system in plagioclase, Anoxic event and the age of the magma emplacement is better represented by U/Pb TIMS ages. Only one of our 40Ar/39Ar anal- yses of plagioclase yielded a statistically valid age that is in line with the recently published U/Pb TIMS ages of 122–125 Ma. The new data clearly document that relying on published data from the K/Ar system can lead to er- roneous conclusions on the age of crystallization in this province without a careful use of additional 40Ar/39Ar degassing data (i.e., K/Ca). We propose that the magmatism on Svalbard and Franz Josef Land represents a dis- tinct magmatic event near the Barremian/Aptian boundary (125 Ma) in the Barents Sea. This Early Cretaceous Barents Sea magmatism resulted in the formation of the BSSC (Barents Sea Sill Complex). BSSC age rocks are also present in Arctic Canada (Sverdrup Basin) and on Bennett Island ( Islands). The massive injec- tion of hot magma into potentially organic-rich sediments in the eastern and northern Barents Basin caused rapid organic matter maturation and formation of thermogenic gas and oil in contact aureoles. We estimate that up to 20,000 Gt of carbon were potentially mobilized, corresponding to 175 trillion barrels of oil equivalent. The production rates and fate of the carbon gases are uncertain. However, we speculate that rapid release of au- reole greenhouse gases (methane) may have triggered the Oceanic Anoxic Event 1a (OAE1a) and the associated negative δ13C excursion in the Early Aptian. Some of the methane may also be trapped in the vast hydrocarbon gas accumulations found in the east Barents Basin. © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

1. Introduction epicontinental seas, supercontinental fragmentation, and several epi- sodes of ocean anoxia (Fluteau et al., 2007; Hay, 2008; He et al., 2008; The Cretaceous is a remarkable period in the Phanerozoic with gen- Keller, 2008). Several large igneous provinces (LIPs) were also emplaced erally warm climates, high sea-level associated with large shallow during this period, e.g., Parana-Etendeka and the Ontong Java oceanic plateau in the southern hemisphere, and the High Arctic LIP (HALIP) in northern hemisphere (Buchan and Ernst, 2006; CoffinandEldholm, ⁎ Corresponding author. 1994; Coffin and Eldhom, 1992; Ernst, 2014; Janasi et al., 2011). E-mail address: [email protected] (S. Polteau). 1 fi Present address: Statoil Forskningssenter, Arkitekt Ebbells veg 10, 7053 Ranheim, When Tarduno et al. (1998) rst coined the term HALIP, they consid- Norway. ered the Late Cretaceous volcanic event on Axel Heiberg Island in the

http://dx.doi.org/10.1016/j.palaeo.2015.07.007 0031-0182/© 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 84 S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95

Sverdrup Basin, Arctic Canada. Later, the HALIP included lava flows, sills methods (Corfu et al., 2013) on samples sometimes from the same sill. and dykes identified in Svalbard, northern Greenland, and the Sverdrup The data reveal an Early Cretaceous sill complex in the Barents Sea, re- Basin (see Fig. 1, Maher, 2001). Senger et al. (2014b) recently published ferred to as BSSC (Barents Sea Sill Complex), and distinct in age from a comprehensive review of the HALIP magmatism in Svalbard. The rec- the Late Cretaceous volcanics dominantly found in Arctic Canada and ognition of HALIP-related rocks is more uncertain in offshore areas Northern Greenland. We suggest that this igneous event had a major (Grogan et al., 1998), but the biggest challenge similar to other well- impact on the Barents Sea basin development, on the evolution of petro- known LIPs is the determination of the accurate age and duration of leum systems, and possibly triggering the global paleoenvironmental the magmatic event. Radiometric dating of HALIP rocks indicates ages changes in the Early Cretaceous. between 130 and 80 Ma (40Ar/39Ar) or 192–34 Ma (K/Ar) (see Table 1, Buchan and Ernst, 2006; Estrada and Henjes-Kunst, 2013; 2. Data and methods Evenchick et al., 2015; Nejbert et al., 2011). These geochronology results suggest an extended period of magmatism punctuated by several volca- The main data available for this study are summarized below and nic pulses (Fig. 1 for distribution and Table 1 for age summary; Buchan shown in Fig. 2. A very dense grid of industry 2D-seismic data was avail- and Ernst, 2006). However, the robustness of the results can be able in the southwest Barents Sea, whereas an open grid of vintage and questioned when considering published K/Ar ages, which give a recent seismic data was interpreted in the eastern Barents Sea. Only po- 50 Ma difference between samples from the same sill intrusion tential field data were available in the formerly disputed zone between (Birkenmajer et al., 2010; Corfu et al., 2013; Nejbert et al., 2011). Norway and Russia. Dolerite samples were collected during Svalbard The aims of this publication are (1) to document the distribution, field work in 2004 and 2010. Additional samples were obtained from volume, and age of Cretaceous igneous rocks in the Barents Sea and existing collections and boreholes. (2) to briefly assess the impact of the Cretaceous magmatism on the Ba- rents Sea basins, petroleum system, and paleoclimate. We have used a 2.1. Geophysical data and interpretation method variety of industry and published geological and geophysical data to map the distribution of the igneous rocks. In addition, new samples 2.1.1. Seismic, gravity, and magnetic data were obtained from outcrops and wells on Svalbard and in Russia to The multi-channel seismic database consists of more than date the rocks using modern 40Ar/39Ar (this contribution) and U/Pb 300,000 km of line data in the western Barents Sea and about

Fig. 1. Regional setting of the Barents Sea within the North Atlantic–Arctic region. The Barents Sea is a shallow platform area surrounded by deep ocean basins in the NE Atlantic and the Arctic. Ages of major Cretaceous igneous complexes from Table 1. GR: Gakkel Ridge; MJR: Morris Jessup Rise; NSI: ; SZ: Severnaya Zemlya; YP: Yermak Plateau. Bathymetry from IBCAO (International Bathymetric Chart of the Arctic Ocean). S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95 85

Table 1 Summary of published ages and distribution of HALIP rocks (see also Fig. 1). (1) Midtkandal et al. (2008),(2)Dallmann (1999),(3)Grogan et al. (1998),(4)Harland (1997),(5)Parker (1967),(6)Dibner (1998),(7)Koryakin and Shipilov (2009),(8)Filatova (2007),(9)Shipilov et al. (2009), (10) Fedorov et al. (2005),(11)Drashev and Saunders (2003), (12) Kos'ko and Korago (2009),(13)Villeneuve and Williamson (2006),(14)Tarduno et al. (1998),(15)Kowallis et al. (1995),(16)Trettin and Parrish (1987),(17)Williamson (1988), (18) Jokat (2003), (19) Buchan and Ernst (2006),(20)Corfu et al. (2010), (21) Corfu et al. (2013).

Province Region Ages (Ma)

Igneous Rocks Sediments enclosing lava//bentonite References

Barents Sea Svalbard U/Pb: 124.5-124.7 on two onshore sills U/Pb: 123.3 ±0.2 Ma on bentonite in the Barremian [1-3, 21] Sills offshore not above Late Jurassic Helvetiafjellet Fm. Kong Karls Land Sills offshore not above Late Jurassic in Barremian Kong Karls Land Fm. [3-5] Franz Josef Land U/Pb. 122.2 ±1.1 Ma on sill Basalts in Barremian Tikhaya Bay Fm. [3,6,7,21] Ar/Ar. 125.2 ±5.5 Ma on dyke K/Ar: 34-175 Ma on sills K/Ar: 94-192 Ma on dykes Sills offshore not above Late Jurassic

New Siberian Bennett Island K/Ar: 106-124 Ma on basalts Basalts in Barremian unnamed formation [8-12] Islands

Arctic Canada Axel Heiberg Island Ar/Ar: 95.3 ±2.0 on basalts Basalts in the Albian-Cenomanian Strand Fjord Fm. [13-16] Sverdrup Basin Ar/Ar: 126-129 Ma on sills Basalts underlying the Turonian Kanguk Fm. Ellesmere Island U/Pb: 92.0 ±1.0 Ma on gabbroic intrusion Basalts in the Albian-Cenomanian Hassel Fm. [13, 16, 17] Ar/Ar: 94-97.2 Ma on two sills Ar/Ar: 92.0 ±2.0 on basalts in Hassel Fm. Queen Elizabeth Ar/Ar: 127-129 Ma on the Lightfoot River Dyke [13] Islands Swarm Ar/Ar: 80.7-96.1 Ma on basalts

Arctic Ocean Alpha Ridge Ar/Ar: 82.0 ±1.0 Ma on a single altered dredged [18] sample

Northern Peary Land Rb/Sr and Ar/Ar: 82-103 Ma Dykes cross-cut Early Cretaceous strata [19] Greenland

Novaya Zemlya Timanides U/Pb: 704-716 on sill Devonian sediments re-interpreted as Precambrian azoic [20] succession

25,000 km in the eastern Barents Sea (Fig. 2). Ship-track gravity data (DB5 and DB9), Hatten (DB47 and DB51), Gåsøyane (DB31), and were available for most of the seismic profiles in the western Barents Gipsvika (DB38) from the paleomagnetic samples of E. Halvorsen (see Sea (287,000 km). Bouguer corrected 200-km high-pass filtered http://toposvalbard.npolar.no/ for a detailed topography map of satellite-derived gravity data were available in the entire study area Svalbard). The Diabasodden and Hatten localities are 3 km apart and (Sandwell and Smith, 2009). The aeromagnetic compilation is based are from the same sill outcropping 15 km northeast of Longyearbyen, on 252,000 km of line and public-domain gridded magnetic data cover- whereas the Gåsøyane and Gipsvika localities are 10 km north on the ing the entire study area (Verhoef et al., 1996). opposite side of Isfjorden. Previously, five sill localities were visited in Svalbard during a field expedition in April 2010 (samples 2A, 4A, 5C, 6D, 7B). Sample SV04-04 was collected in 2004 from the Dunerbukta 2.1.2. Geophysical interpretation method area. Sample DH4 is a dolerite from 951.2 m depth in the Longyearbyen The seismic interpretation of igneous complexes follows the CO Lab borehole DH4 drilled 4.5 km southeast of Longyearbyen. The methods of seismic volcanostratigraphy for extrusive rocks (Planke 2 field expeditions targeted coarse-grained and pegmatitic dolerites et al., 2014; Planke et al., 2000) and Planke et al. (2005) for intrusive primarily for zircon U/Pb dating and 40Ar/39Ar dating for comparing complexes. The depth to magnetic sources were calculated based on Pe- method results. All collected samples were mainly medium grained. ters' half-slope method (Peters, 1949). Sorting of the source depths The coordinates of the sampling sites are shown in Table 2 and locations while simultaneously checking geological constraints was done during in Fig. 2. The results of the U/Pb TIMS dating are reported in Corfu et al. this process (Grogan et al., 1998; Myklebust, 1994). Furthermore, the (2013). ability to correct for strike direction of the magnetic body is important during the processing. The integrated interpretation of the seismic, gravity, and magnetic data was mainly done using the Kingdom Suite 2.2.2. Analytical protocol seismic interpretation system. High-pass filtered gravity and magnetic The samples were crushed and sieved to isolate grains of data were imported as pseudo-horizons and represent density and 180–250 μm. A portion was set aside for whole rock analysis. Magnetic magnetic attribute maps. The SGM increases the efficiency and quality separation using a Frantz isodynamic separator was followed by heavy of seismic interpretation by facilitating the identification of sedimentary liquid separation to concentrate feldspars. Visual inspection using a basins, basement highs, geological trends, and lithological discrimination, binocular microscope showed moderate degree of alteration in most including increased confidence of the interpretation of igneous rocks. samples, with a mix of altered and fresh plagioclase. The groundmass had an altered character. Fresh inclusion-free mineral grains were 2.2. 40Ar/39Ar geochronology handpicked under the binocular microscope. The samples were washed in 5 M HCl for five minutes at room temperature in an ultrasonic bath 2.2.1. Sampling to remove carbonate and light surface alteration, and subsequently Radiometric dating of dolerites is commonly done using the washed ultrasonically several times in distilled water. 40Ar/39Ar method on whole rock, groundmass, or plagioclase. All collect- The transformation 39 K(n, p)39Ar was performed during irradiation ed samples on Svalbard are from the Diabasodden Suite (Dallmann, in three separate batches: samples 2A–DH4 (2010, IFE in Norway), sam- 1999). We selected seven medium-grained samples from Diabasodden ples DB5–DB51 (2009, IFE), and sample SV04-04 (2005, McMaster in 86 S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95

Fig. 2. Barents Sea bathymetry map showing onshore and offshore distribution of Cretaceous igneous rocks, selected offshore hydrocarbon fields and wells, and interpreted seismic profiles in the Eastern Barents Sea.

Canada). Samples irradiated in 2009 and 2010 have been calculated A plateau is defined according to the following requirements: at against the Taylor Creek Rhyolite standard (28.619 ± 0.034 Ma), least three consecutive steps, each within 95% confidence level, com- which is calibrated directly to U/Pb (Renne et al., 2010). The Tinto bio- prising at least 50% of total 39Ar and mean square of weighted deviates tite (410.3 Ma, but not directly calibrated to U/Pb) was used as a fluence (MSWD) less than the two tailed student T critical value. We calculated monitor standard for SV04-04. The 40Ar/39Ar analytical protocol follow- a weighted mean plateau age (WMPA), weighting by the inverse of the ed by the NGU lab is thoroughly presented in Ganerød et al. (2011). variance. The weighted York-2 method was used to calculate the S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95 87

Table 2 Sample locations and main 40Ar/39Ar results. The superscripts P or WR in the sample column denotes Plagioclase separates or whole rock. 39Ar% is the percentage of the total 39Ar used in the age calculation. Uncertainties on the age are reported as analytical and external errors. MSWD is the mean square weighted deviation for the steps included. TGA denotes Total Gas Age. 39 37 40 36 K/Ca is calculated from Ark/ ArCa. Intercept is the trapped Ar/ Ar ratio from inverse isochron analysis. The bold sample names highlights the results where ages have a high confidence for statistically valid plateau and isochron ages.

Sample Lat. Long. Spectrum analysis MSWD TGA K/Ca ±1.96σ Inverse isochron analysis

39Ar % Steps (N) Age ±1.96σ Age ± 1.96σ MSWD Intercept ± 1.96σ

2AP 78.35 16.26 83.48 2–4 (3) 75.44 1.47/1.52 1.24 76.85 ± 1.59 0.181 ± 0.010 81.5 ± 6.7 0.292 219.7 ± 87.3 4AP 78.14 18.78 74.3 1–7 (7) 120.22 1.79/1.91 0.873 121.23 ± 2.11 0.023 ± 0.001 121.0 ± 2.0 0.610 288.5 ± 13.0 5CP 78.16 18.93 61.4 1–7 (7) 109.34 5.48/5.5 1.984 116 ± 3.82 0.027 ± 0.001 110.9 ± 6.4 2.145 289.7 ± 22.7 6DP 78.22 17.88 97.43 3–8 (6) 82.64 0.63/0.76 0.937 82.3 ± 0.82 2.770 ± 0.1100 82.9 ± 1.0 1.067 290.4 ± 23.5 7BP 78.03 13.88 87.64 3–6 (4) 77.72 3.76/3.78 12.62 80.2 ± 1.46 0.430 ± 0.020 83.2 ± 3.0 2.873 205.1 ± 42.5 7BP 78.03 13.88 7.7 8–16 (9) 115.1 14.21/14.22 3.715 “ 0.067 ± 0.003 113.6 ± 14.6 1.307 395.3 ± 140.0 DH4P 78.20 15.83 69.11 1–3 (3) 60.68 2.72/2.74 5.228 73.87 ± 1.57 0.150 ± 0.008 62.6 ± 1.8 1.363 262.6 ± 24.5 DH4P 78.20 15.83 10.32 5–8 (4) 110.96 6.67/6.7 1.066 “ 0.029 ± 0.002 113.3 ± 7.8 1.410 222.8 ± 217.5 DB31WR 78.46 16.28 33.76 3–6 (4) 119.42 2.24/2.53 1.787 105.45 ± 1.54 0.520 ± 0.033 119.4 ± 5.1 2.625 299.7 ± 44.1 DB38WR 78.45 16.40 32.38 2–6 (5) 118.07 4.31/4.49 5.311 102.0 ± 1.9 0.550 ± 0.032 112.8 ± 10.6 5.058 329.5 ± 55.8 DB38WR 78.45 16.40 17.67 2–4 (3) 122.88 4.67/4.82 1.979 " 0.520 ± 0.040 125.6 ± 17.1 3.547 288.30 ± 63.2 DB5WR 78.36 16.13 100 1–11 (11) 101.05 ± 1.55 0.440 ± 0.033 DB9WR 78.36 16.13 50.35 3–7 (5) 131.11 3.26/3.55 1.78 122.16 ± 2.31 0.510 ± 0.030 136.3 ± 11.9 1.922 277.4 ± 47.7 DB47WR 78.36 16.99 78.36 4–9 (6) 124.09 3.58/3.81 10.30 125.11 ± 1.82 0.470 ± 0.025 129.3 ± 5.1 6.031 216.3 ± 63.1 DB51WR 78.36 16.99 36.99 1–5 (5) 123.8 5.20/5.36 3.61 107.64 ± 2.12 0.260 ± 0.017 128.4 ± 5.7 1.950 275.7 ± 21.2 SV04-04P 78.23 16.28 78.49 4,5,7,8 (4) 128.39 2.44/2.81 0.348 120.13 ± 6.0 0.030 ± 0.0010 127.9 ± 3.7 0.437 300.9 ± 10.9

inverse isochron results, where a valid isochron has an MSWD value less Triassic, Permian, Carboniferous, and Devonian sequences (Ohm et al., than the two tailed F-test critical value. 2008).

4. Results 3. Geological framework 4.1. Extent of igneous rocks The Barents Sea is a more than 1.5 million km2 large epicontinental sea, including the Svalbard and Franz Josef Land archipelagos in the The areal distribution of Mesozoic igneous rocks in the Barents Sea is north and Bear Island (Bjørnøya) in the west. The average water shown in Fig. 3. The map is compiled from new seismic, gravity, and depth is about 250 m, with wide glacial channels up to 450 m deep. magnetic interpretation, borehole data, field data, and adjusted using The Barents Sea consists of a complex system of rift basins and base- previous published local studies (Grogan et al., 1998; Kos'ko and ment highs in the west, which are separated by the Central Barents Korago, 2009; Maher, 2001; Minakov et al., 2012; Myklebust, 1994; Monocline from an ultra-deep north-trending sag basin in the east. Senger et al., 2014b; Werner et al., 2011). The confidence of the inter- The geological history was recently summarized by Smelror et al. pretation is high onshore and in the Southeast Barents Basin where seis- (2009) and Henriksen et al. (2011b). The basin history started near mic imaging is very good. The confidence is moderate north of 74° in the the end of the Caledonian orogeny with formation of Devonian and Norwegian segment of the Barents Sea toward Svalbard and poor north Carboniferous clastic rift basins. This period was followed by the forma- of 76° in the east Barents Basin where seismic data are less abundant or tion of predominantly shallow water evaporitic basins in the middle of lower quality. Carboniferous to the middle Permian, including the deposition of thick Outcrops of Mesozoic igneous rocks from the Diabasodden Suite are carbonate and salt successions. This interval forms a karstified widespread on Svalbard and Franz Josef Land (Figs. 2 and 3). Extrusive carbonate reservoir saturated in oil in the “Gotha” discovery in the basaltic rocks are abundant on Franz Josef Land (Dibner, 1998; southern part of the Loppa Ridge south of the Hoop Area. Middle Perm- Evdokimov and Stolbov, 2004), whereas Mesozoic basaltic lavas are ian uplift of the Uralian mountains to the east led to the deposition of only found in Kong Karls Land on Svalbard (Bailey and Rasmussen, prograding siliciclastic sediments in the Southeast Barents Basin and 1997; Grogan et al., 1998; Harland, 1997). Basaltic sills are common the formation of an unconformity in the west. Renewed uplift in the both on Svalbard (Dallmann, 1999; Maher, 2001; Nejbert et al., 2011; east at end-Permian times, possibly associated with the Siberian Traps, Shipilov et al., 2009) and Franz Josef Land (Dibner, 1998). Numerous led to the formation of a major clastic delta prograding across the plat- sills have been intersected by three boreholes on Franz Josef Land form from southeast to northwest during the Triassic. Marine clastic (Dibner, 1998)andfive wells in Svalbard (Senger et al., 2014a; Senger mudstones were subsequently deposited in a shallow sea during the et al., 2014b; Skola et al., 1980). In outcrop, the sills are normally less Jurassic and earliest Cretaceous, whereas deep Late Jurassic–Early Creta- than 50 m thick and mainly intrude Triassic clastic sequences. A north- ceous rift basins were formed in the southwest. This period was follow- west trending dike swarm has been mapped on Franz Josef Land ed by massive Arctic volcanism in the Early Cretaceous with the (Buchan and Ernst, 2006; Døssing et al., 2013), whereas dikes are rare formation of the Diabasodden Suite (e.g., Dallmann, 1999; Senger on Svalbard (Fig. 3). et al., 2014b), causing uplift of the northern margin and formation of The areal extent of the igneous rocks is also difficult to map in the southward-prograding delta sequences. Most of the Barents Sea was regions around Svalbard and Franz Josef Land. The distribution of igne- high standing in the Late Cretaceous and Cenozoic, whereas continental ous intrusive and extrusive rocks offshore Svalbard has been interpreted breakup to the north and west formed Cenozoic passive continental by Grogan et al. (1998) and Minakov et al. (2012) based on seismic re- margins along the Eurasia and Norway basins. The dominantly Plio- flection and seismic wide-angle data combined with magnetic data. Pleistocene glaciations led to episodic large-scale uplift and erosion of Well-defined sill reflections are identified in Permian to Jurassic age se- the Barents Sea (Henriksen et al., 2011a), with the deposition of major quences, locally terminating or showing a hummocky relief at the BCU. glacial fans along the northern and western margins. The main petro- Erosional igneous remnants are locally identified on the seafloor, com- leum source rock is the Upper Jurassic (Hekkingen Formation), but monly preventing deeper seismic imaging. Finally, volcanic flows are organic-rich rocks are also locally present in Paleogene, Cretaceous, identified on and around Kong Karls Land, and magnetic data clearly 88 S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95

Fig. 3. The 200-km high-pass filtered Bouguer anomaly map combined with the distribution of magnetic anomalies grouped by depths from Myklebust (1994). The region with interpreted Early Cretaceous BSSC is shown as a dark grey transparent polygon, but also contains some flows and dykes. High confidence area for the presence of volcanic rocks correspond to Svalbard (130,000 km2), Franz Josef Land (150,000 km2), and SE Barents Sea (180,000 km2). Satellite gravity data from Wessel and Smith (1991). reveal a north-trending dike swarm (Døssing et al., 2013). In the North- such as the Stockman field. Sill intrusions are interpreted from seismic east Barents Basin, the extent of the igneous complex is delineated by the data as high-amplitude, crosscutting and sometimes saucer-shaped re- potential field data calibrated by published seismic profiles of flections (Figs. 4–6). The sills are mainly present in the central part of Khlebnikov et al. (2011). the basin and are rarely identified in structural highs or inverted basin Seismic mapping revealed a very extensive sill complex of more than segments along the basin flanks. In addition, sills are interpreted not 180,000 km2 in the east Barents Basin, underlying giant gas discoveries only in the Triassic sequence but also in the Upper Permian stratigraphic S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95 89

Fig. 4. Seismic profile across the deep Southeast Barents Basin showing an extensive igneous sill complex. The sill intrusions are absent from the structural highs on the flanks of the basin. Prominent prograding reflections onlap sequences just above the base Cretaceous unconformity (BCU) and represent evidence of uplift, erosion, and transport of sediments during the Early Cretaceous times. P: Permian, T: Triassic, J: Jurassic, C: Cretaceous. Sequence boundaries modified from Henriksen et al. (2011b).SeeFigs. 2 and 3 for location. Data courtesy of TGS. intervals. Up to 7–8 levels of sills are imaged, and two different sill facies traced east beyond Franz Josef Land. (2) Intra-sedimentary magnetic have been mapped on a regional seismic grid (Fig. 4): the deep, domi- sources were found extensively in Paleozoic rocks in the western Barents nantly layer-parallel, and shallower transgressive and saucer-shaped Sea. In the eastern Barents Sea, intra-sedimentary magnetic sources are reflections. The extent of the two facies is similar, with the deep sills typically found in Triassic sequences. (3) Several high-amplitude mag- extending only a few kilometers outside the shallow sills along the netic anomalies, earlier interpreted as basement, have been re-defined western and eastern flanks of the basin. However, the deep sills extend as intra-sedimentary boundaries in Paleozoic sequences. for up to 100 km south of the shallow sills in the Southeast Barents Basin The contoured depth to magnetic sources is shown on top of the (Fig. 3). At least one offshore borehole in the east Barents Basin has pen- sill distribution interpreted from the seismic data in Fig. 3.Ingeneral, etrated two dolerites sills (Ludlow well; Corfu et al., 2013; Komarnitskii the distribution of intra-sedimentary magnetic sources (b4km)cor- and Shipilov, 1991). relates well with the outline of the sill facies in the Southeast Barents Very few seismic profiles were available in the Northeast Barents Basin. However, some shallow magnetic anomalies are also present Basin in the area east and south of Franz Josef Land. The available outside the sill domain, in particular, northwest of Murmansk and profiles are difficult to interpret, with an almost opaque top volcanic re- along the western flank of Novaya Zemlya. These magnetic anoma- flection, which could alternatively represent shallow sills or volcanic lies may have a different origin than intrusive igneous rocks or be re- flows. The magnetic data reveal distinct northwest–southeast trending lated to igneous rocks not identified/interpreted on the seismic data directions in this region, interpreted as representing the signature of a (e.g., near vertical dikes) due to limited seismic coverage. Finally, the possible dike swarm (Fig. 3). trends of anomalies in the 200-km Bouguer data correlate very well The vertical distribution of the igneous rocks was determined by with the distribution of the sill complexes. Two broad, low- correlating the magnetic data with regional seismic profiles, effectively amplitude, and north-trending positive gravity anomalies corre- allowing the separation of magnetic source depths into three strati- spond to the mapped extent of the sill complexes for more than graphic levels (Fig. 3). (1) High-frequency magnetic anomalies related 600 km (Fig. 3). to intrusive and extrusive rocks are closely correlated with Mesozoic Hydrothermal vent complexes are common in many volcanic basins, subcrop patterns. The distribution of this magnetic signature has been e.g., the Vøring Basin of mid-Norway (Planke et al., 2005), onshore in 90 S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95

Fig. 5. Seismic profile in the east-central Barents Sea showing an extensive sill complex and an interpreted crater structure at the Base Aptian level. See Figs. 2 and 3 for location. Data courtesy of TGS. the Karoo Basin (Svensen et al., 2008), and Tunguska Basin in Siberia Sample 4A (Fig. 8B) defines a weighted mean plateau age of 120.2 ± (Svensen et al., 2009). One potential hydrothermal vent complex has 1.9 Ma (2σ). The K/Ca ratio varies around 0.02 throughout the spectrum, been identified in the east Barents Basin with its characteristic crater- suggesting the age is representative of the crystallization age of plagio- shaped structure near the interpreted Base Aptian level (Fig. 5). The clase. The 40Ar/39Ar age for this sample is comparable with the U/Pb crater is present above faulted Jurassic sedimentary rocks and the termi- ages from Svalbard documented by Corfu et al. (2013). nation of transgressive sill segments. Another crater-shaped structure Two of our plagioclase samples (2A and 6D, Figs. 8Band8C) pro- has been interpreted at a similar stratigraphic level above the termina- duced statistically valid plateau ages (Table 2), varying from 75.4 ± tion of a deep transgressive sill further south (Fig. 6). However, the re- 1.5 to 82.6 ± 0.8 Ma. However, the K/Ca ratio (0.18–2.8) is higher flections beneath both craters are not discontinuous but parallel to the than would be expected from unaltered plagioclase in both of base crater reflection. Such good reflection continuity is uncommon these samples (2A and 2D). Such high K/Ca ratio indicates that a for typical hydrothermal vent complexes mapped elsewhere. Senger higher K and/or lower Ca mineral is degassing in addition to plagio- et al. (2013) identified another hydrothermal vent complex in central clase. Sericite [KAl3Si3O10(OH,F)2] is known to replace plagioclase Isfjorden based on magnetic, seismic, and multibeam bathymetry data. by mineralogical substitution and/or fills microfractures within pla- gioclase crystals. This substitution process may occur at low temper- ature conditions (b300 °C), for example, during hydrothermal 40 39 4.2. Ar/ Ar geochronology alteration (Verati and Jourdan, 2014). Given the high K2Ocontentof sericite (~10 wt%), pervasive sericite replacement will override the The 40Ar/39Ar results are summarized in Table 2. The raw data tables low K (b0.1 %) signal from the plagioclase, and the obtained age may (corrected for blanks and baselines), spectrum, and inverse isochron represent a sericitization event. We therefore conclude that the ages plots are presented in Appendices A and B. of 75.4 ± 1.5 to 82.6 ± 0.8 Ma are most likely representative of alter- Samples 2A, 4A, and 6D yield statistically valid plateau and isochron ation, rather than of the original emplacement age of our samples. ages for more than 50% of the cumulative 39Ar released. In all other sam- At the high-temperature end of both spectra, K/Ca ratio drops are asso- ples, the K/Ar system clearly has been disturbed and these data are ciated with climbing step ages. At the high-temperature end of the spec- therefore included as background documentation in the Appendices trum (7.7% of the cumulative 39Ar), the K/Ca ratio varies around 0.067 and not discussed in detail. for sample 7B (Appendix A). These steps produce a weighted mean age S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95 91

Fig. 6. Seismic profile across the southeast Barents Sea showing an extensive sill complex and an example of a possible hydrothermal vent complex. The vent is rooted near the tip of a deep sill intrusion, and the crater terminates at the interpreted Base Aptian horizon representing the paleosurface during BSSC magmatism. See Figs. 2 and 3 for location. Data courtesy of TGS. of 113.6 ± 14.6 Ma (2σ), i.e., overlapping with the 40Ar/39Ar age of plagio- based on the integrated interpretation of seismic, gravity, and magnetic clase sample 4A. The high-temperature step ages for samples 2A, 6D, and data. Accurate volume estimates are much more difficult to determine 7B may relate to the age of crystallization, but due to low gas volume left as both the thickness and number of sill intrusions are commonly very in the sample, statistically valid age cannot be calculated from these steps. difficult to interpret from seismic data. The volume of intrusive rocks is calculated separately for four differ- 5. Discussion ent provinces: (1) the Southeast Barents Basin to about 77° north, (2) the Northeast Barents Basin and Franz Josef Land, (3) the eastern 5.1. BSSC distribution and volume Svalbard and the surrounding offshore shelf, and (4) the remaining platform and formerly disputed areas. The average number of sills in The mapped areal extent of Mesozoic sill intrusions in the Barents each province is estimated from seismic, outcrop, and well data, but Sea is 900,000 km2 (transparent dark grey polygon in Fig. 3)andis the uncertainty is fairly high due to limited data, imaging problems, and scarce outcrops. Well-defined high-amplitude sill reflections typi- Table 3 cally originate from units N50 m thick (Planke et al., 2005), and we Estimated volumes of BSSC. Minimum estimates correspond to the minimum sill layers, have used 50 m as a conservative estimate of the thickness of one sill and maximum volumes for maximum number of sill layers. layer in the BSSC. The estimated volume of intrusive igneous rocks is 3 Province Area, Interpretational Number of Min. volume, Max. volume, in the range from 103,000 to 206,000 km (Table 3). km2 confidence sill layers a km3 km3

1 180,000 High 4–8 36,000 72,000 2 310,000 Moderate 3–6 46,500 93,000 5.2. Age of the BSSC 3 130,000 Moderate 1–2 6,500 13,000 4 280,000 Low 1–2 14,000 28,000 Our 13 samples of intrusive rocks from Svalbard are located along Total 900,000 103,000 206,000 an east–west profile from Festningen to Agarddalen (Fig. 7). The dol- a Average sill thickness of 50 m. erites intrude upper Paleozoic (Linnévatn; Gipshuken), Triassic 92 S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95

Fig. 7. Compilation of new and published radiometric ages along a west–east profile across Svalbard from Festningen to Agardhdalen, and from Franz Josef Land. The 40Ar/39Ar isochron ages are listed in Table 2. The vertical light red and yellow boxes indicate the range of ages for the crystallization and alteration events, respectively. The names in bold indicate statistically reliable 40Ar/39Ar ages. Note that Linnevatn is about 5 km south of Festningen, the Gipshuken/Gåsøyane localities (DB31 and DB38) are about 10 km north of Diabasodden, and the Domen (SV04-04) dolerite was sampled about 15 km north of Agardhdalen. U/Pb TIMS ages (Corfu et al., 2013) are shown. Sample 7A yields two Concordia ages of 124.8 ± 0.4 Ma (zircon + rutile) and 123.9 ± 0.3 Ma (titanite). Sample 1AX has an age of 124.6 ± 0.3 Ma (zircon), whereas the Severnaja age on baddeleyite is 121.5 ± 0.3 Ma, which is likely 1–2 Myr too young due to Pb loss from the platy mineral. Samples DH3 and DH4 are from Longyearbyen CO2 Lab boreholes. Sample DH3 is from a thin silicic tuff in the Helvetiafjell Fm. with an age of 123.3 ± 0.3 Ma (zircon), whereas sample DH4 is from a 2.3 m thick dolerite at 950 m depth. Geological profile modified from Dallmann et al. (2011). Sample DH7 is from Midtkandal et al. (2014).

(Longyearbyen,Diabasodden,Sassendalen), and Jurassic sequences and the K/Ca ratio throughout the entire spectrum. This age of 120.2 ± (Agardh). The westernmost samples are from strongly folded sedi- 1.9 Ma closely approximates the U/Pb TIMS data of Corfu et al. (2013) ments, whereas thrust faults are abundant in the Longyearbyen– for Svalbard, which define a narrow range of ca. 123–125 Ma. Our best Diabasodden region. The 40Ar/39Ar age that reflects crystallization estimate for the age of the Svalbard sills, based on both 40Ar/39Ar and and the two ages reflecting alteration (Fig. 8)areplottedwiththe U/Pb data, thus spans from 120.2 ± 1.9 Ma to 124.7 ± 0.3 Ma. new U/Pb TIMS ages from Corfu et al. (2013) and Midtkandal et al. Alternatively, biostratigraphy can be used to indirectly date (2014) in Fig. 7. extrusive rocks. Sedimentary rocks within and enclosing lava flows on There is a large spread in the raw 40Ar/39Ar ages (Table 2), spanning Kong Karls Land and Franz Josef Land suggest an Early Cretaceous from 135 to 60 Ma, with two peaks of about 130–120 Ma (Early Creta- (Barremian) age of the volcanic rocks (Dibner, 1998; Smith et al., 1976). ceous) and 85–80 Ma (Late Cretaceous). The large spread is in general Additional evidence pointing toward an Early Cretaceous age for the agreement with previously published K/Ar and 40Ar/39Ar ages with a Barents Sea magmatism is related to the stratigraphic position of crater more than 50 Ma age range with peaks around 115, 100, 91, and structures connected to potential hydrothermal vent complexes. 78 Ma (Birkenmajer et al., 2010; Nejbert et al., 2011). Locally, very Hydrothermal vent complexes form by catastrophic release of meta- large age differences exist between the 40Ar/39Ar and the U/Pb ages on morphic gas produced in the metamorphic aureole surrounding sill in- the same sill, e.g., there is a N40 Ma difference at both Linnévatnet and trusions. Therefore, the position of the craters in the stratigraphy Diabasodden (Fig. 7; Table 2). Our results clearly indicate that the indicates the timing of sill emplacement. Even though the craters are large age span observed in the published K/Ar and 40Ar/39Ar data does not well defined, they lie near the Base Aptian paleosurface interpreted not need to reflect a prolonged phase of magmatism but instead proba- from seismic data (Figs. 5 and 6). bly represents other geological processes that led to the alteration of Finally, regional changes in depositional patterns of sedimentary plagioclase. sequences closely associated in time with BSSC can be interpreted in Only one of our 40Ar/39Ar ages, plagioclase sample 4A (Fig. 8A), can terms of lithospheric doming resulting from the rising mantle plume be reliably interpreted as a crystallization age based on the age statistics (He et al., 2003; Saunders et al., 2007). Borehole and outcrop data S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95 93

A) 4A, plagioclase

0.02 0.023 ± 0.0005

0.01 K/Ca

0.00 300 13

250

200 120.22 ± 1.87 Ma 150 11 12 14 Age (Ma) 5 6 7 1 2 3 4 8 9 100 10

50

0 0 20 40 60 80 100 Cumulative %39 Ar released B) 2A, plagioclase Fig. 9. Estimated range of the methane production potential, WC, in metamorphic aureoles in BSSC sedimentary basins following the method of Svensen et al. (2004). WC = FCVA, ρ ρ −3 0.40 where is the organic matter density ( =2400kgm ), VA is the volume of the meta- 0.181 ± 0.01 morphic contact aureole, and FC is the amount of carbon transferred to gas (estimated in 0.30 the range from 0.5 to 2 wt%). The estimated volume of the metamorphic aureole is 0.035 ± 0.001 K/Ca 0.20 taken to be twice the volume of the sill intrusions (see Table 3). WC is mass of carbon pro- 0.10 duced in the metamorphic aureole. The production potential is also given in barrel of oil equivalent, bbl = WC/0.11, where 0.11 (t) is the mass of carbon in one barrel of crude 0.00 300 oil (159 l). An estimate of the amount of gas ultimately produced from the original organic 8 matter during maturation and metamorphism is 50–90% depending on the kerogen type 250 16 and initial basin temperature. The possible contributions to carbon gas formation from sediment decarbonation reactions, magma degassing, or petroleum reservoirs pierced 200 9 by sills or hydrothermal vent complexes are not included.

150 75.44 ± 1.5 Ma 7 Age (Ma) 1215 6 100 10 14 1 2 3 4 Land, and the formation of extensive southward-prograding clinoforms 5 13 50 in the southeast of the Barents Sea (Fig. 4). These sedimentary se- 11 quences have a Barremian to Early Aptian age at ca. 125 Ma that 0 immediately pre-dates the emplacement of the BSSC (Midtkandal et al., 2014). -50 0 20 40 60 80 100 Cumulative %39 Ar released 5.3. Implication for carbon gas generation C) 6D, plagioclase Magmatic intrusions emplaced in carbonaceous sediments involve 8 2.77 ± 0.11 the sudden exposure of organic matter in contact aureoles to tempera- 6 tures as high as 600 °C. The effects of contact metamorphism on 4 organic-rich sediments has been previously investigated in terms of or- K/Ca 2 ganic matter maturation (Aarnes et al., 2011; Golab et al., 2007; Gurba 0 160 and Weber, 2001; Meyers and Simoneit, 1999), volatile generation

13 (Cooper et al., 2007; Jamtveit et al., 2004; Suchý et al., 2004; Wycherley et al., 1999), and global environmental crisis (Aarnes et al., 82.6 ± 0.71 Ma 120 14 2015; Aarnes et al., 2011; Storey et al., 2007; Svensen et al., 2008; 11 10 Svensen et al., 2007; Svensen et al., 2010; Svensen et al., 2004; 12 8 34 567 Svensen et al., 2009). The thermal effect of a thin sill emplaced in Trias- 9 80 2 sic sediments was recently studied using core data from borehole DH4 Age (Ma) located 4.5 km to the southeast of Longyearbyen (Senger et al., 1 2014a). The aureole is associated with higher maturity and decreased 40 carbon content toward the sill intrusion and is approximately the same thickness as the dolerite both above and below the ~2 m thick intrusion. 0 0 20 40 60 80 100 In general, the metamorphic aureole is twice the intrusion thickness 39 Cumulative % Ar released (Aarnes et al., 2015; Aarnes et al., 2010; Svensen et al., 2009), implying – 3 40 39 that the 103,000 206,000 km of sills thermally altered between Fig. 8. Ar/ Ar ages for (A) sample 4A (crystallization age), (B) 2A (sericitization age), 3 – – and (C) 6D (sericitization age). ~200,000 and 400,000 km of Carboniferous Permian Triassic sedi- ments during the massive BSSC magmatic event in the Barents Sea show that uplift of the northern portion of the Barents Sea shelf resulted (Table 3 and Fig. 9). By using an approach similar to that of Svensen in the deposition of the continental Helvetiafjellet Formation on Sval- et al. (2004), the mass of carbon potentially transferred from organic bard, continental volcano-sedimentary accumulations on Franz Josef matter to gas ranges between 2,400 and 19,200 Gt for a sill thickness 94 S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95 between 114 and 228 m over the 900,000 km2 area (Table 3 and Fig. 9). Ando, A., Kakegawa, T., Takashima, R., Saito, T., 2002. New perspective on Aptian carbon isotope stratigraphy: data from d13C records of terrestrial organic matter. Geology These simple calculations assume that 0.5 to 2 wt% of the carbon in the 30, 227–230. aureole is potentially converted into gas, regardless of the kerogen type Bailey, J.C., Rasmussen, M.H., 1997. Petrochemistry of Jurassic and Cretaceous tholeiites and origin (marine vs. continental), initial carbon content, and matura- from Kong Karls Land, Svalbard, and their relation to Mesozoic magmatism in the Arctic. Polar Res. 16, 37–62. tion stage. The end member values correspond to 22 and 174 TBOE (tril- Birkenmajer, K., Krajewski, K.P., Pecskay, Z., Lorenc, M.W., 2010. K-Ar dating of basin lion of barrels of oil equivalent) of hydrocarbon potentially produced intrusions at Bellsund, Spitsbergen, Svalbard. Pol. Polar Res. 31, 3–16. during the magmatic event (0.11 t of carbon in one barrel of crude Buchan, K.L., Ernst, R.E., 2006. Giant dyke swarms and the reconstruction of the Canadian oil). The carbon from the aureole can be partly mobilized as Arctic Islands, Greenland, Svalbard and Franz Josef Land. In: Hanski, E., Mertanen, S., Rämö, O.T., Vuollo, J. (Eds.), Dyke swarms: Time Makers of Crustal Evolution. Taylor & 13 thermogenic methane with a δ C value of about −30‰ (Svensen Francis, Balkema, Netherlands, pp. 27–48. et al., 2015b). Coffin, M.F., Eldholm, O., 1994. Large igneous provinces: crustal structure, dimensions, and external consequences. Rev. Geophys. 32, 1–36. The production rates and the fate of the aureole gases are more dif- fi fi Cof n, M.F., Eldhom, O., 1992. Volcanism and continental break-up: a global compilation cult to determine based on the current study. The release of isotopical- of large igneous provinces. In: Storey, B.C., Alabaster, T., Pankurts, R.J. (Eds.), ly light carbon aureole gases in LIPs may have triggered global warming Magmatism and the causes of continental break-up. Geological Society Special Publi- a number of times during the Earth history (e.g., Svensen et al., 2015a). cation N 68, pp. 17–30. Cooper, J.R., Crelling, J.C., Rimmer, S.M., Whittington, A.G., 2007. Coal metamorphism by The timing of the BSSC is broadly similar to the Oceanic Anoxic Event 1a igneous intrusion in the Raton Basin, CO and NM: implications for generation of (OAE1a) and the release of isotopically light carbon from this event may volatiles. Int. J. Coal Geol. 71, 15–27. have caused the OAE1a negative isotope excursion and associated glob- Corfu, F., Svensen, H., Neumann, E.-R., Nakrem, H.A., Planke, S., 2010. U-Pb and geochem- ical evidence for a Cryogenian magmatic arc in central Novaya Zemlya. Arctic Russia. al warming (Ando et al., 2002; Godet et al., 2006). Some of the aureole Terra Nova 22, 116–124. gases may also still be present in the east Barents Basin, and could Corfu, F., Polteau, S., Planke, S., Faleide, J.I., Svensen, H., Zayoncheck, A., Stolbov, N., 2013. form parts of the gigantic gas fields in the area (Fig. 2). U-Pb geochronology of Cretaceous magmatism on Svalbard and Franz Josef Land. Barents Sea Large Igneous Province. Geol. Mag. 150, 1127–1135. http://dx.doi.org/ 10.1017/S0016756813000162. 6. Conclusions Dallmann, W.K., 1999. Lithostratigraphic lexicon of Svalbard, review and recommenda- tions for nomenclature use, Upper Palaeozoic to bedrock. Norsk Polarinstitutt, Tromsø, Norway. The Early Cretaceous BSSC magmatism in the High Arctic had impor- Dallmann, W.K., Piepjohn, K., Halverson, G.P., Elvevold, S., Blomeier, D., 2011. Geological tant global paleogeographic implications that can be interpreted from map Svalbard 1:100 000, sheet D6G Vaigattbogen. nr. 48. ed. Norsk Polarinstitutt the integrated seismic–gravity–magnetic data, radiochronology results, Temakart. boreholes, and field data. The interaction between a mantle plume with Dibner, V.D., 1998. Geology of Franz Josef Land. Norsk Polarinstitutt, Oslo. Døssing, A., Jackson, H.R., Matzka, J., Einarsson, I., Rasmussen, T.M., Olesen, A.V., Brozena, the lithosphere caused the northern portion of the Barents Sea shelf J.M., 2013. On the origin of the Amerasia Basin and the High Arctic Large Igneous to be uplifted, resulting in regional changes in sedimentary deposi- Province—results of new aeromagnetic data. Earth Planet. Sci. Lett. 363, 219–230. tional patterns during the Barremian and Early Aptian times. The Drashev, S., Saunders, A.D., 2003. The Early Cretaceous Arctic LIP: its geodynamic setting and 2 implications for Canada Basin opening. Fourth International Conference on Arctic mar- massive (900,000 km ) and likely rapid injection of hot magma gins.USDepartmentoftheinterior,Dartmouth,NovaScotia,Canada,pp.216–223. into organic-rich sediments in the Barents Sea caused rapid organic Ernst, R.E., 2014. Large Igneous Provinces. Cambridge University Press. matter maturation and formation of thermogenic gas and oil be- Estrada, S., Henjes-Kunst, F., 2013. 40Ar-39Ar and U-Pb dating of Cretaceous continental rift-related magmatism on the northeast Canadian Arctic margin. Z. Dtsch. Ges. tween 120 and 125 Ma. We estimated that up to 19,200 Gt of carbon Geowiss. 164, 107–130. were potentially mobilized from the contact aureoles to form ther- Evdokimov, A.N., Stolbov, M., 2004. The interior structure of Franz Josef Land basalt mogenic gas. Finally, we have demonstrated that the large range of region. 32nd International Geological Congress, pp. 269–270. Evenchick, C.A., Davis, W.J., Bédard, J.H., Hayward, N., Friedman, R.M., 2015. Evidence for published ages of BSSC related rocks based on the K/Ar system is protracted High Arctic Large Igneous Province magmatism in the central Sverdrup much more likely to be the result of alteration (with a dominant Basin from stratigraphy, geochronology, and paleodepths of saucer-shaped sills. phase in the Early Cretaceous at ca. 80 ± 5 Ma) rather than Geol. Soc. Am. Bull. http://dx.doi.org/10.1130/B31190.1. representing an extended period of magmatism. Fedorov, P.I., Flerov, G.B., Golobin, D.I., 2005. New data on the age and composition of volcanic rocks on the Bennett Island (East Arctic). Dokl. Akad. Nauk 400, 666–670. Supplementary data to this article can be found online at http://dx. Filatova, N.I., 2007. Origin of the De Long Rise, east Arctic. Dokl. Akad. Nauk 413, 520–524. doi.org/10.1016/j.palaeo.2015.07.007. Fluteau, F., Ramstein, G., Besse, J., Guiraud, R., Masse, J.P., 2007. Impacts of palaeogeography and sea level changes on Mid-Cretaceous climate. Palaeogeogr. Palaeoclimatol. Palaeoecol. 247, 357–381. Acknowledgments Ganerød, M., Chew, D., Smethurst, M.A., Troll, V.R., Corfu, F., Meade, F., Prestvik, T., 2011. Geochronology of the Tardree Rhyolite Complex, Northern Ireland: implications for fi We gratefully acknowledge the financial support from the Nor- zircon ssion track studies, the North Atlantic Igneous Province and the age of the Fish Canyon sanidine standard. Chem. Geol. 286, 222–228. wegian Research Council with its Centers of Excellence funding Godet, A., Bodin, S., Föllmi, K.B., Vermeulen, J., Gardin, S., Fiet, N., Adatte, T., Berner, Z., scheme (project number 223272) and with Statoil through the Stüben, D., van de Schootbrugge, B., 2006. Evolution of the marine stable carbon- PETROBAR project. The seismic, magnetic, and gravity data present- isotope record during the Early Cretaceous: a focus on the late Hauterivian and Barremian in the Tethyan realm. Earth Planet. Sci. Lett. 242, 254–271. ed in this study were provided by TGS. The Norwegian Petroleum Golab, A.N., Hutton, A.C., French, D., 2007. Petrography, carbonate mineralogy and Directorate gave access to their seismic data. Erik Halvorsen (Univer- geochemistry of thermally altered coal in Permian coal measures, Hunter Valley, sity of Oslo) supplied his paleomagnetic samples for dating, and Snorre Australia. Int. J. Coal Geol. 70, 150–165. Grogan, P., Nyberg, K., Fotland, B., Myklebust, R., Dahlgren, S., Riis, F., 1998. Cretaceous Olaussen (University of Svalbard) coordinated the sampling of the magmatism south and east of Svalbard: evidence from seismic reflection and Longyearbyen CO2 wells on Svalbard. Finally, we thank Fred Jourdan, magnetic data. Polarforschung 68, 25–34. Kim Senger, and Carol Evenchick for their constructive reviews. Gurba, L.W., Weber, C.R., 2001. Effects of igneous intrusions on coalbed methane poten- tial, Gunnedah Basin, Australia. Int. J. Coal Geol. 46, 113–131. Harland, W.B., 1997. The Geology of Svalbard. The Geological Society, London. References Hay, W.W., 2008. Evolving ideas about the Cretaceous climate and ocean circulation. Cretac. Res. 29, 725–753. Aarnes, I., Svensen, H., Connolly, J.A.D., Podladchikov, Y.Y., 2010. How contact metamor- He, B., Xu, Y.-G., Chung, S.-L., Xiao, L., Wang, Y., 2003. Sedimentary evidence for a rapid, phism can trigger global climate changes: modeling gas generation around igneous kilometer-scale crustal doming prior to the eruption of the Emeishan flood basalts. sills in sedimentary basins. Geochim. Cosmochim. Acta 74, 7179–7195. Earth Planet. Sci. Lett. 213, 391–405. Aarnes, I., Svensen, H., Polteau, S., Planke, S., 2011. Contact metamorphic devolatilization He, H., Pan, Y., Tauxe, L., Qin, H., Zhu, R., 2008. Toward age determination of the M0r of shales in the Karoo Basin, South Africa, and the effects of multiple sill intrusions. (Barremian–Aptian boundary) of the Early Cretaceous. Phys. Earth Planet. Inter. Chem. Geol. 281, 181–194. 169, 41–48. Aarnes, I., Planke, S., Trulsvik, M., Svensen, H., 2015. Contact metamorphism and thermo- Henriksen, E., Ryseth, A.E., Larssen, G.B., Heide, T., Rønning, K., Sollid, K., Stoupakova, A.V., genic gas generation in the Vøring and Møre basins (offshore Norway) during the 2011a. Tectonostratigraphy of the greater Barents Sea: implications for petroleum sys- Paleocene–Eocene thermal maximum. J. Geol. Soc. Lond. http://dx.doi.org/10.1144/ tems. In: Spencer, A.M., Embry, A.F., Gautier, D.L., Stoupakova, A.V., Sørensen, K. (Eds.), jgs2014-098 (in press). Arctic Petroleum Geology. The Geological Society of London, London, pp. 163–195. S. Polteau et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 83–95 95

Henriksen, E., Bjørnseth, H.M., Hals, T.K., Heide, T., Kiryukhina, T., Kløvjan, O.S., Larssen, Senger, K., Roy, S., Braathen, A., Buckley, S.J., Bælum, K., Gernigon, L., Mjelde, R., Noormets, G.B., Ryseth, A.E., Rønning, K., Sollid, K., Stoupakova, A., 2011b. Uplift and erosion of R., Ogata, K., Olaussen, S., Planke, S., Ruud, B.O., Tveranger, J., 2013. Geometries of dol- the greater Barents Sea: impact on prospectivity and petroleum systems. In: Spence, eritic intrusions in central Spitsbergen, Svalbard: an integrated study of an onshore–

D.A.e.a. (Ed.), Arctic Petroleum Geology. Geological Society, London, pp. 271–281. offshore magmatic province with implications on CO2 sequestration. Nor. J. Geol. 93, Jamtveit, B., Svensen, H., Podladchikov, Y., Planke, S., 2004. Hydrothermal vent complexes 143–166. associated with sill intrusions in sedimentary basins. In: Breitkreuz, C., Petford, N. Senger, K., Planke, S., Polteau, S., Ogata, K., Svensen, H., 2014a. Sill emplacement and con- (Eds.), Physical Geology of High-Level Magmatic Systems. Geological Society Special tact metamorphism in a siliciclastic reservoir on Svalbard, Arctic Norway. Nor. Geol. Publication, London, pp. 233–241. Tidsskr. 94, 155–169. Janasi, V.d.A., de Freitas, V.A., Heaman, L.H., 2011. The onset of flood basalt volcanism, Senger, K., Tveranger, J., Ogata, K., Braathen, A., Planke, S., 2014b. Late Mesozoic Northern Paraná Basin, Brazil: a precise U-Pb baddeleyite/zircon age for a Chapecó- magmatism in Svalbard: a review. Earth Sci. Rev. 139, 123–144. type dacite. Earth Planet. Sci. Lett. 302, 147–153. Shipilov, E.V., Karyakin, Y.V., Matishov, G.G., 2009. Jurassic–Cretaceous Barents–Amer- Jokat, W., 2003. Seismic investigation along the western sector of Alpha Ridge, Central asian superplume and initial stage of geodynamic evolution of the Arctic Ocean. Arctic Ocean. Geophys. J. Int. 152, 185–201. Dokl. Akad. Nauk 426, 369–372. Keller, G., 2008. Cretaceous climate, volcanism, impacts, and biotic effects. Cretac. Res. 29, Skola, I.V., Pcelina, T.M., Mazur, V.B., Alter, S.M., 1980. New data on the composition and 754–771. structure of the sedimentary platform cover on the basis of materials from the drilling Khlebnikov, P.A., Belenky, V.Y., Peshkova, I.N., Kazanin, G.S., Shkarubo, S.I., Pavlov, S.P., of a parametric hole at Grumantbyen, Geology of the Sedimentary Platform Cover of Shlykova, V.V., 2011. Geological structure and petroleum potential of the eastern the Archipelago of Svalbard. NIIGA, Leningrad, pp. 13–24. flank of the Northern Barents Basin. In: Spencer, A.M., Embry, A.F., Gautier, D.L., Smelror, M., Petrov, O.V., Larssen, G.B., Werner, S., 2009. ATLAS—Geological History of the Stoupakova, A.V., Sørensen, K. (Eds.), Arctic Petroleum Geology, pp. 261–269 Geol. Barents Sea. Soc. Mem.; London. Smith, D.G., Harland, W.B., Hughes, N.F., Pickton, C.A.G., 1976. The geology of Kong Karls Komarnitskii, V.M., Shipilov, E.V., 1991. New Geological Data on the Magmatism of the Land, Svalbard. Geol. Mag. 113, 193–232. Barents Sea. Dokl. Akad. Nauk SSSR 320, 1203–1206. Storey, M., Duncan, R.A., Swisher III, C.C., 2007. Paleocene–Eocene thermal maximum and Koryakin, Y.V., Shipilov, E.V., 2009. Geochemical specifics and 40Ar/39Ar age of the the opening of the northeast Atlantic. Science 316, 587–589. basaltoid magmatism of the Alexander Land, Northbrook, Hooker, and Hayes Islands Suchý, V., Šafanda, J., Sýkorová, I., Stejskal, M., Machovič, V., Melka, M., 2004. Contact (Franz Josef Land Archipelago). Dokl. Akad. Nauk 425, 213–217. metamorphism of Silurian black shales by a basalt sill: geological evidence and ther- Kos'ko, M., Korago, E., 2009. Review of geology of the New Siberia Islands between Laptev mal modeling in the Barrandian Basin. Bull. Geosci. 79, 133–145. and the East Siberian Seas, North East Russia. Stephan Mueller Special Publication Svensen, H., Planke, S., Malthe-Sørenssen, A., Jamtveit, B., Myklebust, R., Eidem, T., Rey, Series 4. pp. 45–64. S.S., 2004. Release of methane from a volcanic basin as a mechanism for initial Eocene Kowallis, B.J., Christiansen, E.H., Deino, A.L., Kunk, M.J., Heaman, L.M., 1995. Age of the global warming. Nature 429, 542–545. Cenomanian–Turonian boundary in the Western Interior of the United States. Cretac. Svensen, H., Planke, S., Chevallier, L., Malthe-Sorenssen, A., Corfu, F., Jamtveit, B., 2007. Hy- Res. 16, 109–129. drothermal venting of greenhouse gases triggering Early Jurassic global warming. Maher, H.D.J., 2001. Manifestations of the Cretaceous High Arctic Large Igneous Province Earth Planet. Sci. Lett. 256, 554–566. in Svalbard. J. Geol. 109, 91–104. Svensen, H., Bebout, G., Kronz, A., Li, L., Planke, S., Chevallier, L., Jamtveit, B., 2008. Nitro- Meyers, P.A., Simoneit, B.R.T., 1999. Effects of extreme heating on the elemental and gen geochemistry as a tracer of fluid flow in a hydrothermal vent complex in the isotopic compositions of an Upper Cretaceous coal. Org. Geochem. 30, 299–305. Karoo Basin, South Africa. Geochim. Cosmochim. Acta 72, 4929–4947. Midtkandal, I., Nystuen, J.P., Nagy, J., Mørk, A., 2008. Lower Cretaceous lithostratigraphy Svensen, H., Planke, S., Polozov, A.G., Schmidbauer, N., Corfu, F., Podladchikov, Y., Jamtveit, across a regional subaerial unconformity in Spitsbergen: the Rurikfjellet and B., 2009. Siberian gas venting and the end-Permian environmental crisis. Earth Planet. Helvetiafjellet formations. Nor. J. Geol. 88, 287–304. Sci. Lett. 277, 490–500. Midtkandal, I., Planke, S., Polteau, S., Corfu, F., Faleide, J.I., Svensen, H., Selnes, H., 2014. Svensen, H., Planke, S., Corfu, F., 2010. Zircon dating ties NE Atlantic sill emplacement to Arctic OAE1a Triggered by the Barents Sea Large Igneous Province? Fall Meeting. initial Eocene global warming. J. Geol. Soc. 167, 433–436. American Geophysical Union, San Francisco (abstract #PP13A-1400). Svensen, H.H., Planke, S., Neumann, E.-R., Aarnes, I., Marsh, J.S., Polteau, S., Harstad, C.H., Minakov, A., Mjelde, R., Faleide, J.I., Flueh, E.R., Dannowski, A., Keers, H., 2012. Mafic intru- Chevallier, L., 2015a. Sub-volcanic intrusions and the link to global climatic and envi- sions east of Svalbard imaged by active-source seismic tomography. Tectonophysics ronmental changes, Advances in Volcanology. Springer International Publishing, 518–521, 106–118. Switzerland. Myklebust, R., 1994. Magnetic Mapping of the Barents Sea (MMBS). Interpretation report. Svensen, H., Fristad, K., Polozov, A.G., Planke, S., 2015b. Volatile generation and release Amarok, A.S, Oslo, p. 214. from continental large igneous provinces. In: Schmidt, A., Fristad, K., Elkins Tanton, Nejbert, K., Krajewski, K.P., Dubinska, E., Pecskay, Z., 2011. The dolerites from Svalbard, L.T. (Eds.), Volcanism and global environmental change. Cambridge University NW Barents Sea Shelf: age, geochemistry and significance for geotectonic interpreta- Press, pp. 177– 192. tion of the High-Arctic large igneous province. Polar Res. 30, 24. Tarduno, J.A., Brinkman, D.B., Renne, P.R., Cottrell, R.D., Scher, H., Castillo, P., 1998. Evi- Ohm, S.E., Karlsen, D.A., Austin, T.J.F., 2008. Geochemically driven exploration models in dence for extreme climatic warmth from Late Cretaceous Arctic vertebrates. Science uplifted areas: examples from the Norwegian Barents Sea. AAPG Bull. 92, 1191–1223. 282, 2241–2243. Parker, J.R., 1967. The Jurassic and Cretaceous sequence in Spitzbergen. Geol. Mag. 104, Trettin, H.P., Parrish, R., 1987. Late Cretaceous bimodal magmatism, northern Ellesmere 487–505. Island: isotopic age and origin. Can. J. Earth Sci. 24, 257–265. Peters, L.J., 1949. The direct approach to magnetic interpretation and its practical applica- Verati, C., Jourdan, F., 2014. Modeling effect of sericitization of plagioclase on the 40K/40Ar tions. Geophysics 14, 290–320. and 40Ar/39Ar chronometers : implication for dating basaltic rocks and mineral Planke, S., Symonds, P.A., Alvestad, E., Skogseid, J., 2000. Seismic volcanostratigraphy of deposits. Geol. Soc. Lond., Spec. Publ. 378, 155–174. large-volume basaltic extrusive complexes on rifted margins. J. Geophys. Res. 105, Verhoef, J., Macnab, R., Roest, W., et al., 1996. Magnetic anomalies, Arctic and North Atlan- 19335–19351. tic Oceans and adjacent land areas. Geological Survey of Canada Open File 3282c. Planke, S., Rasmussen, T., Rey, S.S., Myklebust, R., 2005. Seismic characteristics and distri- Villeneuve, M., Williamson, M.-C., 2006. 40Ar-39Ar dating of mafic magmatism from the bution of volcanic intrusions and hydrothermal vent complexes in the Vøring and Sverdrup Basin Magmatic Province. In: Scott, R.A., Thurston, D.K. (Eds.), Proceedings Møre basins. In: Doré, A.G., Vining, B.A. (Eds.), Petroleum Geology: Northwest of the Fourth International Conference on Arctic Margins. US Department of the Inte- Europe and Global Perspectives. Geological Society, London, pp. 833–844. rior, pp. 206–215. Planke, S., Svensen, H., Myklebust, R., Bannister, S., Manton, B., Lorenz, L., 2014. Geophys- Werner, S.C., Ebbing, J., Litvinova, T.P., Olesen, O., 2011. Chapter 11 Structural interpreta- ics and Remote Sensing. IAVCEI, Springer International Publishing, Switzerland, Ad- tion of the Barents and Kara Seas from gravity and magnetic data. In: Spencer, vances in Volcanology. A.M.e.a. (Ed.), Arctic Petroleum Geology. Geological Society, London, pp. 197–208. Renne, P.R., Mundil, R., Balco, G., Min, K., Ludwig, K.R., 2010. Joint determination of 40 K Wessel, P., Smith, W.H.F., 1991. Free software helps map and display data. EOS Trans. Am. decay constants and 40Ar*/40 K for the Fish Canyon sanidine standard, and improved Geophys. Union 72, 445–446. accuracy for 40Ar/39Ar geochronology. Geochim. Cosmochim. Acta 74, 5349–5367. Williamson, M.-C., 1988. The Cretaceous igneous province of the Sverdrup Basin, Canadi- Sandwell, D.T., Smith, W.H.F., 2009. Global marine gravity from retracked Geosat and ERS- an Arctic: field relations and petrochemical studies PhD Thesis. Dalhousie University, 1 altimetry: ridge segmentation versus spreading rate. J. Geophys. Res. 114, B01411. Halifax, Nova Scotia, p. 417. Saunders, A.D., Jones, S.M., Morgan, L.A., Pierce, K.L., Widdowson, M., Xu, Y.G., 2007. Re- Wycherley, H., Fleet, A., Shaw, H., 1999. Some observations on the origins of large gional uplift associated with continental large igneous provinces: the roles of mantle volumes of carbon dioxide accumulations in sedimentary basins. Mar. Pet. Geol. 16, plumes and the lithosphere. Chem. Geol. 241, 282 –318. 489–494.