NORWEGIAN JOURNAL OF GEOLOG Y Crelaceous post-rift deposils, northern North Sea 169

Forward depositional modell ing of the Cretaceous post-rift deposits in the northern North Sea

Lena K. Øvrebø, Tomas Kjennerud, Stephen J. Lippard, Jan C. Rivenæs & Martin Hamborg

Øvrebø, L., Kjennerud, T., Lippard, S.J., Rivenæs, ] .C., Hamborg, M.:Forward depositional modelling of the Cretaceous post-rift deposits in the northern North Sea. Norsk Geologisk Tidsskrift Vol. 81, pp.169-178 . Trondheim 200 l. ISSN 0029-196X.

The 2D stratigraphic simulation software DEMOSTRAT is used to model the Cretaceous post-rift deposits of the northern North Sea. The aim is to reproduce the geometry of the structural reconstruction of a regional pro file a cross the northern Horda Platform and Viking Gra ben. The geo­metry of the reconstruction is reproduced successfully by the simulation and shows that deposition during the Cretaceous was submarine with mass-flowage being a significant process in the Viking Graben in the early Cretaceous. In contrast, no mass flowage occurred on the Horda Flat­form. The simulations show that subsidence occurred throughout almost the entire Cretaceous in the Viking Graben and on the Horda Platform and that subsidence rates accelerated in the Cenomanian - Turonian. A structural high separating the Viking Graben and the Horda Platform experienced uplift and emergence in the early Cretaceous, but after this it was submerged. The Viking Graben experienced minor subsidence in the Maastrichtian, but only if the palaeowater-depth at this time was more than 180 m, whereas shallower water depths require uplift.

L. K. Øvrebø • & S. J. Lippard, Department of Geology and Mineral Resources Engineering, NTNU, N0-7 491 Trondheim, Norway; T. Kjennerud & M. Hamborg, Basin Modelling Department, SINTEF Petroleum Research, N0-7465 Trondheim, Norway; f. C. Rivenæs, Norsk Hydro, Research Centre, N0-5020 Bergen, Norway. •Present address: Department of Geology, University College Dublin, Belfield, Dublin 4, Ireland.

lntroduction Depositional modelling is an important tool in under­standing the sedimentary development of basins. In modelling, the different geological processes operating in a basin are viewed upon as functions of specific parame­ters. Simple cases can be constructed to see how these parameters affect the processes and the depositional geo­metry. The goal of simulating the evolution of a basin is to determine how the processes act together and, if pos­sible, what processes are the most important in the evol­ution of the basin.

Quantitative stratigraphic models can be divided into two classes, geomettic models and dynamic models (Paola, 2000). In geometric models, the sediment surface is represented by one or more surfaces with predefined geometry. In dynamic models, sediment transport and deposition interact with subsidence to produce the sur­face topography. Dynamic models can further be divided into forward and inverse models. In forward modelling, predictions are made based on causative agents (proces­ses), and these models can be termed process-response models. In inverse models, measurements of present-day parameters are used to establish the evolution of a basin. The software used in the present study is DEMOSTRAT (Rivenæs 1993), which is a 2D forward simulation of sili­ciclastic deposition. The erosion, transport and deposi­tion of sediments in DEMOSTRAT are solved by assig­ning individual transport coefficients to the sediment

BEGIN

l TECTONIC SUBSIDENCE

SET SEA LEVEL

EROSJON AND DEPOSITION

JSOSTA TIC ADJUSTMENT

END?

YES

Fig. l The time-loop algorithm in DEMOSTRAT (Lø�eth, 1999). The processes are modelled sequentially for each time step. All mod u les, except for erosion and deposition, can be bypassed if unwanted.

170 L. K. Øvrebø et al.

types (clay and sand), and then solving a set of coupled diffusion equations for each type simultaneously (Rive­næs, 1992; 1993; 1997). The model predicts downslope sorting of the sediments. This multiple-diffusivity approach is appropriate for sand-mud sorting on conti­nental shelves where it can be argued that the processes are diffusional (Paola, 2000). DEMOSTRAT was there­fore chosen for the present study. All the different pro­cesses that contribute to sediment erosion, transport and deposition in the basin are modelled by adjusting the transport coefficients. Further description of the algo­rithm can be found in Rivenæs (1992; 1993; 1997). Mass flows are modelled separately in DEMOSTRAT (Løseth, 1999), where stability calculations of the slope are per­formed. When slope failure occurs, the slope is eroded and diffusion processes transport the sediments into deep water where they are deposited. In addition to sedi-

-l·

62 !

60

58

NORWEGIAN JOURNAl OF GEOlOG Y

ment transport, erosion and deposition, the following elements are taken into consideration in the simulation of a basin: sediment supply, compaction ( empirical porosity-depth relations), tectonic subsidence history (user-defined subsidence rates), isostasy and eustacy (user-defined sea-leve! curve). All the above described process-elements are linked together in the software in a time-step loop (Fig. 1).

In the present work, DEMOSTRAT is used to model Cretaceous deposition along a profile crossing the Viking Graben and the Horda Platform (Fig. 2). The profile has also been reconstructed with respect to palaeobathyme­try by Kjenneud et al. (in press). This reconstruction, together with palaeowater-depth obtained from micro­palaeontology by Gilmore et al. (in press), forms the basis for the simulations in the present study. Kyrkjebø et al. (2000) modelled the Cenozoic deposition along the

50 km

same profile. The result of their study was that the Cenozoic subsidence rates deviated from the expec­ted post-rift pattern.

The main aim of the simulations of the present study is to reproduce the geometry of the structu­ral reconstruction and obtain information on subsidence patterns and associated sedimentation (for example predict the presence of mass-flows).

Fig. 2 Structural map of the northern North Sea showing the location of the structurally re­stored (Fig. 4) and modelled profile (Fig. 7) and the well 30/10-06 (Fig. 3) in the area. ESB = East Shetland Basin; ESP = East Shetland Platform; VG = Viking Graben; HP = Horda Platform; SB = Stord Basin; UH = Utsira High; UT = Uer Terrace; MgB =Magnus Basin; MrB =Møre Basin; MT = Møre Terrace; WG = Witch Ground Graben; AG = Asta Graben. The map also shows the two locations (50 km and 150 km) of the tec­tonic subsidence curves in Fig. 8.

NORWEGIAN JOURNAL OF GEOLOG Y

Geological setting The northern North Sea is an epicontinental basin, loca­ted between southern Scandinavia and northern Britain and is composed of a number of Mesozoic sub-basins and highs. The main rift basins are the Central Graben, Viking Graben and Outer Moray Firth (Witch Ground Graben), which join at a triple junction. Two major -phases of rifting have been recognized: the first occurred in the (?)late Permian to early Triassic (Badley et al. 1988; Gabrielsen et al. 1990; Færseth 1996) and the second in the mid-late Jurassic (Badley et al., 1988; Gabrielsen et al. ; 1990; Rattey & Hayward 1993; Færseth 1996; Færseth & Ravnås 1998). Most of the major faults active during the Jurassic rifting were probably reactivated faults for­med during the earlier rift phase (Badley et al. 1988; Færseth 1996); however, the axis of Permo-Triassic rif­ting seems to have underlain the Horda Platform, where­as Jurassic rifting was concentrated further west in the Viking Graben (Færseth 1996).

The Cretaceous post-rift episode that followed the mid-late Jurassic rifting has received relatively little attention in the lite­rature, largely because of the predominantly fine­grained nature of the sediments and the resul­ting lack of prospecti­vity. The sediments deposited during the Cretaceous in the nor­thern North Sea are mainly marls, clays and fine-grained silts, with some deep-water sands, best known from the Agat area in the northe­ast (Isaksen & Tonstad 1989; Skibeli et al. 1995). Isaksen & Tonstad (1989) provided the for­mal lithostratigraphic nomenclature for the Cretaceous of the nor­thern North Sea. Skibeli et al. (1995) described the Lower Cretaceous sequence stratigraphy and depositional envir-onments in the northern North Sea. They divided the succession into two

Ma lfieriocl Eooch Grou�

70 -

80 -

en 90 - :l

o w

100 - (.) � w Albian

110 - 0: (.) o

Aptian c �

120 - ... Q)

Bar. E e

130 - Ha ut. ()

Val.

140 - Ryaz.

Cretaceous post-rik deposits, northern North Sea 171

shelfal environment, whereas the upper (Aptian to late Albian) was deposited in a basinal environment.

Several authors, in describing the general develop­ment of the North Sea basins (Hancock 1990; Ziegler 1990), have noted during the Cretaceous a general trend of decreasing water depths, with sedimentation rates generally exceeding the combined effects of subsidence and rising sea levels, characterize the northern North Sea. A parallel trend was the gradual burial and sub­mergence of pre-existing topography generated during the previous rifting phases (Nøttvedt et al. 1995). Nøt­tvedt et al. (1995) noted that "deep water depositional environments dominated along the basin axis in the Cretaceous with progressive onlap against basin margins and intrabasinal highs".

In the present and related studies (Kjennerud et al.) (in press), the Cretaceous of the northern North Sea is divided into 6 informal seismo-stratigraphic units (K l -

K6) (Fig. 3), with the lower two (K l - K2) representing the Lower Cretaceous and the upper four (K3 - K6) the Upper Cretaceous. Gabrielsen et al. (in press) divide the

Åsgard­Mime

K1 Sh

Palaeowater-depth well 30/1 0-6

.':"'e'

:.� 1·=-: --: . - : - ....

l l • • .. - ... - -·- -· l l l l • • f l

.

=-· fs'rT:'a 1 l ' : 1 : l Micropal. l : r : . :• :• :. -:- -r�--+1--!--lt

0: : l: • l ' : l : : , : .. . .

-_ t---i--; t: • : , : : : , : : : . ' . # l ' '

i...!.. : : ' ' . ' . . ' ' l � : .

- Most likely range (micropalaeontology)

• • • • • Additional max/min range (micropalaeontology) supersequences and showed that the lower supersequence (Ryazanian to late Barremian) was deposited in a mainly

Fig. 3 The stratigraphic outline of the Cretaceous from well 30/1 0-06, showing the time sea le, the lithostra­tigraphy, lithology data, the palaeowater-depths obtained from the structural reconstruction (Kjennerud et al., in press) and the palaeowater-depth obtained from micropalaeontology (Gilmore et al. in press). Cl= Clay. Lst = Limestone. Sh = shale. The time scale is taken from Gradstein et al. 1995.

172 L. K. Øvrebø et a l .

Cretaceous of the northern North Sea, on the basis of palaeo-topography, into three stages of post-rift deve­lopment, which they call the incipient, middle and mature stages, respectively. The incipient post-rift stage (K l - K2, Ryazanian - Albian) is characterized by a highly irregular topography inherited from the rifting p hase, with local basins and the crests of fault blocks as emergent islands or shoals. The middle post-rift stage (K3 - K4, Cenomanian - Turonian) is characterized by progressive burial and submergence of the pre-existing relief, and the mature post-rift stage (KS - K6, Coniacian - Maastrichtian) by widening of the basin, complete drowning of the pre-existing relief and overstepping of the rift margins.

Model lnput The main inputs to the modelling are the initial basin shape, sediment input, transport coefficients, parameters to calculate mass flows, compaction, subsidence, isostasy and eustacy (Table 1).

Two recent studies on Cretaceous - Cenozoic palaeo­bathymetry are used as input for the basin shapes. Kjen-

NORWEGIAN JOURNAL OF GEOLOGY

nerud et al. (in press) restored palaeobathymetry along four regional sections in the northern North Sea, using structural restoration techniques, sequence geometries and zero water-depth information. Transect 2 in Kjenne­rud et al. (in press) (Fig. 4) is modelled in the present study. Gilmore et al. (in press) reconstructed water­depth using micropalaeontology for twelve wells. Well 30/10-06 (Fig. 3) is most convenient, located almost on the pro file and is used as input in the present study.

The late Ryazanian basin shape is based upon the structural reconstruction performed by Kjennerud et al. (in press) (Fig. SA). In the earliest Cretaceous, the basin consisted of two sub-basins (the Viking Graben and the Horda Platform) separated by an intrabasinal high. Published data (e.g. Isaksen & Tonstad, 1989) show that the Cretaceous deposits that fllled in the basin were dominantly shales. Isaksen and Tonstad (1989) report Upper Cretaceous chalk on the Horda Platform. The seismic data show an infilling pattern in the Viking Gra­ben and on the Horda Platform, indicating that the main sedimentation process was probably pelagic. The sedi­ment input is calibrated to match the thickness distribu­tion (Fig. SB-C) of the sediments in the basin. The depo­sit thickness varies along the profile with the thickest deposits in the Viking Graben and the thinnest on the

Table l s ·�· y of the input por'Clmeters used in the modelling. Parameter Value Initial basin shape - Late Ryazanian Structural Reconstruction of the pro file (Kjennerud et al., in press) (Fig. 4A)

Transport Coefficients The values that best reproduced the depositional geometry.

Lithology Clay (Isaksen & Tonstad, 1989; well 30/1 0-06)

Compaction of clay Helland Hansen et al. ( 1988)

Compaction of sand Baldwin & Butler (1985)

Slope gradient when mass-flow occurs 3.0 o

Slope gradient when deposition occurs 0.5 o

Tectonic subsidence The subsidence rates that best reproduced the geometry (Fig. 8)

Flexural rigidity of the crust 1.10" Nm (Thorne &Watts, 1989; Fjeldskaar, 1994; Kyrkjebø et al. 2000)

Eustacy Long term curve ofHaq et al. (1987) (Fig. 6)

ESP VG 30/10-06

km

Fig. 4 The present day profile that is modelled in the present study. The profile has been structurally restored by Kjennerud et. (in press) (Fig. 5).

NORWEGIAN JOURNAL Of GEOLOGY

o

1

2

3

4

Cretaceous post-rift deposits, northern North Sea 173

200 k

Fig. 5 Structural reconstructions of the cross section (Fig. 4) performed by Kjennerud et al. (in press). Kl-K6 are the Cretaceous seismic units (Fig. 3 ). A) Late Ryazanian (start point for the simulation). B) Late Cenomanian. C) Latest Cretaceous (end point for the simulation).

Horda Platform. The deposits thin towards the Shetland Platform and the intrabasinal high. Sediments are sup­plied to the basin both through erosion of the Shetland Platform and from an external source in the east (sout­hern Norway). The sediments are transported into the basin in suspension. Sediment was probably also sup­plied from the north and the south, but these sources are not considered in the present study. For compaction of the sediments the empirical porosity-depth relationships of Baldwin & Butler (1985) for sand and Helland-Han­sen et al. (1988) for shale are used.

The transport coefficient reflects the energy conditi­ons dominant at different times, and the transport effici­ency of the sediments (Rivenæs, 1993). The parameter is difficult to quantify and verify. The values used in the present study are those that best reproduce the pattern observed in the structural reconstruction. The transport coefficients are set to be high above sea level where ero­sion occurs and to decrease exponentially with water­depth. Mass-flowage is set to occur when the slope gradi­ent is steeper than 3.0oo . The gradient of the slope where the mass-flow deposition starts is set to 0.5oo .

The mass-flow diffusivities are chosen so that the mass­flows are small, but frequent.

The criteria for the tectonic subsidence curves are that large enough accommodation space must be created for the deposits and the water masses throughout time. Simulations for two different cases were run. In Case l the accommodation space is based on the structural reconstruction of the cross section (Kjennerud et al., in press) (Fig. 5). In Case 2 the sediment thickness and the basin shape are based on the structural reconstruction, while the absolute palaeowater-depths are based on micropalaeontological interpretations of Gilmore et al. (in press) (Fig. 3). In both cases the flexural rigidity of the crust is set to 1023 Nm. This value was used by Kyrkje­bø et al. (2000) in their modelling of the Cenozoic depo­sits in the northern North Sea. The value is in the lower range of what Fjeldskaar (1994) suggested. Fjeldskaar & Pallesen (1989) used a slightly higher value (5·1023 Nm) in their study of the isostatic subsidence in the Viking Graben. Thorne & Watts (1989) estimated the flexural rigidity to 1023 Nm for the Viking Graben based on gra­vity studies. Lower values have been suggested for the

17 4 L. K. Øvrebø et al.

Jurassic rifting period (e.g. Marsden et al. , 1990; ter Voorde et al., 1997). The flexural rigidity is, however, thought to generally increase during the post-rift period (e.g. Watts et al., 1980; Watts, 1982).

The tectonic movements are modelled as vertical movements in DEMOSTRAT. The cross section shows several faults in the Cretaceous, but only the western border faults of the Viking Graben are considered to be significant for the evolution of the rift basin. In the Cre­taceous, Viking Graben faulting is an important subsi­dence mechanism (Badley et al., 1988). According to Badley et al. (1988) two processes appear to have driven the faulting: syrnmetric subsidence, implying a degree of local isostatic compensation, and asyrnmetric subsi­dence.

The underlying Jurassic and older deposits have a non-uniform thickness along the cross-section (Fig. 4), being thickest in the Viking Graben and thinning towards the flanks. This probably caused differential compaction and base-Cretaceous subsidence, which is not possible to represent in the DEMOSTRAT sirnula­tion. The compaction of the underlying sediments is the­refore added to the tectonic subsidence in the DEMOS­TRAT simulation. The error introduced by including the compaction of underlying sediment is corrected for by calculating the compaction of Jurassic sediments. This compaction is subtracted from the tectonic subsidence curves.

For the global variations of the sea level, the lang­term eustatic sea level curve (Fig. 6) of Haq et al. ( 1987) is used, since the modelling considers long-term pro-

Maastrichtian K6

Campanian KS

Conlaclan

Turonlan K4

Cenomanian K3

Al bi an K2

Aptian

Barremian

Hauterivian K1

Valangian

Ryazanian

Sea Level (m) -50 50 150 250

\

( )

)

( � /

70

80

90

100

110

120

130

140

Fig. 6 Long term sea-level curve of Haq. et al. (1987) for the Cretaceous. Present day sea-level is used as reference. The Cretaceous seismic units (Kl-K6) are shown on the curves. The lang term sea­leve! curve is used as input in the simulation.

NORWEGIAN JOURNAL OF GEOLOGY

cesses. As the depositional styles show an infllling pattern rather than a progradational pattern, a lang­term global sea-level change of 50- 150 m is not reflec­ted in the depositional style. The global sea level curve influences the tectonic subsidence. If the global sea-level is low, high tectonic subsidence rates will be modelled in order to reproduce the basin shape, and vice versa.

Modelling results During the early Cretaceous the basin was divided into two sub-basins separated by an intrabasinal high. In the late Ryazanian the high was located between 65 km and 90 km in the profile (Fig. 4A). Sediments were supplied to the sub-basins through erosion of the highs and from external sources. The simulation results of the present study are plotted in Fig. 7 A-D, while the tectonic sub­sidence curves are shown in Fig. 8.

Case l: Palaeowater-depth from structural reconstruction

The output of the simulation of Case l is shown in Fig. 7 A-C. The deposits onlap the Shetland Platform and the intrabasinal high in the Ryazanian to the late Cenoma­

nian (K l-K3). In the Turonian to the Maastrichtian (K4-K6) the high was submerged and covered by sediments.

The sedimentation rate increased in the Cenomanian (K3), especially on the Horda Platform, where the rate was almost three times higher than in the previous period (Ryazanian to Albian, K l-K3), as shown by incre­ased distance between the time lines (Fig. 7 A).

On the western margin of the Viking Graben active faulting occurred and this caused steep depositional slo­pes. The simulation suggests that mass-flow deposition was a major process in the Viking Graben in the early Cretaceous (K l-K4). It is not possible to recreate the geometry of the basin with no mass-flow activity. The process brought sediments eroded from the highs out into deep water in the graben (Fig. 7A-C). In the Aptian (K2) the turbidite deposits shifted slightly to the east in the Viking Graben according to the simulation. This can be explained by shifts in the subsidence pattern. On the Horda Platform the gradients were low and no mass­flows occurred in the Cretaceous. The sand occurring at the western margin of the Horda Platform (Fig. 7 A) is a shallow marine sand owing to erosion of the intrabasinal high separating the Viking Graben and the Horda Plat­form.

The present study showed that the Viking Graben and the Horda Platform experienced subsidence (Fig. 8) in the Ryazanian to Cenomanian (K l-K3), while the intrabasinal high was subjected to uplift. Two areas of maximum subsidence existed on the Horda Platform, one in the east and one in the west. The subsidence rate was higher in the Viking Graben than on the Horda Plat­form, with maximum subsidence in the centre (Fig. 8). The depositional basin in the Viking Graben widened

NORWEGIAN JOURNAL OF GEOLOGY Cretoceous post-rift de�sits, northern North Seo 175

Fig. 7 The results of the simulations. SAND FRACTION (%) O m =present day sea level. For ages of the seismic units see Fig. 3. A) -C): Case 1: Palaeowater- o 10 20 30 40 50 60 70 80 90 100 depth based on the structural restorations 800 (Fig. 5 ). Note that sand is deposited as mass-flows 400 in deep water in units K1-K4 in the east of the 200

o Viking Graben. A) Late Cenomanian ( top K3 ).

�:: Shallow marine sand is deposited on the Horda Platform due to erosion of the intrabasinal high. 1: B) Latest Cretaceous (top K6). Palaeowater-depth )i of 30 m is assumed, and uplift occurs in the � ·1000 Viking Graben. C) Latest Cretaceous (top K6). ·1200

·1400 Zero subsidence in the Viking Graben gives a ·1800 palaeowater-depth of 180 m. D) Late Turonian ·1800 (top K4) for Case 2: Palaeowater-depth based on ·2000 the micropalaeontology. Less mass-flowage occurs ·2200 in the Viking Graben. DISTANCE (KM)

400 200

o ·200

(-400 �-800 li-eoo �-1000

-1200 ·1400 ·1800 ·1800

400 200

o ·200

(-400 z-800 �-800 §-1000 Ill

·1200 ·1400 ·1800 -1800 o 20 -40 60 80 100 120 140 160 180 200

DISTANCE (KM) 400

300

200 100

i o

� ·100 c 5·200

-300 -400 ·500 -800

o 20 -40 eo 80 100 120 140 180 180 200 DIST ANCE (KM)

176 L. K. Øvrebø et al .

with time, so that at the end of the Aptian (K l) it was about 35 km wide, while at the end of the Cenomanian (top K3) it was 55 km wide. This is because the western boundary fault backstepped with time, and larger parts of the intrabasinal high in the east of the Viking Graben subsided. By the end of the early Cretaceous (K3) almost the entire intrabasinal high had subsided. In the Cenom­anian (K3) the subsidence rate accelerated in the Viking Graben. On the Horda Platform, the westernmost part experienced uplift in the Cenomanian (K3), while on the rest of the platform subsidence accelerated.

After the latest Cenomanian (K4- K6) the intrabasi­nal high became submerged and covered with sediments (Fig. 7B). Mass-flows continued to be active in the Viking Graben in the late Cenomanian to Turonian (K4). After the Turonian the basin gradients became gentler and the mass-flow activity ceased.

Except for the East Shetland Platform in the west, subsidence occurred along the entire profile in the late Turonian to Campanian (K4 - KS). The faulting at the western border continued its backstepping nature in the late Turonian to the end of the Campanian. The fault movements ceased in the Maastrichtian. The subsidence rates were higher in the Viking Graben than on the Horda Platform during the late Cenomanian to the earli­est Maastrichtian (K4 - K6) (Fig. 8). The subsidence rates over the earlier intrabasinal high were roughly the same as the rates affecting the Horda Platform. The subsidence

A)

·200

·400

·600

g ·800

5 -1000 :;:; � Cll ·1200 iii

·1400

·1600

·1800

·1700

Tectonlc subsidence at 50 km

Viking Graben

125 105 Ma

85 65

NORWEGIAN JOURNAL OF GEOLOGY

rates decreased in the Coniacian to the Campanian (KS). On the Horda Platform the rates were dose to zero in this period.

The subsidence rates in the Maastrichtian (K6) are uncertain, and depend on how large the accommodation space was at this time. The structural reconstruction indicates low relief, and possible shallow water-depth (Kjennerud et al., in press). Water-depths at the end of the Cretaceous are uncertain. Case l is therefore divided into two sub-cases, Cases la and lb (Fig. 7B- C). In the simulation of Case la the water-depth is set to 35 m, causing uplift (Fig. 8) in the Viking Graben. The purpose of Case lb is to find out what the water-depth was in the Viking Graben if zero subsidence occurred. The result of the simulation of Case lb is a water-depth of 180 m by the end the of the Cretaceous.

Case 2: Palaeowater-depth from micro-palaeontology

In Case 2 the sediment supply, transport coefficients and the eustatic sea level are unchanged from the earlier set of simulations. The subsidence rates are the only para­meter that is changed from Case l . A significant diffe­rence is that for Case 2 less mass-flows occurred in the Turonian (K4) (Fig. 7D). This can be explained by the changes in water-depth and subsidence rates, which cause lower depositional slope gradients in the Turonian. In contrast to Case l, in Case 2 the Viking Graben expe­rienced subsidence in the Maastrichtian (Fig. 8).

B)

·100

·200 e �-300 o �-400 � iii ·500

-600

·700

Tectonic subsldence at 150 km

Horda Platform

·800 ......_ ______________ � 125 105 85 65

Ma

Case 1a ---- Case 1b --- Case2

Fig. 8 Tectonic subsidences at two locations along the profile. See Fig. 2 for location. The tectonic subsidence is corrected for compaction of the underlying sediments. Case la and lb: The palaeowater-depth in the Cretaceous is based on the structural restorations. Case la: The palaeo­water-depth is 35 m in the Viking Graben by the end of the Cretaceous, resulting in uplift of the Viking Gra ben in the Maastrichtian. Case lb: The subsidence is zero to minor in the Viking Graben causing the palaeowater-depth to be 180 m at top Cretaceous. Case 2: The palaeowater­depth is based on micro-palaeontology. A) Centre of the Viking Gra ben B) Horda Platform. Note that the vertical scale is different in the two graphs. Ma =Million years befare present.

NORWEGIAN JOURNAl OF GEOLOG Y

Discussion and condusion

Simulation compared to reconstruction

The sediment transport is assumed to be in an east­west direction, transverse to the basin axis. It is possible to reproduce the geometry of the pro file without consi­dering sediment transport along the basin axis. The dif­ferences in deposited thicknesses between the simula­ted and reconstructed profiles are less than l 00 m. The distribution of mass-flow deposits can explain any small thickness difference between the simulation and the reconstruction. The mass-flow deposits may in reality have another distribution than shown in the simulation. Depending on the bathymetry they will spread out in three dimensions, and will not be found along a straight line from the source.

Mass-flows

The modelling shows that mass-flowage is more likely in the Viking Graben in the Ryazanian to the Turonian, than in the latest Cretaceous. On the Horda Platform no mass-flow activity occurred in the Cretaceous due to low gradients. Cretaceous mass-flows in the North Sea have been reported in the Agat area by Guldbrand­sen et al. (1987) and Shanmugan et al. (1994). The early Cretaceous margins were steep due to inherited rift­relief, mass-flowage is therefore likely to have occurred. The mass-flows were not necessarily sandy, as modelled in the present study. The lithology of the mass-flows depends on the source area. Well 30/10-06 is located at 65 km, where no mass-flows are deposited in the simu­lations. There is no silt or sand in the well that could indicate the presence of coarse-grained mass-flows.

Subsidence

The tectonic subsidence curves obtained in the present study are a result of reproducing the accommodation space for each seismic unit observed on the reconstruc­tion. Due to uncertainties regarding the compaction of the underlying sediments (see discussion under model input), the precise tectonic subsidence rates cannot be extracted. The shape of the subsidence curves is, howe­ver, believed to be correct.

Nadir & Kusznir (1995) modelled the latest Cretace­ous bathymetry to range from 500 m to 800 m in the northern North Sea. Their results were based on flexure models for the crust and that the subsidence followed the McKenzie model. The study by Gradstein et al. (1999) indicated shallower water-depths than the " extreme" water-depths suggested by Nadir & Kusznir {1995). Gradstein et al. (1999) proposed a maximum water-depth of 450 m and that shallowing occurred during the late Cretaceous to 200 m by the end of the Cretaceous. In the present study the water-depth by the end of the Cretaceous was 35 m in Case la, 180 m in Case lb, and 200 m in Case 2. However, it should be

Cretoceous post;ift deposits, northern North Sea 177

noted that in the present study the water-depth is used as an input in the modelling, and not as an output.

The simulations for both Cases l and 2 show accele­ration in the subsidence in the Cenomanian - Turo­nian. This acceleration deviates from the McKenzie (1978) model where the subsidence rates decay with time following a rifting episode. Gabrielsen et al. (in press) also report acceleration in the subsidence rates in the Cenomanian - Turonian. They relate this to "a lit­hosphere-scale thermal event and corresponding trans­formation of the lithospheric rheological properties". Lippard and Liu (1990) used the software RIFT (based on the McKenzie model) to model the subsidence along a profile crossing the northern Viking Graben and the northern Horda Platform. Their results showed a Cen­omanian - Turonian acceleration in subsidence and higher subsidence rates in the Viking Graben than on the Horda Platform, as was the case for the present study.

All the simulations of the present study and the structural restorations show a smooth relief at the Cre­taceous-Tertiary boundary. Kjennerud et al. (in press) based the assumption of a smooth top Cretaceous relief on three observations: l) The K6 seismic reflectors are parallel to the Palaeocene reflectors in the central part of the basin. 2) The K6 isopach indicates that the rift relief was filled in. 3) Gilmore et al.'s (in press) micro­palaeontological estimates of the palaeowater-depth are typically 150- 300 m across the whole northern North Sea. Case l in the present modelling shows that the Viking Graben subsided in the Maastrichtian if the palaeowater-depth at the end of the Cretaceous was more than 180 m, otherwise uplift occurred. Gabrielsen et al. (in press) also suggested uplift in the Viking Gra­ben during the Maastrichtian, if shallow water-depths are assumed. Seismic and well data in the area show that the Danian sediments are missing (Isaksen & Ton­stad, 1989), an uplift in the Maastrichtian or early Pale­ocene could explain this.

Both Cases l and 2 show that the intrabasinal high experienced uplift in the early Cretaceous, while it sub­sided in the late Cretaceous. The behaviour of this high has not been reported in any previous studies. It is a northern continuation of the Utsira High, which fur­ther south remained as a positive high until the Maas­trichtian (Gabrielsen et al., in press).

Acknowledgements. We thank Mai Britt E. Mørk for constructive comments on an earlier version of the manuscript, and the NGT re­viewers for the constructive comments.

178 L. K. Øvrebø et a l .

References Badley, M. E., Price, J. D., Dahl, C. R. & Agdestein, T. 1988: The struc­

tural evolution of the northern Viking Graben and its bearing upon extensional modes of basin formation. journal of the Geolo­gical Society, London 145, 455-4 72 .

Baldwin, B. & Butler, C. O. 1985: Compaction curves. American Associ­ation of Petroleum Geologists Bulletin 69, 6641-6679 .

Fjeldskaar, W. & Pallesen, S . 1989: The application of a viscoelastic lit­hosphere model to isostatic subsidence in backstripping. In Corre­lation to Hydrocarbon Exploration. Norwegian Petroleum Society, 53-59 .

Fjeldskaar, W. 1994: Viscosity and thickness of the asthenosphere detected from the Fennoscandian uplift. Earth and Planetary Sci­ence Letters 126, 931-944 .

Færseth, R. B. 1996: Interaction of Permo-Triassic and Jurassic exten­sional fault-blocks during the development of the northern North Sea. journal of the Geological Society, London 153, 931-944 .

Færseth, R. B. & Ravnås, R. 1998: Evolution of the Oseberg Fault­Block in context of the northern North Sea structural framework. Marine and Petroleum Geology 15, 467-490.

Gabrielsen, R. H., Færseth, R. B., Steel, R. J. & Kløvjan, O. S. 1990: Architectural styles of basin fill in the northern Viking Graben. In Blundell, D. & Gibbs, A. (eds.): Evolution of the North Sea Rifts, 158-179, Oxford Science Publication.

Gabrielsen, R. H., Kyrkjebø, R., Faleide, J. 1., Fjeldskaar, W. & Kjenne­rud, T., in press: The Cretaceous post-rift basin configuration of the northern North Sea. Petroleum Geoscience.

Gillmore, G. K., Kjennerud, T. & Kyrkjebø, R., in press: The recon­struction and analysis of palaeowater depths: a new approach and test of micropalaeontological approaches in the post-rift (Creta­ceous to Quaternary) interval of the Northern North Sea. In Mar­thinsen, O. et al. (eds.): Sedimentary Environments offshore Nor­way - Palaeozoic to Recent. Norwegian Petroleum Society Special Publication, X, xxx-xxx.

Gradstein, F.M., Agterberg, F. P., Ogg, J. G., Hardenbol, J., van Veen, P., Thierry, J., Huang, Z. 1995: A Triassic, Jurassic and Cretaceous time scale. SEPM (Society of Sedimentary Geology) 54, 95-126 .

Gradstein, F. M., Kaminski, M. A. & Agterberg, F.P. 1999: Biostrati­graphy and paleo-oceanography of the Cretaceous seaway bet­ween Norway and Greenland. Earth-Science Reviews 46, 27-98 .

Guldbrandsen, A., 1987: Agat. In Spencer, A. M. et al. (eds.): Geology of the Norwegian Oil and Gas Fields, 363-370. Graham & Trotman.

Hancock, J. M. 1990: Cretaceous. In Glennie, K. W. (ed): Introduction to the Petroleum Geology of the North Sea, 255-272 . Blackwell Sci­entific Publications.

Haq, B., Hardenbol, J. & Vail, P.R. 1 987: Chronology of fluctuating sea leve! since the Triassic. Science 235, 1156-1167 .

Helland-Hansen, W., Kendall, C. G. St. C., Lerche, I. & Nakayama, K. 1988: A simulation of continental basin margin sedimentation in response to crustal movements, eustatic sea leve! change, and accumulation rates. Mathematical Geology 20, 777-802.

Isaksen, D. & Tonstad, K. 1989: A revised Cretaceous and Tertiary lit­hostratigraphic nomenclature for the Norwegian North Sea. NPD Bulletin no 5 .

Kjennerud, T., Faleide, J. 1., Gabrielsen, R . H., Gillmore, G . K., Kyr­kjebø, R., Lippard, S. J. & Løseth, H., in press: Structural restora­tion of Cretaceous-Cenozoic (post-rift) palaoebathymetry. In Marthinsen, O. et al. (eds. ): Sedimentary Environments offshore Norway - Palaeozoic to Recent. Norwegian Petroleum Society Spe­cia! Publication X, xxx-xxx.

NORWEGIAN JOURNAl OF GEOLOGY

Kyrkjebø, R., Hamborg, M., Faleide, J. l., Jordt, H. & Christiansson, P. 2000: Cenozoic tectonic subsidence from 2D depositional simula­tions of a regional transect in the northern North Sea basin. In Nøttvedt, A. (ed.): Dynamics of the Norwegian Margin. Geological Society London, Special Publication 167, 273-294 .

Lippard, S. J. & Liu, G. 1990: Tectonic modelling of the northern North Sea using program RIFT. In Larsen, R. M., Brekke, H., Lar­sen, B. T. & Talleraas, E. (eds.): Structural and Tectonic Modelling and its Applications to Petroleum Geology. Norwegian Petroleum Society Special Publication l, 43-54 .

Løseth, T. M. 1999: Submarine Mass-flow Sedimentation. Computer Modelling and Basin Fill Stratigraphy. Springer Verlag. 156 p.

Marsden, G., Yielding, G., Roberts, A.M. & Kuznir, N. J. 1990: Applica­tion of a flexural cantilever simple-shear/pure-shear model of continental lithosphere extension to the formation of the nor­thern North Sea basin. In Blundell, D. S. & Gibbs, A.D. ( eds.): Tec­tonic Evolution of the North Sea Rifts, 240-261. Oxford Science Publication.

McKenzie, D. P. 1978: Some remarks on the development of sedimen­tary basins. Earth and Planetary Science Letters 40, 25-32 .

Nøttvedt, A., Gabrielsen, R . H . & Steel, R. J . 1995: Tectonostratigraphy and sedimentary architecture of rift basins, with reference to the northern North Sea. Marine and Petroleum Geology 12, 881-901 .

Paola, C. 2000: Quantitative models of sedimentary basin filling. Sedi­mentology 47, Suppl.l, 121-178 .

Rattey, R. P. & Hayward, A. B. 1994: Sequence stratigraphy of a failed rift system: the Middle Jurassic to Early Cretaceous basin evolut­ion of the Central and Northern North Sea. In Parker, J. P. (ed.): Petroleum Geology of Northwest Europe: Proceedings of the 4th Con­ference, 215-249 . Geological Society, London.

Rivenæs, J. C. 1992: Application of dual-lithology, depth-dependent diffusion equation in stratigraphic simulation. Basin Research 4, 133-146.

Rivenæs, J. C. 1993: A computer simulation model for siliciclastic basin stratigraphy. Dr. ing. Thesis, Department of Petroleum Geology and Applied Geophysics, University of Trondheim. 133 p.

Rivenæs, J. C. 1997: Impact of sediment transport efficiency on large­scale sequence architecture: results from stratigraphic computer simulation. Basin Research 9, 91-105 .

Skibeli, M, Barnes, K., Straume, T., Syvertsen, S. E. & Shanmugan, G. 1995: A sequence stratigraphic study of Lower Cretaceous depo­sits of the northernmost North Sea. In Steel, R. J. et al. (ed.) : Sequence Stratigraphy on the Northwest European Margin. Norwe­gian Petroleum Society Special Publication 5 , 389-400.

Thorne, J. A. & Watts, A. B. 1989: Quantitative analysis of North Sea subsidence. The American Association of Petroleum Geologists Bul­letin 73,88-161 .

ter Voorde, M., Ravnås, R., Færseth, R. & Cloething, S. 1997: Tectonic modelling of the Middle Jurassic synrift stratigraphy in the Ose­berg-Brage area, northern Viking Graben. Basin Research 9, 133-150.

Watts, A. B. 1982: Tectonic subsidence, flexure and global changes of sea-leve!. Nature 297,469-474 .

Watts, A. B., Bodine, J. H. & Steckler, M. S. 1980: Observations of fle­xure and the state of stress in the oceanic lithosphere. journal of Geophysical Research 85, 6369-6376.

Ziegler, P . A. 1990: Tectonic and palaeogeographic development of the North Sea rift system. In Blundell, D. S. & Gibbs, A. D. ( eds.) Tec­tonic Evolution of the North Sea Rifts, 1-36 . Oxford Science Publi­cation.