当前位置:首页 >> 电力/水利 >>

Sedimentology and sequencestratigraphy of a pronounced Early Ordovician sea-level


Sedimentary Geology 224 (2010) 1–14

Contents lists available at ScienceDirect

Sedimentary Geology
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s e d g e o

Sedimentology and sequence stratigraphy of a pronounced Early Ordovician sea-level fall on Baltica — The Bj?rk?sholmen Formation in Norway and Sweden
Sven Egenhoff a,?, Chris Cassle a, J?rg Maletz b, ?sa M. Frisk c, Jan Ove R. Ebbestad d, Konstanze Stübner e
a

Department of Geosciences, Colorado State University, 322 Natural Resources Building, Fort Collins, CO 80523-1482, USA St. Francis Xavier University, Department of Earth Sciences, P. O. Box 5000, Antigonish, Nova Scotia, Canada B2G 2W5 Department of Earth Sciences, Palaeobiology, Uppsala University, Villav?gen 16, SE 752 36 Uppsala, Sweden d Museum of Evolution, Uppsala University, Norbyv?gen 16, SE 752 36 Uppsala, Sweden e Department of Earth Sciences, Dalhousie University, Room 3006, Life Science Center, Halifax, Nova Scotia, Canada B3H 4J1
b c

a r t i c l e

i n f o

a b s t r a c t
The Bj?rk?sholmen Formation consists of interbedded carbonates, shales, and glauconitic beds and is characterized by heavy bioturbation and few preserved sedimentary structures. The unit shows ?ve facies shale, glauconitic packstone, and three predominantly mud-dominated carbonate facies. Carbonates and shales are arranged in small-scale deepening-upward cycles. A minimum of fourteen of these small-scale cycles are recognized in the Bj?rk?sholmen Formation. They are arranged in stacks of 3 to 5, forming a total of four medium-scale cycles separated by decimeter-thick shale units. Based on the predominance of mud-rich facies the succession is interpreted to have been deposited in an overall tranquil setting during one mayor sea-level fall and subsequent initial rise of third order. Timeestimates suggest that the 14 small-scale cycles fall into the Milankovitch band of precessional forcing, and the overriding medium-scale cycles likely represent short eccentricity. The sequence stratigraphic interpretation shows that the Bj?rk?sholmen Formation is characterized by falling stage, lowstand and initial transgressive systems tracts. Consequently, the contact between the Bj?rk?sholmen and the underlying Alum Shale Formation represents the basal surface of forced regression. The maximum regressive surface is de?ned by a hiatus in the ?land sections and by shallow-marine packstones within mud-rich distal ramp carbonates in Norway. The top of the Bj?rk?sholmen Formation represents a ?ooding surface at the base of the transgressive systems tract. A comparison of time-equivalent successions worldwide suggests that the Bj?rk?sholmen Formation represents a tectonically-enhanced lowstand with two overriding short-term Milankovitch eustatic signals. Although deposition of the Bj?rk?sholmen Formation coincides with the initiation of a foreland basin in the Caledonides of Norway it remains unclear how these tectonic movements may have lead to the widespread Bj?rk?sholmen lowstand during the Early Ordovician. It is suggested in this study that a combination of compressional forces from Avalonia and the Caledonian margin may have acted in concert to produce an uplift of larger parts of the Baltica plate for a time-span of approximately 0.5 Myr. ? 2009 Elsevier B.V. All rights reserved.

Article history: Received 24 October 2008 Received in revised form 1 December 2009 Accepted 7 December 2009 Available online 16 December 2009 Keywords: Deepening-upward cycles Cool-water carbonate Tectonically-enhanced lowstand Baltica Sequence stratigraphy

1. Introduction The Bj?rk?sholmen Formation (formerly termed “Ceratopyge Limestone”) is one of the most prominent lowstand units exposed in the Scandinavian Ordovician: sandwiched between two siliciclastic mudstones, the Alum Shale Formation below and the T?yen Formation on top, it can be traced for several hundreds of kilometers from the Oslo–Asker area in the west to southern ?land in the east (Tjernvik, 1956; Jaanusson, 1982). The relevant facies of the easternmost
? Corresponding author. Tel.: +1 970 491 0749; fax: +1 970 491 6307. E-mail addresses: sven@warnercnr.colostate.edu (S. Egenhoff), jmaletz@stfx.ca (J. Maletz), asa.frisk@geo.uu.se (?.M. Frisk), jan-ove.ebbestad@evolmuseum.uu.se (J.O.R. Ebbestad), konstanze@dal.ca (K. Stübner). 0037-0738/$ – see front matter ? 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.sedgeo.2009.12.003

outcrops on ?land encompass partly the Djupvik and K?pingsklint formations (Van Wamel, 1974) as de?ned by Stouge (2004), but will for simplicity be referred to the Bj?rk?sholmen Formation during the remainder of this paper. Several studies on principally trilobites deal in depth with the paleontology of the succession (Dalman, 1827; Angelin, 1854; Moberg and Segerberg, 1906; Tjernvik, 1956; Ebbestad, 1999; Frisk and Ebbestad, 2008), however, the sedimentology of this distinct carbonate unit has been left aside to date. The Bj?rk?sholmen Formation records unusual sedimentary conditions that have major implications for reconstructing the Early Paleozoic history of Scandinavia: it is the oldest carbonate unit present in the Ordovician succession, and it also records the most pronounced sea-level fall during the entire Lower Ordovician on Baltica. The reasons for this extreme sea-level lowstand have not been

2

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

explored to date. If this regression was predominantly caused by a eustatic fall such a strong signal should be (a) traceable worldwide and (b) well de?ned also in other areas, especially as the Bj?rk?sholmen Formation is short-lived and encompasses just one trilobite, one conodont and not even an entire graptolite Biozone (Tjernvik, 1956; Ebbestad, 1999; Frisk and Ebbestad, 2008; Maletz and Egenhoff, 2001). The Bj?rk?sholmen Formation also displays hitherto not recognized deepening-upward cycles which are rare in carbonate systems (Lukasik and James, 2003). The cycles are likely triggered by orbital forcing and therefore may re?ect a climate control on sedimentation in the Scandinavian Ordovician succession. The present study aims at exploring the causes of this extreme sealevel lowstand by carefully examining the sedimentology of the Bj?rk?sholmen Formation and comparing it to time-equivalent successions worldwide. Our results are based on detailed facies analysis and stacking patterns of two key regions in Scandinavia, Oslo–Asker in Norway, and southern ?land in eastern Sweden (Fig. 1), and a total of 10 lithological logs and 30 thin sections. 2. Baltica in the Early Paleozoic In the Early Ordovician, Scandinavia was situated at the western rim of the Baltica plate, located approximately between 40 and 50°S (Cocks and Torsvik, 2002, 2005). During the entire Ordovician, large parts of the Baltica microcontinent were covered by an epicontinental sea (Lindstr?m, 1971). Southern Sweden and the Oslo region of Norway formed part of a ramp system on the western ?ank of this terrane facing Laurentia (Egenhoff, 2004). This ramp was inclined to the west with predominantly carbonate production in shallowmarine areas such as ?land and Billingen (V?sterg?tland) in Sweden, and more predominant shale deposition towards the western, deeper side represented by the Oslo region in Norway (Jaanusson, 1976). During the Early Ordovician Baltica was characterized by temperate climatic conditions based on its paleogeographic position which is

also re?ected in faunal assemblages (Cooper et al., 1991). Sedimentation rates on Baltica are low throughout the Lower Ordovician, and the succession is condensed in nearshore as well as in distal ramp settings (cf. Lindstr?m, 1971). 3. Biostratigraphy Three groups of biostratigraphically relevant fossils can be used to de?ne a precise relative age of the upper Tremadocian Bj?rk?sholmen Formation (Fig. 2): trilobites (Ebbestad, 1999; Frisk and Ebbestad, 2008), conodonts (Tjernvik, 1956; Erdtmann and Paalits, 1994; Stouge 2004) and graptolites (Erdtmann, 1965; Maletz, 1999; Maletz and Egenhoff, 2001). Conodonts and trilobites occur in the Bj?rk?sholmen Formation itself as well as in the under- and overlying shales, while the graptolites are restricted to the shales. The Bj?rk?sholmen Formation encompasses the trilobite Biozone of Apatokephalus serratus and represents the top of the Paltodus deltifer conodont Biozone. Based on Scandinavian biostratigraphic data, the Bj?rk?sholmen Formation is delimited by the Kiaerograptus stoermeri Biozone below and the Kiaerograptus supremus Biozone on top. On a worldwide correlation scheme this would correspond to the upper portion of the Aorograptus victoriae Biozone of eastern North America and Bolivia. 4. Sedimentology of the Bj?rk?sholmen Formation The Bj?rk?sholmen Formation consists of predominantly carbonate mud-rich strata and some shales which in the present study are subdivided into ?ve facies discussed in detail below. The focus of this study, however, is on the carbonates and not the shales. In general, all facies with the exception of the packstones of facies D are highly bioturbated leading to gradual boundaries and mixing of originally more distinct depositional facies. The Bj?rk?sholmen lithologies are characterized by only a very limited number of grain types

Fig. 1. Location of the study areas in southern Norway (A and B) and on ?land, Sweden (C); Norwegian localities: Bj = Bj?rk?sholmen, ?v = ?vre ?ren, Sj = Sj?strand, Sl, Sl2 and St = Slemmestad Rortunet Center, Sm = Slemmestad cement factory, Se = Slemmestad harbor; Swedish sections: Dg = Degerhamn, Ot = Ottenby. See also Owen et al. (1990) and Ebbestad (1997) for detailed locality information and additional localities of the Bj?rk?sholmen Formation in the Oslo Region.

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

3

Fig. 2. Biostratigraphy of the Bj?rk?sholmen Formation in its westernmost development based on Ebbestad (1999), Maletz and Egenhoff (2001) and Frisk and Ebbestad (2008). The Bj?rk?sholmen Formation corresponds to the Apatokephalus serratus trilobite Biozone and represents the upper part of the Paltodus deltifer conodont Biozone. The over- and underlying graptolite shales indicate that the Bj?rk?sholmen Formation is delimited by the Kiaerograptus stoermeri Biozone below and the Kiaerograptus supremus Biozone on top which corresponds to the upper portion of the Aorograptus victoriae Biozone of eastern North America and Bolivia.

making a differentiation of facies challenging, especially when thoroughly mixed. Therefore, the Bj?rk?sholmen facies have been grouped into more generalized categories such as mud- to wackestones or wacke- to packstones in order to account for the postdepositional impact of bioturbation. 4.1. Shale (facies A) The shales associated with the Bj?rk?sholmen Formation are gray, greenish or black in color and contain varying amounts of organic matter and carbonate. Shales with high amounts of organic matter form some millimeter-thick laminae intercalated in more grayish or greenish shales; these carbonaceous shales are referred to as subfacies A1. Carbonate occurs in sub-millimeter-thick horizontal layers and on cracks, but also as ?nely dispersed material within greenish and gray shales; such facies with signi?cant carbonate content but without laminae of organic material is referred to as subfacies A2. The gray and greenish shales in both subfacies contain fossil fragments of trilobites, phosphatic and carbonate shells of brachiopods and rare phyllocarids. Graptolites occur but are generally restricted to the black shale laminae within subfacies A1. The shales also contain varying amounts of glauconite and may preserve some bioturbation features. 4.1.2. Interpretation The shales represent relatively quiet water deposition in a distal setting, likely dominated by suspension fall-out although some bedload processes and weak currents may have been involved (see Schieber et al., 2007). The carbonate present in subfacies A1was probably also brought in through suspension clouds and settled in a quiet depositional environment, forming the thin horizontal layers. It was, however, in part remobilized during diagenesis ?lling the fractures in the rock. The environment was generally well ventilated as indicated by the abundance of faunal elements and bioturbation features. However, the accumulation of abundant organic matter with exclusively planktic graptolites preserved in the millimeter-thick dark laminae within subfacies A2 argues for at least temporary more dysoxic and slightly deeper water conditions. 4.2. Trilobite-brachiopod mud- to wackestone (facies B) This facies consists of heavily bioturbated carbonate muds (bioturbation index between 4 and 6 following Taylor and Goldring, 1993) with

minor amounts of grains and is devoid of sedimentary structures. Carbonate grains are mostly trilobite and brachiopod debris (Fig. 3A); less frequently, glauconite occurs (less than 1%) as well as some unknown centimeter-long cone-shaped fossils (calpionellids?) in the ?land sections. Burrows within this facies may be vertically or horizontally oriented. Some are completely ?lled with carbonate mud and therefore distinct in wackestones. However, mud-?lled burrows in mudstones are in places visible through a concentration of submillimeter iron mineral precipitations. Other bioturbations are only partly ?lled with internal sediment and open space is occluded by cement (Fig. 3B), or completely ?lled with cement. Rarely, carbonate pellets are preserved ?lling individual burrows. Within some beds, bioturbation increases towards the top parallel with the shale content and diagenetic overprint of the carbonate. Shell fragments in this facies are generally millimeter to sub-millimeter in size with very few larger particles and often randomly oriented within the matrix, but in places show a rough orientation parallel to bedding. Single beds may show a normal grading from wackestones at the base to mudstone at the top (Fig. 3A). The trilobite-brachiopod mud- to wackestones either show a gradual top contact to overlying shales, or sharp contacts if the overlying lithology consists of glauconite packstones (facies E). In ?land, this facies often displays some centimeter-long vertical and horizontal cracks. The trilobite mud- to wackestones are often heavily recrystallized in the Norwegian sections and the original facies is only preserved within concretions. 4.2.1. Interpretation This facies represents a very low energy environment which is re?ected in the dominance of carbonate mud. It was probably deposited at signi?cant distance from the shore below or close to storm wave base. The random orientation of grains is a product of intense bioturbation which points to abundant benthic life and therefore well oxygenated conditions at the sea?oor. Sedimentation rates during deposition of this facies were probably relatively low. This is re?ected in the gradual transition of the mud- to wackestones into shales, open to only partially ?lled burrows, as well as the intense bioturbation of this facies. 4.3. Trilobite-brachiopod wacke- to packstone (facies C) This facies consist of mostly trilobite debris with varying amounts of brachiopod shells, minor echinoderm remains, very few articulate

4 S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14 Fig. 3. Carbonate facies of the Bj?rk?sholmen Formation; scale bar is 1 mm in all photos (A) trilobite-brachiopod mud- to wackestone (facies B) from Slemmestad, Norway, showing the two dominant grain types; orientation of grains is roughly parallel to bedding; (B) Bioturbation structure in mud- to wackestone (facies B) partly ?lled with internal sediment which is slightly lighter gray than the surrounding carbonate matrix, and in its upper part occluded by calcite cement; ?vre ?ren, Norway; (C) Highly bioturbated wacke- to packstone (facies C) from the Ottenby locality, Sweden; note the dominance of trilobite over brachiopod remains; (D) Stromatactis-type cavity in wacke- to packstones (facies C), Sweden; (E) Mud-poor-packstone (facies D) composed of mostly trilobite debris oriented parallel to bedding, suggesting constantly agitated conditions during deposition, Degerham, Sweden; (F) Trilobite-dominated mud-poor packstone of facies D with shelter pores at the base, and glauconite packstone (facies E) at the top; note local glauconite content in the carbonate, and the amount of shell fragments accumulated as lags in the basal part of the glauconite packstone; (G) Glauconitization of carbonate shell fragments within glauconitic packstone (facies E); the groundmass may have been mostly micrite, but is completely recrystallized, and shows some ghost structures of shell fragments; (H) Rock fragments in clastic dyke (within facies B wackestones), angular with slightly rounded edges, several millimeter in diameter; note strong recrystallization of carbonate matrix within the dyke; (I) Boring in ?rm- to hardground (in facies C wacke- to packstones), ?lled with clay and recrystallized carbonate mud; Ottenby, Sweden; (J) Recrystallized ?ne-grained carbonate with ghost structures of former trilobite remains; Slemmestad, Norway.

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

5

ostracods and some glauconite. The trilobite and shell fragments can still be roughly oriented parallel to bedding or are randomly distributed because of burrowing organisms (Fig. 3C). The degree of bioturbation is high (bioturbation index 4 to 5, Taylor and Goldring, 1993). Bioturbation is predominantly diffuse, but some burrows are distinct with internal sediment at the base and cement ?lling the upper part. The facies does generally not preserve sedimentological structures. The trilobite-brachiopod wacke- to packstones also occur as in?ll of burrows in mud-rich lithologies. Stromatactis-like cavities are common in places (Fig. 3D), and some shelter pores (Fig. 3F) occur. Glauconite occurs as sand-sized grains which are often more common within burrows, and also as ?nely dispersed crystals in a roughly bedding-parallel wavy band crossing shell particles and bioturbations. This facies shows little recrystallization in the Swedish localities in contrast to Norway where patches and zones of enhanced recrystallization obliterates much of the original fabric. Despite ?nely dispersed pyrite some beds of this facies show millimeter-scale vugs likely caused by the preferential dissolution of particles or burrow ?lls. This facies may make up entire beds or show a vertical transition into mud- to wackestones (facies B) and packstones (facies D). 4.3.1. Interpretation This facies was deposited under ?uctuating high and moderate to low energy conditions. High-energy events are recorded in the packstones and moderate to low energy deposition in the wackestones. Even though bioturbation is high, re?ecting abundant benthic life the primary sedimentary differentiation of this facies into shell-rich and mud-rich beds is in many places still clearly distinguishable. The shell-rich beds forming the packstones represent lag accumulations of coarser-grained material during high-energy events, probably storms, when most of the ?ne-material was held in suspension. A remnant of these storm events are the shelter pores preserved below convexupward biogenic grains which have not been tilted by bioturbation. The wackestones represent carbonate deposition from suspension during waning storms and background sedimentation (cf. Kreisa, 1981; Aigner, 1985). The stromatactis pores occurring in some of the thin sections may be related to burrowing (Shinn, 1968) or are the result of microbial growth and decay (Pratt, 1982; Flajs and Hüssner, 1993; Flügel, 2004). It is likely that the sand-sized glauconite grains were brought in by burrowing organisms from mostly overlying strata which is indicated by their preferential presence within burrows. The irregular glauconite bands consisting of sub-millimeter-sized crystals, however, re?ect diagenetic fronts developed through post-depositional precipitation processes as they cross individual grains and bioturbations. 4.4. Mud-poor trilobite-brachiopod packstone These packstones consist of abundant shell fragments of trilobites and brachiopods, cement ?lling the original pore spaces and around 10% carbonate mud (Fig. 3E). It is restricted to bed 2 within the Degerhamn locality in Sweden (cf. Frisk and Ebbestad, 2008). In this section the facies shows an irregular geometry pinching out laterally within the bed. The trilobite-brachiopod packstones show a gradual transition through upward coarsening from underlying wacke- to packstones of facies C, and are overlain by the same facies with a gradual contact. Internally, they also show a decrease in the amount of mud, and a subsequent increase within only a few centimeters vertically. Shells in this facies are oriented in both a convex and concave manner (Fig. 3E) and show varying amounts of carbonate mud on top. Minor amounts of glauconite are present in this facies as sand-sized grains associated with the carbonate mud. Bioturbation features are absent. 4.4.1. Interpretation The dominance of grains in this facies together with the low amount of carbonate mud and the absence of bioturbation indicates

high-energy deposition in a constantly agitated sedimentary environment. Waves and currents re?ected in the preferred convex-up orientation of the shells (Menard and Boucot, 1951; Johnson, 1957; Nagle, 1967) did not distribute shells evenly but lead to rather patchy accumulations so that this facies pinches out laterally. Vertically, the gradual transition of packstones with slightly higher mud content into facies containing less mud and back re?ects an increase and a subsequent decrease in depositional energy. It is likely that this facies transition was caused by longer term base level ?uctuations rather than high-energy events such as storms which would have produced sharp boundaries at the base (cf. Aigner, 1985; Myrow 1992). The sand-sized glauconite has likely been brought in as grains together with the carbonate mud accumulating in the voids between the shells. 4.5. Glauconite wacke- to packstone (facies E) The glauconite wacke- to packstones are characterized by a carbonate matrix that contains varying amounts of shale. The grains in this facies are generally more than 95% glauconite with very few shell fragments. Only in direct contact with underlying, shell-rich carbonates shell concentration in this facies can increase to about 10% of the rock volume (Fig. 3F), and in that case the shells are concentrated in small depressions mixed with the glauconite. The few shell fragments elsewhere in this facies are either recrystallized carbonate or are replaced by glauconite (Fig. 3G). Beds of this facies generally show a knife-sharp, but undulating base. Within individual beds the matrix often contains more shale at the base and increases in carbonate content towards the top. Within the more shale-dominated portions of the matrix some remnant sub-millimeter-scale layering is visible, mostly in the alignment of small shells and subtle variations in carbonate and shale contents. The glauconite grains occasionally show bed-parallel arrangement of several millimeter- to centimeter-thick zones dominated by larger or smaller grains, often slightly disturbed by bioturbation. In some Norwegian localities, clear coarseningupward trends within single beds can be recognized that are likely paralleled by an increase in maturity of the grains from slightly evolved to evolved (following Odin, 1988). Some ?land samples, however, show an opposite trend with highly evolved glauconite at the base and a decrease in maturity up section within a single bed, grading into a mix of evolved and slightly evolved grains near the top. 4.5.1. Interpretation The glauconite wacke- to packstones represent a condensed setting (Amorosi, 1995; Kitamura, 1998) in an overall tranquil environment. This is re?ected in the matrix with its small grain sizes in shales as well as carbonates, the presence of the relic sub-millimeter-scale laminations and the absence of sedimentary structures produced by bed-load transport. Where glauconite is interbedded with carbonate-rich shales deposition took place likely below normal, but above storm wave base, whereas calcite-poor shales likely represent constantly quiet deposition below the reach of storm waves. Because of the overall quiet conditions the glauconite grains have probably not been transported and mostly moved by bioturbation, and therefore have grown at or only slightly below the surface during early diagenesis (Odin, 1988). This is indicated by the absence of erosional features, the lack of sorting within individual layers, the presence of some glauconized shell fragments and the fact that some glauconite grains are ?oating in areas where shaly matrix dominates. The layers of glauconite grains present in certain samples, some of which are characterized by ?ner and some by coarser aggregates, therefore likely represent zones of concurrent growth. It is suggested here that after each of these layers was deposited a time of relative quiescence followed in which the glauconite grew. Following Odin's (1988) stage model the zones with larger glauconite grains therefore re?ect longer quiescence periods than the layers with smaller aggregates.

6

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

The presence of varying amounts of shale and carbonate in the matrix re?ects the position of the sedimentary environment which is transitional between carbonate deposition and offshore mudstones. The carbonate mud is brought into this setting from a more proximal environment mostly in suspension. The more carbonate the matrix contains, the closer this facies was located to the paleo-shoreline. The shale-rich glauconite wacke- to packstones therefore represent the distal portions of this facies grading laterally into non-glauconite containing offshore shales. 5. Post-depositional sediment movement and diagenetic overprint Clastic dykes occur exclusively in the Norwegian locality of Sj?strand (section Sj in Fig. 5). They cut through wackestone and mudstone facies and are ?lled with recrystallized carbonate mud and fragments of the surrounding lithologies (Fig. 3H). Orientation of the dykes is sub-horizontal to vertical. The clasts within these dykes are angular to rounded and still re?ect their original facies, or are preserved as ghost structures within the matrix. The walls con?ning the dykes laterally are smooth in some places and show a signi?cant roughness in others. The dykes formed as a result of dewatering and liquefaction of ?ne-grained carbonate sediment (cf. Allen, 1980; Pratt, 1998). The in part smooth walls delineating the dykes and the rounded clasts ?oating within the matrix indicate that the formation of these dykes must have happened not long after deposition while a larger part of the carbonates was only slightly compacted and still relatively soft. The fact that these dykes are exclusively present in one locality in Norway points to local small earthquakes from movements along a minor fault as a cause for their formation. All carbonate facies of the Bj?rk?sholmen Formation may show diagenetic overprint, mostly in the form of heavy recrystallization and neomorphism of carbonates (Folk, 1965) to a degree that the original facies cannot be discerned any more with con?dence. The degree of overprint is generally stronger in the Norwegian localities than in the ?land sections. In places, however, remnants of shell fragments can still be identi?ed because they have escaped recrystallization, or in the arrangement of newly formed calcite crystals as ghost structures (Fig. 3J) (Flügel, 2004). Other recrystallized carbonates, however, do not preserve any original depositional features. Crystal sizes may vary within one bed which in places likely represents original bedding while in others probably does not re?ect primary lithological differences. From partially recrystallized samples it seems that small crystal sizes represent originally ?ne-grained carbonates such as mud- and wackestones whereas coarsely recrystallized carbonates tend to re?ect packstones. 6. Facies associations The Bj?rk?sholmen Formation is characterized by three facies associations. Facies association 1 consists of shale facies A1 and A2 as well as the carbonate mud- to wackestones and wacke- to packstones of facies B and C (Fig. 4A). Facies association 2 shows shale facies A2 and the mud- to wackestones and wacke- to packstones of carbonate facies B and C, as well as glauconite wacke- to packstones of facies E (Fig. 4B). However, wacke- to packstones of facies C are very rare in facies association 2. Facies association 3 is characterized by shale facies A2, the carbonate mud- to packstones of facies B and C, glauconite wacke- to packstones of facies E, and additionally contains the mud-poor carbonate packstones of facies D (Fig. 4C). Wacke- to packstones of facies C make for about a third of the rock volume in facies association 3. The three facies associations show a distinct stratigraphic and/or geographic distribution pattern: facies association 1 occurs exclusively in the lower two thirds of the Norwegian sections while the upper third is formed by facies association 2 sedimentary rocks. The Swedish sections consist entirely of facies association 3.

7. Stacking patterns and facies distribution The stacking patterns show a wide lateral variation within the Bj?rk?sholmen Formation, not only between the Swedish and the Norwegian successions which are located more than 500 km apart, but also from one location to the next within both study areas. All three facies associations show an internal arrangement in cycles re?ected in repetitive stacks of distinct facies. Three orders of cycles characterize the Bj?rk?sholmen Formation: the thinnest ones are here referred to as “small-scale cycles”, while the intermediate ones consisting of several of the small-scale cycles are called “mediumscale cycles”. The largest one refers to the entire Bj?rk?sholmen Formation consisting of a carbonate unit bounded above and below by shales. Each of the small-scale cycles consists of a sharp, in many cases erosional base overlain by a carbonate bed with a shale on top, and in places glauconite in-between. The top contact between the carbonate and the shale/glauconite may be gradual or sharp. These small-scale cycles are between 1 and a maximum of 25 cm thick; however, some of them most likely represent amalgamation of several small-scale cycles into thicker beds. Many, but not all small-scale cycles can be correlated laterally relatively well between outcrops in Norway, but with much less con?dence between the two Swedish sections. It is far beyond biostratigraphic resolution to know whether any of the smallscale cycles in Norway is equivalent to the ones exposed in Sweden. Within the small-scale cycles, especially in facies association 1 carbonate and shale facies are vertically arranged in a distinct ?ning-upward fashion (Fig. 4A). When well developed, carbonate beds of facies association 1 are characterized by some millimeter- to several centimeter-thick facies C wacke- to packstones overlying an erosional base. These grade vertically into mud- to wackestones of facies B. The upper part of this idealized cycle is formed by calcareous shales (facies A2) overlain by dark, organic-rich shales (facies A1). In the Sj section in Norway, however, a symmetric pattern developed with light calcareous shale grading vertically into dark shale and back into calcareous shale which is overlain by the base of the next cycle (Fig. 5). Many small-scale cycles, though, are not complete, often missing the basal coarse-grained carbonates or the shale portion, the latter resulting in amalgamation of cycles into thicker beds. The thickness of each of the facies varies from cycle to cycle, and also within each cycle laterally; however, the mud- to wackestone portion (facies B) is generally the most pronounced while facies C intervals are generally relatively thin. In facies association 2, the near-absence of coarse-grained carbonate wacke- to packstones of facies C accounts for a less clear ?ning-upward pattern (Fig. 4B). Furthermore, most of the sand-sized grains present in this facies association are glauconite. The glauconite wacke- to packstones of facies E, however, can be intercalated either at the base or between the carbonate and shale part of individual small-scale cycles which also alters an otherwise more pronounced ?ning-upward pattern. The same holds true for facies association 3 (Fig. 4C). Although it contains higher amounts of grain-supported facies than facies association 2, a distinct ?ningupward pattern is often not well developed due to intercalations of glauconite packstones in different levels of the small-scale cycles. Furthermore, the exact position of the mud-poor packstones of facies D within the cycle is unclear as it exclusively occurs in the Swedish Degerhamn locality, and only associated with facies C packstones in one incomplete cycle. Based on the observations from Degerhamn facies D is shown intercalated into facies C packstones in the idealized facies association 3 succession (Fig. 4C). Several of the small-scale cycles are arranged into medium-scale cycles which show thicknesses of between 20 and 60 cm. The medium-scale cycles consist of up to ?ve small-scale cycles. In the Si and Se sections only a single small-scale cycle seems to also represent a medium-scale cycle; however, it s assumed that (1) either the carbonate bed consists of several amalgamated small-scale cycles,

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

7

Fig. 4. The three facies associations characterizing the Bj?rk?sholmen Formation, their vertical facies succession, character of contacts, overall grain size trends (black triangles) and relative sea-level curve to deposit each of the three cycles types; (A) Facies association 1 occurs in the lower portion of the Norway sections, and does not contain any glauconite; (B) facies association 2 characterizes the upper part of the Norway succession, is ?ner grained than the underlying facies association 1, and shows signi?cant amounts of glauconite; glauconite facies can also occur in different stratigraphic positions within a cycle; (C) The ?land sections show coarser facies than Norway; the mud-poor packstones of facies D, however, only occur in the Degerhamn locality (Sweden), and therefore its position within the cycle is not well known, but most likely as shown. One of the Swedish cycles therefore shows a thin shallowing-upward below the more pronounced deepening-upward.

or (2) small-scale cycles are missing due to erosion. Many of the medium-scale cycles are characterized by a distinct pattern within the carbonate portion of the cycle. The carbonate beds generally show a thinning-upward throughout medium-scale cycles. However, a basal cm-thick carbonate bed may exist that accounts for a thin thickeningupward portion underlying the dominant thinning-upward succession. In the lower part of the medium-scale cycle, the shale portion of small-scale cycles is often missing, likely caused by local erosion of shale prior to carbonate deposition, and carbonate beds are amalgamated. In the upper part of medium-scale cycles, however, the shales are more often preserved. Each of the medium-scale cycles culminates in an 8 to 20 cm thick shale unit at its very top that separates it from the

overlying cycle. The Bj?rk?sholmen Formation is characterized by a total of four of these medium-scale cycles (numbered units/mediumscale cycles 1 to 4 in Fig. 5). The medium-scale cycles are only well de?ned in the Norwegian localities while it is dif?cult to recognize them in the Swedish sections. As both, the small-scale and the medium-scale cycles are better re?ected and more easily recognized in the Norwegian sections, an estimate of cycle numbers is exclusively based on the Norwegian succession, not on the Swedish data set. The de?nition of both, the small-scale and the medium-scale cycles, is based on the vertical facies succession. Cycle numbers vary laterally within the Bj?rk?sholmen Formation, especially as many carbonate beds are amalgamated

8 S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14 Fig. 5. Eight measured sections of the Bj?rk?sholmen Formation from Norway (left) and two from ?land (right); the sections show facies (in parenthesis), Dunham classi?cation of carbonates of individual beds or “rk” for recrystallized, two types of possible Milankovitch cyclicity and the sequence stratigraphic interpretation of the succession. The Bj?rk?sholmen Formation consists of decimeter-thick shale intervals that separate carbonate-dominated intervals, and this pattern has been used for lateral correlation of the Norway sections. Each decimeter- to several decimeter-thick shale-carbonate interval represents one medium-scale cycle, while the smaller carbonate beds with shale partings form one small-scale cycle. Centimeter-thick lenticular carbonate beds may represent storm beds and therefore have not been counted as cycles. In the Swedish sections only the small-scale cycles are well developed, while medium-scale cycles are not well developed. Locality abbreviations are the same as in Fig. 1; MS cycle = medium-scale cycle; LST = lowstand systems tract, FSST = falling stage systems tract.

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

9

and may or may not show a clear ?ning-upward trend forming the basis for recognizing individual cycles. Therefore, the maximum amount of well de?ned cycles present in the laterally correlated Norwegian sections was selected for estimating cycle numbers. Based on this approach, the total number of small-scale cycles in the Bj?rk?sholmen Formation is estimated to be around 14, two in unit I (?v and St sections), ?ve in unit II (sections Sl and Bj), four in unit III (section Sj), and three in unit IV (sections Sm and St) (Fig. 5). 7.1. Interpretation 7.1.1. Depositional model — facies belts The Scandinavian succession of the Bj?rk?sholmen Formation shows a separation into two broad facies belts, the nearshore carbonates and the more distally deposited shales (Fig. 6; see also “confacies belts” of Jaanusson, 1976). Despite heavy bioturbation, the carbonate facies belt can be subdivided into three zones which are the mud-poor packstones (facies D) representing the highest energy regime and likely shoreface deposition, with the adjacent wacke- to packstones (facies C) re?ecting an upper offshore setting and the mud-to wackestones (facies B) which characterize the distal portion of the carbonate facies belt adjacent to the offshore shales (Fig. 6). The glauconite packstones (facies E) in places mark the transition between the carbonate and the shale facies belts. The Norwegian sections suggest that glauconite is present only in an overall transgressive regime and not during regressions (see Discussion below), while in the Swedish sections it occurs throughout and cannot be assigned to an overall transgressive regime. However, energy and therewith proximity to shoreline is not always re?ected in facies, but sometimes in the contact surface between carbonates and underlying shales. In sections Se and Bj (Fig. 5) the sharp basal contact of medium-scale cycle 3 as well as several other small-scale cycles in the Sm and Sj sections show signi?cant erosion and scouring into underlying shales. This documents a high-energy environment in contrast to the facies making up the mud-rich carbonate bed directly overlying the erosive base (facies B) deposited under low energy conditions. This surface is therefore very likely a bypass surface showing that some carbonate may have been transported through this area and deposited further basinward. The ?rm- to hardgrounds developed on some of the ?land carbonate beds indicate that the eastern part of the study area underwent marine transgressions and condensation periods (see also Tjernvik, 1956). The carbonate dykes are a local phenomenon likely caused by minor synsedimentary tectonic movements. These and

other tectonically-induced features can occur in all facies belts and have no implication on proximal–distal relationships. Although it occurs just in this one locality it clearly indicates that the study area was characterized by at least some tectonic activity. 7.1.2. Cyclicity and sea-level changes The three order of cyclicity observed in the Bj?rk?sholmen Formation mirror sea-level ?uctuations of different length and magnitude. The small-scale cycles are strongly asymmetric and with few exceptions show exclusively sediment deposited during sea-level rise. Their facies succession therefore re?ects a deepening-upward, and a decrease in depositional energy towards the top. This is characteristic for temperate to cool-water carbonate ramps where sea-level rise easily outpaces carbonate production rates (Lukasik and James, 2003). Sediments deposited during sea-level fall are rarely preserved as they are mostly eroded prior to lowstand carbonate deposition. However, centimeter-thick nodular carbonate beds at the very base of the Bj?rk?sholmen Formation may represent strong storm events that predate continuous carbonate deposition, and therefore re?ect a shallowing portion in some of the cycles (see Discussion). The base of most of the small-scale cycles is sharp and often strongly erosional caused by an at least temporarily high-energy environment above storm wave base that cut into the underlying shales through either wave action or currents. The often ?ne-grained nature of the carbonate sediments in many of the cycles, especially in the upper part of the Norway succession shows that if high-energy events were responsible for the formation of the sharp basal surface they did only rarely deposit any coarse material such as lags, or such larger grains have been mixed in with the mud by later bioturbation. The shales overlying the carbonates re?ect a further decrease in energy and transition from ramp into offshore facies (cf. Burchette and Wright, 1992). While the calcite-rich shales (facies A2) still received some input of carbonate mud from the nearshore facies belt, the shales with organic matter (facies A1) represent deposition in an even further distal environment out of reach of carbonate-rich suspension clouds. The medium-scale cycles are asymmetric and consist of stacked small-scale cycles. Their bases are often de?ned by centimeter-thick nodular carbonate beds intercalated into shales, which develop upsection within a few centimeters into thick carbonate beds with or without millimeter-thick shale intercalations and transition back to thinner carbonate beds and shales (Fig. 5). The thickening and subsequent thinning of limestone beds records the reduction and successive increase of accommodation space (cf. Goldhammer et al., 1987).

Fig. 6. Depositional scheme for the Bj?rk?sholmen Formation modi?ed from Posamentier and Morris (2000). The lateral facies succession on this Scandinavian ramp during the Ordovician shows nearshore mud-poor packstones (facies C and D) grading basinward into wackestones (facies B and C), then mudstones (facies B), and shales (facies A). In contrast to their and other similar schemes for disconnected falling stage systems tract geometries, the Bj?rk?sholmen sedimentary system does not ?ll accommodation space by progradation but passively follows even the slightest sea-level variations which results in deepening-upward small-scale cycles (Fig. 5). The scheme shows the formation of two of these small-scale cycles during an overall loss of accommodation space. Sea-level is at position 1 in the beginning, and then rises slightly to position 2. The Bj?rk?sholmen sedimentary system retrogrades as sea-level rises. The next sea-level fall pushes the system down the ramp pro?le to position 3, and the subsequent slight sea-level rise again results in a retrogradation of the system. As a consequence individual largely disconnected decimeter-thick wedges form.

10

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

The sequence stratigraphic interpretation is based on the facies trends observed within the three cycle types present in the Bj?rk?sholmen Formation, and the amount of time available for deposition of the Bj?rk?sholmen Formation based on biostratigraphic data (see discussion). Whereas the small- and medium-scale cycles represent parasequences, the large-scale cycle forms part of a thirdorder sequence/million-year-scale cycle. In the ?land sections, the lower portion consisting of intercalated shales and limestones shows a shift of facies belts basinwards. This is re?ected in distal ramp mudto wackestone facies B in the carbonates of the lowermost cycle at Degerhamn (Dg in Fig. 5) in comparison to the proximal ramp packstones of the cycle on top (facies D). The overlying carbonate and glauconite lithologies which do not contain signi?cant amounts of offshore shales again represent mud-rich distal ramp facies. The lowest position of sea-level re?ected in facies is therefore present in the second cycle from the base at Degerhamn containing the mudpoor packstones (facies D). The signi?cantly lower number of preserved cycles in the ?land localities in comparison to Norway, however, makes it likely that ?land got exposed at some point during the deposition of the Bj?rk?sholmen Formation. As the mud-poor packstones represent the shallowest facies recorded in the ?land sections, it is well possible that the hiatus is located at the top of the second cycle from the base at Degerhamn (section Dg), which likely corresponds to the top of the ?rst well de?ned cycle at Ottenby, Sweden (section Ot in Fig. 5; cf. trilobite-based correlation in Frisk and Ebbestad, 2008). If this interpretation is correct, then the lower two cycles in Degerhamn and the basal cycle in Ottenby represent the shallowing of the environment during a loss of accommodation space. From a sequence stratigraphic standpoint, this initial phase of deposition of the Bj?rk?sholmen Formation would be interpreted as the early Lowstand Systems Tract (LST) following Posamentier et al. (1988) or the Falling Stage Systems Tract (FSST) according to Coe (2003; based on Plint and Nummedal, 2000). The loss of accommodation space likely exposed the ?land area and deposited sediment wedges further basinwards. The increase of carbonate bed thicknesses on top of the hiatus, in contrast, is interpreted to re?ect slowly increasing accommodation space (cf. Goldhammer et al., 1987; Read and Goldhammer, 1988; Drummond and Wilkinson, 1993). The upper, carbonate portion of the ?land section therefore probably shows the late LST (Posamentier et al., 1988) equivalent to the LST of Coe (2003) characterized by backstepping facies belts. In the Oslo area of Norway, the same general trends are observed as in the Swedish sections. As the onset of carbonate deposition on top of the distal shelf Alum Formation shales represents a basinward shift of facies belts the basal surface of the Bj?rk?sholmen Formation is the sequence boundary, following the traditional sequence stratigraphic nomenclature (Posamentier et al., 1988). As for the Swedish localities, the lower portion of the Norwegian succession forms part of the FSST (Coe 2003) or early LST (Posamentier et al., 1988). The shallowest facies preserved are wackeand packstones of facies C. However, the recrystallized portion in the central part of unit 2 originally likely represented a permeable, graindominated and therefore similar or even more proximal sediment as facies C. The top of this recrystallized unit is therefore interpreted to show the maximum regressive surface (MRS; Catuneanu, 2006; Helland-Hansen and Martinsen, 1996) indicating the lowest position of sea-level recorded in facies within the Norway sections. On top of the MRS, small-scale cycles thicken and successively contain more offshore shales. Together with the massive occurrence of glauconites in the upper part of the succession this clearly indicates a slight increase of accommodation space and re?ects transgressive conditions. This portion of the succession is therefore interpreted as the late LST (Posamentier et al., 1988), or the entire LST following Coe (2003). The top of the Bj?rk?sholmen Formation showing the transition of glauconitic packstones into offshore shales represents a ?ooding surface with the T?yen Shale above representing the Transgressive Systems Tract (TST).

8. Discussion 8.1. Cycle deposition and bounding surfaces The interpretation of the small-scale cyclicity in the Bj?rk?sholmen Formation suggests that most carbonate beds re?ect a lowering of base level and therefore the lowstand of a short-term sea-level ?uctuation. Individual carbonate beds, however, could also have been deposited by one or a series of storms which holds true especially for the only centimeter-thin layers, or represent at least in part fairweather sedimentation which only applies to some of the several centimeter- to decimeter-thick carbonate beds. Similar to tempestites, the nodular centimeter-scale limestone beds show a sharp, in places clearly erosive base and in some cases lag deposits at the very bottom. They are generally strongly bioturbated which often blurs original grain-size trends. Despite the bioturbation two different types of centimeter-thick and mostly nodular carbonate beds can be observed: nodules with an internal ?ning-upward trend, and without any clear trends, the latter being by far the most common. Beds with a ?ning-upward likely represent individual storms or high-energy events, while beds without any preserved structures could represent either event sedimentation or higher in?ux of carbonate mud during a sea-level lowstand. It therefore remains unclear whether massive thin carbonate beds represent one end member of depositional cyclicity, or if they are the product of a more randomly acting process. The several centimeter- to decimeter-thick and laterally traceable carbonate beds, however, generally show clear and repetitive vertical facies trends and therefore more likely represent cycles. Both these clear cycles as well as the possible thinner nodular carbonates interpreted as storm event beds are generally characterized by a sharp basal boundary. However, cycles can show a gradational or sharp upper boundary towards shales, while storm beds seem to be characterized exclusively by sharp upper boundaries (Fig. 7). Both boundary types are heavily overprinted by diagenesis and were often sites of carbonate dissolution which is re?ected in dissolution seams and the abrupt termination of half-dissolved bioclasts at the carbonate–shale boundary. The gradual top contact of carbonate beds is developed in two forms: (a) as millimeter-scale intercalated carbonate and shale layers, or (b) as successively more shale-rich carbonate in a centimeter-thick transition zone (Fig. 7). In both cases, carbonate-dominated layers are intercalated with shale-rich layers. This suggests pulses of sedimentation during either fair-weather or storms, and therefore further supports the cyclic deposition of these several centimeter- to decimeter-thick carbonate beds. The cycles in this Ordovician succession from Scandinavia are deepening-upward cycles which is rarely reported for small-scale carbonate cyclicity (Barnett et al., 2002; Lukasik and James, 2003; Spence and Tucker, 2007). The deepening-upward is a result of the low carbonate production potential of the Scandinavian Ordovician system: The very limited thickness of the Bj?rk?sholmen Formation and also of the entire Scandinavian Lower and Middle Ordovician succession (Jaanusson, 1976, 1982) shows that this cool-water ramp did not form enough carbonate to signi?cantly prograde by itself at any stage of the base level curve. Facies shifts within this system therefore seem to be a pure response to base level variations so that one of the main factors governing carbonate platform successions — sediment supply (Schlager, 1993, 2003) — is not an important controlling parameter on platform architecture. Consequently, this study does not follow the classical sequence stratigraphic de?nition of Van Wagoner et al. (1988) who use ?ooding surfaces to delimit individual parasequences. Instead, the surfaces selected in this study to delimit small-scale cycles are “maximum regressive surfaces” (Catuneanu, 2006) or “sequence boundaries” of 5th order. This approach is more comparable to the alternative de?nition by Spence and Tucker (2007) where individual parasequences do not have to be

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

11

Fig. 7. (A) Upper contact of the carbonate part of a cycle (facies B) to overlying shales, showing abundant dissolution seams; note relatively gradual transition between carbonate and shale, and high carbonate content of shale re?ected in relatively light colors; (B) basal very sharp contact between carbonate-rich shale (facies A2) with much less carbonate content than in (A) and overlying mud- to wackestones (facies B); both SM section, Slemmestad, Norway.

bound by marine ?ooding surfaces but may be de?ned by abrupt facies changes related to either shallowing or deepening. Our study suggests that systems with very low carbonate production rates such as cool-water settings will more likely produce exclusively small-scale deepening-upward cycles where parasequences are bound by regressive surfaces than tropical carbonate systems where the response can be deepening or shallowing depending on the overall sea-level trend (cf. Spence and Tucker, 2007). 8.2. Water depth and magnitude of sea-level changes The Bj?rk?sholmen sediments have likely been deposited in a wide range of water depth from shoreface to below storm wave base. Using the Persian Gulf as an analogue being the most thoroughly studied modern ramp (Purser, 1973; Wagner and Van der Togt, 1973) the shoreface mud-poor packstones in the Bj?rk?sholmen Formation (facies D) likely represent water depth of around 15 m or above (cf. Sindhu et al., 2007), while the wacke- to packstones (facies C) as well as mud- to wackestones (facies B) would equal water depth below ? 15 m, but above ? 40 m where the transition into argillaceous shales (facies types A1 and A2) occurs. The entire carbonate portion of the ramp must therefore have been deposited in water depth of ? 40 m and above based on the comparison with the Persian Gulf. However, the Persian Gulf is not an open ocean environment as the Bj?rk?sholmen Formation, and therefore the real water depth values for the Ordovician example may have been somewhat higher than for the recent counterpart. Using these approximate water depth values, the amplitude of sealevel changes can be roughly estimated. The third-order/million-year ?uctuation encompassing the entire Bj?rk?sholmen Formation shows a transition from shales into likely shoreface carbonates — represented by heavily recrystallized facies in the Norway sections — and back to shales. This argues for amplitudes of at least 40 m using the above estimates. The medium-scale cycles should be considered of having approximately the same amplitude range as the million-year cycle as e.g. the second cycle from the base in the Norway sections also consists of recrystallized shoreface packstones, offshore mud-, wacke- and packstones, and shales. Only the small-scale cycles seem to have lower magnitudes in the 10 to 20 m range as most of them only vary between mud-rich mud- to wackestones (facies B) and shales (facies A). 8.3. Duration and causes of sea-level changes The average length of one cycle can be roughly estimated by combining absolute ages for the Tremadocian and stratigraphic information from the Scandinavian and other time-equivalent successions. The Tremadocian has a total duration of approximately 9.7 Myr (488.3 +/? 1.7–478.6 +/? 1.7; International Commission on Stratigraphy, 2008). In Scandinavia, the Tremadocian is subdivided

into 10 to 11 graptolite biozones (Maletz, 1998; Maletz and Egenhoff, 2001). The average length of each of these biozones is therefore approximately 0.97 to 0.88 Myr. The Bj?rk?sholmen Formation occupies a portion (a third to half?) of the Kiaerograptus kiaeri Biozone (cf. Maletz and Egenhoff, 2001) and may reach up into the overlying Kiaerograptus supremus Biozone (Fig. 2). Therefore, the stratigraphic interval represented by the Bj?rk?sholmen Formation is estimated to be between a third and half a graptolite biozone which is equivalent to approximately 0.29–0.485 Myr. Our study shows that the Bj?rk?sholmen Formation is characterized by about 14 small-scale cycles in the Norway localities (Fig. 5) which have an average duration of 20,710–34,640 years based on the estimates above. These values are roughly in the range of Milankovitch-type precessional cycles (19,000–23,000) even when taking into account that these types of cycles were shorter in the Ordovician than today (Berger and Loutre, 1994). It is very probable, though, that the cycle numbers present in the Norway sections do not represent all the cycles that have been deposited during Bj?rk?sholmen times especially as a detailed correlation of small-scale cycles between the Norway and Swedish sections is not feasible. The Bj?rk?sholmen Formation probably consists of numerous sedimentary wedges stepping down the ramp during sea-level fall. As these wedges are likely in part disconnected from each other not all cycles are expected to show up in a single locality. Therefore the number of small-scale cycles really present within the Bj?rk?sholmen interval is likely signi?cantly higher than 14, maybe above 20 which would still agree with the assumption that the small-scale cycles represent precessional cycles in the Milankovitch band. Signals with Milankovitch periodicities can be produced by both tectonics and/or eustasy (e.g. Cisne, 1986; Goldhammer et al., 1987). Based on our current knowledge, however, only fault-controlled settings in extensional regimes seem to produce small-scale regular cycles similar to their eustatic Milankovitch counterparts (De Benedictis et al., 2007). These tectonically-controlled cycles modeled by De Benedictis et al. (2007) are always shallowing-upward. Therefore, they are not comparable to the Bj?rk?sholmen cycles which also have been deposited on a low relief carbonate ramp, probably not close to any large extensional fault. The vast majority of studies favor a eustatic control for cyclicity with periodicities within the Milankovitch band (e.g. Goldhammer et al., 1987; Grotzinger, 1986; Hinnov and Goldhammer, 1991; Preto et al., 2001; Zühlke et al., 2003, among many others). This suggests that deposition of smallscale cycles within the Bj?rk?sholmen Formation is likely controlled by a climate and not a tectonic signal. The medium-scale cycles consist of between 3 and 5 small-scale cycles that show an overall asymmetrical stacking pattern. While the base of the medium-scale cycles is often sharp and erosive the deepening trend in the upper cycle portion can be gradual, mostly

12

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

expressed in a successive thinning of carbonate beds upsection, but often with sharp contacts between shales directly overlying carbonates. As discussed above, it is likely that not all short-term sea-level ?uctuations have been recorded in the Bj?rk?sholmen Formation (“missing beats” of Goldhammer et al., 1987) and therefore the amount of small-scale cycles building the medium-scale cycles in the Norway sections must be considered a minimum number. Based on the time-span available to deposit the Bj?rk?sholmen Formation (0.29–0.485 Myr, see above), the medium-scale cycles must have had a duration between 72,500 and 121,250 years. This estimate falls into the duration of the short eccentricity Milankovitch cycle. It is therefore concluded that the Bj?rk?sholmen stacking patterns show interwoven precession and eccentricity cycles superimposed on a third-order sea-level ?uctuation similar to the Latemàr model of Goldhammer et al. (1987; however, for a contrasting view on the Latemàr see Kent et al., 2004; Mundil et al., 2003; Zühlke et al., 2003). The reason why the Bj?rk?sholmen Formation preserved the orbital cycles so well is likely a result of the low productivity of this cool-water carbonate ramp (cf. James and Clarke, 1997). In a ramp environment as the Scandinavian Ordovician the carbonate is produced in inner to mid-ramp settings, but not transported into the offshore shale-dominated realm because of distance and overall low energy conditions (cf. Burchette and Wright, 1992). As the carbonate production is very low, the system does not ?ll up accommodation space and therefore cannot prograde in contrast to rimmed carbonate platforms such as the Latemàr. Shifts of facies are therefore a clear indication of base level changes as the system does not have the capacity to produce autocycles (see Strasser, 1991). The Scandinavian hinterland was characterized by very low relief (Cocks and Torsvik, 2005, p. 51) so that the basin received no signi?cant siliciclastic input. Therefore, neither siliciclastic nor carbonate turbidites disturbed the eustatic signal as in other deep shelf and basinal successions (e.g. Maurer et al., 2004). 8.4. Correlation of Norway and Sweden sections The present study only proposes a correlation of large-scale facies trends between the two study areas and does not intent to match neither the medium- nor the small-scale cycles between Norway and ?land. The reasons are that (1) the biostratigraphic resolution within the Bj?rk?sholmen Formation is by far not good enough to allow unequivocal correlation down to such a scale, and (2) the sections are nearly 500 km apart which is beyond reasonable lithostratigraphic correlation limits for an only one meter-thick unit. The following discussion will therefore concentrate on the portion of the thirdorder/million-year cycle the Bj?rk?sholmen Formation represents. When comparing the succession below and above the maximum regressive surface (MRS) in the Norwegian and the ?land sections they have very little in common. In ?land, the section is extremely condensed below the maximum regressive surface (MRS) and consists of more glauconitic packstones and shales than carbonates with some ?rm- to hardground development. This clearly re?ects the little amount of accommodation space available in the ?land localities to deposit sediment during the prograding phase of the system or overall diminishing accommodation space. Once sea-level started to rise again, however, successively more accommodation space was made available. Consequently, the carbonate beds on ?land are thicker above the MRS, and are characterized by mud-rich middle to outer ramp facies (facies B and C). In Norway which represents a more distal location on the platform, the MRS is located in the lower half of the sections. In contrast to ?land the carbonate beds are thickest in the interval around the MRS (medium-scale cycle 2 in Fig. 5), and indications of exposure are lacking. It is therefore well possible that sea-level fall stopped approximately in this location before it reversed. As it did not expose the study area in Norway relatively more accommodation space was available to deposit coarse-grained

nearshore facies than in ?land. Therefore the Norwegian sections were likely characterized by locally more than 10 cm thick beds interpreted to represent former packstones which have been diagenetically altered to coarse sparite. The trend of generally thinner beds above the MRS re?ects less and less input of carbonate material from nearshore areas as facies belts successively moved eastwards up the platform. In both Norway and ?land, the evolution of the Bj?rk?sholmen Formation terminates with thick glauconite beds re?ecting starved sedimentary conditions during the ?nal transgression. 8.5. The Ceratopyge Regressive Event-controlled by tectonics or eustasy? Traditionally, the Bj?rk?sholmen Formation is interpreted as representing the deposits of a sea-level fall abbreviated as “CRE” (Ceratopyge Regressive Event) by Erdtmann (1986). Erdtmann (1986) and Erdtmann and Paalits (1994) recognized that this unit was not the product of a single sea-level oscillation, but represented “a swarm” of lowstands; however, they did not de?ne how many lowstands they recognized. Although their studies do not rely on sedimentological data they argue that the regressive event reached its maximum at the top of the Bj?rk?sholmen Formation. Our detailed sedimentological study shows that the Bj?rk?sholmen Formation indeed consists of several, according to our data set a minimum of 14 small-scale cycles and hence at least 14 lowstands. The top of the Bj?rk?sholmen Formation, however, is clearly de?ned by a transgression, and not as envisioned by Erdtmann (1986) and Erdtmann and Paalits (1994) by a lowstand exposure surface. The major lowstand, in contrast, lies within the formation, in ?land close to its base and in Norway within the lower third to half of the succession. As outlined above the Bj?rk?sholmen Formation consists of a “swarm” of basinwardstepping units overlain by a number of retrograding units. As the Bj?rk?sholmen Formation is prominent and ubiquitous throughout Scandinavia it was thought to represent a eustatic sealevel fall (e.g. Nielsen, 2004), and several authors have recognized a presumably time-equivalent sea-level fall in other basins worldwide: Haq and Schutter (2009) in the Georgina Basin of Australia, Goldhammer et al. (1993) in western Texas, US, and Demicco and Hardie (1994) in the Grove Limestone of the central Appalachians, Maryland, US. It should be noted, however, that it remains unclear to date how well these lowstands really match the exact time of Bj?rk?sholmen deposition as the biostratigraphic precision varies signi?cantly between all of these Lower Ordovician basins. In all reconstructions of Ordovician sea-level changes Tremadocian and Floian sea-level ?uctuations show roughly the same amplitudes (e.g. Haq and Schutter, 2009). This is in signi?cant contrast to the Scandinavian succession where the Bj?rk?sholmen Formation represents the strongest drawdown of sea-level in the Tremadocian. While the Bj?rk?sholmen lowstand may still contain a eustatic component, its extent suggests that this sea-level fall was ampli?ed by regional Scandinavian tectonics. If the Bj?rk?sholmen Formation formed as a consequence of regional Scandinavian tectonics it should be possible to constrain these examining known plate tectonic events. During the Early Ordovician, a foreland basin formed in the Caledonides of Norway related to the loading of the Baltica margin (Greiling and Garfunkel, 2007). This process which re?ects the commencement of signi?cant crustal reorganization started at approximately 479 Myr (Greiling and Garfunkel, 2007) which is slightly older than the Tremadocian–Floian boundary (International Commission on Stratigraphy, 2008). The forebulge of this initial foreland trough formed in late Early to midOrdovician times and clearly postdates the Bj?rk?sholmen lowstand. Also, forebulges are long-lived features that migrate cratonwards during the evolution of foreland basins (e.g. Covey, 1986). The Bj?rk?sholmen lowstand, in contrast, had a duration of less the 0.5 Myr (see above). It is envisioned in this study that compressional movements associated with the initiation of the Early Ordovician Caledonide

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14

13

foreland basin may be responsible for an initial uplift of large parts of the Baltica craton, producing the widespread Bj?rk?sholmen lowstand. One possible mechanism may be intraplate stress which can produce base level changes in the range of over a hundred meters in a few million years (Cloetingh et al., 1985; Cloetingh, 1988). Although sediment load was de?nitely lower than assumed by the Cloetingh model it should be kept in mind that during the Lower Ordovician Baltica was in?uenced by convergent margins on two sides, towards Laurentia and associated island arcs in the northwest and Avalonia in the southwest (both present orientation; Cocks and Torsvik, 2005; Greiling and Garfunkel, 2007). Therefore, changes in stress ?elds affecting both sides of Baltica may have overlapped resulting in a short-lived uplift of the Swedish and Norwegian portions of the craton. Although the causes of the Bj?rk?sholmen lowstand are not well understood, a major change in basin con?guration is mirrored in the composition of the shales under- and overlying this carbonate unit. The Alum Shale Formation underlying the Bj?rk?sholmen has TOC values of 4–10% and more, whereas the T?yen Formation overlying the Bj?rk?sholmen carbonates shows TOC values of 2% and less (Schovsbo, 2001). Deposition of the Bj?rk?sholmen Formation therefore terminated widespread deposition of the most prominent Lower Paleozoic source rock in Scandinavia. This also argues against a purely eustatic ?uctuation as a cause for deposition of the Bj?rk?sholmen Formation and favors interpretation towards a tectonic mechanism that changed the style of the entire basin. 9. Conclusions (1) The Bj?rk?sholmen Formation consists of mostly carbonates, some shale intercalations and glauconite. The carbonate facies is heavily bioturbated throughout and preserves few sedimentary structures. The Bj?rk?sholmen Formation shows ?ve facies: shales, trilobite-brachiopod mud- to wackestones, trilobite-brachiopod wacke- to packstones, mud-poor trilobite-brachiopod packstones, and glauconite wacke- to packstones. The Norwegian localities are characterized by heavier recrystallization of carbonate facies than the ones in Sweden. (2) Carbonate beds generally show a sharp base and varying amounts of grains from bottom to top. The carbonates transition with a sharp or gradual contact into overlying shales. This couplet of carbonates and shales is regarded as one smallscale deepening-upward cycle, 14 of which build the Bj?rk?sholmen Formation in Norway. These small-scale cycles are arranged into stacks of 3 to 6, forming medium-scale cycles separated by decimeter-scale shale beds. (3) The mud-rich nature of the Bj?rk?sholmen sediments re?ects deposition in an overall tranquil setting with a low gradient. In this scenario, ?land represents the proximal and Norway the distal ramp environment. (4) The Bj?rk?sholmen Formation consists of a falling stage (FSST) and a lowstand (LST) systems tract. The contact between the Bj?rk?sholmen and the underlying Alum Shale Formation is de?ned by the very diachronous basal surface of forced regression. The presence of the shallowest facies indicates the position of the maximum regressive surface (MRS). The overlying retrograding parasequences belong to the LST terminated by a transgressive surface (TS) marking the top of the Bj?rk?sholmen Formation. (5) Intercalation of the Bj?rk?sholmen Formation between two thick shale units, the Alum Formation below and the T?yen Formation on top, shows that the succession has been deposited during one mayor sea-level fall of the third order. A time-estimate on the duration of Bj?rk?sholmen sedimentation indicates that two superimposed Milankovitch cyclicities can be observed in the Bj?rk?sholmen Formation in Norway,

the small-scale cycles likely representing precessional, and the units of 3 to 6 small-scale cycles probably representing short eccentricity forcing. (6) The Bj?rk?sholmen Formation represents a regional tectonically-enhanced lowstand. Although deposition of this unit likely coincides with the initiation of a foreland basin in the Caledonides of Norway it remains unclear how these tectonic movements may have lead to an uplift of the Scandinavian part of the Baltica plate for a time-span of about 0.5 Myr. A combination of compressional forces from the Avalonian and the Caledonian margins may have acted in concert resulting in uplift and consequently the widespread Bj?rk?sholmen lowstand during the Early Ordovician.

Acknowledgements We would like to thank David Bruton, Oslo, for introducing us to suitable Bj?rk?sholmen Formation localities in Norway, and Mark Harris and an anonymous reviewer for very detailed and helpful suggestions. References
Aigner, T., 1985. Storm depositional systems. Lecture Notes in Earth Sciences, vol. 3. Springer Verlag, Berlin, Heidelberg. 174 pp. Allen, J.R.L., 1980. Sedimentary structures, their character, and physical basis, volume II. Developments in Sedimentology, vol. 30. Elsevier. 663 pp. Amorosi, A., 1995. Glaucony and sequence stratigraphy: a conceptual framework of distribution in siliciclastic sequences. Journal of Sedimentary Research B65, 419–425. Angelin, N.P., 1854. Palaeontologica Scandinavica. Iconographia Crustacea Formationis Transitionis Fasc II, 21–92. Barnett, A.J., Burgess, P.M., Wright, V.P., 2002. Icehouse world sea-level behaviour and resulting stratal patterns in late Visean (Mississippian) carbonate platforms: integration of numerical forward modelling and outcrop studies. Basin Research 14, 417–438. Berger, A., Loutre, M.F., 1994. Astronomical forcing through geological time. In: de Boer, P.L., Smith, D.G. (Eds.), Orbital forcing and cyclic sequences: IAS Special Publication, vol. 19, pp. 15–24. Burchette, T.P., Wright, V.P., 1992. Carbonate ramp depositional systems. Sedimentary Geology 79, 3–57. Catuneanu, O., 2006. Principles of Sequence Stratigraphy. Elsevier, New York. 386 p. Cisne, J.L., 1986. Earthquakes recorded stratigraphically on carbonate platforms. Nature 323, 320–322. Cloetingh, S., 1988. Intraplate stresses: a tectonic cause for third-order cycles in apparent sea level? In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea-Level Changes — An Integrated Approach. SEPM Special Publication, vol. 42. Tulsa, Oklahoma, pp. 19–29. Cloetingh, S., McQueen, H., Lambeck, K., 1985. On a tectonic mechanism for regional sea level variations. Earth and Planetary Science Letters 75, 157–166. Cocks, L.R.M., Torsvik, T.H., 2002. Earth geography from 500 to 400 million years ago: a faunal and palaeomagnetic review. Journal of the Geological Society 159, 631–644. Cocks, L.R.M., Torsvik, T.H., 2005. Baltica from the late Precambrian to mid-Palaeozoic times: the gain and loss of a terrane's identity. Earth Science Reviews 72, 39–66. Coe, A.L., 2003. The Sedimentary Record of Sea-Level Change. Cambridge University Press, Cambridge UK. 287 pp. Cooper, R.A., Fortey, R.A., Lindholm, K., 1991. Latitudinal and depth zonation of early Ordovician graptolites. Lethaia 24, 199–218. Covey, M., 1986. The evolution of foreland basins to steady state: evidence from the western Taiwan foreland basin. In: Allen, P.A., Homewood, P. (Eds.), Foreland Basins. International Association of Sedimentologists, Special Publication, vol. 8, pp. 77–90. Dalman, J.W., 1827. Om Palaeaderna eller de sa kallade Trilobiterna. Kungliga Svenska Vetenskaps-Akademiens Handlingar (Separat Stockholm 1828). De Benedictis, D., Bosence, D., Waltham, D., 2007. Tectonic control on peritidal carbonate parasequence formation: an investigation using forward tectono-stratigraphic modelling. Sedimentology 54, 587–605. Demicco, R.V., Hardie, L.A., 1994. Sedimentary structures and early diagenetic features of shallow marine carbonate deposits. Society for Sedimentary Geology. SEPM Atlas Series, vol. 1. Tulsa, Oklahoma. 265 pp. Drummond, C.N., Wilkinson, B.H., 1993. Carbonate cycle stacking patterns and hierarchies of orbitally forced eustatic sealevel change. Journal of Sedimentary Petrology 63, 369–377. Ebbestad, J.O.R., 1997. Bj?rk?sholmen Formation (Upper Tremadoc) in Norway: regional correlation and trilobite distribution. In: Stouge, S. (Ed.), WOGOGOB-94 Symposium. Rapport Danmarks og Gr?nlands Geologiske Unders?gelse 1996/98, pp. 27–35. Ebbestad, J.O.R., 1999. Trilobites of the Tremadoc Bj?rk?sholmen Formation in the Oslo Region, Norway. Fossils and Strata 47, 1–118.

14

S. Egenhoff et al. / Sedimentary Geology 224 (2010) 1–14 Menard, H.W., Boucot, A.J., 1951. Experiments on the movement of shells by water. American Journal of Science 249, 131–151. Moberg, J.C., Segerberg, C.O., 1906. Bidrag till k?nnedomen om Ceratopygeregionen med s?rskild h?nsyn till dess utveckling i Fogelsongstrakten. Meddelande Lunds Geologiska F?ltklubb B2, 1–113. Mundil, R., Zühlke, R., Bechst?dt, T., Peterh?nsel, A., Egenhoff, S., Oberli, F., Meier, M., Brack, P., Rieber, H., 2003. Cyclicities in Triassic platform carbonates: synchronizing radio-isotopic and orbital clocks. Terra Nova 15, 81–87. Myrow, P.M., 1992. Bypass-zone tempestite facies models and proximality trends for an ancient muddy shoreline and shelf. Journal of Sedimentary Petrology 62, 99–115. Nagle, J.S., 1967. Wave and current orientation of shells. Journal of Sedimentary Petrology 37, 1124–1138. Nielsen, A.T., 2004. Ordovician sea-level changes: a Baltoscandian perspective. In: Webby, B.D., Droser, M.L. (Eds.), The Great Ordovician Biodiversi?cation Event. Columbia University Press, New York, pp. 84–93. Odin, G.S., 1988. Green Marine Clays. Elsevier, Amsterdam. 445 pp. Owen, A.W., Bruton, D.L., Bockelie, J.F., Bockelie, T.G., 1990. The Ordovician successions of the Oslo region, Norway. Norges Geologiske Unders?kelse (NGU), Special Publication, vol. 4, p. 54. Plint, A.G.,Nummedal,D., 2000. The falling stage systems tract: recognition and importance in sequence stratigraphic analysis. In: Hunt, D., Gawthorpe, R.L. (Eds.), Sedimentary Responses to Forced Regressions. Geological Society London, Special Publication, vol. 172, pp. 1–17 Posamentier, H.W., Morris, W.R., 2000. Aspects of the strata architecture of forced regressive deposits. In: Hunt, D., Gawthorpe, R.L. (Eds.), Sedimentary Responses to Forced Regressions: Geological Society London, Special Publication, vol.172, pp. 19–46. Posamentier, H.W., Jervey, M.T., Vail, P.R., 1988. Eustatic controls on clastic deposition I — conceptual framework. In:Wilgus, C.K., Hastings, B.S., Kendall, C.G.S.T.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea-Level Changes — An Integrated Approach. Society of Economic Paleontologists and Mineralogists (SEPM), Special Publication, vol. 42, pp. 109–124. Pratt, B.R., 1982. Stromatolite decline — a reconsideration. Geology 10, 512–515. Pratt, B.R., 1998. Syneresis cracks: subaqueous shrinkage in argillacous sediments caused by earthquake-induced dewatering. Sedimentary Geology 117, 1–10. Preto, N., Hinnov, L.A., Hardie, L.A., De Zanche, V., 2001. Middle Triassic orbital signature recorded in the shallow-marine Latemar buildup (Dolomites, Italy). Geology 29, 1123–1126. Purser, B.H., 1973. The Persian Gulf. Springer-Verlag, New York. 471 p. Read, J.F., Goldhammer, R.K., 1988. Use of Fischer plots to de?ne third-order sea-level curves in Ordovician peritidal carbonates, Appalachians. Geology 16, 895–899. Schieber, J., Southard, J., Thaisen, K., 2007. Accretion of mudstone beds from migrating ?occule ripples. Science 318, 1760–1763. Schlager, W., 1993. Accommodation and supply — a dual control on stratigraphic sequences. Sedimentary Geology 86, 111–136. Schlager, W., 2003. Benthic carbonate factories of the Phanerozoic. International Journal of Earth Sciences 92 (4), 445–464. Schovsbo, N.H., 2001. Why barren intervals? A taphonomic case study of the Scandinavian Alum Shale and its fauna. Lethaia 34, 271–285. Shinn, E.A., 1968. Practical signi?cance of birdseye structures in carbonate rocks. Journal of Sedimentary Petrology 38 (1), 215–223. Sindhu, B., Suresh, I., Unnikrishnan, A.S., Bhatkar, N.V., Neetu, S., Michael, G.S., 2007. Improved bathymetric data sets for the shallow water regions in the Indian Ocean. Journal of Earth System Science 116, 261–274. Spence, G.H., Tucker, M.E., 2007. A proposed integrated multi-signature model for peritidal cycles in carbonates. Journal of Sedimentary Research 77, 797–808. Stouge, S., 2004. Ordovician siliciclastics of ?land, Sweden. Erlanger geologische Abhandlungen Sonderband 5, 91–111. Strasser, A., 1991. Lagoonal–peritidal sequences in carbonate environments: autocyclic and allocyclic processes. In: Einsele, G., Ricken, W., Seilacher, A. (Eds.), Cycles and Events in Stratigraphy. Springer Verlag, Berlin, Heidelberg. 1040 pp. Taylor, A.M., Goldring, R., 1993. Description and analysis of bioturbation and ichnofacies. Journal of the Geological Society 150, 141–148. Tjernvik, T.E., 1956. On the early Ordovician of Sweden. Bulletin of the Geological Institutions of Uppsala 35, 108–284. Van Wagoner, J.C., et al., 1988. An overview of the fundamentals of sequence stratigraphy and key de?nitions. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.S.T.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea-Level Changes: An Integrated Approach: Society of Economic Paleontologists and Mineralogists (SEPM), Special Publication, vol. 42, pp. 39–45. Van Wamel, W.A., 1974. Conodont biostratigraphy of the Upper Cambrian and Lower Ordovician of northwestern ?land, south-eastern Sweden. Utrecht Micropaleontological Bulletins 10, 1–126. Wagner, C.W., Van der Togt, C., 1973. Holocene sediment types and their distribution in the southern Persian Gulf. In: Purser, B.H. (Ed.), The Persian Gulf. Springer-Verlag, New York. p. 123–155. Zühlke, R., Bechst?dt, T., Mundil, R., 2003. Sub-Milankovitch and Milankovitch forcing on a model Mesozoic carbonate platform — the Latemar (Middle Triassic, Italy). Terra Nova 15, 69–80.

Egenhoff, S., 2004. The Ordovician succession on Baltica — carbonate ramp or epicontinental sea? The 26th Nordic Winter Meeting, Uppsala, Abstract Volume: GFF, vol. 126, pp. 160–161. Erdtmann, B.D., 1965. Eine sp?t-Tremadocische Graptolithenfauna von T?yen in Oslo. Norsk Geologisk Tidsskrift 45, 97–112. Erdtmann, B.D., 1986. Early Ordovician eustatic cycles and their bearing on punctuations in early Nematophorid (planktic) graptolite evolution. Lecture Notes in Earth Sciences 8, 139–152. Erdtmann, B.D., Paalits, I., 1994. The Early Ordovician “Ceratopyge Regressive Event” (CRE): its correlation and biotic dynamics across the East European Platform. Geologija 17, 36–57. Flajs, G., Hüssner, H., 1993. A microbial model for the Lower Devonian stromatactis mud mounds of the Montagne Noire (France). Facies 29, 179–194. Flügel, E., 2004. Microfacies of Carbonate Rocks, Analysis, Interpretation and Application. Springer, Heidelberg. 976 pp. Folk, R.L., 1965. Some aspects of recrystallization in ancient limestones. In: G.M. Friedman and S.A. Ali (Eds.), Diagenesis of Carbonate Rocks: Cement–Porosity Relationships. SEPM, Tulsa, Reprint series number 10, p. 28–62; reprinted 1982 from SEPM Special Publication 13. Frisk, A., Ebbestad, J.O., 2008. Trilobite bio- and ecostratigraphy of the Tremadocian Djupvik and K?pingsklint formations (A. serratus trilobite Zone) on southern ?land, Sweden: GFF, vol. 130, pp. 153–160. Goldhammer, R.K., Dunn, P.A., Hardie, L.A., 1987. High frequency glacio-eustatic sea level oscillations with Milankovitch characteristics recorded in Middle Triassic platform carbonates in northern Italy. American Journal of Science 287, 853–892. Goldhammer, R.K., Harris, M.T., Dunn, P.A., Hardie, L.A., 1993. Sequence stratigraphy and systems tract development of the Latemar Platform, Middle Triassic of the Dolomites (northern Italy); outcrop calibration keyed by cycle stacking patterns. In: Sarg, J.F. (Ed.), Carbonate sequence stratigraphy; recent developments and applications. American Association of Petroleum Geologists Memoir, 57. Tulsa, Oklahoma, pp. 353–387. Greiling, R.O., Garfunkel, Z., 2007. An Early Ordovician (Finnmarkian?) foreland basin and related lithospheric ?exure in the Scandinavian Caledonides. American Journal of Science 307, 527–553. Grotzinger, J.P., 1986. Cyclicity and paleoenvironmental dynamics, Rocknest platform, northwest Canada. Geological Society of America Bulletin 97, 1208–1231. Haq, B.U., Schutter, S.R., 2009. A chronology of Paleozoic sea-level changes. Science 322, 64–68. Helland-Hansen, W., Martinsen, O., 1996. Shoreline trajectories and sequences: description of variable depositional-dip scenarios. Journal of Sedimentary Research 66, 670–688. Hinnov, L.A., Goldhammer, R.K., 1991. Spectral analysis of the Middle Triassic Latemar Limestone. Journal of Sedimentary Petrology 61, 1173–1193. International Commission on Stratigraphy 2008. http://www.stratigraphy.org/cheu. pdf. Accessed 30 November 2009. Jaanusson, V., 1976. Faunal dynamics in the Middle Ordovician (Viruan) of BaltoScandia. In: Bassett, M.G. (Ed.), The Ordovician System. Proceedings of the Palaeontological Association. University of Wales Press, Cardiff, pp. 301–326. Jaanusson, V., 1982. Introduction to the Ordovician of Sweden. In: Bruton, A., Williams, L. (Eds.), Fourth International Symposium on the Ordovician System. University of Oslo, Oslo, Norway, pp. 1–9. James, N.P., Clarke, J.A.D., 1997. Cool-water carbonates. Society for Sedimentary Geology (SEPM), Special Publication, vol.56, 440 pp. Johnson, R.G., 1957. Experiments on the burial of shell. Journal of Geology 65, 527–535. Kent, D.V., Muttoni, G., Brack, P., 2004. Magnetostratigraphic con?rmation of a much faster tempo for sea-level change for the Middle Triassic Latemar platform carbonates. Earth and Planetary Science Letters 228, 369–377. Kitamura, A., 1998. Glaucony and carbonate grains as indicators of the condensed section: Omma Formation, Japan. Sedimentary Geology 122, 151–163. 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, 823–848. Lindstr?m, M., 1971. Vom Anfang, Hochstand und Ende eines Epikontinentalmeeres. Geologische Rundschau 60, 419–438. Lukasik, J.J., James, N.P., 2003. Deepening-upward subtidal cycles, Murray Basin, south Australia. Journal of Sedimentary Research 73, 653–671. Maletz, J., 1998. Das Ordovizium von Rügen und seine pal?ogeographische Stellung aufgrund der Graptolithenfaunen. Habilitation thesis, Ernst-Moritz-Arndt Universit?t, Greifswald, Germany, 185 pp. Maletz, J., 1999. Late Tremadoc graptolites and the base of the Tetragraptus approximatus Zone. In: Kraft, P., Fatka, O. (Eds.), 8th Symposium on the Ordovician System. Acta Universitatis Carolinae, Prague, pp. 25–28. Maletz, J., Egenhoff, S.O., 2001. Late Tremadoc to Early Arenig graptolite faunas of southern Bolivia and their implications for a worldwide biozonation. Lethaia 34, 47–62. Maurer, F., Hinnov, L.A., Schlager, W., 2004. Statistical time-series analysis and sedimentological tuning of bedding rhythms in a Triassic basinal succession (southern Alps, Italy). In: D'Argenio, B., Fischer, A.G., Premoli Silva, I., Weissert, H., Ferreri, V. (Eds.), Cyclostratigraphy, Approaches and Case Histories. Society for Sedimentary Geology (SEPM), Special Publication, vol. 81, pp. 83–99.


赞助商链接
相关文章:
更多相关标签: