1. Introduction
The Late Devonian is an epoch which saw significant changes in the biosphere. For instance, the Frasnian/Famennian boundary (F/F) extinction was one of the five most important events during Phanerozoic times. The Devonian–Carboniferous Hangenberg event represents one of the largest biotic disturbances (for reviews, see House, Reference House2002; Caplan & Bustin, Reference Caplan and Bustin1999). Late Devonian global events, such as the punctata event (Pisarzowska, Sobstel & Racki, Reference Pisarzowska, Sobstel and Racki2006; Yans et al. Reference Yans, Corfield, Racki and Préat2007; Marynowski, Filipiak & Pisarzowska, Reference Marynowski, Filipiak and Pisarzowska2008), Lower and Upper Kellwasser (F/F) events (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001; Racki et al. Reference Racki, Racka, Matyja and Devleeschouwer2002; Bond & Zatoń, Reference Bond and Zatoń2003; Bond, Wignall & Racki, Reference Bond, Wignall and Racki2004; Hartkopf-Fröder et al. Reference Hartkopf-Fröder, Kloppisch, Mann, Neumann-Mahlkau, Schaefer, Wilkes, Becker and Kirchgasser2007; Bond & Wignall, Reference Bond and Wignall2008), the Annulata event (Marynowski, Narkiewicz & Grelowski, Reference Marynowski, Narkiewicz and Grelowski2000; Bond & Zatoń, Reference Bond and Zatoń2003; Racka & Marynowski, Reference Racka and Marynowski2008) or the Hangenberg event (Marynowski & Filipiak, Reference Marynowski and Filipiak2007; Trela & Malec, Reference Trela and Malec2007) are relatively well recognized from the palaeontological and geochemical points of view in the Holy Cross Mountains. All are characterized by the presence of organic-rich deposits, particularly distinctive black shale horizons (sensu Wignall, Reference Wignall1994), with elevated concentrations of organic carbon, in some cases exceeding 20% of total organic carbon (TOC). The Dasberg event, known as a worldwide hypoxic event (Hartenfels & Becker, in press, and references therein), has been relatively little studied. The present work is a continuation of the multidisciplinary research (geochemistry and palynology) concerning development and environmental perturbations recorded in Upper Devonian black shale (Hangenberg and Dasberg) exposed on the northern wall of the Kowala Quarry (see Fig. 1; Filipiak & Racki, Reference Filipiak and Racki2005; Racki, Reference Racki, Over, Morrow and Wignall2005; Marynowski & Filipiak, Reference Marynowski and Filipiak2007).
Figure 1. Simplified geological map of the western and central part of the Holy Cross Mountains (a) with location of the Kowala Quarry (b).
Here we report the first palynological (palynostratigraphy and palynofacies) and geochemical study of the Upper Devonian Dasberg Event section (Fig. 2), especially focused on depositional environments in the Holy Cross Mountains (HCM) area (Fig. 1). The study is particularly interesting due to the immature character of the organic matter (OM) in the Kowala quarry section (Racki, Reference Racki, Over, Morrow and Wignall2005). The relative immaturity of Devonian sedimentary rocks provides an insight into a wide range of sedimentary and biotic aspects which enrich our knowledge of Palaeozoic events. The results have been compared to other Late Devonian event intervals recorded in the Holy Cross Mountains such as the F/F event (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001), D/C (Marynowski & Filipiak, Reference Marynowski and Filipiak2007) and the Early–Middle Frasnian transitions (Marynowski, Filipiak & Pisarzowska, Reference Marynowski, Filipiak and Pisarzowska2008).
Figure 2. The general lithology of Middle/Upper Devonian section of northern wall of the Kowala Quarry, with currently analysed section enlarged. (a) View of the Dasberg part of the section and samples location (VI 2005). (b) Enlarged part of the section with Dasberg black shales and sample location.
2. Geological setting
The well-known large active Kowala Quarry is geologically located in the southern limb of the Gałęzice–Kowala syncline of the Holy Cross Mountains (Fig. 1) (see e.g. Racki et al. Reference Racki, Racka, Matyja and Devleeschouwer2002). Because detailed data concerning the previous stratigraphical and lithological studies of the deep-shelf (basinal) Kowala section have already been presented by Marynowski & Filipiak (Reference Marynowski and Filipiak2007), in the present paper we have described only the lithology of the currently analysed part of the section (Fig. 2a, b). Lithological divisions of the Famennian sequence have been presented by, for example, Szulczewski (Reference Szulczewski1971) and Berkowski (Reference Berkowski2002), and stratigraphical data come mainly from Żakowa, Nehring-Lefeld & Malec (Reference Żakowa, Nehring-Lefeld and Malec1985), Turnau (Reference Turnau1990) and Dzik (Reference Dzik1997, Reference Dzik2006).
2.a. Lithology and lithostratigraphy
The currently analysed part of the Middle/Upper Famennian section at Kowala belongs to sets K and L (Szulczewski, Reference Szulczewski1971; Berkowski, Reference Berkowski2002). Generally, its lithology consists of a rhythmic succession of marly shales, marly limestones and limestones distinctly intercalated with two black shale horizons (Fig. 2a, b).
Above two metres of light grey limestones rhythmically intercalated with darker marly shales, a lower black shale horizon (~55 cm thick), divided into two parts, characterizes an abrupt sedimentary change (Fig. 2b). A nodular limestone parting, 10 cm thick, is visible sandwiched between the black shale. Above this, one metre of another package of rhythmically, thinly bedded limestones and marly shales occurs. It is overlain by a much thinner (~4 cm in thickness), upper black shale horizon. The first, thicker and stratigraphically older black shales, divided by nodular limestone into two parts, are an equivalent of the internationally known Dasberg Event horizon (Fig. 2).
2.b. Dasberg Shales (DBS) in Western Europe: similarities and differences
The most comprehensive data on the lithological development and biostratigraphy of the DBS have been recently presented by Hartenfels & Becker (in press) from the European and North African areas. During bed by bed investigation of several sections, they established the chronostratigraphy, biostratigraphy and eustasy of the global event. Taking the lithological development into account, the sections described from the Rhenish Massif (Germany) (e.g. Effenberg and Oese) are the most similar to the Middle/Upper Famennian sequences of the Kowala Quarry. Generally, a rhythmic succession of marly limestones and marly shales with sudden occurrence of ~30 cm thick black shales is present in the German area. Moreover, the second, thinner (~5 cm thick) dark layer above the DBS in the Effenberg Quarry (Korn, Reference Korn2004; Hartenfels & Becker, in press), similar to that exposed in the Kowala Quarry (Fig. 2), occurs as well. It is notable that unlike in German sections, the DBS in the Kowala Quarry is thicker (~55 cm) and divided into two parts by the nodular limestone layer. However, the general lithological similarities, supported by palynological data, indicate that these Famennian sediments exposed in the Kowala Quarry represent the Dasberg Event interval. Our biostratigraphical data of the section are based exclusively on miospores, which are not as precise as the conodonts or ammonoids used by Hartenfels & Becker (in press); however, a tentative correlation based on standard miospore zonation by Streel et al. (Reference Streel, Higgs, Loboziak, Riegel and Steemans1987) and Avkhimovitch et al. (Reference Avkhimovitch1993) indicates more or less precisely the uppermost postera–lowermost praesulcata interval in the conodont zonation (Fig. 3).
Figure 3. Correlation of the miospore and conodont zonal schemes for the Middle/Upper Famennian of western Europe and Russian Platform with the T/R curve.
3. Materials and Methods
All information about samples and methods is presented in the online Supplementary Material at http://www.cambridge.org/journals/geo.
4. Palynostratigraphy
The palynological study was conducted in order to establish a palynostratigraphical framework and to analyse the relative abundances of particular kerogen components. The two Middle/Upper Famennian miospore Zones were recognized: Diducites versabilis–Grandispora famenensis (VF) and Retispora lepidophyta–Apiculiretusispora verrucosa LV (Fig. 3). Due to changing similarities of investigated miospore assemblages to those characterizing different standard miospore zones, two miospore schemes were used. The older recognized miospore zone (VF) has been assigned to the eastern European zonation by Avkhimovitch et al. (Reference Avkhimovitch, Tchibrikova, Obukhovskaya, Nazarenko, Umnova, Raskatova, Mantsourova, Loboziak and Streel1993), but the younger (LV) assemblage has been assigned to the western European scheme proposed by Streel et al. (Reference Streel, Higgs, Loboziak, Riegel and Steemans1987). The assignment to a particular zone was based on the presence of the zonal index species and characteristic assemblage.
4.a. Diducites versabilis–Grandispora famenensis (VF) Zone
The VF miospore Zone has been recognized in the first three analysed samples D-3, D-2 and D-1 (Fig. 4). The assemblage is taxonomically diverse and contains the index species Diducites versabilis and Grandispora famenensis accompanied by Retispora macroreticulata, Rugospora radiata, Hymenospora intertextus, Diducites poljessicus, Endoculeospora gradzinskii and Grandispora facilis (Fig. 5). Additionally, in the sample D-2, Lophozonotriletes proscurrus has been found, which, according to Avkhimovitch et al. (Reference Avkhimovitch, Tchibrikova, Obukhovskaya, Nazarenko, Umnova, Raskatova, Mantsourova, Loboziak and Streel1993), together with G. facilis mentioned above, is an important taxon for the SP (Spelaeotriletes papulosus; the younger part of the VF Zone) Subzone in the Eastern European miospore scheme. Other, more frequent miospores noticed in this assemblage are: Corbulispora cancellata, Endoculeospora setacea, Raistrickia baculata, Gorgonispora crassa, Knoxisporites hederatus and Cymbosporites acutus. It is of note that Grandispora cornuta, the index taxon for the VCo miospore Zone for Western Europe (see Streel et al. Reference Streel, Higgs, Loboziak, Riegel and Steemans1987; Fig. 3) has been found only in sample D-1.
Figure 4. Stratigraphical range of important miospore and acritarcha species (for lithology explanations see Fig. 2). a – chronostratigraphy; b – palynostratigraphy; c – lithological section; d – samples. LV* indicates a transitional interval to the typical LV Zone.
Figure 5. Famennian miospores (VF-LV Zones). (a) Lophozonotriletes proscurrus Kedo. Kowala sample D1. (b) Spelaeotriletes papulosus (Sennova) Avkhimovitch. Kowala sample D-3. (c) Endoculeospora gradzinskii Turnau, Reference Turnau1975. Kowala D-1. (d) Grandispora famenensis Streel, 1974 (in Becker et al. Reference Becker, Bless, Streel and Thorez1974). Kowala sample D-3. (e) Diducites mucronatus (Kedo) Van Veen, Reference Van Veen1981. Kowala sample D1A. (f) Diducites versabilis (Kedo) Van Veen, Reference Van Veen1981. Kowala sample D-3. (g) Diducites poljessicus (Kedo) Van Veen, Reference Van Veen1981. Kowala sample D1. (h) Rugospora radiata (Jushko) Byvscheva Reference Byvsheva, Menner and Byvsheva1985. Kowala sample D-3. (i) Retispora lepidophyta (Kedo) Playford, Reference Playford1976. Kowala sample D1. (j) Raistrickia baculata Filipiak, Reference Filipiak1996. Kowala sample D-3. (k) Raistrickia baculata Filipiak, Reference Filipiak1996. Kowala sample D1A. (l) Grandispora facilis (Kedo) Avkhimovitch, Reference Avkhimovitch, Byvsheva, Higgs, Streel and Umnova1988 (in Avkhimovitch et al. Reference Avkhimovitch, Byvsheva, Higgs, Streel and Umnova1988). Kowala sample D1. (m) Retispora lepidophyta (Kedo) Playford, Reference Playford1976. Kowala sample D2A. (n) Knoxisporites dedaleus (Naumova) Moreau-Benoit, Reference Moreau-Benoit1980. Kowala sample D-3. (o) Endoculeospora setacea (Kedo) Avkhimovitch & Higgs, Reference Avkhimovitch, Byvsheva, Higgs, Streel and Umnova1988 (in Avkhimovitch et al. Reference Avkhimovitch, Byvsheva, Higgs, Streel and Umnova1988). Kowala sample D-3. (p) Hymenospora intertextus (Nekriata & Sergeeva) Avkhimovitch & Loboziak, Reference Avkhimovitch, Tchibrikova, Obukhovskaya, Nazarenko, Umnova, Raskatova, Mantsourova, Loboziak and Streel1993 (in Avkhimovitch et al. Reference Avkhimovitch, Tchibrikova, Obukhovskaya, Nazarenko, Umnova, Raskatova, Mantsourova, Loboziak and Streel1993). Kowala sample D2A. (q) Retispora lepidophyta? (Kedo) Playford, Reference Playford1976 or R. macroreticulata? (Kedo) Byvscheva, Reference Byvsheva, Menner and Byvsheva1985. Kowala sample D1A. (r) Miospore tetrad. Kowala sample D-3. (s) Hymenospora intertextus (Nekriata & Seergeeva) Avkhimovitch & Loboziak, Reference Avkhimovitch, Tchibrikova, Obukhovskaya, Nazarenko, Umnova, Raskatova, Mantsourova, Loboziak and Streel1993 (in Avkhimovitch et al. Reference Avkhimovitch, Tchibrikova, Obukhovskaya, Nazarenko, Umnova, Raskatova, Mantsourova, Loboziak and Streel1993). (t) Retispora macroreticulata (Kedo) Byvscheva, Reference Byvsheva, Menner and Byvsheva1985. Kowala sample D-3.
In addition to numerous terrestrial components, phytoplankton is very rich and diverse although without stratigraphical value.
4.b. Retispora lepidophyta–Apiculiretusispora verrucosa (LV) Zone
The LV miospore Zone has been recognized in the remaining analysed samples, from D1 to 1KN (Fig. 4). The regular occurrence of the first index species Retispora lepidophyta (Fig. 5) characterizes the beginning of the next LV Zone (Streel et al. Reference Streel, Higgs, Loboziak, Riegel and Steemans1987). Apiculiretusispora verrucosa, a nominal species, was recorded in the D1 and D1A samples only. The miospore assemblage, observed in the samples D1–D3 (Fig. 4), consists of similar species as previously observed in the proceeding VF Zone: Retispora macroreticulata, Grandispora facilis, G. famenensis, Diducites mucronatus, D. versabilis, D. poljessicus, Endoculeospora gradzinskii, E. setacea and Hymenospora intertextus (Fig. 5). Generally, Diducites spp. and Grandispora spp. are the most common taxa among miospores recorded in this level. According to Avkhimovitch et al. (Reference Avkhimovitch, Tchibrikova, Obukhovskaya, Nazarenko, Umnova, Raskatova, Mantsourova, Loboziak and Streel1993) and the previous results (Filipiak, Reference Filipiak2004), the last occurrence of Hymenospora intertextus is noted at the top of the VF Zone, while this taxon is constantly present here, together with ‘younger’ Retispora lepidophyta (D1–D3 samples; Fig. 4). The very rare presence of both index species Retispora lepidophyta and Apiculirtusispora verrucosa (<1%) and constant occurrence of Hymenospora intertextus may indicate that between samples D1 and D3 there is a ‘transitional interval’ (named LV* here; see Fig. 4) between the typical VF and LV miospore Zones. The typical LV miospore Zone (but without A. verrucosa) has been identified in the last two analysed samples (1KN and 2KN). However, it is notable that samples from the upper part of the analysed section (D4–1KN) are poorer in taxa and generally possess miospore assemblages that are less diversified (see Fig. 4). The echinata Subzone (the youngest part of the LV Zone) was previously recognized in this section (see fig. 4 in Marynowski & Filipiak, Reference Marynowski and Filipiak2007), five metres above the currently investigated samples (Fig. 2).
The organic remains are rich in phytoplankton as well but without stratigraphical value. Among the acritarch taxa, Gorgonisphaeridium ohioense is the most common. The frequent occurrence of this species has been noticed previously by Filipiak (Reference Filipiak2005) and Marynowski & Filipiak (Reference Marynowski and Filipiak2007) at the same LV biostratigraphical level.
5. Palynofacies
5.a. Observations
Countable kerogen components have been divided into four categories. Acritarchs together with prasinophytes and leiospheres indicate a marine environment, but miospores and plant tracheids are land-derived (e.g. Batten, Reference Batten, Jansonius and McGregor1996). Leiospheres, as distinct and very common palynomorphs, are excluded from the rest of the prasinophyta group and treated as a separate component here. The relative abundance of acritarchs, along with prasinophytes (Figs 6, 7), and versus miospores, is distinctly variable in the investigated succession, especially when comparing its lowermost and upper parts.
Figure 6. Famennian phytoplankton. (a) Unellium winslowiae Rauscher, Reference Rauscher1969. Kowala sample D1. (b) Unellium cf. piriforme Rauscher, Reference Rauscher1969. Kowala sample D-3. (c) Polyedryxium pharaone Deunf, 1961. Kowala sample 2KN. (d) Veryhachium polyester Staplin, Reference Staplin1961. Kowala sample 2KN. (e) Stellinium micropolygonale (Stockmans & Willière) Playford, Reference Playford1977. Kowala sample D-3. (f) Gorgonisphaeridium discissum Playford, Reference Playford and Dring1981 (in Playford & Dring, Reference Playford and Dring1981). Kowala sample 2KN. (g) Gorgonisphaeridium sp. Kowala sample D-3. (h) Gorgonisphaeridium ohioense (Winslow) Wicander, Reference Wicander1974. Kowala sample 1KN. (i) Solisphaeridium(?) sp. and Gorgonisphaeridium ohioense (Winslow) Wicander, Reference Wicander1974. Kowala sample 2KN. (j) Leiosphaeridia sp. Kowala sample D1. (k) Hemiruptia sp. Kowala sample 1KN. (l) Leiosphaeridia sp. Kowala sample D1A. (m) Cymatiosphaera perimembrana Staplin, Reference Staplin1961. Kowala sample 2KN. (n) Tasmanites sp. Kowala sample 1KN. (o) Maranhites mosesii (Sommer) González, Reference Gonzalez2009. Kowala sample 2KN.
Figure 7. Environmental scanning electron microscope photo of phytoplankton and palynofacies. (a) Gorgonisphaeridium ohioense (Winslow) Wicander, Reference Wicander1974. Kowala sample 1KN. (b) Micrhystridium stellatum Deflandre, Reference Deflandre1945. Kowala sample 1KN. (c) Stellinium micropolygonale (Stockmans & Willière) Playford, Reference Playford1977. Kowala sample 2KN. (d) Maranhites mosesii (Sommer) González, Reference Gonzalez2009. Kowala sample 1KN. (e) Leiosphaeridia sp. Kowala sample 2KN. (f) Gorgonisphaeridium discissum Playford, Reference Playford and Dring1981 (in Playford & Dring, Reference Playford and Dring1981); the small picture in the left upper corner shows ornamentation details. Kowala sample 2KN. (g–j) Palynofacies from the Kowala 1KN sample composed mainly by Leiosphaeridia (L), large Leiosphaeridia (LL) with small amount of miospores (S), Maranhites (M), acritarchs (A), plant tracheid (T), amorphous organic matter (AOM) and organic debris (OD).
Detailed relative changes among the above mentioned components are compiled in Figure 8, but a more precise description of the palynofacies component fluctuations is in the online Supplementary Material at http://www.cambridge.org/journals/geo.
Figure 8. Percentages of kerogen components and abundance of AOM (amorphous organic matter). The dashed area (1KN sample) indicates presences of leiospheres larger than 180 μm.
5.b. Palynofacies interpretation
Taking into account the proportions of various components of the palynofacies, we may suggest that depositional conditions changed slightly with time from a more inshore to offshore environment in the Chęciny–Zbrza intrashelf basin (e.g. Racki et al. Reference Racki, Racka, Matyja and Devleeschouwer2002). Excluding three topmost samples, within the thin black shale at the top of the section (1KN), there is no sharp change in palynofacies composition between the DBS and the rest of the underlying samples. Generally, the amount of terrestrial particles (Fig. 9) is smaller here. However, this could be connected with a deepening phase of the basin and/or diluting effect connected with a great amorphous organic matter (AOM) content (Tyson, Reference Tyson and Jenkins1993; Chow, Wendte & Stasiuk, Reference Chow, Wendte and Stasiuk1995; Fig. 10). The main difference between palynofacies of the DBS (sample D2) and those from the other samples is a complete lack of acritarchs and the smaller-sized leiospheres in the DBS (Figs 8, 10). Thanks to the present geochemical analysis (Fig. 11) it is clear that DBS sedimentation prevailed under anoxic or even euxinic conditions in the water column, which probably restricted phytoplankton growth (mostly acritarchs) in a stressed habitat (e.g. Habib & Knapp, Reference Habib and Knapp1982). The common occurrence of prasinophytes, even in oxygen-depleted environments, can be explained by their presumed green algal nature (e.g. Tappan, Reference Tappan1980; Tyson, Reference Tyson1995). Possessing chlorophyll, they could have thrived in the near-surface photic zone of the sea (Guy-Ohlson, Reference Guy-Ohlson, Jansonius and McGregor1996), far above the unfavourable environmental conditions, in which no other planktic biota could have thrived. Thus, blooming of prasinophycean algae and their subsequent deposition in the greater depths are regarded as good indications of a stressed environment. On the other hand, however, prasinophytes are also more tolerant of stressful conditions than other phytoplankton (Tappan, Reference Tappan1982; Hartkopf-Fröder et al. Reference Hartkopf-Fröder, Kloppisch, Mann, Neumann-Mahlkau, Schaefer, Wilkes, Becker and Kirchgasser2007). Their phycoma permits survival under unfavourable conditions, such as a lack of sunlight, or other environmental/hydrochemical changes, such as anoxia (see Tyson, Reference Tyson1995). Data have also been retrieved from the upper part of the section (D4–2KN) and from the second thinner black shale horizon (1KN). The relative high amount of tracheids in samples D4 and 2KN, together with decreasing amount of similarly land-derived miospores, indicates rather offshore conditions during sedimentation and can be connected with a transgression episode (e.g. Summerhayes, Reference Summerhayes, Brooks and Fleet1987; Tyson, Reference Tyson and Jenkins1993). This hypothesis is supported by the more taxonomically differentiated marine phytoplankton present in samples from this part of the section. According to Johnson, Klapper & Sandberg (Reference Johnson, Klapper and Sandberg1986), the transgression began at the base of the Lower expansa conodont Zone, known as an initiation of the IIf cycle on the Euramerican sea-level curve (see Fig. 3).
Figure 9. Environmental scanning electron microscope photo of charcoal. (a, b, d–g) Kowala sample 2KN; (c) Kowala sample 1KN; internal cellular structure perfectly preserved (see, for example, a–c, g).
Figure 10. Transmitted light microscope pictures of eight selected palynofacies. In the lower left corner is the sample code. Dashed lines divide some picture (c–e) on two parts. On the right side is palynofacies before HNO3 treatment, on the left the same sample after AOM partly removed. Abbreviations: A – acritarcha, AOM – amorphic organic matter, L – leiosphere, LL – large leiosphere (>180 μm), S – miospore, Sc – scolecodont, T – tracheid.
Figure 11. Composite plot of the Dasberg event section showing organic carbon, biomarkers, U/Th and pyrite analyses data. (a) Total organic carbon content – TOC (%). (b) Isorenieratane concentration (μg/g TOC). (c) C13–C22 aryl isoprenoids concentration (μg/g TOC). (d) Gammacerane/C30 17α(H)-hopane ratio. (e) Uranium/thorium (U/Th) ratio. (f) Box-and-whisker plots of the pyrite framboid diameters. Shaded levels responds to the lower and upper Dasberg black shales.
In the last sample (1KN), phytoplankton is the most differentiated, rich in thick-walled and large species of Leiosphaeridia, Maranhites and Tasmanites (Figs 6, 7, 10) with more limited terrestrial components indicating a more open marine environment, characterized by suitable trophic conditions, which could have enhanced bio-productivity. Similar large forms of sphaeromorphs have previously been observed from the F/F boundary interval (Filipiak, Reference Filipiak2002) and from the Middle Frasnian strata (Filipiak in Marynowski, Filipiak & Pisarzowska, Reference Marynowski, Filipiak and Pisarzowska2008) in the Kowala succession. In those intervals their elevated amount was connected with a positive carbon isotopic anomaly (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001; Racki et al. Reference Racki, Racka, Matyja and Devleeschouwer2002; Pisarzowska, Sobstel & Racki, Reference Pisarzowska, Sobstel and Racki2006). The biogeochemical perturbations could indicate exceptionally favourable environmental conditions, increasing nutrient supply for phytoplankton growth (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001; Yans et al. Reference Yans, Corfield, Racki and Préat2007). The slight decrease in terrestrial components is generally observed in samples possessing AOM concentration. According to Tyson (Reference Tyson and Jenkins1993), the percentage of terrestrial organic matter particles is strongly dependent on AOM concentration. In cases where AOM and prasinophytes are well-preserved and richly present, terrestrial components are usually significantly diluted. It is interesting that three samples possessing high concentrations of AOM (D1, D2 and D3; Fig. 10) yield scolecodonts as well (see Fig. 8). However, the abundance of AOM is strongly correlated with areas of low bottom-water oxygen content (Tyson, Reference Tyson and Jenkins1993). Then, the infrequent scolecodonts present in dysoxic–anoxic deposits (e.g. samples D1, D2 and D3) may be explained by their secondary origin, such as from fish fecal pellets (Tyson, Reference Tyson1995), or they are the remains of some polychaete annelids that were well-adapted to an oxygen-depleted environment (Courtinat, Reference Courtinat1998). The miospore tetrads appear rather infrequently as single elements, especially in samples from the lower part of the investigated section (D-2 and D-3), and they are completely absent in all samples from the top of the section. It is obvious that their presence generally increases landward, because they are heavier and often most abundant closer to the parental flora sources (Tyson, Reference Tyson and Jenkins1993).
In summary, all palynofacies components, especially those from the three topmost samples (D4–1KN), show relative changes connected to a progressive deepening (Johnson, Klapper & Sandberg, Reference Johnson, Klapper and Sandberg1986). Generally, the currently analysed palynofacies demonstrate rather small changes in composition of the phytoplankton community from the DBS and underlying strata. The main difference concerns the complete lack of acritarchs in the D2 sample with the constant presence of leiosphere ‘disaster’ taxa (Tappan, Reference Tappan1986), which can be interpreted as an effect of anoxia in the water column. The large sizes of prasinophyta in the 1KN sample can most probably be explained by the onset of favourable trophic ocean conditions (Marynowski, Filipiak & Pisarzowska, Reference Marynowski, Filipiak and Pisarzowska2008).
5.c. Comparison with the Hangenberg Black Shale
See Section 3 of online Supplementary Material text at http://www.cambridge.org/journals/geo.
6. Organic geochemistry
6.a. Bulk geochemical and petrographical data
Total organic carbon (TOC) content differs significantly between samples (Table 1, Fig. 11). Organic-poor limestones and marls (samples D0, D1A, D2A and D4) contain 0.03 to 0.48% TOC (Table 1). Some marls, marly shales and greenish shales contained an average TOC amount (D-2, D-3 and 2KN) ranging from 0.63 to 1.74% TOC (Table 1). Two dark-grey shale samples (D1 and 1KN), characterized by relatively high TOC amounts, ranged from 3.21 to 4.58%, and the main Dasberg black shales (D2 and D3) contain 9.14 and 15.44% TOC, respectively (Table 1). Comparable high TOC amounts, exceeding 10%, were noted from the Hangenberg and Kowala black shales (Marynowski & Filipiak, Reference Marynowski and Filipiak2007), whereas Middle Frasnian black shales and F/F boundary shales and limestones contain generally lower TOC concentrations in the range of 0.5 to 8% (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001; Marynowski, Filipiak & Pisarzowska, Reference Marynowski, Filipiak and Pisarzowska2008). Black shales (samples D2 and D3) are characterized by relatively high HI and low OI (Table 1), which is characteristic for kerogen type II (e.g. Armstroff et al. Reference Armstroff, Wilkes, Schwarzbauer, Littke and Horsfield2006). Such values are typical for Upper Devonian organic-rich facies from the Holy Cross Mountains (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001; Marynowski & Filipiak, Reference Marynowski and Filipiak2007; Marynowski, Rakociński & Zatoń, Reference Marynowski, Rakociński and Zatoń2007).
Table 1. Bulk geochemical data, percentage yields of fractions and basic molecular parameters
TOC – total organic carbon; Carb – carbonates; EOM – extractable organic matter; HI – hydrogen index; OI – oxygen index; AL – aliphatic, AR – aromatic, POL – polar; CPI – carbon preference index; Pr – pristane; Ph – phytane; SCh/LCh – short chain to long chain n-alkanes ratio: (nC17+nC18+nC19)/(nC27+nC28+nC29).
Vitrinite and fusinite reflectance values were measured for sample 1KN in order to obtain the maturity range and compare it to the literature data, and to calculate the approximate temperature of wildfires based on fusinite reflectance (Jones & Lim, Reference Jones and Lim2000).
The vitrinite reflectance values, based on measurements of 70 vitrinite particles, are identical as average values (Marynowski, Czechowski, & Simoneit, Reference Marynowski, Czechowski and Simoneit2001), attaining 0.53% Rr. Fusinite fragments (105 particles measured) showed a relatively wide range of reflectance values, but average values are much higher than that of vitrinite. The values range from 0.62% to 1.88% with mean value of 0.86%. Calculated temperatures of charcoal formation, based on the formula presented by Jones & Lim (Reference Jones and Lim2000), ranged from approximately 257 to 406°C, with a mean value of 286°C. Some additional information on the maturity data and distribution of n-alkanes, isoprenoids and steroids is available in the online Supplementary Material text at http://www.cambridge.org/journals/geo.
6.b. Triterpanes
Hopanes are important constituents of the aliphatic fraction of all samples. The most abundant hopane is C30-17α,21β-hopane, but in some samples, especially from the upper part of the section (1KN and 2KN), the most abundant are 17α-22,29,30-trisnorhopane (Tm) and C29-17α,21β-norhopane (Fig. 12).
Figure 12. Partial mass chromatogram m/z 133 of the three samples from Dasberg section showing the distribution of isorenieratane (filled triangle), 2,3,6-/3,4,5-TM substituted diaryl isoprenoid (open triangle) and their related derivatives including aryl isoprenoids (numbers identify individual carbon number pseudohomologues) in the black shale sample (D3) and lack or trace amount of these compounds in the lower and upper horizons of Dasberg black shale (a) and mass chromatogram m/z 191 showing hopanes distribution in the Dasberg samples (b).
Values for the ratio of the βα-hopane to the sum of βα + αβ-hopane (Peters, Walters & Moldowan, Reference Peters, Walters and Moldowan2005) in almost the total sample set are 0.29±0.1, which is characteristic for low to moderately mature organic matter. Lower values (~0.1) were noted only for samples D0 and D1A, most probably due to their partial weathering (Elie et al. Reference Elie, Faure, Michels, Landais and Griffault2000; see also Fig. 8).
The distribution of homohopanes (C31 to C35 αβ-hopanes) differs significantly between samples. All samples are characterized by a predominance of C31(S+R) homologues but the presence of both C35(S+R) and C34(S+R)-homohopanes was affirmed in only two black shale samples (Fig. 12). Moreover, C33(S+R)-homohopanes are relatively abundant in the black Dasberg shales and almost absent in the rest of the samples (Fig. 12). The lack of higher molecular weight homohopanes is not typical for marine basinal samples (Marynowski, Narkiewicz & Grelowski, Reference Marynowski, Narkiewicz and Grelowski2000) but much more typical for coals and terrestrial organic matter (e.g. Disnar & Harouna, Reference Disnar and Harouna1994).
Values of hopanes/n-alkanes ratios for organic rich samples are comparable to those obtained for the Early–Middle Frasnian transition (Marynowski, Filipiak & Pisarzowska, Reference Marynowski, Filipiak and Pisarzowska2008) and higher than for the F/F transition (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001), showing enhanced bacterial and/or cyanobacterial contributions to the organic material (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001). One exception is the sample D1A, which is partially oxidized/biodegraded with significant participation of the unresolved complex mixture (see Gough, Rhead & Rowland, Reference Gough, Rhead and Rowland1992) and therefore contains lower relative concentrations of n-alkanes.
Organic-rich black shales contain gammacerane (Figs 11, 12) in moderate to minor amounts, indicated by the gammacerane/C30-17α-hopane ratio presented in Table 2. Besides black shales, relatively low amounts of gammacerane were detected in grey shale (sample D1), just below the Dasberg shales (Fig. 11). In the organic-poor samples, gammacerane was not detected.
Table 2. Aryl isoprenoids, isorenieratane, benzo[a]anthracene, benzo[e]pyrene, benzo[b]fluoranthene and sum of pyrolytic PAHs concentrations and molecular parameters based on n-alkanes, hopanes, steranes and aryl isoprenoid distributions
AIR – Aryl isoprenoid ratio: (C13 – C17)/(C18 – C22) (Schwark & Frimmel, Reference Schwark and Frimmel2004)
Ster/17α-hop – regular steranes consist of the C27, C28, C29 ααα(20S + 20R) and αββ(20S + 20R), 17α-hopanes consist of the C29 to C33 pseudohomologue (including 22S and 22R epimers)
Σhop/Σalk – hopanes / n-alkanes ratio
B[a]A – benzo[a]anthracene; B[e]Py – benzo[e]pyrene; B[b]Fl – benzo[b]fluoranthene
Pyrolytic PAHs 1 = sum of benzo[a]anthracene, benzo[e]pyrene, benzo[a]pyrene, benzo[b]fluoranthene and coronene
Pyrolytic PAHs 2 = sum of fluoranthene, pyrene, benzo[a]pyrene, benzo[b]fluoranthene, benzo[ghi]perylene and coronene (Finkelstein et al. Reference Finkelstein, Pratt, Curtin and Brassell2005)
The high abundance of 17α-22,29,30-trisnorhopane (Tm) in comparison to the more thermodynamically stable 18α-22,29,30-trisnorneohopane (Ts) is again consistent with the low maturity of the samples. However, as shown by Bakr & Wilkes (Reference Bakr and Wilkes2002), this parameter is strongly dependent on facies influences.
Values of regular sterane to 17α-hopane ratios are lower than 1, but the highest values are observed for the black shales and their intercalation (0.5–0.6, Table 2). Relatively low values of this parameter are rather untypical in the Late Devonian shelf-basin in the Holy Cross Mountains (Marynowski, Narkiewicz & Grelowski, Reference Marynowski, Narkiewicz and Grelowski2000), particularly in the case of organic-rich samples. For example, the values of ster/17α-hop ratio for the Hangenberg shale ranged between 1.5 and 3.8 (Marynowski & Filipiak, Reference Marynowski and Filipiak2007). Low values of this parameter may suggest intensive bacterial activity after sedimentation (see Marynowski, Narkiewicz & Grelowski, Reference Marynowski, Narkiewicz and Grelowski2000). However, the highest ster/17α-hop ratio noted in the black shales versus rest of the samples, together with high amounts of AOM affirmed by palynological observations (Fig. 8), implies increased input of algal material and primary productivity during the Dasberg shale deposition.
6.c. Isorenieratane derivatives
Isorenieratane and other diagenetic products of isorenieratene are excellent proxies for the assessment of past photic zone euxinia present in the water column (Summons & Powell, Reference Summons and Powell1986; Sinninghe Damsté & Schouten, Reference Sinninghe Damsté and Schouten2005; Sinninghe Damsté & Hopmans, Reference Sinninghe Damsté and Hopmans2008) and are usually used as indicators of past sedimentary conditions (e.g. Köster et al. Reference Köster, Rospondek, Schouten, Kotarba, Zubrzycki, Sinninghe Damsté, Horsfield, Radke, Schaefer and Wilkes1998; Behrens et al. Reference Behrens, Wilkes, Schaeffer, Clegg and Albrecht1998; Pancost et al. Reference Pancost, Freeman, Patzkowsky, Wavrek and Collister1998; Brown & Kenig, Reference Brown and Kenig2004; Kenig et al. Reference Kenig, Hudson, Sinninghe Damsté and Popp2004; Grice et al. Reference Grice, Cao, Love, Böttcher, Twitchett, Grosjean, Summons, Turgeon, Dunning and Jin2005; Marynowski & Filipiak, Reference Marynowski and Filipiak2007; Schwab & Spangenberg, Reference Schwab and Spangenberg2007; Hays et al. Reference Hays, Beatty, Henderson, Love and Summons2007; Heimhofer et al. Reference Heimhofer, Hesselbo, Pancost, Martill, Hochuli and Guzzo2008).
Two Dasberg black shale samples (D2 and D3) contain isorenieratene derivatives (Figs 11, 12), such as isorenieratane, 2,3,6-/3,4,5-trimethyl-substituted diaryl isoprenoid, 2,3,6-trimethyl-substituted aryl isoprenoids and other diagenetic products of isorenieratene (Figs 12, 13, identified based on Koopmans et al. Reference Koopmans, Köster, Van Kaam-Peters, Kenig, Schouten, Hartgers, De Leeuw and Damsté1996; Clifford, Clayton & Sinninghe Damsté, Reference Clifford, Clayton and Sinninghe Damsté1998). These compounds are diagenetic and catagenetic products of strictly anaerobic green sulphur bacteria (Chlorobiaceae), which are photosynthetic and require both light and H2S (Hartgers et al. Reference Hartgers, Sinninghe Damsté, Requejo, Allan, Hayes, Ling, Xie, Primack, De Leeuw and Telnæs1994; Koopmans et al. Reference Koopmans, Köster, Van Kaam-Peters, Kenig, Schouten, Hartgers, De Leeuw and Damsté1996). Interestingly, the samples which contain isorenieratane (D2 and D3) also contain gammacerane. This compound is absent in the rest of the sample set, except sample D1 (grey shale), where this compound was detected in low amounts (Table 2). According to Sinninghe Damsté et al. (Reference Sinninghe Damsté, Kenig, Koopmans, Köster, Schouten, Hayes and De Leeuw1995), gammacerane is an indicator of water column stratification. This compound is derived from bacterivorous ciliates grazing on bacteria, including green sulphur bacteria, thus it explains the co-occurrence of these compounds in the Dasberg shales.
Figure 13. Summed mass chromatogram (m/z 235 + 237 + 448 + 538) of the Dasberg black shale sample (D3) showing distribution of long chain dia- and catagenetic products of isorenieratane and 2,3,6/3,4,5-TM substituted diaryl isoprenoid. Identification after Clifford, Clayton & Sinninghe Damsté (Reference Clifford, Clayton and Sinninghe Damsté1998).
Similarly, as shown in the Hangenberg section (Marynowski & Filipiak, Reference Marynowski and Filipiak2007), differences in the isorenieratane concentration are compatible with differences in the sum of isorenieratane plus 2,3,6-/3,4,5-TM substituted diaryl isoprenoids (Table 2, Fig. 11), suggesting a common origin for both compounds, proposed on the basis of similar δ13C values (Hartgers et al. Reference Hartgers, Sinninghe Damsté, Koopmans and De Leeuw1993, Reference Hartgers, Sinninghe Damsté, Requejo, Allan, Hayes, Ling, Xie, Primack, De Leeuw and Telnæs1994). The abundance of aryl isoprenoids in the Dasberg shales and the rest of the samples is variable (Table 2, Figs 11, 12). Two black shale samples are characterized by high concentrations of aryl isoprenoids (Table 2), in contrast to all remaining samples that either do not contain aryl isoprenoids (2KN) or contain a minor amount of these compounds (Table 2). It is notable that values of the aryl isoprenoid ratio (AIR), calculated from the ratio of C13–17/C18–22 (Schwark & Frimmel, Reference Schwark and Frimmel2004), are relatively high for black shale samples and low for some organic-poor samples (Table 2). We interpret this phenomenon as being due to the influence of secondary processes such as oxidation and/or water washing on the samples, causing degradation/dilution of low molecular weight aryl isoprenoids in the organic-poor samples. However, more detailed study is needed to explain this problem.
6.d. Polycyclic aromatic compounds
Samples from the upper part of the section are characterized by the occurrence of high concentrations of unsubstituted polycyclic aromatic hydrocarbons (PAHs) and moderate concentrations of oxygen-containing aromatic compounds (Figs 14, 15). The major identified compounds are: phenanthrene, 2-phenylnaphthalene, fluoranthene, pyrene, benzo[a]anthracene, chrysene, triphenylene, benzofluoranthenes, benzo[a]pyrene, benzo[e]pyrene, indeno[1,2,3-cd]pyrene, benzo[g, h, i]perylene and coronene (Fig. 14). The same samples contain low to medium concentrations of alkylated derivatives of the above-mentioned structures (Fig. 14), indicating a lack of microbial or sedimentary methylation reactions (e.g. Bastow et al. Reference Bastow, Alexander, Fisher, Singh, Aarssen and Kagi2000). Such a distribution is characteristic for rapid high-temperature processes, such as combustion, taking place during sedimentation (Venkatesan & Dahl, Reference Venkatesan and Dahl1989; Killops & Massoud, Reference Killops and Massoud1992; Kruge et al. Reference Kruge, Stankiewicz, Crelling, Montanari and Bensley1994; Jiang et al. Reference Jiang, Alexander, Kagi and Murray1998; Arinobu et al. Reference Arinobu, Ishiwatari, Kaiho and Lamolda1999; Finkelstein et al. Reference Finkelstein, Pratt, Curtin and Brassell2005) or hydrothermal petroleum formation at sea-floor spreading centres (Kawka & Simoneit, Reference Kawka and Simoneit1990; Simoneit & Fetzer, Reference Simoneit and Fetzer1996).
Figure 14. Distribution of total ion chromatogram of an aromatic fraction of the 1KN sample showing high concentration of unsubstituted PAHs. Circles denote methyldibenzofurans, square denote dimethyldibenzofurans, triangles denote benzo[b]naphthofuran isomers. DMN – dimethylnaphthalene, TMN – trimethylnaphthalene, TeMN – tetramethylnaphthalene, DBT – dibenzothiophene, MePh – methylphenanthrenes, MePy – methylpyrenes, MeFl – methylfluoranthenes. IS – internal standard. DB-35MS column was used.
Figure 15. Composite plot of the Dasberg section showing individual and summed PAHs concentrations. For explanation of abbreviations see Table 2.
The concentration of individual polycyclic aromatic hydrocarbons and sum of the pyrolytic PAHs are four to ten times higher in the upper part of the section (samples 1KN, 2KN and D4) than in the remaining section samples (Table 2, Fig. 15). Two different sums of PAHs were calculated: the first proposed by Finkenstein et al. (Reference Finkelstein, Pratt, Curtin and Brassell2005) (Table 2, PAHs2) and the second by us in this paper (Table 2, PAHs1). We are using the additional PAH parameter since we believe that in the case of Fammenian samples it better emphasizes the pyrolytic character of organic matter. But in fact, comparison of concentration results for two PAHs parameters are similar for all the samples (Table 2; Fig. 15). Quantitatively, the most important compounds in the sample 1KN, the most enriched in PAHs, are benzo[b]fluoranthene (23.6 μg/g TOC), fluoranthene (19.5 μg/g TOC) and pyrene (12.71 μg/g TOC) with somewhat lower concentrations of high molecular weight PAHs (Fig. 14, Table 2). The distribution of aromatic hydrocarbons differs from the Hangenberg section where the high molecular weight, peri-condensed polycyclic aromatic hydrocarbons, for example, coronene, benzo[g, h, i]perylene, benzofluoranthenes and benzo[e]pyrene, are significantly dominant (Marynowski & Filipiak, Reference Marynowski and Filipiak2007). Moreover, in the samples from the Dasberg section, dibenzofuran, methyldibenzofurans, dimethyldibenzofurans and benzo[b]naphthofurans were detected (Fig. 14). These oxygen-containing aromatic compounds are most possibly derived from the terrestrial organic matter (Radke, Vriend & Ramanampisoa, Reference Radke, Vriend and Ramanampisoa2000; Watson et al. Reference Watson, Sephton, Looy and Gilmour2005; Sephton et al. Reference Sephton, Looy, Brinkhuist, Wignall, De Leeuw and Visscher2005; Wang & Visscher, Reference Wang and Visscher2007).
Taking into account the lack of evidence of other PAH sources like ocean spreading centres and active hydrothermal venting in the latest Devonian of the Kowala region, the most likely source for peri-condensed PAH is forest wildfires. As suggested by Grice, Nabbefeld & Maslen (Reference Grice, Nabbefeld and Maslen2007), some PAHs, including chrysene, triphenylene and benzo[e]pyrene, may have originated from algal OM. However, the co-occurrence of terrestrial OM with undoubtedly identified charcoal fragments (Fig. 9) and PAHs in the Dasberg section, as well as relatively high concentrations of the ‘true combustion marker’ benzo[a]pyrene (Grice, Nabbefeld & Maslen, Reference Grice, Nabbefeld and Maslen2007), indicate a wildfire origin for the PAHs. Palynological evidence affirming this interpretation is supplied by the elevated concentration of terrestrial organic debris in samples from the upper part of the section (Figs 8, 9).
7. U/Th ratio
The calculated values of the U/Th ratio range from 0.3 to 1.2 (Fig. 11). In the lower part of the section (samples D-2 and D0), the U/Th ratios (0.77 and 0.8, respectively) point to dysoxic conditions in the bottom environment. Values drop well below 0.75 (0.56 to 0.6) above this (samples D1–D2A, see Fig. 11), indicating a shift to persistent oxygenated bottom waters. However, in sample D3, the highest U/Th value of 1.2 appears, indicating the occurrence of anoxic conditions. Above this level, the U/Th ratios drop again to lower values (0.3 to 0.6), indicating the return of oxygenated bottom water conditions.
Interestingly, the highest U/Th value (1.2) is coincident with the highest values of other organic geochemical parameters that indicate anoxic conditions in the water column (see Fig. 11). On the contrary, below this horizon (during deposition of the shales D2), the U/Th value is very low (0.59) in comparison to other organic geochemical parameters which still indicate water column anoxia (Fig. 11).
8. Pyrite framboid diameter analysis
Pyrite framboids are rather widely scattered in the samples studied. Only in two samples (D2 and D3) was it possible to measure 100 framboids. In the rest of the samples, only 21 to 33 framboids have been measured. The diameters of pyrite framboids vary widely from 1.1 to 22.1 μm, with mean values of 4.0 to 10.6 μm. In all the samples, large framboids (12.0 to 22.1 μm) occur.
The lowest two samples (D-2 and D1) and the highest one (1KN) are dominated by large framboids with mean values around 10 μm. Small-sized framboids (<6 μm), which may have formed in a euxinic water column (see Wilkin, Barnes & Brantley, Reference Wilkin, Barnes and Brantley1996; Wignall & Newton, Reference Wignall and Newton1998), occur as single specimens only (one specimen in samples D-2 and D1, and two specimens in sample 1KN). They have presumably formed within the sediments under an oxic or dysoxic, at most, water column (see Wignall & Newton, Reference Wignall and Newton1998). In the samples D2 and D3, on the other hand, the mean values of framboid diameters are the lowest, being 5.7 and 4.0, respectively (Table 3). Furthermore, the standard deviations are the lowest of all the samples investigated, attaining 2.4 and 1.7, respectively. In sample D2, only a few large framboids above 10 μm are found, and the majority of framboids in that sample are characterized by diameters less than 6 μm. Higher in the section, in sample D3, framboids of less than 4 μm in diameter are most numerous, and the largest framboids, up to 12 μm in diameter, occur as single specimens only. The dominance of small-sized (<6 μm in diameter) framboids in sample D2 suggests that the anoxic conditions prevailed in the water column, while the presence of a few larger ones may point to a brief oxygenation of the bottom waters. This is also supported by the low value of the U/Th ratio (see Table 3). The framboid diameters (dominance of very small, <4 μm, specimens and single large framboids only) in sample D3 are in agreement with both the biomarker data and the highest U/Th ratio, which indicate anoxic conditions in the water column and bottom waters, respectively.
Table 3. Uranium (U) and thorium (Th) concentrations, U/Th ratio and pyrite framboid diameter values
n – number of measured framboids, min – minimum value, max – maximum value, mean – mean value, sd – standard deviation.
9. Discussion
9.a. Productivity and anoxia
The comprehensive application of independent palynological, geochemical and petrographical methods has shown the presence of photic zone anoxia during deposition of the Dasberg black shales in the Kowala section. Although part of the investigated sequence (D-2, D1, 1KN) is characterized by high TOC values, anoxia is only undoubtedly present in the black shale samples (D2 and D3).
All the indices used, such as TOC, isorenieratane and aryl isoprenoid concentrations, gammacerane to hopane ratio, U/Th ratios and the pyrite framboid diameter study indicate that the deepest oxygen deficiency prevailed during sedimentation of the upper Dasberg shale (sample D3; Fig. 11). Anoxic conditions began in the course of the lower Dasberg shale sedimentation (D2) and were interrupted by an oxic episode (sample D2A), as also seen during the somewhat older Annulata event, and anoxia reached its maximum during deposition of the upper Dasberg shale.
Our data show that the redox conditions during the deposition of the shales fluctuated.
The presence of green sulphur bacteria biomarkers and dominance of small-sized framboids are indicative of water column anoxia in the photic zone, while the presence of large framboids and low values of the U/Th ratio are characteristic of oxygenated bottom-waters (Tables 2, 3; Fig. 11). Such conflicting data retrieved from a single sample may mean that in fact variable conditions are preserved. Thus, it may be interpreted, that (1) during sedimentation of the lower Dasberg shale (sample D2), the anoxic conditions intermittently occurred in the water column, or (2) the anoxia prevailed in the upper part of the water column, while the bottom waters were oxygenated, at least briefly. Recent findings of oxygen-requiring benthic foraminifera in many Late Devonian black shales of eastern North America (Schieber, Reference Schieber2009) confirm our geochemical observations: despite the presence of anoxia in the water column, the bottom waters were oxygenated.
Somewhat later (sample D2A), anoxic conditions in the water column vanished completely, but during the sedimentation of the upper Dasberg black shales (sample D3), anoxia returned and encompassed both the bottom (high U/Th value and domination of small, <4 μm, pyrite framboids) and higher part of the water column environments, before another re-oxygenation during the sedimentation of the overlying deposits (Fig. 11).
This model of the Dasberg shale deposition is in agreement with general observations of Late Devonian events, where pulses of anoxia are typically observed (Murphy, Sageman & Hollander, Reference Murphy, Sageman and Hollander2000; Racki et al. Reference Racki, Racka, Matyja and Devleeschouwer2002; see Racki, Reference Racki, Over, Morrow and Wignall2005 for review), with the exception of the entirely anoxic upper Dasberg black shale.
As palynological results have shown, both black shale horizons are almost free from acritarchs. Only some stress-indicating ‘disaster species’ like leiospheres have been found in any abundance. Interestingly, in the sample D3, which was deposited in a completely anoxic environment, Lophosphaerium sp. and single Unellium winslowiae additionally appear with the common leiosphere taxa. Moreover, acritarchs are present in moderate but constant amounts in the rest of the samples, deposited under oxic conditions. The common presence of AOM in the samples from the DBS also partially supports geochemical results concerning anoxic conditions.
The data obtained enable the interpretation of the sedimentation of the Dasberg black shales as the result of sea-level fluctuations and blooming of primary producers (e.g. Algeo & Scheckler, Reference Algeo and Scheckler1998). A more or less similar situation occurred during sedimentation of the Middle to Upper Frasnian sediments (Marynowski, Filipiak & Pisarzowska, Reference Marynowski, Filipiak and Pisarzowska2008), Kellwasser horizons (Joachimski et al. Reference Joachimski, Ostertag-Henning, Pancost, Strauss, Freeman, Littke, Sinninghe Damsté and Racki2001; but see Racki et al. Reference Racki, Racka, Matyja and Devleeschouwer2002), Hangenberg shales (Marynowski & Filipiak, Reference Marynowski and Filipiak2007) and Annulata shales (Racka & Marynowski, Reference Racka and Marynowski2008).
The occurrence of periodically recurrent anoxic events during the Late Devonian may be easily interpreted as multi-stage rapid sea-level rise caused by eustatic pulses (eustatic oscillations) (see Hallam & Wignall, Reference Hallam and Wignall1999; also Racki, Reference Racki1998 and Sageman et al. Reference Sageman, Murphy, Werne, Ver Straeten, Hollander and Lyons2002). The similarity of particular small- and large-scale Upper Famennian events has already been underlined previously from the viewpoint of ammonoid evolution (Becker, Reference Becker and House1993; see also House, Reference House2002), as ‘multiphased sudden eustatic and anoxic events. . .’. Such rapid sea-level changes may have co-occurred with Upper Devonian recurrent periods of global warmth which additionally promoted anoxia due to reduced solubility of oxygen (see Meyer & Kump, Reference Meyer and Kump2008).
9.b. Terrestrial organic matter and wildfires
In the upper part of the investigated section the higher PAH concentrations and the presence of small charcoal particles has been documented, which point to the occurrence of wildfires during sedimentation of these deposits. This is the first evidence of wildfire detected in the vicinity of VF/LV zone boundary. So far, reports on the Upper Famennian wildfires have rarely been correlated with detailed stratigraphical data. The typical marine-basinal character of sedimentation (Szulczewski, Reference Szulczewski1971; Marynowski, Narkiewicz & Grelowski, Reference Marynowski, Narkiewicz and Grelowski2000) and large amount of charcoal in the samples investigated, suggests that wildfires were common and on a wide-scale.
Charcoal is very buoyant due to its porous structure and therefore capable of prolonged flotation (Batten, Reference Batten, Jansonius and McGregor1996). On the other hand, however, during fires, very small pieces of charcoal or ash can be carried high into the atmosphere and then deposited far away from the source. These wind-blown charcoal particles, usually do not exceed 20 μm in diameter (Tyson, Reference Tyson and Jenkins1993; Collinson et al. Reference Collinson, Steart, Scott, Glasspool and Hooker2007), but a recent report showed that intensive crown fires cause long-distance transport of macroscopic (up to 1.3 cm) charcoal fragments (Tinner et al. Reference Tinner, Hofstetter, Zeugin, Conedera, Wohlgemuth, Zimmerman and Zweifel2006) (Fig. 9). Moreover, charcoal is not attractive as a food for animals because it is indigestible. Therefore, its concentration in sediments indirectly may be a result of its simple non-digestion, especially in well-oxygenated bottom habitats (e.g. 2KN sample; Fig. 10).
Temperatures of charcoal formation calculated on the basis of inertinite reflectance measurements ranged from 257 to 406°C, with a mean value of 286°C. These results suggest that the charcoal was formed in the low-temperature ground and/or surface fires (Scott, Reference Scott2000; Jones & Lim, Reference Jones and Lim2000). Interestingly, our results differ from those presented by Rimmer, Scott & Cressler (Reference Rimmer, Scott and Cressler2006), who described ‘moderately hot (around 550°C) fires from the Upper Devonian (Famennian 2c) lowland sites (Red Hill, PA, and Elkins, WV)’. It may suggest that the Upper Devonian charcoal formed during a different wildfire temperature regime.
It is notable that forest wildfire evidence was also found recently for the Hangenberg event horizon from Kowala (Marynowski & Filipiak, Reference Marynowski and Filipiak2007). Fairon-Demaret & Hartkopf-Fröder (Reference Fairon-Demaret and Hartkopf-Fröder2004) recognized very well-preserved and strongly taxonomically differentiated charcoalified plant mesofossils in the samples from the Refrath borehole, from the Ardennes-Rhenish Massif (Germany), dated to the LL miospore Zone (sensu Maziane, Higgs & Streel, Reference Maziane, Higgs and Streel1999; see also Hartkopf-Fröder, Reference Hartkopf-Fröder2004). The frequency of such remains in many samples from over 30 m of the section indicates vegetation that has been rather frequently swept by fire (Fairon-Demaret & Hartkopf-Fröder, Reference Fairon-Demaret and Hartkopf-Fröder2004). These and the previous reports (Rowe & Jones, Reference Rowe and Jones2000; Cressler, Reference Cressler2001; Scott, Reference Scott2000; Rimmer et al. Reference Rimmer, Thompson, Goodnight and Robl2004; Rimmer & Scott, Reference Rimmer and Scott2006; Rimmer, Scott & Cressler, Reference Rimmer, Scott and Cressler2006; Scott & Glasspool, Reference Scott and Glasspool2006; Prestianni et al. Reference Prestianni, Decombeix, Thorez, Fokan and Gerrienne2009) provide further evidence of extensive wildfires during Late Famennian times. The main cause of intensive wildfires was a distinct increase of O2 in the atmosphere, calculated as high as 23% (Scott & Glasspool, Reference Scott and Glasspool2006), and the intensive development of land plants including Archeopteris, ferns, and some of the earliest seed plants (Rowe & Jones, Reference Rowe and Jones2000; Cressler, Reference Cressler2001; Dąbrowska & Filipiak, Reference Dąbrowska and Filipiak2006; Scott & Glasspool, Reference Scott and Glasspool2006).
Some calculations of the O2 in the Famennian atmosphere are much lower, showing results from ~13 to 15% (Bergman, Lenton & Watson, Reference Bergman, Lenton and Watson2004; Algeo & Ingall, Reference Algeo and Ingall2007). However, such values are unrealistic, taking into account the common occurrence of charcoals in the Upper Devonian sediments (see above) and recent burn experiments which undoubtedly showed that the lower O2 limit for combustion in a natural environment should be increased from 12 to at least 15% (Belcher & McElwain, Reference Belcher and McElwain2008).
Moreover, common tuffite intercalations in the Upper Devonian strata (see Marynowski & Filipiak, Reference Marynowski and Filipiak2007) suggest strong volcanic activities during this time. Such intensive explosive volcanism could strongly influence the atmospheric acidification, resulting in the formation of miospore tetrads and abnormal miospores found in the uppermost Devonian sediments (Marynowski & Filipiak, Reference Marynowski and Filipiak2007).
10. Conclusions
Geochemical proxies reveal the presence of the photic zone anoxia during Dasberg black shales deposition recorded in the Kowala quarry section. The Lower Dasberg black shale sedimentation was characterized by anoxic conditions in the upper part of the water column, but the bottom waters were oxygenated. Throughout deposition of the upper Dasberg black shale, anoxic conditions encompassed both the bottom and water column environments. These dynamically changing redox regimes during sedimentation of the black shales were strictly connected with the sea-level changes, causing enhanced supply of land-derived nutrient remains, and in consequence, blooming of primary marine producers. The palynofacies evolution through the succession suggests a transgression signal corresponding to the IIf sea-level curve sensu Johnson, Klapper & Sandberg (Reference Johnson, Klapper and Sandberg1986).
Above the Dasberg black shales, charcoal and elevated concentrations of polycyclic aromatic hydrocarbons were detected, providing evidence of the presence of wildfires just above the boundary of VF/LV palynological zones. Temperatures calculated from the fusinite reflectance values indicate that the charcoal was formed during low-temperature ground and/or surface fires. The high levels of charcoal in the typically marine sedimentary rocks indicate large-scale wildfires and intensive transport of land particles.
Acknowledgements
We are grateful to Professor Grzegorz Racki and Professor Elżbieta Turnau for giving their useful comments and suggestions on the first version of this manuscript. Ewa Teper, M.Sc. (Faculty of Earth Sciences, Sosnowiec) is acknowledged for her help during analyses using ESEM. This work has been financed in part by an MNISW grant: N N307 2379 33 (to LM).
