Surface and subsurface expressions of gas seepage to the seabed—examples from the Southern North Sea

Marine and Petroleum Geology 22 (2005) 499–515 www.elsevier.com/locate/marpetgeo

Surface and subsurface expressions of gas seepage to the seabed— examples from the Southern North Sea Barthold M. Schroot*, Gerard T. Klaver, Ruud T.E. Schu¨ttenhelm Netherlands Institute of Applied Geoscience TNO, National Geological Survey, P.O. Box 80015, 3508 TA, Utrecht, The Netherlands Received 4 June 2003; accepted 31 August 2004

Abstract Expressions of gas seepage observable within North Sea seismic and acoustic data include seabed pockmarks, seepage plumes in the water column, acoustic blanking, shallow enhanced reflectors, and shallow seismic chimneys. Three areas were selected for a marine survey in which 60 vibrocores were taken. Gas content of the seabed sediment samples (methane and ethane concentrations of 2 to10,395 vppm and 0.5 to 2.2 vppm respectively were observed in the headspace gas) and carbon stable isotope ratios of methane (K88.3 to K30.5 ‰) were determined, in order to examine if geochemical anomalies confirm the interpretation of geophysical anomalies. In a few cases they did, whereas in other cases it was questionable whether only slightly elevated concentrations may be considered as a confirmation of the interpretation of geophysical anomalies. In the case of a seabed pockmark methane concentrations in headspace gas were elevated (122.6 vppm) in the samples taken from the centre of the pockmark compared with 5–10 vppm background values away from the pockmark. Very high concentrations of methane (up to 10,395 vppm) were found near an active vent, which had previously been identified via a seepage plume in the water column. Shallow enhanced reflectors in a 2D seismic profile confirm the presence of gas in the subsurface at this site. The methane concentrations (5 to 172 vppm) and carbon isotope ratios ranging from K88.9 to K30.5 ‰ found in samples taken above a Jurassic condensate field provide indications for leakage of a mixture of thermogenic and biogenic hydrocarbons to the seabed. q 2005 Elsevier Ltd. All rights reserved. Keywords: Gas seeps; Pockmarks; North Sea

1. Introduction In most of the world’s hydrocarbon basins phenomena related to the migration of hydrocarbons to the near-surface environment or to seepage into the biosphere can be found. We present observations from the southern North Sea basin, where various expressions of seepage to the seabed have been examined. The study of hydrocarbon seepage in general may be carried out for different reasons. First, the observation and measurement of seepage (through the detection of seismic anomalies and the determination of concentrations and stable isotope ratios from geochemical analysis of soil or seabed samples) can be used as an exploration tool. Second, hydrocarbon seepage and related accumulations of shallow gas may constitute a geohazard * Corresponding author. Tel.: C31 30 256 4640; fax: C31 30 256 4605. E-mail address: [email protected] (B.M. Schroot).

0264-8172/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpetgeo.2004.08.007

for drilling operations. Furthermore, natural seepage of hydrocarbons into the atmosphere may contribute to the greenhouse gases and therefore to global climate change. Finally, geological sequestration of CO2 is being considered as a means for mitigating climate change caused by humaninduced release of CO2 into the atmosphere. The seepage of gas (primarily methane) in the southern North Sea may be regarded as a natural analogue to potential leakage of CO2 from future subsurface storage locations.

2. Geological setting The present day southern North Sea basin can be viewed as a sedimentary basin which was dominated by rifting during most of the Mesozoic, with an acceleration in rifting activity at the transition from the Jurassic to the Cretaceous and which was basically in a post-rift sag phase during the Cenozoic (Ziegler, 1990). Occasionally rifting or thermal subsidence was interrupted by compressional tectonic events, such as those related to the Alpine deformation

500

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

phases. Salt tectonics, mainly involving mobilization of the Permian Zechstein evaporates, play a major role in most of the basin. The principal source rocks for hydrocarbons in the Netherlands part of the southern North Sea basin are the Upper Carboniferous coal beds and the Lower Jurassic Posidonia Shale, containing both marine algal sapropel and land derived organic material (type II kerogen). In the study area in the northern part of the Netherlands North Sea sector (Fig. 1), there are in addition a number of potential source rocks within the Kimmeridgian-Ryazanian sequences. These include the shales of the Middle Graben Formation and the bituminous Clay Deep Formation, which contains marine algal sapropel of the type I kerogen (Wong et al., 1989). The extent to which the Jurassic source rocks may be mature strongly depends on the location with respect to the main structural feature in the area, the N–S trending Dutch Central Graben. For instance the eastern section of block F2 and the western part of block F3 (Fig. 2) are in the inner part of the Central Graben where maximum subsidence took place and the Lower Jurassic shales entered the wet gas generation phase during the Jurassic. Even in the adjacent Outer Graben the base of the Zechstein is buried at depths of more than 5000 m and thick Triassic and Jurassic series are present (Schroot, 1991). Basin modelling in the area is

Fig. 1. The Netherlands North Sea sector showing the location of the study area consisting of quadrants A, B, E and F. Each full quadrant consists of 18 license blocks.

Fig. 2. Part of the study area with the locations of oil, gas and condensate fields; triangles show the positions of the 60 vibrocores taken in 2002.

somewhat complicated by the tectonic inversion which took place during the Late Cretaceous. While the Netherlands part of the Southern North Sea basin (Fig. 1) is typically a gas basin, with mainly Carboniferous source rocks, there are a number of oil and condensate fields holding hydrocarbons from Jurassic source rocks. In addition, there is a group of shallow gas fields, notably in the A and B blocks, with reservoirs of Pliocene to Pleistocene age, typically at depths of 600–700 m, where the origin of the gas has been subject to debate. The gas consists almost exclusively of methane. The sedimentary section in which shallow gas and the migration to the seabed are observed postdates the MidMiocene unconformity, a surface which is buried at 1000 to 1500 m depth in the north of the Netherlands North Sea. From the end of the Miocene onwards a complex fan delta system, with associated pro-delta deposits, gradually evolving into a fluvial delta and alluvial plain, prograded from the east over the Mid-Miocene unconformity (Sha, 1991; Overeem, 2002). These wedge-shaped units represent material from a large Baltic River System mainly consisting of mature sands, and fining to the west and upwards. Fluctuations in eustatic sea-level together with tectonic movements and shifting depocenters resulted in regressive and transgressive deposits. The marine facies was initially situated west, later northwest of the terrestrial facies. During the latest parts of the Early Pleistocene and the earlier parts of the Middle Pleistocene coastlines were situated periodically around the northern tip of or to the north of the Netherlands sector (Jeffery et al., 1991). However, occasional transgressions, interrupting the prevailing alluvial plain conditions, reached as far south as the present Dutch north coast. Sediments are predominately sandy with

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

the presence of minor clays and peat. Channelling is common and continuous reflectors scarce. The first glacial event reaching the present Dutch North Sea was the Elsterian (Marine Isotope Stage 12) glaciation (Long et al., 1988; Laban, 1995). Scandinavian and British ice masses coalesced and spread over most of the Netherlands sector, except the very south. Glacial channels up to 400 m deep were being excavated, mainly in an E–W belt crossing the Netherlands sector between 538 and 548 20 0 N (Cameron et al., 1986). Sediments generally consist of planar deposits of glacial clays and sandy outwash, while within the channels a coarse basal fill is overlain by laminated, clayey, lacustrine deposits with sandy transgressive deposits on top (Praeg, 1996). Ice-loading affected pre-existing faulting and salt tectonics, while the glacial channels disrupted sediment continuity and created pathways for fluids and gases. Next, transgressions resulted in sheets of marine sands with some clay near the transgression limits. The subsequent Saalian (M.I.S. 6) glaciation brought Scandinavian ice to the eastern part of the Netherlands sector where tills, glacial clays and sandy and gravelly outwash were laid down. Glacial channels were fewer and much shallower, but ice-pushing and tongue basins more common (Joon et al., 1990; Laban, 1995). The following interglacial again resulted in transgression sands. Falling sea-level at the end of that interglacial in combination with a glacial-conditioned seabed morphology resulted in sheetlike clays deposited in depressions, the largest of which centres around the P blocks (Cameron et al., 1989). British ice of the youngest (M.I.S. 2) glacial, the Weichselian, covered the NW of the Dutch North Sea sector resulting in glacial deposits and some glacial channels (Long et al., 1988; Laban, 1995). Outside the ice limit, discontinuous eolian sands and fluvial channel-fills were deposited. The Holocene saw a drowning of the Netherlands sector resulting in scattered, thin, muddy, lagoonal and tidal flat deposits overlain in most places by transgressive, reworked sand sheets (Cameron et al., 1989). The large saucer-shaped depression in the central-northern part of the sector has muddy sand and mud at the seabed. Sonar data indicate (through the observation of acoustic blanking and acoustic turbidity) that the sands immediately below these muddy sediments frequently contain gas. Present water depths in the study area vary from 30–45 m.

3. Data collection and analyses Our project aimed at a better description and understanding of expressions of shallow gas and its migration to the near-surface environment in the Netherlands North Sea sector. Traditionally most reports of such expressions have come from the northern part (A, B, E and F blocks) of this sector where this project was also focussed (Figs. 1 and 2). At an early stage of the project an inventory was made of observations from archive data thought to be related to

501

shallow gas or to seepage of hydrocarbons to the seabed (Schroot and Schu¨ttenhelm, 2003). Industrial seismic data (both 2D and 3D) and high frequency sub-bottom profiler data and sonar data were examined. Subsequently, during marine surveys in the summer of 2002, a few selected sites were visited, sub-bottom profiler records and 3D bathymetric data were acquired and vibrocores of the seabed sediments (maximum length 5.5 m) were taken at 60 sites (Fig. 2). 3.1. Survey and sampling method Based on the geophysical anomalies found earlier, a sampling survey was carried out with the RV Zirfaea from 19–23 August, 2002. For positioning a digital global positioning system (Sercel NR-103) was used with an accuracy of a few meters (water depths range from 30 to 45 m). Sediment sampling was carried out using a hydraulic vibrocorer. The corer was operated through the Zirfaea’s moon pool and produced undisturbed cores to a maximum penetration subsurface depth of 5.5 m, depending on the local conditions of the seabed sediments. Once on deck, the core was subdivided into one meter sections from the top downwards. Each of the core sections was subsampled immediately after retrieval on deck. Square holes were made halfway down the length of each section, large enough to subsample the centre of the core with a 60 or 30 ml syringe. Two subsamples were taken from each subsampling point. The first sample consisting of approximately 60 ml of sediment was transferred into a 1000 ml Schott Duran bottle. The bottle was closed with an aluminium screw cap fitted with a gas tight rubber plug in the middle. The rubber plug was used as a septum to extract gas for the geochemical analyses. The second sample was put in module jars provided by W.L. Gore and Associates. All subsamples were kept refrigerated at 4 degrees Celsius until they could be analyzed. The primary objective of the geochemical measurements was to determine whether or not the interpreted geophysical anomalies in the area could be validated using light hydrocarbon gas analyses on vibrocores, either through the headspace analysis or with the GOREe Survey for Exploration technique (described below). A known complication with both techniques is that lower levels of seepage are difficult to detect in the uppermost shallow sediments, because of gas desorption during core recovery and analysis as well as processes like sulfate reduction and methane oxidation in the zone of maximum disturbance (Abrams, 1996a). In that zone the composition of migrated hydrocarbons is likely to have been altered. No data are available over the depth distribution of that zone in this part of the North Sea. We assumed that the greater the depth below the seabed surface, the lower the rates of sulphate reduction and methane oxidation. Therefore, only the deepest samples at each site were sent to the laboratory

502

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

for headspace gas analyses. In total 86 samples were analyzed for the concentration of methane and ethane (results in Table 2) and only the samples with a methane concentration higher than 5 vppm were analyzed for the d13C of the methane. 3.1.1. The GOREe survey for exploration method The second set of samples was sent to W.L. Gore and Associates, Inc. to test the GOREe Survey for Exploration technique and examine how their results would relate to the results from the headspace gas analysis. The GOREe Survey for Exploration is a commercial method for the detection of organic compounds in the soil zone, at sensitivities in the ppb (10K9 g) range. Geochemical data acquisition is achieved using proprietary sample devices, termed modules, which are exposed to seabed core material for some period of time. A module contains selected adsorbents having affinities for organic compounds (hydrocarbons) in the light gaseous (C2) to light oil range (C18). The method is not optimal for the retention of methane. The adsorbents of the module are held within an inert, permeable membrane housing, which allows vapor diffusion to the adsorbents, and prevents liquid and solid material from coming into contact with the adsorbents (which would act to reduce the sensitivity of the module). Because this method of sampling relies on adsorbents to collect organic compounds present in the sample matrix, exposure times are set to several days. This allows equilibrium to be established between ambient diffusing organics and adsorbed organics, and yields a more representative composition of the organic compound mix to be measured. Analysis of modules is accomplished by thermal desorptiongas chromatography-mass spectrometry. Target compounds for this method are identified by simultaneous analysis of calibration standards, both for gaseous and liquid phase compounds. The method currently includes 87 organic compounds in the target list. The evaluation and interpretation of the resultant geochemical data is accomplished using standard multivariate statistical techniques, with an emphasis on detecting and mapping changes in the hydrocarbon compound pattern in the range of C2 to C18, across a survey of samples. For the geochemical data derived from seabed core samples of this program, 27

organic compounds were detected at levels substantially above inherent system levels (Table 1). The method includes certain types of compounds in the target list which are more likely to occur in onshore environments (for instance, the monoterpene class of compounds); these classes of compounds were not retained for the offshore application. The geochemical data were then processed with hierarchical cluster analysis, to determine the natural groupings of samples across the geochemical survey. Specific clusters of samples were evaluated as exhibiting prominent aliphatic compound response, and were flagged as areas of interest for hydrocarbon emanation. More information on this method may be found on the Gore website (www.gore.com/surveys). 3.2. Headspace gas analyses The Schott Duran bottles with the rubber stops were used for headspace gas analyses. In these bottles all the gas that evolved from the sediment after it was transferred into the bottle was collected and could be analyzed. As the exact amount of sediment (G60 ml) put into the Schott Duran bottles (headspace G940 ml) was not determined, the determination of gas concentrations must be considered as semi-quantitative. When the samples were taken out of the cooling, they were allowed to warm up to room temperature and were shortly shaken on a high speed shaker. During the analyses of the samples it became clear that the amount of sediment in the Schott bottles was too low to measure the C3 to C5 concentrations, or alternatively they are naturally too low to analyze in this part of the North Sea. All analyses were performed by Isolab, a commercial laboratory which specialises in geochemical services for the oil and gas industry (www. Isolab.com). Methane and ethane concentrations were analyzed using a GC fitted with a capillary column and a flame ionization detector (FID). Samples from the headspace were injected via a sample loop of 50 ml. The methane concentrations in the headspace of the sampling bottles were generally too low to measure isotope ratios directly, therefore methane concentrations were enriched. Methane enrichment was done in a short packed

Table 1 Significant detections of organic compounds by the GOREe Survey for Exploration method from the seabed sediment core samples 1-Butene Pentane Furan Carbon disulfide Cyclopentane 3-Methylpentane Hexane Methylcyclopentane 2,4-Dimethylpentane

Cyclohexane cis-1,3-Dimethylcyclopentane trans-1,2-Dimethylcyclopentane Heptane Methylcyclohexane 2,5-Dimethylhexane 3-Methylheptane cis-1,3/1,4-Dimethylcyclohexane Cycloheptane

cis-1,2-Dimethylcyclohexane trans-1,3/1,4-Dimethylcyclohexane trans-1,2-Dimethylcyclohexane Ethylcyclohexane Ethylbenzene m,p-Xylenes 1,3,5-Trimethylbenzene 1-Ethyl-4-methylbenzene Indane

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

503

Table 2 Summary of analytical results. The location of the three areas F3/F6, B13 and A11 is shown in Fig. 2 Area

F3/F6

Sample no.

02DW212M2 02DW213M3 02DW214M4 02DW214M5 02DW214M6 02DW215M3 02DW216M4 02DW216M5 02DW216M6 02DW217M4 02DW218M4 02DW218M5 02DW219M2 02DW220M3 02DW221M4 02DW221M5 02DW221M6 02DW222M4 02DW222M5 02DW223M3 02DW224M2 02DW225M4 02DW225M5 02DW226M4 02DW227M3 02DW228M2 02DW229M4 02DW230M1 02DW230M2 02DW230M3 02DW230M4 02DW231M4 02DW232M3 02DW232M4 02DW233M3 02DW234M3 02DW235M4 02DW236M3 02DW237M4 02DW237M5 02DW238M3 02DW239M4 02DW240M3 02DW241M4 02DW241M5 02DW242M4 02DW243M4 02DW244M2 02DW245M3 02DW246M3 02DW247M3 02DW248M3 02DW249M3 02DW250M3 02DW251M3 02DW252M1 02DW252M2 02DW252M2

Coordinates [m]

Water depth [m]

Sample depth below seabed 1.5 2.6 3.5 4.5 5.2 2.6 3.5 4.5 5.2 3.4 3.5 4.5 1.5 2.6 3.5 4.5 5.2 3.5 4.3 2.3 1.3 3.5 4.4 3.6 2.4 1.6 3.4 0.5 1.5 2.5 3.2 3.6 2.5 3.3 2.2 2.6 3.2 2.2 4.0 4.9 2.4 3.6 2.3 3.5 4.5 3.2 3.4 1.5 2.4 2.5 2.2 2.6 2.3 2.5 2.4 0.5 1.5 1.7

(UTM zone 31, X

ED50) Y

624500 624500 621498

6070251 6072252 6074249

43.8 43.1 42.9

624501 627502

6074250 6074249

42.6 42.6

625502 624496

6075250 6075253

42.3 42.3

623502 627499 625499

6075252 6076248 6076252

42.4 42.1 42.2

624496

6076249

42.2

623500 621500 617499

6076250 6076252 6076251

42.2 42.2 42.5

617502 621500 623498 624498 625500

6078250 6078250 6077251 6077248 6077250

42.4 41.8 41.9 41.9 41.9

627500 624500

6078250 6078252

41.7 41.9

624501 613502 610501 608502 608501

6080250 6078249 6078249 6078249 6077251

41.6 43.1 43.4 43.8 43.8

608502 624499 624502 624501

6075252 6082249 6084250 6086999

44.3 41.4 41.3 41.2

624500 624500 626499 626500 627500 627499 628499 628500 627501 627499 606502 clay sand

6087994 6089001 6088999 6087999 6087000 6088000 6087998 6088999 6088999 6089997 6078249

41.1 41 41.1 41.2 41.3 41.1 41.1 41 40.8 40.9 43.7

Gas in Headspace C1vppm (vppm) 3 3 4 3 4 3 3 3 3 5 4 3 3 4 9 4 3 5.9 9 4 4 4 3 7 3 6 3 13 15 2 2 3 2.9 4 3 6 3 5 4 4 9 3.8 4 8 6 5 4 6 7 2 5 4 5 5 3 25 66 12

13

d C1 ‰

C2vppm (vppm)

Cluster Family interpreted by GORE

Lithology of subsample

A D D F

K60.8

K59.3 K61.1

K59

K55.2 K52.5 K46.1 K45.9 K54.4 K46.9

1

F F B C C D D

1

D D F D F

1

F F F D F F F F

0.5 1

A F F

K50.3

Sand Sand Sand Silty sand Sand Sand Sandy silt

1 F

K51.8 K51.2

Silt, sand Silt, sand Silty sand Sand Sand Sand Sand Sand Sand Sandy silt Sand Sand

2 1

K57.6

K58.3

1

K59.3 K71.1 K51.9

0.9 0.9 0

F

Sand Sand Clayey silt Clayey silt Sand Sand

D F F F F

Sand

B A

Sand Peat C B

Sand Silty clay Sand (continued on next page)

504

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

Table 2 (continued) Area

Sample no.

Coordinates [m] (UTM zone 31, X

B13

A11

02DW252M3 02DW252M4 02DW253M3 02DW254M3 02DW255M3 02DW256M3 02DW257M4 02DW258M3 02DW259M2 02DW260M3 02DW261M1 02DW261M2 02DW261M3 02DW261M4 02DW262M4 02DW263M2 02DW264M4 02DW264M5 02DW265M4 02DW265M5 02DW266M3 02DW266M4 02DW267M1 02DW267M2 02DW268M3 02DW269M3 02DW270M2 02DW271M2

ED50) Y

Water depth [m]

608498 608500 608500 608501 613501 613503 620501 620500 569331

6079248 6081247 6083250 6085250 6088001 6088998 6088002 6089001 6129650

43.2 42.8 42.1 41.9 41.3 41.4 40.3 40.3 43.9

569331 570331 569330

6130651 6129650 6128658

43.8 44 44

568328

6129648

43.7

531884

6138096

34.3

531884

6138045

33.9

531883 531884 531885 531935

6137998 6137896 6138195 6138013

33.5 32.4 32.5 34

column at K90 8C. The enriched sample was then injected into a capillary plot column and analyzed with a GC-C-IRMS (a standard GC-CombustionCMat 253 set-up of Thermo, Bremen-Germany). The injection volume was typically 5 ml. The results are based on 3 to 5 measurements and reproducibility is generally within 0.5 ‰. Ethane isotopes could not be measured because of the high CO2 concentration in the headspace samples. In order to increase the methane and ethane concentrations from sandy samples, the bottles were heated to cause higher desorption of the methane and ethane from the sediment. However, heating of these samples did not result in significantly higher methane and ethane concentrations. Heating did not cause the d13C values of methane to change. 3.3. Analytical results The geochemical analyses were performed to examine possible correlations between the occurrence of geophysical anomalies and geochemical anomalies. Secondly, in case elevated methane concentrations, the question was raised whether the sorbed gas in the sediments is of biogenic or thermogenic origin. From the literature it is obvious that the best way to establish this is to use C1/C2CC3 ratios and carbon isotope ratios of methane and ethane.

Sample depth below seabed 2.5 3.4 2.4 2.6 2.4 2.5 3.2 2.4 1.5 2.3 0.5 1.5 2.5 3.5 3.5 1.2 3.5 4.3 3.5 4.4 2.5 3.2 0.5 1.5 2.6 2.5 1.5 1.6

Gas in Headspace 13

C1vppm (vppm)

d C1 ‰

C2vppm (vppm)

58 172 14.3 15 5.1 7.1 8 7 7 3 14 23 13 39 55 7 6 2 8983 10395 11.5 5 80 122.6 6.5 13.7 10.1 59

K72.5 K88.9 K30.5 K55.4 K41.4 K41.3

1.9 2.2

K41.6

1

Cluster Family interpreted by GORE A

Lithology of subsample Silty clay Sand Sand Sand Sand Sand

E F K56.9 K61.5 K63.1 K65.2 K58.8 K47.3 K59.9

0.9

2.2 2 1 1

K88.3 K83.8 K54.7

F F F B A F B A B A F

K36.8 K42.3 K46.1 K55.0 K50.5 K44.5

1

1

Silty sand Sand Sand Silty sand (Silty)sand Sand Silty clay Stiff clay Sandy clay Sand Sand Sand Sand Sand Sand (Silty)sand Sand Silty sand

The results of the geochemical analyses, together with locations, depth below sea bottom and lithology are given in Table 2. Many of the methane concentrations are below 10 vppm. We consider the ‘background’ value for the methane concentration in the study area, when applying this particular method, to be within the range of 3–10 vppm. We were not able to determine the C3 composition because the C3 concentration was below our detection limits. From Table 2 it can be seen that the ethane concentrations in most of the samples are below the detection limit (0.5 vppm). In 21 (out of 86) samples ethane could be detected. The highest concentration found is only 2.2 vppm. We decided not to use the ethane results in C1/C2 plots, because: 1. The variation between the highest and lowest measured ethane concentration is so low that we don’t know whether or not we did detect the background value in this area for ethane. The difference in background concentration may be caused by preferential loss of methane from the sediment column or during the coring and subsequent direct sampling. 2. We were not able to measure the d13C of the ethane, and could therefore not use the d13C of the ethane as an

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

independent criterion to test the meaning of the C1/C2 ratio. As far as the results of the GOREe Survey for Exploration method are concerned, the 10th column of Table 2 shows the cluster family for each sample which was analyzed by GORE resulting from the hierarchial cluster analysis. It is important to note that only the cluster families A and B are significant in terms of the detection of thermogenic compounds. Cluster family A has a geochemical signature evaluated as reasonably light thermogenic compound influence, including prominent alkane compounds in the C5 and C6 range. Cluster family B has a geochemical signature evaluated as slightly altered light thermogenic compound influence, with alkane compounds in the C6–C8 range. The other cluster families (C, D, E and F) were characterized as essentially devoid of thermogenic compound response.

4. Surface expressions 4.1. Pockmarks Seabed pockmarks are concave, crater-like depressions, which are commonly associated with the release of gas or fluids from the subsurface (King and MacLean, 1970; Hovland and Judd, 1988). These morphological features were first reported by King and MacLean (1970) to occur in profusion on muddy parts of the Scotian Shelf. They found diameters ranging from 15 to 45 m and depths of 5 to 10 m. The expression on echograms is that of a V-shaped notch along the sediment-water interface. During the nineteenseventies and nineteen-eighties pockmarks were also reported in large quantities from the North Sea (Hovland and Judd, 1988; McQuillin and Fannin, 1979). At that time it was thought that pockmarks did not occur in the shallower Southern North Sea. Later on, however, closer line spacing and better observation techniques led to discovery of some pockmarks to the south as well. In the Netherlands sector

Fig. 3. A 3.5 kHz sub bottom profiler record showing a seabed pockmark in block A5; diameter about 40 m and depth 2 m. Acoustic blanking suggests the presence of gas in the shallowest sediments.

505

pockmarks have been found on archive 3.5 kHz sub-bottom profiler records at two locations, in blocks A5 and F10 respectively (Fig. 2). These features are typically about 40 m in diameter and 2 m deep (Fig. 3). 4.2. Pockmark in Block A11 A side-scan sonar record taken by Hr. NlMS Tydeman in 1998 first showed a pockmark in block A11 (for location see Fig. 2). Its diameter was estimated to be 140–150 m (Schroot and Schu¨ttenhelm, 2003). Multi-beam records collected during our 2002 marine surveys gave further information about the morphological feature (Fig. 4). It turned out to be asymmetric in shape, and about 2 m deep at its deepest point near the eastern rim. Samples were collected from cores taken at six different locations in and near the pockmark. Maximum core length was 3.4 m. The samples were used for headspace analysis. Fig. 4 shows that the highest CH4 concentration (122.6 vppm) is found in the core from the very centre of the feature. This value is considered to be significantly higher than background values (Fig. 4). The location of the methane anomaly almost coincides with the presence of a fresh, smaller unit pockmark. Unit pockmarks are smaller features of a few meters in diameter, occurring within the larger feature (Hovland and Judd, 1988). Within the larger feature the unit pockmark probably represents the most recent spasmodic seepage. Fig. 5 shows the relationship between methane isotopic composition and methane concentration in the headspace gas for the six core locations sampled in block A11. Although the number of measurements is low, it is noted that all isotopic composition values are heavier than or equal to K55 ‰, and that the two highest methane concentrations correspond to the heaviest isotope values. These two samples are from within the pockmark. Moreover, ethane was detected in these two samples (1 vppm). Abrams (1996b) proposed three categories of methane isotopic composition based on similar cross-plots. The A11 measurements would belong to his Type A, with compositions in the thermogenic range. It must be noted however, that bacteria can alter the carbon isotopic composition of methane. Barker and Fritz (1981) demonstrated that bacterial oxidation leaves residual methane enriched in 13C. Therefore, the interpretation that the gas is of thermogenic origin is only justified if it is assumed that bacterial alteration did not play a significant role. The spatial distribution of the values relative to the pockmark feature (Fig. 4) together with the occurrence of ethane provides some indication that the A11 site shows seepage of thermogenic hydrocarbons or of a mixture of biogenic and thermogenic hydrocarbons. However, concentrations are too low to be very certain about such an interpretation and bacterial oxidation cannot be ruled out as an alternative explanation.

506

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

Fig. 4. Multi-beam image of pockmark in block A11 (Fig. 2), with six vibrocore locations. At each location the methane concentrations (in bold) and methane isotopic composition (d13C of CH4) measured in the headspace of samples, generally taken at 2–3 m below the seabed, are shown. Water depth is about 33 m.

The core sediments were characterized according to their depositional environment and incorporated into the regional litho- and chronostratigraphic system. The samples selected for geochemical analysis were all taken from sandy intervals without significant amounts of organic particles. In the cores from this particular pockmark no carbonate cement was found. The cores contain eolian sand from the later part of the last glacial overlain by marine sand from the Holocene transgression interbedded with some clayey intervals. Escape and injection structures accompanied by fractured and broken-up layers were observed in various places (Fig. 6). On this basis, pockmark formation was interpreted to have started towards the end of the glacial period and to have continued well into the marine transgression phase, whilst present activity is limited as outlined above.

gas analyses are shown in Table 2 and plotted on the map of Fig. 7. The most striking observation from the block B13 data is the very high anomaly of 10,395 vppm methane at site # 265, at a depth of 4.4 m below seabed. This site is about 1000 m west of the suspected seepage location, which was close to location # 261. The E–W sub-bottom profile crossing both these locations (Fig. 8) shows that the

4.3. Seepage plumes in the water column in block B13 A second site selected for further investigation was a rumoured seepage feature in block B13, located right over a known shallow gas field (Fig. 2). A survey company reported presumed gas venting in the water column around this location. Our survey resulted in 5 vibrocores (this small survey was designed to be centred around the reported seepage location) and 4 seismic, sub-bottom profiler and multi-beam lines. The results of the headspace

Fig. 5. Cross-plot of methane isotopic composition versus methane concentration in seabed sediment samples taken in and near the pockmark of block A11. The isotopic composition ranges from K42.3 ‰ to K55.0 ‰. The higher methane concentrations are from two sites within the pockmark; these samples also show the heaviest isotopic composition and they both contained 1 vppm ethane.

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

Fig. 6. Core photograph (vibrocore # 268 from the A11 pockmark) showing mud-filled injection structures in sand due to the escape of gas and fluids. Core width is 10 cm.

507

10,395 vppm CH4 value was found in sediments very close to a plume in the water column at site # 265. It should be noted that the profiles do not necessarily cross the precise seepage spots, but that the latter may be projected onto the 2D profiles from a small distance. The most prominent gas plume in the water was found about 60 m east of core location # 261 (Fig. 8). This core still shows an anomaly of 39 vppm CH4 at a depth of 3.5 m. In addition to the two plumes referred to above, the E–W profile across the two sites (Fig. 8) also shows three smaller plumes. All observed plumes are plotted in Fig. 7 as circles. The SW–NE running 2D seismic profile SNST87-03 (Fig. 9) shows the subsurface characteristics related to the seepage observed at the surface. First of all, there is a clear bright spot at about 550 msec (a depth of about 600 m). This bright spot corresponds to the gas reservoir of Plio-Pleistocene age, the outline of which is shown in Fig. 7. The reservoir is obviously leaking. In addition to the gas plumes of Fig. 8, so-called shallow enhanced reflectors (further discussed in one of the next sections) are visible in Fig. 9 as three distinct patches over the field at shallower levels. These shallow enhanced reflectors are indications for gas saturation of the near-surface sediments and are frequently observed in the area seabed (Schroot and Schu¨ttenhelm, 2003). From the configuration in block B13 it can be concluded that the field is leaking at individual and isolated locations, probably corresponding to pathways provided by locally concentrated fractures in the shallow subsurface. The d13C values of methane in each of the five cores from B13 are light in isotopic composition (Table 2). It can be

Fig. 7. Methane measurement in block B13. The survey is over a shallow gas field. C1 concentrations in the headspace gas of the deepest sample of each of the vibrocores are shown, units are vppm. Circles show locations where gas plumes in the water column were observed on two of the high frequency sub-bottom profiler records.

508

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

Fig. 8. High frequency sub-bottom profiler record running W–E across the B13 gas accumulation, showing gas plumes in the water column.

argued that the value of site # 263 should be disregarded, because the deepest available sample at that location is from a depth of 1.2 m below seabed only. We suspect that this one value of K47.3 ‰ measured at a depth of only 1.2 m at location # 263 is probably the result of shallow methane bacterial oxidation. This interpretation is supported by the observation in nearby core # 261, where a shift is observed from K65.2 ‰ at a depth of 3.5 m to K56.9 ‰ at 0.5 m, implying a change to methane enriched in 13C of 8.3 ‰ over 3 m. All remaining values in block B13 are lighter than K55 ‰ and have all been measured at depths of at least 3.5 m. However, the deepest samples of each core from block B13 (except the shallow # 263) were all classified by the GOREe Survey for Exploration method to belong to either

cluster family A or B (see Table 2). Cluster family A represents light thermogenic influence (including prominent alkane components in the C5–C6 range) and cluster family B to altered light thermogenic influence (with compounds in the C6–C8 range). This apparent contradiction between light carbon isotopic composition of methane and the presence of components with a higher molecular weight is very common in oceanic sediments containing only traces of sorbed gases (Abrams, 2005 this volume; Whelan et al., 1988). No isotopic compositions from reservoir fluids in B13 were available, but from a very similar shallow PlioPleistocene gas field in adjacent block B16 a value of K70.3 ‰ for the d13C of methane from reservoir depth

Fig. 9. Seismic section showing the bright spot corresponding to the Plio-Pleistocene gas field in block B13 and overlying patches of shallow enhanced reflectors indicating gas saturation of the near-surface environment.

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

509

5. Subsurface expressions

Fig. 10. Acoustic blanking in block F3.

(about 600 m) was reported (well B16-01). This indicates that the gas in the reservoir is of the same biogenic origin as would be interpreted from the headspace analysis. The gas composition in that well is 99.6% methane. The detailed 3D bathymetric information obtained did not show any pockmarks in or around the sites where the gas venting was observed in the water column. At most, only inconspicuous seabed phenomena (relief w0.1 m) were seen on an otherwise flat seabed. Gas originating from a steady overflow of multiple, stacked, shallow reservoirs, is apparently emanating from fluidized sand without causing any seabed expression.

Subsurface expressions related to the presence of gas can be observed on both standard industry seismic data (2D and 3D) and on higher frequency seismic and acoustic surveys, such as sub-bottom profiler records. The acoustic surveys reveal anomalies in the uppermost 25 m or so below seabed, whereas the industry seismic data displays the deeper subsurface. Because of the shallow water depth (30–45 m), standard industry seismic cannot image the uppermost 100 m or so properly, nor is there a clear seabed reflection in this data. Recently an inventory was made of the various subsurface gas expressions observed on archive data available from the Netherlands offshore (Schroot and Schu¨ttenhelm, 2003). In the very shallow realm (up to 25 m below sea bed) the expressions observed were predominantly acoustic blanking, with some acoustic turbidity. On standard and high frequency seismic data shallow enhanced reflectors were commonly observed. In a 1989 3D seismic survey from block F3 (position shown in Fig. 2) a shallow gas chimney over a salt dome and elsewhere an apparently leaking fault structure, also over a salt dome, were found. These features were selected for further investigation during the 2002 marine surveys. 5.1. Acoustic blanking In the Netherlands North Sea sector the most commonly observed gas-related, near-seabed feature is acoustic blanking. Acoustic blanking appears as patches where nearseabed reflections are suddenly faint or absent below a certain level. This may result from the disruption of

Fig. 11. Seismic profile from a 3D survey in block E17 (in the south of Quadrant E, see Fig. 1) showing shallow enhanced reflectors.

510

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

sediment layering by the migration of pore fluids or gas, or from the reflection of acoustic energy by overlying hard sediment (Judd and Hovland, 1992). However, in the Netherlands sector with its thick series of unconsolidated sands and clays, blanking is most likely caused by the absorption or scattering of acoustic energy in overlying, locally gas-charged sediments. High-frequency profilers show acoustic blanking clearly. Distribution and intensity is highly variable and uneven. Acoustic blanking is locally very common and widespread underneath or within clay caps and seabed muds. An example from block F3 is shown in Fig. 10. In the northern part of the Netherlands offshore blanking also occurs in channel-fill settings, in particular in Pleistocene glacial valley-fills (Cameron et al., 1986, 1989; Laban, 1995). Acoustic blanking there is largely within the upper, laminated, lacustro-glacial clayey fill. We found that acoustic blanking was particularly widespread (c. 50% of the records) above the condensate field in the SW of block F3. Cores taken there were also more prone to liquefaction during the coring process than elsewhere, suggesting an elevated gas content. On the other hand, acoustic blanking was not very clear around the A11 pockmark (location in Fig. 2), nor was it apparent directly underneath the active gas vents in block B13. However, to

the south and west of the vents large patches of acoustic blanking do occur. 5.2. Shallow enhanced reflectors Shallow enhanced reflectors at typical ‘depths’ from 100 to 300 msec two-way-travel time (TWT), corresponding to actual depths up to about 200 m below seabed, were observed on both 2D and 3D seismic surveys. A regional 2D survey covering the entire Netherlands offshore shows this phenomenon to occur quite frequently. Fig. 11 displays an example from a 3D survey from block E17, about 60 km to the south of our study area. The patches of very high amplitude reflectors are interpreted to relate to the presence of gas. In map view, on timeslices, the anomaly in E17 turns out to relate to an elongated (O10 km) and broad (O1 km) band crossing most of the block. A possible interpretation is that this band reflects the presence of glacial channels, in which the glacial clays constitute a local seal for upward migration to the seabed. We have recently confirmed this correlation between the occurrence of shallow enhanced reflectors and mapped glacial channels in more 3D seismic surveys in the area. The features visible in Fig. 11 do not show any clear indications for migration pathways from the deeper subsurface.

Fig. 12. Map of block F3, showing the location of the investigated seismic anomalies and locations of profiles and of the timeslice, shown in Figs. 13, 14 and 16 respectively; triangles mark core locations; methane concentrations shown in black, d13C of methane shown in red and numbers of key cores shown in blue.

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

511

Fig. 13. Seismic profile inline 695 of 3D survey across a leaking fault system.

5.3. Leaking fault systems Fig. 12 shows a map of block F3 with a number of features selected for the 2002 surveys. One of those features is a N–S trending fault system in the NE of block F3, which seems to provide a migration pathway for gas to the seabed. The extensional faults are related to a deeper salt structure (Fig. 13). At various points where the faults intersect porous layers, such as the Plio-Pleistocene sands known to bear gas elsewhere in the area, small bright spots are visible. These reflect localised gas pockets. From the seismic profile it is not clear if the gas actually migrates all the way to the surface. The faults do seem to continue up to the seabed. It is interesting to observe that about 5 km more to the west on the same profile (Fig. 13) a larger bright spot is visible right over a second and slightly less pronounced salt structure. This second salt structure does not show associated faults over its crest, and there are no indications on the seismic profile for upward movement of hydrocarbons. It is likely that the gas trapped in this bright spot migrated up dip (eastward) from the deeper part of the Central Graben in the west. The concentrations of methane in the headspace gas from the samples collected along the profile (Fig. 13) (see NE of map, Fig. 12) are too low to confirm any seepage to the seabed. 5.4. Shallow seismic chimneys Seismic chimneys, also called gas chimneys, are seismic anomalies associated with the upward movement of fluids or

free gas. In many published examples the chimneys are characterized by the loss of seismic signal, i.e. by low amplitudes and low seismic coherency (e.g. Heggland et al., 2000; Meldahl et al., 2001). The exact causes of these changes in seismic response are not always immediately obvious. Explanations may depend on the local circumstances. For instance, shallow velocity perturbations, which may or may not be related to gas saturation, can result in inadequate seismic processing of the underlying sections. Secondly, in case of a rather homogeneous fine-grained lithology the effect of the overall decrease in acoustic impedance, resulting from gas saturation on either side of an acoustic impedance interface, can be that the amount of impedance contrast also decreases. An alternative explanation for the loss of seismic signal would be the actual break-up of layering resulting from fluid escape processes. The latter is only likely to occur in a ’high energy’ environment, which implies active seepage. In the southeast of the 3D survey covering blocks F3 and F6 (Fig. 12) a shallow gas chimney was detected. The overall setting is shown in the seismic profile of Fig. 14. Between about 150 and 450 msec TWT the chimney is visible directly over a bright spot in Plio-Pleistocene gas bearing sands. Immediately adjacent to the chimney and cutting through the bright spot, there is a fault, which appears to continue all the way up to the seabed. This extensional fault is associated with the crest of a N–S trending salt structure. Fig. 15 shows the chimney in more detail and in Fig. 16 the feature is also visible as a high seismic amplitude anomaly in a time slice at 300 msec TWT. As is obvious both in the profile and in

512

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

Fig. 14. Seismic profile inline 190 of 3D survey across Pliocene-Pleistocene bright spots and a shallow chimney.

Fig. 15. Part of seismic profile inline 190 showing the high amplitude chimney.

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

513

Fig. 16. Time slice at 300 msec from the southeast of the F3 3D survey showing the expression of the shallow chimney and the adjacent fault.

the time slice, this particular chimney is characterized by an increase in seismic amplitude with respect to the surrounding sediments, and so differs from low amplitude chimneys shown in other papers. This difference may also imply a difference in migration processes responsible for the anomaly. It was noted earlier (Schroot, 2002) that high amplitudes together with the maintenance of reflector continuity must imply the preservation of sedimentary bedding. This particular chimney and its general geological setting resemble that of the chimney over the Machar field in the Central North Sea shown by Trasher et al. (1996). They interpret the chimney to represent ‘weak and localized’ seepage, a seepage style which is often related to salt diapirs. The salt structures act as foci for hydrocarbon migration. The overpressure in the Machar reservoir would be insufficient for fluids to induce fracturing, and thus the primary leaking mechanism must have been capillary failure of the top seal. Our attempt to validate seepage from this chimney by analyzing the headspace gas from the seabed sediment samples did not result in a positive confirmation. Fig. 12 shows both the methane concentrations and the isotopic compositions of methane measured in some of the samples. The methane concentrations are not significantly higher than background values, with the possible exception of one sample showing 15 vppm. But it is questionable if such a minor elevation is significant in view of the limitations of the sampling method. The isotopic composition of methane in the samples close to the chimney ranges from K50.3 to K61.1 ‰.

5.5. Geochemical measurements over the F3 condensate field A third part of the F3/F6 area on which our 2002 marine survey focused was the SW corner of block F3 over a producing Jurassic condensate field (Fig. 12). There are a few measurements over this field which show methane anomalies. One core, # 252 in the SW of the field, has samples with higher concentrations of methane (up to 172 vppm in the deepest sample from 3.4 m below seabed), together with a very light isotopic composition of methane at greater depth (K88.9 ‰). The isotopic composition gradually becomes heavier in the shallower samples (up to K51.9 ‰), while the methane concentration decreases, consistent with increasing bacterial oxidation in the shallower samples. The origin of the gas at site # 252 must be considered biogenic based on this range of methane d13C values. Closer to the crest of the field there are two sampling sites with slightly elevated methane concentrations (14 and 15 vppm) and heavier carbon isotope ratios (up to K30.5 ‰ in core # 253). 6. Discussion and conclusions In the northern part of the Netherlands North Sea sector we found a variety of surface and subsurface expressions of hydrocarbon seepage on geophysical data. Only some of the previously identified geophysical and bathymetric anomalies showed geochemical anomalies in sediment core samples.

514

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

The geochemical survey was successful in collecting samples from 60 sites in only 2.5 working days. However, the method needs to be improved. In the future the volume of sediment in the bottles, from which the headspace gases were sampled, must be increased in order to be able to determine the d13C of hydrocarbons heavier than methane. Because of the lack of these d13C measurements from the heavier hydrocarbons we were not able to test whether it is possible with the GOREe Survey for Exploration method to detect gas of thermogenic origin. The seabed pockmark in block A11 (Figs. 2 and 4) shows minor seepage from its centre. Formation of the pockmark probably occurred quite some time ago, from the end of the last glacial into the marine transgression phase, when seepage was more active and episodic. On the other hand, there is active and present-day seepage observed (e.g. through seepage plumes in the water column and shallow enhanced reflectors) over a shallow gas field in block B13. The gas has a very light isotopic composition of methane. However, there are also indications for the presence of higher molecular-weight hydrocarbons. Prinzhofer and Pernaton (1997) provide a possible mechanism for the very light isotopic composition of CH4 at the sites where there are also indications for thermogenic gas. These authors propose carbon isotope fractionation resulting from diffusion and argue that segregation of 12C and 13 C of methane resulting from diffusive migration has been an underestimated isotopic fractionation mechanism. Zhang and Krooss (2001) observed a significant depletion of 13C in methane during diffusion experiments, after having started with a shaley source rock containing methane with a d13C of K39.1 ‰. However, diffusion could only play a significant role under very specific conditions. Migration must be dominated by molecular transport, i.e. volume flow should not occur over an extended period of time and there should not be a steady supply of hydrocarbons (Krooss and Leythaeuser, 1996). Based on the observations in B13 and the general geological setting in the area we must conclude that such specific conditions do not prevail in this area. Hence, the isotopic composition in B13 can only be explained by a predominantly biogenic origin of the gas. There are a few independent indications that seepage from the B13 shallow gas field is far from laterally continuous. First, we observe a few discrete and separate seepage plumes. Second, methane concentrations decrease rapidly when moving away a small distance from the plumes. Finally, the shallow enhanced reflectors, indicating gas saturation in the near-surface sediments, appear in separate patches. From the fact that accumulations of shallow gas in the F3 and F6 blocks seem to be concentrated over salt structures that act as focal structures for migration, we conclude that there has been a significant lateral component in migration. This could either apply to biogenic gas moving towards the traps or to the mainly upward passage of subsurface fluids along the salt, providing nutrients and possibly substrates

for microbiological communities producing methane at shallower levels. Comparing the observations from blocks A11 and B13, we conclude that pockmarks and active gas venting are not necessarily mutually linked. The B13 area with present active venting does not contain pockmarks. A closer examination of three selected areas in blocks F3 and F6 results in an ambiguous impression of seepage in the area. The geochemical data mostly indicates a biogenic signature for the headspace gas. Sub-bottom profiler records from this area show acoustic blanking to be almost omnipresent. This means that any weak indication of seeping gas with a thermogenic signature is hidden behind the overwhelming biogenic signal. Seepage from the two geophysical anomalies in the east (a supposedly leaking fault system and a shallow seismic chimney) cannot be confirmed positively from the geochemical data. From the seismic data it is concluded that both these geophysical anomalies are related to salt tectonics. The seepage can be classified as weak and localized. The seismic character of the shallow chimney corresponds to a situation where sedimentary bedding inside the chimney has been preserved also indicative that seepage has been weak. The third selected area in block F3, the Jurassic condensate field, shows a few elevated methane concentrations related to the seepage of predominantly biogenic gas. At some locations a heavier carbon isotopic composition of methane suggests an admixture with thermogenic fractions. However, this isotopic composition may also be attributable to bacterial alteration. Bacterial oxidation of methane leaves the residual methane enriched in 13C (Barker and Fritz, 1981). Clayton et al. (1997) quantify this effect of biodegradation in the case of a North Sea field as an increase in d13C for both methane and ethane of about 10 ‰. If this observation may be generalised, our observation of a d13C of K30.5 ‰ over the condensate field could mean the seepage of some thermogenic hydrocarbons. However, evidence is not strong because the methane concentration in that sample was only 14.3 vppm.

Acknowledgements The support of Rijkswaterstaat-North Sea Directorate in realising the sea-going surveys, in particular of Messrs. T. Krijthe and P. Pronk (chief surveyors) and Messrs. S. Bicknese and R. Lambij (data processing) is gratefully acknowledged. We thank Dirk Maas for his assistance in collecting the samples. Also, the geochemical analyses performed by W. L. Gore and Associates, Inc., USA (Al Silliman) and by BGR, Germany, (Dr. Manfred Teschner) are gratefully acknowledged. Financial support to the NASCENT project was provided by DG Research of the European Commission.

B.M. Schroot et al. / Marine and Petroleum Geology 22 (2005) 499–515

References Abrams, M.A. (1996a). Distribution of subsurface hydrocarbon seepage in near-surface marine sediments. In: Schumacher, D., Abrams, M.A. Editors, 1999. Hydrocarbon migration and its near-surface expression: AAPG Memoir, 66, 1-14. Abrams, M.A. (1996b). Interpretation of methane carbon isotopes extracted from surficial marine sediments for detection of subsurface hydrocarbons In: Schumacher, D., Abrams, M.A. Editors, 1999. Hydrocarbon migration and its near-surface expression: AAPG Memoir, 66, 309-318. Abrams, M.A. (2005, this volume). Significance of hydrocarbon seepage relative to sub-surface petroleum generation and entrapment. Barker, J.F., Fritz, P., 1981. Carbon isotope fractionation during microbial methane oxidation. Nature 293, 289–291. Cameron, T.D.J., Laban, C., Mesdag, C.S., Schu¨ttenhelm, R.T.E., 1986. Quaternary Geology, Indefatigable Sheet 538 N-028 E. 1:250 000 map series, British Geological Survey and Rijks Geologische Dienst, Ordnance Survey, Southampton 1986. Cameron, T.D.J., Schu¨ttenhelm, R.T.E., Laban, C., 1989. Middle and Upper Pleistocene and Holocene stratigraphy in the Southern North Sea between 528 and 548 N, 28 to 48 E. In: Henriet, J.P., de Moor, G. (Eds.), The Quaternary and Tertiary Geology of the Southern Bight, North Sea. Belgian Geological Survey, Brussels, pp. 119–135. Clayton, C.J., Hay, S.J., Baylis, S.A., Dipper, B., 1997. Alteration of natural gas during leakage from a North Sea salt diapir field. Marine Geology 137, 69–80. Heggland, R., Meldahl, P., de Groot, P., Aminzadeh, F., 2000. Chimney cube unravels subsurface. American Oil and Gas Reporter, February 2000. Hovland, M., Judd, A.G., 1988. Seabed pockmarks and seepages, impact on geology, biology and the marine environment. Graham and Trotman, London p. 293. Jeffery, D.H., Laban, C., Mesdag, C.S., Schu¨ttenhelm, R.T.E., 1991. Map sheet Dogger, 1:250,000, Quaternary Geology, British Geological Survey/Rijks Geologische Dienst, Southampton 1991. Joon, B., Laban, C., van der Meer, J.J.M., 1990. The Saalian glaciation in the Dutch part of the North Sea. Geologie and Mijnbouw 69, 151–158. Judd, A.G., Hovland, M., 1992. The evidence of shallow gas in marine sediments. In: Davis, A.M. (Ed.), Methane in marine sediments Continental Shelf Research, 12, pp. 1081–1097. King, L.H., MacLean, B., 1970. Pockmarks on the Scotian Shelf. Geological Society of America Bulletin 81, 3141–3148. Krooss, B.M., Leythaeuser, D. (1996). Molecular diffusion of light hydrocarbons in sedimentary rocks and its role in migration and dissipation of natural gas. In: Schumacher, D. and Abrams, M.A. Editors, 1999. Hydrocarbon migration and its near-surface expression: AAPG Memoir, 66, 173-183. Laban, C. (1995). The Pleistocene glaciations in the Dutch sector of the North Sea. PhD Thesis, University of Amsterdam, 200 pp.

515

Long, D., Laban, C., Streif, H., Cameron, T.D.J., Schu¨ttenhelm, R.T.E., 1988. The sedimentary record of climatic variation in the southern North Sea. Philosophical Transactions of the Royal Society London, 318B 1988; 523–537. McQuillin, R., Fannin, N.G.T., 1979. Explaining the North Sea floor. New Scientist 1163, 90–92. Meldahl, P., Heggland, R., Bril, B., de Groot, P., 2001. Identifying faults and gas chimneys using multi-attributes and neural networks. The Leading Edge 20 (5), 474–482. Overeem, I., (2002). Process-response simulation of fluvio-deltaic stratigraphy, PhD. Thesis, Delft Technical University, 170 pp. Praeg, D. (1996). Morphology, stratigraphy and genesis of buried Elsterian tunnel valleys in the southern North Sea basin, PhD. Thesis, University of Edinburgh, 207 pp. Prinzhofer, A., Pernaton, E., 1997. Isotopically light methane in natural gas: bacterial imprint or diffusive fractionation? Chemical Geology 142, 193–200. Schroot, B.M., 1991. Structural development of the Dutch Central Graben. In: Michelsen, O., Frandsen, N. (Eds.), The Jurassic in the Southern Central Trough, Danmarks Geologiske Undersøgelse, DGU serie B-16, pp. 32–35. Schroot, B.M., 2002. North Sea shallow gas as a natural analogue in feasibility studies on CO2 sequestration, In: Extended Abstracts of the 64th EAGE Meeting and Technical Exhibition, Florence, paper H010 2002 p. 4. Schroot, B.M., Schu¨ttenhelm, R.T.E., 2003. Expressions of shallow gas in the Netherlands North Sea. Netherlands Journal of Geosciences/Geologie en Mijnbouw 82 (1), 91–105. Sha, L.P. (ed.), (1991). Quaternary Sedimentary Sequences in the southern North Sea basin, Final discipline report of the project: The Modelling And Dynamics Of The Quaternary Geology Of The Southern North Sea And Their Applications To Environmental Protection And Industrial Developments, CEC DGXII, Brussels, Sci. Progr. Contract No. SCI*-128-C 9EDB: 135 pp. Trasher, J., Fleet, A.J., Hay, S.J., Hovland, M., Du¨ppenbecker, S., (1996). Understanding geology as the key to using seepage in exploration: spectrum of seepage styles In: Schumacher, D. and Abrams, M.A. Editors, 1999. Hydrocarbon migration and its near-surface expression: AAPG Memoir, 66, 223-241. Whelan, J.K., Simoneit, B.R.T., Tarafa, M.E., 1988. C1–C8 hydrocarbons in sediments from Guaymas Basin, Gulf of California-Comparison to Peru Margin, Japan Trench and Califronia Borderlands. Org. Geochem. 12 (2), 171–194. Wong, T.E., van Doorn, T.H.M., Schroot, B.M., 1989. “Late Jurassic” Petroleum Geology of the Dutch Central North Sea Graben. Geologische Rundschau 78 (1), 319–336. Zhang, T., Krooss, B.M., 2001. Experimental investigation on the carbon isotope fractionation of methane during gas migration through sedimentary rocks at elevated temperature and pressure. Geochimica et Cosmochimica Acta 65, 2723–2742. Ziegler, P.A., 1990. Geological Atlas of Western and Central Europe, 2nd ed. Elsevier for Shell Internationale Petroleum Maatschappij, Amsterdam p. 239.