marine and petroleum geology...marine and petroleum geology 78 (2016) 516e535 strata further north,...

lable at ScienceDirect

Marine and Petroleum Geology 78 (2016) 516e535

Contents lists avai

Marine and Petroleum Geology

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

Research paper

Provenance of Triassic sandstones on the southwest Barents Shelf andthe implication for sediment dispersal patterns in northwest Pangaea

Edward J. Fleming a, *, Michael J. Flowerdew a, Helen R. Smyth a, 1, Robert A. Scott a, 2,Andrew C. Morton a, b, Jenny E. Omma c, Dirk Frei d, Martin J. Whitehouse e

a CASP, 181A Huntingdon Road, Cambridge, CB3 0DH, UKb HM Research Associates, St Ishmaels, Haverfordwest SA62 3TG, UKc Rocktype Limited, Oxford, OX4 1LN, UKd Stellenbosch University, Victoria Street, Stellenbosch, South Africae Swedish Museum of Natural History, Frescativ€agen 40, 114 18, Stockholm, Sweden

a r t i c l e i n f o

Article history:Received 21 June 2016Received in revised form24 August 2016Accepted 3 October 2016Available online 5 October 2016

Keywords:Barents seaTriassicPalaeogeographyProvenanceU-Pb zircon geochronologyReservoir qualitySediment compositionPetroleum basin analysis

* Corresponding author.E-mail address: [email protected] (

1 Present address: Neftex Exploration Insights, HallUnited Kingdom.

2 Deceased.

http://dx.doi.org/10.1016/j.marpetgeo.2016.10.0050264-8172/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

Thick Triassic siliciclastic units form major reservoir targets for hydrocarbon exploration on the BarentsShelf; however, poor reservoir quality, possibly associated with variation in provenance, remains a keyrisk factor in the area. In this study, sandstone dispersal patterns on the southwest Barents Shelf areinvestigated through petrographic and heavy mineral analysis, garnet and rutile geochemistry and zirconU-Pb geochronology. The results show that until the Early Norian Maximum Flooding Surface, twocontrasting sand types were present: (i) a Caledonian Sand Type, characterised by a high compositionalmaturity, a heavy mineral assemblage dominated by garnet and low chrome-spinel:zircon (CZi) values,predominantly metapelitic rutiles and mostly Proterozoic and Archaean detrital zircon ages, interpretedto be sourced from the Caledonides, and (ii) a Uralian Sand Type, characterised by a low compositionalmaturity, high CZi values, predominantly metamafic rutiles and Carboniferous zircon ages, sourced fromthe Uralian Orogeny. In addition, disparity in detrital zircon ages of the Uralian Sand Type withcontiguous strata on the northern Barents Shelf reveals the presence of a Northern Uraloid Sand Type,interpreted to have been sourced from Taimyr and Severnaya Zemlya. As such, a coincidental system isinferred which delivered sand to the Northern Barents Shelf in the late Carnian/early Norian. Followingthe Early Norian Maximum Flooding Surface, a significant provenance change occurs. In response to LateTriassic/Early Jurassic hinterland rejuvenation, supply from the Uralian Orogen ceased and the northernScandinavian (Caledonian) source became dominant, extending northwards out on to the southwestBarents Shelf. The data reveal a link between reservoir quality and sand type and illustrate how prov-enance played an important role in the development of clastic reservoirs within the Triassic of theBarents Shelf.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

The Triassic succession on the Barents Shelf is currently ofconsiderable importance for hydrocarbon exploration (Henriksenet al., 2011; Lundschien et al., 2014). During the Triassic, the areaformed part of a large epicontinental seaway in the northwestern

E.J. Fleming).iburton, Abingdon OX14 4RW

corner of the supercontinent Pangaea (Worsley, 2008). This seawaywas progressively infilled by prograding delta complexes whichdeposited up to 7 km of siliciclastic sediment in a series oftransgressive-regressive sequences (Glørstad-Clark et al., 2011;Klausen et al., 2015; Riis et al., 2008). The deposition of thinorganic-rich shales in front of these prograding systems, followedby extensive paralic sandstones of the delta front and top have ledauthors to postulate prolific hydrocarbon play potential (e.g.Lundschien et al., 2014). These ideas were reinforced by the dis-covery of the Goliat field (174 MMbbl estimated) in 2000, whichfound good reservoir quality in mid-late Triassic sandstones in thesouthern Hammerfest Basin. The considerable thickness of Triassic

mailto:[email protected]

http://crossmark.crossref.org/dialog/?doi=10.1016/j.marpetgeo.2016.10.005&domain=pdf

www.sciencedirect.com/science/journal/02648172

http://www.elsevier.com/locate/marpetgeo

http://dx.doi.org/10.1016/j.marpetgeo.2016.10.005



E.J. Fleming et al. / Marine and Petroleum Geology 78 (2016) 516e535 517

strata further north, combined with extensive play developmenthas prompted a significant drilling campaign in recent years.However, outside the southern Hammerfest Basin, exploration re-sults within the Triassic have been less successful, with only smallaccumulations and/or residual hydrocarbons being encountered(Henriksen et al., 2011).

The reservoir quality of Triassic sandstones on the Barents Shelfcan be variable (Henriksen et al., 2011) which is a key risk factor forexploration in the area. The cause of this variability is likely to becomplex, but differences in the original sandstone compositionlikely play a significant role. One of the fundamental controls ofsandstone composition is provenance (Johnsson, 1993). Previousresearch has shown this to be inconsistent on the Barents Shelf inthe Triassic (Mørk, 1999). Northern Scandinavia is thought to havebeen an important sediment source throughout the Mesozoic,particularly during the Jurassic. However, in the Triassic, seismicanalysis has revealed that major deltaic complexes migrated fromthe southeast across the shelf (Glørstad-Clark et al., 2010; Klausenet al., 2015; Riis et al., 2008) in response to the developing UralianOrogeny. Additional sources have been inferred from Greenland(Mørk, 1999; P�ozer Bue and Andresen, 2014) and local topographichighs (e.g. Glørstad-Clark et al., 2010). The delineation and possibleinteraction between these systems on the southwest Barents Shelfis largely unknown.

This paper presents provenance data from Triassic core mate-rials that form part of the CASP Barents Shelf provenance dataset.The selection includes 39 petrographic analyses, 72 heavy mineralanalyses, 10 rutile geochemical analyses, 8 garnet geochemicalanalyses and 8 U-Pb zircon age analyses. Using an integratedapproach, the aims of this study are to (i) characterise the prove-nance signatures of Triassic sandstones on the southwest BarentsShelf; (ii) map the distribution of sand types to reconstruct thedrainage distribution; and finally (iii) compare the provenance datawith published data from the greater Barents Shelf in order toprovide insight into the wider tectonic evolution and location ofsedimentary source regions. The outcomes of this work enablemore precise constraints on the location and composition ofTriassic sandstones on the Barents Shelf, which may be used toreduce uncertainty in reservoir prediction.

2. Geological background

The Barents Shelf is subdivided into numerous basins, platformsand highs (Fig. 1a; Gabrielsen et al., 1990). To the east, the regionalstructure is dominated by a major basin, located west of NovayaZemlya, called the East Barents Basin. Westwards, there is a sig-nificant change in structural style, and a series of relatively narrow,predominantly northeast-southwest trending, elongate exten-sional basins are present (Gernigon et al., 2014). Deposition wasinitiated due to Late Palaeozoic collapse of the Caledonian moun-tain belt. This was followed by rift-related subsidence and thedevelopment of the North Atlantic rift system (Dor�e, 1992). Rapidsubsidence occurred in the North and South Barents basins to theeast during the Late Permian and continued through the Triassicresulting in the creation of a large epicontinental seaway (Smelroret al., 2009).

On the southwest Barents Shelf, a total thickness in excess of2.5 kmwas deposited in the Triassic, which has been divided into 5lithostratigraphic formations (Havert, Klappmyss, Kobbe, Snaddand Fruholmen formations; Fig. 1b). Each of these can be related toa transgressive-regressive cycle, bounded by a maximum floodingsurface (MFS) (Glørstad-Clark et al., 2010; Henriksen et al., 2011;Klausen et al., 2015). Significant seaward shifts of the shorelinecan be recognised seismically (Glørstad-Clark et al. 2010, 2011;Klausen et al., 2015). These reflect regression and the

progradation of major delta complexes which gradually infilled theremnant end-Permian palaeotopography (Glørstad-Clark et al.,2010; Lundschien et al., 2014). These are separated by periods oftransgression where offshore marine deposits onlap against deltaicfacies. Of particular note is the Early Norian MFS (Klausen et al.,2015) which separates the Snadd and Fruholmen formations.Whilst being poorly defined seismically (Glørstad-Clark et al.,2010), the surface coincides with a profound change in miner-alogy and depositional environments (e.g. Bergan and Knarud,1993; Mørk, 1999; Ryseth, 2014).

Traditionally, northern Scandinavia was thought to have beenthe major source area for Palaeozoic and Mesozoic successions onthe southwest Barents Shelf (Rønnevik et al., 1982). In the Triassic,Mørk et al. (1999) identified quartzofeldspathic sandstones in theHammerfest Basin, interpreted to be derived from the Caledonidesof Northern Scandinavia and the Baltic margin. However, since theearly days of hydrocarbon exploration, the recognition of north-westerly prograding clinoforms and an increasing thickness ofthe sediment pile to the east (Jacobsen and Veen, 1984; Nøttvedtet al., 1993; van Veen et al., 1993) implied the existence of a ma-jor source area to the southeast of the Barents Shelf. This wassupported by the work of Mørk (1999), who also identified distinctepidote-bearing, lithic-rich sandstones which were interpreted tobe sourced from the developing Uralian Orogeny. Today, the seismicevidence for delta progradation to the northwest is well docu-mented (Glørstad-Clark et al., 2010; Høy and Lundschien, 2011;Lundschien et al., 2014; Riis et al., 2008) and the shoreline isinterpreted to have reached Svalbard in the late Carnian (P�ozer Bueand Andresen, 2014). A source from Greenland has been suggestedfor some parts of the Triassic on Svalbard (Mørk, 1999) which mayalso have provided sediment to the southwest Barents Shelf(Glørstad-Clark et al., 2010).

Shelfal environments, such as those that characterise theTriassic on the Barents Shelf, can have a complex source-to-sinksystem as they are capable as acting both as sediment sources(e.g. wave ravinement), transporters (e.g. longshore drift) and sinks(Sømme et al., 2009). These processes have the ability to cause localprovenance variation and are therefore an important considerationduring analysis. However, the majority of the samples in this studywere collected from deltaic facies formed during the rapid pro-gradation of the clinoform belts (Klausen et al., 2015). As such,significant mixing and erosion of different source areas either bytides, waves or through long shore drift was probably minor at thescale examined.

3. Methods

Point counting of thin sections, stained for porosity and K-feldspar, and conventional heavy mineral analysis followed themethods outlined by Folk (1980) and Mange and Wright (2007),respectively. The provenance-sensitive heavy mineral ratios ATi(apatite:tourmaline index), GZi (garnet:zircon index), RuZi (ruti-le:zircon index), MZi (monazite:zircon index) and CZi (chrome-spinel:zircon index) were determined following Morton andHallsworth (1994). Garnet geochemical analyses were performedusing a Link Systems AN 10/55S energy-dispersive x-ray analyserattached to a Cambridge Instruments Microscan V electronmicroprobe housed at the University of Aberdeen. Garnet assem-blages have been compared by determining the relative abun-dances of garnet types A (low-Ca, high-Mg), B (low-Mg, variable-Ca) and C (high-Mg, high-Ca), as defined by Mange and Morton(2007). Rutile chemical analyses were carried out at the School ofEarth, Ocean and Planetary Sciences at Cardiff University, using aThermo Elemental X(7) series inductively coupled mass spec-trometer (ICP-MS) coupled to a NewWave Research UP213 Nd:YAG

Fig. 1. (a) Map of the Barents Shelf showing the location of wells used in this paper (tectonic elements after Gabrielsen et al., 1990). (b) Chronostratigraphy and facies summary ofthe Triassic stratigraphy on Svalbard and the western Barents Shelf. Facies interpretations are compiled and adapted after Henriksen et al. (2011), Lundschien et al. (2014), Mørk andElvebakk (1999), Rød et al. (2014) and Vigran et al. (2014). Lithostratigraphy is after Mørk et al. (1999) with adaptations fromMørk et al. (2013). Maximum flooding surfaces are afterGlørstad-Clark et al. (2010) and Klausen et al. (2015). The sea level curve is modified after Haq et al. (1987) and Golonka and Ford (2000).


213 nm UV laser. The rutile chemical data were used to determinethe proportion of metamafic and metapelitic rutiles followingMeinhold et al. (2008) and their temperature of formation usingthe Zr-in-rutile thermometer of Watson et al. (2006).

Detrital zircon U-Pb geochronology was performed using eitherlaser ablation single collector magnetic sector field inductivelycoupled plasma mass spectrometry (LA-SF-ICP-MS) or secondaryionisation mass spectrometry (SIMS). Geochronology using LA-SF-ICP-MS was carried out at the Central Analytical Facility of Stel-lenbosch University with a Thermo Finnigan Element 2 massspectrometer coupled to a NewWave UP213 laser ablation system.SIMS dating was carried out by using either a Cameca 1270 ionmicroprobe at the NORDSIM facility housed at the SwedishMuseum of Natural History, Stockholm or a SHRIMP RG instrumentat the Australian National University in Canberra, Australia. TheICP, Cameca 1270 and SHRIMP analytical methods closelyfollowed those outlined by Frei and Gerdes (2009) andWhitehouseand Kamber (2005) and Williams (1998), also see supplementarymaterial.

4. Results

4.1. Petrography

Thirty-nine samples were selected for petrographic analysis(Table 1). A wide range of compositions are encountered whichshow both spatial and temporal variation (Fig. 2). In the Havert,Klappmyss, Kobbe and Snadd formations (Fig. 1b), stronggeographic variation is seen in the petrographic characteristics.Samples from the southern part of the Hammerfest Basin areclassified as sub-arkoses (Fig. 2a) and are generally compositionallymature. The relative proportions of quartz, plagioclase, K-feldsparand rock fragments are shown in Fig. 2c. The samples contain anaverage of 79% quartz, withminor plagioclase (10%), K-feldspar (5%)and rock fragments. These rock fragments are predominantlyplutonic with minor metamorphic and sedimentary grains. Thereservoir quality is generally good or excellent, with visible po-rosities ranging from 0 to 30%, with a mean (x) of 14%. In contrast,away from the southern part of the Hammerfest Basin, samples arecompositionally immature and can be classified as arkoses, lithicarkoses and arkose litharenites (Fig. 2a). On average, samplesfrom these areas contain a lower proportion of quartz (54%) andK-feldspar (2%), but a higher proportion of plagioclase (19%)and rock fragments (25%) (Fig. 2c). The rock fragments aremostly sedimentary (mudstone and sandstone) or metamorphic,with subordinate plutonic grains. The reservoir quality isvariable, with visible porosities ranging from 0 to 20%, with anaverage of 7%.

A major petrographic shift occurs across the boundary betweenthe Snadd and the Fruholmen formations (Fig. 2) at the Early NorianMFS. Above this boundary, a marked increase in compositionalmaturity is seen in all areas. The sandstone compositions becomedominated by quartz (94%), with low proportions of K-feldspar(1%), plagioclase feldspar (3%) and rock fragments (2%) (Fig. 2c). Ofparticular note are samples from the south Hammerfest Basinwhich show extreme compositional maturity. Elsewhere, maturityis variable, but extreme compositional maturity is also seen in somesamples from the Nyslepp/Måsøy Fault Complex (e.g. samples7124_3-1_1349.35 and 7124_3-1_1355.4). Others display slightergreater immaturity (e.g. 7125_4-1_904.5). Rock fragments, wherepresent, are mostly sedimentary (65%; mudstone and sandstone)and plutonic and volcanic fragments are low in abundance (9%).The reservoir quality is generally good or excellent, with porositiesranging from 5 to 30%, with an average of 19%.

4.2. Heavy mineral data

Seventy-two samples were selected for conventional heavymineral analysis (Table 1). The results are alignedwith the temporaland spatial changes recognised in the petrographic data. In theHavert, Klappmyss, Kobbe and Snadd formations, stronggeographic variation is seen in the heavy mineral characteristics.Samples collected from the southern part of the Hammerfest Basinare commonly dominated by garnet (77%), withminor zircon (12%),tourmaline (5%) and apatite (4%) (Fig. 3c). In contrast, samples fromelsewhere contain a more diverse heavy mineral assemblagedominated by apatite (49%), including relatively high proportionsof tourmaline (7%), chrome-spinel (6%), chloritoid (5%), epidote(3%) and titanite (2%). Garnet is only a minor constituent (14%) andis absent in samples from the Bjørnøya Basin. Zircon contents arevariable (x ¼ 14%) and increase up stratigraphy. Samples from theHavert Formation from one well in the Nordkapp Basin displayunusually high proportions of epidote and titanite.

Cross plots of the provenance-sensitive heavy mineral ratios forthe Havert, Klappmyss, Kobbe and Snadd formations are shown inFig. 3a i-a.iii. Distinct differences in these ratios are observed be-tween samples from the southern part of the Hammerfest Basinand those from elsewhere. Of particular note, CZi is uniformly lowin samples from the southern part of the Hammerfest Basin (<1). Incontrast, CZi ratios from other areas are variable but normallysignificantly higher (e.g. 55 in Sample 7228_7-1A_2059.3 from theNordkapp Basin). Similarly, the southern Hammerfest Basin dis-plays high GZi (x ¼ 84) and RuZi (x ¼ 42) ratios but relatively lowATi (x ¼ 47) compared to the remaining areas.

A distinct change in the heavymineral assemblage is seen acrossthe boundary between the Snadd and the Fruholmen formations.Across this boundary, an increase in the average proportion ofultrastable heavy minerals e.g. zircon (42%), and tourmaline (20%)and fewer less stable minerals such as apatite (11%) and garnet (6%)are apparent (Fig. 3d). Geographic variation is also seen within theFruholmen Formation. Samples from Well 7122/7-4s and 7122/7-1in the Hammerfest Basin (south) display an unusually high pro-portion of kyanite. In addition, samples in the Bjørnøya Basindisplay large amounts of apatite and chrome-spinel.

Cross plots of the provenance-sensitive heavy mineral ratios forthe Fruholmen Formation are shown in Fig. 3b i-b.iii. In contrast tothe underlying formations, the samples are characterised by lowATi (x ¼ 14) and CZi (x ¼ 4) values across the study area. Thenotable exception to this pattern comes from the Bjørnøya Basinwhere high CZi and ATi values are retained. These Bjørnøya BasinFruholmen Formation samples fall into distinct fields on theRuZi:ATi, RuZi:CZi and GZi:ATi cross plots.

4.3. Garnet geochemistry

A subset of 8 samples was selected for garnet geochemicalanalysis (Table 1). Garnet geochemistry can be a particularly usefulmineral-chemical tool due its common occurrence in heavy min-eral assemblages, relative stability during both weathering anddiagenesis, and its wide range of major element compositions (seeMange and Morton (2007) for review). Although the results are notas distinct as the other provenance techniques, some spatial vari-ability can be seen between the samples. Samples from thesouthern part of the Hammerfest Basin (Fig. 4a, b and c) are typi-cally moderately pyrope-rich and spessartine-poor and have a highproportion of Type Aii garnets, suggesting a derivation from high-grade granulite-facies metasediments. In contrast, samples fromother areas such as the Finnmark Platform (Fig. 4e) and NordkappBasin (Fig. 4f, g and h) are typically more pyrope-poor andspessartine-rich and most grains tend to plot within in the Type Bi

Table 1Samples data table.

Sample No Formation Latitude Longitude Region Provenance Petrography HM garnet rutile zircon

7120_12-1_3521.7 Kobbe 71.11353 20.75559 Hammerfest Basin (south) Caledonian C C

7121_5-1_3097.4 Snadd 71.59858 21.40605 Hammerfest Basin (north) Uralian C

7122_6-1_2065.4 Fruholmen 71.63870 22.81189 Hammerfest Basin (north) Caledonian C C

7122_6-1_2093.4 Fruholmen 71.63870 22.81189 Hammerfest Basin (north) Caledonian C C

7122_6-2_2499.9 Snadd 71.58711 22.85598 Hammerfest Basin (north) Uralian C

7122_7-1_1141.3 Fruholmen 71.28637 22.31862 Hammerfest Basin (south) Caledonian C C

7122_7-1_1148.5 Fruholmen 71.28637 22.31862 Hammerfest Basin (south) Caledonian C


7122_7-1_1162.4 Fruholmen 71.28637 22.31862 Hammerfest Basin (south) Caledonian C C C


7122_7-1_1169.5 Fruholmen 71.28637 22.31862 Hammerfest Basin (south) Caledonian C

7122_7-1_1173.6 Fruholmen 71.28637 22.31862 Hammerfest Basin (south) Mixed C

7122_7-3_1187.8 Snadd 71.25487 22.26794 Hammerfest Basin (south) Caledonian C C

7122_7-3_1813.2 Kobbe 71.25487 22.26794 Hammerfest Basin (south) Caledonian C



7122_7-3_2520.2 Havert 71.25487 22.26794 Hammerfest Basin (south) Caledonian C C C C

7122_7-4S_1185.5 Fruholmen 71.25363 22.31817 Hammerfest Basin (south) Caledonian C C

7122_7-4S_1194.3 Fruholmen 71.25363 22.31817 Hammerfest Basin (south) Caledonian C

7122_7-4S_1198.5 Fruholmen 71.25363 22.31817 Hammerfest Basin (south) Caledonian C C

7122_7-4S_1210.5 Fruholmen 71.25363 22.31817 Hammerfest Basin (south) Caledonian C C C

7122_7-4S_1805.5 Kobbe 71.25363 22.31817 Hammerfest Basin (south) Caledonian C C C C

7122_7-4S_1811.4 Kobbe 71.25363 22.31817 Hammerfest Basin (south) Caledonian C

7122_7-4S_1817.9 Kobbe 71.25363 22.31817 Hammerfest Basin (south) Caledonian C C C

7122_7-4S_1885.4 Kobbe 71.25363 22.31817 Hammerfest Basin (south) Caledonian C C

7122_7-4S_2057.9 Klappmyss 71.25363 22.31817 Hammerfest Basin (south) Caledonian C C C



7124_3-1_1316.4 Fruholmen 71.76001 24.78055 Nyslepp/Måsøy Fault ComplexNyslepp/Måsøy Fault ComplexNyslepp/MåsøyFault ComplexNyslepp/Måsøy FaultComplex

Caledonian C C C C

7124_3-1_1337.5 Fruholmen 71.76001 24.78055 Nyslepp/Måsøy Fault ComplexNyslepp/Måsøy Fault ComplexNyslepp/MåsøyFault ComplexNyslepp/Måsøy FaultComplex

Mixed C C

7124_3-1_1349.35 Fruholmen 71.76001 24.78055 Nyslepp/Måsøy Fault ComplexNyslepp/Måsøy Fault ComplexNyslepp/MåsøyFault Complex

Caledonian C C C

7124_3-1_1350 Fruholmen 71.76001 24.78055 Nyslepp/Måsøy Fault ComplexNyslepp/Måsøy Fault ComplexNyslepp/MåsøyFault Complex

Caledonian C


Caledonian C C


Mixed C


Caledonian C


Caledonian C


Mixed C C

7131_4-1_1080.5 Snadd 71.69472 31.01138 Finnmark Platform Uralian C C C

7131_4-1_1100.1 Snadd 71.69472 31.01138 Finnmark Platform Uralian C C

7131_4-1_920.1 Fruholmen 71.69472 31.01138 Finnmark Platform Caledonian C

7131_4-1_941.5 Fruholmen 71.69472 31.01138 Finnmark Platform Caledonian C C C

7219_9-1_2747.4 Fruholmen 72.40022 19.95324 Bjørnøya BasinBjørnøya BasinBjørnøyaBasin

Mixed C C C C

7219_9-1_2760.4 Fruholmen 72.40022 19.95324 Bjørnøya BasinBjørnøya BasinBjørnøyaBasin

Mixed C

7224_7-1_1725.8 Kobbe 72.28509 24.30083 Bjarmeland Platform Uralian C C

7224_7-1_1734.3 Kobbe 72.28509 24.30083 Bjarmeland Platform Uralian C C

7226_11-1_2141.7 Kobbe 72.23838 26.47911 Nordkapp Basin Uralian C C

7226_11-1_3058.3 Havert 72.23838 26.47911 Nordkapp Basin Uralian C

7226_11-1_3067.5 Havert 72.23838 26.47911 Nordkapp Basin Uralian C C C

7226_11-1_3079.2 Havert 72.23838 26.47911 Nordkapp Basin Uralian C C

7228_2-1_4190.4 Havert 72.85271 28.44156 Bjarmeland Platform Uralian C

7228_7-1A_2059.3 Snadd 72.25803 28.14992 Nordkapp Basin Uralian C C C C

7228_7-1A_2062.1 Snadd 72.25803 28.14992 Nordkapp Basin Uralian C C C C

E.J. Fleming et al. / Marine and Petroleum Geology 78 (2016) 516e535520

Table 1 (continued )

Sample No Formation Latitude Longitude Region Provenance Petrography HM garnet rutile zircon

7228_7-1A_2062.1b Snadd 72.25803 28.14992 Nordkapp Basin Uralian C

7228_7-1A_2073.4 Snadd 72.25803 28.14992 Nordkapp Basin Uralian C

7228_7-1A_2090 Snadd 72.25803 28.14992 Nordkapp Basin Uralian C

7228_7-1A_2096.2 Snadd 72.25803 28.14992 Nordkapp Basin Uralian C

7228_7-1A_2846.6 Klappmyss 72.25803 28.14992 Nordkapp Basin Uralian C C

7228_7-1A_2858.7 Klappmyss 72.25803 28.14992 Nordkapp Basin Uralian C C

7228_9-1S_2871.6 Havert 72.39677 28.71908 Finnmark Platform Uralian C C

7321_7-1_2387.7 Snadd 73.43210 21.07549 Bjørnøya Basin Uralian C C

7321_8-1_1480 Fruholmen 73.33666 21.41591 Bjørnøya BasinBjørnøya Basin Mixed C

7321_8-1_1510 Fruholmen 73.33666 21.41591 Bjørnøya BasinBjørnøya Basin Mixed C

7321_8-1_1538.1 Fruholmen 73.33666 21.41591 Bjørnøya Basin Mixed C

7321_8-1_2840.6 Snadd 73.33666 21.41591 Bjørnøya Basin Uralian C C

7324_10-1_1664.4 Kobbe 73.16374 24.31323 Bjarmeland Platform Uralian C

7324_10-1_1666.8 Kobbe 73.16374 24.31323 Bjarmeland Platform Uralian C

7324_10-1_2640.4 Havert 73.16374 24.31323 Bjarmeland Platform Uralian C C

7324_10-1_2653.7 Havert 73.16374 24.31323 Bjarmeland Platform Uralian C

7324_10-1_2656.3 Havert 73.16374 24.31323 Bjarmeland Platform Uralian C C


or Type B fields on the discrimination diagram of Mange andMorton (2007), suggesting more of a derivation from amphibolitefacies metasediments. In the Fruholmen Formation, sample7124_3-1_1349.35 (Fig. 4d) from the Nyslepp/Måsøy Fault Complexhas a mixed garnet geochemistry, with garnets plotting within theType Ai, Aii B and Bi fields.

4.4. Rutile geochemistry

Detrital rutile geochemical data were obtained from 10 samples(Table 1 and Fig. 5). In the southern part of the Hammerfest Basin,samples from the Havert (7122_7-3_2520.2) and Kobbe formations(7122_7-4s_1805.5 and 7122_7-4s_1817.9) have predominantlymetapelitic rutiles (x ¼ 68%) and have an average crystallisationtemperature of 640 �C. In contrast, samples from the Kobbe(7224_7-1_1725.8) and Snadd formations (7228_7-1A_2059.3)situated away from the southern part of the Hammerfest Basin havedominantly metamafic rutiles (x ¼ 79%) and a slightly lowercalculated average crystallisation temperature of 615 �C. Samplesfrom the Fruholmen Formation display variable rutile varieties withmean crystallisation temperatures similar to those from otherTriassic formations.

4.5. Detrital zircon U-Pb geochronology

U-Pb detrital zircon geochronology was carried out on 8 sam-ples (Table 1). In addition to the histogram and probability dia-grams displayed on Fig. 6, the data are given as the percentageswithin key age ranges are given in the supplementary materials,and these further highlight similarities and differences between thesamples.

Two samples were analysed from the Kobbe Formation in thesouthern part of the Hammerfest Basin. Sample 7122_7-4s_1805.5and 7122_7-4s_1817.9 (Fig. 6a and b) display a broad Proterozoiczircon age distribution. A significant Late Neoproterozoic (c. 600Ma) population is recorded in both samples. OtherMesoproterozoicpopulations are recorded between 1000 Ma and 1250 Ma and abroad Palaeoproterozoic age distribution are seen between 1500Ma and 1800 Ma. Finally, a minor Archaean population is observedin both samples.

In the Snadd Formation, detrital zircon age analysis was carriedout on 2 samples 7228_7-1A_2059.25 and 7228_7-1A_2062.1 fromthe Nordkapp Basin (Fig. 6c and d). Both yield broadly similardetrital zircon age spectra. In contrast to the Kobbe Formation inthe southern part of the Hammerfest Basin, the samples contain

abundant grains with Permian and Carboniferous ages between280 Ma and 370 Ma. A minor population of Late Neoproterozoic/Early Cambrian ages between 520 Ma and 570 Ma was also recor-ded. Grains with older Proterozoic and Archaean ages are rare.

In the Fruholmen Formation, the ages of detrital zircons weremeasured on 4 samples: Sample 7219_9-1_2747.4 (Fig. 6e) is fromthe Bjørnøya Basin, Sample 7122_7-1_1162.4 (Fig. 6f) is from thesouthern part of the Hammerfest Basin and samples 7124_3-1_1316.4 (Fig. 6g) and 7124_3-1_1349.35 (Fig. 6h) are from theNyslepp/Måsøy Fault Complex. A diverse zircon age spectrumcontaining significant Carboniferous c. 340 Ma and Late Neo-proterozoic 550e600 Ma populations was observed in the samplefrom the Bjørnøya Basin (Sample 7219_9-1_2747.4). In contrastgrains younger than 900Ma are rare or absent in the other samples.Instead, these samples have grains which yield a spread of Meso-proterozoic and Palaeoproterozoic ages with a significant zirconpopulation at c. 1640 Ma.

5. Provenance interpretation of the Barents Shelf Triassicsuccession

Through the integration of the petrographic analyses, heavymineral analyses, single grain garnet and rutile geochemical ana-lyses and detrital zircon U-Pb geochronology, it is possible toidentify different sand types. Within the Havert, Klappmyss, Kobbeand Snadd formations, two contrasting groups of samples are seenin the provenance results (Fig. 7), as summarised and interpreted inthe following sections.

5.1. Pre-Early Norian Triassic sand types

5.1.1. The Caledonian Sand TypeA distinct group of samples is seen fromwells that are located in

the southern Hammerfest Basin. They are characterised by a highdegree of compositional maturity (Fig. 7c). They have a low pro-portion of rock fragments, which are mostly plutonic in origin, andthe reservoir quality is generally good or excellent, with meanvisible porosity at 14%. The heavy mineral assemblage in thesesamples is dominated by garnet, and chrome-spinel contents arelow resulting in high GZi values and CZi values at or close to zero(Fig. 7a). Rutiles have compositions consistent with a metapeliticsource (Fig. 7d) and most of the detrital garnets plot within theType A fields (Fig. 7b). The detrital zircon spectrum (Fig. 7e) dis-plays a significant Late Neoproterozoic population, in addition toolder Meso and Palaeoproterozoic and Archaean populations;

Q

P K

Arko

se

Lith

ic a

rkos

e

Arkose litharenite

Litharenite

Q

F K

Sub-

arko

se

QuartzArenite

Sub-litharenite

Q

KP

Arko

se

Lith

ic a

rkos

e

Arkose litharenite

Litharenite

Q

F R

Sub-

arko

se

QuartzArenite

Sub-litharenite

Hammerfest Basin (south)Hammerfest Basin (central and north)Nyslepp/Måsøy Fault ComplexBjarmeland Pla ormNordkapp BasinBjørnøya BasinFinmark Pla orm

FruholmenHavertKlappmyssKobbeSnadd

0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 % 100 %

7122_7-1_1141.37122_7-1_1153.57122_7-1_1163.7

7122_7-4S_1185.57122_7-4S_1198.57122_7-4S_1210.5

7122_6-1_2065.47122_6-1_2093.47124_3-1_1337.5

7124_3-1_1349.357124_3-1_1355.47124_3-1_1380.3

7125_4-1_904.57219_9-1_2747.4

7131_4-1_941.5

Hammerfest Basin (south)

Bjørnøya BasinFinnmark Pla orm

Fruholmen Fm.Hammerfest Basin (central and north)

Nyslepp/Måsøy Fault Complex

7122_7-3_2520.27324_10-1_2656.37226_11-1_3067.57226_11-1_3079.27324_10-1_2640.47228_9-1S_2871.6

Hammerfest Basin (south)Bjarmeland Pla orm

Nordkapp Basin


Havert Fm.

7122_7-4S_2057.97228_7-1A_2846.67228_7-1A_2858.7


Nordkapp BasinKlappmyss Fm.

7120_12-1_3521.77122_7-4S_1885.4

7122_7-5_1902.37122_7-5_1908.67224_7-1_1734.3

7226_11-1_2141.7


Bjarmeland Pla ormNordkapp Basin

Kobbe Fm.

7122_7-3_1187.87121_5-1_3097.47122_6-2_2499.9

7228_7-1A_2062.17228_7-1A_2096.2

7321_7-1_2387.77321_8-1_2840.67131_4-1_1080.57131_4-1_1100.1


Nordkapp Basin

Bjørnøya Basin

Finnmark Pla orm

Snadd Fm.

Hammerfest Basin (central and north)

Loca

Forma

(a) (b)

(c)

QuartzQuartz Plagioclase Alkali Feldspar Rock FragmentsRock Fragments

Pre-Early Norian MFS

Post-Early Norian MFS

Fig. 2. Petrographic analysis of sandstone samples from the Triassic formations. a) Ternary diagram showing the relative proportions of quartz (Q), feldspar (F) and rock fragments(R) with fields representing the sandstone classification of Folk (1980). b) Ternary diagram showing the relative proportions of quartz (Q), plagioclase feldspar (P) and K-feldspar (K).The upper plots show data from the Fruholmen Formation only, whereas the lower plots shows data from the other formations. c) Normalised stacked bar charts showing therelative proportions of quartz, plagioclase, K-feldspar and rock fragments.

ATi vs. RuZi

ATi

ATi

RuZi

RuZi

0

0

20

20

40

40

60

60

80

80

100

100

100

100

80

80

60

60

40

40

20

20

0

0

ATi vs. GZi

ATi

ATi

GZi

GZi

0

0

20

20

40

40

60

60

80

80

100

100

100

100

80

80

60

60

40

40

20

20

0

0

CZi vs. RuZi

CZi

CZi

RuZi

RuZi

0

0

10

10

20

20

30

30

40

40

50

50

60

60

100

100

80

80

60

60

40

40

20

20

0

0

7122_7-1_1141.37122_7-1_1148.57122_7-1_1153.57122_7-1_1162.47122_7-1_1163.77122_7-1_1169.57122_7-1_1173.6

7122_7-4S_1185.57122_7-4S_1194.37122_7-4S_1198.57122_7-4S_1210.5

7122_6-1_2065.47122_6-1_2093.47124_3-1_1316.47124_3-1_1337.5

7124_3-1_13507124_3-1_1355.47124_3-1_1403.7

7125_4-1_889.07125_4-1_904.5

7219_9-1_2747.47219_9-1_2760.4

7321_8-1_14807321_8-1_1510

7321_8-1_1538.17131_4-1_920.17131_4-1_941.5


Bjørnøya Basin

Finnmark Pla orm

Fruholmen Fm.


Nyslepp/Måsøy Fault Complex

7122_7-3_2520.27228_2-1_4190.4

7324_10-1_2653.77324_10-1_2656.37226_11-1_3058.37226_11-1_3067.57226_11-1_3079.27324_10-1_2640.47228_9-1S_2871.6


Bjarmeland Pla orm

Nordkapp Basin


Havert Fm.

7122_7-4S_2057.97228_7-1A_2846.67228_7-1A_2858.7

Hammerfest Basin (south)Nordkapp BasinKlappmyss Fm.

7120_12-1_3521.77122_7-3_1813.27122_7-3_1821.77122_7-3_1828.8

7122_7-4S_1805.57122_7-4S_1811.47122_7-4S_1817.97122_7-4S_1885.4

7122_7-5_1902.37122_7-5_1908.67224_7-1_1725.87224_7-1_1734.3

7324_10-1_1664.47324_10-1_1666.87226_11-1_2141.7


Bjarmeland Pla orm

Nordkapp Basin

Kobbe Fm.

7122_7-3_1187.87121_5-1_3097.4

7228_7-1A_2059.37228_7-1A_2062.1

7228_7-1A_2062.1b7228_7-1A_2073.4

7228_7-1A_20907321_7-1_2387.77321_8-1_2840.67131_4-1_1080.57131_4-1_1100.1


Nordkapp Basin

Bjørnøya Basin

Finnmark Pla orm


Snadd Fm.

(a. i) (a. ii) (a. iii)

(b. i) (b. ii) (b. iii)

(c)

FruholmenHavertKlappmyssKobbeSnadd

Forma

0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 % 100 %

Apa e Anatase Calcic amphibole Clinopyroxene Chrome spinel Chloritoid EpidoteGahnite Garnet Kyanite Monazite Titanite Staurolite Tourmaline Zircon

Pre-Early Norian MFS

Post-Early Norian MFS

Fig. 3. Heavy mineral characteristics of the Triassic formations. Cross plots of the heavy mineral ratios for samples from the pre (a) and post (b) Early Norian MFS showing (i)RuZi:ATi, (ii) RuZi:CZi and (iii) GZi:ATi. c) Normalised stacked bar charts showing the heavy mineral assemblage.

Fig. 4. eTernary diagram of garnet compositions showing the relative proportions of Pyrope (XMg), Grossular (XCa) and Almandine- Spessartine (XFeþMg) garnets. Filledsymbols ¼ XMn > 5%. Also shown are pie charts with the relative portion of the different garnet types according to the subdivisions of Mange and Morton (2007).


however, Palaeozoic zircons are absent.The broad detrital zircon age spectrum, combined with the

geographic distribution of these samples, suggests a source from

Northern scandinavia. The grouping is referred to as the CaledonianSand Type despite the absence of grains with ages that coincidewith the main Caledonian magmatic and tectonic events (Bingen

Fig. 5. Rutile geochemistry of the analysed samples showing the relative abundance of metapelitic and metamafic rutile and their peak metamorphic temperatures. Discriminationof rutile provenance (metamafic or metapelitic) is after Meinhold et al. (2008) and the temperature calculations are from Watson et al. (2006).


and Solli, 2009). This is because in northern Scandinavia the orogencontains few felsic Palaeozoic intrusions but is instead formed ofthick Neoproterozoic and Early Palaeozoic successions of theCaledonian Allochthons and Autochthons (Sturt and Austrheim,1985). It is thought that many of the grains were recycled fromthese units because the combined signal from the Caledoniansuccessions yields a detrital zircon pattern that is most similar tothe offshore samples (Fig. 8).

5.1.2. The Uralian Sand TypeA second group of samples is present with provenance charac-

teristics that are distinct from those from the Caledonian SandType. These samples come from wells that are located in areasnorth of the southern part of the Hammerfest Basin and are char-acterised by low compositional maturity (Fig. 7c). The reservoirproperties are quite variable, with an average visible porosity of 7%.The proportion of rock fragments is high and they are mostly of

Fig. 6. Histogram of U-Pb ages and relative probability curve for detrital zircon. (i) 0-3.5Ga (50 Ma bin width) and (ii) 0e800 Ma (25 Ma bin width). Numbers above the probabilitycurve equate to the approximate ages of the most significant zircon populations. The grey bars represent periods of magmatic and metamorphic activity within the FennoscandianShield and Caledonides of northern Norway. The red bars represent periods of magmatic and metamorphic activity within the Uralian Orogeny whilst the blue bars represent theTimanian Orogeny. The deeper colours represent higher zircon fertility. (For interpretation of the references to colour in this figure legend, the reader is referred to the web versionof this article.)

Fig. 7. Summary of the different characteristics of the Caledonian and Uralian sand types seen in the Havert, Klappmyss, Kobbe and Snadd Formations. a) Cross plots of the heavymineral ratios for (i) RuZi:ATi, (ii) RuZi:CZi and (iii) GZi:ATi. b) Ternary diagram showing the relative proportions of Pyrope (XMg), Grossular (XCa) and Almandine- Spessartine(XFeþMg) garnets, using the subdivisions of Mange and Morton (2007). Filled symbols ¼ XMn > 5%. c) Ternary diagram showing the relative proportions of quartz (Q), feldspar (F) androck fragments (R) with fields representing the sandstone classification of Folk (1980). d) Rutile geochemistry of the analysed samples showing the relative abundance of metapeliticand metamafic rutile and their peak metamorphic temperatures. e) Histogram of U-Pb ages and relative probability curve for detrital zircon. (i) Caledonian Sand Type and (ii)Uralian Sand Type (20 Ma bin width). The grey bars represent periods of magmatic and metamorphic activity within the Fennoscandian Shield and Caledonides of northern Norway(deeper colours represent higher zircon fertility). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)


Fig. 8. Histogram of U-Pb ages and relative probability curve for detrital zirconscomparing the data from this study on the southwest Barents Shelf for (a) the Cale-donian Sand Type deposited after the Norian Flooding surface with (b) CaledonianSand Type deposited prior to the Early Norian MFS, (c) the Barents Sea Group (Zhanget al., 2015), (d) the Neoproterozoic e Early Palaeozoic Paraautochthon and Autoch-thon of northern Norway (Andresen et al., 2014; Kirkland et al., 2008; Zhang et al.,2015) and (e) the Kalak Nappe Complex of north Norway (Kirkland et al., 2007, 2008).


sedimentary origin (mudstone and sandstone). These samples alsohave a distinct heavy mineral assemblage characterised by a lowabundance of garnet but a high proportion of apatite and chrome-spinel, which results in high CZi and ATi values (Fig. 7a). Most of thedetrital rutiles have metamafic compositions and a higher portionof the garnets fall in the Type B field compared with those from theCaledonian Sand Type. The detrital zircons are predominantly LateCarboniferous and Permian in age (Fig. 7e). Proterozoic andArchaean grains are much less abundant.

The occurrence of Permian and Carboniferous zircons and highchrome-spinel contents, combined with the extensive seismic ev-idence of north-easterly prograding clinoforms in these areas

(Glørstad-Clark et al., 2010; Klausen et al., 2014; Lundschien et al.,2014; Riis et al., 2008) are consistent with a source for this sandtype from the Uralian Mountains and magmatic rocks associatedwith the Uralian Orogeny. The high chrome-spinel contents arefrom ultramafic rocks, present along the length of the origin (e.g.Ivanov et al., 2013). The important population of c. 310 Ma zirconsmay have originated from the continental arc rocks which occurwithin tectonic units which were accreted to the Siberian Craton(Fershtater, 2013; Fershtater et al., 2007; Smirnov et al., 2014).These units were exhumed during the suturing of Baltica withSiberia and the final phases of the Uralian Orogeny. The exactsource area along the length of the Uralian Orogeny remainsuncertain.

5.2. Post-Early Norian Triassic sand types

In the Havert to Snadd formations, a clear distinction is seenbetween the Caledonian and Uralian sand types. However, abovethe Early Norian MFS, a significant change in sandstone texture andcomposition occurs in the Fruholmen Formation. All geographicareas display an increase in compositional maturity and extremecompositional maturity is seen in some samples (Fig. 2a). Rockfragments are mostly sedimentary and reservoir properties aregenerally good or excellent. Samples display a large increase in theaverage proportion of ultra-stable heavy minerals (e.g. Zircon andtourmaline; Fig. 3c). As a consequence, ATi and CZi values are low(Fig. 3b i - b.iii). There is some geographical variation in composi-tion. Samples from the Bjørnøya Basin, in particular, are less matureand have a greater proportion of apatite and chrome-spinel andthus high CZi and ATi values, akin to those recorded from theUralian Sand Type. The garnet geochemistry (Fig. 4) of samplesfrom the Fruholmen Formation is a mixture between those seen inthe Caledonian and Uralian sand types in older units with garnetsfalling in the Type Ai, Aii, B and Bi fields (Fig. 4d). Rutiles are mostlymetapelitic (Fig. 5).

The zircon signature of the Fruholmen Formation is variable(Fig. 6). Samples from the Hammerfest Basin and Nyslepp/MåsøyFault Complex display abundant Mesoproterozoic and Palae-oproterozoic populations (Fig. 6f, g and h) which are typical of theCaledonian Sand Type and the (meta) sedimentary successions ofnorthern Scandinavia. The sample from the Bjørnøya Basin (Fig. 6e)displays amuchmoremixed zircon age spectrum. In addition to theolder populations, the sample displays two dominant youngerpeaks: one corresponding to the main Uralian Orogen seen withinthe Uralian Sand type and one resembling the Neoproterozoicpopulation evident within the Caledonian Sand Type of thesouthern Hammerfest Basin.

The conspicuous textural and compositional changes that occurbetween the Snadd and Fruholmen formations suggest a sharpincrease in the degree of weathering. Climate plays a major role incontrolling weathering (Morton and Hallsworth, 1999); however,whilst there is a general trend tomore humid climatic conditions inthe Late Triassic Boreal Realm (Decou et al., 2016; Hochuli andVigran, 2010), the changes in the zircon age spectra cannot easilybe attributed to climatic variations. Instead the data suggest thatthe Early Norian MFS marks the start of a change in provenance.

Major tectonic activity, such as the uplift of Novaya Zemlya(Scott et al., 2010), is postulated in areas surrounding the BarentsShelf in the Late Triassic. In addition, uplift and hinterland rejuve-nation of the Finnmark Platform has been suggested by variousauthors (Mørk, 1999; Ryseth, 2014; Worsley, 2008). This wouldhave had a pronounced effect on the source-to-sink relationship ofthe southwest Barents Shelf. A continued sediment input from theCaledonian source area combined with the recycling of the un-derlying Uralian and/or Caledonian sand types in the Triassic due to


the hinterland rejuvenation provides a likely explanation for themixed signature seen in the Fruholmen Formation from theBjørnøya Basin.

By comparing the zircon spectrum from the Caledonian SandType with that from the Fruholmen Formation (excluding thesample with a mixed signal from the Bjørnøya Basin), some dif-ferences are evident (Fig. 8). The zircon pattern from the FruholmenFormation (Fig. 8a) closely resembles that from the Kalak NappeComplex (Fig. 8e), whereas the samples of the Caledonian SandType deposited before the Early Norian MFS (Fig. 8b) have a patternmore akin to a mixture of the signatures that characterise KalakNappe Complex and the Para-autochthon and Autochthon of thenorthern Fennoscandian Shield (Fig. 8d). A Kalak Nappe Complexsource for the Fruholmen Formation samples from the HammerfestBasin is consistent with their unusually high proportion of kyanitebecause this nappe complex contains kyanite-bearing rocks(Kirkland et al., 2007). By contrast, the increased proportions ofapatite and chrome-spinel and Uralian-aged zircons in samplesfrom the Bjørnøya Basin could be considered consistent with a firstcycle origin from the Uralian Orogeny. However, these samples aremore compositionally mature than the typical Uralian Sand Typeand therefore more consistent with a recycled origin.

6. Reconstruction of Triassic dispersal patterns

Samples from each of the 5 main transgressive-regressive cyclesare assigned a sand type and in combinationwith sedimentologicaldata from offshore boreholes and published seismic data (Høy andLundschien, 2011; Klausen et al., 2014; Lundschien et al., 2014; Riiset al., 2008) sand dispersal reconstructions are made in Fig. 10.These reconstructions show the interpreted palaeogeography atmaximum shoreline progradation with the shoreline positionsfrom the Induan to Norian are after Glørstad-Clark et al. (2010).

The Induan (Fig. 10a) marks the start of the basin organisationthat continued to the early Norian and developed following LatePermian tectonic movements associated with the Uralian Orogeny.Most of the wells analysed during this period display the UralianSand Type, associated with the major deltaic complexes that orig-inated from the Uralian Orogen, depicted seismically from north-westerly prograding clinoforms (Anell et al., 2014; Glørstad-Clarket al., 2010; Høy and Lundschien, 2011; Riis et al., 2008; van Veenet al., 1993). One well (7122/7-3), restricted to the southernmargin of the Hammerfest Basin, displays the Caledonian SandType suggesting a second fluvial system delivered sand fromnorthern Scandinavia, supported by the presence of northwardprograding clinoforms in the Induan (Glørstad-Clark et al., 2011;van Veen et al., 1993).

The basin organisation continued during the Olenekian regres-sion (Fig. 10b). Clinoform geometries suggest the coastline reachedas far as the eastern margins of the Loppa High (Glørstad-Clarket al., 2011). There is a slight shift of the depocentre towards theBjarmeland Platform and minor progradation of the delta com-plexes towards the northwest. The provenance results reveal thatthe two major fluvial systems continued to be active during thistime. Most of thewells display the Uralian Sand Type. However, onewell (7122/7-4s), in the southern Hammerfest Basin, displays theCaledonian Sand Type implying a second fluvial system deliveredsand from northern Scandinavia, probably corresponding to thenorthward prograding clinoform belts seen in the HammerfestBasin (e.g. Glørstad-Clark et al., 2010). Further east, northwardprograding clinoforms are also observed on the Finnmark Platform(Glørstad-Clark et al., 2011) and these systems probably alsodelivered the Caledonian Sand Type to these areas although thiscannot be proven with the present sample and well coverage.

The Anisian to early Ladinian sequence (Fig. 10c) was

accompanied by a westward shift in the depocentre and subse-quently an eastward-thinning of sediment thickness, representinga stepwise infilling of the western Barents Shelf (Glørstad-Clarket al., 2010). The shoreline extended over the top of the LoppaHigh and migrated westwards. The provenance results reveal thatthe two major fluvial systems continued to be active during thistime. Most of the western Barents Shelf was dominated by themajor delta complexes that prograded from the southeast, resultingin the widespread deposition of the Uralian Sand Type. However,the Caledonian Sand Type is seen in various samples in the Goliatfield area and further west in well 7120/12-1. These match theclinoform geometries in the Hammerfest Basin (Glørstad-Clarket al., 2010) and imply a second fluvial system delivered the Cale-donian Sand Type to the southern Hammerfest Basin.

Progressive, stepwise infilling of the basin continued throughthe Ladinian and early Carnian (Fig. 10d). The depocentre shiftswest- and northwards, towards Hopen and the Edgeøya Platform(Glørstad-Clark et al., 2010). Further south, a significant depocentreoccurred over the top of the Loppa High and a westward migrationof the shoreline is observed. The exact position of the shorelineduring this period is poorly constrained due to problems inter-preting clinoform geometries in the latest Carnian because ofincreasingly shallow water deposition. Whilst we place the shore-line immediately west of the Loppa High (Fig. 10d), it is plausiblethat the epicontinental sea may have closed completely at times.The provenance results reveal that the two major fluvial systemscontinued to be active during this time; however, the systemassociated with the Uralian was clearly dominant as almost all ofthe samples from wells during this period show the Uralian SandType. One well from the Goliat field area (7122/7-3) displays theCaledonian Sand Type, indicating that the second fluvial systemstill delivered sediment to the southern Hammerfest Basin.

A major provenance change occurs across the early Norian MFS(Fig. 10e) in response to basin reorganisation which coincided withthe development of the Slottet Bed in Svalbard (Mørk et al., 1999).Up until this boundary, the Triassic was dominated by the UralianSand Type and the Caledonian Sand Type was restricted to thesouthern Hammerfest Basin. However, above this boundary theCaledonian Sand Type becomes dominant and supply of the UralianSand Type is impeded. Hinterland rejuvenation resulted in wide-spread sediment recycling and fluvial systems, sourced from theKalak Nappe Complex in Finnmark, supplied sediment to extensiveregions over the Finnmark Platform and into more northerly areas.Some samples display elevated CZi values and zircon ages associ-ated with the Uralian Orogeny. In most cases, these samples can beexplained by mixing of the two sand types by recycling of an un-derlying Uralian signature.

7. Comparisons with the De Geerdalen Formation in Svalbard

The De Geerdalen Formation in Svalbard is often used as ananalogue for the Snadd Formation on the Barents Shelf (e.g.Harishidayat et al., 2015; Klausen and Mørk, 2014; Lord et al.,2014b). It was initially thought to have been sourced from thenortheast, possibly from an area of land not currently exposed butthought to have existed in the Mesozoic, known as Crockerland(Embry, 1993). However, recent seismic evidence has identifiednorth-westerly prograding clinoforms offshore Svalbard and else-where on the Barents Shelf (Høy and Lundschien, 2011; Riis et al.,2008). As such, deposition of the De Geerdalen Formation in Sval-bard is widely interpreted to have been associated with thesesouth-easterly sourced prograding units (Lundschien et al., 2014;Rød et al., 2014).

The detrital zircon age data for the De Geerdalen Formation(Fig. 9b) reveals a ~300 Ma population which corresponds with the

Fig. 9. Histogram of U-Pb ages and relative probability curve for detrital zircons comparing the data from this study on the southwest Barents Shelf for (a) the Uralian Sand Typewith (b) data from Svalbard (Festningen, Hopen and Edgeøya; P�ozer Bue and Andresen, 2014) and (c) data from Franz Josef Land (Soloviev et al., 2015). Numbers above theprobability curve equate to the approximate ages of the most significant zircon populations.


main Uralian Orogeny and matches the dominant populationrecorded in the Snadd Formation from the southwest Barents Shelf.However, significant c. 235e240 Ma and c. 420e470 Ma zircon agepopulations are also seen. Zircons of this age are unusual as they arelargely absent from the samples on the southwest Barents Shelf(Fig. 9a) and sources for these zircons are not currently evidentwithin the Uralian Orogen or Novaya Zemlya. Similar populationsare observed in Late Triassic samples from Franz Josef Land (Fig. 9c)and the age spectra in these areas can be considered cognate.

A possible explanation is that zircons of this age were sourcedfrom the northeast. Syenitic to granitic magmatism of this age hasbeen reported in Taimyr within the western part of the Taimyr fold-and-thrust belt and this magmatism was possibly associated withthe late stages of the Siberian Traps (Vernikovsky et al., 2003). Inaddition, intrusive rocks with ages between 411 and 470 Ma occuron Severnaya Zemlya (see Lorenz et al., 2008 and referencestherein) which could provide an explanation for the Ordovician andSilurian zircon populations.

The seismic evidence for progradation to the northwest towardsSvalbard in the Carnian is well documented (Glørstad-Clark et al.,2010; Høy and Lundschien, 2011; Lundschien et al., 2014; Riiset al., 2008) and is not in dispute. The provenance data from theSnadd Formation presented here is supportive of this interpreta-tion. However, the zircon signature of samples from the De Geer-dalen Formation on Svalbard and Late Triassic sandstones on FranzJosef Land are not cognate with the Snadd Formationwhich impliesthe two areas had different sources. As such, an additional system isproposedwhich delivered sand to Svalbard from source areas to the

northeast.

8. Implications for the tectonic evolution of the Barents Shelf

The data are placed within a wider sedimentary provenancediagram for maximum regression in the early Norian in Fig. 11. Theresults have several important implications for the tectonic devel-opment of the Barents Shelf, particularly surrounding the timingand development of Novaya Zemlya. Until the early Norian, sedi-ment sourced from the Uralian Orogen must have crossed the EastBarents Basin to reach the southwest Barents Shelf. This impliesthat the East Barents Basin and any possible embayment areas tothe east prior to the development of Novaya Zemlyawere not actingas effective sediment sinks. This suggests that sediment supplymust have been greater than subsidence in the basin allowing theUralian Sand Type to fill and bypass these basins.

The provenance change which occurs across the Early NorianMFS is interpreted to mark a major change in basin organisation.The Caledonian Sand Type which was previously restricted to thesouthern margins of the Hammerfest Basin became extensive andflowed northwards over the Finnmark Platform and HammerfestBasin and beyond into more northerly basin areas. This boundaryalso marks a change in sedimentary style from one dominated bydelta progradation to one of shallow shelf conditions. This marked anew phase of basin organisation that continued into the Jurassicand facilitated the deposition of mature sandstones of such as thoseof the Stø Formation (main reservoir of the Snøhvit gas field). Thesechanges coincide with major tectonic activity in areas surrounding

Fig. 10. Reconstruction of the palaeogeography and sandstone dispersal patterns for the southwest Barents Shelf during (a) the Induan (Havert Formation), (b) Olenekian(Klappmyss Formation), (c) Ladinian (Kobbe Formation), (d) Carnian (Snadd Formation) and (e) Rhaetian (Fruholmen Formation).

Fig. 11. Summary of the wider implications of the provenance results for northwest Pangaea in the late Carnian at maximum regression. The tectonic plate configuration wasconstructed using GPlates for plate reconstruction at 220 Ma using data from Müller et al. (2016). The palaeogeography is adapted after Scott et al. (1996) and Smelror et al. (2009).Also shown are the locations of samples used in Fig. 9.


the Barents Shelf, such as the uplift of Novaya Zemlya (Scott et al.,2010). In Late Triassic to Early Jurassic times, uplift is also docu-mented in Eastern Finnmark and the Kola Peninsula (Hendriks andAndriessen, 2002; Veselovskiy et al., 2015) and it is inferred to haveoccurred in other areas on the central and northern Barents Shelf.

Such tectonic activity would have had two significant conse-quences. Firstly, uplift of the Finnmark Platform and Kola Peninsulawould have created a barrier and prevented sediment derived fromthe Urals reaching the southwest Barents Shelf. Secondly, as Triassic

tectonism on Novaya Zemlya progressed, the East Barents Basinwould have transitioned into a forebulge basin resulting inincreased subsidence and the deposition of thick Jurassic andCretaceous successions in that region. The barrier created by upliftof the Finnmark Platform combined with increased subsidence inthe East Barents Basin would have acted to divert sedimentnorthward, away from thewestern Barents Shelf possibly into areasadjacent to the developing Novaya Zemlya thrust belt. Uplift of theFinnmark Platform would have also resulted in the rejuvenation of


the hinterlands in the northern Scandinavian source area. Thesandstones contain few Archaean and Palaeoproterozoic zircons,suggesting that the crystalline basement of the FennoscandianShield contributed few zircons. Instead, the zircon signal may eitherhave been controlled by grains recycled from Caledonian molassedeposits, which are thought to have extended across the area(Larson et al., 1999) or were derived directly from the extensionKalak Nappe Complex on to the Finnmark Platform (Gernigon et al.,2014). The hinterland rejuvenation would have allowed fluvialunits to extend further north onto the southwest Barents Shelf thanthey had done previously. At the same time, the actions of sedi-mentary recycling of the pre-existing Caledonian and Uralian sandtypes and the erosion of coarser grained older units would haveincreased compositional maturity and grain size.

These tectonic changes may also provide an explanation for theapparent contradiction between the zircon spectra from the DeGeerdalen Formation in Svalbard and the offshore seismic data. Wesuggest that, in addition to the well-documented northwest cli-noform progradation that reached offshore Svalbard in the Carnian(Glørstad-Clark et al., 2010; Høy and Lundschien, 2011; Lundschienet al., 2014; Riis et al., 2008), a second delta complex was alsopresent that delivered a Northern Uraloid Sand Type, sourced fromTaimyr and routed via Severnaya Zemlya to the northern BarentsShelf (Fig. 11). Although this system may have initially been minor,by the late Carnian, sediment restriction in the southwest associ-ated with initial stress change during the development of NovayaZemlya combined with the Late Triassic uplift of Taimyr in thenortheast Barents may have resulted in renewed progradation ofthis system. This system dominated sedimentation on Svalbard inthe late Carnian and delivered the unusual Mid Triassic populationfrom sources in Taimyr and Silurian/Ordovician from sources inSevernaya Zemlya.

The poor documentation of this system to date may be due toproblems interpreting clinoform geometries in the latest Carnian,due to the increasingly shallow water deposition, along with a lackof publically available seismic coverage from the northeast BarentsShelf. If confirmed, this hypothesis may provide an explanation forthe evidence of marine conditions in Hopen (Hopen Member; Lordet al., 2014a) at the same time as continental deposition on Spits-bergen (Vigran et al., 2014) despite the former being in a moreproximal to the northwest prograding systems. In addition, thenorthern system may have continued past Svalbard to the north-eastern Sverdrup Basin which provides an explanation for Triassic-aged zircons in the area (Anfinson et al. 2016; Omma et al., 2011).Future field, provenance and geophysical data, especially fromSvalbard and the northern Barents Shelf offshore, are needed to testthese hypotheses.

9. Implications for petroleum exploration

As provenance has a fundamental control on the composition ofclastic reservoir units, the variation highlighted in this study hasimplications for hydrocarbon exploration on the Barents Shelf. TheTriassic sandstone reservoirs of the Goliat field display highporosity and good reservoir quality (Rodrigues Duran et al., 2013).With the considerable thickness of Triassic sediments furthernorth, in areas postulated to be favourable for play development(NPD, 2014), underpinned by the proven Triassic petroleum systemin the Goliat field, one would expect significant opportunity forhydrocarbon exploration. Despite this exploration outcomes havebeen mixed.

Whilst the cause of the poor reservoir quality is likely to becomplex, it is not coincidental that these more northerly explora-tion wells sit within the area that is dominated by the Uralian SandType. The poor reservoir quality is at least in part due to the

presence of less mature sandstones, deposited at a considerabledistance to the Uralian source area. These are in contrast to the highporosity, clean sandstones seen around the Goliat field area in theCaledonian Sand Type. This shows that until the Early Norian MFS,the reservoir characteristics of the Goliat discoveries are anexception to the dominant provenance pattern seen on thesouthwest Barents Shelf. As such, the existence of Triassic reser-voirs with similar characteristics in areas further north should notbe expected. After the Early Norian MFS, the more extensivedevelopment of depositional systems supplied by the composi-tionally mature Caledonian Sand Type means that the reservoirpotential of Norian, and where preserved Rhaetian, sedimentsshould not be overlooked.

10. Conclusions

In this study, Triassic sandstones on the southwest Barents Shelfhave been investigated through petrographic and heavy mineralanalysis, garnet and rutile geochemistry and U-Pb zircon geochro-nology. Using these provenance techniques, in combination withpublished sedimentological, provenance and seismic data, sand-stone dispersal patterns on the Barents Shelf have been recon-structed. The main outcomes are as follows:

� Until the Early NorianMFS, two sand types are recognised on thesouthwest Barents Shelf. The Caledonian Sand Type is charac-terised by a high compositional maturity, a heavy mineralassemblage dominated by garnet, a low CZi value, predomi-nantly metapelitic rutiles and mostly Proterozoic and Archaeandetrital zircon ages. It is interpreted to be sourced fromnorthernScandinavia. The Uralian Sand Type is characterised by a lowcompositional maturity, high CZi value, predominantly meta-mafic rutiles and Carboniferous zircon ages which are consistentwith an origin from the Uralian Orogen.

� During the Induan to Early Norian stages (Havert, Klappmyss,Kobbe and Snadd formations), most of the southwest BarentsShelf was dominated by the Uralian Sand Type. These resultscorrespond to the well documented northeasterly progradingTriassic clinoforms seen in the seismic data. In contrast, theCaledonian Sand Type is restricted to the southern HammerfestBasin.

� A significant provenance change occurs across the Early NorianMFS. In response to Late Triassic/Early Jurassic hinterland reju-venation and an increase in accommodation space in the EastBarents Basin, the Caledonian Sand Type becomes dominant andextends northwards out onto the southwest Barents Shelf. Themain Uralian supply from the southeast is impeded and divertedinto the East Barents Basin by the uplift of Novaya Zemlya.Recycling of Triassic units occurs at the basin margins, leading toa mixed Uralian/Caledonian signature in some samples.

� Discrepancies exist between the provenance signature of theUralian-sourced Snadd Formation and age-equivalent strata inSvalbard and Franz Josef Land, which contain abundant zirconswith mid-Triassic and Silurian-Ordovician ages. Sources forthese zircons within the Uralian Orogen or Novaya Zemlya areelusive; the most likely sources are Taimyr and SevernayaZemlya. This system became increasingly dominant in the lateCarnian and early Norian because the developing orogeny onNovaya Zemlya choked sediment supply from the southwestand resulted in a renewed phase of uplift in Taimyr in the LateTriassic/Early Jurassic.

� These results are important for reservoir quality prediction. Theprovenance data reveal that the sandstone reservoirs of theGoliat field are an exception to the provenance signature forTriassic sandstones seen further north on the Barents Shelf at


this time. Until the Early Norian MFS, the Caledonian Sand Type,with often good reservoir potential, is restricted to the southernHammerfest Basin. The remaining areas are dominated by theUralian Sand Type which displays more variable reservoirquality.

Acknowledgments

This work was undertaken as part of the CASP Barents ShelfProvenance Project. The sponsoring companies are thanked fortheir continued support of this work. In addition, the authors wouldlike to thank Stephen Vincent, two anonymous reviewers and theeditor, Sven Egenhoff, for their constructive comments. The Nord-sim ionmicroprobe facility operates as a joint Nordic infrastructure.This is Nordsim publication 475 and University of Cambridge,Department of Earth Sciences contribution esc.3758.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.marpetgeo.2016.10.005.

References

Andresen, A., Agyei-Dwarko, N.Y., Kristoffersen, M., Hanken, N.M., 2014. A TimanianForeland Basin Setting for the Late Neoproterozoic-early Palaeozoic Cover Se-quences (Dividal Group) of Northeastern Baltica. Geological Society SpecialPublication, pp. 157e175.

Anell, I., Midtkandal, I., Braathen, A., 2014. Trajectory analysis and inferences ongeometric relationships of an Early Triassic prograding clinoform succession onthe northern Barents Shelf. Mar. pet. Geol. 54, 167e179.

Anfinson, O.A., Embry, A.F., Stockli, D.F., 2016. Geochronologic constraints on thepermian-Triassic Northern source region of the Sverdrup Basin, Canadian Arcticislands. Tectonophysics.

Bergan, M., Knarud, R., 1993. Apparent changes in clastic mineralogy of the Tri-assiceJurassic succession, Norwegian Barents Sea: possible implications forpalaeodrainage and subsidence. In: Vorren, T.O., Bergsager, E., Dahl-Stamnes, Ø.A., Holter, E., Johansen, B., Lie, E., Lund, T.B. (Eds.), Norwegian Pe-troleum Society Special Publications. Elsevier, pp. 481e493.

Bingen, B., Solli, A., 2009. Geochronology of magmatism in the Caledonian andSveconorwegian belts of Baltica: synopsis for detrital zircon provenancestudies. Nor. J. Geol. 89, 267e290.

Decou, A., Andrews, S.D., Alderton, D.H.M., Morton, A., 2016. Triassic to EarlyJurassic climatic trends recorded in the Jameson Land Basin, East Greenland:clay mineralogy, petrography and heavy mineralogy. Basin Res.

Dor�e, A.G., 1992. Synoptic palaeogeography of the Northeast Atlantic seaway: latepermian to cretaceous. Geol. Soc. Lond. Spec. Publ. 62, 421e446.

Embry, A.F., 1993. Crockerland e the northwest source area for the Sverdrup Basin,Canadian Arctic islands. In: Vorren, T., bergsager, E., Dahl-Stamnes, Ø.A.,Holen, L.H., Johansen, B., Lie, E., Lund, T.B. (Eds.), Arctic Geology and PetroleumPotential. Elsevier, Amsterdam, pp. 205e216.

Fershtater, G.B., 2013. Magmatism in the epochs of ural paleocean closure andmobile belt development: composition, distinctive evolutionary features, andsources. Petrology 21, 181e202.

Fershtater, G.B., Krasnobaev, A.A., Bea, F., Montero, P., Borodina, N.S., 2007. Geo-dynamic settings and history of the Paleozoic intrusive magmatism of thecentral and southern Urals: results of zircon dating. Geotectonics 41, 465e486.

Folk, R.L., 1980. Petrology of Sedimentary Rocks, 2 ed. Hemphill PublishingCompany.

Frei, D., Gerdes, A., 2009. Precise and accurate in situ UePb dating of zircon withhigh sample throughput by automated LA-SF-ICP-MS. Chem. Geol. 261,261e270.

Gabrielsen, R.H., Frerseth, R.B., Jensen, L.N., Kalheim, J.E., Riis, F., 1990. Structuralelements of the Norwegian continental shelf. NPD Bull. No. 6, 33.

Gernigon, L., Br€onner, M., Roberts, D., Olesen, O., Nasuti, A., Yamasaki, T., 2014.Crustal and basin evolution of the southwestern Barents Sea: from Caledonianorogeny to continental breakup. Tectonics 33, 347e373.

Glørstad-Clark, E., Birkeland, E.P., Nystuen, J.P., Faleide, J.I., Midtkandal, I., 2011.Triassic platform-margin deltas in the western Barents Sea. Mar. pet. Geol. 28,1294e1314.

Glørstad-Clark, E., Faleide, J.I., Lundschien, B.A., Nystuen, J.P., 2010. Triassic seismicsequence stratigraphy and paleogeography of the western Barents Sea area.Mar. pet. Geol. 27, 1448e1475.

Golonka, J., Ford, D., 2000. Pangean (Late CarboniferouseMiddle Jurassic) paleo-environment and lithofacies. Palaeogeogr. Palaeoclimatol. Palaeoecol. 161, 1e34.

Haq, B.U., Hardenbol, J., Vail, P.R., 1987. The new chronostratigraphic basis ofCenozoic and Mesozoic sea level cycles. Cushman Found. Foraminifer. Res.

Special Publ. 24, 7e13.Harishidayat, D., Omosanya, K.d.O., Johansen, S.E., 2015. 3D seismic interpretation of

the depositional morphology of the middle to late Triassic fluvial system ineastern Hammerfest Basin, Barents sea. Mar. pet. Geol. 68, 470e479.

Hendriks, B.W.H., Andriessen, P.A.M., 2002. Pattern and timing of the post-Caledonian denudation of northern Scandinavia constrained by apatitefission-track thermochronology. Geol. Soc. Lond. Spec. Publ. 196, 117e137.

Henriksen, E., Ryseth, A.E., Larssen, G.B., Heide, T., Rønning, K., Sollid, K.,Stoupakova, A.V., 2011. Tectonostratigraphy of the greater Barents Sea: impli-cations for petroleum systems. In: Spencer, A.M., Embry, A.F., Gautier, D.L.,Stoupakova, A., Sørensen, K. (Eds.), Arctic Petroleum Geology, vol. 1. GeologicalSociety, London, pp. 163e195. Memoirs.

Hochuli, P.A., Vigran, J.O., 2010. Climate variations in the boreal Triassic d inferredfrom palynological records from the Barents sea. Palaeogeogr. Palaeoclimatol.Palaeoecol. 290, 20e42.

Høy, T., Lundschien, B.A., 2011. Triassic deltaic sequences in the northern BarentsSea. In: Spencer, A.M., Embry, A.F., Gautier, A.V., Stoupakova, A., Sørensen, K.(Eds.), Arctic Petroleum Geology, vol. 35. Geological Society, London,pp. 249e260. Memoirs.

Ivanov, K.S., Puchkov, V.N., Fyodorov, Y.N., Erokhin, Y.V., Pogromskaya, O.E., 2013.Tectonics of the Urals and adjacent part of the West-Siberian platform base-ment: main features of geology and development. J. Asian Earth Sci. 72, 12e24.

Jacobsen, V., Veen, P., 1984. The Triassic offshore Norway north of 62�N. In: Petro-leum Geology of the North European Margin. Springer Netherlands, Spencer,A.M, pp. 317e327.

Johnsson, M.J., 1993. The system controlling the composition of clastic sediments.Geol. Soc. Am. Special Pap. 284, 1e20.

Kirkland, C.L., Daly, J.S., Whitehouse, M.J., 2007. Provenance and terrane evolution ofthe Kalak nappe complex, Norwegian Caledonides: implications for neo-proterozoic paleogeography and tectonics. J. Geol. 115, 21e41.

Kirkland, C.L., Daly, J.S., Whitehouse, M.J., 2008. Basement-cover relationships of theKalak nappe complex, Arctic Norwegian Caledonides and constraints on neo-proterozoic terrane assembly in the North Atlantic region. Precambrian Res.160, 245e276.

Klausen, T., Mørk, A., 2014. The upper Triassic paralic deposits of the De GeerdalenFormation on Hopen: outcrop analog to the subsurface Snadd formation in theBarents sea. AAPG Bull. 98, 1911e1941.

Klausen, T.G., Ryseth, A.E., Helland-Hansen, W., Gawthorpe, R., Laursen, I., 2014.Spatial and temporal changes in geometries of fluvial channel bodies from theTriassic Snadd formation of offshore Norway. J. Sediment. Res. 84, 567e585.

Klausen, T.G., Ryseth, A.E., Helland-Hansen, W., Gawthorpe, R., Laursen, I., 2015.Regional development and sequence stratigraphy of the middle to late TriassicSnadd formation, Norwegian Barents sea. Mar. pet. Geol. 62, 102e122.

Larson, Tullborg, Cederbom, Stiberg, 1999. Sveconorwegian and Caledonian fore-land basins in the Baltic Shield revealed by fission-track thermochronology.Terra nova. 11, 210e215.

Lord, G.S., Solvi, K.H., Ask, M., Mørk, A., Hounslow, M.W., Paterson, N.W., 2014a. TheHopen member: a new member of the Triassic De Geerdalen Formation, Sval-bard. Nor. Pet. Dir. Bull. 11, 81e96.

Lord, G.S., Solvi, K.H., Klausen, T.G., Mørk, A., 2014b. Triassic channel bodies onHopen, Svalbard: their facies, stratigraphic significance and spatial distribution.Nor. Pet. Dir. Bull. 11, 41e59.

Lorenz, H., M€annik, P., Gee, D., Proskurnin, V., 2008. Geology of the SevernayaZemlya Archipelago and the North Kara terrane in the Russian high Arctic. Int. J.Earth Sci. 97, 519e547.

Lundschien, B.A., Høy, T., Mørk, A., 2014. Triassic hydrocarbon potential in theNorthern Barents Sea; integrating Svalbard and stratigraphic core data. Nor. Pet.Dir. Bull. 11, 3e20.

Mange, M.A., Morton, A.C., 2007. Chapter 13 Geochemistry of Heavy Minerals. In:Maria, A.M., David, T.W. (Eds.), Developments in Sedimentology. Elsevier,pp. 345e391.

Mange, M.A., Wright, D.T., 2007. Heavy Minerals in Use. Elsevier Science.Meinhold, G., Anders, B., Kostopoulos, D., Reischmann, T., 2008. Rutile chemistry

and thermometry as provenance indicator: an example from Chios Island,Greece. Sediment. Geol. 203, 98e111.

Mørk, A., Dallmann, W.K., Dypvik, H., Johannessen, E.P., Larsen, G.B., Nagy, J.,Nøttvedt, A., Olaussen, S., Pchelina, T.M., Worsley, D., 1999. Mesozoic Lithos-tratigraphy. In: Dallmann, W.K. (Ed.), Lithostratigraphic Lexicon of Svalbard,Upper Palaeozoic to Quaternary Bedrock Review and Recommendations forNomenclature Use. Norsk Polarinstitutt, Tromsø, pp. 127e214.

Mørk, A., Elvebakk, G., 1999. Lithological description of subcropping lower andmiddle Triasic rocks from the Svalis Dome, Barents sea. Polar Res. 18.

Mørk, A., Lord, G.S., Solvi, K.H., Dallmann, W.K., 2013. G14G Hopen, Svalbard 1:100,000. Temakart Nr. 50.

Mørk, M.B.E., 1999. Compositional variations and provenance of Triassic sandstonesfrom the Barents shelf. J. Sediment. Res. 69, 690e710.

Morton, A.C., Hallsworth, C.R., 1999. Processes controlling the composition of heavymineral assemblages in sandstones. Sediment. Geol. 124, 3e29.

Morton, A.C., Hallsworth, C., 1994. Identifying provenance-specific features ofdetrital heavy mineral assemblages in sandstones. Sediment. Geol. 90,241e256.

Müller, R.D., Seton, M., Zahirovic, S., Williams, S.E., Matthews, K.J., Wright, N.M.,Shephard, G.E., Maloney, K., Barnett-Moore, N., Hosseinpour, M., 2016. Oceanbasin evolution and global-scale plate reorganization events since pangeabreakup. Annu. Rev. Earth Planet. Sci. 44.



http://refhub.elsevier.com/S0264-8172(16)30343-9/sref1













































































































































































































Nøttvedt, A., Cecchi, M., Gjelberg, J., Kristensen, S., Lønøy, A., Rasmussen, A.,Rasmussen, E., Skott, P., Van Veen, P., 1993. Svalbard-Barents Sea correlation: ashort review. In: Vorren, T., Bergsager, E., Dahl-Stamnes, Ø.A., Holter, E.,Johansen, B., Lie, E., Lund, T.B. (Eds.), Arctic Geology and Petroleum Potential:Norwegian Petroleum Society (NPF), Special Publication, vol. 2. Elsevier,Amsterdam, pp. 363e375.

NPD, 2014. Geological Plays. World Wide Web Address. http://www.npd.no/en/Topics/Geology/Geological-plays/.

Omma, J.E., Pease, V., Scott, R.A., 2011. Chapter 37 UePb SIMS zircon geochronologyof Triassic and Jurassic sandstones on northwestern Axel Heiberg Island,northern Sverdrup Basin, Arctic Canada. Geol. Soc. Lond. Memoirs 35, 559e566.

P�ozer Bue, E., Andresen, A., 2014. Constraining depositional models in the BarentsSea region using detrital zircon UePb data from Mesozoic sediments in Sval-bard. Geol. Soc. Lond. Spec. Publ. 386, 261e279.

Riis, F., Lundschien, B.A., Høy, T., Mørk, A., Mørk, M.B.E., 2008. Evolution of theTriassic shelf in the northern Barents sea region. Polar Res. 27, 318e338.

Rød, R.S., Hynne, I.B., Mørk, A., 2014. Depositional environment of the upper TriassicDe Geerdalen Formation - an EW transect from Edgeøya to central Spitsbergen,Svalbard. Nor. Pet. Dir. Bull. 11, 21e40.

Rodrigues Duran, E., di Primio, R., Anka, Z., Stoddart, D., Horsfield, B., 2013. 3D-basinmodelling of the Hammerfest Basin (southwestern Barents Sea): a quantitativeassessment of petroleum generation, migration and leakage. Mar. pet. Geol. 45,281e303.

Rønnevik, H., Beskow, B., Jacobsen, H.P., 1982. Structural and stratigraphic evolutionof the Barents sea. Proc. Third Int. symp. Arct. Geol. 8.

Ryseth, A.L.F., 2014. Sedimentation at the JurassiceTriassic boundary, south-westBarents sea. From Depositional Systems to Sedimentary Successions on theNorwegian Continental Margin. John Wiley & Sons, Ltd, pp. 187e214.

Scott, R.A., Howard, J.P., Guo, L., Schekoldin, R., Pease, V., 2010. Offset and curvatureof the Novaya Zemlya fold-and-thrust belt, Arctic Russia. Geol. Soc. Lond. Pet.Geol. Conf. Ser. 7, 645e657.

Scott, R.A., Turton, M.A.K., Stewart, J.C., 1996. Triassic tectonostratigraphy of theNorth Atlantic and Arctic. CASP Rep. Reg. Arct. Ser. Triassic Subser. 655, 78(unpublished).

Smelror, M., Undersøkelse, N.G., Petrov, O.V., Basov, V.A., Larssen, G.B., 2009. Atlas:geological history of the Barents sea. Geological Survey of Norway, 134 pages.

Smirnov, V.N., Ivanov, K.S., Larionov, A.N., 2014. The U-Pb SIMS zircon age andgeodynamic conditions of formation of granitoides of the Verkhisetsk batholith,the eastern slope of the Middle Urals. Stratigr. Geol. Correl. 22, 576e593.

Soloviev, A.V., Zaionchek, A.V., Suprunenko, O.I., Brekke, H., Faleide, J.I.,Rozhkova, D.V., Khisamutdinova, A.I., Stolbov, N.M., Hourigan, J.K., 2015. Evo-lution of the provenances of Triassic rocks in Franz Josef land: U/Pb LA-ICP-MSdating of the detrital zircon from well Severnaya. li Miner. Resour. 50, 102e116.

Sømme, T.O., Helland-Hansen, W., Martinsen, O.J., Thurmond, J.B., 2009. Relation-ships between morphological and sedimentological parameters in source-to-sink systems: a basis for predicting semi-quantitative characteristics in sub-surface systems. Basin Res. 21, 361e387.

Sturt, B., Austrheim, H., 1985. Age of gneissic rocks in the Caledonian nappes of theAlta district, northern Norway. Norges Geol. Unders. Bull. 403, 179e181.

van Veen, P.M., Skjold, L.J., Kristensen, S.E., Rasmussen, A., Gjelberg, J., Stølan, T.,1993. Triassic Sequence Stratigraphy in the Barents Sea. In: Vorren, T.,Bergsager, E., Dahl-Stamnes, Ø.A., Holter, E., Johansen, B., Lie, E., Lund, T.B. (Eds.),Norwegian Petroleum Society Special Publications. Elsevier, pp. 515e538.

Vernikovsky, V.A., Pease, V.L., Vernikovskaya, A.E., Romanov, A.P., Gee, D.G.,Travin, A.V., 2003. First report of early Triassic A-type granite and syenite in-trusions from Taimyr: product of the northern Eurasian superplume? Lithos 66,23e36.

Veselovskiy, R.V., Thomson, S.N., Arzamastsev, A.A., Zakharov, V.S., 2015. Apatitefission track thermochronology of Khibina Massif (Kola Peninsula, Russia):implications for post-devonian tectonics of the NE Fennoscandia. Tectonophy-sics 665, 157e163.

Vigran, J.O., Mangerud, G., Mørk, A., Worsley, D., Hochuli, P.A., 2014. Palynology andgeology of the Triassic succession of Svalbard and the Barents sea. Geol. Surv.Nor. Special Publ. 14.

Watson, E.B., Wark, D.A., Thomas, J.B., 2006. Crystallization thermometers for zirconand rutile. Contributions Mineralo. Petrol. 151, 413e433.

Williams, I., 1998. U-Th-Pb Geochronology by Ion Microprobe. In: McKibben, M.A.,Shanks III, W.C., Ridley, W.I. (Eds.), Applications of Microanalytical Techniquesto Understanding Mineralizing Processes, pp. 1e35. Reviews in EconomicGeology.

Whitehouse, M.J., Kamber, B.S., 2005. Assigning Dates to Thin Gneissic Veins inHigh-Grade Metamorphic Terranes: A Cautionary Tale from Akilia, SouthwestGreenland. J. Petrol. 46, 291e318.

Worsley, D., 2008. The post-Caledonian development of Svalbard and the westernBarents Sea. Polar Res. 27, 298e317.

Zhang, W., Roberts, D., Pease, V., 2015. Provenance characteristics and regionalimplications of Neoproterozoic, Timanian-margin successions and a basalCaledonian nappe in northern Norway. Precambrian Res. 268, 153e167.












http://www.npd.no/en/Topics/Geology/Geological-plays/

http://www.npd.no/en/Topics/Geology/Geological-plays/











































































































marine and petroleum geology...marine and petroleum geology 78 (2016) 516e535 strata further north,...

Documents