at SciVerse ScienceDirect

Marine and Petroleum Geology 43 (2013) 208e221

Contents lists available

Marine and Petroleum Geology

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

Distribution of subsurface fluid-flow systems in the SW Barents Sea

Sunil Vadakkepuliyambatta a,*, Stefan Bünz a, Jürgen Mienert a, Shyam Chand b

aDepartment of Geology, University of Tromsø, Dramsveien-201, 9037 Tromsø, NorwaybGeological Survey of Norway (NGU), P.O. Box 6315 Sluppen, 7491 Trondheim, Norway

a r t i c l e i n f o

Article history:Received 28 July 2012Received in revised form28 January 2013Accepted 1 February 2013Available online 26 February 2013

Keywords:Fluid flowGas chimneysGas hydratesBarents SeaSeismic

* Corresponding author. Tel.: þ47 77623290.E-mail address: sunil.vadakkepuliyambatta@uit.no

0264-8172/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.marpetgeo.2013.02.007

a b s t r a c t

The SW Barents Sea is a large hydrocarbon-prone epi-continental Sea of the Norwegian Arctic region. Asignificant portion of the hydrocarbon gases generated in deep source rocks has leaked or migrated intothe shallow subsurface and is now trapped in gas hydrate and shallow gas reservoirs. The evolution ofsedimentary basins of this region has controlled the leakage of these fluids through marine sediments.Understanding the distribution of various fluid-flow systems may enhance our knowledge of the evo-lution of different basins in the SW Barents Sea and could help find potential targets for future hydro-carbon exploration. We analyze approximately 3000 2D multi-channel seismic profiles and data from 60wells covering the entire SW Barents Sea, to identify and classify fluid-flow features, and study theirrelationship to tectonic elements and geological history. Gas chimneys are the most abundant featureamong various other fluid-flow features such as fluid leakage along faults and fractures, seepage pipes,and high amplitude anomalies potentially indicating trapped fluids. Large fluid-flow features, coveringareas as large as 600 km2, occur close to known hydrocarbon fields such as Snøhvit, Skrugard, and Havis.The fluid-flow features occur above major deep-seated faults in the area suggesting a close relation to it.The number of fluid-flow features in the western part of the study area is significantly higher than in theeastern part. The amount of net erosion in the study area shows no direct control over the distribution offluid-flow features, suggesting that the faults and distribution of mature source rocks control the fluidflow. The strong correlation between the locations of fluid-flow features and structural elements in-dicates that extensional tectonics, uplift and glaciations could have played major roles in the timing andactivity of the fluid leakage, although erosion might have had an added effect.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The vertical flow of fluids through marine sediments is awidespread and dynamic geological process that occurs on passiveand active continental margins worldwide. Fluid migration isassociated with excess pore-fluid pressure, and is attributed totemporally and spatially varying processes, such as rapid sedimentloading (e.g. Dugan and Flemings, 2000), uplift and erosion (e.g.Doré and Jensen, 1996), dissociation of gas hydrate (e.g. Mienertet al., 2005), polygonal faulting (e.g. Cartwright et al., 2007), andhydrocarbon generation and leakage from deep and shallow sourcerocks and reservoirs (e.g. Heggland, 1998; Solheim and Elverhøi,1985; Hovland and Judd, 1988). Presence of shallow gas accumu-lations associated with fluid leakage are of interest for severalreasons: (1) Shallow gas accumulations may reduce the shear

(S. Vadakkepuliyambatta).

All rights reserved.

strength of sediments, and pose a hazard to hydrocarbon explora-tion and development (Andreassen et al., 2007a), (2) the occur-rence of shallow gas and indications of fluid flow underlying it maypoint toward deeper prospective reservoirs (e.g. Heggland, 1998)and (3) shallow gas accumulations could be of commercial interestin the future (Carstens, 2005).

Vertical migration of gas through subsurface strata can causewidely distributed acoustic low-velocity zones. These low-velocityzones can deteriorate the seismic signal and create regions ofchaotic seismic signals. The nature and shape of this chaotic regionof acoustic signals can vary depending on the process of formationof these zones. Chaotic regions in seismic data can also result frommud mobilization triggered by vertical migration of fluids (Løsethet al., 2003). Mud mobilization can modify the structure of sedi-ments to a disrupted succession and form a low-density sediment-fluid mixture. Vertical fluid-flow features and chaotic seismicreflection zones are commonly observed with most types of sedi-ment mobilizations (Graue, 2000; Hurst et al., 2003; Løseth et al.,2003; Jackson and Stoddart, 2005). Fluid flow can also alter the

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221 209

formation, cause local sediment remobilization, and appear aschaotic reflections in the seismic profile (Ligtenberg, 2007). Ge-ometry of subsurface fluid-flow systems is hard to constrain bydirect observations (Talukder, 2012) and characterization of them isdifficult because flow can be highly transient and can vary in timeand space along complex and changing conduit systems (Hornbachet al., 2007).

The SW Barents Sea is a part of the Arctic Ocean located north ofNorway (Fig. 1). Occurrence of shallow gas, gas hydrates and sea-floor expulsion features has been reported from several areas of theSW Barents Sea (e.g. Andreassen et al., 1990; Perez-Garcia et al.,2009; Chand et al., 2009, 2012; Ostanin et al., 2012). Migration offluids into shallow sediments and seepage into the ocean throughthe seafloor was probably the result of spillage and migration ofhydrocarbons in response to uplift and erosion processes in theCenozoic (e.g. Doré, 1995; Doré and Jensen, 1996; Henriksen et al.,2011).

Uplift and erosion is known to have affected the SW BarentsSea during Cenozoic times (Faleide et al., 1996; Dimakis et al., 1998;Anell et al., 2009). This process is thought to have a very strongimpact on petroleum systems (e.g. Doré and Jensen, 1996; Bjørkumet al., 2001; Cavanagh et al., 2006; Ohm et al., 2008; Henriksen et al.,2011). The negative effects of uplift and associated erosion on hy-drocarbon systems include tilting and opening of hydrocarbon-filledtraps resulting in spillage of oil and gas (Kjemperud and Fjeldskaar,1992), gas expansion and ex-solution from oil (Skagen, 1993; Doréand Jensen, 1996; Cavanagh et al., 2006), seal failure (e.g., Sales,1993), reduction in temperature due to uplift resulting in imma-ture source rocks (Doré et al., 2000; Ohm et al., 2008), and porosityand permeability reduction due to diagenetic processes (Berglundet al., 1986; Walderhaug, 1992).

Figure 1. General bathymetric map of SW Barents Sea with major basins and hydrocarbonfigures are also shown.

Whereas, positive effects include the occurrence of thermallymatured source rocks at shallow levels (Bjørkum et al., 2001),liberation of dissolved methane from formation water (Doré andJensen, 1996) due to decreases in pressure (e.g. Maximov et al.,1984; Nesterov et al., 1990), ex-solution of light oil or condensatefrom gas (e.g. Piggott and Lines, 1991), fracture enhancement of lesspermeable reservoirs (e.g. Aguilera, 1995) and remigration toshallower structures (Waylett and Embry, 1992). In addition, localre-deposition under a heavy overburden associated with erosioncan result in rapid maturation of source rocks and generation ofhydrocarbons (e.g. Dahl and Augustson, 1993).

Numerous glaciations also affected the SW Barents Sea regionduring the late Cenozoic. Rapid buildup and removal of ice load, asoccurred in the SW Barents Sea, may have less impact on theevolution of the basin (Lerche et al., 1997). However, distortions tothe thermal regime of sub-ice sediments caused by spatial andtemporal variations of ice thickness influence the generation,migration and present-day accumulation of hydrocarbons (Lercheet al., 1997; Cavanagh et al., 2006). Ice loading can cause struc-tural distortions leading to a redistribution of gas and oil in thereservoir and spill of hydrocarbons (Lerche et al., 1997). Major ice-loading effects are present in the Hammerfest Basin; where manydrilled wells found only a small amount of residual oil in rotatedand tilted structures (Rasmussen et al., 1993; Nyland et al., 1992).

Often, large gas anomalies overlie hydrocarbon discoveries inSW Barents Sea indicating leakage of gas from the deeper forma-tions (Chand et al., 2008, 2009, 2012; Heggland, 1998, 2004;Meldahl et al., 2001). A recent study from the Loppa High (Chandet al., 2012) reported seepage of gas into the water column indi-cating that the gas migration is still active through open faults. Thelocations of these anomalies may indicate possible targets for

discoveries. The location of the study area (red star) and seismic lines in the following

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221210

future hydrocarbon exploration. Therefore, it is important to un-derstand the distribution of fluid-flow anomalies and their gov-erning controls. Here, we analyze 2D seismic data covering thewhole SW Barents Sea (Fig. 2) in order to delineate and classifyvarious fluid-flow features, to characterize their distribution, and tobetter understand their relation to the structural setting, glacia-tions, uplift, and erosion during the Cenozoic.

2. Geological setting

2.1. Geological history of the area

The Barents Sea is a large epi-continental sea, bound to the westand north by young passive margins that developed during theCenozoic opening of the NorwegianeGreenland Sea and Eurasiabasin, respectively (Faleide et al., 1993). The SW Barents Sea con-tains some of the deepest sedimentary basins worldwide withsediment cover exceeding 10 km at places (e.g. Nordkapp Basin,Sørvestsnaget Basin) (Smelror et al., 2009). These basins formed inresponse to several phases of regional tectonism within the NorthAtlanticeArctic region, culminating with continental separation ofEurasia and Greenland, and accretion of oceanic crust, in the earlyTertiary (Faleide et al., 1993).

The western part of the Barents Sea shows much more complextectonic development than the eastern part, with a mosaic of ba-sins, platforms, and structural highs (Faleide et al., 1984; Gabrielsenet al., 1990; Gudlaugsson et al., 1998). The SW Barents shelf formeda central part of the northern Pangean margin from the lateDevonian (Worsley, 2008) and is underlain by early Devonianmetamorphic basement formed during the Caledonian Orogeny(Smelror et al., 2009). Extensional tectonic movements during

Figure 2. Coverage of 2D seismic data (thin black lines) used for the stu

earlyemiddle Devonian, Carboniferous, Permian, Triassic and lateJurassiceearly Cretaceous (Johansen et al., 1993) dominate the latePaleozoic and Mesozoic tectonic history of the SW Barents Sea.Extensional faulting also affected the region during late Cretaceousand Paleogene (Faleide et al., 1993). In late Devonian to earlyCarboniferous times, most of the Barents Seawas affected by crustalextension (Gjelberg, 1981, 1987; Faleide et al., 1993). These riftingevents caused the formation of Hammerfest Basin as a NE-trendinghalf graben (Cavanagh et al., 2006). Evaporites were deposited inthe Nordkapp Basin and possibly also in the Tromsø Basin (Worsley,2008). Extensional faulting affected the Loppa and the StappenHighs during late Carboniferous to early Permian period (Brekkeand Riis, 1987), while northeastern part of Bjarmeland Platformand Nordkapp Basin were stable (Riis et al., 1986).

The western part of the Barents Sea has been the most tecton-ically active sector throughout Mesozoic and Cenozoic times;although Triassic to early Jurassic was relatively quiet tectonically(Gabrielsen et al., 1990). However, reactivated rifting in thePermian-early Triassic led to tilting in the Loppa High and Stappenhigh region (Fig. 1) (Gudlaugsson et al., 1998). The early Triassic wascharacterized by regional subsidence in eastern areas and sedimentinflux into the regional sag basin from the East (Gudlaugsson et al.,1998; Worsley, 2008). Salt tectonics influenced the depositionalpatterns in the Nordkapp andMaud Basins (Fig.1) (Gabrielsen et al.,1990).

Rifting and associated block (extensional) faulting started againin the Mid-Jurassic, related to the opening of Central Atlantic(Ziegler, 1988), increased during the period from late Jurassic intoearly Cretaceous, with the opening of southern North Atlantic(Faleide et al., 1993), and terminating with the formation of thenow well-known major basins and highs (Fig. 1). Structural

dy with major basins and boundaries (thick black lines) in the area.

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221 211

development in this period was complicated. On one hand,extreme rates of subsidence are seen in the Tromsø Basin and thewestern part of the Bjørnøya Basin during the early Cretaceous(Aptian to Albian). On the other hand, indications of local earlyCretaceous inversion are found, e.g. along the Ringvassøy-LoppaFault Complex (RLFC) and its junction with the Asterias FaultComplex (Fig. 1) (Gabrielsen et al., 1990). During the early Creta-ceous, Harstad, Tromsø, and Bjørnøya basins underwent large-scale subsidence and become major depocenters (Faleideet al., 1993). Uplift and erosion in the northeastern Barents Seabrought more sediments into these depocenters (Smelror et al.,2009). Throughout the Cretaceous, rifting prevailed and causedformation of pull-apart basins such as Sørvestsnaget Basin andVestbakken Volcanic Province (Fig. 1) (Smelror et al., 2009). In thewestern part of the area, especially in the Vestbakken VolcanicProvince, there was abundant magmatic activity, probably in thePaleocene and Eocene (Jebsen and Faleide, 1998; Faleide et al.,2008).

The opening of the North Atlantic was associated with upliftduring the Paleocene-Eocene (Eldholm et al., 1987; Faleide et al.,1996). Inversion and folding reached a maximum in Eocene toOligocene times (Talwani and Eldholm, 1977; Myhre et al., 1982;Eldholm et al., 1987).

The Barents Sea was affected by extensive erosion, related todeglaciation and uplift in the late PlioceneePleistocene (Vorrenet al., 1991; Nyland et al., 1992). During late PlioceneePleistocene,the entire Barents Shelf was uplifted and eroded, and largeamounts of sediment were deposited along the western margin(Nyland et al., 1992). Particularly large accumulations are found inthe Bjørnøya trough mouth fan (Fig. 1), with up to 6 km-thickpackages of glacigenic sediments, a main depocenter during thelate Cenozoic (Hjelstuen et al., 2004; Vorren et al., 1991). In thesouthern part of the SW Barents Sea, the Hammerfest Basin, theNordkapp Basin and the Loppa High suffered lesser amounts ofuplift and erosion (Smelror et al., 2009). The Upper Regional Un-conformity (URU), formed during the late PlioceneePleistocene atabout 2.5 Ma, separates the glacigenic sediments of the Barents Seafrom the deeper pre-glacial sedimentary rocks (Tertiary and older)and is a major seismic reflector in the study area (Solheim andKristoffersen, 1984; Eidvin et al., 1993). Criss-crossing glacial line-ations and iceberg plough marks are a major feature of the BarentsSea seabed and paleo seabeds affected by glacial erosion (Elverhøiand Solheim, 1983; Andreassen et al., 2007b).

2.2. Major source and reservoir rocks

Numerous source-rock formations are present in the SWBarents Sea (Fig. 3) (Doré,1995). The amount of total organic carbon(TOC), hydrocarbon generative potential, and hydrogen index ofvarious source rocks in the Norwegian Barents Sea are discussed indetail by Ohm et al. (2008). The most widely distributed and mostprolific source rock is the Hekkingen Formation of late Jurassic age,which consists of dark, organic rich shales (Dalland et al., 1988).Despite being widespread over most of the southern Barents Sea,these shales have not realized their full hydrocarbon-generationpotential because of varying depth of burial and thereby maturity.The unit is thought to be mature for oil and gas generation in anarrow belt at the western margin of the Hammerfest Basin andalong the western fringe of the Loppa High (Doré, 1995). Hekkingenshale also has the highest TOC, highest hydrocarbon generativepotential, and hydrogen index among the source rocks in theNorwegian Barents Sea (Ohm et al., 2008).

Triassic shales such as, Snadd Formation and Havert Formation,are also shown to have potential for generating hydrocarbons(Bjorøy et al., 2009). The cumulative generation potential from the

thick Triassic (and possibly Lower Jurassic) sedimentary pile islarge, and it is widely assumed that the major gas discoveries of theSouth Barents Basin emanate from this source (Johansen et al.,1993). Permian and Carboniferous shales are the major source ofhydrocarbons in the Finnmark Platform (Ohm et al., 2008).

Figure 3 provides a summary by geological age of all majorsource rocks and reservoirs, proven and postulated, in the Nor-wegian Barents Sea. The most significant proportion of the hydro-carbon resources proven to date in both the Norwegian and RussianBarents Sea are trapped within strata of Jurassic age (Doré, 1995).The major discoveries in the SW Barents Sea have principal reser-voirs consisting of loweremiddle Jurassic sandstone called the StøFormation (Dalland et al., 1988). Larsen et al. (1993) estimate thatabout 85% of the Norwegian Barents Sea resources lay within thisformation. Minor Triassic gas accumulations have been found in theHammerfest Basin and on the margins of the Nordkapp Basin.

3. Data used

The fluid-flow features were mapped using approximately 30002D multi-channel seismic profiles covering the SW Barents Sea(Fig. 2). In addition to the seismic data, well logs with formationtops from 60 wells in the SW Barents Sea aided the interpretation.The 2D lines were from different surveys with different horizontaland vertical resolutions.

4. Observation and interpretation of fluid-flow anomalies

Multi-channel seismic data from different basins of the SWBarents Sea are used for mapping various types of fluid-leakageanomalies. Mapping of the source of fluid-flow anomalies wasoften difficult because of varying survey specifications in resolutionand seismic-signal quality. The fluid-flow features interpreted onseismic data were subdivided into three main categories; 1) fluid/gas chimneys; 2) leakage along faults, and 3) all other features, notincluded in the first two categories.

4.1. Fluid/gas chimneys

Many of the seismic profiles from the study area show zones ofchaotic acoustic signals and/or acoustic masking (Fig. 4). The sizeand shape of these acoustically anomalous zones varies widely.In most parts, these anomalous zones terminated at differentshallow stratigraphic levels showing high-amplitude anomalies,suggesting the presence of gas. Such features are interpreted asgas chimneys, which can be defined as a region of distortedseismic signals resulting from irregularly distributed low-velocitygas-charged zones, formed due to an upward migration of gas/fluids (Meldahl et al., 2001; Løseth et al., 2002; Heggland, 2004;Judd and Hovland, 2007; Arntsen et al., 2007; Connolly et al.,2008). Figure 4(a) shows gas chimneys located in the BjørnøyaBasin west of the Loppa High. They show well-defined zones ofchaotic seismic signals and terminate at the Top Cretaceous level.Patchy high-amplitude reflections were observed above bothchimneys. The small depressions on the seafloor were inconclu-sive for pockmarks, since the area was glaciated in the past andthus the depressions may be glacial plough marks. The gaschimneys are located in the Bjørnøyrenna Fault Complex (BFC)area (Fig. 1), and the small gas chimney is located right above afault (Fig. 4a). We hypothesize that the gas migrates upwardthrough this fault. Seismic data from the Polheim Sub-platformwest of the Loppa High shows an exceptionally large gas chimney(Fig. 4b). A very large zone, approximately 12 km wide, of acousticmasking characterizes it. Shallow high-amplitude reflections markthe upper termination of the chimney. High-amplitude anomalies

Figure 3. Some major proven and potential reservoir and source rocks in the Barents Sea (modified after Doré, 1995; Ohm et al., 2008).

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221212

are patchy, polarity-reversed and crosscut some parts of thelithology. The seismic signal is highly chaotic inside the chimney,making it difficult to identify the source of gas leakage. However,the area is part of the BFC and the fluid migration could be alongthese faults.

Vertical and lateral migration of fluids is visible in SørvestsnagetBasin close to Veslemøy High (Fig. 4c). High-amplitude anomalieslocated at the Plio-Pleistocene boundary (URU) mark the uppertermination of a columnar zone of fluid migration. The high-amplitude reverse polarity reflections at this stratigraphic level

suggest accumulation of gas. This feature is interpreted as a gaschimney. High-amplitude reflections are also visible on otherstratigraphic boundaries adjacent to the gas chimney and awayfrom the termination point of the chimney, suggesting lateralmigration in the up-dip direction along the strata.

The seismic line located on the southeast part of Loppa High,close to the junction of Asterias FC and RLFC (Fig. 1), shows anotherexample of a gas chimney (Fig. 4d). The region of fluid migration ismarked by a zone, approximately 1-km wide, of seismic reflectorsthat are chaotic and deteriorated in amplitude. Faults may be

Figure 4. a) Show fluid flow in the southeast part of Bjørnøya Basin. The leakage zone of large fluid flow feature isw8 km at its widest region. Patchy high-amplitude anomalies arepresent on top of the leakage zone. The leakage zone covers an area of w36 km2. The small chimney sits on top of a fault and shows bright spots on top of the termination. b) A12 km wide fluid migration feature SW of the Loppa High. It covers an area of w130 km2. The shallow high amplitude reflections show reverse polarity with respect to the seafloorindicating presence of gas. They also cross cuts lithological reflectors suggesting formation of gas hydrates. c) Seismic profile from Sørvestsnaget Basin close to Veslemøy High showschaotic acoustic signals, which terminates in Late Tertiary sediments leaving bright spots. This feature is interpreted as a gas chimney, and the high-amplitude anomalies indicatethe presence of gas. d) Shows a gas chimney from the SW part of Loppa High. Highly chaotic seismic reflectors define the zone of fluid flow, and a very bright reflector is visible onthe termination point of the chimney, suggesting accumulation of gas.

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221 213

present inside the chimney, but are difficult to identify in theseismic profile due to loss of seismic energy inside the chimneyzone. A very strong, polarity reversed reflector at the terminatingpoint of the chimney just beneath the URU reflector suggests

accumulation of gas. The Snadd Formation, one of the major sourcerocks in the area (Fig. 3, Section 2.2) and other deeper source rockscould be the source of fluids here; however, it is very difficult toidentify it accurately from the seismic data.

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221214

4.2. Leakage along faults

2D seismic data show chaotic reflections and high-amplitudeanomalies associated with faults in many parts of the SW BarentsSea (Fig. 5). A seismic profile from the northern part of the Loppa

Figure 5. a) Seismic profile from northern part of the Loppa High shows leakage along faulAccompanying minor faults also shows signs of fluid flow with high amplitude anomalies onup covering an area ofw290 km2. b) Shows fluid leakage along faults from the southern partthe underlying Paleozoic strata. c) Fluid leakage along faults from the northwest part of thformation, show high-amplitude anomalies along them and at their termination in the shallosource of the upward migrating gas. d) Seismic profile across the Samson Dome, NE of the Hzone is on the southern part of the dome, covering an area of w141 km2.

High shows two major faults, cutting through the Hekkingen For-mation and extending up to the URU, associated with several minorfaults (Fig. 5a). Highly chaotic and low-amplitude reflections occurclose to the root of themajor faults, suggesting fluidmigration froma much deeper source and branching of fluid migration along the

ts in the Hekkingen formation. The two major faults act as main pathways for leakage.their flanks. The fluid flow feature is narrow at the base (w2 km) and spreads as it goesof the Loppa High. The source of the leakage could be early Triassic Havert Formation ore Hammerfest Basin. Two faults, one of them cutting through the Jurassic Hekkingenw strata, indicating gas migration and accumulation. Hekkingen formation could be theammerfest Basin, show w7 km wide region of fluid migration along faults. The leakage

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221 215

faults. High-amplitude anomalies are visible along the fault planesclose to the termination of the faults. Occurrences of high-amplitude anomalies in seismic data along the faults indicate thatgas is present (Løseth et al., 2009; Cartwright et al., 2007;Ligtenberg, 2005). Minor faults also show high-amplitude re-flections along fault planes. No high-amplitude reflections areobserved above the URU, suggesting trapping of upward migratinggas at or beneath URU. Hekkingen formation could be the source ofshallow gas in this area, although deeper formations could alsocontribute to the gas leakage (See Section 2.2).

Figure 5(b) shows a heavily faulted region from the southernpart of the Loppa High. Noisy seismic reflections are visible close toand along the faults. It also shows high-amplitude reflections onthe fault planes close to the termination point of faults, suggestingpresence of gas. A few faults extend almost all the way to theseafloor and their termination coincides with the occurrence ofsmall depressions on the seafloor. Based on the 2D seismic data wecannot resolve whether these depressions are pockmarks or not.The fluidmigration is along twomajor faults in the deeper part, andbranches out throughmany smaller faults in the shallow strata. TopHavert formation of early Triassic age could be the major source ofhydrocarbons. Deeper formations could also be contributing to theleakage (See Section 2.2). Seismic profile from northwestern part ofthe Hammerfest Basin shows a perfect example of fluid leakagealong a fault (Fig. 5c). High-amplitude reflections occur along thestratigraphic horizons over close distances away from the faults.The fluid leakage is mainly along the bigger fault until Top Creta-ceous level and then branches along another fault. High-amplitudereflections are visible close to the terminating point of the faults,suggesting shallow gas accumulation. The main fault cuts acrossthe Top of Hekkingen formation, which could be the source ofleaking gas (Doré, 1995). The Samson Dome area in the BjarmelandPlatform, northeast of the Hammerfest Basin (Fig. 1), shows fluidleakage and shallow gas accumulation over an anticlinal structure(Fig. 5d). Noisy zones of seismic signals and high-amplitudeanomalies close to the seafloor suggest upward migration offluids. The top of the anticlinal structure marked by the late JurassicHekkingen Formation and formations below show many smallfaults, formed probably due to extensional forces. These faults maybe acting as migration pathway for hydrocarbons. Much largerfaults are also present on the flanks of the anticlinal structure andmay also play a role in fluid migration. The Hekkingen formation isa unit with hydrocarbon potential, but is not mature in this area(Doré, 1995). Much deeper Triassic formations, such as Havert for-mation, could be the source of upward migrating gas (Fig. 3).

4.3. Other features

Many seismic expressions associated with fluid flow other thangas chimneys and leakages along faults are present in the SWBarents Sea. These features include seepage pipes, pockmark-likedepressions, acoustic pull downs related to fluid migration, andbottom-simulating reflectors (BSR). They are rare and widelydistributed in the study area. Seepage pipes are columnar zones ofdisrupted reflections with localized amplitude anomalies(Cartwright et al., 2007). Pockmarks are depressions on the seabedresulting from the seepage of gas and pore fluids in soft sediments(Judd and Hovland, 2007). Pockmarks of different sizes and den-sities are common in some parts of the Barents Sea (Solheim andElverhøi, 1985; Hovland and Judd, 1988). The central Barents Seashows seabed depressions up to 1000 m in diameter (Solheim andElverhøi, 1993; Lammers et al., 1995). BSRs are high-amplitudereverse-polarity reflections resulting from abrupt change inacoustic impedance at the boundary between a hydrate-bearinglayer and underlying gassy sediments (Sloan and Koh, 2007; Judd

and Hovland, 2007). Figure 6(a) shows a vertical, narrow zone ofacoustic disturbance from the southern part of the BjarmelandPlatform north of Norsel High. The zone of chaotic seismic signalsterminates in shallow strata. However, faults (one of which extendsto the seafloor) are visible above the termination point and thesefaults may represent pathways for fluidmigration upwards throughthe strata. High-amplitude reflections are visible above the termi-nation of the vertical zone of fluid leakage. The zone is approxi-mately 800-m wide and such regions of narrow, vertical fluidmigration, and is interpreted as a seepage pipe. The source of fluidleakage could be Triassic or older formations, since the leakagezone go deeper than the Havert formation representing earlyTriassic. A seismic profile from central part of the Finnmark Plat-form shows chaotic zones of seismic reflectors and high-amplitudeanomalies indicating migration of fluids (Fig. 6b). The reflectorsshow acoustic pull down suggesting presence of low-velocity ma-terial, possibly gas. The source of fluid flow could be early Permian/late Carboniferous Ørn formation or much deeper Carboniferousshales (Fig. 3, Ohm et al., 2008). Also visible is a crater-likedepression, 1.2 km wide, which can be interpreted as a paleochannel.

The Bjørnøyrenna Fault Complex, west of the Loppa High,shows a gas chimney, 6 km-wide, with high-amplitude reflectionsat its shallow terminating point (Fig. 6c). The chaotic region in theseismic data suggests upward migration of fluids. High-amplitudereflections observed above the chimney are polarity reversed withrespect to the seafloor, suggesting accumulation of gas beneath it.These enhanced reflections are also discontinuous and crosscutstratigraphic boundaries near it. This crosscutting event can beinterpreted as a gas-hydrate-related BSR (Holbrook et al., 1996;Kvenvolden, 1998). Higher-order hydrocarbon gases could bemigrating along with other fluids through the chimney and canform gas hydrates in this part of the Barents Sea (Chand et al.,2008). BSRs have been observed in seismic data earlier in thisarea and are suggested to have gas compositionwith few percents ofethane and propane along with methane (Laberg and Andreassen,1996).

Although no high-resolution 3D-seismic multibeam data wereused for this study, wewere able to identify possible pockmark-likedepressions in some areas. It should be kept in mind that the SWBarents Sea was heavily glaciated in the Cenozoic and glacialplough marks are very common in the study area. This makesidentification of surface fluid-expulsion features difficult with 2Ddata. However, seismic profiles from the Bjørnøya fan showedpockmark-like depressions on the seafloor, approx. 600 m wide(Fig. 6d). These depressions showed vertically stacked high-amplitude reflections underneath them. Reflectors underneaththe channel showed acoustic pull down. The high-amplitude re-flections suggest the presence of gas and fluid flow. The gasescaping to the seafloor may be attributed to the formation of thesedepressions. It is also possible that these are erosional channels astheir axis is along the slope of Bjørnøya fan.

5. Distribution of fluid-flow features

Fluid-flow features, shallow gas accumulations, and associatedgas hydrates are characterized from different parts of the SWBarents Sea (e.g. Andreassen et al., 1990; Laberg and Andreassen,1996). Fluid-flow features, especially large gas chimneys, are hy-drocarbon migration pathways and the location of these features isimportant for the hydrocarbon industry (Heggland, 1998) sincethese chimneys could pose a drilling hazard due to high pore-fluidpressures that may be present (Løseth et al., 2002). Our studyshows that the fluid flow is abundant and widespread in the SWBarents Sea (Fig. 7).

Figure 6. a) A vertical, narrow zone of fluid leakage from the southern part of the Bjarmeland Platform north of the Nyslepp Fault Complex. The leakage zone is approximately800 m wide, which is interpreted as a seepage pipe. b) Shows chaotic seismic reflectors, and high amplitude reflector from the central part of the Finnmark Platform, suggestingmigration of fluids. A depression, approximately 600 m wide, is also visible above the fluid migration feature. The downward bending reflectors indicate acoustic pull-down due tolow-velocity material, probably gas. c) Cross-cutting high-amplitude reflectors associated with the gas chimney from the Bjørnøyrenna Fault Complex. Huge region of maskedacoustic reflections and polarity reversed high-amplitude anomalies show fluid flow and gas accumulation. The patchy, crosscutting high-amplitude reflector mimics the seafloorand is interpreted as the base of gas hydrate stability zone. d) Seismic profile from the Bjørnøya fan showing pockmark-like depressions and associated high amplitude feeders. Thedepressions were few hundred meters across. Acoustic pull-down and presence of bright spots suggest the presence of gas.

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221216

Gas chimneys are the most common fluid-flow feature in thestudy area and appear in most parts with various sizes and shapes.The Ringvassøy-Loppa and Bjørnøyrenna Fault Complexes, northernpart of the Tromsø Basin, the Polheim Sub-platform and theVeslemøy High show a distinctly higher density of gas chimneys

compared to other regions in the SW Barents Sea. Fewer gaschimneys are present in the Tromsø Basin and the easternpart of thestudy area, especially the Finnmark Platform, Nordkapp Basin andthe Bjarmeland Platform. The Loppa High and the heavily faultedareas around it also show a high concentration of gas chimneys. In

Figure 7. Map showing the distribution of gas chimneys and leakage along faults in the SW Barents Sea. The black lines are the major fault boundaries. Most of the features werelocated right on top of major faults in the area.

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221 217

the Hammerfest Basin, gas chimneys are present close to theSnøhvit reservoir. Many of the gas chimneys were located rightabove major faults in the study area (Fig. 7). Sørvestsnaget Basinshowsmany gas chimneys, which are not related to anymajor faults.However, they are small and widely distributed.

Leakages along faults are present in almost all parts of the SWBarents Sea. Their density, however, was higher in thewestern part,especially in the Loppa high and areas surrounding it. RLFC and BFCshow high concentrations of leakage along faults. The major faultboundaries between different basins also show numerous faultleakages (Fig. 7). Our observations indicate that the fluid leakagesare mainly related to major faults in the area.

Although the distribution of seismic data used for the study isuneven, the western part of the study area clearly shows a higherconcentration of fluid-flow features. This can be attributed tovarious reasons. Presence of high number of faults in the westernpart increases the odds of fluid flow since fault reactivation duringuplift and glacial cycles could negatively affect the sealing ability.The upper Jurassic, Triassic and Permian/Carboniferous sourcerocks are oil mature in the western part and can exsolve gas duringuplift and erosion (Ohm et al., 2008). Triassic source rocks are over-mature or gas-mature in the western part of SW Barents Sea (Ohmet al., 2008). In addition, evaporites in the eastern part of the studyarea might have negatively affected hydrocarbon migration to theshallow strata (Doré and Jensen, 1996).

6. Large fluid-flow features

The high density of 2D seismic profiles in some areas allowed ustomap the aerial extent of some of the fluid-flow features in the SWBarents Sea. The size and shape of these features are approximate,

since they are derived from 2D data. Areas such as the Loppa High,northern part of Hammerfest Basin, northern part of Tromsø Basinand the RLFC have good coverage of 2D seismic data, which allowedthe mapping of large features in these locations.

In all, 93 large fluid-flow features were mapped (Fig. 8). Theyhave peculiar shapes and cover huge areas. Most of them terminatein the shallow strata, with high amplitude reflections at their crest,suggesting shallow gas accumulation. The area covered by theselarge fluid-flow features varied from 1 to 600 km2. Of the 93 largefeatures mapped, 81% were comparatively smaller in area, varyingfrom 1 to 50 km2. The largest mapped fluid-flow feature is a gaschimney, located on the RLFC and northern part of Tromsø Basin,covering approximately 600 km2. The total area covered by theselarge fluid-flow features is approximately 3000 km2, which isapproximately one percent of the SW Barents Sea. The shape ofthese features varies widely from circular to elongate. Elongatedfluid-flow features are seen located above major faults, while cir-cular fluid-flow features are present on all parts.

The Ringvassøy-Loppa and Bjørnøyrenna Fault Complexes, thenorthern part of the Tromsø Basin, the Polheim Sub-platform, andthe Veslemøy High contain most of the large fluid-flow features.Sørvestsnaget Basin and Finnmark Platform show few small fluid-flow features. The Veslemøy High has a denser distribution offluid-flow features. These features are also larger than in otherareas and mostly occur above major fault boundaries. The northernpart of Tromsø Basin, devoid of salt domes, also shows large fluid-flow features especially along the boundaries with Veslemøy High,Loppa High and Polheim Sub-Platform. The Hammerfest Basinshows large fluid-flow features close to the Snøhvit reservoir andon its boundary with the Loppa High. The Bjørnøya Basin showstwo of the largest fluid-flow features, covering 260 km2and

Figure 8. Distribution of large fluid flow features in the area. They vary in area fromw1 km tow600 km2. Histogram plot of area covered by these large features show 81% of themfalling in the category with an area of 1e50 km2.

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221218

200 km2, above the Bjornøyrenna Fault Complex. The Loppa Highshows a very large gas chimney on the northern part, which coversan area of 290 km2. The largest fluid-flow feature in the eastern partof the study area is located above the Samson Dome and covers150 km2. The Nordkapp Basin shows no large fluid-flow features.

7. Discussion

We observed subsurface fluid-flow systems on all parts of theSW Barents Sea. The observed features were of various types,interpreted as gas chimneys, leakage along faults and fractures andother related features. However, gas chimneys were dominant inmost parts of the study area (Fig. 7). A wide variety of fluid-flowexpressions probably indicates the variable response to glacialcycles, uplift and erosion in the different basins. Fluid-flow featuresare spatially related to the structural elements of the study area,especially with faults (Fig. 7) separating the major basins and plat-form in the SW Barents Sea. Most of the fluid-flow features werelocated above major deep-seated faults located in hydrocarbon-richsource/reservoir rocks in the study area. Fluid flow along faults isalsowidespread in the study area (Figs. 5 and 7). Faults and fracturescan act as migration pathways for fluids since they are generallymore permeable than the surrounding rock (Berndt et al., 2003;Cartwright et al., 2007). Removal of sedimentary overburden duringuplift and erosion, and the consequent decrease in pressure, willcause the gas in a gas accumulation to expand. The effect of thisexpansion, assuming that the pre-existing structure was filled tospill, will be the expulsion of hydrocarbons from the closure. Thiswill result in shorter oil legs and/or a less dense, and hence smaller,gas accumulation (c.f. Nyland et al., 1992). The same effect couldoccur during repeated loading and unloading of glaciers (Lerche

et al., 1997). Thus glacial cycles, isostatic uplift, and erosion duringthe Cenozoic can be responsible for the majority of the fluid leakageobserved throughout the SW Barents Sea.

Near the Loppa High, where large fluid-flow features areconcentrated, it is evident that the leakage is along the rotated faultblocks (Fig. 4a), which might have been reactivated during glacialcycles. These rotated fault blocks contain one of the prolific reser-voir rocks of the Barents Sea, the Stø formation. The Snadd for-mation with shale and sandstone inter-bedding could also be asource for hydrocarbons in this area. Fracturing of the cap rock canbe a major cause of fluid leakage (Arntsen et al., 2007), which couldbe the reason for fluid leakage in the Samson Dome (Fig. 4c).Sudden release of pressure from a highly pressurized fluid accu-mulation could have resulted in the formation of seepage pipe inthe Bjarmeland Platform (Fig. 6a) (Løseth et al., 2011; Cartwrightet al., 2007). Structural traps, especially faults, are thought to behighly sensitive to the effects of uplift and erosion (Henriksen et al.,2011). Extensional fault-dominated areas such as the HammerfestBasin, Loppa High and Veslemøy High showed a high density offluid-flow features. Salt-related structural traps, which areconsidered stable against considerable amounts of uplift anderosion, present in the Nordkapp and Tromsø basins showed rela-tively less fluid-flow features. However, numerous small gaschimneys occur along the southern and western rims of theNordkapp Basin (Fig. 7). Thus, the structural evolution of SWBarents Sea might have played a major role in the fluid flow historyof the area.

The large fluid-flow features represent significant areas of hy-drocarbon leakage. Many of the largest ones were located close tothe two major fault complexes in the area, RLFC and BFC. Largefluid-flow zones suggest unfocussed migration of gas from the

Figure 9. The amount of net erosion (black lines, after Henriksen et al., 2011) does not seem to affect the distribution of fluid flow (in pink).

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221 219

rotated fault blocks, which could provide multiple points of hy-drocarbon spillage. Such extensive areas of leakage suggest thepresence of significant hydrocarbon reservoirs in the subsurface.Some of the fluids from these reservoirs have clearly leaked fromtraps. There still, however, may be economic quantities of hydro-carbon beneath these fluid-leakage areas, as evidenced by discov-eries in the Snøhvit and Goliat fields in Hammerfest Basin (Doré,1995).

The source of the observed fluid-flow features was difficult toidentify from seismic data due to their deep origin and acousticmasking caused by upward migrating gas. The tentative maturity ofvarious source rocks in the Norwegian Barents Sea suggested byOhm et al. (2008) shows oil mature upper Jurassic HekkingenFormation in the Hammerfest basin and the northern part of theLoppa High. Triassic and Permian/Carboniferous formations (Snadd,Havert) are oil mature in major basins in the SW Barents Seaincluding the Hammerfest Basin. The presence of multiple sourcerocks, especially in the western part of study area, increases thechanceof hydrocarbon generation anddissolution of gas fromoil andcan also contribute to mixing of hydrocarbons from different strat-igraphic intervals, as observed in somewells (e.g. 7120/2-1, 7125/1-1,Goliat oil well 7122/7-3) in the Barents Sea (Ohm et al., 2008).

Isostatic uplift and associated erosion is considered to haveaffected the hydrocarbon accumulation and migration in the SWBarents Sea. Many workers (Vorren et al., 1991; Riis, 1992; Nylandet al., 1992; Henriksen et al., 2011) have discussed net erosion inthe Barents Sea. Their studies suggest net erosion values ofapproximately 900e2000 m in the SW Barents Sea and concludethat different amounts of uplift and erosion in the area could affectthe petroleum systems. We correlated the net erosion of theSW Barents Sea (Henriksen et al., 2011) with the distribution offluid-flow features. We could not find any relation between the

amount of net erosion and the distribution of fluid-flow features(Fig. 9). The areas with a dense distribution of fluid-flow features inthe western half of the study area show an erosion of approxi-mately 1000 m, whereas the eastern half, with a net erosion ofaround 2000 m, shows comparatively less fluid-flow features.

This suggests that the fluid leakage in the study area is primarilycontrolled by the presence of mature source rocks and structuraltraps, especially faults. Tectonic faulting and fracturing is oftensuggested as the controlling mechanism for the location and dis-tribution of fluid-flow observations worldwide (Gay et al., 2007;Barnes et al., 2010; Talukder et al., 2003). In the western part of thestudy area, most of the fluid-flow features terminate in youngersediments (late Tertiary-Quaternary) (Figs. 4, 5c, 6c, and d) whichsuggests that they might have been triggered by uplift and glacia-tions during the late PlioceneePleistocene.

Gas chimneys terminate at different stratigraphic levels in thestudy area (Figs. 4e6). This suggests that the timing of fluid flowcould be different from basin to basin. However, regional tectonicevents, such as rifting, uplift or glaciations would affect the wholeregion and could trigger fluid leakage at the same time regionally.Difference in the lithology of shallow sediments in the area couldcause different termination point for chimneys, even though thesame tectonic event triggered them. Since the structural evolutionof the SW Barents Sea is complex and different from basin to basin,a more focused study on individual basins may be needed toaccurately analyze the origin and timing of these fluid-flowfeatures.

8. Conclusions

Fluid flow is abundant and widespread in the SW Barents Seawith some very large gas chimneys covering approximately one

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221220

percent of the study area. Different types of fluid-flow featuresoccur in the area, the most dominant being gas chimneys. Thedistribution of fluid-flow features showed a direct relationshipwith the structural setting of the SW Barents Sea. This could bedue to the inability of deep-seated faults to effectively seal thetrapped hydrocarbons. Fault traps could be highly sensitive tothe effects of uplift, erosion and glaciations. The fluid-flow fea-tures were highly dense in the western part of the SW BarentsSea. The difference in distribution could be due to a) presence ofmultiple source rocks, which are over-mature or gas mature inthe western part, b) availability of structural traps, especiallyfaults and c) presence of evaporites in the eastern part. Theamount of net erosion in the study area did not seem to becontrolling the distribution of fluid leakage features. Given therelation of fluid flow to the structural setting, extensional tec-tonics, uplift and glaciations in the Plio-Pleistocene time couldhave played major roles in the timing and activity of the fluidleakage, although erosion might have had an added effect.However, due to the structural complexity of the study area andthe number of fluid-flow features observed, a more focusedstudy on the timing and origin of fluid-flow features in indi-vidual basins is necessary.

Acknowledgments

The authors are thankful to TGS-NOPEC Geophysical Companyfor providing some of the seismic data. We thank Schlumberger forproviding the Petrel Interpretation software. We are grateful toAndrew Smith for correcting the English language. Reviews byMartin Hovland and Herald Ligtenberg significantly improved themanuscript.

References

Aguilera, R., 1995. Naturally Fractured Reservoirs, second ed. PennWell, Tulsa, OK.Andreassen, K., Hogstad, K., Berteussen, K.A., 1990. Gas hydrate in the southern

Barents Sea, indicated by a shallow seismic anomaly. First Break 8, 235e245.Andreassen, K., Nilssen, E., Ødegaard, C., 2007a. Analysis of shallow gas and fluid

migration within the Plio-Pleistocene sedimentary succession of the SWBarents Sea continental margin using 3D seismic data. Geo-Marine Letters 27,155e171.

Andreassen, K., Ødegaard, C.M., Rafaelsen, B., 2007b. Imprints of former ice streamsimaged and interpreted using industry three-dimensional seismic data fromthe south-western Barents Sea. In: Geological Society, London, Special Publi-cations, vol. 277, pp. 151e169.

Anell, I., Thybo, H., Artemieva, I.M., 2009. Cenozoic uplift and subsidence inthe North Atlantic region: geological evidence revisited. Tectonophysics 474,78e105.

Arntsen, B., Wensaas, L., Loseth, H., Hermanrud, C., 2007. Seismic modeling of gaschimneys. Geophysics 72, Sm251eSm259.

Berglund, L.T., Augustson, J., Færseth, R., Gjelberg, I., Ramberg-Moe, H., 1986. Theevolution of the Hammerfest Basin. In: Spencer, A.M., et al. (Eds.), Habitat ofHydrocarbons on the Norwegian Continental Shelf: Proceedings of an Inter-national Conference (Habitat of Hydrocarbons, Norwegian Oil and Gas Finds).Graham & Trotman for the Norwegian Petrolium Society, London, pp. 319e338.

Barnes, P.M., Lamarche, G., Bialas, J., Henrys, S., Pecher, I., Netzeband, G.L.,Greinert, J., Mountjoy, J.J., Pedley, K., Crutchley, G., 2010. Tectonic and geologicalframework for gas hydrates and cold seeps on the Hikurangi subductionmargin, New Zealand. Marine Geology 272, 26e48.

Berndt, C., Bünz, S., Mienert, J., 2003. Polygonal fault systems on the mid-Norwegianmargin: a long-term source for fluid flow. In: Geological Society, London,Special Publications, vol. 216, pp. 283e290.

Bjorøy, M., Hall, P.B., Ferriday, I.L., Mørk, A., 2009. Triassic Source Rocks of theBarents Sea and Svalbard. AAPG Convention, Denver, Colorado, USA.

Bjørkum, P.A., Walderhaug, O., Nadeau, P.H., 2001. Thermally driven porosityreduction: impact on basin subsidence. In: Geological Society, London, SpecialPublications, vol. 188, pp. 385e392.

Brekke, H., Riis, F., 1987. Tectonics and Basin Evolution of the Norwegian ShelfBetween 62-Degrees-N and 72-Degrees-N. In: Norsk Geol Tidsskr, vol. 67.pp. 295e321.

Carstens, H., 2005. Gas found in glacial, shallow sands. GeoExPro, Geoscience andTechnology Explained 4, 24e25.

Cartwright, J., Huuse, M., Aplin, A., 2007. Seal bypass systems. Aapg Bulletin 91,1141e1166.

Cavanagh, A.J., Di Primio, R., Scheck-Wenderoth, M., Horsfield, B., 2006. Severity andtiming of Cenozoic exhumation in the southwestern Barents Sea. Journal of theGeological Society, London 163, 761e774.

Chand, S., Mienert, J., Andreassen, K., Knies, J., Plassen, L., Fotland, B., 2008. Gashydrate stability zone modelling in areas of salt tectonics and pockmarks of theBarents Sea suggests an active hydrocarbon venting system. Marine andPetroleum Geology 25, 625e636.

Chand, S., Rise, L., Ottesen, D., Dolan, M.F.J., Bellec, V., Bøe, R., 2009. Pockmark-likedepressions near the Goliat hydrocarbon field, Barents Sea: morphology andgenesis. Marine and Petroleum Geology 26, 1035e1042.

Chand, S., Thorsnes, T., Rise, L., Brunstad, H., Stoddart, D., Bøe, R., Lågstad, P.,Svolsbru, T., 2012. Multiple episodes of fluid flow in the SW Barents Sea (LoppaHigh) evidenced by gas flares, pockmarks and gas hydrate accumulation. Earthand Planetary Science Letters 331e332, 305e314.

Connolly, D.L., Brouwer, F., Walraven, D., 2008. Detecting fault-related hydrocarbonmigration pathways in seismic data: implications for fault-seal, pressure, andcharge prediction. Gulf Coast Association of Geological Societies Transactions58, 191e203.

Dahl, B., Augustson, J.H., 1993. The influence of Tertiary and Quaternary sedimen-tation and erosion on hydrocarbon generation in the Norwegian offshorebasins. In: Doré, A.G., et al. (Eds.), Basin Modeling: Advances and Applications.NPF Special Publications 3. Elsevier, Amsterdam, pp. 419e432.

Dalland, A., Worsley, D., Ofstad, K., 1988. A Lithostratigraphical Scheme for theMesozoic and Cenozoic Succession Offshore Mid- and Northern Norway.Norwegian Petroleum Directorate Bulletin 4.

Dimakis, P., Braathen, B.I., Faleide, J.I., Elverhøi, A., Gudlaugsson, S.T., 1998. Cenozoicerosion and the preglacial uplift of the SvalbardeBarents Sea region. Tectono-physics 300, 311e327.

Doré, A.G., 1995. Barents Sea geology, petroleum resources and commercialpotential. Arctic 48, 207e221.

Doré, A.G., Jensen, L.N., 1996. The impact of late Cenozoic uplift and erosion onhydrocarbon exploration: offshore Norway and some other uplifted basins.Global and Planetary Change 12, 415e436.

Doré, A.G., Scotchman, I.C., Corcoran, D., 2000. Cenozoic exhumation and predictionof the hydrocarbon system on the NW European margin. Journal ofGeochemical Exploration 69e70, 615e618.

Dugan, B., Flemings, P.B., 2000. Overpressure and fluid flow in the New Jerseycontinental slope: implications for slope failure and cold seeps. Science 289,288e291.

Eidvin, T., Jansen, E., Riis, F., 1993. Chronology of Tertiary fan deposits off thewestern Barents Sea: implications for the uplift and erosion history of theBarents Shelf. Marine Geology 112, 109e131.

Eldholm, O., Faleide, J.I., Myhre, A.M., 1987. Continent-ocean transition at theWestern Barents Sea Svalbarrd continental-margin. Geology 15, 1118e1122.

Elverhøi, A., Solheim, A., 1983. The physical environment western Barents Sea, 1:1,500,000, sheet A, surface sediment distribution. Norsk Polar Institutt Rap-portserie 179A, 23.

Faleide, J.I., Solheim, A., Fiedler, A., Hjelstuen, B.O., Andersen, E.S., Vanneste, K., 1996.Late Cenozoic evolution of the western Barents Sea-Svalbard continentalmargin. Global and Planetary Change 12, 53e74.

Faleide, J.I., Tsikalas, F., Breivik, A.J., Mjelde, R., Ritzmann, O., Engen, O., Wilson, J.,Eldholm, O., 2008. Structure and evolution of the continental margin of Norwayand Barents Sea. Episodes 3, 82e91.

Faleide, J.I., Gudlaugsson, S.T., Jacquart, G., 1984. Evolution of the western BarentsSea. Marine and Petroleum Geology 1, 123e150.

Faleide, J.I., Vågnes, E., Gudlaugsson, S.T., 1993. Late Mesozoic-Cenozoic evolution ofthe south-western Barents Sea in a regional rift-shear tectonic setting. Marineand Petroleum Geology 10, 186e214.

Gabrielsen, R.H., Færseth, R.B., Jensen, L.N., Kalheim, J.E., Riis, F., 1990. StructuralElements of the Norwegian Continental Shelf Part I: the Barents Sea Region.Norwegian Petroleum Directorate Bulletin 6.

Gay, A., Lopez, M., Berndt, C., Séranne, M., 2007. Geological controls on focused fluidflow associated with seafloor seeps in the Lower Congo Basin. Marine Geology244, 68e92.

Gjelberg, J.G., 1981. Upper Devonian (Famennian) e Middle Carboniferous succes-sion of Bjørnøya: a study of ancient alluvial and coastal marine sedimentation.Norsk Polarinstitutt Skrifter 174, 1e67.

Gjelberg, J.G., 1987. Early Carboniferous graben style and sedimentation response,Svalbard. In: Miller, J., Adams, A.E., Wright, V.P. (Eds.), European DinantianEnvironments. John Wiley & Sons, Chichester, pp. 93e113.

Graue, K., 2000. Mud volcanoes in deepwater Nigeria. Marine and PetroleumGeology 17, 959e974.

Gudlaugsson, S.T., Faleide, J.I., Johansen, S.E., Breivik, A.J., 1998. Late Palaeozoicstructural development of the South-western Barents Sea. Marine and Petro-leum Geology 15, 73e102.

Heggland, R., 1998. Gas seepage as an indicator of deeper prospective reservoirs. Astudy based on exploration 3D seismic data. Marine and Petroleum Geology 15,1e9.

Heggland, R., 2004. Definition of geohazards in exploration 3-D seismic data usingattributes and neural-network analysis. Aapg Bulletin 88, 857e868.

Henriksen, E., Bjørnseth, H.M., Hals, T.K., Heide, T., Kiryukhina, T., Kløvjan, O.S.,Larssen, G.B., Ryseth, A.E., Rønning, K., Sollid, K., Stoupakova, A., 2011. Chapter17 Uplift and erosion of the greater Barents Sea: impact on prospectivityand petroleum systems. In: Geological Society, London, Memoirs, vol. 35,pp. 271e281.

S. Vadakkepuliyambatta et al. / Marine and Petroleum Geology 43 (2013) 208e221 221

Hjelstuen, B.O., Sejrup, H.P., Haflidason, H., Berg, K., Bryn, P., 2004. Neogene andQuaternary depositional environments on the Norwegian continental margin,62�Ne68�N. Marine Geology 213, 257e276.

Holbrook, W.S., Hoskins, H., Wood, W.T., Stephen, R.A., Lizarralde, D., 1996. Methanehydrate and free gas on the Blake Ridge from vertical seismic profiling. Science273, 1840e1843.

Hornbach, M.J., Ruppel, C., Van Dover, C.L., 2007. Three-dimensional structure offluid conduits sustaining an active deep marine cold seep. Geophysical ResearchLetters 34.

Hovland, M., Judd, A.G., 1988. Seabed Pockmarks and Seepages. Graham and Trot-man, London.

Hurst, A., Cartwright, J., Huuse, M., Jonk, R., Schwab, A., Duranti, D., Cronin, B., 2003.Significance of large-scale sand injectites as long-term fluid conduits: evidencefrom seismic data. Geofluids 3, 263e274.

Jackson, C., Stoddart, D., 2005. Temporal constraints on the growth and decay oflarge-scale mobilized mud masses and implications for fluid flow mapping insedimentary basins. Terra Nova 17, 580e585.

Jebsen, C., Faleide, J.I., 1998. Tertiary rifting and magmatism at the western BarentsSea margin (Vestbakken volcanic province). In: 3rd International Conference onArctic Margins, p. 92.

Johansen, S.E., Ostisty, B.K., Birkeland, Ø., Federovsky, Y.F., Martirosjan, V.N., BruunCristensen, O., Cheredeev, S.I., Ignatenko, E.A., Margulis, L.S., 1993. Hydrocarbonpotential in the Barents Sea region: play distribution and potential. In:Vorren, T.O., Bergsager, E., Dahl-Stamnes, Ø.A., Holter, E., Johansen, B., Lie, E.,Lund, T. (Eds.), Arctic Geology and Petroleum Potential. Elsevier, Amsterdam,pp. 273e320.

Judd, A., Hovland, M., 2007. Seabed Fluid Flow. Cambridge University Press,Cambridge.

Kjemperud, A., Fjeldskaar, W., 1992. Pleistocene glacial isostasy-implications forpetroleum geology. In: Larsen, R.M., Brekke, H., Larsen, B.T., Talleraas, E. (Eds.),Structural and Tectonic Modeling and its Application to Petroleum Geology. NPFSpecial Publications 1. Elsevier, Amsterdam, pp. 187e195.

Kvenvolden, K.A., 1998. A primer on the geological occurrence of gas hydrate. In:Geological Society, London, Special Publications, vol. 137, pp. 9e30.

Laberg, J.S., Andreassen, K., 1996. Gas hydrate and free gas indications within theCenozoic succession of the Bjornoya Basin, western Barents Sea. Marine andPetroleum Geology 13, 921e940.

Lammers, S., Suess, E., Hovland, M., 1995. A large methane plume east of Bear Island(Barents Sea): implications for the marine methane cycle. Geologische Run-dschau 84, 59e66.

Larsen, R.M., Fjæran, T., Skarpnes, O., 1993. Hydrocarbon potential of the NorwegianBarents Sea based on recent well results. In: Vorren, T.O., Bergsager, E., Dahl-Stamnes, Ø.A., Holter, E., Johansen, B., Lie, E., Lund, T. (Eds.), Arctic Geology andPetroleum Potential. Elsevier, Amsterdam, pp. 321e331.

Lerche, I., Yu, Z., Tørudbakken, B., Thomsen, R.O., 1997. Ice loading effects in sedi-mentary basins with reference to the Barents Sea. Marine and Petroleum Ge-ology 14, 277e338.

Ligtenberg, J.H., 2005. Detection of fluid migration pathways in seismic data: im-plications for fault seal analysis. Basin Research 17, 141e153.

Ligtenberg, J.H., 2007. Indications for Pressure Release from Zechstein Rafts duringLate Kimmerian e Implications for Reducing Drilling Risks, EAGE 69th Con-ference & Exhibition, London.

Løseth, H., Wensaas, L., Arntsen, B., 2002. Gas Chimneys e Indication of FracturedCap Rocks, AAPG Hedberg Conference, Vancouver, BC, Canada.

Løseth, H., Gading, M., Wensaas, L., 2009. Hydrocarbon leakage interpreted onseismic data. Marine and Petroleum Geology 26, 1304e1319.

Løseth, H., Wensaas, L., Arntsen, B., Hanken, N.M., Basire, C., Graue, K., 2011. 1000 mlong gas blow-out pipes. Marine and Petroleum Geology 28, 1047e1060.

Løseth, H., Wensaas, L., Arntsen, B., Hovland, M., 2003. Gas and fluid injectiontriggering shallow mud mobilization in the Hordaland Group, North Sea. In:Geological Society, London, Special Publications, vol. 216, pp. 139e157.

Maximov, S.P., Zolotov, A.N., Lodzhevskaya, M.I., 1984. Tectonic conditions for oiland gas generation and distribution on ancient platforms. Journal of PetroleumGeology 7, 329e339.

Meldahl, P., Heggland, R., Bril, B., de Groot, P., 2001. Identifying faults and gaschimneys using multiattributes and neural networks. The Leading Edge 20,474e482.

Mienert, J., Vanneste, M., Bünz, S., Andreassen, K., Haflidason, H., Sejrup, H.P., 2005.Ocean warming and gas hydrate stability on the mid-Norwegian margin at theStoregga slide. Marine and Petroleum Geology 22, 233e244.

Myhre, A.M., Eldholm, O., Sundvor, E., 1982. The margin between Senja and Spits-bergen fracture zones: Implications from plate tectonics. Tectonophysics 89,33e50.

Nesterov, I.I., Salvamanov, F.K., Kontorovich, A.E., Kulakhmetov, N.K., Surkov, V.S.,Trofimuk, A.A., Shpilman, V.I., 1990. West Siberian oil and gas superprovince. In:Geological Society, London, Special Publications, vol. 50, pp. 491e502.

Nyland, B., Jensen, L.N., Skagen, J., Skarpnes, O., Vorren, T.O., 1992. Tertiary uplift anderosion in the Barents Sea: magnitude, timing and consequences. In: Larsen, R.M.,Brekke, H., Larsen, B.T., Talleraas, E. (Eds.), Structural and Tectonic Modeling andits Application to Petroleum Geology. Elsevier, Amsterdam, pp. 153e162.

Ohm, S.E., Karlsen, D.A., Austin, T.J.F., 2008. Geochemically driven explorationmodels in uplifted areas: examples from the Norwegian Barents Sea. AapgBulletin 92, 1191e1223.

Ostanin, I., Anka, Z., di Primio, R., Bernal, A., 2012. Identification of a large UpperCretaceous polygonal fault network in the Hammerfest basin: implications onthe reactivation of regional faulting and gas leakage dynamics, SW Barents Sea.Marine Geology 332e334, 109e125.

Perez-Garcia, C., Feseker, T., Mienert, J., Berndt, C., 2009. The Håkon Mosby mudvolcano: 330 000 years of focused fluid flow activity at the SW Barents Seaslope. Marine Geology 262, 105e115.

Piggott, N., Lines, M.D., 1991. A case study of migration from the West Canada Basin.In: Geological Society, London, Special Publications, vol. 59, pp. 207e225.

Rasmussen, A., Kristensen, S.E., Van Veen, P.M., Stelan, T., Vail, P.R., 1993. Use ofsequence stratigraphy to define a semi-stratigraphic play in Anisian sequences,southwest Barents Sea. In: Vorren, T.O., Bergsager, E., Dahl-Stamnes, Ø.A.,Holter, E., Johansen, B., Lie, E., Lund, T. (Eds.), Arctic Geology and PetroleumPotential. Elsevier, Amsterdam, pp. 439e455.

Riis, F., Vollset, J., Sand, M., 1986. Tectonic development of the western margin of theBarents Sea and adjacent areas. In: Halbouty, M.T. (Ed.), Future PetroleumProvinces of the World. AAPG Memoirs, pp. 661e676.

Riis, F., 1992. Dating and Measuring Uplift and Subsidence in Norway and theNorwegian Shelf in Glacial Periods. In: Norsk Geol Tidsskr, vol. 72. pp. 325e331.

Sales, J.K., 1993. Closure vs. seal capacity-a fundamental control on the distributionof oil and gas. In: Doré, A.G., et al. (Eds.), Basin Modeling: Advances and Ap-plications. NPF Special Publications 3. Elsevier, Amsterdam, pp. 399e413.

Skagen, J.I., 1993. Effects of hydrocarbon potential caused by tertiary uplift anderosion in the Barents Sea. In: Vorren, T.O., et al. (Eds.), Arctic Geology and Pe-troleum Potential. NPF Special Publications 2. Elsevier, Amsterdam, pp. 711e719.

Sloan, E.D., Koh, C.A., 2007. Clathrate Hydrates of Natural Gases, third ed. CRC Press,Boca Raton, FL.

Smelror, M., Petrov, O., Larssen, G.B., Werner, S., 2009. Geological History of theBarents Sea. Geological Survey of Norway, Trondheim, Norway.

Solheim, A., Elverhøi, A., 1985. A pockmark field in the central Barents Sea; gas froma petrogenic source? Polar Research 3, 11e19.

Solheim, A., Elverhøi, A., 1993. Gas-related sea floor craters in the Barents Sea. Geo-Marine Letters 13, 235e243.

Solheim, A., Kristoffersen, Y., 1984. Sediments above the upper regional unconfor-mity; thickness, seismic stratigraphy and outline of the glacial history. NorskPolar Institutt Rapportserie 179B, 26.

Talukder, A.R., 2012. Review of submarine cold seep plumbing systems: leakage toseepage and venting. Terra Nova 24, 255e272.

Talukder, A.R., Comas, M.C., Soto, J.I., 2003. Pliocene to recent mud diapirism andrelated mud volcanoes in the Alboran Sea (Western Mediterranean). In:Geological Society, London, Special Publications, vol. 216, pp. 443e459.

Talwani, M., Eldholm, O., 1977. Evolution of Norwegian Greenland Sea. GeologicalSociety of America Bulletin 88, 969e999.

Vorren, T.O., Richardsen, G., Knutsen, S.-M., Henriksen, E., 1991. Cenozoic erosionand sedimentation in the western Barents Sea. Marine and Petroleum Geology8, 317e340.

Walderhaug, O., 1992. Magnitude of Uplift of the Stø and Nordmela Formations inthe Hammerfest Basin e a Diagenetic Approach. In: Norsk Geol Tidsskr, vol. 72.pp. 321e323.

Waylett, D.C., Embry, A.F., 1992. Hydrocarbon loss from oil and gas fields of theSverdrup Basin, Canadian Arctic Islands. In: Vorren, T.O., et al. (Eds.), ArcticGeology and Petroleum Potential. NPF Special Publications 2. Elsevier,Amsterdam, pp. 195e204.

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

Ziegler, P.A., 1988. Evolution of the Arctic-North Atlantic and the Western Tethys.Am. Assoc. Geol. Mem. No. 43, 198 pp.