Paralic sedimentation on an epicontinental ramp shelf ... · Paralic sedimentation on an epicontinental ramp shelf during a full cycle of relative sea-level fluctuation; the ... sedimentary

343

Paralic sedimentation on an epicontinental ramp shelf during a full cycle of relative sea-level fluctuation; the Helvetiafjellet Formation in Nordenskiöld land, Spitsbergen

Ivar Midtkandal, Johan Petter Nystuen and Jenö Nagy

Ivar Midtkandal, Johan Petter Nystuen & Jenö Nagy. Norwegian Journal of Geology, vol. 87, pp. 343-359.Trondheim 2007. ISSN 029-196X.

A depositional model for the development of the Helvetiafjellet Formation in Nordenskiöld Land, Spitsbergen is presented. The formation was deposited into a segment of the large epicontinental Boreal basin that rimmed northern Pangaea during the Early Cretaceous. A wide range of depositional subenvironments are recorded within the succession; including fluvial braidplain, shallow marine bay, delta, coastal plain and fluvial channel. The depositional model approaches a layer-cake style for this part of the basin, caused by the rapid rates of progradation and retrograda-tion made possible by the gentle depositional gradient. An initial period of fluvial deposition arose in response to an early rise in relative sea-level. Following a regional flooding, the progradational to aggradational architecture in the area reflects a balanced rate of increase in accommodation vs. rate of sedimentation (A/S) ratio. This resulted in a heterolithic stacking of sandstone and mudstone. Autogenic variables are thought to have domi-nated the lateral facies variations recorded in the upper and middle parts of the succession.

Ivar Midtkandal ([email protected]), Johan Petter Nystuen ([email protected],) Jenö Nagy ([email protected]). Department of Geoscien-ces, University of Oslo, P.O Box 1047 Blindern, NO-0316 Oslo Norway

IntroductionBasin physiography is a major factor controlling sedi-mentary infill patterns. The relative influence of basin physiography varies between different basin types; tec-tonically active or passive, deep or shallow, large or small (Allen & Allen 2005). The present study deals with sedi-ment infill and variation in facies and sedimentary archi-tecture in an epicontinental basin with a low-sloping, wide ramp shelf setting dominated by paralic deposi-tional environments. The bathymetric profile of an epicontinental basin gen-erally shows a gradual deepening towards the basin cen-tre. Shelf breaks may be absent or only weakly developed. However, any topography inherited from the submerged continental plate of such a basin can potentially affect drainage and depositional architecture (Fagherazzi et al. 2004). Slope gradients may be an inherited property, or may be caused by tectonic activity during basin devel-opment. Variations in relative sea-level on low-gradi-ent shelves may give rise to long-distance basinward or landward shifts in facies, coupled with total emergence or complete submergence of the shelf. Moderate, high-frequent sea-level variations do not necessarily cause sig-nificant changes in overall facies stacking patterns, due to the relatively constant basin floor gradient and tectonic stability of the shelf area. Minor fluctuations in sea-level may trigger changes in coastal morphology, and varia-tions in relative impact on sediment partitioning by flu-

vial, tidal and wave-induced currents. Autogenic variables such as river avulsions and delta lobe shifts, together with variation in sedimentary regime parameters may con-tribute to lateral, as well as vertical variation in deposi-tional environment and facies (Swift & Thorne 1991). The relative influence of allogenic and autogenic fac-tors on the depositional architectural style is considered important in very low-sloping ramp shelf basins, as it influences modelling procedures significantly for this type of sedimentary succession. The horizontal scale of variation in depositional architectural elements is fur-thermore considered particularly crucial for modelling epicontinental basin successions, as emphasised for other types of depositional environments (Miall 1988; Walker 1992). Modelling studies have also shown that a wide variety of depositional styles may develop with identical sea-level, tectonic and sediment supply histories (Carey et al. 1999). This complicates any attempt to reconstruct sedimentary relationships in basin settings with very low resolution of the depositional relief. Several large epicontinental basins developed as a result of a global sea-level rise during the Cretaceous (Grocke et al. 2003). One of these was the Boreal basin at the northern margin of Pangaea, predating the opening of the Polar basin and the northeasternmost Atlantic in the Late Cretaceous to Early Cenozoic (Torsvik et al. 2002). The object of the present study is to describe deposi-tional patterns and facies variation within the Barremian

NORWEGIAN JOURNAL OF GEOLOGY Paralic sedimentation on an epicontinental ramp shelf

344

Helvetiafjellet Formation of Spitsbergen. This formation developed in a fluvial to paralic setting during a period of fall and rise in relative sea-level at the ramp margin of the Boreal epicontinental basin in the Svalbard domain. Vertical stacking patterns and lateral variations in facies associations, as controlled by processes and factors men-tioned above, form the central theme of this paper.

Geologic framework and depositional models of the Helvetiafjellet Formation

The study area lies in Nordenskiöld Land on Spitsber-gen, the largest island in the Svalbard archipelago, Nor-way. Here, several hundred meters of Mesozoic strata are intermittently exposed. Svalbard was located at roughly 60˚N in the Late Jurassic to Early Cretaceous (Steel & Worsley 1984; Torsvik et al. 2002), and formed part of

1000 m

500 m

250 m

Longyearbyen

Sassenfjorden

15 2

0’

78 10’ 15’’

78 22’ 08’’

15 4

0’

Janus-fjellet

N3 km

Konusen

Wimanfjellet

Kong Karls LandSPITSBERGEN

Helvetiafjellet

Forkastnings-fjellet

Hanaskogdalen

Criocerasaksla138˚

1

2

Section 1, Fig. 5

Sassenfjorden

Log trace positions

log traces

1 Km

100 m

Section 2, Fig. 6

Hanaskogdalen

Nordenskiold Land

N

S

EW

n=124

mean=138˚

Fig. 1. Map of Svalbard with the studied field area in central Spitsbergen enlarged. The profiles marked by 1 and 2 correspond to Figs. 5 and 6, respectively and illustrate the difference in scales between the two correlation panels. The rose diagram shows the palaeocurrent directions for the area, with a mean direction of 138˚.

I. Midtkandal et al. NORWEGIAN JOURNAL OF GEOLOGY

345

the Boreal basin, which also included the Sverdrup Basin, the Alaskan Basin and basins at northern Greenland. The area was a shallow epicontinental sea (Harland et al. 1984; Torsvik et al. 2002; Nagy 1970), with a low-angle, gradually deepening shelf from shoreline to basin centre. The Boreal basin was, during the Early Cretaceous, affected by tectonic and magmatic processes associ-ated with the development of a Large Igneous Province (LIP) to the north and east, with subaerial lava flows interfingering with Helvetiafjellet Formation sands on Kong Karls Land (Mørk et al. 1999; Maher 1999). This is a small group of islands about 100 km offshore eastern Spitsbergen, exposing the easternmost exposures of the formation (Fig. 1). Volcanic ashes (bentonites) have been recorded in the upper part of the Helvetiafjellet Forma-tion at Festningen, western Spitsbergen (Pers. comm. Hans Amundsen, Physics of Geological Processes, Uni-versity of Oslo) and elsewhere in Spitsbergen (Mørk et al. 1999). Dolerite sills intruded the underlying shale suc-cession (the Janusfjellet Subgroup) in eastern Spitsber-gen in the Barremian, possibly during the development of the Helvetiafjellet Formation (Miloslavskij et al. 1992). During the Barremian, the climate was humid, support-ing peat (coal) accumulation and a dinosaur population (Nøttvedt et al. 1992; Hurum et al. 2006). Temperature gradients were low, and did not change much during the Barremian stages (Fischer 1981; Ziegler et al. 1987; Nemec 1992).

The Helvetiafjellet Formation is a 12-155 m thick shallow-marine to fluvial sandstone-dominated succession which

overlies an up to 800 m thick mudstone-rich interval, the Janusfjellet Subgroup (Fig. 2). The Rurikfjellet For-mation (Parker 1967), being the upper mudstone unit in this subgroup, was deposited under open marine oxic shelf conditions. In its upper part, the Rurikfjellet For-mation generally shallows upwards through repeated sets of upward-coarsening parasequences of shore-face and delta lobe deposits (Dypvik 1980; Dypvik 1985; Dypvik et al. 1991; Mørk et al. 1999). The boundary towards the overlying Helvetiafjellet Formation has been a mat-ter of discussion and is of critical importance for the depositional system of the Helvetiafjellet Formation (see below). Stratigraphically above the Helvetiafjellet Forma-tion, the Carolinefjellet Formation (Parker 1967; Nagy 1970; Mørk et al. 1999) was deposited after a regional transgression that brought about a return to the condi-tions of an open marine shelf environment.

Several studies discuss the Helvetiafjellet Formation, its stratigraphy, facies and depositional conditions (Róžycki 1959; Parker 1967; Birkenmajer 1975; Edwards 1976; Steel 1977; Edwards 1978; Steel et al. 1978; Edwards 1979; Edwards et al. 1979; Nemec et al. 1988a, b; Grosfjeld 1992; Nemec 1992; Nøttvedt et al. 1992; Gjelberg & Steel 1995; Prestholm & Walderhaug 2000; Steel et al. 2001; Maher et al. 2002; Midtkandal 2002; Maher et al. 2004; Ahokas et al. 2005; Midtkandal et al. 2005).

The Helvetiafjellet Formation was named in a description of the Jurassic and Cretaceous stratigraphy of Spitsbergen by Parker (1967); it is this description that is currently in use (Mørk et al. 1999). Dealing with the formation in Nordenskiöld Land, Steel (1977) described and discussed the variability of sandstone deposits in deltaic settings, including a wide range of subenvironments. The local variability of sandstone bodies, and how depositional subenvironments interfinger laterally in a deltaic coast setting, was demonstrated. Steel and Worsley (1984) and Nemec (1992) presented general depositional models for the Helvetiafjellet Formation, with emphasis on the flu-vial character of the Festningen sandstone member and the variability of depositional environments that form the middle to upper parts of the formation.

The original model for the internal stratigraphy of the Helvetiafjellet Formation (Parker 1967; Nagy 1970) was of layer-cake type, whereas models advocated by Steel and Worsley (1984) and Nemec at al. (1988a, b) outlined the development of a retrogradational delta complex. Gjelberg and Steel (1995), reviewing these models, sug-gested a third depositional model for the sequence strati-graphic development of the Helvetiafjellet Formation. Their model comprises a backstepping palaeo-shoreline and an overall regressive development linked to a rise in relative sea-level throughout the temporal and spa-tial extent of the Helvetiafjellet Formation. An impor-tant feature in the model of Gjelberg and Steel (1995) is that all fluvial sandstone bodies are clastic wedges that lithostratigraphically can be linked upstream to the Fest-

Cre

tace

ou

sJu

rass

icLa

teEa

rly

Barremian

Berriasian

Hauterivian

Aptian

Albian

Valanginian

Volgian

Kimmeridgian

Oxfordian

Callovian

Ad

ven

tdal

en G

rou

p

AgardhfjelletFm

RurikfjelletFm

Open marine shelf

Regressive openmarine shelf

Fluvial and paralic

Shallow to distalmarine shelf

HelvetiafjelletFm

CarolinefjelletFm

Thickness(m)

Lithostratigraphicsubdivision

Depositionalenvironment

Lithology(shale/sst)

Age

90-

350

190-

?1200

110-

400

12-

155

Fig. 2. Svalbard’s Mesozoic succession, compiled from Mørk et al. (1999).


346

SB

FS1

FS2

E

W

20 m

log trace, Konusen W

log trace, Konusen E

FA1

FA2 FA3

FA4FA6

FA5

FA7

FA7

Fig. 3. Northern slope of Konusen, showing a typical Helvetiafjellet Formation outcrop on the southern shore of Sassenfjorden, Nordenskiöld Land. The log traces marked on either side of the cliff section correspond to the logs with the same name in Fig. 5. The lowest sandstone body is the Festningen sandstone (FA1), a fluvial braidplain deposit that rests unconformably on the regional sequence boundary.

FA7

FA1

FA3

FA5

FA7

FA1

FA1

FA1

FA2

FA3

FA4

FA6

FA5

FA5

FA5

FA4

FA2FA3

FA4FA2

15 m

20 c

m

15 m

15 m

2 m

FA4

FA4

8 m

5 m

SB

FS1

FS1

FS1

SB

Rurikfjellet Formation

Helvetiafjellet Formation

Carolinefjellet Formation

a b

c d

e f

g h

E W

E

E W

E W

E W

W

EW

EW

Fig. 4. Facies associations recorded in the Helvetiafjellet Formation (and Rurikfjel-let Formation). a) FA1; braided fluvial channel belt overlying the regional sequ-ence boundary (SB) at Konusen. Notice the sandstone bar forms that prograde towards the left (east), and the progres-sive thinning of sandstone beds towards the upper boundary of FA1. b) FA2; Interdistributary bay or lagoon with thin interbeds of FA3 (delta lobe toeset) in the upper part (at Konusen). The FA3 beds thin and pinch out towards the left (east). c) FA3; delta lobe toesets at Hanaskogda-len west. The FA3 deposits are stacked to form a 6-7 m thick sandstone-rich unit at this locality (Hanaskog 1 west). Notice the sandstone-sandstone contact to the underlying FA1 deposit. d) FA4; delta lobe front and top at Hanaskogdalen. The locality displays the conformable transition from lagoon/interdistributary bay, into delta lobe toeset and further on to delta front deposits. Notice the muds-tone wedges that separate the lower toe-set beds. e) Coastal floodplain deposits at Janusfjellet. The image shows thin-bed-ded sandstone with mud- and coal dra-pes. Photo by Kjell Ove Storvik. f) FA6; fluvial channel deposits dissecting coastal floodplain sediments at Konusen. The fluvial channels are laterally discontinu-ous and show significant thickness varia-tions. g) FA7; open marine deposits at Hanaskogdalen. The layered sandstone beds behind the person are Facies F, hum-mocky cross stratified sandstone beds at the top of the Rurikfjellet Formation. The regional sequence boundary (SB) marks the upper contact of FA7. Notice the bar forms that downlap onto the SB in FA1. h) The Wimanfjellet locality, Sassenfjor-den. The Festningen sandstone forms an easily recognizable cliff, but makes acces-sibility problematic.


347

ningen sandstone member, within the Spitsbergen domain. This was due either to a fall in relative sea-level or by coastal plain progradation within an overall backstep-ping transgressive systems tract. The model also assumed a conformable contact existed between the Helvetiafjellet Formation and the underlying Rurikfjellet Formation in the south-eastern part of Spitsbergen. Parker (1967) and later studies (Edwards 1976; Nøttvedt et al. 1992; Mørk et al. 1999; Midtkandal et al. in review) have described the lower boundary of the Helvetiafjellet Formation gen-erally as an erosional unconformity.

The Nordenskiöld Land study areaThe Nordenskiöld Land study area lies to the NE of Longyearbyen, Svalbard (Fig. 1). The outcrops included in this study are grouped into two cliff sections oriented roughly ENE along the northern side of Hanaskogdalen and along the southern shore of Sassenfjorden (Fig. 1). A moderately well exposed 1000 m long cliff section is exposed in Hanaskogdalen, distributed over 3 locali-

ties, whereas the Sassenfjorden area provides excellent exposures over a distance of 7 km, here represented by 6 localities. Accessibility is hazardous; especially in the Sas-senfjorden localities, where the majority of the cliff sec-tions cannot be visited due to the steepness of the slopes (Fig. 3). Palaeocurrent measurements of beds in the Hel-vetiafjellet Formation in this area indicate a mean flow direction of 324˚, roughly SSE. Thus, the field data allow for the construction of two sub-parallel oblique-dip sec-tions based on semi-continuous cliff exposures, and a 3D reconstruction of the depositional development in the area can be inferred from a correlation between the Sas-senfjorden and Hanaskogdalen outcrops.

Facies and facies associationsIn Nordenskiöld Land, thirteen sedimentary facies, des-ignated A to M, have been defined in the Helvetiafjellet Formation and in the adjacent parts of the under- and overlying formations, the Rurikfjellet and Carolinefjel-let formations, respectively (Table 1). These facies occur

Table 1. Facies recorded in the Helvetiafjellet Formation in Nordenskiold Land

Facies Description Grain size Interpretation

A Extrabasinal clasts (chert, quartzite, crystal-line quartz) in sandstone matrix. Matrix supported. Lower boundary erosive, grada-tional upper boundary into Facies B.

Medium to very coarse sand matrix, max. clast size 12 cm.

High energy fluvial channel base conglomerate.

B Trough cross-stratified sandstone with extrabasinal clasts at foreset basal bounda-ries and upward fining foresets. Tangential bottomset terminations. Scattered mudstone clasts.

Medium to very coarse sand. Extrabasinal clasts up to 1.5 cm in diameter.

Fluvial channel deposition as 3D dunes with sufficient energy to cause flow separation over dune crests.

C Tabular cross-stratified sandstone with extrabasinal clasts at some foreset basal boundaries. Angular foreset terminations.

Medium to coarse sand. Extra-basinal clasts up to 1 cm in diameter.

Fluvial channel deposit formed as 2D dunes. No flow separation at dune crests.

D Plane-parallel stratified sandstone Medium to coarse sand Upper flow regime fluvial channel deposition

E Massive structureless sandstone Fine and medium sand Rapid deposition from suspension in a fluvial system

F Hummocky cross- stratified sandstone Very fine to fine sand Offshore transition to open shelf sedimentation. Deposition below normal wave base.

G Asymmetrical ripple laminated sandstone Very fine to fine sand Lower flow regime deposition in unidirectionally flowing water

H Symmetrical ripple laminated sandstone, chevron type

Very fine to fine sand Deposition in shallow, standing body of water during horizontal oscillation

I Plane parallel laminated siltstone and sandstone

Silt and very fine sand Deposition from suspension in very low energy conditions

J Interbedded sandstone and mudstone Mud and very fine to fine sand Deposition in alternating low and zero energy at any water depth. Can be marine or lacustrine in origin

K Mudstone, black and grey with plane- parallel lamination

Mud Deposition from suspension in standing water

L Coal Mud Formed in areas with low sediment influx and high organic production, such as marshes or swamps

M Cross-stratified sandstone with double mud drapes

Very fine to fine sand Fluvially deposited bedforms modified by tidal currents.


348

within seven facies associations (Table 2). The facies asso-ciations FA1 to FA7, here listed according to their order of stratigraphic appearance, thus reflect roughly how the sedimentary environment changed over time (Table 2).

FA1 Braided fluvial channel beltDescription

A sandstone body, ranging from 15-25 m in thickness, forms the lowermost lithostratigraphic unit of the Hel-vetiafjellet Formation in Nordenskiöld Land. The sand-stone has traditionally been referred to the Festnin-gen sandstone member (Mørk et al. 1999). The lower boundary is clearly erosive, marking an abrupt change in lithology from mudstone (Facies K) to conglomerate (Facies A), the latter dominating the lower 1-3 m of this stratigraphic unit. The unit furthermore includes beds of facies B (trough cross-stratified sandstone), C (tabu-lar cross-stratified sandstone), D (plane-parallel strati-fied sandstone) and E (massive structureless sandstone) (Tables 1 and 2). FA1 is a complex of numerous interfin-gering multi-storey, multi-lateral lens-shaped bodies of sandstone, 2-5 m thick and several tens of meters long. Lens thicknesses vary as both upper and lower boundar-ies undulate and pinch out in all directions (Fig. 4 a). Two orders of surfaces are observed within the FA1 sand-stone unit at the Konusen locality (Fig. 3); the highest order surface dips gently with a component in the down-stream direction, and forms low-angle, large scale clino-themes that span the length of the outcrop (~1 km). The lower order surface marks boundaries between discrete bars that comprise most of the lower half of this unit. Both boundaries are sandstone-sandstone contacts, and can be identified at a distance from the outcrops. Palaeo-current directions generally deviate less than ±45˚ from a mean direction of 138˚ (Fig. 1). Vertical trends in the sandstone succession are upward fining in the lower third of the unit, and upward thinning of beds in the upper third. The uppermost beds of FA1 are generally 40-60% thinner than those in the lower half of the unit, with cor-respondingly reduced bar form sizes, both in the vertical and lateral directions.

The lateral extent of FA1 is far beyond the outcrop area of Nordenskiöld Land and has been recorded in several outcrop areas across Spitsbergen (Birkenmajer 1975; Edwards 1976; Steel et al. 1978; Edwards 1979; Nemec 1992; Maher et al. 2004; Midtkandal et al. in review).

Interpretation

Based on lithology, sedimentary structures, low palaeo-current direction divergence and absence of fine grained beds, the FA1 sandstone succession is interpreted to rep-resent a braided fluvial channel belt, as has been previ-ously documented by Nemec (1992). The lens-shaped sandstone bodies represent mid- and lateral channel bars, mainly accreting in the downstream direction, second-

arily in lateral directions (Figs. 4, 5 and 6). The undulat-ing upper and lower bar boundaries reflect how the braid bars merge into each other and coalesce in all directions due to the presence of sand everywhere on the channel floor. The highest order of surfaces that cut through the unit at a low angle probably represents bypass surfaces that developed while the unit was being formed. High influx rates of sediment coupled with limited vertical accommodation space is thought to have caused inter-mittent bypass and erosion as sediments were being transported into the lowest topographic areas available at any given time. The second order of intra-sandstone sur-faces represent braid bar boundaries, formed in episodes of active downstream and/or lateral accretion followed by scouring, when the local channel thalweg shifted, thus changing the main direction of sediment transport. The complete absence of mudstone beds suggests high channel mobility, making peat accumulation impossible (Nemec 1992). Upward thinning of beds and bedforms in the upper third of FA1 is here ascribed to an overall waning of current energy. Channel abandonment due to lateral migration, reduction of runoff rates, or increased rate of rise in base level during deposition are possible explanations for the upward thinning of beds.

The lower boundary of FA1 represents a significant regional basinward shift in facies, from open marine mud accumulation in the underlying Rurikfjellet Forma-tion, to subaerial erosion, followed by fluvial deposition. This boundary surface is suggested to represent a type 1 sequence boundary in the sense of Posamentier & Vail (1988), Van Wagoner et al. (1988) and Van Wagoner et al. 1990).

FA2 Interdistributary bay or lagoonDescription

FA2 overlies FA1, with an abrupt lithological change from Facies C (tabular cross stratified sandstone) to Facies K (mudstone) (Table 1). The boundary appears to be hori-zontal with no visible signs of erosion. FA2 is dominated by Facies K (Fig. 4 b), and forms a 5-15 m thick, dark-coloured stratigraphic unit that extends laterally beyond the outcrop area of Nordenskiöld Land. Accurate thick-ness measurements of this unit are difficult to obtain due the unconsolidated nature of the mudstone beds, and scree-covered nature of the outcrops (Fig. 3).

At various stratigraphic levels within this mudstone suc-cession, sandstone intercalations up to 1.5 m thick occur, representing four facies: G (asymmetrical ripple lami-nated sandstone), H (symmetrical ripple laminated sand-stone), I (plane-parallel laminated silt- and sandstone) and J (interbedded sandstone and mudstone) (Tables 1 and 2). Individual beds are 10-50 cm thick and appear vertically stacked. The beds can commonly be traced along cliff sections for hundreds of meters, but correla-tion between log traces is difficult due to poor exposures between localities. At Konusen, (Fig. 3) the beds decrease


349

in thickness eastwards. FA2 varies internally and contains both very finely lami-nated, pure mudstone and intervals with a coarser clas-tic component. In a sample taken from finely laminated dark mudstone at the Konusen site, the following three foraminiferal taxa have been documented: Trochammina sp., Ammobaculites sp. and Verneuilinoides sp.

Interpretation

The low diversity, agglutinated nature of the foraminiferal assemblage occurring in the FA2 mudstone indicates a restricted marine environment. This means that condi-tions diverged from those of a normal marine shelf. The small size of the observed specimens supports this inter-pretation. The main restricting factor was apparently reduced salinity in an interdistributary bay or lagoon. In addition, moderately hypoxic conditions might have occurred in the lower part of the water column owing to salinity-induced stratification in a shallow depositional area with high organic supply. The finely laminated dark mudstone lithology of the FA2 beds is in accordance with such stagnant bottom conditions

Further support for this interpretation is provided by comparisons with modern and ancient analogues, although comparisons with modern faunas are compli-cated owing to major evolutionary changes in foraminif-eral assemblages during the Late Cretaceous and Tertiary. The dominant species in the FA2 sample is Trochammina sp., which is similar to small-sized representatives of Tro-chammina previously reported as dominant in Jurassic and Paleogene mudstones and indicative of a hyposa-line, delta-influenced environments (Nagy et al. 1990; Nagy 2005). The same genus is also dominant in several modern brackish lagoons, with a salinity-stratified water column where hypoxic bottom conditions develop sea-sonally (Takata et al. 2005; Takata et al. 2006). Ammo-baculites is reported as common to dominant in Jurassic brackish prodelta deposits (Nagy & Seidenkrantz 2003). In modern faunas this genus is often associated with low salinity estuarine conditions (Busas 1974; Murray 1991). Verneuilinoides and related serial forms are reported to be as common in Paleogene lagoonal sections (Nagy 2005), and in modern hyposaline waters (Alve 1990).

From these micropaleontological data it is apparent that FA2 was deposited in a restricted, shallow, marginal marine environment such as an interdistributary bay or lagoon. Palaeo water depths were shallow, as in modern lagoons or interdistributary bays, allowing variations in clastic detritus to mix with the mud. The sandstone beds within FA2 may represent a range of coastal shallow marine deposits. Some or all of these beds probably form distal parts of the succeeding facies association, FA3, delta lobe toe set (see below). During periods of increased sed-imentation rates, deltaic sands may have been deposited further out beyond the shoreline than is the case under

average conditions. Alternatively, a portion of the sand-stones may have formed as sand spits or as a result of any combination of processes that form sand drapes in shallow marine environments, such as storm surges and neap/spring tides.

An abrupt shift from a coarse grained fluvial (FA1) to a marginal marine environment indicates flooding of the area. The lower boundary of this facies association is henceforth termed FS1 (Figs. 3, 4, 5, and 6), and repre-sents a potentially regional significant flooding surface. In this low-gradient setting, this flooding is inferred to have been rapid and widespread, thus making this sur-face applicable as a datum plane for correlation, as is dis-cussed below.

FA3 Delta lobe toe setDescription

In the western, proximal, Hanaskogdalen localities (Fig. 6), a 6-7 m thick interval of thin-bedded (5-15 cm) sandstone directly overlies the lowest fluvial unit in the Helvetiafjellet Formation (FA1). The entire interval con-sists of Facies D (plane parallel stratified sandstone) and J (interbedded sandstone and mudstone) (Tables 1 and 2). The lower boundary is lithologically sharp, but with-out any observed erosional features. The sandstone beds are stacked conformably with a thin (mm) mudstone drape marking bed boundaries (Fig. 4 c). Bed thickness decreases slightly from 10-15 cm upwards to about 5-10 cm, and grain size is reduced from fine to very fine sand-stone in the upper 2 m of the unit. The western, proxi-mal, continuation of FA3 is located beyond the outcrop area in Hanaskogdalen, and is thus not exposed here for observation. The eastern, distal extension of FA3 is also covered but at the next log trace downstream from the western Hanaskogdalen localities (Hanaskogdalen 1 West), this interval is replaced by a mudstone dominated succession (FA2, interdistributary bay or lagoon) of cor-responding thickness. The boundary to the succeeding stratigraphic unit, FA4 (described below), is an abrupt shift into Facies C (tabular cross stratified sandstone), with no visible sign of erosion.

Similar sandstone successions are recorded within the stratigraphic interval marked by FA2 in the Sassenfjor-den area, and may correlate with this unit (see discussion under FA2, above).

Interpretation

FA3 is interpreted to represent delta toe set deposits. The distal facies transition into FA2 mudstone facies sug-gests a shallow water hyposaline environment where thin sheets of sand at the front of a delta lobe pinch out basin-wards (east). The lower boundary of FA3 is identified as the same flooding surface, FS1, as previously mentioned (Figs. 3, 4, 5 and 6).


350

FA4 Delta lobe front and topDescription

FA4 is a succession of variable thickness (2-9 m) and is composed of Facies C (tabular cross-stratified sand-stone), D (plane-parallel stratified sandstone), E (mas-sive structureless sandstone), G (asymmetrical ripple-laminated sandstone), H (symmetrical ripple-laminated sandstone), I (plane-parallel laminated silt- and sand-stone), and J (interbedded sandstone and mudstone), K (mudstone), and L coal (Tables 1 and 2). The lower boundary of FA2 is conformable, recorded as an upward coarsening from FA4 mudstone into fine sand (facies D and E) over an interval of less than 1 m. In the entire out-crop area, the unit forms a sandstone body of relatively constant thickness that exceeds the aerial limits of out-crop included in this study.

A series of eastward dipping clinothemes is recorded in outcrop sections that are sub-parallel to the depositional dip (SSE) (Fig. 4 d). Here, beds can be traced downwards to a position where they tangentially meet the lower boundary, at which point the bed thickness is zero. These pinch-outs appear as thin sandstone wedges protruding into the FA2 mudstone below. Correspondingly, mud-stone drapes mark the boundaries between individual clinothemes with a maximum thickness at the base (2-3 cm) and tapering upwards into a single mud drape hori-zon, generally within less than 2 m. Up to 50 cm thick mudstone beds separate clinothemes at a few places. Bur-rows of unidentified type are recorded in places at the lower boundary of FA4.

Upward coarsening motifs characterize the FA4 depos-its exposed along outcrops orientated as vertical sections perpendicular or sub-perpendicular to the depositional dip. Coal beds (5 cm thick) along with plant fragments mark sandstone bed boundaries in the upper 2 m of FA4. Ornithopod dinosaur tracks (Pers. comm. J. H. Hurum, Natural History Museum, University of Oslo, Norway) embedded in Facies J were discovered in Hanaskogdalen, near the top of FA4. Planolites and Anchorichnus (?) bur-rows are observed on Facies H bed surfaces. At the Konusen locality (Fig. 1), relatively thin (maxi-mum 30-50 cm thick) sandstone beds of Facies G, H, I and J appear roughly in the middle of FA2, preceded and succeeded by 5-10 m of mudstone. The total thickness is about 2 m in the western Sassenfjorden area, and pinches out to the west of Criocerasaksla. The upper bound-ary of FA4 is a conformable transition into FA5, and is described further below.

Interpretation

The deposits described above are interpreted as repre-senting a prograding delta lobe, as illustrated in Fig. 7-5, in accordance with the delta lobe interpretation illus-

trated in Fig. 11, p. 581 in Gjelberg & Steel (1995).The basinward dipping clinothemes and sandstone pinch-out geometries reflect sandstone deposition into stand-ing water. The presence of coal seams and dinosaur footprints indicate deposition of the beds in which they occur as subaerial or in very shallow water coastal envi-ronments. This suggests they were formed in the most proximal part of the delta, as topset deposits. At pres-ent, it is even possible that the bioturbation occurring in these is of lacustrine origin. Palaeowater depth did prob-ably not exceed 9 m (the maximum sandstone thickness) for this facies association. The mud drapes on each cli-notheme unit may reflect tidal influence during depo-sition. Restricted, shallow bays, as is suggested for FA2, are prone to amplify tidal currents, hence increasing the probability of a tidal modulation of the prograding delta lobe geometry and facies. The laterally extensive deltaic sandstone bodies suggest the delta lobes developed as elongated features extending into the basin, as seen in modern “bird-foot” deltas, an interpretation supported by Steel (1977).

FA5 Coastal floodplainDescription

FA5 is largely dominated by Facies K (mudstone), but also includes Facies E (massive structureless sandstone), G (asymmetrical ripple laminated sandstone), H (sym-metrical ripple-laminated sandstone), I (plane-parallel laminated silt- and sandstone), J (interbedded sandstone and mudstone) and L (coal) (Tables 1 and 2). The lower boundary towards FA4 is conformable, and the absence of basinward dipping clinothemes and increased internal heterogeneity are the main features that distinguish FA5 from the delta top deposits (FA4). The upper bound-ary of FA5 is marked either by an erosive contact to an overlying fluvial channel, or as a conformable transition into overlying open marine deposits. This conformable boundary is generally a mudstone to mudstone contact, and is by nature difficult to pinpoint. The upper, fine grained sediments of the Helvetiafjellet Formation are commonly covered by talus, which is a further compli-cating factor in determining the exact position of this boundary. FA5 is characterized by mudstone and coal intervals with rootlet horizons and desiccation cracks, plant fragments and palaeocurrent directions with no apparent directional trend. Sandstone beds of facies E, G and H range from 0.1 to 2 m in thickness (Fig. 4 e), with both erosive and non-erosive lower boundaries, and can be traced up to 150 m laterally. Sandstone facies I, J and L occur randomly within this stratigraphic unit, and cannot as a rule be traced laterally beyond a few meters. FA5 units on the whole are generally laterally persistent beyond the outcrop limits in Nordenskiöld Land, and occur at several stratigraphic levels, separated by FA6 (described below) (Figs 5 and 6). Intervals which are obscured by talus are inferred to consist of fine grained mudstone. No further details are forthcoming.


351

Interpretation

FA5 strata are interpreted to have formed on a coastal floodplain, laterally adjacent to the main sediment trans-port channels (Fig 6-5). It is suggested here that they have conformably succeeded fluvial channels (in the case of lateral migration and abandonment) and delta top envi-ronments (in the case of progradation), as has also been discussed by Gjelberg & Steel (1995). As a result of fur-ther, renewed lateral migration on the coastal plain, flu-vial (and distributary-) channels eroded the upper part of these deposits before depositing fluvial channel sedi-ments. This is a highly dynamic, and thus varied, depo-sitional environment characterized by pools of stand-ing water, vegetation growth, creeks, crevasse splays and channels, and wet marshes. Low-gradient coastal plain environments are easily flooded, and basinward of the bay-line position (cf. Posamentier & Vail 1988), the plain may be intermittently affected by spring tides or storm surges. It is thus possible that a portion of the mudstone content in FA5 may be marine or brackish in origin, and possibly a tidal flat during the final stages of its develop-ment.

FA6 Fluvial channel Description

Several 2-10 m thick, very clean sandstone units com-posed of Facies B (trough cross-stratified sandstone), C (tabular cross-stratified sandstone), D (plane paral-lel stratified sandstone), E (massive structureless sand-stone) and M (cross-stratified sandstone with double mud drapes) comprise FA6 (Table 1). FA6 units occur at several stratigraphic levels in the upper part of the Hel-vetiafjellet Formation in Nordenskiöld Land; their distri-bution is shown in Figs. 5 and 6. The lowest unit is the most laterally extensive, extending beyond the outcrop area included in this study. Above this unit, in the Sas-senfjorden section, one or both lateral terminations of individual FA6 units are recorded within the study area. In the significantly shorter Hanaskogdalen section, all of the FA6 units extend beyond the outcrop area.

The lower part of the FA6 units displays a similar inter-nal organization as FA1, being mainly composed of trough cross-stratified sandstone, but without extraba-sinal conglomerate clasts. The lowest order of internal surfaces described in FA1 is also observed within FA6, separating individual bar forms up to 3 m in thickness. The bar-bounding surfaces are less clearly defined com-pared to FA1. The lower boundaries of FA6 units are generally abrupt, with an erosive contact to underly-ing coastal floodplain deposits (Fig. 4 f). Palaeocurrent directions are consistent with FA1, having a mean direc-tion of roughly 138˚ (Fig. 1). Facies M is only recorded in the highest stratigraphic levels of FA6 in Hanaskogdalen. These observations are made on very poor outcrops, and can thus not be confidently placed on a log trace. Corre-lation and stratigraphic context for the highest FA6 units

in both Hanaskogdalen and Sassenfjorden are problem-atic due to deteriorating outcrop quality with increasing stratigraphic height, and a reduction of FA6 unit thick-nesses.

Interpretation

FA6 is interpreted to have been formed as fluvial channel infill sediments. The basal erosive boundaries and high energy facies assemblage reflect channelized transport of sand across the FA5 coastal plain (Fig. 7-6). The trans-port direction was to the SSE, suggesting that the palaeo-coastline was oriented roughly WSW-NNE during this phase of development and largely unaltered since deposi-tion of FA1. The apparent lenticular shape of the upper fluvial channel sandstone units, seen in sections oriented semi-perpendicularly to the regional palaeocurrent direction (Sassenfjorden, Fig. 5), may imply that these sandstone bodies represent narrow distributary fluvial channels on a distal part of the coastal plain. The double mud drapes recorded in the upper stratigraphic levels of FA 6 reflect tidal influence with an increase in current energy, possibly heralding the transgressive event that followed. The abrupt upper boundary of FA 6 may be the result of a rapid rise in relative sea-level, a local shift in depositional environment, or a combination of both factors. This upper boundary is identified as a flooding surface, as it marks an abrupt lithological change, but the palaeo-water depths above this boundary have not been estimated using biostratigraphic methods. Hence, the flooding surface is in reality the only identifiable surface within a flooding interval that is probably represented by a few metres (?3-10 m) of siltstone and mudstone.

FA7 Marine shelfDescription

The Rurikfjellet Formation and the Carolinefjellet For-mation, located respectively below and above the Helve-tiafjellet Formation, are here both included in FA7, based on their mudstone dominated lithological composition, and because they represent the adjacent formations of the study object (Fig. 4 g). The main facies in FA7 is Facies K (mudstone), but Facies E (massive structureless sandstone), F (hummocky cross-stratified sandstone), G (asymmetrical ripple- laminated sandstone) and I (plane-parallel laminated silt and sandstone) are also recorded in laterally extensive beds up to 1 m in thickness (Tables 1 and 2). The sandstone beds contain more mud than the previously described sandstone facies, and are greyish brown in colour. Bioturbation is common and the bur-rows Skolithos and Diplocraterion are recorded in Facies E and G. The Rurikfjellet Formation is part of an up to 400 m thick mudstone succession (Fig. 2). Several upward coarsening shale-sandstone successions are recorded in the upper part of this formation which terminates at the erosional regional unconformity mentioned above, beneath the Helvetiafjellet Formation.


352

The Carolinefjellet Formation is an up to ?1200 m thick mudstone succession with a varying number of sand-stone beds (Parker 1967; Nagy 1970) (Fig. 2). Its lower boundary is conformable with the upper parts of coastal plain mudstone units of the Helvetiafjellet Formation, and is thus difficult to define where the exposures are poor.

Interpretation

The Rurikfjellet Formation was deposited as a regres-sive, upward shallowing shelf succession on the Boreal epicontinental shelf of Spitsbergen (Dypvik 1985; Mørk et al. 1999). The Carolinefjellet Formation, capping the Helvetiafjellet Formation, was formed in response to the transgression that finally terminated the paralic depo-sitional environment of the Helvetiafjellet Formation (Nagy 1970).

Table 2. Facies associations recorded in Nordenskiöld Land.

Formation Member Architectural element

Depositional Environment

Facies included (table 1)

Sequence strati-graphic surface

Carolinefjellet Formation FA 7 Marine shelf K, E, F, G, I FS 2FA 6 Fluvial channel B, C, D, E

FA 5 Coastal floodplain K, E, G, H, I, J, L

Mid- to upper Helvetiafjellet Formation

FA 4 Delta lobe front and top

C, D, E, G, H, I, J, K, L

Helvetiafjellet Formation FA 3 Delta lobe toeset D, J

FA 2 Interdistributary bay or lagoon

K, G, H, I, J

Festningen sandstone member

FA 1 Braided fluvial channel belt

A, B, C, D, E FS 1

Rurikfjellet Formation FA 7 Marine shelf K, E, F, G, I SB

Forkastningsfjellet Janusfjellet Konusen W Konusen E Criocerasaksla Wimanfjellet

2500 m 1000 m 600 m 1100 m 2300 m

SBSB

FS1 FS1

FS2FS2

1 Km10 m

12

3

4

5

6

7

W EFluvial channel sandstone

Coastal plain mudstone

Delta sandstone

Paralic sandstone

palaeocurrent direction

plant fossilcoal bedbioturbation

Dinosaurfootprint

Restricted marine mudstone C

Open marine mudstone

Shallow marine sandstone

Fig. 5. Correlation panel from the southern shore of Sassenfjorden. The numbered stratigraphic intervals on the right correspond to the deposi-tional environments in Fig. 7.


353

Log correlationSedimentological log traces were recorded at a total of 11 localities; 5 in Hanaskogdalen, and 6 along the southern shore of Sassenfjorden. A correlation panel has been con-structed for each of the two main semi-continuous cliff sections, using identifiable sediment bounding surfaces acting as links between log traces (Figs. 5 and 6).

Across the entire outcrop area in Nordenskiöld Land, the lower boundary of the Helvetiafjellet Formation is an unconformity due to erosion(Fig. 4 g). As described above, it marks an abrupt shift from a marine shelf environment (FA 7) into a fluvial channel environment (FA1). As one of the most recognizable sequence strati-graphic surfaces within the Helvetiafjellet Formation, it is nevertheless a diachronous surface, and is thus not used as the main datum plane for log correlation in this study. The upper boundary of the Helvetiafjellet Forma-tion is not easily identifiable in the field, and it is conse-quently impractical to use it as a chronostratigraphical or lithostratigraphical correlation surface.

The flooding surface that marks the upper boundary of FA 1 (FS1, Table 1), is used as the main datum plane, as it represents a practically isochronous surface with only an insignificant amount of associated topographic relief. Correlation between the panels of two cliff sections

involves greater uncertainties, partly because of the sig-nificantly shorter cliff section along the northern side of Hanaskogdalen (1500 m) compared to the section along the southern shore of Sassenfjorden (7000 m). Thus, three-dimensional control of the depositional environ-ments that developed in the area is somewhat limited, but can be inferred within reason. Figures 5 and 6 show the constructed correlation panels for the Sassenfjorden and Hanaskogdalen cliff sections, respectively. Notice the difference in scale between the two panels.

DiscussionDevelopment of depositional environmentsFall in relative sea-level; development of a regional sequ-ence boundary

Open marine shelf conditions prevailed during depo-sition of the Rurikfjellet Formation (Edwards 1976; Dypvik 1980; Dypvik 1985; Mørk et al. 1999; Århus 1991) (Fig. 7-1). Upward coarsening successions in the upper part of the Rurikfjellet Formation reflect a regres-sive and shallowing shelf. Forced regressive features, such as detached shorelines or sharp-based shoreline succes-sions are not documented at the boundary between the Rurikfjellet and Helvetiafjellet formations in Norden-skiöld Land or anywhere else in Spitsbergen, suggesting

C

Hanaskogdalen 2 EastHanaskogdalen 2 West

Hanaskogdalen 1 East Hanaskogdalen 1 WestHanaskog 3 West

5 m100 m

W E

12

3

4

5

6

7FS2

FS1

SB

Fig. 6. Correlation panel from Hanaskogdalen. Notice the difference in scale compared to Fig. 5. See figure 5 for legend. The numbered stratigraphic intervals on the right correspond to the depositional environments in Fig. 7.


354

that the regression marked by the boundary between these two formations was non-accretionary (cf. Helland-Hansen & Gjelberg 1994). Tectonic uplift to the north-west caused a fall in relative sea-level across an area larger than the gross outcrop area on Spitsbergen (Nøttvedt et al. 1992). During this regional fall in relative sea-level, an unknown amount of the Rurikfjellet Formation was eroded as denudation patterns were established (Fig. 7-2). The erosion resulted in a palaeotopography of mod-erate relief, truncating the Rurikfjellet Formation (Figs. 5 and 6) variably.

Early rise in relative sea-level; deposition of a fluvial braidplain

Deposition of the Helvetiafjellet Formation commenced at the onset of a rise in relative sea-level that led to base level elevation and the generation of accommodation space. Facies association 1 (FA1) is the stratigraphically lowest depositional unit in the Helvetiafjellet Formation,

and represents a widespread system of braided fluvial channel infill (Fig. 7-3), lithostratigraphically termed the Festningen sandstone member (Mørk et al. 1999). At the onset of the rise in relative sea-level, vertical accom-modation space was limited, causing the braided stream channels to migrate laterally during frequent avulsions, thus forming a sheet-like, multilateral sandstone body. Consequently, the entire emerged shelf area was filled with braidplain sand. High channel mobility and erosive capacity removed any fine-grained overbank facies, and major underlying irregularities were smoothed out as the fluvial channel courses were continuously re-directed in search of steeper gradients, as described for this type of depositional setting by Holbrook (1996). The result was a laterally persistent, high net/gross sandstone body with limited thickness variations. The fluvial sandstone blanket was likely deposited in a backstepping way in response to rising sea-level and base level, as also suggested by Gjelberg and Steel (1995). The

1: Upper Rurikfjellet Formation

2: Erosional stage

3: Earliest rise in relative sea-level

4-6: Aggadational paralic stage

7: Flooding stage

Aggradation of paralic environments:

4

5

6

Fig. 7. Successive development of depositional environments in the Nördenskiold Land area. The numbered timeslices correspond to the stratigraphic intervals marked in Figs. 5 and 6 and illustrate a possible depositional scenario for the different stages of Helvetiafjellet Formation development. See discussion in text for details.


355

Festningen sandstone member is thus a typical diachron-ous depositional unit. However, the lateral extent of the fluvial sandstone is remarkable, and the backstepping mode may have been very rapid, with numerous bypass surfaces contained within the unit. The river systems on the emerged shelf may have acted as important routes for sediment transport, depositing fine grained components in the deeper part of the epicontinental basin in the adja-cent Barents Sea region to the southeast. In this part of the basin, a shelf edge may have developed, but this has not yet been recognized in any Spitsbergen outcrop.

Continued rise in relative sea-level; aggradation of coastal plain environments

The entire Nordenskiöld Land area was flooded after deposition of the Festningen sandstone member and the upper part of the Helvetiafjellet Formation, previously named the Glitrefjellet Member (Parker 1967; Mørk et al. 1999), was deposited. The stratigraphic interval is char-acterized by its heterolithic composition and frequent facies changes. The flooding was caused by increase in the rate of relative sea-level rise, a reduction in the rate of sediment supply, or a combination of the two. Restricted shallow marine conditions were established, such as interdistributary bays and lagoons in a setting where small, continuously changing deltas developed along the shoreline. Infrequent sandstone intercalations within the mudstone-dominated succession reflect pulses of increased sediment supply. The delta toe sand beds (FA3) that prograded from the west into the otherwise mud-dominated depositional environment (Fig. 7-4) indicate an approaching palaeo-shoreline located somewhere to the west of Nordenskiöld Land, here inferred to have been oriented roughly NNE-SSW, judging from palaeo-current data that show dominant flow towards the SSE, and sandstone pinch-out directions towards the east or southeast.

The deltaic deposits that succeed FA3 reflect a renewed shoreline progradation in this part of the basin (Fig. 7-5). The deltaic environment was modified by tidal and wave currents. Continued regression shifted the shore-line further to the southeast, establishing coastal plain conditions in Nordenskiöld Land (FA5). This terrestrial environment was characterized by relatively shallow and narrow distributary rivers dissecting a wet, marine-influenced coastal floodplain, where organic matter and fine-grained overbank sediments accumulated (Fig. 7-6). The final stages of fluvial deposition in the area became increasingly influenced by tidal currents as the rise in rel-ative sea-level increased.

Increased rate of rise in relative sea-level; transgression of the coastal plain

As the rate of rise in relative sea-level increased, a trans-gression flooded the entire area (forming FS2), initiating

deposition of the Carolinefjellet Formation, as is reflected in the succession that represents the transition from coastal plain to shallow marine conditions. The deposi-tional environment returned to a progressively deepen-ing, open marine shelf condition (Fig. 7-7), the trans-gression culminating in the deposition of fine-grained sediments represented by the Inkjegla Member of this formation (Parker 1967; Nagy 1970).

Basin physiography as a factor of depo-sitional controlThe epicontinental Boreal basin had a very low deposi-tional gradient, as is reflected in the predominantly mud-stone deposits contained within the Rurikfjellet Forma-tion. Gjelberg & Steel (1995) estimated the gradient to have been about 0.09˚ in the Spitsbergen domain dur-ing deposition of the Helvetiafjellet Formation. Despite the very low-gradient depositional dip, river compe-tency during lowstand was sufficient to transport extra-basinal clasts up to 10 cm in diameter into the basin in Nordenskiöld Land, a minimum distance of 60 km from the hinterland in the west or northwest. In this setting, without a marked gradient increase represented by a shelf break, a significant fall in relative sea-level simply resulted in a repositioning of the facies belts in relation to the new shoreline position. Depending on the gradi-ent and elevation of a basin’s equilibrium profile, sedi-ments may or may not accumulate during forced regres-sion (Helland-Hansen & Martinsen 1996; Posamentier 2001). Independently of whether forced regressive sedi-ments accumulated or not, the basin floor gradient was the most important factor in determining distance and rate of regression; the slope gradient controls how far into the basin the shoreline is shifted following a fall in relative sea-level (Posamentier & Morris 2000). The Hel-vetiafjellet Formation shoreline was shifted to a position southeast of present day Spitsbergen, without experienc-ing any abrupt increase in basin floor gradient. On an emerged shelf, river incision takes place in response to increase in slope gradient relative to the flu-vial equilibrium profile, along which the river path devel-ops during a fall in relative sea-level (Schumm 1993). If the sea-level remains above a shelf edge, the gradient of the shelf must be greater than that of the coastal plain for incision to take place (Talling 1998). Rivers may not incise at all if they enter a broad, low-angle shelf which is merely an extension of the equilibrium profile of the alluvial and coastal plains (Woolfe et al. 1998).

Because of the factors discussed above, it is suggested here that distinct incised valleys may have formed dur-ing the erosion of the upper Rurikfjellet Formation in areas upstream of Nordenskiöld Land, where the depo-sitional gradient is believed to have been steeper, as also suggested by Nøttvedt et al. (1992), Gjelberg & Steel (1995) and Midtkandal et al. (in review). The incised val-leys widened down-slope to the southeast and coalesced


356

into a broad erosional plain, where several kilometres of broad, shallow stretches were bypassed by fluvial systems. This plain of fluvial entrenchment is thought to have formed an incised valley complex, several hundred of kilometres wide, like that delineated by the Pleistocene maximum sea-level fall in the East China Sea (Wellner & Bartek 2003).

Paralic sedimentation on a low-gradient shelf ramp: critical variablesParalic environments are generally highly susceptible to any changes in factors controlling depositional facies and sandstone body architecture (e.g. Emery & Myers, 1996). The transgressive systems tract is in the present Nor-denskiöld Land example, characterised by an overall aggradational stacking pattern of paralic sediments, thus showing that the rate of sediment supply closely matched the rate of increase in accommodation space during this time interval (Fig. 8). Low-amplitude, high-frequency changes in relative sea-level would, in such a very low-sloping ramp shelf setting, cause very rapid and laterally extensive facies changes during a rise, as well as during a fall in sea-level. The mechanisms that controlled facies changes within the upper, heterolithic part of the Helve-tiafjellet Formation in Nordenskiöld Land may have been avulsion events or high-frequency sea-level changes. This is interpreted as a type of accretionary transgressive sys-tems tract, a depositional setting that according to Hel-land-Hansen & Gjelberg (1994, p. 43) requires “a subtle balance between rate of sediment supply and rate of sea-level rise”.

Lateral and downstream facies changes during intervals between high-frequent changes in relative sea-level are suggested to be the result of highly dynamic autogenic sediment dispersal processes. Limited vertical accom-modation space coupled with a high clastic input to the coastal areas gave rise to frequent avulsions (cf. Jones & Schumm 1999), which in turn altered the positioning of local depocentres. Combined with irregular coastal mor-phology and variations in wave and tidal energy, highly heterogenic and irregular sediment stacking patterns developed.

The LIP-volcanism in the Boreal realm and sill emplace-ments in eastern Spitsbergen may have influenced dep-ositional processes and the resulting stacking patterns tectonically (Midtkandal et al. in review). No faults that might have been related to Early Cretaceous tectonics and volcanism have been recorded within the Norden-skiöld Land area; however, the influence of such allogenic factors on sedimentation patterns can not be dismissed for the Nordenskiöld Land area either.

The depositional model of Gjelberg & Steel (1995) for the Helvetiafjellet Formation within the Spitsbergen domain of the Boreal basin suggests retrogradation of

the shoreline towards the northwest, as a response to transgression, intermittently broken and reset by high-frequent, low-amplitude changes in sea-level. Their model includes fluvial sandstone wedges linked to the basal Festningen sandstone member, formed in response to repeated high-frequent falls in sea-level. The present data set does not support this model for Spitsbergen, as shown by the overall aggradational architectural style of paralic facies associations in the upper part of the Hel-vetiafjellet Formation. Though the present model for the Helvetiafjellet Formation outcrops approaches the old “layer-cake model” reviewed by Gjelberg & Steel (1995), it is founded on the dynamic behaviour of a very low-sloping shelf, and sequence stratigraphical principles of aggrading facies belt associations during the fall and rise in relative sea-level.

Conclusions• The Barremian Helvetiafjellet Formation in Norden-

skiöld Land, Spitsbergen, reveals an internal stratigra-phy and facies distribution formed in response to a full cycle of fall and rise in sea-level on a very low-gradient ramp shelf of the Boreal epicontinental basin.

• During a fall in sea-level fluvial erosion took place,

open marine

open marine

braidplain

deltaprogradation

coastalplain

?

?

Aggradation

Aggradation

Aggradation/retrogradation

Progradation

Retrogradation/progradation

A/S

1<1 >1

FS1

FS2

Hel

veti

afje

llet F

m

Caroline-fjellet

Fm

Rurik-fjellet

Fm

SB

Architecturalstyle

Depositionalenvironment

Fig. 8. Outline of architectural trends linked to the A/S ratio during deposition of the Helvetiafjellet Formation in Nordenskiold Land. The low-frequency, high amplitude A/S trends are relatively clear compared to the high- frequency, low amplitude variations. The interplay between allogenic and autogenic factors is the main source of uncertainty.


357

truncating the emerged shelf succession forming the top of the open-marine Rurikfjellet Formation, and possibly forming incised valleys in the proximal parts of the emerged shelf. A wide plain of fluvial scouring in its distal part, may be part of a very broad incised valley complex.

• The basal, subaerially formed sequence boundary acted as a bypass surface for fluvial coarse-clastic sediment transport during lowstand, in accordance with the opinion that fluvial systems on low-gradient emerged shelves have the capacity and competency to transport large sand volumes into the deeper part of epiconti-nental basins.

• Due to the early rise in sea-level, fluvial sediments started to accumulate in a backstepping pattern, leav-ing behind a sheet-like fluvial sandstone body covering the whole sequence boundary.

• A transgressive systems tract, in the middle and upper parts of the Helvetiafjellet Formation, developed with a dominantly aggradational stacking pattern of paralic facies associations, including restricted bay, delta lobes, lagoons, distributary rivers and coastal plain deposits. The aggradational style was controlled by an overall balance between rate of accommodation versus rate of sediment influx, minor high-frequent sea-level fluctua-tions as well as autogenic variables such as differential compaction, stream and delta lobe avulsions, and a dynamically changing coastal morphology.

• An apparent “layer-cake” stratigraphy with very gen-tly sloping shelf ramps is the result of rapid and lateral changes in facies belt positions, brought about by mod-erate, high-frequent fluctuations in relative sea-level, combined with variations in factors such as A/S, differ-ential subsidence, delta lobe shifts and river avulsion.

Acknowledgement: The authors thank Atle Mørk and Paul Atle Midt-kandal for insightful comments and reviews. Reviews by Tine Lærdal and Morten Smelror, appointed by the Norwegian Journal of Geology, were very helpful in improving the clarity and contents of the manu-script. The University Centre of Svalbard (UNIS) is gratefully acknow-ledged for generous support in preparations for field work. Espen Tal-laksen is also thanked for field assistance.

ReferencesAhokas, J.M., Midtkandal, I. & Nystuen, J.P., 2005. Depositional envi-

ronment, processes and sequence stratigraphy of the paralic Hel-vetiafjellet Formation in Ullaberget, Southern Spitsbergen. In: Nakrem, H.A. (ed.): Abstracts and Proceedings of the Geological Society of Norway, 1.

Allen, P.A. & Allen, J.R., 2005. Basin Analysis: Principles and Applica-tions. Blackwell Publishing, Oxford, 549 pp.

Alve, E., 1990. Variations in estuarine foraminiferal biofacies with diminishing oxygen conditions in Drammensfjord, SE Norway. In: Hemleben, C., Kaminski, M.A., Kuhnt, W. and Scott, D.B. (eds.): Palaeoecology, Biostratigraphy, Palaeoceanography, and Taxonomy of Agglutinated Foraminifera, 661-694. Kluwer Academic Publishers, Dordrecht.

Birkenmajer, K., 1975. Jurassic and Lower Cretaceous sedimentary for-mations of SW Torell Land, Spitsbergen. Studia Geologica Polonika, 44, 7-43.

Busas, M.A., 1974. Vertical distribution of Ammobaculites in the Rhode River, Maryland. Journal of Foraminiferal Reseach, 4, 144-147.

Carey, J.S., Swift, D.J.P., Steckler, M., Reed, C.R. & Niedoroda, A.W., 1999. High-resolution sequence stratigraphic modeling 2: Effects of sedimentation processes. In: Harbaugh, J.W., Watney, W.L., Ran-key, E.C., Slingerland, R., Goldstein, R.H. and Franseen, E.K. (eds.): Numerical Experiments in Stratigraphy: Recent Advances in Strati-graphic and Sedimentologic Computer Simulations, 151-164. SEPM Special Publication 62.

Dypvik, H., 1980. The sedimentology of the Janusfjellet Formation, central Spitsbergen (Sassenfjorden and Agardhfjellet areas). Norsk Polarinstitutt Skrifter, 172, 97-134.

Dypvik, H., 1985. Jurassic and Cretaceous black shales of the Janusfjel-let Formation, Svalbard, Norway. In: Hesse, R. (ed.): Sedimentology of siltstone and mudstone, 235-248. Sedimentary Geology.

Dypvik, H., Nagy, J., Eikeland, T.A., Backer, O.K. & Johansen, H., 1991. Depositional conditions of the Bathonian to Hauterivian Janusfjel-let Subgroup, Spitsbergen. Sedimentary Geology, 72(1-2), 55-78.

Edwards, M.B., 1976. Depositional environments in Lower Cretaceous regressive sediments, Kikutodden, Sørkapp Land, Svalbard. Norsk Polarinstitutt Årbok 1974, 35-50.

Edwards, M.B., 1978. A regional survey of composition, provenance and diagenesis of sandstones in the Lower Cretaceous Helvetiafjel-let Formation, Svalbard. Norsk Polarinstitutt Årbok 1977, 343-345.

Edwards, M.B., 1979. Sandstone in Lower Cretaceous Helvetiafjellet Formation, Svalbard: Bearing on Reservoir Potential of Barents Shelf. American Association of Petroleum Geologists Bulletin, 63(12), 2193-2203.

Edwards, M.B., Bjaerke, T., Nagy, J., Winsnes, T.S. & Worsley, D., 1979. Mesozoic stratigraphy of eastern Svalbard; a discussion. Geological Magazine, 116(1), 49-54.

Emery, D. & Myers, K. (eds.): 1996. Sequence stratigraphy. Blackwell Science, London, United Kingdom, 297.

Fagherazzi, S., Howard, A.D. & Wiberg, P.L., 2004. Modeling fluvial erosion and deposition on continental shelves during sea level cycles. Journal of Geophysical Research, 109(F03010), doi:10.1029/2003JF000091.

Fischer, A.G., 1981. Climatic oscillations in the biosphere. In: Nitecki, M.H. (ed.): Biotic Crises in Ecological and Evolutionary Time, 103-131. Academic Press, New York.

Gjelberg, J. & Steel, R., 1995. Helvetiafjellet Formation (Barremian-Aptian), Spitsbergen: characteristics of a transgressive succession. In: Steel, R., Felt, V.L., Johannesen, E.P. and Mathieu, C. (eds.): Sequence Stratigraphy on the Northwest European Margin, 571-573. Norwegian Petroleum Society (NPF). Special Publication 5.

Grocke, D.R., Price, G.D., Ruffell, A.H., Mutterlose, J. & Baraboshkin, E., 2003. Isotopic evidence for Late Jurassic-Early Cretaceous cli-mate change. Palaeogeography, Palaeoclimatology, Palaeoecology, 202(1), 97-118.

Grosfjeld, K., 1992. Palynological age constraints on the base of the Helvetiafjellet Formation (Barremian) on Spitsbergen. Polar Rese-arch, 11(1), 11-19.

Harland, W.B., Gaskell, B.A., Heaffard, A.P., Lind, E.K. & Perkins, P.J., 1984. Outline of Arctic post-Silurian continental displacements: Petroleum geology of the North European margin, 137-48. Graham and Trotman, London, United Kingdom.

Helland-Hansen, W. & Gjelberg, J.G., 1994. Conceptual basis and vari-ability in sequence stratigraphy: a different perspective. Sedimen-tary Geology, 92(1-2), 31-52.

Helland-Hansen, W. & Martinsen, O.J., 1996. Shoreline trajectories and sequences; description of variable depositional-dip scenarios. Journal of Sedimentary Research, 66(4), 670-688.

Holbrook, J.M., 1996. Complex fluvial response to low gradients at maximum regression: a genetic link between smooth sequence-boundary morphology and architecture of overlying sheet sands-tone. Journal of Sedimentary Research, 66(4), 713-722.

Hurum, J.H., Milàn, J., Hammer, Ø., Midtkandal, I., Amundsen, H.


358

& Sæther, B., 2006. Tracking polar dinosaurs - new finds from the Lower Cretaceous of Svalbard. Norwegian Journal of Geology, 86, 397-402.

Jones, L.S. & Schumm, S.A., 1999. Causes of avulsion: an overview. Fluvial Sedimentology VI: International Association of Sedimentolo-gists, Special Publication, 28, 171-178.

Maher, H.D., Hays, T., Shuster, R. & Mutrux, J., 2002. Signature of a mafic volcanic source in Barremian to Albian sandstones on Spits-bergen, Norway: Geological Society of America, 2002 annual mee-ting, 208. Geological Society of America (GSA), Boulder, CO, Uni-ted States.

Maher, H.D., Jr., 1999. Character of Cretaceous High Arctic LIP from a Svalbard perspective: Geological Society of America, 1999 annual meeting, 430. Geological Society of America (GSA), Boulder, CO, United States.

Maher, H.D., Jr., Hays, T., Shuster, R. & Mutrux, J., 2004. Petrography of Lower Cretaceous sandstones on Spitsbergen. Polar Research, 23(2), 147-165.

Miall, A.D., 1988. Reservoir heterogeneities in fluvial sandstones: les-sons from outcrop studies. American Association of Petroleum Geo-logists Bulletin, 72(6), 682-697.

Midtkandal, I., 2002. Depositional environment, sandstone architec-ture and sequence stratigraphy of the lower Cretaceous Helvetia-fjellet Formation, Spitsbergen. Cand. Scient. Thesis, University of Oslo, Norway, 162.

Midtkandal, I., Ahokas, J. & Nystuen, J.P., 2005. The lower Cretace-ous Helvetiafjellet Formation, Spitsbergen: paralic environment at a ramp shelf from falling to rising sea level: Vinterkonferansen. Geological Society of Norway (NGF), Abstracts and Proceedings, Røros, Norway, 74-75.

Miloslavskij, M.J., Birjukov, A.S., Slenskij, S.N., Hansen, S., Larsen, B.T., Dallmann, W.K. & Andresen, A., 1992. Agardhfjellet, Geologi-cal map, Svalbard, sheet D9G. Norsk Polarinstitutt.

Murray, J.W., 1991. Ecology and Palaeoecology of Benthic Foramini-fera. Longman Scientific and Technical, London, 329 pp.

Mørk, A., Dallmann, W.K., Dypvik, H., Johannesen, E.P., Larssen, G.B., Nagy, J., Nøttvedt, A., Olaussen, S., Pchelina, T.M. & Worsley, D., 1999. Mesozoic lithostratigraphy. In: Dallmann, W.K. (ed.): Lithos-tratigraphic lexicon of Svalbard, Review and recommendations for nomenclature use. Upper Palaeozoic to Quaternary bedrock, 127-214. Norsk Polarinstitutt, Tromsø.

Nagy, J., 1970. Ammonite faunas and stratigraphy of Lower Cretaceous (Albian) rocks in southern Spitsbergen. Norsk Polarinstitutt Skrifter, 1 - 58.

Nagy, J., Pilskog, B. & Wilhelmsen, R.M., 1990. Facies controlled dis-tribution of foraminifera in the Jurassic of the North Sea Basin. In: Hemleben, C., Kaminski, M.A., Kuhnt, W. and Scott, D.B. (eds.): Paleoecology, Biostratigraphy, Paleoceanography and Taxonomy of Agglutinated Foraminifera, 621-657. Kluwer Academic Publishers, Dordrecht.

Nagy, J. & Seidenkrantz, M.S., 2003. New foraminiferal taxa and revi-sed biostratigraphy of Jurassic marginal marine deposits on Anholt, Denmark. Micropaleontology, 49(1), 27-46.

Nagy, J., 2005. Delta-influenced foraminiferal facies and sequence stratigraphy of Paleocene deposits in Spitsbergen. Palaeogeography, Palaeoclimatology, Palaeoecology, 222, 161-179.

Nemec, W., Steel, R.J., Gjelberg, J., Collinson, J.D., Prestholm, E. & Øxnevad, I.E., 1988a. Anatomy of Collapsed and Re-established Delta Front in Lower Cretaceous of Eastern Spitsbergen: Gravita-tional Sliding and Sedimentation Processes. American Association of Petroleum Geologists Bulletin, 72(4), 454 - 476.

Nemec, W., Steel, R.J., Gjelberg, J., Collinson, J.D., Prestholm, E., Øxne-vad, I.E. & Worsley, D., 1988b. Exhumed rotational slides and scar infill features in a Cretaceous delta front, eastern Spitsbergen. Polar Research, 6, 105-112.

Nemec, W., 1992. Depositional controls on plant growth and peat accumulation in a braidplain delta environment: Helvetiafjel-let Formation (Barremian - Aptian), Svalbard. In: McCabe, P.J.

and Parrish, J.T. (eds.): Controls on the Distribution and Quality of Cretaceous Coals, 209-226. Boulder, Colorado. Geological Society of America Special Paper 267.

Nøttvedt, A., Livbjerg, F., Midtbøe, P.S. & Rasmussen, E., 1992. Hydro-carbon potential of the Central Spitsbergen Basin. In: Vorren, T.O., Bergsager, E., Dahl-Stamnes, Ø.A., Holter, E., Johansen, B., Lie, E. and Lund, T.B. (eds.): Arctic Geology and Petroleum Potential, 333-361. Norwegian Petroleum Society (NPF) Special Publication.

Parker, J.R., 1967. The Jurassic and Cretaceous sequence in Spitsber-gen. Geological Magazine, 104(5), 487-505.

Posamentier, H.W. & Vail, P.R., 1988. Eustatic controls on clastic depo-sition II-Sequence and systems tract models. In: Wilgus, C.K., Has-tings, B.S., Kendall, C.G., Posamentier, H.W., Ross, C.A. and Van Wagoner, J.C. (eds.): Sea-Level Changes: An Integrated Approach, 125–154. SEPM Special Publication 42, Tulsa.

Posamentier, H.W. & Morris, W.R., 2000. Aspects of the stratal archi-tecture of forced regressive deposits. In: Hunt, D. and Gawthorpe, R.L. (eds.): Sedimentary Responses to Forced Regressions, 19-46. Geological Society of London, Special Publication 172, London.

Posamentier, H.W., 2001. Lowstand alluvial bypass systems; incised vs. unincised. American Association of Petroleum Geologists Bulletin, 85(10), 1771-1793.

Prestholm, E. & Walderhaug, O., 2000. Synsedimentary faulting in a Mesozoic deltaic sequence, Svalbard, Arctic Norway; fault geome-tries, faulting mechanisms, and sealing properties. American Asso-ciation of Petroleum Geologists Bulletin, 84(4), 505-522.

Róžycki, S.Z., 1959. Geology of the north-west part of Torell Land, Vestspitsbergen. Studia Geologica Polonica, 2, 1-98.

Schumm, S.A., 1993. River response to baselevel change: implications for sequence stratigraphy. Journal of Geology, 101(2), 279-294.

Steel, R., 1977. Observations on some Cretaceous and Tertiary sands-tone bodies in Nordenskiöld Land. Norsk Polarinstitutt Årbok 1976, 43-67.

Steel, R., Gjelberg, J. & Haarr, G., 1978. Helvetiafjellet Formation (Bar-remian) at Festningen, Spitsbergen a field guide. Norsk Polarinsti-tutt Årbok 1978, 111-128.

Steel, R. & Worsley, D., 1984. Svalbard’s post-Caledonian strata; an atlas of sedimentational patterns and paleogeographic evolution: Petroleum geology of the North European Margin, 190-235. Norwegian Petroleum Society (NPF).

Steel, R.J., Crabaugh, J., Schellpeper, M., Mellere, D., Plink-Björklund, P., Deibert, J. & Løseth, T., 2001. Deltas and Rivers on the shelf edge: Their relative contributions to the growth of shelf margins and basin-floor fans (Barremian and Eocene, Spitsbergen). In: Wei-mer, P., Slatt, R.M., Coleman, J., Rosen, N.C., Nelson, H., Bouma, A.H., Styzen, M.J. and Lawrence, D.T. (eds): Transactions Gulf Coast SEPM Annual Conference, Houston, 981-1009.

Swift, D.J.P. & Thorne, J.A., 1991. Sedimentation on continental mar-gins, I: a general model for shelf sedimentation. Shelf Sand and Sandstone Bodies: International Association of Sedimentologists, Special Publication, 14, 3-31.

Takata, H., Seto, K., Sakai, S., Tanaka, S. & Takayasu, K., 2005. Corre-lation of Virgulina fragilis Grindell & Collen (benthic foraminife-rid) with near-anoxia in Aso-Kai Lagoon, central Japan. Journal of Micropaleontology, 24, 159-167.

Takata, H., Takayasu, K. & Hasegawa, S., 2006. Foraminifera in an orga-nic-rich, brackish-water lagoon, Lake Saroma, Hokkaido, Japan. Journal of Foraminiferal Research, 36(1), 44-60.

Talling, P.J., 1998. How and where do incised valleys form if sea level remains above the shelf edge. Geology, 26(1), 87-90.

Torsvik, T.H., Carlos, D., Mosar, J., Cocks, L.R.M. & Malme, T.N., 2002. Global reconstructions and North Atlantic paleogeography 440 Ma to Recent. In: Eide, E.A. (coord.): BATLAS - Mid Norway plate reconstruction atlas with global and Atlantic perspective, 18-39. Geo-logival Survey of Norway.

Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M., Vail, P.R., Sarg, J.F., Loutit, T.S. & Hardenbol, J., 1988. An overview of sequence stratigraphy and key definitions. In: Wilgus, C.K., Hastings, B.S.,


359

Posamentier, H.W., Ross, C.A., van Wagoner, J.C. and Kendall, C.G. (eds.): Sea-Level Changes-An Integrated Approach, 39-45. SEPM Special Publication 42.

Van Wagoner, J.C., Mitchum, R.M., Campion, K.M. & Rahmanian, V.D., 1990. Siliciclastic sequence stratigraphy in well logs, cores, and outcrops; concepts for high-resolution correlation of time and facies. American Association of Petroleum Geologists Methods in Exploration Series, 7, 55 pp.

Walker, R.G., 1992. Facies, facies models and modern stratigraphic concepts. In: Walker, R.G. and James, N.P. (eds.): Facies Models; response to sea level change, 1-14. Geological Association of Canada, Ontario.

Wellner, R.W. & Bartek, L.R., 2003. The Effect of Sea Level, Climate, and Shelf Physiography on the Development of Incised-Valley Complexes: A Modern Example From the East China Sea. Journal of Sedimentary Research, 73(6), 926-940.

Woolfe, K.J., Larcombe, P., Naish, T. & Purdon, R.G., 1998. Lowstand rivers need not incise the shelf; an example from the Great Bar-rier Reef, Australia, with implications for sequence stratigraphic models. Geology, 26(1), 75-78.

Ziegler, A.M., Raymond, A.L., Gierlowski, T.C., Horrell, M.A., Rowley, D.B. & Lottes, A.L., 1987. Coal, climate and terrestrial productivity; the present and early Cretaceous compared. In: Scott, A.C. (ed.): Coal and coal-bearing strata: recent advances, 25-49. Geological Society of London, Special Publication 32.

Århus, N., 1991. Dinoflagellate cyst stratigraphy of some Aptian-Albian sections from Northern Greenland, Southeast Spitsbergen and Barents Sea. Cretaceous Research, 12, 209-225.


Documents

Paralic sedimentation on an epicontinental ramp shelf ... · Paralic sedimentation on an epicontinental ramp shelf during a full cycle of relative sea-level fluctuation; the ... sedimentary