Introduction

The Ludfordian (Upper Silurian) stage was the time of the most pronounced positive stable carbon isotope excursion (CIE) in the Phanerozoic, known as the Lau excursion (Wenzel and Joachimski 1996; Bickert et al. 1997; Wigforss-Lange 1999; Calner et al. 2004; Martma et al. 2005; Kaljo and Martma 2006; Jeppsson et al. 2007; Kaljo et al. 2007; Lehnert et al. 2007; Eriksson and Calner 2008; Barrick et al. 2010; Kozłowski and Munnecke 2010; Munnecke et al. 2010; Cramer et al. 2011; Loydell and Frýda 2011; Jeppsson et al. 2012; Kaljo et al. 2012; Kozłowski and Sobień 2012). Very good geochemical, biostratigraphical, and paleoecological documentation exists for the Lau excursion, and it is often used as a model to study positive CIEs, which are widely used as chemostratigraphic markers and as indicators of perturbations of the global carbon cycle (Jeppsson 1990; Aldridge et al. 1993; Bickert et al. 1997; Jeppsson and Aldridge 2000; Munnecke et al. 2003; Calner 2005a; Melchin and Holmden 2006; Cramer and Saltzman 2007a, b; Fanton and Holmden 2007; Calner et al. 2008; Małkowski et al. 2009; Kozłowski and Sobień 2012). Studies of recent and sub-fossil carbonate sediments indicate that sea-level forcing, restricted circulation and laterally variable sediment composition can significantly influence the δ13Ccarb values, leading to amplification of positive excursions in carbonate platform interiors during flooding intervals (Immenhauser et al. 2003; Swart and Eberli 2005; Swart 2008; Gischler et al. 2009; Oehlert et al. 2012). The small-scale spatial variability of δ13C values results from a combination of generic marine-water signal and local factors such as facies, restricted circulation, and diagenetic overprint. Although it is difficult to attribute the final δ13C values to the influence of a particular factor, in some cases it can be demonstrated (e.g., Holmden et al. 1998; Harzhauser et al. 2007; Colombié et al. 2011). Some local factors seem to have been involved in the record of the Lau excursion, which is mostly documented by δ13Ccarb of limestones from the epicontinental basin developed during the early Paleozoic on Baltica. This Ludfordian CIE has been recorded in Sweden (+11.2 ‰, Wigforss-Lange 1999; +8.80 ‰, Samtleben et al. 2000), Lithuania (+8.2 ‰, Martma et al. 2005), Latvia (+5 ‰, Kaljo and Martma 2006; Kaljo et al. 2012), Poland (+8.9 ‰, Kozłowski and Munnecke 2010; +6.7 ‰, Kozłowski and Sobień 2012) and Ukraine (+6.9 ‰, Kaljo et al. 2007; +6.0 ‰, Kaljo et al. 2012). These localities represent different settings across the carbonate platform and enable the detection of local variability in the CIE record and its dependence on local facies development. Kaljo et al. (2012) reported substantial differences in the record of the Ludfordian CIE in closely spaced shallow-water sections located in the Podolia region of western Ukraine, including strongly depleted values (down to −5.0 ‰) within or slightly above the positive peak. In the present paper, five lithological sections in the Zbruch River Valley in Podolia, representing the Lau excursion interval, have been investigated. We provide a detailed microfacies analysis, review, and update descriptions of lithostratigraphic units, identify their sequence stratigraphic positions and their potential correlation with depositional sequences reported from other sections within and outside the basin. An explanation is proposed for the unusual δ13Ccarb values occurring in this interval in the Zbruch River valley and elsewhere in Podolia.

Materials and methods

Sections investigated in the present study are exposed along the lower reaches of the Zbruch River, ca. 8 km upstream of its outlet into Dniester, and the stretch between the villages of Chernokozyntsi and Zavalya (Fig. 1). Outcrops are located on both banks of the Zbruch River in deeply incised slopes. In view of the lack of detailed topographical maps of the area, localization of the outcrops was primarily based on satellite photographs and GPS coordinates (given in the supplementary online material 1). The sections Kudryntsi–Castle and Kudryntsi–Cowpath have been additionally correlated using a dumpy level and the results have been integrated in the final correlation chart (Fig. 2, with a larger version in the supplementary online material 2).

Fig. 1
figure 1

Map of the study area: the lower reaches of the Zbruch River, south of the city of Skala Podilska with positions of described sections. Grey areas indicate continuous belts of Silurian outcrops in the river valley slopes

Full size image
Fig. 2
figure 2

Correlation of sections based on lithology, topographic position, and dumpy leveling, with an assignment to lithostratigraphic units discussed in the text

Full size image

For stable carbon isotope measurements of carbonates, 16 unweathered samples and two brachiopods were selected. Vein-free pelitic matrix and unaltered shell fragments were powdered using a steel needle and analyzed in the Stable Isotope Laboratory of the Polish Academy of Sciences in Warsaw. Sample powder was treated with phosphoric acid in a Kiel IV preparation system and analyzed in a conjunct Finnigan Delta + mass spectrometer. Values are reported in Table 1 using the conventional delta notation with respect to the Vienna Pee Dee Belemnite (VPDB). Reproducibility for the isotopic analysis was better than ±0.1 %.

Table 1 Stable carbon and oxygen isotope ratios (VPDB)
Full size table

Study area

Silurian outcrops in the region of Podolia, western Ukraine, expose the sedimentary rocks deposited on the south-western shelf of Baltica in a basin that stretched from present-day southern Sweden through to Moldova (Fig. 3). Surface exposures in Podolia represent shallow-water carbonate facies developed on a vast epeiric platform, which extended basinward into the shelf facies recognized today in the subsurface of Poland (Teller 1997; Kozłowski and Sobień 2012), Baltic countries (Kaljo 1970; Kaljo et al. 1997; Martma et al. 2005), as well as in Sweden (Samtleben et al. 2000; Calner et al. 2006).

Fig. 3
figure 3

Paleogeographic map of the study area and its position on the SW shelf of Baltica in the Upper Ludfordian, based on Einasto et al. (1986) and Teller (1997). USB Upper Silesian Block, PC Pomeranian Caledonides, T-TL Teisseyre-Tornquist Line, red dots coring localities and sections mentioned in the text, black dots present-day cities

Full size image

Geological context

Lithostratigraphy

Two main lithostratigraphical schemes exist for the Silurian of Podolia: one developed by the research group of Nikiforova (Nikiforova and Predtechenskij 1968; Nikiforova et al. 1972), and the second by Tsegelnyuk (Tsegelnyuk 1980a, 1980b) and Tsegelnyuk et al. (1983). Herein the latter scheme is followed, as it corresponds more closely with our own observations on the facies development within the studied interval. Tsegelnyuk et al. (1983) distinguished four main “series” in the Silurian of Podolia: Bolotyn, Yaruga, Malinovtsy, and Rukshin, and related them, respectively, to the Llandovery, Wenlock, Ludlow, and Downton series of the Anglo-Welsh Basin.

The outcrop belt of the Malinovtsy Series is spread between the villages of Bolshaya Slobodka to Isakivtsy. It reaches a thickness of 90–141 m and is subdivided into three suites: the Konovka, Tsviklivtsy, and Rykhta suites. Of particular interest for the present study is the Rykhta suite, which is divided into the Grinchuk sub-suite, represented by marly nodular limestone and marlstone, and the overlying Isakivtsy sub-suite, represented by dolomitized grainstone with remains of a diverse fauna, including tabulate corals, stromatoporoids, algae, crinoids, gastropods, and brachiopods. The sub-suites of Grinchuk and Isakivtsy reach thicknesses of 18–19 and 5–6 m, respectively.

The lower boundary of the Rukshin series is drawn at the base of a black shale or marly dolomite of the Prygorodok suite that rest on an erosion surface at the top of the dolomite of the Isakivtsy sub-suite. The total thickness of the Rukshin series reaches 250 m in the outcrop belt from Khotin to Melnytsya-Podilska. Of the five suites of the Rukshin series, namely Prygorodok, Varnytsya, Trubchyn, Dzvenygorod, and Khudykivtsy, the first two are relevant to the interval studied in the present work. An additional help in local correlation is the presence of six K-bentonite beds numbered C1–C6 in the Prygorodok suite, and two more (C7, C8) in the Varnytsya suite, characterized geochemically by Huff et al. (2000) and Kiipli et al. (2000).

Abushik et al. (1985) have reintroduced the term “Skala series”, originally employed by Alth (1874), and described the Skala Series deposits in the Zbruch River Valley. Abushik et al. (1985) distinguished only two “formations” (used interchangeably with “suites”) within the Skala series: Rashkov and Dzvenygorod. The lower boundary of the Skala Series was placed by these authors at the abrupt lithological change from argillaceous dolomite devoid of fossils, which is attributed to the upper sub-suite of the Isakivtsy suite, to the shallow-marine deposits associated with the Rashkov suite. Based on lithological profiles of key sections employed to characterize these units, it can be inferred that the Isakivtsy Formation sensu Abushik et al. (1985) and Koren’ et al. (Koren’ et al. 1989) includes the Isakivtsy sub-suite and the Prygorodok suite of Tsegelnyuk et al. (1983) in the rank of sub-suites (Table 2). The Rashkov suite was distinguished on the basis of depositional cycles, which are identified in the present study as peritidal. As the Skala Series has been subsequently used to denote Přídolí in the regional stratigraphic scheme for Podolia (Koren’ et al. 1989), this unit is adopted here as a synonym for the Rukshin Series defined by Tsegelnyuk et al. (1983).

Table 2 Correlation between lithostratigraphical units in the Ludfordian to Přídolí of Podolia, distinguished by Abushik et al. (1985) and Tsegelnyuk et al. (1983)
Full size table

Biostratigraphy

In the standard regional scheme for the Silurian of Podolia by Koren’ et al. (Koren’ et al. 1989) counterparts of the Isakivtsy and Prygorodok Fms are placed entirely in the Ludfordian Stage, and the boundary between the Prygorodok Fm. and the Varnytsya Fm. is placed at the Ludlow-Přídolí boundary. Abushik et al. (1985) reported occurrences of Ozarkodina crispa in the lower member of the Rashkov suite, corresponding to the Varnytsya Fm., which indicates its Ludfordian age. These authors reported the O. eosteinhornensis assemblage from the middle Rashkov sub-suite, corresponding to the upper part of the Varnytsya Fm. and establishing its Přídolí age. However, Paris and Grahn (1996) reported Eisenackitina barrandei in the Dzvenygorod Formation, suggesting that the entire underlying Varnytsya Fm. is of Ludfordian age. This disagreement may result from diachronous boundaries between the Varnytsya and Dzvenigorod Fms. While the studies do not completely agree, both place the lower part of the Varnytsya Fm. in the Ludfordian Stage (see correlation of flooding surfaces in the discussion).

These findings constrain the age of the studied rocks from the top. From the bottom, it has been constrained by Paris and Grahn (1996), who identified Sphaerochitina sphaerocephala and E. barrandei in the top part of the Isakivtsy Fm. The Ludfordian age of these deposits is also supported by the presence of Daya navicula in the uppermost Grinchuk, Isakivtsy and Prygorodok Fms. (Tsegelnyuk et al. 1983; Nikiforova et al. 1985) and of Homoeospira baylei in the last two formations (Nikiforova et al. 1985).

Stable carbon isotope stratigraphy

Of potential utility for constraining the stratigraphic position of the interval investigated in the present study is the positive stable CIE recognized globally in the Ludfordian (Wenzel and Joachimski 1996; Bickert et al. 1997; Kaljo et al. 1997; Azmy et al. 1998; Wigforss-Lange 1999; Wenzel et al. 2000; Saltzman 2001; Calner 2005b; Martma et al. 2005; Kaljo and Martma 2006; Jeppsson et al. 2007; Kaljo et al. 2007; Lehnert et al. 2007; Eriksson and Calner 2008; Barrick et al. 2010; Kozłowski and Munnecke 2010; Munnecke et al. 2010; Cramer et al. 2011; Loydell and Frýda 2011; Kaljo et al. 2012; Kozłowski and Sobień 2012; Manda et al. 2012). The beginning of the excursion coincides with an abrupt decrease in abundance of Polygnathoides siluricus, shortly before its Last Appearance Datum (LAD) (Martma et al. 2005; Kaljo and Martma 2006; Jeppsson et al. 2007; Lehnert et al. 2007; Eriksson and Calner 2008). The peak interval is constrained within the O. snajdri zone (coeval with the top of the Neocucullograptus kozlowskii zone), and in many sections there is lithological evidence for a sub-stratigraphic (contained entirely within one biozone) gap or even erosion (Kaljo et al. 1997; Calner 2005b; Lehnert et al. 2007; Eriksson and Calner 2008; Kozłowski and Munnecke 2010; Loydell and Frýda 2011; Kaljo et al. 2012; Kozłowski and Sobień 2012). The position of the falling limb of the excursion is less consistent. The CIE ends clearly below the First Appearance Datum (FAD) of O. crispa in Gotland (Calner et al. 2004; Jeppsson et al. 2007; Eriksson and Calner 2008), as well as in the Ventspils (Latvia) and Vidukle-61 (Lithuania) cores (Martma et al. 2005; Kaljo and Martma 2006; Kaljo et al. 2012), but it persists over a much longer interval in the Ohesaare core (Saarema, Estonia). Data available from the Bohemian Basin (Lehnert et al. 2007) and from the Anglo-Welsh Basin (Loydell and Frýda 2011) also suggest that the return to values close to 0 ‰ predates the first appearance of O. crispa.

The excursion has been confirmed by Kaljo et al. (2007, 2012) in the Dniester River valley, where it begins entirely in the Prygorodok Fm. (Isakivtsy-45 locality, max. δ13C = 6.0 ‰) or well within the Isakivtsy Fm. (Zhvanets-39 locality, max. δ13C = 6.6 ‰). Under the assumption that metabentonite beds in the Prygorodok Fm. are isochrones, the return to base-level values is clearly diachronous, e.g., between the Ataki-117 + Braga-119 section (Kaljo et al. 2007; below the C2 metabentonite bed) and the Isakivtsy-45 section (Kaljo et al. 2012; below the C3 metabentonite bed). These authors also introduced the “top Ludfordian twin excursion” based on the analysis of the Vidukle-61 and Ohesaare cores, where this smaller double peak is developed entirely within the O. crispa zone. This separate twin peak is, however, not visible in most sections, but elevated δ13C values seem to decay at various rates (Kaljo et al. 1997; Kaljo and Martma 2006; Kozłowski and Munnecke 2010; Kaljo et al. 2012). According to Kaljo et al. (2012), the Ludfordian excursion in Podolia is entirely contained within the Isakivtsy and Prygorodok Fms and the δ13Ccarb values are close to 0 ‰ in the Varnytsya Fm.

Results

Characterization of lithostratigraphic units and depositional environments

The Isakivtsy Formation

Dolomite belonging to the Isakivtsy Formation is exposed in the Zbruch River Valley along a continuous belt of outcrops stretching from Milivtsi along the eastern bank of the Zbruch River, up to the floodplain below the village of Chernokozyntsi, where they disappear under vegetation.

Macroscopically, the Isakivtsy dolomite forms a massive unit, the top of which is clearly marked in the topography of the valley slopes (Fig. 4a). The topmost 3–6 m of the unit is formed by unbedded dolosparite with the degree of recrystallization decreasing towards the bottom of the section, where the dolomite is argillaceous and less resistant to weathering. Weathered surfaces reveal coquinas formed by brachiopods and ostracods (Fig. 4b). The total thickness of the deposits of the Isakivtsy Fm. could not be determined, but it clearly reaches more than 10 m (Fig. 2).

Fig. 4
figure 4

The Milivtsi–North section. a–c The Isakivtsy Fm. a Top part of the dolomite unit. Outcrop height is approx. 7 m. The overlying deposits of the Prygorodok Fm. form a gentle slope and are usually overgrown with thick vegetation. b Shell accumulation on weathered surfaces of the Isakivtsy dolomite. c Dolomite rhombohedra in the matrix exhibit cloudy cores, which may be the remnants of the original mineralogical phase; scale 100 μm. d–m Prygorodok Fm. d Thick-bedded to massive silty dolomite (hammer for scale). e Brecciated laminated marly dolomite in the basal part of the Prygorodok Fm. Clasts are overturned, but remain in situ; scale 5 mm. f Cross section of a burrow within the marly dolomicrite, formed in bioturbated ostracod wackestone; scale 5 mm. g Laminated, highly argillaceous dolomicrite in the basal part of the Prygorodok Fm; scale 5 cm. h Deformation in microbial mats resulting from the formation of desiccation cracks or from uneven growth; scale 1 cm. i Undulating dolomitic crusts (marked with dashed line). j Silicified octracod carapaces in marly dolomite; scale 200 μm. k Bioclastic-peloidal grainstone in the upper part of an event shell accumulation. Note individual recrystallized shells (convex-up orientation) and abundant ostracod carapaces; scale 4 mm. l Polished slab of a nodular limestone bed at the transition to the Varnytsya Fm.; scale 1 cm. m Vertical structure of an event shell accumulation; scale 5 mm

Full size image

The Isakivtsy dolosparite, characterized in detail as microfacies RD in Table 3 and in Fig. 4c, is inequigranular and planar-subhedral (classification of Sibley and Gregg 1987, which is also used in the descriptions that follow). Cap-in-cap structures formed of disarticulated ostracod carapaces indicate that the original sediment may have been subjected to wave action. Based on these characteristics, it is proposed that the original rock was formed in a relatively restricted, but marine environment, e.g., a shallow lagoon.

Table 3 Microfacies of the Ludfordian deposits in the Zbruch River Valley
Full size table

The Prygorodok Formation

Deposits representing the Prygorodok Fm. reach a thickness of 17 m in the studied area (Fig. 2) and the most complete profile is accessible in the Milivtsi–North section (Fig. 4d). Auxiliary profiles are available in the Milivtsi–South (Fig. 5a) and Kudryntsi–Cowpath sections (Fig. 6a). The dominant lithology is argillaceous dolomite.

Fig. 5
figure 5

The Milivtsi—South section. a The decimeter-scale bedded marly dolomicrite forming the bulk of the Prygorodok Fm; hammer for scale. b Orange-colored metabentonite bed; measuring tape shows 20 cm. c Wrinkle structures on a bedding plane within marly dolomicrite. d Angular pores occurring in one horizon within marly dolomite, interpreted here as halite casts; scale 1 cm; e dense crinkled structures resembling mud cracks, abundant on bedding planes within the upper part of the Prygorodok Fm.; measuring tape shows 10 cm

Full size image
Fig. 6
figure 6

The Prygorodok Fm. in the Kudryntsi—Cowpath section. a General view of one of the adjacent ravines which form the exposure, hammer for scale. b Laminated volcanogenic bed, coin diameter 18 mm. c Cross bedding in laminated dolomicrite, indicating current action; scale 5 mm. d Microfacies of the laminated volcanogenic sediment: quartz layers contain angular, moderately sorted grains belonging to the very fine sand fraction, together with biotite flakes (arrows) aligned parallel to bedding; plane polarized light, scale 500 μm. e As in d, with gypsum plate inserted

Full size image

In the basal part, a breccia is present composed of angular intraclasts up to 10 cm in diameter (microfacies IWR in Table 3). The dominant clast lithology is dark brown, centimeter-scale laminated dolomicrite, argillaceous and rich in organics (Fig. 4e, g, h), which binds fine sand-sized angular quartz grains with admixtures of biotite. The process of brecciation affected sediment, which was already partly stabilized by lithified dolomitic crusts (Fig. 4i).

Microbial carbonates are present throughout the formation, forming massive dolomicrite beds with an admixture of quartz (microfacies LMQ in Table 3), or undulating, deformed sheets covered with mudcracks (Fig. 4h, i). Wrinkle structures (Fig. 5c) and load casts (Figs. 4h, 5e) indicate that lamination of the dolomite is stromatolitic, but locally affected by current action (Fig. 6c). Microbial crusts occur upon and stabilize yellow siltstone beds ranging from 1 to 15 cm in thickness, formed of unlithified quartz grains with a high carbonate content (Fig. 4i). Typically, the lower boundary of a siltstone bed is sharp, and the top undulates as it was stabilized by microbial mats. In the Milivtsi–South section there are two horizons of cube-shaped pseudomorphs, 1–5 mm in diameter (Fig. 5d).

In the upper part of the Prygorodok Fm., the proportion of laminated sediments decreases and bedding in the dolomite thickens and becomes massive in the uppermost part. The color is lighter, probably indicating a more oxygenated environment. The thick-bedded marly dolomite is poor in fossils, except for burrows (Fig. 4f, microfacies BUM in Table 3). Silicified leperditicopid ostracods are concentrated in several horizons (microfacies OWP in Table 3; Fig. 4j). The silica was likely derived from metabentonite beds, which are present in substantial thicknesses (up to tens of centimeters) and in vivid colors (Figs. 2, 5b, 6b, d, e). The Kudryntsi–Cowpath section contains a distinctive 4-cm-thick bed of laminated fine-grained volcanogenic sandstone-claystone (Fig. 6b, d, e), consisting of sub-millimeter-scale alternations of fine quartz and clay layers, which indicates sedimentation from suspension.

The massive dolomite contains several shell beds composed of normally graded shells with convex-up orientation (microfacies BRG in Table 3; Fig. 4k, m). The lower boundary of each shell bed is erosional (Fig. 4m). The crushed and reworked fauna bioclasts contain evidence of organisms not occurring in the Prygorodok Fm., such as tabulate corals and bryozoans.

On the upper bedding planes of shell beds, gastropod and bivalve shells are current-aligned. Although the shell beds represent sediment transported shoreward, the allochthonous faunal assemblage is poorly diversified, pointing to a restricted lagoonal environment as a source area for the redeposited material. These shell beds are interpreted as either tempestites or washover-fan deposits. The lateral extent of these beds could not be determined, as all examined outcrops did not exceed 5 m in width. The proportion of this type of event bed increases towards the top of the Prygorodok Fm., along with the diversity of bioclasts (qualitative observation).

Stable carbon isotope development

The laminated dolomite just above the breccia level in the Milivtsi-North outcrop has a δ13Ccarb value of −7.11 ‰ and the minimum value of −10.53 ‰ is reached within the thin-bedded laminated dolomite in the Kudryntsi–Cowpath section (the position of carbon isotope samples is indicated on the lithological profiles in Fig. 2). Above the laminated dolomite, the δ13Ccarb values in the uppermost part of the Prygorodok Fm. rise to −2.54 ‰ within the Milivtsi–North section and to −2.05 ‰ in the Milivtsi–South section just below the nodular limestone bed marking the boundary with the Varnytsya Fm.

Interpretation of the environment of deposition

The proposed interpretation of the environment of deposition of the Prygorodok Fm. is a system of extremely restricted water bodies, likely cut off from open-marine waters following a drop in sea level. This interpretation is based on the following features of the studied deposits, selected from known features of carbonate lakes (summarized in Freytet and Verrecchia 2002; Gierlowski-Kordesch 2010):

  1. 1.

    laminated sediments;

  2. 2.

    deposition in conditions of oxygen depletion (as indicated by abundant organic matter);

  3. 3.

    massive carbonates;

  4. 4.

    dominance of microbialites;

  5. 5.

    marginal or reworked basal deposits;

  6. 6.

    an apparent lack of well-defined sequences or sedimentary subenvironments (Tucker 1978; Valero-Garcés and Aguilar 1992);

  7. 7.

    lack of fauna or restricted, monospecific assemblages, composed predominantly of ostracods and soft-bodied burrowing fauna;

  8. 8.

    lack of evidence of tidal activity in spite of shallow-water environment (Tucker 1978);

  9. 9.

    high lateral variability.

It is difficult to confirm the lacustrine nature of a pre-Devonian basin, given that indicative organisms specifically adapted to such environments had not yet evolved. Coastal lakes and ponds, however, are a ubiquitous (although individually ephemeral) element in the geomorphology of tropical carbonate platforms (e.g., Dix et al. 1999). Strongly depleted δ13Ccarb values (up to −10.53 ‰) indicate intensive biological fractionation, which is in agreement with the dominance of stromatolites throughout the formation and the high content of organic matter (especially in the basal part). The environment is envisaged as an enclosed lagoon or coastal lake rather than an open-marine (although extremely epeiric) environment, because the latter would require that both sea-floor anoxia and a benthic microbial carbonate factory developed in the subtidal zone (compare Calner 2005a). Small bodies of water are much more prone to develop restricted conditions and respond to seasonal climatic changes, which are reflected in depositional cyclicity, as observed in the Prygorodok Fm. with respect to redox conditions. An extremely calm environment, which allowed sedimentation from suspension and preservation of bentonite beds, as well as the lack of evidence for tidal or wave activity in spite of the very shallow depth, also support the proposed interpretation.

The Varnytsya Formation

Deposits representing the Varnytsya Fm. were recognized in the Kudryntsi-Castle outcrop. The boundary with the underlying Prygorodok Fm. can be found in the Kudryntsi-Cowpath and Milivtsi–South and North (Fig. 4l) sections as a nodular limestone bed in the lowermost part of the Kudryntsi–Castle section (Figs. 2, 7), marking the beginning of fully developed peritidal deposition. In the most continuous section, Kudryntsi–Castle, 6-m-scale cycles have been recognized and an idealized cycle is presented in Fig. 8. Since the top of the formation is not exposed in the area, the exact thickness of the Varnytsya Fm. could not be determined, but it is at least 24 m.

Fig. 7
figure 7

The Kudryntsi–Castle section. a View of the quarry, total height is approx. 24 m. b Nodular limestone bed, hammer for scale. c Nodular limestone—cemented contact between a nodule (top) and marly matrix (bottom), scale 500 μm. d Peloidal-bioclastic grainstone (PBWG in Table 3) with two types of calcareous microproblematica: one funnel-shaped and one ostracod-like (compare Fig. 9f), scale 200 μm. Supratidal zone. e Collapse breccia resulting from dissolution of evaporites, scale 5 mm. f Organic-rich mudstone with reworked clasts of underlying rock and coalified stems of Primochara calvata(?), overlain by claystone; scale 5 cm. g Lithoclastic-bioclastic floatstone-rudstone with peloids (LBF in Table 3); scale 1 cm. h Stromatolitic boundstone breccia (SBB in Table 3), scale 5 mm

Full size image
Fig. 8
figure 8

Idealized peritidal cycle of the Varnytsya Fm

Full size image

Subtidal facies

Two different facies have been distinguished in the subtidal deposits of the Varnytsya Fm.: deeper-water, nodular limestones, and shallower-water, stromatoporoid-tabulate coral biostromes. In the lower part of the Varnytsya Fm., biostromes are absent and nodular limestones are depleted in shelly fauna. Faunal diversity increases along with the development of biostromes towards the top of the section.

Nodular limestones (NL in Table 3) are characterized by dark, hard nodules embedded in a lighter, marly matrix. The nodules are formed by bioclastic mudstone to packstone, commonly with peloids (Fig. 7c). They form continuous layers in the lower part of each bed and become disrupted towards the top (Fig. 7b), where they are penetrated by burrows lined with bioclasts, which are commonly selectively dolomitized.

The shallow subtidal facies is represented by stromatoporoid-tabulate biostromes and bioturbated bioclastic limestones containing abundant brachiopods, ostracods, crinoids, bryozoans, rugose corals, and rare fragments of trilobites (BW and PBGW in Table 3). The thickness of biostromes does not exceed 1.5 m. They intercalate laterally with bioclastic limestone, which exhibits an increasing degree of environmental restriction, reflected by a decreasing diversity of fossils and increase in numbers of beyrichid ostracods (OWP and OM in Table 3; Fig. 9k–m). Bryozoans and tabulate corals are fragmented, indicating shoreward or off-biostrome transport. The transition to the intertidal facies (see below) is associated with peloidal-bioclastic wackestone-grainstone (PBWG in Table 3; Fig. 7d), which is interpreted as lagoonal sediment formed in conditions of increasing salinity and restricted circulation, but under wave action. The boundary with the overlying laminites is often developed as an erosional surface in partially lithified mud, covered with a flat-pebble conglomerate (LBFR in Table 3; Fig. 7g).

Fig. 9
figure 9

The Zavalya 1 section. a Overview of the outcrop. The height is ca. 21 m. b Horizons of cavities filled with clayey residue rich in iron oxide formed in dolomitic laminites; tape shows 20 cm. c Collapse breccia composed of angular clasts of the surrounding dolomite, cemented by sparry calcite and red clay. d Nodular limestone bed; hammer for scale. e–h Bioclastic wackestone (microfacies BW in Table 3). e Bioturbation indicated by the arrangement of bioclasts. f Microproblematicum resembling an unusual ostracod shell associated with Microcodium; scale 100 μm. g Cross section through a conodont element, scale 200 μm. h Calcified cyanobacterial tubes resembling Girvanella; scale 100 μm. i Cross section through a bundle of calcified tubes of Tuxekanella simplex Riding and Soja (1993), scale 100 μm. j Longitudinal section through a bundle of T. simplex; scale 100 μm. k Polished slab of an ostracod mudstone (OM in Table 3) from the intertidal zone; scale 1 cm. l, m Ostracod wackestone-packstone (OWP in Table 3). l Cap-in-cap structures formed by ostracods; scale 200 μm. m Accumulation of leperditicopid and beyrichid ostracods; scale 500 μm

Full size image

Intertidal facies

The intertidal facies is represented by stromatolitic dolomite and limestone (DSB in Table 3; Fig. 9b) and ostracod mudstone (OM in Table 3; Fig. 9k). Almost no trapping of detrital sediment was observed in the laminites. This indicates very high rates of accumulation of autochthonous carbonate. The presence of desiccation cracks and flat-pebble conglomerates (LBFR in Table 3; Fig. 7g) indicates that the sediment was subjected to periodic subaerial exposure, and that lithification within the microbial mat was rapid. The intertidal deposits are barren, except for rare eurypterids and ostracods (Fig. 9k).

Supratidal facies

The boundary between intertidal and supratidal deposits is gradual and placed within the laminite facies. The following features distinguish the supratidal facies: (1) internal breccia composed of angular intraclasts embedded in orange, clayey matrix (SBB in Table 3; Fig. 7e, h); this breccia is interpreted as resulting from sediment fracturing during evaporite growth and collapse following its dissolution; (2) abundant mudcracks; and (3) horizons of cavities filled with an easily disintegrating argillaceous, porous marl. These horizons are usually conformable with bedding, but locally affect several adjacent beds. The porosity seems to result from dissolution of evaporite minerals.

Additionally, two horizons exhibit a sequence indicative of regolith formation. The sequence starts with 1–2 cm of dark brown mudstone (Fig. 7f), capped with a thin, creamy claystone crust with iron oxide spotting, devoid of carbonate. Up to 10-cm-long coalified stems, tentatively identified as Primochara calvata T. Ishchenko, 1975, are preserved in these beds, which are otherwise devoid of macroscopic fossils.

The δ13Ccarb values in the Varnytsya Fm. have been measured only in nodular limestones and were all negative (ranging from −0.33 to −3.20 ‰).

The Zavalya section

The Zavalya section consists of two outcrops located at a distance of ca. 600 m apart on the same valley slope. The basal part of the Zavalya 2 section (Figs. 2, 3) is developed as massive bioclastic-lithoclastic grainstone (BLG in Table 3; Fig. 10b–d). The grain-supported fabric is formed by fragmented and slightly abraded shells, some intensively bored (Fig. 10c). Sorting changes gradually within a given bed, from very well sorted and aligned shell fragments to a moderately sorted mixture of shells and rounded mudstone and peloidal grainstone extraclasts. There are also rare tangential and slightly eccentric ooids with sparry or micritized cortices (Fig. 10b). In view of the small size of the outcrop, it was not possible to observe macroscopic sedimentary structures, such as cross-bedding. These facies would be readily interpreted as shoal or levee deposits, if not for horizons of very high secondary porosity due to dissolution vugs, which expand primary pores and also affect bioclasts (Fig. 10d). At least two generations of blocky calcite cement are present. The first generation of cement, which surrounded original shells, preserves the empty voids left after their dissolution. The second generation fills these empty voids, and is gravitational, indicating a vadose diagenetic environment. The rocks were therefore deposited either as bioclastic shoals which became emergent, or were originally formed as beachrocks.

Fig. 10
figure 10

The Zavalya 2 section. a Outcrop view; the height is approx. 8 m. b–d Bioclastic-lithoclastic grainstone (BLG in Table 3). b A tangential and slightly eccentric ooid; scale 500 μm. c Scale 200 μm. d Scale 5 mm. e Stromatoporoid-tabulate packstone (STP in Table 3). The stromatoporoid is encrusted and penetrated by Girvanella tubes; scale 5 mm

Full size image

The grainstone is overlain by several low-relief stromatoporoid-tabulate biostromes (STP in Table 3; Fig. 10e). The interval directly above the biostromes is not exposed (Fig. 2, 3). The upper part of the outcrop begins with bioturbated bioclastic wackestone to packstone (BW in Table 3) with a diverse fully marine fauna, including bryozoans, nautiloids, brachiopods, gastropods, calcareous algae (Fig. 9h–j), crinoids, and ostracods (Fig. 9e). This facies represents the subtidal zone and is succeeded by thick beds of dolomitic laminites of the intertidal zone, with dissolution breccia of the same type as described above from the Varnytsya Fm. (Fig. 9b, c). In the upper part of the section, the fauna of the subtidal zone is reduced almost exclusively to beyrichid ostracods, commonly forming dense shell beds (Fig. 9k–m).

In the lower part (Zavalya 2), δ13C values are close to 0 (−0.64 and −0.84 ‰). In the upper part of the section (Zavalya 1), δ13C values reach +2.77 ‰ in the nodular limestone bed (Fig. 9d), and drop back to low negative (−1.03 ‰) in the marls above.

Discussion

Depositional environment of the Prygorodok Formation

The thickness of the deposits belonging to the Prygorodok Fm. in the studied area reaches 16 m. Low-energy sedimentation and the lack of a proper benthic carbonate factory of Paleozoic-type raise questions about the main sediment source. As no facies models for Silurian carbonate lakes exist, this environment will be discussed here in more detail.

Typically, for carbonate coastal lakes, the following mechanisms of sediment accumulation are considered (Gierlowski-Kordesch 2010):

  1. 1.

    fluvial input and surface runoff;

  2. 2.

    aeolian input;

  3. 3.

    material transported shoreward from the sea by hurricanes and tsunamis;

  4. 4.

    carbonate precipitation through biogenic mediation;

  5. 5.

    evaporation.

Tempestites occur in the Prygorodok Fm. as a subordinate component, and the evidence for evaporite growth is limited to cubic voids, interpreted as pseudomorphs left after halite dissolution (Fig. 5d). The presence of evaporites and lack of sedimentary structures characteristic of fluvial facies also makes the fluvial input less likely.

The bulk of the Prygorodok Fm. is dolomicrite. Dolomite precipitation from sea water through the activity of microorganisms is a likely mechanism. Their role in precipitation of dolomite has been attributed to the removal of sulphate by sulphate-reducing bacteria (Vasconcelos and McKenzie 1997; Van Lith et al. 2003; Sanz-Montero et al. 2008), but recent studies point to the organic matrix provided by microbial mats as the crucial factor. The site where microbial dolomite precipitation was originally observed (Lagoa Vermelha, Brazil) closely resembles the environmental model proposed here for the deposition of the Prygorodok Fm. in terms of climatic setting and hydrologic conditions. The site is an isolated lagoon with salinity that fluctuates with the evaporation/precipitation ratio in the wet and dry season, and depends also on the proportions of meteoric and marine waters in the groundwater supply (Vasconcelos and McKenzie 1997). In view of the relative novelty of modern analogue, few fossil examples of such a lacustrine-lagoonal environment have been recognized. Many ancient dolomite occurrences exhibit features associated with syndepositional or early diagenetic formation of dolomite under anoxic conditions and with a distinct cyclic microfacies pattern, which may correspond to the development and demise of microbial communities or seasonal climate rhythms (Vasconcelos and McKenzie 1997). Therefore, it seems likely that more fossil analogues may be identified in the future.

Laminae of organic matter enriched in silt- and fine sand-sized quartz grains in the lower part of the unit indicate that in the early stages of deposition the basin underwent stratification with respect to redox conditions, with repeated episodes of sediment oxidation associated with the influx of silt-sized fluvial or aeolian clastic material. This stratification regime was probably not maintained during deposition of the massive, argillaceous dolomite with bioturbation.

In the mid- to upper part of the Prygorodok Fm., leperditicopid ostracods in the massive dolomite facies, together with bioturbation, indicate that the factor(s) which limited the development of a more diverse fauna were not dysoxia. Salinity is another likely restrictive factor. Ephemeral coastal lakes are typically unstable environments subject to drastic hydrological fluctuations. The cubic casts found in the Milivtsi–South section (Fig. 5d) are the only sedimentary evidence for elevated salinity. Westwards across the strike, Tsegelnyuk et al. (1983, p. 68) reported the occurrence of translucent gypsum and anhydrite in boreholes corresponding to the Prygorodok Fm. In younger deposits, the nature of the biota is often a useful reflection of the salinity regime, but no proper freshwater ecosystems existed in the Silurian. The association of ostracods and microbialites in the absence of other organisms may indicate elevated or reduced salinity. Monospecific leperditicopid ostracod assemblages are typical indicators of restricted settings, such as confined lagoons and environments affected by periodic subaerial exposure (Vannier et al. 2001). It is not possible at this point to confirm the salinity regime of the basin during deposition of the Prygorodok Fm. Lack of fauna or low-diversity assemblages could be explained by any one of three factors: unfavorable redox conditions, a yet to be filled Silurian ecospace, or finally, nutrient deficiency. The latter has been evoked as a possible explanation for the barren character of Paleozoic lake environments (Park and Gierlowski-Kordesch 2007).

In addition to biologically mediated dolomite precipitation, a subordinate aeolian admixture is represented by the bentonite beds and by the laminated fine-grained sandstones-claystones of volcanic origin (Fig. 6d, e). Silt-sized quartz admixture visible in the laminated mudstone, which locally forms sheet-like beds several centimeters thick, cemented with carbonate and stabilized by undulating dolomitic microbial crusts, is also a possible candidate for an aeolian origin. Over-representation of wind-derived sediment may reflect increased dustiness coincident with the Ludfordian CIE, as proposed by Kozłowski and Sobień (2012, see discussion therein).

The record of sea-level changes

Studied sections are presented in Fig. 2 along with a correlation based on topographic position, lithology and stable carbon isotope trends (Fig. 11). In spite of the facies model of Einasto et al. (1986), where the Silurian carbonate platform of Podolia is treated as rimmed, it is assumed here that a carbonate ramp is a more appropriate model due to the very low-relief and spatial discontinuity of the biostromes and reef mounds and the epeiric character of the Podolian part of the basin (compare Eriksson and Calner 2008; Skompski et al. 2008).

Fig. 11
figure 11

Summary of sections with sequence-stratigraphical interpretation for the Kudryntsi–Castle section. HST highstand systems tract, TST transgressive systems tract, SB sequence boundary, MFS maximum flooding surface, FS flooding surface

Full size image

The first recognizable element of stratal architecture is the sequence boundary (SB) marking the top of the Isakivtsy Fm. The second element is the stacking pattern recorded in the peritidal deposits of the Varnytsya Fm. Figure 11 shows changes in cycle thickness, which, it is assumed, reflect the amount of accommodation space available. The first three cycles show an aggradational stacking pattern, which is a characteristic feature for early highstand systems tract (HST). The next three cycles are substantially thinner and exhibit in their upper parts extensive evidence of subaerial exposure, indicating a gradual loss of accommodation, typical of the late HST. The entire cyclic sequence is interpreted as HST, with the first three cycles having been deposited during a sea-level stillstand, and the following three, during a relative sea-level fall.

The Prygorodok Fm. does not record the depositional processes, which took place following the sea-level fall and subaerial exposure of the Isakivtsy Fm. It is likely that the sedimentary gap above the sequence boundary corresponds with deposition of a falling stage systems tract farther down the ramp, followed by a lowstand systems tract. The Prygorodok Fm. deposits would therefore represent the transgressive systems tract (TST), ending in the maximum flooding surface (MFS) clearly distinguishable in all sections as a condensed, fossiliferous nodular limestone bed (Fig. 11). Subsequent to this peritidal facies represent the HST and the transition from the Prygorodok Fm. to the Varnytsya Fm. In the present study, no sedimentological features were found which would point to when the open-marine waters transgressed into the Prygorodok lagoon or lake basin. The upper part of the formation is developed as microbial laminites with numerous traces of exposure, resembling the supratidal zone deposits of the Varnytsya Fm.

The position of the Zavalya section

The dolosparite of the Isakivtsy Fm., exposed north of Milivtsi as a single, nearly continuous outcrop over a distance of 1.3 km, does not outcrop elsewhere in the studied area. Geometrically, the Zavalya 1 section corresponds to the topographic level at which the Isakivtsy dolomite is found, however, the facies in the Zavalya 1 section are entirely different. They represent peritidal deposits resembling the Varnytsya Fm. described above, with the topmost part dominated by marly, stromatolitic dolomite with a monospecific ostracod fauna. The combined Zavalya 1 and 2 sections and the Isakivtsy-Prygorodok transition show opposite sea-level trends: the Zavalya section starts with the facies deposited in a shallow, but marine environment with fully developed biostromes (Fig. 10), and grades into marginal-marine deposits with evidence of exposure and evaporite formation (Fig. 9b, c), and can be therefore summarized as regressive, whereas the Prygorodok-Varnytsya transition represents late transgressive to highstand conditions. Moreover, δ13C values in the Varnytsya deposits, both in this study (Fig. 2) and reported from neighboring sections (Kaljo et al. 2007, 2012), are close to 0 or slightly negative, whereas the data here record a positive peak (+2.77 ‰, value from brachiopod shell) in the upper part of the section, supporting its position at the level of the CIE at the Isakitvtsy-Prygorodok boundary.

As the Isakivtsy Fm. is clearly defined based on diagenetic features, it cannot be ruled out that the deposits in Zavalya represent the same original facies that have not undergone dolomitization. The lateral distance between the two sections is ca. 6.5 km (approximately along the strike). A possible reason for patchy spatial distribution of dolomitization might be a reflection of lateral variability of the overlying Prygorodok Fm., supporting the presented environmental interpretation.

Relationships with the global facies record and eustatic trends

The sequence boundary present at the Isakivtsy-Prygorodok transition, recording prolonged exposure of the carbonate platform, may be correlated with the widely observed sea-level fall coincident with the base of the mid-Ludfordian CIE (see review in Loydell and Frýda 2011). In the carbonate succession of Gotland this event is recorded as an erosion surface at the Hemse/Eke boundary (Jeppsson et al. 2007) and the occurrence of grainstone and stromatolites, and dissolution pipes, indicating karstification in the overlying Lower Eke Fm. (Cherns 1982; Eriksson and Calner 2008). In the clastic carbonate succession of the Rzepin section in the Holy Cross Mountains (Kozłowski and Munnecke 2010) the SB interval is developed as the sandy-oolitic Jadowniki Member, which contains numerous erosive surfaces and signs of subaerial exposure in the top of the oolite-bearing unit (e.g., keystone vugs, Kozłowski 2003; Kozłowski and Munnecke 2010). In the Shropshire area (UK), the pronounced positive excursion in organic carbon isotopes occurs near the Ludlow Bone Bed and is followed by progradation of the Downtonian facies (Loydell and Frýda 2011).

In the case of more distal depositional settings, common shallowing-upward facies trends are observed near the base of the CIE (Scania—Wigforss-Lange 1999; Australia—Jeppsson et al. 2007); however, the shallowing trends in these cases continues also above the onset of the CIE (Martma et al. 2005; Kozłowski and Sobień 2012), indicating prolonged progradation during the sea-level lowstand (Kozłowski and Munnecke 2010).

It is important to note that the sections mentioned above represent foreland basin settings with both high rates of subsidence and sedimentation; hence the sedimentary record includes the sequence boundary and early transgressive deposits. In other cases the gap is more extensive, often encompassing the lower part or the entire CIE interval. In the case of the Prague basin, the shallower facies belt is characterized by erosive-paleokarst surfaces, until the decline of the CIE in the Mušlovka Quarry and probably after the CIE decline in the case of the Požáry Quarry (Lehnert et al. 2007). Similar situations are observed in the proximal shelf of Baltica (Ohesaare drillcore—Kaljo et al. 1997) and Laurentia (Barrick et al. 2010).

In the case of Podolia, regarding the stable carbon isotope stratigraphic data of Kaljo et al. (2007, 2012), the SB is diachronous with respect to the main positive shift of the CIE. It is suggested here that this delay in the placement of the SB results from a deeper-water setting inherited after the flooding interval preceding the regression associated with the Lau excursion and represented by the Grinchuk Fm. deposited in an open-shelf environment. Hence, the regression resulted from both a sea-level fall and filling up of the accommodation space by Isakivtsy sediments.

Sedimentation seems to have been re-established relatively early given the proximal position of studied sections with respect to the Baltica shore, resulting in a more complete record. The stable carbon isotope stratigraphic data of Kaljo et al. (2007, 2012) indicate that the first flooding surface after regression, recorded as the Isakivtsy/Prygorodok boundary is diachronous. The base of the Prygorodok Fm. in the western part of the Dniester valley outcrop area (Isakivtsy-45 of Kaljo et al. 2012) corresponds to the end of the rising limb of the CIE, whereas in the eastern part (5 km to the east, in Braga) the Prygorodok base records CIE decline (Kaljo et al. 2007). Regarding these data and the facies development of the formation in this study, with a slow deepening-upward trend and exposure levels within the succession, the sea-level rise may be interpreted as relatively slow with successive internal flooding surfaces and filling events. This interpretation agrees with the record of the CIE maximum interval in the Holy Cross Mountains, where it is represented by lagoonal sediments of the Bełcz Member and interpreted as early transgressive deposits (Kozłowski and Munnecke 2010). In the case of Gotland, the beginning of the transgression seems to be more prominent (middle and upper Eke Formation, Eriksson and Calner 2008), but it is followed by progradation of the Burgsvik sandstone, which causes the loss of accommodation. In the more distal setting represented by the sediments in the Vidukle drillcore in Lithuania (Martma et al. 2005) biofacies data also suggest the predominance of a progradational-regressive trend, probably accompanied by slowly rising sea level. Spectral gamma ray logs from the open-shelf Mielnik IG-1 core (Kozłowski and Sobień 2012) also indicate the domination of progradational conditions during the entire CIE maximum interval, with terrigenous influx towards the end, consistent with the record from Gotland, Holy Cross Mountains, and the Vidukle core.

The next stage of sedimentation, associated with the end of the CIE decline, which relates to the upper part of the Prygorodok Fm., records in all discussed sections a predominance of transgressive conditions. The facies record in this study area preserves the maximum flooding interval of this transgression in the lowermost part of the Varnytsya Fm., marked by an up to 3-m-thick dark nodular limestones with a more open-marine fauna.

At this level Kaljo et al. (2012) noted minimal δ13C values, referenced as the “post-Prygorodok isotope low”. A similar low in the C-isotope record is observed in the Mielnik IG-1 section (Kozłowski and Sobień 2012) and corresponds with the most pronounced MFSs with condensations of the Monograptus balticus–Pseudomonoclimacis latilobus fauna in the offshore setting. Above this level, geophysical and facies data in the Mielnik IG1 section reflect a longer highstand interval. A similar sequence development above the CIE is recognized in Gotland (see Eriksson and Calner 2008, Fig. 10e and related text); therefore the peritidal Varnytsya Fm. may be correlated with the Hamra-Sundre highstand interval.

Conclusions

  1. 1.

    The record of the Ludfordian positive CIE known as the Lau excursion in the Isakivtsy-Prygorodok deposits in the Zbruch River Valley is highly influenced by biogenic carbonate precipitation; the δ13Ccarb values in this interval shift strongly towards the negative and constitute an example of facies overprint on the carbon isotope record of sea-water composition.

  2. 2.

    The Prygorodok Fm. is represented by laminated dolomicrite with bentonite and siltstone intercalations, deposited in enclosed lagoons or coastal lakes dominated by microbial mats, which were the main sediment producers, as indicated by δ13Ccarb values as low as −10.53 ‰.

  3. 3.

    The Prygorodok Fm. represents TST deposits following a sequence boundary, which resulted from a very rapid regressive event recorded in the Ludfordian of Baltica and other paleocontinents.

  4. 4.

    The Varnytsya Fm. is represented by peritidal facies, encompassing deeper subtidal nodular limestones, shallow subtidal stromatoporoid-tabulate biostromes and bioclastic limestones, intertidal stromatolitic laminites and supratidal regoliths. The δ13Ccarb values remain close to 0 ‰ or slightly negative.

  5. 5.

    The sequence boundary in Podolia, which can be correlated across the entire basin (e.g., with Gotland, Holy Cross Mountains), is diachronous with respect to maximum positive δ13Ccarb values in the area, reflecting inherited bottom morphology and proximity to the shoreline.

  6. 6.

    The MFS marking the onset of peritidal deposition of the Varnytsya Fm. corresponds to the “post-Prygorodok isotope low” of Kaljo et al. (2012) and to Monograptus balticus–Pseudomonoclimacis latilobus condensation surfaces in the open-shelf Mielnik-IG1 section (Kozłowski and Sobień 2012).