AUTHORS

R. Jonk � Department of Geology and Petro-leum Geology, University of Aberdeen, AB243UE, Aberdeen, United Kingdom; presentaddress: ExxonMobil Exploration Company,222 Benmar Drive GP8-448 Houston, Texas77060; rene.jonk@exxonmobil.com

Rene Jonk received his M.Sc. degree in structuralgeology from the Free University Amsterdam(1999) and a Ph.D. from the University ofAberdeen (2003) studying the origin and diagen-esis of sand injectites. He is currently workingwith ExxonMobil in Houston.

A. Hurst � Department of Geology and Petro-leum Geology, University of Aberdeen, AB243UE, Aberdeen, United Kingdom;a.hurst@abdn.ac.uk

Andrew Hurst holds the chair of Production Geo-science at the University of Aberdeen, UnitedKingdom. He has a B.S. degree from Aberdeenand a Ph.D. from Reading (United Kingdom).Prior to joining the academia in 1992, he workedfor more than 12 years in the international oiland gas exploration and production industry. Hiscurrent research includes sand injectites, deep-water clastic systems, sediment composition, andthe nondestructive analysis of porous media.

D. Duranti � Badley Ashton America, Hous-ton, Texas; davideduranti@baai-houston.com

Davide Duranti received a Ph.D. in sedimentologyfrom the University of Bologna (Italy). He wasa research fellow for 4 years at the University ofAberdeen (United Kingdom) and a geological con-sultant for various oil companies. His researchfocused primarily on the injected sandstones as-sociated with the deep-water reservoirs of the NorthSea. He recently joined Badley Ashton America.

J. Parnell � Department of Geology andPetroleum Geology, University of Aberdeen,AB24 3UE, Aberdeen, United Kingdom;j.parnell@abdn.ac.uk

John Parnell is a professor in the Department ofGeology and Petroleum Geology at the Universityof Aberdeen, where he has taught since 1999. Hereceived his B.A. degree from the University ofCambridge and his Ph.D. from Imperial College,London. He is an editor of the journal Geofluids.His research is focused on the composition,evolution, and migration of fluids in sedimen-tary basins.

Origin and timing of sandinjection, petroleum migration,and diagenesis in Tertiaryreservoirs, south VikingGraben, North SeaR. Jonk, A. Hurst, D. Duranti, J. Parnell,A. Mazzini, and A. E. Fallick

ABSTRACT

Petrographic, fluid-inclusion, and carbon and oxygen stable isotope

studies of Tertiary injectite reservoirs in the south Viking Graben of

the North Sea allow an understanding of the origin and timing of

sand injection, petroleum migration, and diagenesis. Injection from

shallowly (<400 m; <1300 ft) buried Paleocene and Eocene uncon-

solidated sandstones occurred at the end of the Eocene, probably in

response to earthquake activity. Liquid oil was already present in

the parent sands prior to injection and leaked from the injectites to

the seabed. Upward-migrating oil and basinal brines mixed with

downward-invading mixed meteoric-marine pore fluids in the in-

jectites, causing extensive biodegradation of the oil. Biodegradation

of oil provided the driver for early carbonate cementation in inject-

ites, causing diminished reservoir quality. However, early carbonate

cementation also sealed off the injectites as potential escape routes

for petroleum from the underlying parent sands. Oil (and gas) con-

tinued to migrate into the reservoir (parent) sands upon increased

burial, causing a mixing of high-API oil with the early charged,

extensively biodegraded low-API oil. The study of early diagenetic

imprints reveals the evolution of injectite reservoirs, which forms

the basis for understanding how to explore and develop them.

INTRODUCTION

Outcrops of injected sandstones have been an intermittent subject

of interest for almost 200 yr (e.g., Murchison, 1827; Diller, 1890;

Newsome, 1903; Waterston, 1950; Winslow, 1983; Jolly et al., 1998).

Typically, field descriptions are given of what is considered to be a

AAPG Bulletin, v. 89, no. 3 (March 2005), pp. 329–357 329

Copyright #2005. The American Association of Petroleum Geologists. All rights reserved.

Manuscript received February 12, 2004; provisional acceptance June 24, 2004; revised manuscriptreceived October 15, 2004; final acceptance October 26, 2004.

DOI:10.1306/10260404020

geological curiosity, sometimes offering explanations for their origin

(Murchison, 1827; Diller, 1890; Newsome, 1903; Waterston 1950)

and, more recently, adding paleostress studies of the geometries

of sandstone dikes, which aids in understanding the stress regime

at the time of dike formation (Huang, 1988; Boehm and Moore,

2002; Jonk et al., 2003). During the past decade, numerous de-

scriptions have emerged of large-scale (hundreds of meters verti-

cally and several square kilometers areally) sand injectite complexes

(all sandy facies associated with sand injection [see Duranti et al.,

2002] are hereafter referred to as sand injectites [after Hurst et al.,

2003]) associated with Tertiary petroleum reservoirs in the north-

ern North Sea (Jenssen et al., 1993; Dixon et al., 1995; Lonergan

and Cartwright, 1999; Lonergan et al., 2000; Bergslien, 2002;

Duranti et al., 2002; Purvis et al., 2002; Hurst et al., 2003; Duranti

and Hurst, 2004; Huuse et al., 2004). These studies have largely

used seismic, core, and wire-line-log data, and although they assist

in the recognition of sand injectites, they shed little light on the

timing and genesis of sand injection.

Fluid overpressuring in a sand body is believed to be required

for sand injection to occur (Jolly and Lonergan, 2002; Duranti

and Hurst, 2004), and three mechanisms of fluid overpressuring

and subsequent sand injection are proposed:

� disequilibrium compaction when pore-fluid expulsion from sedi-

ments upon burial is reduced because of rapid loading and/or

good seal integrity (Osborne and Swarbrick, 1997)� seismically induced liquefaction (Obermeier, 1996)� addition of overpressured (petroleum) fluids (Lonergan et al.,

2000)

Each of these mechanisms is invoked to varying extents to ex-

plain the sand injectites in the Tertiary petroleum reservoirs of the

northern North Sea (Lonergan et al., 2000; Jolly and Lonergan, 2002;

Duranti and Hurst, 2004; Huuse et al., 2004).

Because sand injectites form as a product of sand fluidization

and, once formed, continue to act as fluid conduits, study of their

diagenetic evolution should elucidate the relative timing of sand

injection, the involvement of petroleum, and the cementation his-

tory, which are vital for understanding sand injectite petroleum plays

(Figure 1). For example, injectites and underlying depositional sand-

stones (parent and parent sand body, hereafter used to describe the

depositional sand bodies from which sand injectites have emanated)

in Paleogene fields in the northern North Sea are strongly affected by

early diagenetic cements, particularly carbonate (Watson et al., 1995).

Hence, the presence or absence of petroleum fluid inclusions in dia-

genetic carbonate helps in understanding the relative timing of sand

injection and petroleum migration (Figure 1). Here, we present core

and diagenetic data from four petroleum accumulations in the south

Viking Graben of the northern North Sea (Figure 2) and attempt to

unravel the timing and origin of sand injection, petroleum migration,

and cementation in the Paleogene petroleum system.

A. Mazzini � Department of Geology and Pe-troleum Geology, University of Aberdeen, AB243UE, Aberdeen, United Kingdom;a.mazzini@abdn.ac.uk

Adriano Mazzini received his M.Sc. degree in ma-rine geology from the University of Genoa (1997)and a Ph.D. from the University of Aberdeen (2004)studying methane-related authigenic carbonatesand hydrocarbon-plumbing systems. He is cur-rently a research assistant in the Department ofGeology and Petroleum Geology at the Universityof Aberdeen.

A. E. Fallick � Scottish Universities Environ-mental Research Center, G75 0QF, East Kilbride,United Kingdom; T.Fallick@surrc.gla.ac.uk

Anthony E. Fallick graduated from Glasgow Uni-versity with degrees in physics (B.Sc.) and chem-istry (Ph.D.). He held postdoctoral positions ingeology and geography at McMaster University(Ontario) and in mineralogy and petrology atCambridge University before moving in 1980to East Kilbride. He is currently director of Scot-tish Universities Environmental Research Centerand professor of isotope geosciences in GlasgowUniversity.

ACKNOWLEDGEMENTS

An AAPG Grant-in-Aid donated in 2002 to thefirst author assisted in performing fluid-inclusionand carbon and oxygen stable isotope studies.Kristine Holm (TotalFinaElf), Karen Martin (BP),and Gerhard Templeton (Kerr-McGee) arethanked for assisting with core examinationand sampling. Mads Huuse is thanked for hisscientific support throughout the duration ofthe Injected Sands Project at the Universityof Aberdeen (2000 –2002). J. R. Boles, D. W.Houseknecht, and D. A. Pivnik are thanked forconstructive reviews that helped improve themanuscript.

DATASHARE 17

Datashare Tables 1–3 are accessible in anelectronic version on the AAPG Web siteas Datashare 17 at <www.aapg.org/datashare/index.html> .

330 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Figure 1. Cartoon illus-trating two injectite reser-voirs, with a different rela-tive timing of sand injection,petroleum migration, and(carbonate) cementation.Scenario A: Sand injectionoccurs at shallow (a fewhundred meters) burialdepth, followed by local-ized carbonate cementa-tion. After subsequentburial (indicated with theadded dashed box ofmudstone in the last car-toon of scenario A), petro-leum migration occurs.Patches of pervasive car-bonate cementation pre-vent the migration ofpetroleum into some ofthe injectites, whereassome others, along withthe top part of the under-lying parent sandstone,form petroleum reser-voirs. Scenario B: Petro-leum migrates into theunconsolidated sand-stones at shallow (a fewhundred meters) burialdepth. Sand injection post-dates petroleum migra-tion and petroleum escapesthrough those injectitesthat reach the seabed(dashed arrows). Somepetroleum may be trappedin injected sandstonesthat do not reach theseabed. Late carbonatecementation has no con-sequence as possible bar-riers to petroleum flow,apart from diminishingthe reservoir quality insome of the petroleum-bearing injectites.

Jonk et al. 331

THE TERTIARY PETROLEUM SYSTEM OF THESOUTH VIKING GRABEN

During the Cenozoic, the North Sea was an intracra-

tonic sag basin developed over a series of failed Meso-

zoic rift structures (Ziegler, 1990).The studied petroleum

accumulations have reservoirs of Paleogene age located

in the south Viking Graben (Figure 2), which, at that

time, was situated in a marine basinal setting with wa-

ter depths of 400–600 m (1300–2000 ft) (Joy, 1992).

Although the Tertiary section in the northern North

Sea is relatively unaffected by tectonic deformation, tec-

tonic activity did have a major influence on sedimen-

tation (Bowman, 1998). Rifting of the Greenland–

European plate in the early Paleocene caused the thermal

uplift of Scotland and the East Shetland platform (Bow-

man, 1998; Haaland et al., 2000; Ahmadi et al., 2003).

Clastic sediments derived from these uplifted areas

were deposited as deltaic and deep-water systems in

the south Viking Graben (Dixon et al., 1995); hence,

the majority of the Paleogene reservoirs are turbidite

systems sourced from the west.

The Paleogene of the northern North Sea is divided

into several basinwide sequences (Deegan and Scull,

1977; Mudge and Copestake, 1992; Bowman, 1998)

(Figure 3). Most of the examined core is from the lower

Eocene Balder and Frigg formations. The Balder For-

mation comprises laminated pyrite-bearing smectitic

mudstones interbedded with numerous, centimeter-

and decimeter-thick tuff beds, and thick (up to some

hundred meters), largely structureless, poorly cemented,

fine- to medium-grained sandstones. These thick sand-

stone units are mostly lobate in plan view and severely

mounded, with flanks as steep as 20j, in cross section

(Dixon et al., 1995; Huuse et al., 2004). The overlying

Frigg Formation consists of smectitic mudstones inter-

bedded with thin (<10-m [<33-ft]-thick) sandstones.

METHODS

Core intervals containing injectites were examined,

after which samples were taken for diagenetic studies.

Broken surfaces of samples were first examined using

Figure 2. Left: location of the south Viking Graben to the east of the East Shetland platform. Dashed box indicates the location of detail on right.Right: detail of dashed box in left diagram showing the outlines of the petroleum accumulations (gray) and the locations of cores from wellsstudied (indicated in black). The license blocks are also indicated, along with the Norwegian–United Kingdom license boundary (dashed line).

332 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

an ISI ABT-55 scanning electron microscope with a

Link analytical AN10155S electron-dispersive system

with a beam current of 30–70 mA and an acceleration

voltage between 15 and 25 kV. Following this first ex-

amination, samples were selected for thin sectioning.

Samples were mounted on 75 � 23-mm (3 � 1-in.)

glass slides and ground to a thickness of 30 mm. Samples

were prepared as uncovered polished sections and first

examined using a polarization microscope. Following

this, uncovered polished sections were examined for

different types of carbonates using a Cambridge CITL

Cold Cathodoluminescence 8200 mk3 with operating

beam current conditions of 200–300 mA and an ac-

celeration voltage of 15 kV. Carbon-coated polished

sections were investigated for authigenic quartz and

feldspar using an Oxford Instruments Hot Cathodolu-

minescence detector on a Link analytical AN10155S

electron-dispersive system. Operating beam current

conditions were 30–70 mA with an acceleration voltage

of 25 kV. On the same system, at beam current con-

ditions between 3 and 5 nA, the chemical composition

of diagenetic minerals was determined using a back-

scattered electron detector. Following petrographic stud-

ies, samples were selected for fluid-inclusion studies.

Samples for microthermometric analysis were prepared

as doubly polished wafers and examined using a Linkam

THM600 heating-freezing stage attached to a Nikon

Optiphot2-POL microscope. Finally, carbon and ox-

ygen stable isotope analysis of authigenic carbonates

were carried out at the Scottish Universities Environ-

mental Research Center. Samples of about 1 mg were

drilled from polished slabs with a spatial resolution

of about 1 mm. Sample powders were placed in a

temperature-controlled block at 70jC. Ultrapure he-

lium was used to purge all atmospheric gases. Samples

were digested in 103% H3PO4 (Wachter and Hayes,

1985) and were left for 8 hr to react. Gas was extracted

on a AP2003 Gas Prep Interface by overpressurizing

Figure 3. Lithostratigra-phy and nomenclature ofthe northern North Sea(modified after Mudgeand Copestake, 1992).

Jonk et al. 333

with helium to 5 mbar. The resulting gas was analyzed

using an AP2003 triple-collector mass spectrometer with

a 1s reproducibility of ±0.1xfor both d13C and d18O

values. All data are reported as per-mil deviation from

the Vienna Peedee belemnite (V-PDB) standard unless

stated otherwise.

CORE DESCRIPTION

A total of 904 m (2965 ft) of core from 11 wells from

the 4 studied fields were examined (Figure 2; also

see Datashare Table 1 at AAPG’s Datashare Web site

[Datashare 17], www.aapg.org/datashare/index.html).

The A, B, and C fields are characterized by 10–100-m

(33–330-ft)-thick, largely structureless, unconsolidated

(parent) sands in the Balder Formation, overlain by up

to 100-m (330-ft)-thick sections of interbedded sand-

stone and mudstone, mainly in the upper part of the

Balder Formation and the overlying Frigg Formation.

Only the uppermost 10 m (33 ft) of a 50-m (160-ft)-

thick sandstone body enclosed in the lower part of the

Balder Formation were recovered from the D field.

This thick unit is overlain by generally centimeter-thick

sandstones encased in mudstones and abundant tuffs

and is underlain by a sand-rich (Hermod Member) Sele

Formation.

Four sandy facies, described in the subsequent sec-

tion, are distinguished in the core: depositional sand-

stones, depositional sandstones affected by sand remo-

bilization, injected sandstones, and injection breccias.

Depositional Sandstones

Surprisingly few sandstones show distinct primary de-

positional structures; however, a few thick (individual

beds typically on the order of 1–2 m [3.3–6.6 ft]), un-

cemented depositional sandstones occur in the Balder

intervals of the A, B, and C fields. These beds are char-

acterized by parallel lamination, erosional bases, and

concordant tops and have mudstone clasts aligned

parallel to bedding (Figure 4A). These are believed to

be typical structures of sediment gravity flow depos-

its (Middleton and Hampton, 1973; Leeder, 1982).

Figure 4. Core examples of de-positional sandstones. (A) Relativelythick (1 – 2-m; 3.3 –6.6-ft), unce-mented, oil-bearing depositionalsandstones (dark) encased in mud-stone (light) containing alignedmudstone clasts (arrow 1), parallellamination (arrow 2), discordantbases (arrow 3) and concordanttops (arrow 4). Example is fromwell 9. (B) Relatively thin (up-to-20-cm [8-in.]), well-cemented (light)depositional sandstones containingparallel lamination (arrows). Exam-ple is from well 10.

334 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Thin (up to 20 cm [8 in.] thick) carbonate-cemented,

horizontally laminated sandstone units, which are

interpreted as depositional sandstones, are found in

the Balder section of the D field (Figure 4B).

Depositional Sandstones Affected by Sand Remobilization

The A, B, and C fields comprise 10–100-m (33–330-ft)-

thick sections of uncemented, structureless sandstones

that preserve no depositional structures. Their contacts

with the overlying mudstones, where preserved, are

commonly sharp and discordant. Thick sections of struc-

tureless sandstones in the lower Tertiary section of the

North Sea have been controversially attributed to the

deposition of sandy debris flows (Shanmugam et al.,

1995; Hiscott et al., 1997). The paucity of depositional

structures, the characteristics of the upper contacts, and

the association with injected sandstones in the overlying

mudstones support formation by severe postdeposi-

tional sand fluidization and remobilization, as inferred

for similar sandstones in the nearby Eocene Alba field

(Duranti and Hurst, 2004). The large-scale fluidization

and remobilization required for the upward injection

of significant amounts of sand is likely to have oblit-

erated any depositional structures originally present

(Duranti and Hurst, 2004). Moreover, the steep-sided

mounded geometry of these sand bodies is likely to have

been, at least partially, produced by large-scale sand

remobilization (Dixon et al., 1995; Purvis et al., 2002;

Duranti and Hurst, 2004).

Injected Sandstones

Injected sandstones comprise sandstone dikes (discor-

dant to bedding) and sills (concordant to bedding), with

thicknesses ranging from subcentimeter to meter scale.

Numerous sandstone dikes occur in the up-to-100-m

(330-ft)-thick Balder and Frigg intervals of interbedded

sandstone and mudstone that overlie thick, structureless

sandstones (the parent) in the A, B, and C fields, where-

as the Balder interval of the D field contains only a few

injected sandstones. Sandstone dikes display varied de-

grees of postinjection cementation, with uncemented,

oil-bearing sandstone dikes (Figure 5A) occurring in the

vicinity of pervasively carbonate-cemented sandstone

dikes (Figure 5B). Some uncemented sandstone dikes

display ptygmatic folding related to differential compac-

tion (Figure 5A), whereas other carbonate-cemented

sandstone dikes display overlaps, which may have origi-

nated from brittle failure upon differential compaction

(Figure 5B). The distribution of carbonate cement is

highly variable but tends to be more pervasive in thinner

sandstone dikes and is generally concentrated along the

margins of sand bodies (Figure 5C). Sandstone sills may

be difficult to discriminate from thin, depositional sand-

stones. However, the association of numerous, struc-

tureless, sharply bounded, thin concordant sandstones

with sandstone dikes favors an injected origin for most

of these sandstones because injectite complexes tend to

consist of an interconnected network of sandstone dikes

and sills (Parize, 2001; Jonk et al., 2003).

Injection Breccias

Up to 10-m (33-ft)-thick, sand-supported, mud-clast

breccias commonly occur associated with sandstone

dikes and sills in intervals that overlie thick, structure-

less sandstones. Individual units generally have discor-

dant tops and bases, highly variable amounts and sizes

of mud clasts, and highly variable degrees of carbonate

cementation of the sandy matrix (Figure 6). Mud clasts

tend to be concentrated along the margins of sand bodies

(Figure 6), and carbonate cement tends to occur along

the margins of mud clasts (Figure 7). Mud clast size is

varied, ranging from a few millimeters to larger than

the core width (Figure 6). Sometimes, mud clasts form

a jigsaw texture, and clast rotation is sparse. We infer

that most mud clasts are unlikely to have been trans-

ported; rather, a depositional mudstone was pervasively

fractured and invaded by fluidized sand, thus creating

the brecciated appearance.

PETROGRAPHY AND THE DIAGENETIC SEQUENCE

A total of 53 samples were taken for petrographic

studies ( see Datashare Table 2 at AAPG’s Datashare

Web site [Datashare 17], www.aapg.org/datashare/index

.html). These include samples from injected sandstones,

injection breccias, and depositional sandstones.

Detrital Texture and Composition

Injected and depositional sandstones from all locations

have similar detrital composition, suggesting a common

source. They are generally subarkoses, with K-feldspar

dominating over plagioclase (plagioclase never exceeds

more than 20% of the total feldspar). Up to 2% musco-

vite may be present, along with minor glauconite, bio-

tite, tourmaline, and zircon. The sandstones are fine to

medium grained, moderately to well sorted, and grains

are subangular to subrounded. The textural observations

Jonk et al. 335

support the interpretation that the sandstones were

fed from well-sorted sediments at a shallow, wave- or

storm-dominated, sandy shelf delta (Dixon et al., 1995).

Factors Controlling the Present-Day Porosityof the Sandstones

Figure 8 shows the relationship between cementa-

tion and compaction as the factors controlling the po-

rosity evolution of the sandstones analyzed. A large

variation in the porosity evolution of individual sam-

ples is observed; however, one can roughly distinguish

two groups.

Group 1 consists of samples with intergranular vol-

umes between 23 and 43% and cement volumes be-

tween 0 and 26%, where mechanical compaction is

the dominant factor causing porosity reduction but

where a component of cementation occurs as well. This

group includes both depositional and injected sand-

stones and the majority of the injection breccias. Group 2

Figure 5. Core examples ofinjected sandstones. (A) Oil-bearing (dark) ptygmaticallyfolded sandstone dike. Exam-ple is from well 6. (B) Well-cemented sandstone dike,displaying brittle fracturing inresponse to differential com-paction (arrow). Example fromwell 8. (C) Core images (left)and interpretive drawing (right)showing the distribution ofcement in sandstone dikesencased in mudstone (bed-ding in mudstone indicatedwith dashed lines). Examplesfrom well 9.

336 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Figure 6. Core example(top) and interpretivedrawing (bottom) of a6-m (20-ft)-thick, oil-bearing, sand-supportedinjection breccia. The sandis generally poorly cement-ed and oil bearing (givingit a brownish color), al-though some patches arepervasively carbonatecemented (arrows). Mud-stone clasts are angularand display a range ofsizes from a few milli-meters to tens of centi-meters in width. Beddingin mudstone is indicatedwith dashed lines. Exam-ple is from well 9.

Jonk et al. 337

consists of samples with intergranular volumes be-

tween 34 and 55% and cement volumes between 34

and 55%, where cementation is the dominant factor

causing porosity reduction. This group is dominated

by injected sandstones.

Diagenesis

Given the broadly similar diagenetic sequences for de-

positional sandstones, injected sandstones, and injection

breccias, we describe the general diagenetic sequence for

all sandstones together.

The General Diagenetic Sequence for All Sandstones

Ankerite is the earliest diagenetic cement and occurs

in modest amounts (never exceeding 20% of the total

rock volume) in samples from the A and C fields. Do-

lomite only occurs in limited amounts in samples from

well 2 in the A field, and its timing cannot be deduced

with certainty. Nonferroan calcite is volumetrically by

far the most important diagenetic mineral in all fields

and postdates ankerite (Figure 9A). It has a uniform, bright

luminescent character (Figure 9B) and makes up almost

100% of the cement volume and between 34 and 55% of

the total rock volume in the completely cemented sand-

stone belonging to group 2 (Figure 8). Pyrite is associated

with nonferroan calcite and occurs as small nodules (mil-

limeter to centimeter size) in all four fields (Figure 9C, D).

Dull-luminescent ferroan calcite postdates nonferroan

calcite and only occurs in modest amounts in the

present-day water leg of the C and D fields (Figure 9E).

Authigenic quartz, K-feldspar, and kaolinite post-

date all authigenic carbonates and occur only in samples

from group 1. K-feldspar occurs in trace amounts (never

exceeding 2% of the total cement volume) as nonlumi-

nescent overgrowths on detrital K-feldspar (Figure 9F)

and predates authigenic quartz and kaolinite. Quartz

and kaolinite always occur together (Figure 9G) and

appear to be (partly) coeval (Figure 9H). Partly dissolved

K-feldspar grains commonly occur in the vicinity of authi-

genic quartz and kaolinite. Minor occurrences of diage-

netically late, mixed authigenic illite-smectite are pre-

sent in some sandstones from the C and D fields.

Figure 7. Core (left) and inter-pretive drawing (right) showingthat carbonate cement in injec-tion breccias is concentrated inthe sand adjacent to mudstoneclasts. Example is from well 9.

338 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

The Distribution of Authigenic Carbonates

Authigenic dolomite only occurs in modest amounts

in one sample from the A field, and thus, little can be

said about its distribution. Nonferroan calcite, ferroan

calcite, and ankerite occur in numerous samples from

all fields (although nonferroan calcite is, by far, the

most important volumetrically), and some trends are

noted.

Ankerite completely overgrows and/or replaces de-

trital smectitic mudstones and pervades only a few mil-

limeters to centimeters into adjacent sand (Figure 9I).

Because ankerite solely occurs associated with host mud-

stones, it occurs along the margins of injected and depo-

sitional sandstones and along mudstone clasts in injection

breccias. As such, ankerite is more pervasive in injection

breccias, where abundant mudstone clasts are present

(Figure 10).

Calcite postdates ankerite and also nucleates along

the contacts between sandstones and detrital mud-

stone and mudstone clasts (Figure 7). In contrast to an-

kerite, calcite is more widely distributed, and relatively

thick (tens of centimeters), injected and depositional

sandstones are commonly pervasively calcite cemented.

Thicker (meter-scale) sand bodies remain relativelyporous

away from their margins. Given the fact that only a few

injected sandstones reach meter-scale thickness compared

to the source and host depositional sandstones, the latter

provide more promising reservoir rocks at the present day.

Late ferroan-calcite only occurs in modest amounts

in the (present-day) water leg (Figure 10).

Figure 8. The relationship be-tween compaction and cementa-tions as the factors controllingthe porosity evolution of thesamples analyzed (method afterHouseknecht, 1987).

Jonk et al. 339

Figure 9. Micrographs illustrating the diagenetic sequence for both depositional sandstones and injectites. (A) Backscatteredelectron image showing early disc-shaped ankerite (bright) enclosed by later nonferroan calcite (gray). Quartz grains are black.Example is from sample A_4_6. (B) Cold cathodoluminescence micrograph showing nonferroan calcite constitutes a single, bright-orange luminescent phase. Also note the high minus-cement porosities (around 50%). Example is from sample A_3_1. (C) Backscatteredelectron image showing authigenic pyrite (bright) occurring as small concretions along the margin (dashed black line) between aninjected sandstone (left) and a host mudstone (right). Authigenic calcite (light gray) is associated with pyrite and encloses it. Exampleis from sample C_9_3. (D) Micrograph taken in plane-polarized light showing porosity in blue, grains in white, and authigenic pyritein black. Note that authigenic quartz (white arrow) only occurs outside the pyrite concretion, suggesting it postdates authigenic pyrite.Example from sample C_9_3. (E) Micrograph pair (plane polarized on left, corresponding image in cold cathodoluminescence onright) showing bright luminescent nonferroan calcite postdated by dull luminescent ferroan calcite. Also note that the former containsabundant petroleum inclusions (arrows). Example is from sample D_10_6.

340 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Figure 9. Continued. (F ) Micrograph pair (backscattered electron image left, hot cathodoluminescence image right) showingK-feldspar grain (light grain) containing authigenic, nonluminescent K-feldspar overgrowths (arrows 1). Also note pore-fillingauthigenic kaolinite postdating authigenic K-feldspar (arrow 2). Example from sample C_9_2. (G) Micrograph pair (backscatteredelectron image left, hot cathodoluminescence image right) showing nonluminescent quartz overgrowths on quartz grains (arrows 1).Also note pore-filling kaolinite (arrow 2) that always occurs in the vicinity of authigenic quartz. Example is from sample C_9_2. (H)Scanning electron micrograph showing interlocking authigenic kaolinite and quartz, suggesting they are coeval. Example is from sampleC_9_2. (I) Backscattered electron image showing smectitic mudstone clasts almost entirely overgrown with white, disc-shapedauthigenic ankerite. Ankerite also occurs in more modest amounts in the surrounding sand and is postdated by nonferroan calcite(light gray). Quartz grains are black, and K-feldspar grains are dark gray. Example is from sample D_1_3.

Jonk et al. 341

DIAGENETIC CONDITIONSDURING CEMENTATION

Fluid-Inclusion Studies

Fluid-Inclusion Petrography

Fluid-inclusion studies were performed on 13 samples

of depositional sandstones, injected sandstones, and

injection breccias (Table 1). Fluid inclusions were ex-

amined in authigenic ankerite, calcite, Fe-calcite, quartz

overgrowths, and in trails in detrital grains. In most

hosts, coexisting aqueous (type A) and petroleum in-

clusions (type B) occur; however, in Fe-calcite, only

aqueous fluid inclusions are found (Table 1). This is

consistent with the observation that Fe-calcite is found

only in the water leg. Petroleum inclusions are gen-

erally yellow fluorescent (Figure 11A), suggesting a low

to moderate API gravity (Lang and Gelfand, 1985),

although some blue fluorescent petroleum inclusions

(type C), suggesting high API gravity, also occur, par-

ticularly in diagenetically late authigenic quartz (Table 1;

Figure 11B).

Fluid-Inclusion Microthermometry

Fluid inclusions in quartz overgrowths were too small

to allow microthermometric measurements. Primary

fluid inclusions in authigenic carbonate consist pre-

dominantly of monophase liquid inclusions coexisting

with a population of two-phase inclusions with varied

liquid-to-vapor ratios. This suggests partial reequil-

ibration of a population trapped at temperatures be-

low 50jC (Goldstein, 2001). Salinities calculated from

64 final ice-melting temperatures (Bodnar, 1993) of

aqueous fluid inclusions (Table 1; Figure 12) show

the coexistence of a low-salinity (2.3 ± 0.8% NaCl

weight equivalent) aqueous fluid with a moderate-

salinity (9.5 ± 1.4% NaCl weight equivalent) aqueous

fluid.

Figure 10. Cartoon illustrating the distribution of authigenic carbonates in Tertiary injectite reservoirs in the northern North Sea.OWC = oil-water contact. See text for explanation.

342 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Carbon and Oxygen Stable Isotope Studies

A total of 89 carbon and oxygen stable isotope pairs

were collected from pore-filling ankerite, calcite, and

Fe-calcite in injectites and depositional sandstones

(Figure 13A; Datashare Table 3, Datashare 17, at AAPG’s

Datashare Web site, www.aapg.org/datashare/index

.html/)). The data show a large spread with d18O val-

ues ranging from �12.62 to �0.73xand d13C values

ranging from �38.32 to +13.07x. No clear differ-

ences are noted between the carbon and oxygen val-

ues for injected sandstones and depositional sand-

stones (Figure 13A); however, a clear division in d13C

values is noted for the three different types of car-

bonate cement (Figure 13B). d18O values show no

clear differences between subgroups; however, a fre-

quency histogram of d18O values shows a skewed bi-

modal distribution with a dominance of d18O values

of about �10xand a smaller group with d18O values

of about �2x(Figure 13C).

DISCUSSION

Sand injectites formed in a subaqueous environment

provide a fascinating window into shallow crustal pro-

cesses because they are formed by the release of over-

pressured fluids and then continue to act as preferential

conduits for fluid escape throughout burial. The diage-

netic data presented here allow us to constrain both

how and when sand injection occurred.

The Timing of Sand Injection, Petroleum Migration,and Calcite Cementation

The Timing of Sand Injection

A first estimate of the burial depth below the sea bed at

which sand injection occurred comes from the thickness

of strata penetrated by injectites above the parent (res-

ervoir) sandstones, which is about 200 m (660 ft) for all

four investigated fields (Duranti et al., 2002; Jolly and

Lonergan, 2002; Huuse et al., 2004). Thus, if the in-

jectites reached the contemporaneous sea bed (Huuse

et al., 2004), the minimum (postcompaction) burial

depth at which sand injection occurred is 200 m (660 ft).

However, at the present day, this thickness of 200 m

(660 ft) is observed at burial depths between 1400 and

2000 m (4600 and 6600 ft). If we consider the compaction

endured by the host strata (which are predominantly

smectitic mudstones), this thickness of 200 m (660 ft)

between 1400 to about 2000 m (4600 to 6600 ft) would

equate to approximately 400 m (4600 ft) from the sea bed

downward (Velde, 1996) during injection. Thus, from

the present-day thickness of the interval containing

injectites, the minimum burial depth at which sand

injection occurred is about 400 m (4600 ft), which

equates to sand injection occurring at the end of the

Eocene (Figure 14).

Burial Depth of Calcite Cementation

Although calcite cementation obviously postdates sand

injection, the very high minus-cement porosities (most

samples between 45 and 55%; Figure 8) suggest calcite

cementation occurred prior to significant compaction.

Estimations of minus-cement porosities may be slight

overestimations of true precementation porosities be-

cause of displacive growth of carbonate (Tucker, 1981)

and grain dissolution during cementation (Burley and

Kantorowicz, 1986). However, only very minor grain

dissolution at the expense of calcite cementation is ob-

served (at most 2–3% of the rock volume), and we have

no unequivocal evidence for displacive calcite growth;

thus, the minus-cement porosities between 45 and

55% are likely close to true precementation porosi-

ties. Using these values, along with a representative com-

paction curve for sandstones in the northern North Sea

(Figure 14), the depth range over which carbonate ce-

mentation commences is between about 0 and 800 m

(0 and 2600 ft) below the sea bed. Obviously, with a

minimum depth of 400 m (1300 ft) when sand injec-

tion occurred, carbonate cementation must have com-

menced between 400- and 800-m (1300- and 2600-ft)

burial, with the former being more likely than the latter

(because the former [400 m; 1300 ft] represents the

mean of all depth estimates from the minus-cement

porosity range). This implies that sand injection occurred

close to the minimum estimate of burial depth of 400 m

(1300 ft) (and reached the contemporaneous sea bed;

Huuse et al., 2004), and that carbonate cementation

commenced almost immediately (in geological time)

after sand injection occurred (Figure 14).

The Timing of Petroleum Migration

Direct evidence of the timing of petroleum migration

comes from fluid-inclusion studies. Primary petroleum

inclusions are found in ankerite and calcite cement in

pervasively cemented injectites and depositional sand-

stones alike (Table 1). Thus, petroleum migration oc-

curred prior to ankerite and calcite cementation. Given

the short time span between sand injection and carbonate

cementation, it is plausible to suggest that petroleum

was present in the parent (reservoir) sandstones during

Jonk et al. 343

Tab

le1

.Re

sults

ofFl

uid

Incl

usio

nSt

udie

s*

Sam

ple

Hos

t

Con

tent

s

atRo

om

Tem

pera

ture

Fluo

resc

ent

Col

orO

ccur

renc

e

Size

Rang

e

(mm

)M

orph

olog

y

Deg

ree

ofFi

ll

Tm

Rang

e

(Num

ber

of

Mea

sure

men

ts)

A_1

_1de

trita

l

grai

ns

L AN

.A.

S;in

trag

ranu

lar

plan

es

2–

7ov

oid

to

elon

gate

1.00

N.D

.

L Bye

llow

S;in

trag

ranu

lar

plan

es

2–

5ov

oid

to

elon

gate

1.00

N.A

.

Part

lyce

men

ted

inje

ctio

nbr

ecci

a

anke

rite

L AV A

N.A

.P;

isol

ated

2–

6(s

ub-)

nega

tive

crys

tal

shap

e

0.70

–0.

95N

.D.

L AN

.A.

P;is

olat

ed2

–4

nega

tive

crys

tal

shap

e

1.00

N.D

.

L BV B

yello

wP;

isol

ated

4–

12(s

ub-)

nega

tive

crys

tal

shap

e

0.80

–0.

95N

.A.

L Bye

llow

P;is

olat

ed3

–6

nega

tive

crys

tal

shap

e

1.00

N.A

.

calc

iteL A

V AN

.A.

P;is

olat

ed2

–8

irre

gula

r;ne

gativ

e

crys

tal

shap

e

0.80

–0.

90N

.D.

L AN

.A.

P;is

olat

ed1

–3

irre

gula

r;ne

gativ

e

crys

tal

shap

e

1.00

�0.

9to

�6.

5(1

2)

L BV B

yello

wP;

isol

ated

2–

5ir

regu

lar

0.80

–0.

90N

.A.

L Bye

llow

P;is

olat

ed1

–4

irre

gula

r;ne

gativ

e

crys

tal

shap

e

1.00

N.A

.

L CV C

blue

P;is

olat

ed2

–5

irre

gula

r0.

80–

0.95

N.A

.

L Cbl

ueP;

isol

ated

2–

7ir

regu

lar

1.00

N.A

.

L AV C

blue

P;is

olat

ed2

–9

irre

gula

r1.

00N

.A.

A_3

_2de

trita

l

grai

ns

L AV A

N.A

.S;

intr

agra

nula

r

plan

es

3–

12el

onga

te0.

80–

0.95

N.D

L AN

.A.

S;in

trag

ranu

lar

plan

es

3–

10ov

oid

to

elon

gate

1.00

N.D

.

Perv

asiv

ely

cem

ente

d

inje

cted

sand

ston

e

L Bye

llow

-gre

enS;

intr

agra

nula

r

plan

es

3–

7el

onga

te1.

00N

.A.

L AL B

yello

w-g

reen

S;in

trag

ranu

lar

plan

es

2–

4ov

oid

to

elon

gate

1.00

N.A

.

calc

iteL A

V AN

.A.

P;is

olat

ed

and

clus

tere

d

3–

11ir

regu

lar;

nega

tive

crys

tal

shap

e

0.80

–0.

95N

.D.

344 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

L AN

.A.

P;is

olat

ed

and

clus

tere

d

1–

3(s

ub-)

nega

tive

crys

tal

shap

e

1.00

�2.

0to

�7.

8(1

2)

L BV B

yello

wP;

isol

ated

and

clus

tere

d

2–

7ir

regu

lar

0.75

–0.

90N

.A.

L Bye

llow

P;is

olat

ed

and

clus

tere

d

1–

4(s

ub-)

nega

tive

crys

tal

shap

e

1.00

N.A

.

L AL B

yello

wP;

isol

ated

and

clus

tere

d

1–

4ir

regu

lar

1.00

N.A

.

A_4

_5de

trita

l

grai

ns

L AN

.A.

S;in

trag

ranu

lar

plan

es

4–

12El

onga

te1.

00N

.D.

L AV A

N.A

.S;

intr

agra

nula

r

plan

es

4–

9el

onga

te0.

75–

0.95

N.D

.

Part

lyce

men

ted

inje

ctio

nbr

ecci

a

L Bye

llow

S;in

trag

ranu

lar

plan

es

2–

4el

onga

te1.

00N

.A.

L AL B

yello

wS;

intr

agra

nula

r

plan

es

3–

6ov

oid

to

elon

gate

1.00

N.A

.

anke

rite

L BV B

yello

wP;

isol

ated

2–

10(s

ub-)

nega

tive

crys

tal

shap

e

0.80

–0.

90N

.A.

L Bye

llow

P;is

olat

ed2

–5

nega

tive

crys

tal

shap

e

1.00

N.A

.

calc

iteL A

V AN

.A.

P;is

olat

ed3

–8

irre

gula

r0.

85–

0.95

N.D

.

L AN

.A.

P;is

olat

ed1

–3

nega

tive

crys

tal

shap

e

1.00

�0.

4to

�1.

5(4

)

L Bye

llow

P;is

olat

ed2

–4

nega

tive

crys

tal

shap

e

1.00

N.A

.

L Cbl

ueP;

isol

ated

2–

5(s

ub-)

nega

tive

crys

tal

shap

e

1.00

N.A

.

A_4

_7de

trita

l

grai

ns

L AN

.A.

S;in

trag

ranu

lar

plan

es

5–

12ov

oid

1.00

N.D

.

L Bye

llow

S;in

trag

ranu

lar

plan

es

2–

8ov

oid

1.00

N.A

.

Perv

asiv

ely

cem

ente

d

inje

cted

sand

ston

e

L AL B

yello

wS;

intr

agra

nula

r

plan

es

2–

8ov

oid

to

elon

gate

1.00

N.A

.

calc

iteL A

V AN

.A.

P;is

olat

ed

and

clus

tere

d

3–

9ir

regu

lar

0.80

–0.

90N

.D.

L AN

.A.

P;is

olat

edan

dcl

uste

red

1–

4ne

gativ

ecr

ysta

l

shap

e

1.00

N.D

.

Jonk et al. 345

Tab

le1

.C

ontin

ued

Sam

ple

Hos

t

Con

tent

s

atRo

om

Tem

pera

ture

Fluo

resc

ent

Col

orO

ccur

renc

e

Size

Rang

e

(mm

)M

orph

olog

y

Deg

ree

ofFi

ll

Tm

Rang

e

(Num

ber

of

Mea

sure

men

ts)

L Bye

llow

P;is

olat

ed

and

clus

tere

d

2–

10(s

ub-)

nega

tive

crys

tal

shap

e

1.00

N.A

.

B_1

detr

ital

grai

ns

L AN

.A.

S;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.D

.

L Bye

llow

S;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.A

.

Poor

lyce

men

ted

inje

cted

sand

ston

e

quar

tz

over

grow

th

L AN

.A.

PS;

intr

agra

nula

r

plan

es

1–

3ov

oid

1.00

N.D

.

L Cbl

uePS

;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.A

.

B_3

calc

iteL A

N.A

.P;

isol

ated

1–

3ir

regu

lar

to

elon

gate

1.00

N.D

.

L AV A

N.A

.P;

isol

ated

1–

3ir

regu

lar

to

elon

gate

0.85

–0.

95N

.D.

Poor

lyce

men

ted

inje

cted

sand

ston

e

L Bye

llow

P;is

olat

ed1

–3

irre

gula

rto

elon

gate

1.00

N.A

.

L BV B

yello

wP;

isol

ated

1–

3ir

regu

lar

to

elon

gate

0.75

–0.

90N

.A.

B_6

detr

ital

grai

ns

L AN

.A.

S;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.D

.

L Bye

llow

S;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.A

.

Poor

lyce

men

ted

depo

sitio

nal

sand

ston

e

quar

tz

over

grow

th

L AN

.A.

PS;

intr

agra

nula

r

plan

es

1–

3ov

oid

1.00

N.D

.

L Cbl

uePS

;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.A

.

B_7

detr

ital

grai

ns

L AN

.A.

S;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.D

.

L Bye

llow

S;in

trag

ranu

lar

plan

es

1–

3ov

oid

1.00

N.A

.

Poor

lyce

men

ted

depo

sitio

nal

sand

ston

e

calc

iteL A

N.A

.P;

isol

ated

1–

7ir

regu

lar

1.00

N.D

.

346 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

L Bye

llow

P;is

olat

ed1

–7

irre

gula

r1.

00N

.A.

B_8

quar

tz

over

grow

th

L AN

.A.

PS;

intr

agra

nula

r

plan

es

1–

3ir

regu

lar

to

elon

gate

1.00

N.D

.

L Cbl

uePS

;in

trag

ranu

lar

plan

es

1–

3ir

regu

lar

to

elon

gate

1.00

N.A

.

Poor

lyce

men

ted

inje

cted

sand

ston

e

C_8

_1ca

lcite

L AV A

N.A

.P;

isol

ated

3–

6(s

ub-)

nega

tive

crys

tal

shap

e

0.85

–0.

95N

.D.

L AN

.A.

P;is

olat

ed1

–4

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

�1.

0to

�5.

6(6

)

Perv

asiv

ely

cem

ente

d

inje

cted

sand

ston

e

L BV B

oran

ge-y

ello

wP;

isol

ated

3–

12(s

ub-)

nega

tive

crys

tal

shap

e

0.90

–0.

95N

.A.

L Bor

ange

-yel

low

P;is

olat

ed2

–5

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

N.A

.

C_9

_1ca

lcite

L AV A

N.A

.P;

isol

ated

3–

8(s

ub-)

nega

tive

crys

tal

shap

e

0.85

–0.

95N

.D.

L AN

.A.

P;is

olat

ed1

–5

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

�2.

2to

�5.

8(3

)

Perv

asiv

ely

cem

ente

d

inje

cted

sand

ston

e

L BV B

oran

geP;

isol

ated

1–

10(s

ub-)

nega

tive

crys

tal

shap

e

0.90

–0.

95N

.A.

L Bor

ange

P;is

olat

ed1

–3

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

N.A

.

D_1

0_3

calc

iteL A

V AN

.A.

P;is

olat

ed5

–12

(sub

-)ne

gativ

e

crys

tal

shap

e

0.80

–0.

95N

.D.

L AN

.A.

P;is

olat

ed2

–6

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

�0.

3to

9.1

(14)

Part

lyce

men

ted

inje

cted

sand

ston

e

L BV B

yello

wP;

isol

ated

2–

8(s

ub-)

nega

tive

crys

tal

shap

e

0.80

–0.

95N

.A.

quar

tz

over

grow

th

L AN

.A.

PS;

intr

agra

nula

r

plan

es

1–

3ov

oid

1.00

N.D

.

L Bye

llow

PS;

intr

agra

nula

r

plan

es

1–

3ov

oid

1.00

N.A

.

D_1

0_6

calc

iteL A

V AN

.A.

P;is

olat

ed3

–7

(sub

-)ne

gativ

e

crys

tal

shap

e

0.85

–0.

95N

.D.

L AN

.A.

P;is

olat

ed1

–3

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

�0.

4to

8.0

(13)

Jonk et al. 347

the injection process and, thus, migrated through the in-

jectites during their formation. The Upper Jurassic source

rock (Kimmeridge Clay) in this area reached oil maturity

at the end of the Paleocene and oil migrated into the

overlying Cretaceous and Tertiary strata throughout the

Eocene to the present day (Kubala et al., 2003). Given the

fact that sand injection occurred in the late Eocene

(Figure 14), oil was present and continued to migrate in

the early Eocene Age, shallowly buried sand bodies (being

buried to about 400 m [1300 ft] at the end of the Eocene)

when sand injection occurred (Figure 14).

The Relation between Sand Injection, Petroleum Migration,and Carbonate Cementation

The Cause of Sand Injection

A full account of the mechanisms of sand injection is

beyond the scope of this paper. The development of

fluid overpressures in unconsolidated (parent) sand

bodies is generally accepted to occur prior to (hydrau-

lic) fracturing of the sealing (mudstone) and fluidized

transport of grains into these fractures (sand injection)

to dissipate fluid pressures (Jolly and Lonergan, 2002;

Hurst et al., 2004; Huuse et al., 2004).

Because of the close timing between petroleum mi-

gration and sand injection, some workers have suggested

that overpressure may be generated by charge of over-

pressured, migrating petroleum in the unconsolidated

sands (Lonergan et al., 2000; Mazzini et al., 2003). In

particular, the presence of methane gas may cause over-

pressuring in the parent sandstones and breaching of the

seal (Huuse et al., 2004). However, during sand injection,

the Kimmeridgian source was not gas mature (Kubala

et al., 2003), and fluid-inclusion studies of early carbon-

ate cements in depositional and injected sandstones do

not reveal the presence of methane. Furthermore, the

charge of oil into the Tertiary sandstones is thought to be

a buoyancy-driven process instead of a rapid charge of

overpressured petroleum (Kubala et al., 2003). The ad-

dition of petroleum may cause overpressuring in small,

laterally poorly connected sandstone bodies. However,

the parent sandstones exhibit good lateral connectivity at

the present day, which would have been even better at

the shallow burial depths when sand injection occurred,

allowing any pressure buildup at one location to be

evenly transferred throughout the whole body.

Another mechanism by which overpressure may

develop is through rapid burial of sand that is encased

in low-permeability lithologies (mudstones). Pore fluids

may not be expelled quickly enough to allow for normal

sediment compaction (Maltman, 1994; Osborne andTab

le1

.C

ontin

ued

Sam

ple

Hos

t

Con

tent

s

atRo

om

Tem

pera

ture

Fluo

resc

ent

Col

orO

ccur

renc

e

Size

Rang

e

(mm

)M

orph

olog

y

Deg

ree

ofFi

ll

Tm

Rang

e

(Num

ber

of

Mea

sure

men

ts)

Perv

asiv

ely

cem

ente

d

depo

sitio

nal

sand

ston

e

L BV B

yello

wP;

isol

ated

3–

7(s

ub-)

nega

tive

crys

tal

shap

e

0.85

–0.

95N

.A.

L Bye

llow

P;is

olat

ed1

–3

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

N.A

.

Fe-c

alci

teL A

V AN

.A.

P;is

olat

ed3

–9

(sub

-)ne

gativ

e

crys

tal

shap

e

0.85

–0.

95N

.D.

L AN

.A.

P;is

olat

ed1

–4

(sub

-)ne

gativ

e

crys

tal

shap

e

1.00

N.D

.

*Ter

min

olog

yaf

ter

Van

den

Ker

khof

and

Hei

n(2

001)

.L

=liq

uid;

V=

vapo

r;P

=pr

imar

y;PS

=ps

eudo

seco

ndar

y;S

=se

cond

ary;

N.A

.=

not

appl

icab

le;

N.D

.=

not

dete

rmin

ed.

348 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Figure 11. (A) Micrograph pair (plane polarized on left, corresponding image in fluorescent light on right) showing the occurrenceof bright fluorescent primary petroleum inclusions in authigenic calcite (arrows). Example is from sample A_4_7. (B) Micrograph pair(plane polarized on left, corresponding image in fluorescent light on right) showing the occurrence of bright fluorescent petroleuminclusions trapped along the quartz grain-quartz overgrowth boundary (arrows). Example is from sample B_6.

Figure 12. Frequency histogramof fluid-inclusion salinities in authi-genic calcite showing a bimodaldistribution.

Jonk et al. 349

Swarbrick, 1997), and pore pressures may reach values

higher than the hydrostatic gradient. However, the sedi-

mentation rates of the Eocene and overlying strata in

the northern North Sea are much lower (50–100 m/m.y.;

16–330 ft/m.y.; Jones et al., 2003) than the high sedimen-

tation rates associated with this process (>600 m/m.y.;

>2000 ft/m.y.; Mann and Mackenzie, 1990). Further-

more, at the shallow burial depth where sand injection

occurred (about 400 m [1300 ft]), the encasing mud-

stones would not be compacted sufficiently to be an

effective pressure seal (Deming, 1994).

The most commonly quoted cause for sand injection

is earthquake activity (Obermeier, 1996; Jolly and Loner-

gan, 2002). Seismic shaking can cause the development

of short-lived fluid overpressures and subsequent sand

remobilization through cyclic liquefaction (Obermeier,

1996). Sand injection occurred during a period when tec-

tonic activity in relation to the North Atlantic breakup and

Figure 13. (A) Crossplot of d13C –d18Ostable isotope pairs of authigenic car-bonates. (B) Frequency histogram ofd13C values of the three chemically dif-ferent types of authigenic carbonate.Black is calcite, gray is Fe-calcite, andwhite is ankerite. (C) Frequency histo-gram of d18O values of all authigeniccarbonates.

350 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Alpine collision caused basin inversion throughout north-

west Europe (Huuse et al., 2004), and thus, it is conceiv-

able that earthquake activity triggered sand injection.

Although we discard the hypothesis that petro-

leum charge directly caused sand injection, the fact

that buoyant petroleum fluids (partly) occupied the

pore space in the parent sand bodies during earth-

quake activity may have facilitated the creation of the

anomalously large-scale sand injectites observed.

The Origin of Carbonate Cement

Fluids Involved in Carbonate Cementation

Fluid-inclusion studies reveal three fluid types that

were present and trapped in the sandstones during car-

bonate cementation. Low to moderate API liquid petro-

leum inclusions are accompanied by two types of aqueous

fluid inclusions; a low-salinity (2.3 ± 0.8% NaCl weight

equivalent) aqueous fluid and a moderate-salinity (9.5 ±

1.4% NaCl weight equivalent) aqueous fluid (Table 1;

Figure 12). The former is untypically low for a marine

pore fluid (typical salinity of 3.5% NaCl wt. equiv.; Gold-

stein, 2001). Although it is possible that the salinity of

seawater has changed significantly through time (Gold-

stein, 2001), the values observed, along with the positive

skew toward values lower than about 2% in the salinity

distribution (Figure 12), is more likely caused by mixing of

marine water with meteoric water. This interpretation

substantiates independent work (Watson et al., 1995;

Stewart et al., 2000) that invoked a charge of meteoric

water from the nearby East Shetland platform into the

shallowly buried sediments to explain extremely depleted

d18O values of carbonate cements in Tertiary (deposi-

tional) sandstones.

The skewed bimodal distribution of d18O values

(Figure 13C) is unsatisfactorily explained by varia-

tions in precipitation temperature (Figure 15), and

mixing of two fluids with mean d18O compositions of

about �10 and �1xVienna standard mean ocean water

(V-SMOW) is invoked (Figure 15). These values cor-

respond well with known compositions of Tertiary

meteoric water in Scotland (about �12xV-SMOW;

Fallick et al., 1985) and Tertiary marine water (about

�1.2xV-SMOW; Shackleton and Kennett, 1975),

respectively.

The moderate-salinity aqueous fluid is typical of

basinal fluids (Warren and Smalley, 1994) and, to-

gether with the petroleum fluids, is thought to have

migrated from the oil-mature Kimmeridgian source

rock upward into the Tertiary parent sandstones. The

low-salinity aqueous fluid is thought to be the ambient

mixed marine-meteoric pore fluid present at the sea

bed and in the shallowly buried sediments. The injectites

created at shallow burial level form ideal pathways for

downward-invading mixed meteoric-marine fluids and

upward-migrating petroleum and basinal brines, hence,

their presence in early cements. Although at the present

day, the studied reservoirs are located far from exposed

Figure 14. Burial history and timing of sand injection, petroleum migration, and carbonate cementation illustrated in a porosity-depth plot. Following deposition of the parent sandstones, the section has experienced steady, progressive burial throughout theTertiary and Quaternary (Kubala et al., 2003). The sandstone compaction curve used equates to $(z) = $0 � e � cz, where $(z) is theporosity at depth z; $0 is the porosity at depth zero (the seabed; assumed to be 50%; Allen and Allen, 1990); c is a constant(assumed to be 0.27 km�1; Allen and Allen, 1990); and z is the burial depth in kilometers.

Jonk et al. 351

land, during the late Paleogene, the northern North Sea

constituted a narrow, deep-marine basin bordered by up-

lifted, exposed land areas of Norway and the East Shet-

land platform in response to the North Atlantic rifting

(Huuse, 2002; Fyfe et al., 2003; Jones et al., 2003). In

addition, a pronounced eustatic lowering has been noted

at the Eocene–Oligocene transition, coinciding with the

timing of sand injection (Lear et al., 2000; Huuse, 2002).

This combined tectonoeustatic sea level fall would have

exposed most of the East Shetland platform west of the

bounding faults of the south Viking Graben, only kilo-

meters away from the reservoirs studied (Figure 2). It is

thus conceivable that meteoric fluids flushed the shallow-

ly buried injectites and their underlying parent sand bodies

during and shortly after injection.

Bicarbonate and Cation Sources for Carbonate Cement

In the shallow subsurface (less than 1-km [0.6-mi]

burial), two main reservoirs of bicarbonate are present;

marine bicarbonate (with a d13C � 0xV-PDB) and bi-

carbonate produced from CO2 via bacterial degrada-

tion processes of organic matter, where four processes

are distinguished, all of which produce CO2 with char-

acteristic d13C isotopic compositions (Irwin et al., 1977):

2CH2O ! CH4 þ CO2 ðfermentation of organic

matter; d13C of CO2 � þ5 to þ 15 Þx

CH2O þ O2 ! H2O þ CO2 ðoxidation of organic

matter; d13C of CO2 � �30 to � 20 Þx

2CH2O þ SO2�4 ! S2� þ 2H2O þ 2CO2

ðsulfate reduction with organic matter

oxidation; d13C of CO2 � �30 to � 20 Þx

CH4 þ 2O2 ! 2H2O þ CO2 ðoxidation of

methane; d13C of CO2 � �80 to � 40 Þx

Whereas the fermentation and sulfate reduction

processes are anaerobic, the two oxidative reactions are

Figure 15. Possible precipitation conditions for authigenic calcite derived from d18O data. The maximum and minimum precipitationtemperatures (38 and 10jC) are determined using the 400–800-m (1300–2600-ft) depth range estimate of Figure 14, along with a25–35jC/km temperature gradient range (Barnard and Bastow, 1991) and a 0–10jC bottom-water temperature range (Shackletonand Kennett, 1975). The total range of possible d18O pore-fluid compositions is given by the measured d18O range of authigeniccalcite, giving a range of possible d18O pore-fluid between �14 and +2xV-SMOW. However, using the two mean values of thebimodal distribution of d18O values of about �10 and �2x(Figure 13C) shows that the two fields of possible pore-fluidcompositions (fields 1 and 2, respectively) do not overlap, and thus, two different pore fluids are invoked; field 1 with a compositionbetween about �13 and �6xV-SMOW (mean about �10xV-SMOW) and field 2 with a composition between about �4 and+2xV-SMOW (mean about �1x).

352 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

clearly enhanced by the presence of oxygen-rich me-

teoric waters, which, in this study, are interpreted to

have charged the injected and underlying depositional

sandstones during shallow (<800-m; <2600-ft) burial.

Distinct differences in the d13C isotopic signatures are

noted between the three types of carbonate (Figure 13).

Early ankerite is characterized by d13C values between

�0.42 and +12.36 with a mean of +8.41 ± 4.52x.

These values suggest a predominant source of carbon

derived from the bacterial fermentation of organic mat-

ter with a contribution of marine bicarbonate (Irwin

et al., 1977; Macaulay et al., 2000). Liquid petroleum

fluid inclusions in ankerite cement show that liquid

petroleum was present during the precipitation of an-

kerite and, as such, provides the most obvious source of

organic matter (given that the host Paleogene mud-

stones are not enriched in unoxidized organic matter).

Early fermentation of organic matter and the introduc-

tion of meteoric waters in the shallowly buried injected

sandstones and underlying (parent) depositional sand-

stones from the nearby East Shetland platform have been

invoked by others (Watson et al., 1995; Macaulay et al.,

2000; Stewart et al., 2000).

The observation of ankerite overgrowing and oc-

curring in the vicinity of host smectitic mudstone clasts

(Figure 9I) suggests that cations may be partly derived

from these mudstones. Smectite occurs abundantly in

the host mudstones and tuffs and formed through the

alteration of volcanic glass (Watson, 1993), a reaction

that releases cations and, as such, is commonly accom-

panied by carbonate cementation (Andreozzi et al.,

1997; Buck and Bates, 1999). We do note, however,

that sand injectites act as foci for fluid flow during

burial and may transmit considerable pore volumes of

fluid per unit volume of sandstone, and thus, external

sourcing of cations may also be a factor in sand inject-

ites, although unrealistic fluid fluxes are required to

account for the full volume of observed carbonates

(Walderhaug and Bjørkum, 1998).

Exhaustion of suitable hydrocarbons for fermen-

tation leads to a (rapid) change in the way CO2 is

generated (Irwin et al., 1977), and bacterial oxidation

and sulfate reduction take over. Instead of ankerite,

calcite precipitates with d13C values of about �25x,

typical of carbon derived through oxidation and/or

sulfate reduction of the petroleum. Some of the more

depleted d13C values (between about �40 and �30x;

Datashare Table 3, Datashare 17, at AAPG’s Datashare

Web site, www.aapg.org/datashare/index.html) are like-

ly caused by minor carbon derived from the oxidation of

methane produced during the fermentation process. The

skew toward d13C values around 0x(Figure 13B) is

caused by additional sourcing from marine-derived

bicarbonate. The precipitated carbonate is low in Fe2+,

because this component is taken up in associated pyrite

precipitating during sulfate reduction.

Fe-calcite is the last diagenetic carbonate to pre-

cipitate. It occurs in minor amounts in the present-day

water leg only, and d13C values of about 0xsuggest an

ultimately marine bicarbonate source, probably de-

rived directly from the mixed meteoric-marine-basinal

pore fluid.

The Evolution of the Tertiary Reservoirs inthe Northern North Sea

Figure 16 shows schematically the evolution of the

reservoirs studied, incorporating all diagenetic data. Oil

migration from the Kimmeridgian source commenced in

the late Paleocene and entered the shallowly buried Paleo-

cene and Eocene strata throughout the Eocene along the

graben-bounding faults (Kubala et al., 2003). Sand in-

jection occurred near the end of the Eocene, probably in

response to earthquake activity, with the buoyant petro-

leum pore fluids facilitating large-scale injection. Oil

migrated upward through the injectites that probably

reached the contemporaneous sea bed (Huuse et al.,

2004). The injectites form conduits at shallow burial

through which mixed marine-meteoric fluids migrate

downward from the nearby exposed East Shetland plat-

form. In the presence of these aqueous fluids, bacteria

ferment some of the oil, and CO2 released from this

reaction is incorporated in minor early ankerite with

d13C of about +10x. With continued upward migra-

tion of oil and downward migration of marine-meteoric

fluids and exhaustion of hydrocarbons suitable for fer-

mentation, a reversal to oxidation and/or sulfate re-

duction occurs (Irwin et al., 1977). The oil is being

extensively biodegraded, and CO2 released through

these reactions is incorporated in extensively occur-

ring calcite with d13C of about �25x. Coeval pyrite

precipitates through sulfate reduction. Calcite ce-

mentation is locally pervasive and may form barriers

that halt continued petroleum migration through some

of the injectites. Continued burial halts the downward

invasion of low-salinity aqueous fluids, and oil and gas

migrating throughout the Pliocene and Quaternary is

not degraded, hence, the presence of both low- and

high-API oils in the fields (Figure 16) and the presence

of high-API fluid inclusions in late authigenic quartz

(Table 1) in particular. Silicate diagenesis (K-feldspar

Jonk et al. 353

Figu

re1

6.

Car

toon

illus

trat

ing

the

evol

utio

nof

the

Tert

iary

inje

ctite

rese

rvoi

rsin

the

sout

hVi

king

Gra

ben

asde

duce

dfr

omdi

agen

etic

stud

ies.

See

text

for

expl

anat

ion.

354 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

dissolution, quartz cementation, and clay mineral ce-

mentation) takes over at increased burial, with kaolinite

and quartz cementation related to K-feldspar dissolution

(Worden and Morad, 2000) and late onset of illitization

of smectite also being accompanied by further quartz

cementation (Worden and Morad, 2003).

CONCLUSIONS

Being able to constrain the timing of sand injection,

petroleum migration, and pervasive (carbonate) cemen-

tation is a major aspect of understanding injectite petro-

leum plays in the northern North Sea and probably

elsewhere in the world. In this case, we have shown that

petroleum migration occurred prior to sand injection

and continued to migrate following the injection pro-

cess. As such, petroleum has been lost to the contem-

poraneous sea bed, and petroleum trapped at very

shallow (0- to about 500-m; 0- to about 1600-ft)

burial depths was extensively biodegraded by invad-

ing low-salinity (mixed meteoric-marine) fluids. Apart

from diminishing the quality of oil, this process had

another important consequence from a petroleum ex-

ploration point of view: it provided an excellent bicar-

bonate source for pervasive, early carbonate cementation.

In fact, it was previously suggested that in those injectite

reservoirs where oil migration occurred more or less co-

eval with sand injection, about 30% of the pore space in

sandstones was filled with carbonate cement, whereas

those where oil emplacement was later only about 4% of

the pore space was occupied by carbonate cements

(Macaulay et al., 2000). In the fields studied here, this

order of cementation is confirmed in depositional (par-

ent) sandstones, with the percentage of pore space

occupied by carbonate cement in injectites even higher,

as much as 50–60% in some of the fields studied

(Figure 8). From this point of view, it is important to

understand the distribution and magnitude of carbon-

ate cementation in injectite reservoirs where biodegra-

dation of shallowly migrated oil may contribute sub-

stantial amounts of bicarbonate to the aqueous pore fluid

and thus may be the driver of carbonate cementation.

REFERENCES CITED

Ahmadi, Z. M., M. Sawyers, S. Kenyon-Roberts, C. W. Stanworth,K. A. Kugler, J. Kristensen, and E. M. G. Fugelli, 2003, Paleocene,in D. Evans, C. Graham, A. Armour, and P. Bathurst, eds., Themillennium atlas: Petroleum geology of the central and northernNorth Sea: The Geological Society (London), p. 235–259.

Allen, P. A., and J. R. Allen, 1990, Basin analysis— Principles andapplications: Oxford, Blackwell Scientific Publications, 451 p.

Andreozzi, M., E. Dinelli, and F. Tateo, 1997, Geochemical andmineralogical criteria for the identification of ash layers in thestratigraphic framework of a foredeep: The early Miocene Mt.Cervarola sandstones, northern Italy: Chemical Geology,v. 137, p. 23–39.

Barnard, P. C., and M. A. Bastow, 1991, Hydrocarbon generation,migration, alteration, entrapment and mixing in the central andnorthern North Sea, in W. A. England and A. J. Fleet, eds.,Petroleum migration: Geological Society (London) SpecialPublication 59, p. 167–190.

Bergslien, D., 2002, Balder and Jotun— Two sides of the samecoin? A comparison of two Tertiary oil fields in the NorwegianNorth Sea: Petroleum Geoscience, v. 8, p. 349–363.

Bodnar, R. J., 1993, Revised equation and table for determining thefreezing point depression of H2O–NaCl solutions: Geochi-mica et Cosmochimica Acta, v. 57, p. 683–684.

Boehm, A., and J. C. Moore, 2002, Fluidized sandstone intrusions asan indicator of paleostress orientation, Santa Cruz, California:Geofluids, v. 2, p. 147–161.

Bowman, M. B. J., 1998, Cenozoic, in K. W. Glennie, ed., Petroleumgeology of the North Sea: Basic concepts and recent advances,4th ed.: Oxford, Blackwell Science, 278 p.

Buck, E. C., and J. K. Bates, 1999, Microanalysis of colloids andsuspended particles from nuclear waste glass alteration: Ap-plied Geochemistry, v. 14, p. 635–653.

Burley, S. D., and J. D. Kantorowicz, 1986, Thin section and SEMtextural criteria for the recognition of cement-dissolution po-rosity in sandstones: Sedimentology, v. 33, p. 587–604.

Deegan, C. E., and B. J. Scull, 1977, A standard lithostratigraphicnomenclature for the Mesozoic of the central and northernNorth Sea, in K. G. Finstad and R. C. Selley, coordinators,Mesozoic; Northern North Sea Symposium 1997 Proceedings:Norwegian Petroleum Society, 24 p.

Deming, D., 1994, Factors necessary to define a pressure seal:AAPG Bulletin, v. 78, p. 1005 – 1009.

Diller, J. S., 1890, Sandstone dikes: Geological Society of AmericaBulletin, v. 1, p. 411–442.

Dixon, R. J., K. Schofield, R. Anderton, A. D. Reynolds, R. W. S.Alexander, M. C. Williams, and K. G. Davies, 1995, Sandstonediapirism and clastic intrusion in the Tertiary submarine fansof the Bruce-Beryl embayment, Quadrant 9, UKCS, in A. J.Hartley and D. J. Prosser, eds., Characterization of deep ma-rine clastic systems: Geological Society (London) Special Pub-lication 94, p. 77–94.

Duranti, D., and A. Hurst, 2004, Fluidisation and injection in thedeep-water sandstones of the Eocene Alba Formation (UKNorth Sea): Sedimentology, v. 51, p. 503–529.

Duranti, D., A. Hurst, C. Bell, S. Groves, and R. Hansen, 2002,Injected and remobilised sands from the Alba field (Eocene,UKCS): Core and wire-line log characteristics: Petroleum Geo-science, v. 8, p. 99–107.

Fallick, A. E., J. Jocelyn, T. Donnelly, M. Guy, and C. Behan, 1985,Origin of agates in volcanic rocks from Scotland: Nature,v. 313, p. 672–674.

Fyfe, J. A., U. Gregersen, H. Jordt, Y. Rundberg, T. Eidvin, D.Evans, D. Stewart, M. Hovland, and P. Andresen, 2003, Oli-gocene to Holocene, in D. Evans, C. Graham, A. Armour, andP. Bathurst, eds., The millennium atlas: Petroleum geology ofthe central and northern North Sea: The Geological Society(London), p. 279–287.

Goldstein, R. H., 2001, Fluid inclusions in sedimentary anddiagenetic systems: Lithos, v. 55, p. 159–193.

Haaland, H. J., H. Furnes, and O. J. Martinsen, 2000, Palaeogenetuffaceous intervals, Grane field (Block 25/11), Norwegian

Jonk et al. 355

North Sea: Their depositional petrographical, geochemicalcharacter and regional implications: Marine and PetroleumGeology, v. 17, p. 101–118.

Hiscott, R. N., K. T. Pickering, A. H. Bouma, B. M. Hand, B. C.Kneller, G. Postma, and W. Soh, 1997, Basin-floor fans in theNorth Sea: Sequence stratigraphic models vs. sedimentaryfacies: Discussion: AAPG Bulletin, v. 81, p. 662–665.

Houseknecht, D. W., 1987, Assessing the relative importance ofcompaction processes and cementation to reduction ofporosity in sandstones: AAPG Bulletin, v. 71, p. 633–642.

Huang, Q., 1988, Geometry and tectonic significance of Albiansedimentary dykes in the Sisteron area, SE France: Journal ofStructural Geology, v. 10, p. 453–462.

Hurst, A., J. A. Cartwright, M. Huuse, R. Jonk, A. Schwab, D.Duranti, and B. Cronin, 2003, Significance of large-scale sandinjectites as long-term fluid conduits: Evidence from seismicdata: Geofluids, v. 3, p. 263–274.

Huuse, M., 2002, Cenozoic uplift and denudation of southernNorway: Insights from the North Sea Basin, in A. G. Dore, J. A.Cartwright, M. S. Stoker, J. P. Turner, and N. White, eds.,Exhumation of the North Atlantic margin: Timing, mecha-nisms and implications for petroleum exploration: GeologicalSociety (London) Special Publication 196, p. 209–233.

Huuse, M., D. Duranti, N. Steinsland, C. G. Guargena, P. Prat, K.Holm, J. A. Cartwright, and A. Hurst, 2004, Seismic char-acteristics of large-scale sandstone intrusions in the Paleogeneof the south Viking Graben, UK and Norwegian North Sea,in R. J. Davies, J. Cartwright, S. A. Stewart, J. R. Underhill,and M. Lappin, eds., 3D seismic data: Application to the ex-ploration of sedimentary basins: Geological Society (London)Memoir 29, p. 257–271.

Irwin, H., C. Curtis, and M. Coleman, 1977, Isotopic evidence forsource of diagenetic carbonates formed during burial of organicrich sediments: Nature, v. 269, p. 209–213.

Jenssen, A. I., D. Bergslien, M. Rye-Larsen, and R. M. Lindholm,1993, Origin of complex mound geometry of Palaeocenesubmarine-fan sandstone reservoirs, Balder field, Norway, inJ. R. Parker, ed., Petroleum geology of northwest Europe: Pro-ceedings of the 4th Conference, Geological Society (London),p. 135–143.

Jolly, R. J. H., and L. Lonergan, 2002, Mechanisms and controls onthe formation of sand intrusions: Journal of the GeologicalSociety (London), v. 159, p. 605–617.

Jolly, R. J. H., J. W. Cosgrove, and D. N. Dewhurst, 1998, Thick-ness and spatial distributions of clastic dykes, northwest Sacra-mento Valley, California: Journal of Structural Geology, v. 20,p. 1663–1672.

Jones, E., R. Jones, C. Ebdon, D. Ewen, P. Milner, J. Plunkett, G.Hudson, and P. Slater, 2003, Eocene, in D. Evans, C. Graham,A. Armour, and P. Bathurst, eds., The millenium atlas: Pe-troleum geology of the central and northern North Sea: TheGeological Society (London), p. 261–277.

Jonk, R., 2003, The origin and diagenesis of intruded sandstones:Ph.D. thesis, University of Aberdeen, Aberdeen, 2 volumes,628 p.

Jonk, R., D. Duranti, J. Parnell, A. Hurst, and A. E. Fallick, 2003,The structural and diagenetic evolution of injected sandstones:Examples from the Kimmeridgian of NE Scotland: Journal ofthe Geological Society (London), v. 160, p. 881–894.

Joy, A. M., 1992, Right place, wrong time: Anomalous post-riftsubsidence in sedimentary basins around the North AtlanticOcean, in B. C. Storey, T. Alabaster, and R. J. Pankhurst, eds.,Magmatism and the causes of continental break-up: Geolog-ical Society (London) Special Publication, v. 68, p. 387–393.

Kubala, M., M. Bastow, S. Thompson, I. Scotchman, and K. Oygard,2003, Geothermal regime, petroleum generation and migra-

tion, in D. Evans, C. Graham, A. Armour, and P. Bathurst, eds.,The millennium atlas: Petroleum geology of the central andnorthern North Sea: The Geological Society (London), p. 289–315.

Lang, W. H., and J. C. Gelfand, 1985, The evaluation of shallowpotential in a deep field wildcat: Log Analyst, v. 26, p. 13–22.

Lear, C. H., H. Elderfield, and P. A. Wilson, 2000, Cenozoic deep-sea temperatures and global ice volumes from Mg/Ca in benthicforaminiferal calcite: Science, v. 287, p. 269–272.

Leeder, M. R., 1982, Sedimentology: Process and products: London,Unwin Hyman, 344 p.

Lonergan, L., and J. A. Cartwright, 1999, Polygonal faults and theirinfluence on deep-water sandstone reservoir geometries, Albafield, United Kingdom central North Sea: AAPG Bulletin,v. 83, p. 410–432.

Lonergan, L., N. Lee, H. D. Johnson, J. A. Cartwright, and R. J. H.Jolly, 2000, Remobilization and injection in deepwaterdepositional systems: Implications for reservoir architectureand prediction, in P. Weimer, R. M. Slatt, J. Coleman, N. C.Rosen, H. Nelson, A. H., Bouma, M. J. Styzen, and D. T.Lawrence, eds., Deep-water reservoirs of the world: GulfCoast Section SEPM Foundation, 20th annual conference,Houston, p. 515–532.

Macaulay, C. I., A. E. Fallick, R. S. Haszeldine, and G. E. Mcaulay,2000, Oil migration makes the difference: Regional distribu-tion of carbonate cement d13C in northern North Sea Tertiarysandstones: Clay Minerals, v. 35, p. 69–76.

Maltman, A. J., 1994, Chapter 1: Introduction, in A. J. Maltman,ed., The geological deformation of sediments: London,Chapman and Hall, p. 1–35.

Mann, D. M., and A. S. Mackenzie, 1990, Prediction of pore fluidpressures in sedimentary basins: Marine and Petroleum Geol-ogy, v. 7, p. 55–65.

Mazzini, A., D. Duranti, R. Jonk, J. Parnell, B. Cronin, A. Hurst,and M. Quine, 2003, Palaeo-carbonate seep structures abovean oil reservoir, Gryphon field, Tertiary, North Sea: Geoma-rine Letters, v. 23, p. 323–339.

Middleton, G. V., and M. A. Hampton, 1973, Sediment gravityflows: Mechanics of flow and deposition, in G. V. Middleton andA. H. Bouma, eds., Turbidites and deep water sedimentation—Short course notes: Pacific Section SEPM, p. 1–38.

Mudge, D. C., and P. Copestake, 1992, Lower Palaeogene litho-stratigraphy of the northern North Sea: Marine and PetroleumGeology, v. 9, p. 287–301.

Murchison, R. I., 1827, Supplementary remarks on the oolitic seriesin the counties of Sutherland and Ross, and in the Hebrides:Transactions of the Geological Society, v. 2, no. ii, p. 353.

Newsome, J. F., 1903, Clastic dikes: Geological Society of AmericaBulletin, v. 14, p. 227–268.

Obermeier, S., 1996, Use of liquefaction-induced features for pa-leoseismic analysis— An overview of how seismic liquefac-tion features can be distinguished from other features and howtheir regional response distribution and properties of source sed-iment can be used to infer the location and strength of Holo-cene paleo-earthquakes: Engineering Geology, v. 44, p. 1–76.

Osborne, M. J., and R. E. Swarbrick, 1997, Mechanisms for gen-erating overpressure in sedimentary basins: A re-evaluation:AAPG Bulletin, v. 81, p. 1023–1041.

Parize, O., 2001, Sand injections in deep marine shale formations inSE France: Association des Sedimentologistes Francais Publi-cation, v. 35, 120 p.

Purvis, K., J. Kao, K. Flanagan, J. Henderson, and D. Duranti, 2002,Complex reservoir geometries in a deep water clasticsequence, Gryphon field, UKCS: Injection structures, geolog-ical modelling and reservoir simulation: Marine and PetroleumGeology, v. 19, p. 161–179.

356 Origin and Timing of Sand Injection, Petroleum Migration, and Diagenesis, South Viking Graben

Shackleton, N. H., and J. P. Kennett, 1975, Palaeotemperaturehistory of the Cenozoic and the initiation of Antarctic glaciation:Oxygen and carbon isotope analysis in DSDP sites 277, 279 and281: Initial Reports of the Deep Sea Drilling Project, v. 24,p. 743–755.

Shanmugam, G., R. B. Bloch, S. M. Mitchell, G. W. J. Beamish, R. J.Hodgkinson, J. E. Damuth, T. Straume, S. E. Syvertsen, and K. E.Shields, 1995, Basin-floor fans in the North Sea: Sequence strat-igraphic models vs. sedimentary facies: AAPG Bulletin, v. 79,p. 477–512.

Stewart, I. J., 1987, A revised stratigraphic interpretation of the earlyPalaeogene of the central North Sea, in J. Brooks and K. W.Glennie, eds., Petroleum geology of NW Europe: London,Graham and Trotman, p. 557–576.

Stewart, R. N. T., R. S. Haszeldine, A. E. Fallick, M. Wikinson, andC. Macaulay, 2000, Regional distribution of diagenetic carbon-ate cement in Palaeocene deepwater sandstones: North Sea:Clay Minerals, v. 35, p. 119–123.

Tucker, M. E., 1981, Sedimentary petrology; an introduction: NewYork, John Wiley, 252 p.

Van den Kerkhof, A. M., and U. F. Hein, 2001, Fluid inclusion pe-trography: Lithos, v. 55, p. 27–47.

Velde, B., 1996, Compaction trends of clay-rich deep sea sediments:Marine Geology, v. 133, p. 193–201.

Wachter, E. A., and J. M. Hayes, 1985, Exchange of oxygen isotopesin carbon dioxide-phosphoric acid systems: Chemical Geology,v. 52, p. 365–374.

Walderhaug, O., and P. A. Bjørkum, 1998, Calcite cement in shallowmarine sandstones: Growth mechanisms and geometry, in S.Morad, ed., Carbonate cementation in sandstones: InternationalAssociation of Sedimentologists Special Publication 26, p. 179–192.

Warren, E. A., and P. C. E. Smalley, eds., 1994, North Sea for-mation waters atlas: Geological Society (London) Memoir 15,208 p.

Waterston, C. D., 1950, Note on the sandstone injections of EathieHaven, Cromarty: Geological Magazine, v. 87, p. 133–139.

Watson, R. S., 1993, The diagenesis of Tertiary sands from theForth and Balmoral fields, northern North Sea: Ph.D. Thesis,University of Aberdeen, Aberdeen, 315 p.

Watson, R. S., N. H. Trewin, and A. E. Fallick, 1995,The formationof carbonate cements in the Forth and Balmoral fields,northern North Sea: A case for biodegradation, carbonatecementation and oil leakage during early burial, in A. J. Hartleyand D. J. Prosser, eds., Characterization of deep marine clasticsystems: Geological Society (London) Special Publication 94,p. 177–200.

Winslow, M. A., 1983, Clastic dike swarms and the structural evo-lution of the foreland fold and thrust belt of the southernAndes: Geological Society of America Bulletin, v. 94, p. 1073–1080.

Worden, R. H., and S. Morad, 2000, Quartz cementation in oil fieldsandstones: A review of the key controversies, in R. H. Wordenand S. Morad, eds., Quartz cementation in sandstones: Inter-national Association of Sedimentologists Special Publication 29,p. 1–20.

Worden, R. H., and S. Morad, 2003, Clay minerals in sandstones:Controls on formation, distribution and evolution, in R. H.Worden and S. Morad, eds., Clay mineral cements in sand-stones: International Association of Sedimentologists SpecialPublication 34, p. 3–41.

Ziegler, P. A., 1990, Geological atlas of Western and Central Europe:Den Haag, Shell Internationale Petroleum Maatschappij BV,(CD-ROM).

Jonk et al. 357