On the lower boundary of the Floian Stage in Estonia

The lower boundary of the Second Stage of the Lower Ordovician Series, the Floian Stage, is defined by the first appearance of the graptolite Tetragraptus approximatus. In the stratotype section of Diabasbrottet at Hunneberg, southern Sweden, the boundary falls within the Megistaspis planilimbata trilobite Zone and within the Oelandodus elongatusAcodus deltatus deltatus conodont Subzone of the Paroistodus proteus Zone, and within the Hunneberg Regional Stage. Graptolites, including the index species T. approximatus, are missing in the terrigenous sediment of Estonia. The earliest trilobites are poorly preserved in this terrigenous succession. They represent an interval older than the Megalaspides dalecarlicus Zone but probably younger than the M. planilimbata Zone, and are correlative to the Prioniodus elegans conodont Zone. Thus the lower boundary of the Second Stage can be detected using mainly conodonts. In accordance with conodont occurrence, the base of the Floian (i.e. somewhat higher than the base of the O. elongatusA. d. deltatus Subzone) falls within the Joa Member of the Leetse Formation. Probable levels have been suggested in some localities. Thus the lower part of the Leetse Formation belongs to the Tremadoc Stage, and to the Hunneberg Regional Stage.


INTRODUCTION
The lower boundary of the Floian Stage in Estonia has been set at different levels.Viira (2011) correlated it with the base of the Oelandodus elongatus-Acodus deltatus conodont Subzone.In a recent overview of conodont diversity in the northern Baltic, Männik & Viira (2012, fig.2) drew that boundary at the lower boundary of the Prioniodus elegans Zone.This position, however, is incorrect or too approximate, considering the graptolite, trilobite and conodont distribution and the correlation data in the stratotype section in Sweden.In some other recent reports on the Ordovician stratigraphy of Estonia, the resolution is lower and a particular level is not bonded with lithostratigraphical units (e.g.Nõlvak et al. 2006Nõlvak et al. , 2007)).The aim of our paper is to specify the position of the lower boundary of the Floian Stage in Estonia.
The GSSP of the Second Stage of the Lower Ordovician Series, the Floian Stage, has been set in the Diabasbrottet section at Hunneberg in Västergötland, southern Sweden, and is defined by the first appearance of the graptolite Tetragraptus approximatus Nicholson (Bergström et al. 2004(Bergström et al. , 2009)).In this stratotype section T. approximatus first appears at the same level with T. phyllograptoides Strandmark -just above the limestone bed with the trilobite Megistaspis planilimbata (Angelin) and conodonts of the Oelandodus elongatus-Acodus deltatus deltatus Subzone of the Paroistodus proteus Zone (Löfgren 1993(Löfgren , 1996;;Maletz et al. 1996;Löfgren & Bergström 2002).Tetragraptus approximatus is an index of the nominal zone of the Pacific Province and is relatively rare in comparison with T. phyllograptoides, which has been the index for the graptolite biozone in Baltoscandia since Törnquist's times (Törnquist 1901).Unlike the latter species, T. approximatus is globally distributed, providing a valuable correlation datum (Cooper & Lindholm 1990).Regionally, these two taxa appear at different levels.For instance, they come in on the same level in the Mossebo section nearby the GSSP (Maletz et al. 1996), but in Scania T. phyllograptoides appears lower than T. approximatus (see, e.g., Maletz & Ahlberg 2011, fig. 2).Only a few graptolites, including T. phyllograptoides, have been recorded in the Leetse Formation in the St Petersburg Region neighbouring Estonia (Tolmacheva et al. 2001;Suyarkova & Koren′ 2009).Tetragraptus phyllograptoides appears there in the middle of the range of O. elongatus (Lindström) where the sediment becomes more clayey.It co-occurs with Didymograptus rigoletto (Maletz, Rushton & Lindholm) (see Tolmacheva et al. 2001, figs 2, 3), which is found somewhat higher in the Diabasbrottet section of Västergötland (Maletz et al. 1996, fig. 4) or together with T. approximatus in the Lerhamn drill core of Scania (Maletz & Ahlberg 2011, fig. 2).Consequently, in the St Petersburg Region, the base of the Floian Stage is marked by the appearance of these graptolites.Graptolites are lacking or very rare in most of the Ordovician succession of Estonia, thus here the base of the Floian cannot be spotted using graptolites.However, the type section as well as other sections nearby exposes both the faunas of the graptolitic facies and of the shelly facies, allowing correlations between regions where only one of those can be found.The correlation of limestone sections can be made via distribution of conodonts and trilobites.

TRILOBITES
The trilobites in Diabasbrottet, Hunneberg and elsewhere in Sweden were studied in detail by Tjernvik (1956).Jaanusson (1982) used his zonation to define the Baltoscandian regional stages.From base to top, the Megistaspis armata and M. planilimbata zones correspond to the Hunneberg Regional Stage the succeeding M. aff.estonica, Megalaspides dalecarlicus and Megistaspis estonica zones to the Billingen Regional Stage (Fig. 1).Megistaspis planilimbata occurs in the thin limestone beds preceding and following the shales with T. approximatus and T. phyllograptoides at the GSSP.Higher, T. approximatus continues alone, but the graptolite zone is still called the T. phyllograptoides Zone despite lack of the index species (see comments in Lindholm 1991; Maletz et al. 1996, p. 139).Based on the correlation of graptolites with trilobites, consequently the lower boundary of the Floian Stage is somewhere above the lower boundary of the M. planilimbata Zone, in the middle of the Hunneberg Regional Stage (Fig. 1).Trilobites have not been investigated in the beds with P. gracilis Lindström in the stratotype section, but from Mossebo we can learn that M. planilimbata appears in the P. gracilis Zone (Maletz et al. 1996, figs 4, 5).The lower boundary of the Oelandodus elongatus-Acodus deltatus deltatus conodont Subzone falls somewhere in the middle of the M. planilimbata Zone in the Diabasbrottet section, Västergötland.The same situation is observed in more calcareous outcrops at Sjurberg, Dalarna (compare sections in Tjernvik 1956, p. 166, fig. 23;Löfgren 1994Löfgren , p. 1352, fig. 3, fig. 3).This fact shows that evidently the base of the Floian is above the lower boundary of the M. planilimbata Zone over the entire basin.
uppermost trilobite zone of the Billingen Regional Stage.In Estonia the first trilobites appear at a level where sandy terrigenous rocks are replaced by calcareous ones, i.e. at the boundary between the Joa and Mäeküla members of the Leetse Formation.This level is evidently geographically diachronous, pending in the Prioniodus elegans and Oepikodus evae zones (Pärnaste 2003(Pärnaste , 2006a;;Viira et al. 2006).In the Uuga section, northwestern Estonia, the boundary between the Joa and Mäeküla members reaches down into the P. proteus Zone (Löfgren et al. 2005, figs 2, 4).The calcareous beds of the Mäeküla Member Leetse Formation) contain very rare, poorly preserved trilobites, which were recently restudied (Pärnaste 2003(Pärnaste , 2006a(Pärnaste , 2006b;;Pärnaste et al. 2009).The taxonomic association of trilobites in Estonia differs from that in Sweden (Pärnaste 2006a;Pärnaste et al. 2012;Bergström et al. 2013) and complicates the correlation between these regions.Mainly, the research (of H. P.) was stuck at the unpublished new species mentioned by Tjernvik as being characteristic of the M. aff.estonica Zone in Sweden (Tjernvik & Johansson 1980).These taxa need to be studied and compared with specimens from Estonia and the St Petersburg Region to improve the correlation based on trilobites.It has been established that in the East Baltic the Megalaspides dalecarlicus Zone (based on the occurrence of its index species) occurs in the Vassilkovo beds in the uppermost Leetse Formation above the Mäeküla beds in NW Russia.The lower Mäeküla beds possibly represent the preceding Megistaspis aff.estonica Zone (Pärnaste 2006a).Note that the M. aff.estonica Zone of Tjernvik & Johansson (1980) corresponds to the topmost part of the M. planilimbata Zone in sense of Tjernvik (1956).This transitional bed was separated from the M. planilimbata Zone of the Hunneberg Stage and included into the Billingen Stage because of the new arrival of the trilobite fauna known from Billingen (Tjernvik & Johansson 1980).However, so far the real M. planilimbata (and that zone) is not known from Estonia and the lower boundary of the Floian cannot be assessed using the trilobite occurrences.

CONODONTS
The conodonts of the type and neighbouring sections at Hunneberg were studied by Löfgren (1993).The conodont zonation of the Hunneberg and Billingen succession, previously united into the Latorp Stage (Tjernvik 1956;Jaanusson 1960), is based on her work not only in the type area, but also in Jämtland, Siljan District, Finngrundet drill core and elsewhere, and is formally defined by its boundaries (Löfgren 1994).Four successive subzones are distinguished within the Paroistodus proteus Zone as follows (from base to top): the Drepanoistodus aff.D. amoenus Subzone, the Tripodus Subzone, the Paracordylodus gracilis Subzone and the Oelandodus elongatus-Acodus deltatus deltatus Subzone.In the Diabasbrottet section the base of the Floian, i.e. appearance of T. approximatus, is above the lower boundary of the O. elongatus-A.deltatus deltatus Subzone, in the lower part of the M. planilimbata trilobite Zone (Bergström et al. 2004, fig. 10).
In Estonia, the Paroistodus proteus Zone is recognized in glauconitic sandstones of the Leetse Formation.It extends from the basal Klooga Member to the middle or upper part of the Joa Member and is followed by the Prioniodus elegans Zone (Viira et al. 2001(Viira et al. , fig. 2, 2006, fig. 3, fig. 3).The species content and the combination of the occurrences of the subzones and their extent vary between localities and differ slightly from those in Sweden, where limestones and shales constitute this interval and four subzones were originally established (Löfgren 1993(Löfgren , 1994)).The lowermost, Drepanoistodus aff.D. amoneus Subzone is missing or the sediment containing fauna characteristic of this subzone was reworked and mixed with that of the following subzone(s) in northern Estonian sections.The second, Tripodus Subzone is recognized in the Cape Pakri, Keila-Joa, Mäekalda, Saka, Ontika and probably some other sections (Löfgren et al. 2005;Viira et al. 2006 and references therein).Findings of the index species are rare, but those of Paltodus subaequalis Pander, another characteristic taxon of this zone, are more common.The third, Paracordylodus gracilis Subzone has been established on the basis of a single specimen in the Saka section only, in a bed with calcareous nodules in the lower third of the Leetse Formation (Viira et al. 2006, fig.2).That species may reach up to the Oepikodus evae Zone in Sweden (Löfgren 1994).The O. elongatus-A.d. deltatus Subzone is defined by the first appearance of O. elongatus (Lindström) within the range of A. d. deltatus Lindström, although the lowest record of the latter conodont comes already from the Tripodus Subzone (e.g. in the Siljan district, Sweden : Löfgren 1994).Oelandodus elongatus appears in the lower part of the Leetse Formation, below the clayey bed with T. phyllograptoides in the St Petersburg Region (Tolmacheva et al. 2001).In Estonia, O. elongatus is very rare and may appear locally as high in the section as in the P. elegans Zone (e.g.Saka section: Viira et al. 2006).In the Mäekalda section (Viira et al. 2001, fig. 4), O. elongatus occurs in sample Mä95-4, marking the lowest possible level for the base of the Floian Stage.In the Saka section, the level above the calcareous nodules (sample A, Viira et al. 2006, fig.2), carrying the P. gracilis elements and below the FAD of P. elegans at about the lowest quarter of the Leetse Formation, may correspond to the O. elongatus-A.d. deltatus Subzone.Possibly, the approximate level for the lower boundary of the Floian lies somewhere there.In the Uuga section (Löfgren et al. 2005, fig. 4), the conodonts defining the uppermost two subzones have not been found so far.The records of Acodus deltatus upwards from sample Pa-96-7 in the lower third of the Leetse Formation gives us only an approximate lowest limit where the base of the Floian could be; it can be anywhere between that sample and sample Pa-96-20, where P. elegans appears.

CONCLUSION
The lower boundary of the Floian Stage in Estonia cannot be detected either by graptolites or by trilobites due to the sedimentation conditions unfavourable for preservation of these fossils.However, conodont distribution gives us an alternative option to define an approximate position of this boundary through the entire Baltoscandia to Estonia.The lower boundary of the Floian lies somewhere close to the lower boundary of the Oelandodus elongatus-Acodus deltatus deltatus Subzone within the Leetse Formation.Thus the lower portion of the Leetse Formation belongs to the Tremadoc Stage, and to the Hunneberg Regional Stage.