Late Ordovician (Hirnantian) depositional pattern ...

c 2004 Cambridge University Press Geol. Mag. 141 (5 ), 2004, pp. 605–616. DOI: 10.1017/S0016756804009446 Printed in the United Kingdom

605

Late Ordovician (Hirnantian) depositional pattern and sea-level change in shallow marine to shoreface cycles in central Sweden PETER DAHLQVIST ∗ Department of Geology, Lund University, Sölvegatan 12, SE-223 62 Lund, Sweden

(Received 18 November 2002; revised version received 13 April 2004; accepted 22 April 2004)

Abstract – The Upper Ordovician Kyrk˚as Quartzite Formation at the Nifs˚asen Quarry (Jämtland, Sweden) exhibits c. 90 m of siliciclastic sedimentary rocks deposited on a shallow shelf at the cratonattached part of the Caledonian foreland basin. Five lithologies are distinguished, including claystone, mudstone, siltstone, subarkose and sublitharenite. Based on these five lithologies, sedimentary structures and biota, three marine facies associations are defined: the Mudstone association (FA1) deposited close to storm wave base, the Sandstone/mudstone association (FA2) formed between storm and fair-weather wave bases, and the Sandstone association (FA3) accumulated above fair-weather wave base. The facies associations are arranged in two sequences, c. 50 and 40 m thick, separated by a transgressive surface, indicating repeated shoreline progradation. Both sequences commence with marine heterolithic shales and siltstones, with upwardly increasing frequency of tempestites. Continued shoaling is indicated by a dominance of hummocky and trough (locally tabular) crossstratified sandstone beds in the upper part of each sequence. Sand beds are increasingly amalgamated up-sequence, reflecting progressively diminishing accommodation space. The depositional style and sedimentary structures indicate that the study area was storm-dominated with an abundant supply of siliciclastic material. Biostratigraphic data tie the depositional changes to the globally recognized Late Ordovician (Hirnantian) glacial interval. These data suggest that the first sequence was formed during the initial phase of regression in the earliest Hirnantian. The lowermost part of the overlying sequence contains elements of a typical Hirnantia fauna followed by beds yielding Normalograptus persculptus, suggesting a second regressive cycle in the Jämtland basin during the early N. persculptus Biozone. Keywords: Late Ordovician, Hirnantia fauna, storm deposition, Jämtland, eustatic sea level, persculptus.

1. Introduction

¨ The Ostersund/Lake Storsjön area is herein divided into a western and eastern part, with the Upper Ordovician Kyrk˚as Quartzite Formation cropping out to the east and northeast (Fig. 1). A detailed analysis of this formation is presented from Nifs˚asen Quarry (701350/144325, Swedish National Grid), 5.5 km NNE ¨ of Ostersund, with comparisons to the successions at the nearby Ljusberg, Rann˚asen and Storhögen quarries (Fig. 1). Nifs˚asen Quarry exposes a 90 m thick succession of siliciclastic sedimentary rocks previously not described. The Kyrk˚as Quartzite Formation represents Late Ordovician deposition, and may contain almost the entire Hirnantian interval. This paper uses sedimentological analysis to interpret the depositional environments and their evolution for the Kyrk˚as Quartzite Formation, and provides an understanding of the Late Ordovician–Early Silurian environmental development in the Caledonian foreland basin. Comparisons with coeval deposits (the Ede Quartzite Formation) in the western part of the area,

∗

E-mail: [email protected]

http://journals.cambridge.org

Downloaded: 08 Oct 2013

where eustatic sea-level changes have been recorded (Dahlqvist & Calner, in press), are also made. 2. Background 2.a. Palaeogeography and geological setting

During the latest Ordovician times the Baltic craton was situated at southern subtropical–tropical latitudes, and Scandinavia formed the western margin of Baltica (Torsvik, 1998). The investigated sedimentary succession was deposited in the Caledonian foreland basin, which developed from the progressive coalescence of the Laurentian and Baltic continental margins (e.g. Greiling, Garfunkel & Zachrisson, 1998). This was a narrow basin along the Caledonian front with the Jämtland area in the northern part and the Oslo area in the south (Baarli, Johnson & Antoshkina, 2003). The deposits were later translocated southeastward to eastward as discrete nappes during the major collisional phase (Scandian) of the Caledonian Orogeny (e.g. Roberts & Gee, 1985). The Upper Ordovician–Lower ¨ Silurian part of the succession belongs to the Ange Group (uppermost Jämtland Supergroup), which forms the main part of the Lower Allochthon (Gee &

IP address: 192.71.158.2

606

P. DA H L Q V I S T 142500

investigations from the Late Ordovician–Early Silurian part of the succession of Jämtland were conducted by Lindström (1872), Wiman (1893), Boucot & Johnson (1964), Karis & Larsson (1982), and Cherns (1994).

145000

705000

2.b. Stratigraphy Edefors

702500

Ljusberg Storsjön Norderön

Storhögen

Nifsåsen

Rannåsen

.. Ostersund

700000 Middle Allochthon L. - M. Silurian Kyrkås and Ede Fm. Kogsta Fm. Cambrian - U. Ord Precambrian crystalline basement

N 0

10 km

¨ Figure 1. Geological map of the Ostersund area, showing the distribution of sedimentary rocks and location of localities referred to in the text. Note that the eastern part with the Kyrk˚as Quartzite Formation occurs to the east of Norderön Island whereas the Ede Quartzite Formation occurs to the west of this island. Coordinates in Swedish National Grid. Ljusberg Quarry (701740/144651, Swedish National Grid) is situated ¨ c. 9 km NNE of Ostersund, c. 3 km NW of the road (45). Nifs˚asen Quarry (701350/144325, Swedish National Grid) is ¨ situated c. 5.5 km NNE of Ostersund, c. 2 km NE of the main road (E14). Rann˚asen Quarry (701020/144425, Swedish ¨ National Grid). c. 2.5 km NE of Ostersund, c. 1 km E of the main road (E14).

Kumpulainen, 1980; Fig. 2). Facies and stratigraphical relationships indicate that most deposits of the Lower Allochthon were deposited on the craton-attached shelf of the foreland basin (cf. Jaanusson, 1982, fig. 3). In Early and Middle Ordovician times, the palaeoslope dipped from the eastern part of the basin towards the west (Karis & Strömberg, 1998, p. 165). Accordingly, the eastern part of the basin was dominated by carbonate deposition, while the western area was dominated by mud and turbidite deposition in deeper water environments (Karis & Strömberg, 1998). The geological history of the area is described in the early works of Thorslund (1940, 1943, 1948) and in Thorslund & Jaanusson (1960). Regional and structural geological development are discussed by Gee & Kumpulainen (1980), Bassett, Cherns & Karis (1982), Jaanusson (1982), Karis (1982), Bassett (1985), Gayer & Greiling (1989), Cherns & Karis (1995), Samuelsson (1997), Greiling, Garfunkel & Zachrisson (1998), Karis & Strömberg (1998) and Dahlqvist & Calner (in press), among others. Palaeontological



The Upper Ordovician Kogsta Siltstone Formation represents uniform marine deposition over large areas. In the eastern part of the area, however, a very finegrained limestone (the Furulund Limestone, Fig. 2), up to 10 m thick, is enclosed in the Kogsta Siltstone Formation approximately 25 m below the contact to the Kyrk˚as Quartzite Formation (Karis & Strömberg, 1998, fig. 76). As demonstrated by Dahlqvist & Calner (in press), the fine-grained clastics of the Kogsta Siltstone Formation, mainly shale and siltstone, were deposited in an offshore environment during a highstand situation, possibly with minor shallowing in the upper part. In the western part of the area its uppermost part has yielded a Hirnantian shelly fauna (Bergström, 1968; Karis & Larsson, 1982), which shows similarities to the Hirnantia fauna (sensu Temple, 1965). Sandstone units follow the Kogsta Siltstone Formation: the Kyrk˚as Quartzite Formation in the eastern and the Ede Quartzite Formation in the western part of ¨ the Ostersund/Lake Storsjön area (Fig. 1). The Kyrk˚as Quartzite Formation conformably overlies the Kogsta Siltstone Formation, whereas the Ede Quartzite Formation rests unconformably on the Kogsta Siltstone Formation (Cherns & Karis, 1995; Dahlqvist & Calner, 2001; Dahlqvist & Calner, in press). The thickness of the Kyrk˚as Quartzite Formation varies regionally due to erosional and tectonic processes. At the Rann˚asen and the no longer accessible Storhögen quarries, the thickness was estimated at about 35–45 m (Cherns & Karis, 1995). New data show that the successions at Nifs˚asen, Rann˚asen and Ljusberg are considerably thicker (see Section 6.a). The Kyrk˚as Quartzite Formation forms two upward-coarsening cycles, as discussed in this paper (see Section 4). Cherns & Karis (1995) interpreted the Kyrk˚as Quartzite Formation as comprising one coarsening-upward cycle. By contrast, Thorslund & Jaanusson (1960, fig. 23c) claimed that the section at Rann˚asen is composed of a lower part of quartzite (c. 25 m) followed by a shaly part (c. 12 m), which is topped by quartzites (c. 11 m). Since formational boundaries are not established at Nifs˚asen or elsewhere, the maximum thickness of the Kyrk˚as Quartzite Formation remains unknown. Biostratigraphical data indicate that the Kyrk˚as strata are of Late Ordovician age. These data include several indicative and important fossils that tie the depositional changes to the globally recognized Late Ordovician (Hirnantian) glacial interval, for instance, the trilobite Tretaspis, brachiopods typical for the Hirnantia fauna, and the zone fossil Normalograptus persculptus.

IP address: 192.71.158.2

Ordovician shallow marine to shoreface cycles SYSTEM

BRITISH SERIES

STAGE

607 .. EASTERN (OSTERSUND) AREA

WESTERN (EDE) AREA

.. JAMTLANDIAN UNITS

S Wenlock

Röde Fm.

(100)

Sheinwoodian

Ekeberg Greywacke Fm.

(40-50)

Telychian

Bångåsen Shale Fm.

(25)

I L U R I A

Aeronian

Berge Limestone Fm.

N

? Rhuddanian

Ede

Llandovery

.. A N G E G R O U P

?

? (90)

Hirnantian

Quartzite Fm.

Hf

D

V

Turbidite sandstone and shale

T Marine shale

L

Micritic shelf limestone

N D

Mixed carbonate-siliciclastic sediments Shelf sand- and mudstone Shoreface sandstone

P

R

O

M

U

Kyrkås

Hf

Continental redbeds

S

Quartzite Fm. O

J .. A

A

(50-75) (6-8)

GENERAL FACIES

Shelf siltstone and shale

E Ashgill

Furulund Limestone

R ? Rawtheyan

I

Kogsta Siltstone Fm.

(75-100) Kogsta Siltstone Fm.

C.

T Å S J Ö N G R.

G

Osmundsberg bentonite

R O

Hf

Hirnantia fauna

U P

Unconformity

Figure 2. Stratigraphy of the Jämtland basin, central Jämtland. Note that the thicknesses of different units are not to scale (given in metres within brackets). Based on figure 2 in Dahlqvist & Calner (in press) and biostratigraphic data presented herein.

According to Thorslund & Jaanusson (1960, fig. 23c), fine clastics at Rann˚asen (as treated herein most probably the lower part of the Kyrk˚as Quartzite Formation) contain the trilobite Tretaspis. Approximately in the middle part of the section at Rann˚asen (in the lower part of the second cycle), the following fossils occur in a dark mudstone: the trilobites Dalmanitina (Mucronaspis) mucronata and Brongniartella platynota, the brachiopod Leptaenopoma trifidium, molluscs and hyolithids (Karis & Larsson, 1982). Together with the brachiopods Eostropheodonta hirnantentis and Dalmanella testudinaria (K. Larsson, unpub. data), they indicate a Hirnantian age of the sediments. This fauna contains elements from the Hirnantia fauna (cf. typical elements of the Hirnantia fauna by Owen, Harper & Jia-yu, 1991), but differs from the Hirnantia fauna in the Kogsta Siltstone Formation at the western localities (L. Karis, pers. comm. 2003). Approximately 2–3 m above this mudstone, graptolites have been found (e.g. Normalograptus persculptus), which first were regarded to be of Early Silurian age (Thorslund & Jaanusson, 1960), but later were referred to the N. persculptus Biozone and Latest Ordovician by R. B. Rickards in Cherns & Karis (1995). These two important fossiliferous intervals are easy to locate, separated by an easily recognized marker bed (see Sections 4.a.1 and 6.a). The clastic deposits of the Ede Quartzite Formation were followed by widespread carbonate deposition (the Lower Silurian Berge Limestone Formation), and succeeding shale deposition (the B˚ang˚asen Shale



Formation; Fig. 2). The mid-Aeronian Berge Limestone Formation was interpreted as being formed in a low-energy offshore environment, during a highstand situation (Dahlqvist & Calner, in press). Probably the expansion of carbonates also indicates a change into a warmer climate, following the cold period (Dahlqvist & Calner, in press). Due to the direction of nappe transportation and later erosion, no records of the Berge Limestone and B˚ang˚asen Shale formations have been found in the eastern part of the area, while they are present to the west. 3. Material and methods

This study is based on facies logging of the Kyrk˚as succession at the northeastern wall of Nifs˚asen Quarry (Fig. 1). The strata are tectonically detached and slightly overturned, allowing detailed sampling along quarry walls. The measured and sampled section is 6–8 m high, permitting moderate control on lateral variations, and encompasses c. 90 m of sandstone, siltstone and shale. A few minor vertical faults and minor differences in dip of strata occur in the section. In total, twenty-nine representative samples were collected. Thin sections made from these samples were point-counted (300 points per section). Sandstones were classified according to Pettijohn, Potter & Siever (1987), and mudstones according to Tucker (2001). Classification of bedding was done according to the scheme of Collinson & Thompson (1989). The classification of trace fossils follows the toponomic

IP address: 192.71.158.2

608

P. DA H L Q V I S T

scheme of Martinsson (1970). Reference samples are stored at the Department of Geology, Lund University, Sweden. 4. Lithofacies descriptions and interpretations

The Kyrk˚as Quartzite Formation is a heterogeneous siliciclastic unit with abundant primary sedimentary structures. On outcrop scale, individual thin- to thickbedded sandstone beds are tabular to wedging. Five lithologies are discerned: claystone, mudstone, siltstone, subarkose and sublitharenite. Based on lithology, sedimentary structures and biota, three Facies Associations were distinguished (Fig. 3): the Mudstone association (FA1), the Sandstone/mudstone association (FA2) and the Sandstone association (FA3). Further, based on the facies associations, their spatial relationships, and the presence and position of a marker bed, two sequences were identified (Fig. 4): sequence 1 (S1) and sequence 2 (S2). The two sequences differ slightly in build-up (see Section 6.b). Bioturbation and marine fossils such as brachiopods, trilobites, gastropods, and cephalopods occur throughout the section at Nifs˚asen Quarry. Body fossils are restricted to the finer lithologies, and no fossils have been found in the sandstone beds.

stratification and ripples. Occasionally, rip-up clasts occur in the lower parts of these sandstone beds. Individual sandstone beds sometimes pinch out laterally. Bioturbation is abundant in the shale as both horizontal and vertical exichnial traces. Silt and sandstone beds are less frequently bioturbated. Horizontal traces are preserved on the lower bedding planes as hypichnial ridges, and more rarely on the upper bedding planes as epichnial ridges. Vertical endichnial traces occur on the lower bedding planes. A useful marker bed occurs within this association (in the lower part of the S2). This is an approximately 0.75 m thick, calcite cemented, rusty brown weathered, siltstone herein termed the ‘Rusty Marker Bed’ (RMB; Figs 4, 5b). This bed is bioturbated, with randomly oriented traces. The RMB also occurs, with approximately the same thickness, in the middle part of the succession at Rann˚asen Quarry (701020/144425, Swedish National Grid), c. 3 km to the SSE (between the two fossiliferous horizons (see Section 2.b)), and at the Ljusberg Quarry (701740/144651, Swedish National Grid) (Fig. 1). The importance of this bed is discussed in Section 6.a. 4.a.2. Interpretation

4.a. Facies Association 1 (FA 1): Mudstone 4.a.1. Description

This association consists of bioturbated, dark grey to black shale, heterolithic mud-siltstone, and subordinate thin- to medium-bedded silt and sandstone beds (Fig. 3). Shale and siltstone are interbedded, forming heterolithic structures, including alternating planar–parallel lamination (mm to cm scale), and wavy and lenticular bedding (cm scale). Lenticular bedding occurs both as single and connected forms, and contains symmetrical and asymmetrical ripple forms. Silt and sandstone beds, 0.04–0.1 m in thickness (rarely 0.2–0.3 m), occur and increase in abundance upwards (Fig. 4). Lower bedding planes are generally sharp and erosive, sometimes associated with load casts, flute casts (Fig. 5a), or tool marks. In outcrop, most silt and sandstone beds appear massive. However, analysis of slabs reveals abundant normal grading, planar to slightly undulating lamination, and smallscale cross-lamination. The upper bedding planes generally grade into mudstone, which in places is bioturbated. Individual beds are generally normally graded, and may be subdivided into a lower massive part, overlain by planar or slightly undulating lamination, superimposed by wave-influenced ripples. The internal structure of the thicker sandstone beds can be subdivided similarly. However, their massive or parallel-laminated lower parts are succeeded by slightly undulating lamination and/or hummocky cross-



Based on fossils and bioturbation, this is an open marine facies association. The predominantly finegrained deposits with interbedded thin, sharp-based sandstones indicate a calm depositional environment, at times affected by event deposition. The primary sedimentary structures within single bedsets typically include a sharp erosional base, sometimes with flute and load-casts, a massively bedded or laminated lower part, and wave-ripples at the upper bedding plane. This succession conforms to facies models for storm deposition (e.g. Dott & Bourgeois, 1982; Brenchley, 1985; Seilacher & Aigner, 1991), as well as to previous studies concerning storm deposits (e.g. Kreisa, 1981; Aigner & Reineck, 1982; Baarli, 1988; Brenchley, 1989). The pinch-out style of sandstone beds also supports a storm-generated origin of these beds (e.g. Kreisa, 1981; Brenchley, Pickerill & Stromberg, 1993). This association shows features characteristic of the upper Offshore to lower Offshore-transition zones (cf. Brenchley, 1985, 1989). The association shoals upward, indicated by the increase in thickness and frequency of sand beds, and by the temporal change in sedimentary structures. 4.b. Facies Association 2 (FA 2): Sandstone/mudstone 4.b.1. Description

This association comprises medium- to thick-bedded sandstones, in places amalgamated, but mostly separated by marine shale and/or heterolithic mud-siltstone.

IP address: 192.71.158.2

Ordovician shallow marine to shoreface cycles

609

2m

FA 3: Sandstone Association Medium to thick bedded, amalgamated sandstone beds. Beds are massive, graded, planar laminated, trough crossbedded, or hummocky cross-stratified. Lower bedding planes are erosional, planar or scouring, and show erosional features; flute casts, tool marks, rip-up clasts. Upper bedding planes are mostly erosive, but may contain ripples, where not cut. Sandstones are fine to medium grained subarkoses and lithic arenites. Sediments were deposited in lower to upper shoreface zone. Storm dominated association with depositional depth above fair-weather wave base.

0.2m

1m

2m

FA 2: Sandstone/mudstone Association Thin to medium bedded sandstone beds interbedded with heterolithic mud and siltstone. Sandstone beds are massive, graded, planar laminated, undulating laminated, or hummocky cross-stratified. Lower bedding planes are erosional, planar or scouring and show erosional features; tool marks, flute casts, gutter casts, load casts, and rip-up clasts. Sandstones are fine grained muddy subarkoses and subarkoses. Heteroliths are moderately bioturbated. Sediments were deposited in Offshore-transition zone. Storm deposited sandstone beds interbedded with low energy shelf muds. Deposition between storm and fair-weather wave bases.

0.2m

1m

2m

FA 1: Mudstone Association Heterolithic mud and siltstone interbedded with thin to mediumbedded sandstone beds. Heteroliths include, alternating lamination, wavy and lenticular bedding. Sandstone beds are planar or undulating laminated, often grading up into ripples at top. Lower bedding planes of sandstone beds show erosional features; load casts and flute casts. Sandstone beds are fine grained muddy subarkoses. Heteroliths are intensely bioturbated. Sediments were deposited in upper Offshore to lower Offshore-transition zones. Dominated by background sedimentation with storm influence. Deposition close to storm wave base.

0.2m

1m

Siltstone, sandstone Trough cross-bedding Planar lamination Hummocky cross-stratification

Ripple cross-lamination Undulating lamination Wavy bedding and alternating lamination

Rip-up clasts

Lenticular bedding

Symmetrical ripples

Mudstone

Figure 3. The three facies associations found at Nifs˚asen, their general characteristics and interpretations. Pictures are shown in original vertical mode with younger strata to the right.



IP address: 192.71.158.2

610

P. DA H L Q V I S T

40 J-00-74

15

J-01-05

J-00-84

Tectonized strata ? ?

J-01-34

60 J-00-76 J-01-26 J-01-32

50

75

35 10

J-01-01

25

RMB

J-00-81

J-01-04

J-00-83

85 J-00-75 J-00-85

55

J-01-30

J-01-03

65 J-00-78

45 J-00-111 J-00-80 J-00-79

30

70

J-01-82

5 J-00-73

J-01-02

20

J-01-31

J-01-33

TRANSGRESSIVE SURFACE

80 J-00-77

J-00-72

SEQUENCE 2 Silt / Sandstone Mudstone / Heteroliths

J-01-27 J-01-29

J-01-28

Sample number/level

Dewatering structures J-01-29

Minor faults SEQUENCE 1

RMB

Rusty Marker Bed

Figure 4. For legend see facing page.



IP address: 192.71.158.2

Ordovician shallow marine to shoreface cycles Sandstone is approximately 75–80 % in this association (Figs 3, 4). Interbedded shale and heteroliths display the same features as in FA1 (see Section 4.a.1), with the exception that bioturbation is less common in FA2. Sandstone beds are 0.1–0.7 m thick. Their lower bedding planes typically scour the underlying mud or sandstone beds; however, planar bases are also present. Where sandstone beds succeed shale, the lower bedding planes of single beds show well-developed load casts or erosional features, such as flute casts, prod marks, groove casts and gutter casts. Gutter casts are U-shaped with decimetre relief and may exhibit smaller, superimposed sole marks. Single beds are normally graded and draped by mud. Some sandstone beds may be subdivided into a massive or parallel-laminated lower part, either succeeded by a slightly undulating lamination and/or hummocky cross-stratified middle part, followed by an undulating or wave-rippled top. A majority of the sandstone beds contain rip-up clasts (up to 0.1 m in diameter) in their lower parts. Lateral pinch out of individual sandstone beds occurs throughout the section. At some levels the bedding is disturbed, showing dewatering and slump structures (Fig. 5c, d). Bioturbation in the muddy heteroliths is preserved as both horizontal and vertical exichnial traces (Fig. 5e). Silt and sandstone beds are less bioturbated. Traces within these beds are horizontal and occur on the lower bedding planes as hypichnial ridges (Fig. 5f, g), and more rarely on the upper bedding planes as epichnial ridges (Fig. 5h). Vertical endichnial traces can be seen on the lower bedding planes. Due to the steeply dipping strata and tectonism, measurements of flute cast and ripple crest palaeodirections are highly uncertain. However, ripple crests are approximately perpendicular to flute cast directions. 4.b.2. Interpretation

The indications of turbulent conditions and high sedimentation rates (e.g. hummocky laminated event beds, flute casts and deformation structures) are in contrast to the interbedded shale deposition and consequently, point to alternating energy conditions and episodic deposition. The idealized make-up of single event beds including sharp, erosive bases and their internal structural subdivisions conforms to facies models for storm deposits (see Section 4.a.2). These deposits are interpreted to be deposited in the Offshoretransition zone (cf. Brenchley, 1985, 1989). Dewatering structures show characteristics of vertical movement of denser sand beds into mud, interpreted as ball and pillow structures (Fig. 5c). Slump

611 structures are convolute bedding (Fig. 5d). Cherns & Karis (1995) recorded sedimentary deformation at Rann˚asen Quarry, which they interpreted as slump horizons resulting from tectonic activity affecting unstable depositional slopes.

4.c. Facies Association 3 (FA3): Sandstone 4.c.1. Description

This facies association is characterized by medium- to thick-bedded, amalgamated sandstone beds. Sandstone beds are 0.1–1.0 m thick with sharp erosional lower bedding planes (Fig. 3). Thin mud or silt drapes are rare and discontinuous. Successions with this association can be subdivided into a lower and upper part. In the lower part, individual beds are normally graded, and show sedimentary structures typical of waning flow (hummocky crossstratification, HCS). Lower bedding planes show the same erosional features (except for gutter casts) as in FA2 (see Section 4.b.1). Due to the amalgamation of beds and associated erosion, most upper bedding planes are sharp. Where erosion did not occur, upper bedding planes occasionally contain symmetrical ripples. Single sandstone beds are typically massively bedded at the base, with rip-up clasts up to 0.4 m in diameter (Fig. 5i), succeeded by parallel/undulating lamination passing over into hummocky cross-stratification at the top. The upper part differs from the lower part in the following: lower bedding planes scour the underlying bed; strata contain low to high angle trough and locally tabular cross stratification; rare quartz gravel lag deposits are present; rip-up clasts also occur on foresets; individual sandstone beds vary in thickness laterally and may pinch out completely. Bioturbation is scarce, and concentrated to the upper, more fine-grained parts of the sandstone beds as epichnial ridges and endichnial traces. No body fossils have been found in this facies association. 4.c.2. Interpretation

This association consists almost entirely of event deposits, suggesting a high-energy depositional environment. The depositional style and sedimentary structures conform to storm deposits (see Section 4.a.2). The low degree of bioturbation indicates frequent reworking and a high sedimentation rate (cf. Aigner & Reineck, 1982). Trough and tabular cross-bedding indicates prograding dunes. This association shoals upward, exemplified by the change in sedimentary structures. The lower part, with amalgamated HCS, indicates

Figure 4. Lithological log from the section at the Nifs˚asen Quarry, with the youngest part to upper right, scale in metres. Possible deepening episodes in the sequences at around 2 m, 14 m, 16 m, 22.5 m, 33.5 m in sequence 1 and 59 m, 66 m, 71 m, 78 m and 83.5 m in sequence 2. Note that the RMB at 57 m separates sediments yielding a Hirnantia fauna (below) from a N. persculptus fauna (above).



IP address: 192.71.158.2

612

P. DA H L Q V I S T

a

b

c

d f

e

g

h

i Figure 5. For legend see facing page.



IP address: 192.71.158.2

Ordovician shallow marine to shoreface cycles lower to middle shoreface depositional environments, whereas the upper half with trough cross-bedding (prograding dunes) indicates mid- to upper shoreface deposition (cf. Brenchley, 1985, 1989; McRory & Walker, 1986). 5. Petrography

Standard petrographic analysis of the Kyrk˚as Quartzite Formation reveals five lithologies, all showing some variation: claystone, mudstone, siltstone, subarkose and sublitharenite. Lithologies often dominate one of the associations, but occur subordinated in the other(s) (Fig. 3). The grain size and matrix content in the subarkoses differ greatly. Therefore a variety termed muddy subarkose is discussed in this paper. This muddy subarkose dominates FA1 and is present in FA2, and the other subarkose dominates FA3 and is present in FA2. The muddy subarkose contains more matrix (mean: 32 %) than the subarkoses (mean: 7 %). The grain size also differs between these, the subarkoses being approximately twice as coarse-grained as the muddy subarkoses. In both cases potassium feldspar (mostly microcline) is the dominating type of feldspar (approximately 95 %). 6. Discussion 6.a. Depositional evolution

The repetition of the three facies associations (FA1–3, Fig. 3) demonstrates the presence of two shallowingupward cycles, sequence 1 (S1) and sequence 2 (S2). These are separated by a transgressive surface (Fig. 4), located between markedly different facies: below are sandstones of mid-upper shoreface origin (FA3); above are mud/silt stones deposited in an offshore environment (FA1). This transgressive surface thereby represents an abrupt change in proximity and depositional energy. Within the two sequences there are gradual transitions between the facies associations reflecting gradual shoaling. The overall shoaling pattern in each sequence is, however, not continuous. Minor deepening episodes might be recognized as a change from sandstone to shale dominance in the section.

613 Possible deepening episodes in the sequences could be at around 2 m, 14 m, 16 m, 22.5 m, 33.5 m in sequence 1 and 59 m, 66 m, 71 m, 78 m and 83.5 m in sequence 2 (Fig. 4). Hence, each sequence contains smaller shoaling-upward elements, probably parasequences. Increased energy and decreased distance to the shoreline are indicated upward in each sequence by increases in: (a) grain size; (b) amalgamation of beds and bed thickness; (c) textural maturity of sandstones; (d) a change to more proximal sedimentary structures; and a decrease in bioturbation. This trend conforms to distal–proximal trends from modern shelf storm deposits (cf. Aigner & Reineck, 1982). Several other characteristics point to a storm-dominated shelf depositional environment for the Kyrk˚as Quartzite Formation, including HCS, normally-graded beds, trough/tabular cross-bedding and wave ripples (cf. Aigner, 1985). The progradation of the shoreline suggests that there was an abundant supply of siliciclastic material. The presence of slump horizons recorded at Storhögen and Rann˚asen (Cherns & Karis, 1995), and Nifs˚asen (this paper), reflects slope instability. Bioturbation and presence of body fossils indicate an oxic bottom environment. The ‘Rusty Marker Bed’ (RMB) occurs, with only minor change in thickness, in facies association 1 (FA1) at all investigated quarries. This may be due to a restriction to a particular depositional environment (in this case the Offshore zone), and/or that the three quarries are aligned along the same strike along the palaeo-shoreline. Its presence makes it possible to re-evaluate the thickness of the Kyrk˚as Quartzite Formation: it indicates the presence of two sequences at Nifs˚asen, as well as in the Rann˚asen and Ljusberg quarries, and thus, approximately doubles the known thickness of the Kyrk˚as Quartzite Formation (cf. Cherns & Karis, 1995). Additionally, the RMB separates sediments yielding fossils characteristic of the Hirnantia fauna, from sediments with graptolites of N. persculptus Biozone (see Section 2.b), that is, at a stratigraphical level with a global oceanographic signature (see Section 6.b). The distinctive appearance of the RMB, the surrounding faunas and globally recognized palaeoceanographic changes during this time (e.g. Sheehan, 2001) suggest that the RMB may

Figure 5. Sedimentary features and bioturbation at Nifs˚asen Quarry. (a) Flute casts on the lower bedding plane of a c. 0.15 m thick sandstone bed, FA1. Coin is 2.4 cm in diameter. (b) The white arrow points to the Rusty Marker Bed, up-section to the right. The RMB is approximately 0.75 m and is present in FA1 in S2, c. 6 m from the uppermost sandstones of S1. (c) Ball and pillow structure in FA2 in S1. Note that laminae are still present and can be followed around inside the bedform. Diameter approximately 30 cm. (d) Convolute bedding in FA2 in S1. Disturbed part is approximately 2 m thick. Note undisturbed sandstone beds above and below the disturbed sequence (arrowed). (e) Horizontal exichnial traces on a laminae plane in a ball structure, FA2. (f ) Branching horizontal hypichnial ridges on the lower bedding plane on a 0.3 m thick sandstone bed, FA2. (g) Horizontal hypichnial traces on the lower bedding plane on a 0.5 m thick sandstone bed, FA2 (scale in cm). (h) Horizontal epichnial ridges on a wave-rippled surface. Sandstone bed is c. 0.15 m thick, FA2. Handle of hammer is 15 cm. (i) Rip-up clasts, up to 0.4 m in diameter, FA3. Hammer is 28 cm long.



IP address: 192.71.158.2

614

P. DA H L Q V I S T

reflect at least a regional marine event. Further analysis of the RMB and surrounding faunas will be undertaken to reveal the importance of this interval. 6.b. Controls on deposition

The build-up of the recorded pattern may be interpreted in several ways. Possible reasons for repeated shoaling include glacio-eustasy due to Hirnantian glaciation and active tectonic control due to early Caledonian movements. Interplay between these causes is also possible. Since the Kyrk˚as strata were deposited during the Hirnantian, a period with major global sealevel changes, eustatic controls must be considered. Evidence for such sea-level change is found in the laterally equivalent Ede Quartzite Formation in the western part of the area. Recent analysis of the sedimentology and the stratigraphical relationships of these strata indicate that they were deposited during a major relative sea-level drop, associated with Late Ordovician glacio-eustasy (Dahlqvist & Calner, 2001). Based on stratigraphical boundaries and facies, Dahlqvist & Calner (in press) interpreted the Ede Quartzite Formation as a downward shift in coastal onlap due to forced regression. Since the western part was affected by Hirnantian glacioeustatic changes, it is probable that the eastern area was also influenced by these changes. The faunas enclosing the RMB (Hirnantia fauna below and N. persculptus fauna above; see Section 2.b) close above the transgressive surface, tie the succession to the stratigraphical level for the Hirnantian glacial interval. The sea-level change associated with that event includes two phases of regression, a first gradual phase in the early Hirnantian (early N. extraordinarius Biozone), and a second, and more rapid fall in sea level, in the early N. persculptus Biozone (e.g. Sutcliffe et al. 2000). The first regressive phase was gradual and represents initial ice-sheet growth, whereas the second defines rapid expansion of the ice-sheet (e.g. Sutcliffe et al. 2000). Major and rapid glacioeustatic sea-level rise began shortly after the second regressive phase (Sutcliffe et al. 2000). According to Brenchley & Marshall (1999), among others, Tretaspis disappears during the first phase of extinction, which was related to the primary regression. Hence, the presence of Tretaspis in the lower part of the first cycle (S1) may correlate with the first phase of regression in early Hirnantian. The second cycle (S2) with Hirnantia and persculptus faunas possibly reflects the second regressive event (cf. Sutcliffe et al. 2000). The recognition of probable parasequences within each sequence points to an episodic shallowing of each sequence, the overall regressive cycles being punctuated by deepening episodes. Similar detailed investigations of more localities could perhaps reveal the same pattern in other sections, and conclusions based on these smaller scale cycles can be made.



The two sequences differ slightly: in comparison to S1, S2 is thinner, and contains more shale and less cross-bedded sandstone in the most proximal facies association (FA3, Fig. 4). This pattern cannot be safely interpreted before the upper contact of the Kyrk˚as Quartzite Formation, and the thickness and content of the whole formation, have been established. The reason is that there is a possibility that there is a missing part (above what can be seen) of the second sequence (S2). This part could be even more proximal than the top of S1 (the strata may even have been exposed above sea level). Although the glacioeustatic control is favoured herein, another possible cause for the revealed pattern is tectonism. Local and regional tectonism can change subsidence patterns, and may be a possible trigger for the pattern. According to Bassett (1985), the Kyrk˚as Quartzite Formation was deposited in a subsidiary trough close to the hinge zone between the platform and outer basin possibly related to basement-controlled faults. Probably minor syn-sedimentary faulting could change the subsidence rate locally and thereby the thickness of the succession. Evidence for larger scale tectonism, possibly influencing the Jämtland area, during the Late Ordovician is discussed in Roberts (2003), Andréasson et al. (2003) and PlinkBjörklund, Roberts & Björklund (2004), among others. Synsedimentary faulting, controlled by plate tectonic activity, is recorded in the Upper Ordovician succession in southern Norway by Stanistreet (1983), among others.

6.c. Correlation

According to biostratigraphic data, Tretaspis in the lower part of S1 in the east, and Hirnantia fauna before the first regressive sign in the west, suggest that the upper parts of the Kogsta Siltstone Formation is most probably younger in the western area (Fig. 2). This points to a preservation of a palaeoslope that dipped from the east towards the west (see Section 2.a). The two different Hirnantia associations do not have to be synchronous, but until a fauna from Edefors is described this is assumed. The differences could be depth related, perhaps with deeper and colder water conditions in the west. This suggests a correlation between the Kyrk˚as and Ede Quartzite formations as shown in Figure 2. There are several similarities between the depositional pattern in the Jämtland area and the Oslo region. In the Oslo region, Brenchley & Newall (1980) recorded shoreline progradation due to the Hirnantian regression on clastic shelves with sufficient supply of sand to maintain continuous sedimentation. These sequences mostly contain synsedimentary deformation structures (Brenchley & Newall, 1984), which also are present in the Kyrk˚as Formation.

IP address: 192.71.158.2


615

7. Conclusions

References

The Upper Ordovician Kyrk˚as Quartzite Formation at Nifs˚asen Quarry contains two superimposed, asymmetrical sequences, indicating repeated relative shallowing of a proximal shelf setting. Each may be subdivided into three depth-related facies associations in ascending order: The Mudstone association. Strata of this association were deposited close to storm wave base in an offshore environment. It is built up by marine heteroliths with rare event beds. Fine grained lithologies, claystone, mudstone and siltstone, dominate, and muddy subarkose occurs sparsely. The Sandstone/mudstone association. This association was deposited between storm and fair-weather wave base in an offshore transition environment, and is built up by event beds and interbedded fines. Muddy subarkose and claystone, mudstone and siltstone are the dominant lithologies. The Sandstone association. Deposits occurred above fair-weather wave base in a shoreface environment, built up by amalgamated event beds and dunes. The dominant lithology is subarkose, with low amounts of muddy subarkose and sublitharenites. A useful marker bed, the ‘Rusty Marker Bed’, makes it possible to correlate localities and supports the presence of two sequences at other localities. The thickness of the Kyrk˚as Quartzite Formation is established as at least 90 m (formerly 45 m). The depositional style and sedimentary structures in Nifs˚asen indicate that the eastern craton-attached shelf area of the Late Ordovician Caledonian foreland basin was storm dominated with an abundant supply of siliciclastic material. The Kyrk˚as Quartzite Formation comprises the most of the Hirnantian interval and reflects glacio-eustatic changes. Biostratigraphic data suggest that the first coarsening-upward cycle (S1) was deposited during a regressive phase caused by a sea-level drop, due to the Hirnantian glaciation. S2 may reflect a second regressive phase in the early N. persculptus Biozone as reported in other areas. These data also enable a correlation to the Ede Quartzite Formation in the ¨ western part of the Ostersund/Lake Storsjön area.

AIGNER, T. 1985. Storm Depositional Systems. Lecture Notes in Earth Sciences. Berlin: Springer-Verlag, 174 pp. AIGNER, T. & REINECK, H. E. 1982. Proximality trends in modern storm sands from the Helgoland Bight (North Sea) and their implications for basin analysis. Senckenbergiana Maritima 14, 183–215. ANDRE´ ASSON, P. G., GEE, D. G., WHITEHOUSE, M. J. & ¨ , H. 2003. Subduction-flip during Iapetus SCHOBERG Ocean closure and Baltica–Laurentia collision, Scandinavian Caledonides. Terra Nova 15, 362–9. BAARLI, B. G. 1988. Bathymetric co-ordination of proximality-trends and bottom-level communities: a case study from the Lower Silurian of Norway. Palaios 3, 577–87. BAARLI, B. G., JOHNSON, M. E. & ANTOSHKINA, A. I. 2003. Silurian stratigraphy and paleogeography of Baltica. In Silurian Lands and Seas – Paleogeography Outside of Laurentia (eds E. Landing and M. E. Johnson), pp. 3–34. New York State Museum Bulletin no. 493. BASSETT, M. G. 1985. Silurian stratigraphy and facies development in Scandinavia. In The Caledonide Orogen – Scandinavia and Related Areas (eds D. G. Gee and B. A. Sturt), pp. 283–92. John Wiley & Sons. BASSETT, M. G., CHERNS, L. & KARIS, L. 1982. The Röde Formation: early Old Red Sandstone facies in the Silurian of Jämtland, Sweden. Sveriges Geologiska Undersökning C793, 1–24. ¨ , J. 1968. Upper Ordovician brachiopods from BERGSTROM Västergötland, Sweden. Geologica et Palaeontologica 2, 1–35. BOUCOT, A. J. & JOHNSON, J. G. 1964. Brachiopods of the Ede Quartzite (Lower Llandovery) of Norderön, Jämtland. Bulletin of the Geological Institutions of the University of Uppsala 42, 1–11. BRENCHLEY, P. J. 1985. Storm influenced sandstone beds. Modern Geology 9, 369–96. BRENCHLEY, P. J. 1989. Storm sedimentation. Geology Today 5, 133–7. BRENCHLEY, P. J. & MARSHALL, J. D. 1999. Relative timing of critical events during the late Ordovician mass extinction – new data from Oslo. Acta Universitatis Carolinae Geologica 43, 187–90. BRENCHLEY, P. J. & NEWALL, G. 1980. A facies analysis of upper Ordovician regressive sequences in the Oslo Region, Norway – a record of galcio-eustatic changes. Palaeogeography, Palaeoclimatology, Palaeoecology 31, 1–38. BRENCHLEY, P. J. & NEWALL, G. 1984. Late Ordovician environmental changes and their effect on faunas. In Aspects of the Ordovician System (ed. D. L. Bruton), pp. 65–79. Palaeontological Contributions from the University of Oslo no. 295. BRENCHLEY, P. J., PICKERILL, R. K. & STROMBERG, S. G. 1993. The role of wave reworking on the architecture of storm sandstone facies, Bell Island Group (Lower Ordovician), eastern Newfoundland. Sedimentology 40, 359–82. CHERNS, L. 1994. A medusoid from the late Ordovician or early Silurian of Jämtland, central Sweden. Journal of Paleontology 68, 716–21. CHERNS, L. & KARIS, L. 1995. Late Ordovician – Early Silurian transgressive sedimentation in the Jämtland basin, central Swedish Caledonides. GFF 117, 23– 30.

Acknowledgements. Kent Larsson (supervisor) and Anders Ahlberg (co-supervisor) are acknowledged for supervision, and help during fieldwork. Mikael Calner is thanked for valuable discussions, comments on earlier drafts of this article, and help during fieldwork in 2001. The referee P. J. Brenchley improved the paper with his advice. Lesley Cherns is acknowledged for linguistic corrections and Lars Karis for discussions and comments. The study was funded by the Royal Physiographical Society (Kungliga Fysiografiska Sällskapet) in Lund, and by the Royal Swedish Academy of Sciences (Kungliga Vetenskapsakademien). I sincerely thank these for their grants, which made it possible to do fieldwork and analysis.



IP address: 192.71.158.2

616


COLLINSON, J. D. & THOMPSON, D. B. 1989. Sedimentary structures. London: Unwin-Hyman, 207 pp. DAHLQVIST, P. & CALNER, M. 2001. A Hirnantian regression recorded in central Sweden. Geological Society of America, annual meeting, Boston, 1–10 Nov. 2001, abstracts with programs 33, A77. DAHLQVIST, P. & CALNER, M. 2004. Late Ordovician palaeoceanographic changes as reflected in the Hirnantian–early Llandovery succession of Jämtland, Sweden. In New Insights into Late Ordovician Climate, Oceanography and Tectonics (eds M. C. Pope and M. T. Harris), pp. 149–64. Palaeogeography, Palaeoclimatology, Palaeoecology 210. DOTT, R. H. & BOURGEOIS, J. 1982. Hummocky stratification: significance of its variable bedding sequences. Geological Society of America Bulletin 93, 663–80. GAYER, R. A. & GREILING, R. O. 1989. Caledonian nappe geometry in north-central Sweden and basin evolution on the Baltoscandian margin. Geological Magazine 126, 499–513. GEE, D. G. & KUMPULAINEN, R. 1980. An excursion through the Caledonian mountain chain in central Sweden ¨ from Ostersund to Storlien. Sveriges Geologiska Undersökning C774, 1–66. GREILING, R. O., GARFUNKEL, Z. & ZACHRISSON, E. 1998. The Orogenic wedge in the central Scandinavian Caledonides: Scandian structural evolution and possible influence on the foreland basin. GFF 120, 181–90. JAANUSSON, V. 1982. Introduction to the Ordovician of Sweden. In Field excursion guide, IV International symposium on the Ordovician System (eds D. L. Bruton and S. H. Williams), pp. 1–9. Palaeontological Contributions from the University of Oslo 279. KARIS, L. 1982. The sequence in the Lower Allochthon of Jämtland. In Field excursion guide, IV International symposium on the Ordovician System (eds D. L. Bruton and S. H. Williams), pp. 55–63. Palaeontological Contributions from the University of Oslo 279. KARIS, L. & LARSSON, K. 1982. Jämtland road-log. In Field excursion guide, IV International symposium on the Ordovician System (eds D. L. Bruton and S. H. Williams), pp. 64–76. Palaeontological Contributions from the University of Oslo 279. ¨ , A. 1998. Beskrivning till KARIS, L. & STROMBERG berggrundskartan o¨ ver Jämtlands län. Del 2: Fjälldelen. Sveriges Geologiska Undersökning Ca53, 1–363. KREISA, R. D. 1981. Storm-generated sedimentary structures in subtidal marine facies with examples from the middle and upper Ordovician of southwestern Virginia. Journal of Sedimentary Petrology 51, 823–48. ¨ , G. 1872. Förteckning p˚a siluriska koraller fr˚an LINDSTROM Jemtland, ssamlade av Dr G. Linnarsson. Geologiska Föreningens i Stockholm Förhandlingar 1, 90–3. MARTINSSON, A. 1970. Toponomy of trace fossils. In Trace fossils (eds T. P. Crimes and J. C. Harper), pp. 323–30. Geological Journal, Special Issue 3. MCRORY, V. L. & WALKER, R. G. 1986. A storm and tidallyinfluenced prograding shoreline – Upper Cretaceous Milk River Formation of Southern Alberta, Canada. Sedimentology 33, 47–60. OWEN, A. W., HARPER, D. A. T. & JIA-YU, R. 1991. Hirnantian trilobites and brachiopods in space and time. In Ordovician Geology (eds C. R. Barnes and



S. H. Williams), pp. 179–90. Geological Survey of Canada 90–9. PETTIJOHN, F. J., POTTER, P. E. & SIEVER, R. 1987. Sand and Sandstone. New York: Springer-Verlag, 553 pp. ¨ ¨ , P., ROBERTS, D. & BJORKLUND , L. 2004. PLINK-BJORKLUND The Baltic Basin stratigraphic record: reflections of tectonothermal and magmatic events in the Scandinavian Caledonides and Trans-European Suture Zone. The 26th Nordic Geological winter meeting, Abstracts, GFF 126, 83. ROBERTS, D. 2003. The Scandinavian Caledonides: event chronology, palaeogeographic settings and likely modern analogues. Tectonophysics 365, 283–99. ROBERTS, D. & GEE, D. G. 1985. An introduction to the structure of the Scandinavian Caledonides. In The Caledonide Orogen – Scandinavia and Related Areas (eds D. G. Gee and B. A. Sturt), pp. 55–68. John Wiley & Sons. SAMUELSSON, J. 1997. Petroleum potential and thermal history of the Caledonian foreland basin, Sweden. Chalmers Tekniska Högskola, Geologiska Institutionen A83, 1–61. SEILACHER, A. & AIGNER, T. 1991. Storm deposition at the bed, facies, and basin scale: the geologic perspective. In Cycles and Events in Stratigraphy (eds G. Einsele, W. Ricken and A. Seilacher). Berlin, Heidelberg: Springer-Verlag, 955 pp. SHEEHAN, P. M. 2001. The Late Ordovician mass extinction. Annual Review of Earth and Planetary Sciences 29, 331–64. STANISTREET, I. G. 1983. Contemporaneous faulting in the Upper Ordovician of the Oslo–Asker district, Norway, and its significance in the development of the Oslo basin. Sedimentary Geology 37, 133–50. SUTCLIFFE, O. E., DOWDESWELL, J. A., WHITTINGTON, R. J., THERON, J. N. & CRAIG, J. 2000. Calibrating the Late Ordovician glaciation and mass extinction by the eccentricity cycles of earth’s orbit. Geology 28, 967–70. TEMPLE, J. T. 1965. Upper Ordovician brachiopods from Poland and Britain. Acta Palaeontologica Polonica 10, 379–427. THORSLUND, P. 1940. On the Chasmops Series of Jemtland and Södermanland (Tvären). Sveriges Geologiska Undersökning C436, 1–191. THORSLUND, P. 1943. Gränsen ordovicium – silur inom Storsjöomr˚adet i Jämtland. Sveriges Geologiska Undersökning C455, 1–19. THORSLUND, P. 1948. De Siluriska lagren ovan Pentameruskalkstenen i Jämtland. Sveriges Geologiska Undersökning C494, 1–37. THORSLUND, P. & JAANUSSON, V. 1960. The Cambrian, Ordovician, and Silurian in Västergötland, Närke, Dalarna and Jämtland, central Sweden. Guide to Excursions Nos. A23 and C18, International Geological Congress 21 Session, Norden 1960, Sweden, Guide Book, 1–51. Stockholm. TORSVIK, T. H. 1998. Palaeozoic Palaeogeography: a North Atlantic viewpoint. GFF 120, 109–18. TUCKER, M. E. 2001. Sedimentary petrology. Oxford: Blackwell Scientific Publications, 262 pp. WIMAN, C. 1893. Ueber die Silurformationen in Jemtland. Bulletin of the Geological Institution of the University of Uppsala 1, 256–76.

IP address: 192.71.158.2