59 Full

Downloaded from http://sp.lyellcollection.
org/ at IFREMER (ISTEX) on January 4,

2022
Repeated lithosphere extension in the northern Viking Graben:

a coupled or a decoupled rheology?
M. T E R V O O R D E , 1 R. B. F L E R S E T H , 2'3
R. H. G A B R I E L S E N 3 & S. A. P. L. C L O E T I N G H 1
l Institute o f Earth Sciences, De Boelelaan 1085, 1081 H V Amsterdam, the Netherlands

2 Norsk Hydro, Exploration, P.O. Box 200, N-1321 Stabekk, Norway
3 Department o f Geology, University o f Bergen, All@aten 41, N-5007 Bergen, Norway
Abstract: The Late Permian-Early Triassic and Jurassic-Cretaceous rift events that influ-
enced the structure of the northern Viking Graben are examined. Here, a 2D forward
numerical model including faults is applied on seismic line NVGT88-04, which crosses the
basin in the E-W direction. We concentrate on (1) the amount of Jurassic basin subsi-
dence that can be explained by thermal contracting resulting from the Permo-Triassic event,
(2) the spatial connection between the two rift events and (3) the apparent mode of flexure
(coupled v. decoupled) of the lithosphere. Modelling results indicate that post-rift subsi-
dence as a result of Permo-Triassic rifting has practically ceased at the onset of Jurassic
rifting. The position of the rift axis is shown to migrate with time, from the Horda Platform
in Permo-Triassic time to the present Viking Graben centre in Jurassic-Cretaceous time.
The major difference between the consequences of a coupled versus a decoupled mode of
flexure is that the latter causes a larger amount of fault-block rotation. On this basis, we
suggest a decoupled mode of flexure for the Permo-Triassic rift phase, a coupled mode of
flexure for the Early Jurassic rift phase, and a decoupled mode of flexure again for the Late
Jurassic period.
The Viking Graben is part of the series of linked Gabrielsen et al. (1990), and a phase of increased
half-grabens that compose the North Sea sedi- subsidence occurred in late Cretaceous time.
mentary basin (Beach et al. 1987; Badley et al. In spite of almost 30 years of research in the
1988; F~erseth et al. 1995a; Christiansson et al. Viking Graben area, several problems remain to
2000; Odinsen et al. 2000; Fig. 1). This basin was be solved:
formed during several extensional events follow-
ing Caledonian collision. The first event occurred Because of overprint of later structuring, the
in Devonian time (e.g. Hossack 1984; McClay Permo-Triassic rift stage was identified only
et al. 1986; Andersen & Jamtveit 1990; Fossen & with difficulty in seismic sections available for
Rykkelid 1992) and affected an area that extends the early studies of the northern North Sea.
far beyond the later margins of the rift system. Because of this and the limited well control,
This was followed by the pronounced Permo- the relative magnitudes of the Permo-Triassic
Triassic(?) and Jurassic-Cretaceous rift phases. and Jurassic-Cretaceous events in the Viking
As well control is sparse for the pre-Triassic Graben have been disputed. Some workers
sequences, the age of the older of these events is (e.g. White 1990; Lippard & Liu 1992)
still a matter of debate (see Table 1). However, a have argued that Jurassic rifting was domi-
Late Permian-Early Triassic age is suggested by nant. However, White (1990) came to this
the occurrence of Permian sediments in the conclusion by focusing on the East Shetland
southern Viking Graben (Lervik et al. 1989) and terrace area in the west, whereas major
in the Unst Basin (Johns & Andrews 1985; Platt Triassic rifting seems to have occurred in
1995), by palaeomagnetic dating of fault rocks the east, below the Horda Platform. Lippard
in the southwestern part of Norway (Torsvik & Liu (1992) inferred non-uniform thin-
et al. 1992), and by the occurrence of Permian ning (3crust < /~manlle) in the Jurassic Horda
dykes in the southwest Norwegian coastal area Platform, to obtain the best fit between
(F~erseth et al. 1976; F~erseth 1978; Furnes et al. observed and calculated basin subsidence.
1982). Additional minor Late Triassic events Others (e.g. Giltner 1987; Marsden et al.
were recognized by Morton et al. (1987) and 1990) derived from numerical modelling that
From: NOTTVEDT, A. et al. (eds) Dynamics of the Norwegian Margin. Geological Society, London, Special
Publications, 167, 59-81. 1-86239-056-8/00/$15.00 © The Geological Society of London 2000.
Downloaded from http://sp.lyellcollection.org/ at IFREMER (ISTEX) on January 4,
2022
60 M. TER VOORDE E T A L .
=,
=. $TAVANGEI
Il
Il
62
I
I
61°
I
Shetland
l
i
i
Platform
i ................ Platform ~ kin,

i , NSDP84-02,, ~
~ lo
__A3o l o
"-'
Fig. 1. Map of Viking Graben and its surroundings, showing the location of the seismic lines NVGT88-04,
NSDP84-01 and NSDP84-02.
the major part of the extension occurred faults of the first rift episode, and that
during the Permo-Triassic phase, and that greatest stretching and subsidence occurred
the basin subsidence of the second phase is above the Permian rift axis. However, they
for a large part due to thermal contraction also observed occasional basement faults in
resulting from the first event. Giltner (1987) places unaffected by the Permian rift stage.
inferred this from a 1D, uniform extension On the other hand, F~erseth (1996) argued
model, including effects of a finite rift phase, that the thickness distribution of the sediment
whereas Marsden et al. (1990) used a 2D pile suggests that the two observed rifting
depth-dependent extension model, assuming stages may have little spatial connection. The
instantaneous stretching. Although the im- effects of the Permo-Triassic rift phase can be
portance of the Permo-Triassic event is now recognized to the east below the Horda
firmly established, by new, deep-penetration Platform (e.g. Steel & Ryseth 1990; Sneider
regional seismic lines (Platt 1995), improved et al. 1995), whereas no unequivocal evidence
interpretations of deep seismic reflection for the rift is found below the Jurassic rifting
data (Christiansson et al. 2000) and basin axis. Therefore, F~erseth (1996) proposed that
modelling studies (Roberts et al. 1993, 1995: the Permo-Triassic extension maximum may
Odinsen et al. 2000), the amount of excess have been located in the eastern part of the
heat 'inherited' by the Jurassic event is still basin, with the main rift situated beneath the
not well established. present Horda Platform. This is supported by
Another matter of debate is the position observations of other dominantly Triassic
of the Triassic rift axis. It is often stated basins in the eastern North Sea, e.g. the Elger-
(e.g. Lippard & Liu 1992; Yielding et al. sund Basin, the Horn Graben and the Asta
1992) that the Triassic rift axis coincided with Graben (Lervik et al. 1989), which are also
that of the Jurassic-Cretaceous basin. Badley found east of the Jurassic rift axis.
et al. (1984, 1988) suggested that most of the Several numerical basin modelling studies
major faults that were active during the latter have been carried out on the northern Viking
episode were reactivated basement-involved Graben (e.g. Giltner 1987; Marsden et al.
2022
LITHOSPHERE EXTENSION IN THE VIKING GRABEN 61
Table 1. Dating of the major rift events, as proposed or used by various workers
Reference First rift phase Second rift phase
Giltner (1987) 248 213 Ma* 169-98 Ma*

Gabrielsen et al. (1990) ?Late Palaeozoic-Scythian Bathonian-Ryazanian
Marsden et al. (1990) 250 213Ma Bathonian Ryazanian
(modelled as instantaneous events: 230 Ma 170 Ma)
White (1990) Triassic 160-100 Ma
Lippard & Liu (1992) 260-235 Mat or 156-131 Mat
225-200 Mat
Roberts et al. (1995) Early Triassic Mid-Jurassic-Earliest Cretaceous
(modelled as instantaneous events: 250 Ma 170 Ma)
Odinsen et al. (2000) Kungurian-Anisian Callovian-Ryazanian
(261-236 Ma~) (165-141 Ma~)
This study 261-236 Mat 165-141 Marc
* Time scale of Harland et al. (1982).

t Time scale of Haq et al. (1987).
:~Time scale of Harland et al. (1990).
1990; Roberts et al. 1995; Odinsen et al. White 1990) statements about the older rift
2000). In most of these studies an estimate of phase were derived from studies on the amount
the lithosphere rigidity, often expressed as an of extension and subsidence of the latter, Rob-
'effective elastic thickness' (EET), was made erts et al. (1995) focused particularly on the
to calculate the flexural response on the mass pre-Jurassic event, using a combined flexural
redistribution as a result of the extension. The backstripping and forward modelling method.
EET values estimated by different workers Following these workers, we investigate the
vary considerably, from 1.5 km (Roberts et al. Permo-Triassic rift phase as well as the younger
1995) to 4-45 km (Odinsen et al. 2000). This ones. We concentrate on the northern part of the
discrepancy is often attributed to the possi- Viking Graben (60°20'-61°N), using a depth-
bility that the crust is decoupled from the converted version of seismic line NVGT88-04
mantle, because of vertical variations in (Figs 1 and 2). A 2D, non-uniform numerical
the strength of the lithosphere. However, model, including faults and incorporating the
little attention has been paid so far to the effect of a finite period of extension, is used.
influence of this strength distribution on The model allows for both coupled and
the extension mechanism. decoupled behaviour (Ter Voorde et al. 1998).
The purpose of this paper is to examine the As the effect of compaction during burial is
nature of the various episodes of rifting in the not included in the model, we first calculate the
Viking Graben, thereby focusing on the follow- decompacted thicknesses of the sediment pack-
ing questions: ages. These thicknesses are used to constrain
the modelling.
• How much post-rift subsidence was induced
by the excess heat, remnant from the Permo-
Triassic stretching event? And how did this
eventually influence the later development of Geological structure
the basin? Seismic line NVGT88-04 is situated between the
• Is there any spatial connection between the deep seismic reflection lines NSDP84-01 and -2
rift events? Where was the region of max- published previously (Beach et al. 1987; Gibbs
imum Permo-Triassic extension, and did 1987) and recently reprocessed and reinterpreted
this influence the position of the Jurassic- (Christiansson et al. 2000; Odinsen et al. 2000;
Cretaceous basin? Fig. 3). The three lines together provide an
• Is the lithosphere likely to be in a 'deco,upled' outline of the cross-sectional structure of the
state, and what does this mean for the northern Viking Graben. Heavily faulted Meso-
flexural response on extension? zoic synrift sediments are unconformably over-
Whereas in early numerical modelling studies lain by a Cenozoic, almost unfaulted post-rift
in the Viking Graben (e.g. Badley et al. 1988; sequence. Pre-Jurassic sediments can clearly
(a) NVGT-88-04 A
120 100o
i
1500 2000
t
2334 3000 3500 36984401 4402 5000 6000 7059 7060 8000 85~ 90009110 10000 11000 12000 12951
, ,
1000
2000
3000
4000
5000
6000
6996
Fig. 2. (a) Seismic line NVGT88-04. Position of the line is indicated in Fig. 1. Vertical axis: two-way travel time (ms). (b) Depth-converted interpretation. OFZ, Oygarden Fault
Zone: HBS, Hild Brent-Statfjord Fault: Os, Oseberg Fault; Br, Brage Fault; BrE, Brage East Fault: Oy, Oygarden Fault.
2022
(b)° ' ,
Viking Graben Oseberg Brage
~Lo -';'~" |:
(...
i~oL ~ Cretaceous V I ~
F ~ U p p e r Jurassic •
~urassic HE~S ~-"B"
l ~ Permo-Triassic , "~ -
I i I
510 -- 1oo 150 200
Distance (km)
Fig. 2. (continued)
NW SE
East Shetland Basin Gullfaks block Viking Graben Horda Platform tDygarden F.Z.
" .If ~ _ ~.... / - _ ~ / - .t-- . ~ ~ /....~- /
, 25 km I ~\N -, "'" "

w /" \
V:H = 2:1 , , -..,.".~'~
NSDP84-01
W E
East Shetland Platform Viking Graben Horda Platform ~)ygarden Fault Zone
~ . . . ~<::........-.. ..... . ...
• . .... :.......... . . . . . . . . . . . . . . . . . . . . . . ............;:.:.: . . . . . . . r . . . . . . -,.-...--:x+..:.:..:.. ............. r l .....
[ ITertiary
~Cretaceous
_ _ . \ / ~_"~- ...
~Jurassic
. . . . . . .
- . I
._
, /
.
":. f ~
,
'~
=--:,_ ; meeTriassicpostrift
\ "G_,=),. ~--
1Permo-Triassic
v-,=2:, FC-:" " " - " ' -
NSDP84-02
Fig. 3. Depth-converted interpretation of lines NSDP84-01 and NSDP84-02. After Christiansson et al. (2000).
Position of the lines is indicated in Fig. 1.
be observed below the Horda Platform, and intervals (pre-Jurassic, early Jurassic, late Jur-
were recently also recognized further to the west assic and Cretaceous) were extracted, and the
(Platt 1995; Faleide et al. 2000). The Permo- decompacted thicknesses were calculated (Bald-
Triassic event extended over a wider area than win & Butler 1985; Dykstra 1987) (Fig. 4). As
the Jurassic rift (e.g. Gabrielsen et al. 1990; Fair- the Bajocian top of the Brent group is clearly
seth et al. 1995a; Roberts et al. 1995; Odinsen visible on the seismic profiles, we choose this to
et al. 2000). be the boundary between the early Jurassic and
the late Jurassic. For the compaction correction
we used the porosity-depth relation
D e c o m p a c t i o n of the s e d i m e n t s 0 = Ooexp(-cz) (1)
From line NVGT88-04, the compacted thick- where 0 is porosity and z is depth. We choose
nesses of the sediments for four seperated time qS0 = 0 . 5 6 and c = 0 . 3 9 k m -z, as suggested by
2022
64 M. TER VOORDE ET AL.
E
t-
8,,, d ) base Cretaceous - base T e r t i a r y

, 50 ' L A I ~ L
100 150 200
C~J
g
J=
¢,O
c) top Brent - base Cretaceous
' 5 ' I , I , I
0 1O0 150 200
CL
8,o b) top Triassic - top Brent
50
,
100
, ,
150
, ,
200
l
,,_. c~J
E : -'9.-.,. ~. . . . . . . . . -'" ~ "",
1
v , ¢
f-
.,,_,
CL
a) top basement - top Triassic

, ! , I = I , ]
50 100 150 200

Distance (km)
Fig. 4. Sediment thickness as measured along profile NVGT88-04, subdivided into four depth intervals. Dashed
lines, before decompaction: continuous lines, after decompaction.
Sclater & Christie (1980) for an averaged sand- Basin subsidence, subdivided into four
shale lithology in the North Sea. Subsequently, time-intervals
the reconstructed profiles for each time interval
were obtained by correcting for the horizontal hTterval top basement-top Triassic" (Fig. 5a)
fault components at each level (Fig. 5). If the
sedimentation rate was high enough to avoid The most striking features in this profile are the
the basin to become underfilled, the resulting large, wedge-shaped basins at the easternmost
profiles represent the basin subsidence during the part of the profile, beneath the present Horda
given time interval. This was not always the case Platform. These wedges are thinning to the west
during the history of the study area. In Cretac- and show the largest subsidence of the Permo-
eous and early Tertiary time, for example, the Triassic basin system in the study area. Decom-
water depth was probably more than 500 m and pacted sediment thicknesses up to 6.2km are
may even have increased to a maximum of derived. To the west, beneath the present rift
1000 m in the central part of the basin (Nelson & axis, a sediment pile with an approximately even
Lamy 1987; Bertram & Milton 1988). As water thickness of 4.5 km is found. However, because
depths are not included in the numerical model, of the relatively poor quality of the seismic
we regard basin subsidence derived for such time image beneath the Brent Group at this depth,
intervals as underestimated. this estimate should be regarded as uncertain.
2022
!i!i i!!iiil ........ ...... ......

04
E
t-
d) base Cretaceous - base Tertiary
Viking Graben Oseberg Brage Horda Platform

_
'~ii~ii~.-i~ ~ ~ili~=~" ~ ' ~?~i~i~;,,'-~,~..':-~.~! .............................. ~:::~..'.:i~.~'.'-~,'.::i;~,'.%'.~-'ii~;:::~i,.'~'..::!:!::=~:i-'.'-.:::::.::.....
c) top Brent - base Cretaceous

.. ~ t ~ .... I , I , l
50 100 150 200
O_
b) top Triassic - top Brent

1 n ] , I , i
V.G. Oseberg Brage Horda Platform
E
r"
t
O_
50 100 150 200

Distance (krn)
Fig. 5. Decompacted sediment thickness along profile NVGT88-04, subdivided into four intervals and corrected
for subsequent faulting. Faults A, B, C and D are described in the text.
We interpreted four major, westerly dipping depth of 3.7 km in a basin of less than 35 km
faults beneath the Horda Platform, whereas width. The position of the present rift axis also
the fault pattern west of fault 'C' (the Brage corresponds to the site where the fault polarity
East Fault) seems to be dominated by easterly shifted from easterly dipping faults in the west to
dipping faults. This is emphasized by the sedi- westerly dipping faults in the east. This implies
mentary wedges slightly increasing in thickness that this site has moved from the Brage area in
towards the west above the Oseberg structure. Permo-Triassic time to the present rift axis
in Early Jurassic time in this part of the basin.
Interval top Triassic-top Brent (Fig. 5b) However, variations along strike in the relation
between the Permo-Triassic and Jurassic centre
At this time, subsidence was shifted towards the of extension exist, possibly influenced by (deacti-
present rift axis, reaching a maximum basin vated?) transfer faults (Fa~rseth 1996).
2022
66 M. TER VOORDE E T AL.
It should be noted that considerable faulting is location of major Jurassic extension. This is
derived for this period. This implies that the also consistent with a shift of the rift axis to the
early Jurassic extensional episode may be more west after Triassic times, which is reflected in
important than might be expected based on lines the reversal of fault polarity in time west of the
farther to the south, where this sequence is seen Oseberg area.
as a general progressive thickening to the west The fault polarities derived from line
with only very limited fault control. NVGT88-04 are not necessarily representative
for the whole Viking Graben area, as the fault
polarity shifts along strike (Gabrielsen et al.
I n t e r v a l top B r e n t - b a s e C r e t a c e o u s 1990; F~erseth et al. 1995a; F~erseth 1996). How-
(Fig. 5c) ever, a shift of the fault polarity in time is
supported by what we regard as the most likely
Although the position of rifting in this period interpretations of lines further to the south
coincided with the Early Jurassic basin centre, (e.g. NVGT88-02). In this area, east-dipping
activity along east-dipping faults in the west faults within the Viking Graben seem to repre-
became more pronounced, reflecting initiation of sent master faults in the Permo-Triassic period
the Shetland Platform. Fault rotation both east as well as in the early stage of Jurassic rifting,
and west of the rift axis appears to have led to but are in places cross-cut by younger west-
footwall uplift and erosion. The maximum basin dipping faults during the late stage of Jurassic
depth was 5.2km, measured in the strongly rifting (F~erseth 1996).
westerly thinning sediment wedge bounded by
fault 'B'. The shallowest point of this wedge
lies + 17 km west of the deepest point, at a depth Lithosphere rheology
of 2.1 kin.
One of the critical factors in a flexural basin
model is the presumed lithosphere rheology.
hTterval base C r e t a c e o u s - b a s e T e r t i a r y In most models, the lithosphere is assumed to
(Fig. 5d) react on a load in the same way as a thin elastic
plate floating on a viscous fluid. The thickness of
Except for a slight offset along two west-dipping this (imaginary) plate, the so-called 'effective
faults at the Horda Platform, the effect of elastic thickness" (EET) determines the flexural
faulting is hardly noticeable in the Cretaceous response of the lithosphere to loading: a small
sequence. Instead, subsidence in a wider area is EET yields a large-amplitude, short-wavelength
observed. Maximum sediment thickness, along response, whereas a large EET leads to small-
the rift axis, is measured to be 3.8 km. Possibly, amplitude, long-wavelength flexure. The EET
thermal st~bsidence and sediment loading, com- can thus be estimated by considering it as a
bined with a general rise in sea level, resulted in factor of the 'response function' describing the
gradual burial of Jurassic fault blocks during lithosphere deformation caused by loading. This
this period. approach has been followed for the Viking
Although we acknowledge the fact that the top Graben by various workers, resulting in supris-
of the basement is difficult to constrain from line ingly large variations in estimates of the EET.
NVGT88-04, and is therefore not undisputed, a Odinsen et al. (2000) derived an EET determined
thick Triassic sediment pile was also suggested by the 4 5 0 C isotherm, which in their model
by Christiansson et al. (2000), who reinterpreted corresponds to a depth of c.45km, using a
the deep structure of the northern Viking Graben kinematic numerical model developed by Kooi
from deep reflection seismic lines and gravimetric ( 1991 ). On the other hand, Kusznir et al. (1991),
and magnetic data (Fig. 3). Bearing this in mind, Roberts et al. (1995) and Ter Voorde et al.
some preliminary conclusions can be drawn. The (1997), using flexural models for footwall uplift,
Permo-Triassic rift phase affected a wider area arrive at EET values between 1.5 and 6 km in the
than did the subsequent Jurassic-Cretaceous same area. We propose that this discrepancy
event. This is consistent with results of F~erseth might be associated with the assumption of a
(1996). Also, Odinsen et al. (2000), using the "coupled' versus a "decoupled' lithosphere rheol-
interpretation of Christiansson et al. (2000), ogy, the latter of which is characterized by a very
derived a regional stretching distribution in the weak and ductile lower crust (i.e. Kusznir et al.
Permo-Triassic compared with the more loca- 1991; Ter Voorde et al. 1998).
lized thinning in Jurassic time. The area of The continental plate may contain dipping or
strongest subsidence associated with the Permo- sub-horizontal weak ductile zones, which cannot
Triassic event is positioned to the east of the support significant bending stresses. It is argued
2022
E
e-
tm
0-17 S-1
~I~ 1 0 -16 s -1
U 1 0 -t5 s -1
1 10 "14 S "1
I 1 0 "t3 S -t
1000 MPa
Strength J
Fig. 6. Strength profiles for a typical continental lithosphere in an extensional regime, assuming a 20 km thick
upper crust, a 15 km thick lower crust, and a total lithosphere thickness of 125 km, for various lithosphere
rheologies. (a) and (b) assume a 'dry' rheology, the upper crust consisting of dry quartzite and the lower crust of
diabase, and (c) assumes a 'wet' rheology, the upper crust consisting of wet quartzite and the lower crust of
diorite. A zero pore pressure was assumed in (a), and a hydrostatic pore pressure in (b) and (e).
(e.g. Le Pichon & Chamot-Rooke 1991; Burov crustal flow is generally not high enough to
& Diament 1995) that the appearance of such compensate for the entire upper-crustal defor-
zones in the lower crust permits mechanical mation (Buck 1988; Ter Voorde et al. 1998).
decoupling of the upper crust from the mantle. Therefore, part of the compensation probably
A redistribution of loads in the upper crust, mostly occurs by deep, very low viscosity mantle
for example as a result of extensional faulting, material causing the lithosphere to be in a
would then be compensated for by elastic defor- 'partly decoupled' mode (see Ter Voorde et al.
mation in the upper crust exclusively, whereas 1998). However, to make the effects of the
the lower crust would adjust by ductile flow. assumption of coupled v. decoupled behaviour
This can explain the very low (1.5-6km) effec- on the modelling results as clear as possible,
tive elastic thicknesses obtained in modelling we will consider only the end-members of the
studies where the basin topography was used as decoupled-coupled spectrum in this study.
the main constraining factor.
Figure 6 shows strength profiles for the
tensional regime in a typical continental crust The numerical model
before extension. The profiles are constructed
from the commonly used, laboratory-derived We use a forward, 2D difference model, describ-
strength equations (e.g. Goetze & Evans 1979; ing the lithospheric deformation resulting from
Brace & Kohlstedt 1980; Carter & Tsenn 1987; extension, and including the effect of faulting in
see Appendix 1). The depth-strength curves the upper crust. A finite syn-rift phase is con-
demonstrate clearly the effect of the strain rate sidered (Waltham 1989, 1990). This is of major
on the crustal configuration. If strain rates are importance when calculating the post-rift ther-
higher than 10 -13 s -1 the shallowest weak ductile mal subsidence, as stressed, for example, by
layer occurs at the base of the lower crust, or Ter Voorde & Cloetingh (1996). A presentation
perhaps even deeper. For lower strain rates, an of the model is given in Fig. 7. Following
additional ductile zone appears at the base of the Kusznir et al. (1991), the model is divided into
upper crust. an upper layer of faulting with vertical shear in
A fully decoupled mode of flexure is likely to the deforming hanging wall, and a lower layer of
be an exceptional case, because the rate of lower more distributed deformation. In both layers the
2022
][iiiiii~ ...... [l|tl!li'~ ' 'iiiiii[
~r~illlll[:lll:ll~l Ir '' ;ll 1 [[ [I lJ I l :1 II [Iii ::i 1 i' ' ;; llt]ll I:,L~,I''" ; ;:::=l]]::',',ll] ..................... LIIlll
~ detachmentlevel ~+++++++t+~++++;++++~+++~++~++!t+++~-+++~ ......................... ~
~!!!!!!!!!!!!!!!!!!!!!!!!!!! ...... iiiii(iiii::i, ---
ilfli
Fig. 7. Schematic representation of the model. Extension is achieved by distributed deformation (i.e. deformation
along faults) in the upper crust, and by distributed thinning in the lower lithosphere.
condition of volume conservation is satisfied. deformation of the crust, temperature changes

Extension rates can be varied per time interval and the mass of deposited sediments, relative to
and per fault. The faults are assumed to flatten that of the undeformed crust:
into the detachment that forms the boundary
between the two layers. The depth of this q(x) J Ap(x, :)[1 - o~T(x, :)] g d: (4)
detachment is an important factor, as it has the
same function in the model as a "necking depth"
(e.g. Braun & Beaumont 1989, Weissel & Karner where c~ is thermal expansion coefficient (K-l).
1989, Kooi et al. 1992). The necking depth is For the coupled mode of flexure, we assume
defined as the level that remains horizontal in the the deflection w to be constant in each vertical
absence of isostatic forces, and is a decisive column through the whole lithosphere, signifying
factor for the existence and amount of footwall that the crust and mantle are forged together by
uplift (e.g. Ter Voorde & Cloetingh 1996). the lower crust. Hence, the integration interval
Temperatures are calculated from the heat for calculating the vertical load q comprises the
transfer equation: whole lithosphere. Pcompl in equation (3) is
the asthenosphere density, and Pcomp2 the sedi-
OT F ment density. For the decoupled mode of flexure,
0--7 = V(~VT) - v V T + p-~ (2) the deflection w is assumed to be different for the
crust and mantle. For the upper crust, q is
where T is temperature (:C)+ t is time (s), tc is calculated by integrating from the surface to the
thermal diffusivity (runs-l), v is velocity, F is lower crust, p.... v~ is taken as the lower-crustal
heat production (Wm-3), p is density (kgm -3) density, Pcomp2 as the sediment density, and the
and C is specific heat (J :C -1 kg-1). rigidity is assumed to be low ( E E T < 6km).
The equation is solved using a finite difference This is different from earlier studies using low
method on a rectangular grid. Thermal proper- EET values (e.g. Kusznir et al. 1991; Roberts
ties on each grid point are obtained by interpo- et al. 1995), where the influence of decoupling
lation from the second, moving grid representing on the compensation materials is not taken
the extending basin. Heat transfer during as well into account.
as after the deformation is calculated. The thermal features of the model have been
Flexure is calculated from the thin-plate described in more detail by Ter Voorde &
approximation (e.g. Bodine et al. 1981): Bertotti (1994), the structural features by Ter
Voorde & Cloetingh (1996), and the effects of
decoupling have been discussed by Ter Voorde
Ox 2 D ( x ) Ox2 j + (Pcompl - P.... p2)gw(x)
e t a / . (1998).
= q(x) (3)
where w is deflection (m), D is rigidity (N m), Model parameters

Pcompl is density of the material underneath the
plate (kgm-3), Pcomp2 is density of the material The model parameters we used in this study are
above the plate (kg m-3), g is gravitational accel- summarized in Table 2. The Moho depth is
eration (m s -2) and q is vertical load (N m-2). assumed to be 35km, based on the crustal
The vertical load is calculated by vertical thickness observed below the platform areas.
integration of the density contrasts caused by The detachment depth is assumed to be 18 kin,
2022
Table 2. Parameters used for the modelling parameters and deviations given in Table 2 and
the same fault configuration as in Fig. 5.
Parameter Value Deviation Because, in our model, a change in necking level
used for implies a change in fault shapes, effects of vary-
sensitivity test
ing this level on the resulting flexural response
EET (coupled) 450°C isotherm see Fig. 8 are difficult to test independently. However, it is
EET (decoupled) 1.5-6 km see Fig. 8 possible to calculate which change in sediment
Density sediments 1900kg m -3 -t-100 kg m -3 density would be equivalent to a change of 1 km
Density upper crust 2700kgm -3 +50 kgm -3 in the level of necking (see Appendix 2), and we
Density lower crust 2950kgm -3 ±50 kgm 3 used this analogy to estimate the model sensi-
Density 3300 kg m -3 - tivity for the level of necking.
asthenosphere
The deviations in densities and Moho depth
Moho depth 35 km +2 km
Detachment depth 18 km +5 km cause maximum variations in the resulting basin
Young's modulus 7 × 10l° Nm - subsidence of 60m for the coupled case and
Poisson ratio 0.25 - 185m for the decoupled case. A reasonable
variation of 5km in necking depth can be
compared with a variation in sediment density
o f - 148 kgm -3 (in the case of a deeper necking
based on observations of Fossen et al. (1998). level) or 262 kgm -3 (in the case of a shallower
They described the flattening of normal faults at necking level) for the coupled case (Appendix 2,
a maximum depth between 18 and 20 kin, in the equation (6)), leading to a maximum variation in
Tampen Spur area on the western flank of the basin subsidence of 100m. For the decoupled
Viking Graben. case the level of necking has no influence
Although Fossen et al. (1999) suggested that (Appendix 2, equation (6)).
these flattened faults are inherited mechanical In Fig. 8 the effects of changes in EET values
weak zones in the basement, related to Devonian are shown for the simulation of the Permo-
extensional structures or Caledonian thrusts, the Triassic time interval (i.e. Fig. 5a). The figure
detachments could also be linked to ductile shear shows clearly that the choice of the EET is of
zones in the middle and lower crust and thus be limited influence for the coupled scenario, but
explained by temperature- and pressure-con- might be important in the case of decoupling. In
trolled changes in the lithosphere rheology the modelling, we used an EET defined by the
(e.g. Fossen & Gabrielsen 1996). The different 450°C isotherm for coupled flexure, and the
detachments may have different causes, and may EET giving the best fit for decoupled flexure.
also have been initiated at different stages during Finally, effects of reasonable errors in the
rifting. Then, truly low-angle structures might estimated fault shapes and deformation mechan-
also develop by flattening of originally steeper ism may be of the order of 1 km (White et al.
fault structures, during repeated phases of 1986; White & Yielding 1991; Withjack &
stretching, and perhaps act as detachments in a Peterson 1993).
later stage (Fossen et al. 2000). Although the In previous studies on the Viking Graben
flattening of the faults with depth may have (Marsden et al. 1990; Roberts et al. 1995; Odin-
no unequivocal explanation, it may indicate sen et al. 2000), modelling results and (seismic)
the existence and position of ductile zones in the data normally fit within 1000 m. Based on this,
lower crust. In the case of the Viking Graben, on the estimated sensitivity of the model, and on
the maximum observed depth of flattening uncertainties in the seismic data, for example as
should then also be interpreted as the depth of a result of possible errors in chosen velocity
mechanical decoupling. models, we regard a fit of the model with the
data within 1000m as a good approximation.
Model sensitivity and constraints

Modelling results
The overall state of flexure depends on the
strength of the layer(s), the density and amount The numerical model was used to simulate the
of sediments, the depth of the Moho, and the sequential basin configurations shown in Fig. 5,
depth of necking during lithospheric extension. in order to relate the amount of basin subsidence
To obtain an impression of the model sensitivity to the amount of extension along various faults,
to these variables for the Viking Graben con- and to the strength and mode of flexure of
figuration, we tested the effects of varying the the lithosphere. In the model, sediments are
EET, the densities and the Moho depth, using assumed to fill the basin up to the surface. This
2022
i [ r :
c) .,..~. EET = 1 5 km
F '.~ .C-'-~, . " . - , - EET = 6 km~ Lv".~, i~
e / s ~ . •
t..-
ia) V ,
s'o , Oo ~so
c)
' ' ' EET~= 450oC_i'sotherm ' r~
- EET = 200°C-isotherm ,
Ib)
50 100 150
Distance (km)
Fig. 8. Modelling results for interval top basement-top Triassic. assuming different values for the EET.
(a) Decoupled rheology: (b) coupled theology.
is a realistic approach, as long as the basin is not major extension occurred on the Horda Plat-
underfilled at the end of the modelled episode. form, provided that the top Triassic interpreted
The modelling parameters we used are summar- in the graben centre is correct.
ized in Table 2. Two scenarios were modelled for this time
interval, one assuming a coupled rheology with
an EET determined by the 450:C isotherm
I n t e r v a l top b a s e m e n t - t o p Triassic ( F~7. 9) (corresponding to an initial depth of 43 km in
the model), the other assuming a decoupled
Triassic sediments are supposed to have been rheology with an EET of 1.5 km. A reasonable
deposited in lacustrine environments, dominated fit between the modelled basin configuration and
by alluvial fans along the graben margins and the one interpreted from the seismic line could
with finer fluvial or lacustrine sediments in the be obtained from both assumptions (Fig. 9b),
"lows" (e.g. Steel & Ryseth 1990). At the end of but with different values for the amount of
Triassic time the northern Viking Graben was extension. In the case of a coupled configura-
approximately at sea level. The sediment thick- tion, the lithosphere is too strong to show a
ness shown in Fig. 5a might thus be interpreted significant flexural response on the extension,
as the total basin subsidence during this time and the basement subsidence is almost entirely
interval. We focused especially on the flanks of due to crustal movements along the faults. How-
the graben, because of the poor seismic resolu- ever, if we adopt a decoupled mode of extension,
tion in the basin centre. Consequently, the much the model shows a syn-rift uplift of the area,
better fit of the model beneath the platform leading to a small amount of subaerial footwall
areas than beneath the rift axis (Fig. 9a) reflects erosion, followed by thermal subsidence (Fig. 9b
the higher confidence of the observed sediment and c).
thickness in that area. The modelled uplift and subsidence patterns
To obtain the best modelling result, a fault for the decoupled scenario agree with the
configuration was required with a change in findings of Roberts et al. (1995), and those of
fault polarity on the flanks of the Viking Graben the coupled scenario with findings of Odinsen
and of the Brage area (Fig. 9a). The thinning of et al. (2000). Roberts et al. (1995) modelled the
the mapped sequence in the Oseberg area pre-Jurassic rifting stage in the Horda Platform
towards the east is then explained by extension region with a flexural cantilever model, using an
along the east-dipping fault 'A', bounding the EET of 1.5 km, and found that a stage of foot-
Viking Graben in the west, whereas the Viking wall uplift and erosion preceded a stage of
Graben centre subsided further along the west- overall thermal subsidence. On the other hand,
dipping fault 'B', 25 km farther to the east. The the results of Odinsen et al. (2000), who used the
2022
0 [ "ESP 'VikingG. OsebergBrage " HordaPlatform ]
50 100 150
' i , i ' I |
,~J
f, , I
50
.,..,c- I
100
I
150
'jj |
l ~ I I ' I
o _ t~
v
c-
50 100 150
Distance (km)
Fig. 9. Modelling results for interval top basement top Triassic. (a) Modelled fault configuration; dashed line
indicates basin depth as derived from profile NVGT88-04. (b) Long dashes, modelling result for coupled
rheology; short dashes, modelling result for decoupled rheology; continuous line, basin depth as derived from
profile NVGT88-04. (e) Modelled basin depth for decoupled theology, at (from top to bottom) 0, 30, 74 and
Ma after the end of rifting.
coupled approach, do not indicate such uplift. Roberts et al. (1995) reported a /3 value of
According to Roberts et al. (1995), the seismic 1.34 in the Horda Platform area, along the same
expression of the better imaged Triassic fault section as in the present study, whereas Odinsen
blocks beneath the Horda Platform gives the et al. (2000) calculated an average ~ factor of
impression of bevelling at the fault-block crests. 1.27 for their northern transect, and 1.19 for
This can be noticed in interpretations of Beach their southern transect. They derived fl Factors
et al. (1987, fig. 2), Roberts et al. (1993, fig. 6a) for the Horda Platform of 1.33 and 1.39, but
and Odinsen et al. (2000, figs 2 and 5), and they did not indicate exactly over which width
supports the decoupled scenario as the most this was measured.
likely one. Figure 9c shows the syn-rift uplift and thermal
The total amounts of extension used in our subsidence with time for the decoupled scenario.
model are 23 km for the coupled scenario and Indicated are the top of the basement at the end
34.5km for the decoupled scenario, signifying of rifting, 30 Ma later (i.e. latest Triassic), 74 Ma
an average/3 value of 1.18 for the coupled case, later (i.e. when the second major rift phase is
and 1.29 for the decoupled case (measured over assumed to start) and after total thermal relaxa-
the entire profile). Major extension in Permo- tion. In contrast to earlier proposals (e.g. Giltner
Triassic time took place on the Horda Platform, 1987; Marsden et al. 1990; Roberts et al. 1995),
the total modelled amount of extension being 14 but in accordance with Odinsen et al. (2000), the
or 22% west of the Brage fault, and 22 or 38% present model suggests that thermal subsidence
east of it, for the coupled or decoupled case, as a result of Permo-Triassic rifting cannot
respectively. have had much influence on the Jurassic event.
2022
(It should be noted that thermal subsidence for coinciding with the site of major subsidence as
the coupled case would have even less effect.) well as the position of shift in fault polarity
Researchers who stated that an inherited ther- along the profile (Fig. 10a). However, the subsi-
mal anomaly influenced Jurassic rifting used 1D dence in the basin centre might be overestimated,
thermal calculations (e.g. Giltner 1987), instan- because, again, the seismic data at this depth in
taneous Triassic rifting (Marsden et al. 1990; the basin centre do not allow for a decisive
Roberts et al. 1995), and/or derived their solution. Compared with the earlier rift phase, a
statements from studies of the Jurassic event change in fault dip was imposed between the
alone (Giltner 1987). Oseberg and Brage area (fault 'C'). Figure 10
shows the modelling results for the coupled and
the decoupled case. For the decoupled case, an
I n t e r v a l top T r i a s s i c - t o p B r e n t G r o u p EET of 6kin was used. The coupled scenario
(Fig. 10) gives results that are more consistent with the
observed basin configuration, as hardly any
The top of the Brent Group horizon is dia- tilting of the fault blocks is observed. For this
chronous on a regional scale, and is situated in reason, thermal subsidence related to this phase
the Bathonian (Middle Jurassic) sequence in the is assumed to be negligible. To model the
northern Viking Graben. The lower Jurassic observed amount of subsidence, modelled offsets
sequence consists predominantly of marine along the faults had to be set up to 2 km larger
shales, whereas the Middle Jurassic sequence was than the observed offsets, suggesting an over-
deposited in non-marine to paralic environments estimation of the amount of extension, and the
(Fjellanger et al. 1996; Ravn~,s et al. 1999). The existence of an additional source of basin subsi-
Tarbert Formation of the uppermost Brent dence, different from fault-related extension.
Group consists of shallow marine sandstones. As shown by the modelling results of the top
As an approximation, the profile shown in basement-top Triassic interval, only a small part
Fig. 5b might thus be regarded to represent the of the additional subsidence (i.e. less then 350 m)
total basin subsidence during early Jurassic time, can be explained by remnant thermal subsidence
until mid-Bathonian time. caused by Permo-Triassic rifting, which was
This time interval could be simulated with a not included in the modelling results. Another
much less complicated fault configuration than explanation for basin subsidence might be the
that of the Permo-Triassic phase. The position existence o f a "proto-rift stage' (Gabrielsen 1986;
of the rift axis was in the Viking Graben centre, Nottvedt et al. 1995). Nettvedt et al. (1995)
ESP VikingG . Oseberg Brage HordaP l a t f o r m 1

0 L_ . . . . T,,- --="'""":~"';;;==":;:;;~;=;= ............ ~'".'"" ,~1 l ~
,,,iiiiii[i[iiiiiiiii[iiiiii[[i[iiiiiii[i[[iii[~r" I[
~!!!!!!!!!!!!![!!!!!!!!!!!!!!!!!!!!!!!~ilb., ..`::~iiiiiii~iii::!iiiiiiiiiiii::::i::::~i::i~ii~Zi![!~[~[~[[~[![![!!!!!![!i[![~[~iiiii.~-~
° i i ii!! ifii ii ii i i
50 100 150
/ ' ' ' /
, i , i , i
50 1 O0 150
Distance (km)
Fig. 10. Modelling results for interval top Triassic top Brent. (a) Modelled fault configuration: dashed line
indicates basin depth as derived from profile NVGT88-04. (b) Long dashes, modelling result for coupled
profile NVGT88-04.
2022
proposed an idealized, three-stage model for rift magnitude of this event cannot be derived from
evolution, involving a proto-rift stage, a syn-rift the modelling results, because of the large uncer-
stage and a post-rift stage. The proto-rift stage is tainty in the basin depth.
characterized by thermally induced domal uplift, This early Jurassic extension, combined with
or by deposition in a wide, slowly subsiding the additional subsidence, caused the deposi-
basin with only minor fault activity. The syn-rift tional environment to convert from continental
stage describes the phase of active stretching and to marine, implying that the creation of new
block rotation, and the post-rift stage consists of accommodation space outpaced sediment sup-
asymptotically decreasing subsidence, caused by ply. Subsequently, as sedimentation continued,
thermal contraction. Interference between post- the depositional environment passed to non-
rift subsidence related to the Permo-Triassic rift marine again, and in latest Bajocian-earliest
phase and proto-rift subsidence related to the Bathonian time the first rotational movements of
Jurassic rift phase (Steel 1993; Nmtvedt et al. the second major rift phase commenced (F~er-
1995) might form an alternative source for the seth et al. 1995b; Ravnfis & Bondevik 1999).
extra subsidence observed.
The total amount of extension in the modelled
scenario (Fig. 10a) is 7.3km, yielding an aver- Interval top Brent Group-base Cretaceous
age/3 value of 1.05. The modelled /3 reaches a (Fig. ]1)
value of 1.16 in the basin centre. Apparently,
a minor extensional event has occurred in The Upper Jurassic sediments in the Viking
earliest Jurassic time, as argued earlier by Roe & Graben consist mostly of marine shales. The
Steel (1985) and Gabrielsen et al. (1990). The shales of the Bathonian-Oxfordian Heather
o [___ " ESP V ~ _ a g e H0~a Platform"]
,°~iiiiiiiiiiiiiiiiiiiiiiill ....
, I , J , iiiiiiiiiiii l iiiiiiiii iiilli

50 100 150
ol _ ___.__
~o.~ observed
' 5~0 ' 100

~ , 150
I t I
o~ m
E
e-
D
:)
t 5 i L __~L-~
0 100 150
Distance (km)
Fig. 11. Modelling results for interval top Brent-base Cretaceous. (a) Modelled fault configuration; dashed line
indicates basin depth as derived from profile NVGT88-04. (h) Long dashes, modelling result for coupled
profile NVGT88-04. (e) Modelled basin depth for decoupled rheology, at (from top to bottom) 0 and vc Ma after
the end of rifting.
2022
formation are overlain by the organic-rich Figure l lc displays the basin configuration
Draupne Formation (equivalent to the Kimmer- immediately after stretching as well as after
idge Clay Formation), and both are interfin- thermal subsidence. From this figure, it is evident
gered by sandstone units (e.g. Ravn~.s et al. that the thermal anomaly caused by the exten-
1999). Based on environmental interpretations, sion alone cannot be responsible for the sub-
the water depth at the end of this episode sequent Cretaceous subsidence. This can partly
(Ryazanian) is estimated to be 200m on the be explained by the fact that the basin was
margins to 300m in the graben centre (Badley underfilled at the end of this rift phase, which was
et al. 1988), although this might be a very con- not included in the modelling. The modelled
servative estimate (e.g. Marsden et al. 1990; amount of extension, and thus the thermal
Yielding et al. 1992). anomaly, is therefore probably an underestima-
The modelling results are shown in Fig. 11. tion. During Cretaceous time, much of the sedi-
The EET used for the decoupled case was 6 km. mentation was accommodated by the infilling of
Broadly the same fault configuration could be previous rift bathymetry, causing additional
used as for the early Jurassic phase, but with a subsidence because of sediment loading. This
few more active faults. The basin now became effect is further exaggerated by the compaction
asymmetric, deepening to the east in this part of of earlier deposited sediments (Yielding et al.
the graben. However, this varies along strike 1992), which is not taken into account in the
(see, e.g. Odinsen et al. 2000). The footwall uplift present model.
in the west is best explained by using the
decoupled approach, as shown in Fig. l lb.
This approach yields an average .3 factor over Modelled Moho configuration
the whole transect of only 1.07 (and a maximum
of 1.2). Together with the early Jurassic phase, To compare the modelled Moho depth with the
this makes a Jurassic z3 of 1.12. This is slightly observations, we superimposed the Moho defor-
lower than the values found for Jurassic stretch- mation of the three time intevals on each other,
ing by Odinsen et al. (2000), who obtained 1.15 the result of which is shown in Fig. 12. The
for the northern transect and 1.19 for the "observed Moho" in Fig. 12 is derived by inter-
southern transect, or by Roberts et al. (1993), polation between the Moho reflectors observed
who reported a ;3 of 1.3 in the basin centre, in seismic lines NSDP84-01 and -2 (e.g. Chris-
of 1.15 in the East Shetland Basin and west- tiansson et al. 2000; Odinsen et al. 2000),
ern Horda Platform and of 1.05 in the eastern although we acknowledge that the Moho is not
Horda Platform. That the 3 value we derived unequivocally imaged east of the Viking Graben
is relatively low can be related to the fact that (see Fig. 3). Starting from a constant initial value
our model assumes a zero water depth, which of 35 kin, the modelled Moho depth is too large
is an underestimation for this time interval. compared with the depth derived from the
' I
50 t O0 150 200
Distance (km)
Fig. 12. Moho uplift as a consequence of rifting. Dashed line, modelled Moho depth resulting from three stages
of rifting; continuous line, Moho depth along line NVGT88-04, as derived from interpolation between
interpretations of line NSDP84-01 and NSDP84-02.
2022
observations. The maximum difference, below thinning, imposed on the thickened crust of
the Viking Graben axis, is 6.5 km. This is in the Caledonian Orogeny (Andersen & Jamtveit
agreement with findings of Odinsen et al. (2000), 1990; Fossen & Rykkelid 1992).
who suggested that the Permian pre-rift Moho
depth might have varied from + 3 5 k m on the
flanks to less than 30 km in the basin areas. To Mode of flexure in the Viking Graben
obtain the best fit for the subsidence data,
Lippard & Liu (1992) needed to assume an End-members: the coupled versus the
initial crustal thickness varying from 28 km in decoupled mode
the centre to 30-32km on the flanks of the
Graben. This pre-Permian variation in Moho An important result of this modelling study is
depth is likely to be caused by Devonian crustal that it is hard to discriminate between the
~ o~
"-'£E~ e brittle
100 200 300 400 500 600
g.-
o dry, 1=0.4
100 200 300 400 500 600
~o
r-
~o
100 200 300 400 500 600

Moho temperature (C)
Fig. 13. Mode diagrams for the lithosphere, indicating the possibility of a decoupled rheology in Moho
depth-Moho temperature space (Spadini & Podladchikov 1996), for three different lithosphere theologies.
Boundaries between the ductile and brittle lower crust are given for strain rates of (from left to right) 10-]7, 10-16,
10-15, 10-14 and 10-13 s 1. Black dots indicate the assumed pre-rift (to) and post-rift (tl) Moho depth (at the
position of major Moho uplift) in the Viking Graben, as derived from Fig. 11. The light grey arrows indicate
the change in the state of the lithosphere in cases of instantaneous rifting and very slow rifting.
2022
coupled and the decoupled mode of flexure on which brittle deformation gives way to ductile
the basis of the fit between modelling results and deformation (Spadini & Podladchikov 1996), i.e.
seismic data alone (e.g. Fig. 9). This might be an
O-b k O'd
explanation for the large variations between
EET values used for previous modelling studies where orb is the brittle strength and Crd is the
of the Viking Graben. Nevertheless, we showed ductile strength needed to cause deformation at
that the presence or absence of footwall uplift is a given strain rate (see Appendix 1). The param-
a good indicator for the mode of flexure: the eters used are given in Table 3. The Moho
suspected presence of eroded fault blocks in the depths and temperatures on the left side are
Permo-Triassic phase and of footwall uplift in associated with an entire brittle crust, whereas
the late Jurassic phase cannot be explained with the conditions on the right side of the curves
the coupled approach (even though we used a yield a ductile lower crust. If we assume the bold
rather deep necking level), whereas the absence line to be the pre-rift geotherm in the northern
of footwall uplift in early Jurassic time is impos- Viking Graben, and the black dot at to to
sible to reconstruct by decoupled flexure. indicate the state of the pre-rift Moho, we can
Although caution should be taken when inter- conclude from the figure that the lower crust is
preting the results (we modelled only the end- at least partly ductile, and that the first condition
members of the coupled-decoupled spectrum, for a decoupled mode of flexure is fulfilled.
whereas the "partly decoupled" mode is most However, the uplift of mantle material combined
likely to occur), the best results were obtained by with syn-rift cooling will eventually result in a
assuming a decoupled mode of flexural response shift from the decoupled to the coupled state of
in the Permo-Triassic phase, a coupled mode of the lithosphere (Spadini & Podladchikov 1996),
flexure in early Jurassic time, and a change back which agrees with our modelling results for the
to decoupled flexure in late Jurassic time. The Permo-Triassic and the Early Jurassic phases.
choice of the mode of flexure is important, as The change back to the decoupled state in
it can have a strong influence on the derived Bathonian time cannot be explained in this way,
amount of extension. For the Permo-Triassic or be related to the change in extension (and
phase, for example, we derived a 3 factor of thus strain) rate. In fact, at lower strain rates,
1.18 for the coupled case and of 1.29 for the decoupled behaviour is more likely to occur than
decoupled case. at high strain rates (e.g. Brun & Tron 1993;
A prerequisite for decoupled behaviour is the Ter Voorde et al. 1998; see Figs 6 and 13). How-
existence of a ductile lower crust. Figure 13 ever, a thermal event could explain the weaken-
shows which Moho depth and temperature ing of the lithosphere and a shift to decoupled
will cause the manifestation of a ductile lower behaviour.
crust according to the theological laws. The A Mid-Jurassic pre-rift thermal dome in the
curves were derived by calculating the depth at so-called 'triple junction' of the North Sea rift is
Table 3. Parameters used for construction of strength pro[iles
Upper crust Lower crust Mantle

olivine
Quartzite(w) Quartzite(d) Diorite (wet) Diabase (dry)
Ep (kJ mol -l ) 172.6 134 212 276 510

Ap(Pa-"s -I) i.26x 10-13 6.03x 10 -24 1.26 x 10 16 3.16 x 10 -~° 7 × 10 -14
n 1.9 2.72 2.4 3.05 3.0
Ed (kJ mol -I) - - - 535
Ad (S-1 ) - 5.7 × 10 I1
O'D (GPa) - - - - 8.5
R (J (tool K) -1) 8.314J (mol K)

Thermal diffusivity sediments 0.75 × 10-6m2 s -I
Thermal diffusivity crust 1 . 0 × 1 0 -6m z s - l
Thickness heat producing layer 15km
Heat production 2.3 x 10-6 W m -3
Material constants for quartzite, diorite, diabase and olivine are adopted from Tsenn &
Carter (1987).
2022
well established (e.g. Ziegler & Van Hoorn 1989; possibly preceded by the 'proto-rift' stage as
Hendrie et al. 1993; Underhill & Partington defined by Nottvedt et al. (1995).
1993), evidenced by basin-wide subaerial erosion The position of the rift axis is shown to change
(the 'Mid-Cimmerian' unconformity), the extru- with time. The Permo-Triassic rift axis was
sion of the Rattray volcanic rocks, and flooding positioned beneath the present Horda Platform,
events on basin margins. This thermal dome whereas the Jurassic rift axis was positioned
is interpreted to be the effect of a short-lived, beneath the present Viking Graben centre. This
anomalously hot asthenospheric pulse (Hendrie result might be dependent on the position along
et al. 1993) or transient mantle plume (Under- strike (F~erseth 1996). For example, Odinsen
hill & Partington 1993). However, although the et al. (2000) found for line NDSP84-I that the
timing of this event is correct to explain the shift maximum/3 factors for both the Permo-Triassic
to decoupled behaviour in Bathonian time, the and the Jurassic phases were in the present
location might be too far away (i.e. ~250 km to Viking Graben axis, whereas for the more
the south) from the northern Viking Graben southern line NSDP84-02 the maximum/3 fac-
to have such a pronounced effect. tor in the Permo-Triassic phase was on the
Horda Platform.
To our knowledge, this basin modelling study
is the first one that attempts to discriminate
The partly decoupled mode of flexure between the coupled and the decoupled mode
of flexure. The major difference between these
As mentioned above, in this study we only two modes is the amount of fault-block rota-
look at the end-members of the spectrum of tion, which is larger when a decoupled rheology
decoupled-coupled modes of flexure. Never- is assumed. The coupled and decoupled mode of
theless, the most probable mechanism to occur flexure are sometimes hard to discriminate on
is the partly decoupled mode of flexure. In this the basis of basin geometries alone (e.g. Fig. 9),
case, the viscosity of the lower crust is too which may explain the large variation in EET
high to allow for an isostatic compensation values found in the literature. Nevertheless, the
level in the lower crust. Therefore, as a first modelling results suggest a decoupled mode of
approximation, the lithosphere will react in a flexure in the Permo-Triassic phase, a coupled
coupled mode, and only short-wavelength lat- mode of flexure in early Jurassic time, and a
eral variation in the load is compensated by change back to decoupled flexure in Late Juras-
lower-crustal flow. sic time. Subsequent research on this topic
In general, modelled surface deflections should focus on further constraints on the
obtained by assuming a partly decoupled litho- mode of flexure, such as the lower-crustal thick-
sphere differ from the fully coupled mode in that ness and material, and the pre-rift geotherm.
an extra short-wavelength component can be
observed. In contrast, the difference from the We would like to thank R. Ravnfis for bringing us
fully decoupled case is that the long-wavelength together, and T. Odinsen for discussions about the
uplift component disappears (Ter Voorde et al. development of the Viking Graben. Reviews by
1998). Translating this to Figs 9-11 raises the R. Boutilier and D. Waltham were greatly appreciated.
suggestion of partly decoupled flexure. This research was supported by the IBS (Integrated
Basin Studies) project, part of the Joule II research
programme funded by the Commission of European
Communities (Contract JOU2-CT92-0110). This paper
is Publication 981001 of the Netherlands Research
Conclusions School of Sedimentary Geology.
The evolution of the Viking Graben has been

studied using a numerical model, constrained by
seismic line NVGT88-04. Although the impor- Appendix 1
tance of the rift phase in Permo-Triassic time
has been affirmed by the modelling results, the The brittle yield strength is given by Byerlee's
post-rift subsidence is shown to have practically law (Byerlee 1978; Brace & Kohlstedt 1980):
ceased at the onset of Jurassic rifting. Hence, O'b = o@g(1 -- A)z
no significant amount of basin subsidence in
the Jurassic phase can be ascribed to thermal where p is the density, g is the gravitational
contraction resulting from the Permo-Triassic acceleration, z is the depth, o~ = 0.75, A : 0 for
event. The Jurassic subsidence should therefore zero pore pressure, and A : 0.4 for hydrostatic
be explained entirely by a new extensional event, pore pressure. Ductile deformation is assumed
2022
to occur by power law creep (Kirby 1983): ~ ...., ,...,..
":':'2:'2'2":" " ""
exp i~i~i:i~!~!iii! iiP uil

O" d : Ap (
::::::: iiiiiiii~i~
where ~ is strain rate, A, n and Ep are empirically N~
......
ii~Pu~i
:::::
.....~
:i:i:i:i:i:i:
, i,~:i,~:i
iiiiiiii:i:iiii P ui - - - ~i;~;~ii
derived material constants, R is the gas con-
stant and T is the absolute temperature. For iiiiiiiiiiiii!i iiiiiiiiii !~!~!~!~!~!~!
olivine with a strength exceeding 200 MPa, duc- ~,. . . . . . . . . . . . . _ . . .
tile deformation is described by Dorn creep --i,

iiiii}ii?i!}!
(Goetze & Evans 1979: Tsenn & Carter 1987):
.............. M-N 2 ...
.............. ; :.::::: ~il~ I::;
ad : °D { 1 -- I2EaT In (~-da) ] .'2 } M-N, ....... ii~iii[il iiiiiiii~i~il
Parameters are given in Table 3.

!iiiiiiiiiiiiii ;iiiiiiiiiiiii
~!!!~!i!i!!!i!i Pm
Appendix 2
The depth of necking is the level of zero vertical
Fig. A2. Thinning of the crust around necking levels
motion in the absence of isostatic forces. This .\'] or X2. assuming the depression of the surface is
level determines the ratio between thinning of the equal (i.e..~ is different) for both cases.
upper crust, where crustal material is replaced by
sediments, and thinning of the lower lithosphere.
where crustal material is replaced by mantle
Reference load Pi caused by thinning around
material. Therefore, the depth of necking is a
necking level Nl, using sediment density P~l (see
decisive factor for the new load distribution
Fig. AI):
caused by crustal extension. For a given amount
of thinning, the change in load A P caused by a
variation in necking depth A N is equal to the
change in load A p caused by a variation in
sediment density Ap~. Ap~ can be calculated
as follows.
where g is gravitational acceleration (m s-2), 3 is

a stretching factor. Pu is density of the upper
.......
....,.......,
........,....
crust (kgm-3), p~ is density of the lower crust
?i:::iii?i! (kgm 3), Pm is density of the mantle (kgm -3)
. . . . . . . .
N2 !iP.uil
• ..........
and M is the Moho depth (kin).
. . . . . . . ............
,.,.-.,...,.,., .............
A P caused by thinning around N2 instead of
!i!iiii!
• .. • ......
. . . . . . .
.., ,.,.w
. . . . . .
..... ......
around Nl:
:i:i!i!i!i!i!~
iii~ii~i!iii!i! AP = g(N2 - Ni) (Psi - Pu - Pm + Pl)
iiiiii!iii
:.:+:<.:.:
,v.-.-.,.....
,.,.....,..,.,
":':':':':
. . . . . - - .
(A.2)
ii~H!!!!!!!!i
-;-c::;::-::
A P caused by using sediment density P2
:-:+:-:
M instead of p~:
iiiiiiiiii i!!!!!i!!!
.:.:.:.:.:
M-N 1 ......
.......
..........
i!i!i!i!i!i!i!i
i!iiiiiiililili
.............
q . ............. . . . . .
iiiiiiiiiiiii
This yields the following equivalence between
A N ( = Nt - N2) and A p ( = p ] - P2):
ili~i~!iiiilili
-~.~-~.~-~.~-
iiiiiiiiiii' . . . . . . .
.,.-.... .....
m
Aps = (P sl - pu - pm + pl ) AN (A.4)
Fig. A1. Thinning of the crust around necking levels
N 1 or N2, assuming/3 is equal (i.e. the depression of the If we assume the depression of the surface as a
surface is different) for both cases. result of crustal thinning to be constant (as it can
2022
L I T H O S P H E R E E X T E N S I O N IN T H E V I K I N G G R A B E N 79
be observed), a change in necking depth implies BUROV, E. & DIAMENT, M. 1995. The effective elastic
a change in the/3 factor (see Fig. A2). Then thickness (Te) of continental lithosphere: what
does it really mean? Journal of Geophysical
(f12 - 1 Research, 100, 3905-3927.
(A.5) BYERLEE, J. D. 1978. Friction of rocks. Pure and
Applied Geophysics, 116, 615 626.
and CARTER, N. L. & TSENN, M. C. 1987. Flow properties
of the continental lithosphere. Tectonophysics,
136, 27 63.
Aps = k, N~N 2 AN (A.6) CHRISTIANSSON, P., FALEIDE, I. J. • BERGE, A. M.
2000. Crustal structure in the northern North Sea:
w h i c h is the e q u a t i o n to be used for the case o f an integrated geophysical study. This volume.
DYRSTRA, 1987. Compaction correction for burial
the Viking G r a b e n .
history curves: application to Lopatin's method
for source rock maturation determination. Geo-
byte, Nov. 87, 16-23.
References FALEIDE, J. I. et al. 2000. Deep basin geometries in the
northern North Sea; constraints from gravity and
ANDERSEN, T. B. & JAMTVEIT, B. 1990. Uplift of deep magnetic data.
crust during orogenic extensional collapse: a FJELLANGER, E., OLSEN, T. R. & RUBINO, J. L. 1996.
model based on field studies in the Sogn- Sequence stratigraphy and palaeogeography of
Sunnfjord region of western Norway. Tectonics, the Middle Jurassic Brent and Vestland deltaic
9, 1097-1112. systems, Northern North Sea. Norsk Geologisk
BADLEY, M. E., EGEBERG, T. & NIPEN, O. 1984. Tidsskrift, 76, 75-106.
Development of rift basins illustrated by the FOSSEN, H. & GABRIELSEN, R. H. 1996. Experimental
structural evolution of the Oseberg structure, modelling of extensional fault systems by use of
Block 30/6, offshore Norway. Journal of the plaster. Journal of Structural Geology, 18, 673-687.
Geological Society, London, 141,639-649. -- & RYKKELID, E. 1992. Post-collisional extension
--, PRICE, J. D., RAMBECH DAHL, C. & AGDE- of the Caledonide orogen in Scandinavia: struc-
STEIN, T. 1988. The structural evolution of the tural expressions and tectonic significance. Geol-
northern Viking Graben and its bearing upon ogy, 20, 737-740.
extensional modes of graben formation, Journal --, ODINSEN, T., FA~RSETH, R. B. & GABRIEL-
of the Geological SocieO', London, 145, 455-472. SEN, R. H. 2000. Detachments and low-angle
BALDWIN, B. & BUTLER, C. O, 1985. Compaction faults in the northern North Sea rift system.
curves. AAPG Bulletin, 69, 622-626. This volume.
BEACH, A., BIRD, T. & GIBBS, A. D. 1987. Extensional FURNES, H., MITCHELL,J. G., ROBINS, B., RYAN, P. D.
tectonics and crustal structure: deep seismic & SKJERLTE, F. 1982. Petrography and geochem-
reflection data from the northern North Sea istry of peralkaline, ultrapotassic syenite dykes of
Viking Graben. In: COWARD, M. P., DEWEY, J. Middle Permian age, Sunnfjord, West Norway.
F. & HANCOCK, P. L. (eds) ContinentalExtensional Norsk Geologisk Tidsskrift, 62, 147-159.
Tectonics. Geological Society, London, Special FA~RSETH, R. B. 1978. Mantle-derived lherzolite xeno-
Publications, 28, 467-476. liths and megcrysts from Permo-Triassic dykes,
BERTRAM, G. T. & MILTON, N. J. 1988. Reconstruct- Sun-nhordland, Western Norway. Lithos, 11,
ing basin evolution from sedimentary thickness: 23-35.
the importance of paleobathymetric control, with - - 1 9 9 6 . Interaction of Permo-Triassic and Jurassic
reference to the North Sea. Basin Research, 1, extensional fault-blocks during the development
247-259. of the northern North Sea. Journal of the
BODINE, J. H., STECKLER,M. S. & WATTS, A. B. 1981. Geological Society, London, 153, 931 944.
Observations of flexure and rheology of the oceanic --, GABRIELSEN, R. H. & HURICH, C. A. 1995a.
lithosphere. Journal of Geophysical Research, 86, Influence of basement in structuring of the North
3695-3707. Sea basin, offshore southwest Norway. Norsk
BRACE, W. F. & KOHLSTEDT, D. L. 1980. Limits Geologisk Tidsskrift, 75, 105-119.
on lithospheric stress imposed by laboratory ex- --, MCINTYRE, R. M. & NATERSTAD, J. 1976.
periments. Journal of Geophysical Research, 85, Mesozoic alkaline dykes in the Sunnhordland
6248 6252. region, western Norway: ages, geochemistry and
BRAUN, J. & BEAUMONT, C. 1989. A physical regional significance. Lithos, 9, 332 345.
explanation of the relation between flank uplifts --, SJOBLOM, T. S., STEEL, R. J., LILJEDAHL, T.,
and the break-unconformity at rifted continental SAUAR, B. E. & TJELLAND, T. 1995b. Tectonic
margins. Geology, 17, 760-764. controls on Bathonian-Volgian syn-rift succes-
BRUN, J. P & TRON, V. 1993. Development of the sions on the Visund Fault Block, northern North
North Viking Graben: inferences from laboratory Sea. In: STEEL, R. J., FELT, V., JOHANNESSEN, E.
modelling. Sedimentary Geology, 86, 31 51. 8¢ MATHIEU, C. (eds) Sequence Stratigraphy on the
BUCK, W. R. 1988. Flexural rotation of normal faults. Northwest European Margin. Norwegian Petro-
Tectonics, 7, 959-973. leum Society, Special Publication, 5, 325 346.
2022
80 M. T E R V O O R D E E T AL.
GABRIELSEN, R. H. 1986. Structural elements in Viking Graben, North Sea. In: ROBERTS, A. M.,
graben systems and their influence on hydrocar- YIELDING, G. & FREEMAN, B. (eds) The Geometo'
bon trap types. In: SPENCER. A. M. et al. (eds) of Normal Faults. Geological Society, London,
Habitat of Hydrocarbons on the Norwegian Con- Special Publications. 56, 41-60.
tinental Shelf Graham and Trotman, London. LE PICHON, X. & CHAMOT-ROOKE, N. 1991. Exten-
55-60. sion of continental crust, hi: MULLER, D. W.,
F,~RSETH, R. B., STEEL. R. J., IDIL, S. & MCKENZIE. J. A. & WEISSERT, H. (eds) Con-
KLOVJAN, O. S. 1990. Architectural styles of basin troversies in Modern Geolog)v Evolution of Geolo-
fill in the northern Viking Graben. ht: BLUNDELL. gical Theories hi Sedimentology, Earth History and
D. J. & GIBBS, A. D. (eds) Tectonic Evoh~- Tectonics, Academic. London. 313-338.
tion of the North Sea R(fts. Clarendon, Oxford. LERVIK. K. S., SPENCER. A. M. & WARRINGTON, G.
158-179. 1989. Outline of Triassic stratigraphy and structure
GIBBS, A. 1987. Deep seismic profiles in the Northern in the central and northen North Sea. In: COLLIN-
North Sea. h;: BROOKS, J. & GLENNIE. K. W. SON, J. (ed.) Correlation in Hydrocarbon Explora-
(eds) Petroleum Geology of North-H'est Europe. tion. Graham and Trotman, London, 173-189.
Graham and Trotman, London, 1925-1928. LIPPARD. S. & Dr:, G. 1992. Tectonic modelling of the
GILTNER, J. P, 1987. Applications of extensional northern North Sea using program RIV'r. In:
models to the Northern Viking Graben. Norsk LARSEN, R. M., BREKKE, H., LARSEN, B. T.
Geologisk Tidsskr(ft, 67, 339-352. & TALLER-XAS. E. (eds) Structural and Tectonic
GOETZE, C. & EVANS, B. 1979, Stress and temperature Modelling and its Application to Petroleum Geol-
in the bending lithosphere as constrained by ogy. Norwegian Petroleum Society, Special Pub-
experimental rock mechanics. Geophysical Journal lication, 1.43-54.
of the Royal Astronomical Society. 59, 463-478. MARSDEN, G.. YIELDING, G., ROBERTS, A. M. &
HAQ, B. U., HARDENBOL, J. & VAIL. P. R. 1987. KLSZNIR, N. J. 1990. Application of a flexural
Chronology of fluctuating sea levels since the cantilever simple-shear pure-shear model of con-
Triassic (250 million years ago to present). tinental extension on the formation of the north-
Science, 235, I 156- i 166. ern North Sea basin. In: BLUNDELL, D. J. &
HARLAND, W. B., ARMSTRONG, R. L.. Cox. A. V., GIBBS, A. D. (eds) Tectonic Evolution of the North
CRAIG, L. E., SMITH. A. G. & SMITH. D. G. 1990. Sea R(fts. Clarendon. Oxford, 236-257.
A Geologic Time Scah,. Cambridge University MCCLAY, K. R.. NORTON, M. G., CONEY, P. & DAVIS,
Press, Cambridge. G. H. 1986. Collapse of the Caledonian orogen
--, Cox, A. V.. LLEWELLYN, P. B.. PICKTON, C. A.. and the Old Red Sandstone. Nature, 323, 147-149.
SMITH, A. G. & WALTERS, R. 1982. A Geological MORTON, N., SMITH, R. M., GOLDEN, M. & JAMES,
Time Scale. Cambridge University Press, Cam- A. V. 1987. Comparative stratigraphic study of the
bridge. Triassic-Jurassic sedimentation and basin evolu-
HENDRIE, D. B., KUSZNIR, N. J. & HUNTER, R. H. tion in the northern North Sea and northwest of
1993. Jurassic extension estimates for the North the British Isles. In: BROOKS, J. & GLENNIE, K. W.
Sea "triple junction" from flexural backstripping: (eds) Petroleum Geology of North-West Europe.
implications for decompression melting models. Graham and Trotman, London, 697-710.
Earth and Planetarv Science Letters, 116, 113 127. NELSON, P. H. H. & LAMY, J. M. 1987. The More/
HOSSACK, J. R. 1984. The geometry of listric growth West Shetlands area: a review, hi: BROOKS, J.
faults in the Devonian basins of Sunnfjord, West & GLENNIE, K. W. (eds) Petroleum Geology
Norway, Journal q/" the Geological Society. Lon- of North-West Europe. Graham and Trotman,
don, 141,629-638. London. 775-784.
JOHNS, C. R. & ANDREWS, I. J. 1985. The petroleum NOTTVEDT, A., GABRIELSEN, R. H. & STEEL, R. J.
geology of the Unst Basin, North Sea. Marine and 1995. Tectonostratigraphy and sedimentary archi-
Petroleum Geology, 2, 361-372. tecture of rift basins, with reference to the north-
KIRBY, S. H. 1983. Rheology of the lithosphere. ern North Sea. Marine and Petroleum Geology, 12,
Review of Geophysics and Space Physics. 21, 881-901.
1458-1487. OD1NSEN, T,. REEMST, P., VAN DER BEEK, P., FALEIDE,
KooI, H. 1991. Tectonic modelling of extensional J. I. & GABRIELSEN. R. H. 2000. Permo-Triassic
basins." the role qf lithosphere flexure, intraplate and Jurassic extension in the northern North Sea:
stress and relative sea-level change. PhD thesis, results from tectonostratigraphic forward model-
Vrije Universiteit, Amsterdam. ling. This vohmw.
, CLOETINGH, S. & BURRUS, J. 1992. Lithospheric PLATT. N. H. 1995. Structure and tectonics of the
necking and regional isostacy at extensional northern North Sea: new insights from deep
basins, 1. Subsidence and gravity modelling with penet/ation regional seismic data. In: LAMBIASE,
an application to the Gulf of Lions margin J. J. (ed.) Hydrocarbon Habitat in Rift Basins.
(SE France). Journal of Geophysical Research, 97. Geological Society, London, Special Publications,
17553-17571. 80, 103-113.
KUSZNIR, N. J., MARSDEN, G. & EGAN. S. S. 1991. A RAVNAS, R. & BONDEVlK, K. 1999. Architecture and
flexural cantilever simple-shear, pure-shear model controls on Bathonian-Kimmeridgian shallow-
of continental lithosphere extension: applications marine syn-rift wedges of the Oseberg~Brage area,
to the Jeanne &Arc-Basin, Grand Banks and northern North Sea. Bash7 Research, in press.
2022
L I T H O S P H E R E E X T E N S I O N IN T H E V I K I N G G R A B E N 81
, , HELLAND-HANSEN, W., LOMO, L., RY- northern Viking Graben. Basin Research, 9,
SETH, R. & STEEL, R. J. 1999. Sedimentation 133-150.
history as an indicator of rift initiation and devel- - - , VAN BALEN, R. T., BERTOTTI,G. & CLOETINGH,
opment: the late Bajocian-Bathonian evolution of S. A. P. L. 1998. The influence of a stratified
the Oseberg-Brage area, northern North Sea. rheology on the flexural response of the litho-
Norsk Geologisk Tidsskrift, in press. sphere to (un)loading by extensional faulting.
ROBERTS, A., YIELDING, G., KUSZNIR, N., WALKER, I. Geophysical Journal International, 134, 721-735.
& DORN-LOPEZ, D. 1993. Mesozoic extension in TORSVIK, T. H., STURT, B. A., SWENSSON, E., ANDER-
the North Sea: constraints from flexural back- SEN, T. B. & DEWEY, J. F. 1992. Palaeomagnetic
stripping, forward modelling and fault popula- dating of fault rocks: evidence for Permian and
tions. In: PARKER, J. R. (ed.) Petroleum Geology of Mesozoic movements and brittle deformation
Northwest Europe: Proceedings of the 4th Con- along the extensional Dalsfjord Fault, western
ference. Geological Society, London, 1123-1136, Norway. Geophysical Journal International, 109,
, , , & 1995. Quantitative 565-580.
analysis of Triassic extension in the northern TSENN, M. C. & CARTER, N. L. 1987. Upper limits of
Viking Graben. Journal of the Geological Society, power law creep of rocks. Tectonophysics, 136,
London, 152, 15-26. 1-26.
ROE, S. L. & STEEL, R. J. 1985. Sedimentation, UNDERHILL, J. R. & PARTINGTON, M. A. 1993.
sealevel rise and tectonics at the Triassic-Jurassic Jurassic thermal doming and deflation in the
boundary (Statfjord Formation), Tampen Spur, North Sea: implications of the sequence strati-
northern North Sea. Journal of Petroleum Geol- graphic eveidence. In: PARKER, J. R. (ed.)
ogy, 8, 163-186. Petroleum Geology of Northwest Europe." Proceed-
SCLATER, J. G. & CHRISTIE, P. A. F. 1980. Conti- ings of the 4th Conference. Geological Society,
nental stretching: an explanation of the post-mid- London, 337-345.
Cretaceous subsidence of the North Sea basin. WALTHAM, D. 1989. Finite difference modelling of
Journal of Geophysical Research, 85, 3711-3739. hanging wall deformation. Journal of Structural
SNEIDER, J. S., DE CLARENS, P. & VAIL, P. R. 1995. Geology, 11,433-437.
Sequence stratigraphy of the Middle to Upper 1990. Finite difference modelling of sandbox
Jurassic, Viking Graben, North Sea. In: STEEL, analogues, compaction and detachment free defor-
R. J., FELT, V., JOHANNESSEN, E. & MATHIEU, C. mation. Journal of Structural Geology, 12, 375-381.
(eds) Sequence Stratigraphy on the Northwest WEISSEL, J. K. & KARNER, G. D. 1989. Flexural uplift
European Margin. Norwegian Petroleum Society, of rift flanks due to mechanical unloading of the
Special Publication, 5, 167-197. lithosphere during extension. Journal of Geophy-
SPADINI, G. & PODLADCHIKOV, Y. 1996. Spacing of sical Research, 94, 13 919-13 950.
consecutive normal faulting in the lithosphere: a WroTE, N. J. 1990. Does the uniform stretching model
dynamical model for rift axis migration (Tyr- work in the North Sea? In: BLUNDELL, D. J.
rhenian Sea). Earth and Planetary Science Letters, & GIBBS, A. D. (eds) Tectonic Evolution of the
144, 21-34. North Sea Rifts. Clarendon, Oxford, 217 240.
STEEL, R. 1993. Triassic-Jurassic megasequence strati- - - & YIELDING, G. 1991. Calculating normal fault
graphy in the Northern North Sea: rift to post-rift geometries at depth: theory and examples. In:
evolution. In: PARKER, J. R. (ed.) Petroleum Geol- ROBERTS, A. M., YIELDING, G. & FREEMAN, B.
ogy qf Northwest Europe. Proceedings of the 4th (eds) The Geometry of Normal Faults. Geolog-
Conference. Geological Society, London, 299-315. ical Society, London, Special Publications, 56,
- - - - & RYSETtl, A. 1990. The Triassic-early Jurassic 251 260.
succession in the northern North Sea: megase- --, JACKSON, J. A. ~¢ MCKENZlE, D. P. 1986. The
quence stratigraphy and intra-Triassic tectonics. relationship between the geometry of normal
In: HARDMAN, R. P. F. & BROOKS, J. (eds) faults and that of the sedimentary layers in their
Tectonic Events Responsible ,for Britain's Oil and hanging walls. Journal of Structural Geology, 9,
Gas Reserves. Geological Society, London, Spe- 789-795.
cial Publications, 55, 139-168. WITHJACK, M. O. & PETERSON, E. T. 1993. Prediction
TER VOORDE, M. & BERTOTTI, G. 1994. Thermal of normal-fault geometries- a sensitivity study.
effects of normal faulting during rifted basin AAPG Bulletin, 11, 1860-1873.
formation. 1: A finite difference model. Tectono- Y~ELD~NG, G., BADLEY, M. E. & ROBERTS, A. 1992.
physics, 240, 133-144. The structural evolution of the Brent Province. In:
- - - 8 ~ ; CLOETINGH, S. 1996. Numerical modelling of MORTON, A. C., HASELDINE, R. S., GILES, M. R.
extension in faulted crust : effects of localized and & BROWN, S. (eds) Geology of the Brent Group.
regional deformation on basin stratigraphy. In: Geological Society, London, Special Publications,
BUCHANAN, P. G. & NIEUWLAND, O. A. (eds) 61, 27-40.
Modern Developments in Structural Interpretation, ZIEGLER, P. & VAN HOORN, B. 1989. Evolution of
Validation and Modelling. Geological Society, North Sea rift systems. In: TANKARD, A. J.
London, Special Publications, 99, 283-296. BALKWILL, H. R. (eds) Extensional Tectonics and
- - - , RAVNA.S, R., FASRSETH, R. & CLOETINGH, S. Stratigraphy of the North Atlantic Margins.
1997. Tectonic modelling of the middle Jurassic American Association of Petroleum Geologists,
synrift stratigraphy in the Oseberg Brage area, Memoir, 46, 471-500.

59 Full

Uploaded by

Copyright:

Available Formats

59 Full

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

59 Full

Uploaded by

Copyright:

Available Formats

Downloaded from http://sp.lyellcollection.

org/ at IFREMER (ISTEX) on January 4,

Repeated lithosphere extension in the northern Viking Graben:

l Institute o f Earth Sciences, De Boelelaan 1085, 1081 H V Amsterdam, the Netherlands

i ................ Platform ~ kin,

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 61

Reference First rift phase Second rift phase

Giltner (1987) 248 213 Ma* 169-98 Ma*

* Time scale of Harland et al. (1982).

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 63

" .If ~ _ ~.... / - _ ~ / - .t-- . ~ ~ /....~- /

, 25 km I ~\N -, "'" "

64 M. TER VOORDE ET AL.

8,,, d ) base Cretaceous - base T e r t i a r y

8,o b) top Triassic - top Brent

E : -'9.-.,. ~. . . . . . . . . -'" ~ "",

a) top basement - top Triassic

50 100 150 200

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 65

!i!i i!!iiil ........ ...... ......

d) base Cretaceous - base Tertiary

Viking Graben Oseberg Brage Horda Platform

c) top Brent - base Cretaceous

50 100 150 200

b) top Triassic - top Brent

V.G. Oseberg Brage Horda Platform

50 100 150 200

66 M. TER VOORDE E T AL.

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 67

68 M. TER VOORDE E T AL.

][iiiiii~ ...... [l|tl!li'~ ' 'iiiiii[

~!!!!!!!!!!!!!!!!!!!!!!!!!!! ...... iiiii(iiii::i, ---

condition of volume conservation is satisfied. deformation of the crust, temperature changes

where w is deflection (m), D is rigidity (N m), Model parameters

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 69

Model sensitivity and constraints

70 M. TER VOORDE E T AL.

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 71

0 [ "ESP 'VikingG. OsebergBrage " HordaPlatform ]

72 M. TER VOORDE E T AL.

ESP VikingG . Oseberg Brage HordaP l a t f o r m 1

/ ' ' ' /

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 73

o [___ " ESP V ~ _ a g e H0~a Platform"]

, I , J , iiiiiiiiiiii l iiiiiiiii iiilli

' 5~0 ' 100

74 M. TER VOORDE E T AL.

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 75

100 200 300 400 500 600

100 200 300 400 500 600

76 M. TER VOORDE E T AL.

Table 3. Parameters used for construction of strength pro[iles

Upper crust Lower crust Mantle

Ep (kJ mol -l ) 172.6 134 212 276 510

R (J (tool K) -1) 8.314J (mol K)

LITHOSPHERE EXTENSION IN THE VIKING GRABEN 77

The evolution of the Viking Graben has been

78 M. TER VOORDE E T AL.