Abstract

New and published data are integrated herein to resolve the age and stratigraphic relationships for problematic strata of the Aeronian and Telychian (Llandovery; Silurian) in Ohio and Kentucky (USA). At least two major depositional sequences were traced along the eastern flank of the Cincinnati Arch; these are separated by a regionally angular unconformity with complex topography. Underlying units are progressively truncated to the northwest while overlying strata change facies, condense, and onlap in the same direction.

The basal unit of the upper sequence is the Waco Member of the Alger Shale Formation in Kentucky and southern Ohio and the Dayton Formation in western Ohio. A persistent, positive carbonate carbon isotope (δ13Ccarb) excursion associated with the mid-Telychian Valgu Event is recognized in the upper subunit of the Waco Member; the absence of a comparable signal in the Dayton Formation corroborates interpretations that it is significantly younger.

The correlations proposed here can be used to understand the nuanced depositional history and chronostratigraphic completeness of the lower Silurian in eastern North America. This framework can be used to characterize sea-level history and local conditions that prevailed during global paleoenvironmental events.

INTRODUCTION

The Aeronian and Telychian Stages of the Silurian record the transition from the Late Ordovician mass extinction and early Silurian recovery (Raup and Sepkoski, 1982; Sheehan, 2001; Krug and Patzkowsky, 2007) to the ecological upheaval of the early Sheinwoodian Ireviken Event (Jeppsson et al., 1995; Munnecke et al., 2003; Lehnert et al., 2010). The richly fossiliferous successions of eastern North America provide an important venue for study of this transition (Zaffos and Holland, 2012), being represented by extensive exposures that have a long history of investigation (Hall, 1852; Orton, 1870; Foerste, 1906). However, the temporal and stratigraphic resolution of this depositional framework has been hindered by a scarcity of index fossils, pervasive dolomitization, and the lack of a uniformly applied nomenclature (see reviews of Berry and Boucot, 1970; McLaughlin et al., 2008b; Brett et al., 2012).

This study integrates carbon isotope chemostratigraphy, facies analysis, and sequence stratigraphy to resolve the depositional history of Llandovery units exposed in Ohio and Kentucky at the transition between the Appalachian foreland basin and Cincinnati Arch (Fig. 1). This new synthesis lays the groundwork for highly refined chronostratigraphic interpretations, a revised sequence stratigraphic framework, and a more complete understanding of the Silurian paleoenvironments and far-field tectonic activity in Laurentia.

GEOLOGIC SETTING

Early Silurian Tectonics and Paleogeography

The study area is situated near the western margin of the Appalachian Basin, which was located at ∼lat 20°S–30°S during the early Silurian (Cocks and Scotese, 1991). Starting in the Late Ordovician, accretion of island arcs onto the eastern margin of Laurentia produced several episodes of mountain building called the Taconian orogeny (Ettensohn and Brett, 2002). This produced structural loading and subsidence in the Appalachian Basin, which was rapidly filled with clastics flushed off the newly formed Taconic highlands (Beaumont et al., 1988; Ettensohn and Brett, 1998; Ettensohn, 2008). This event also created a structural arch on the western margin of the basin, produced by flexure of the crust in response to the strain of accretion, and facilitated by deep-seated basement faults (Quinlan and Beaumont, 1984; Root and Onasch, 1999).

Toward the middle of the Rhuddanian, orogenic activity had begun to taper, but renewed tectonism and uplift (the Salinic orogeny) is recorded by the thick packages of strata spanning the Llandovery to Ludlow Series (Goodman and Brett, 1994; Ettensohn and Brett, 1998; Brett et al., 1998; van Staal et al., 2009; Ettensohn et al., 2013).

The study area is within a transitional zone between deep-water, mud-dominated depositional environments of the central Appalachian Basin and the shallower, carbonate-dominated systems in the northwest (Fig. 2; Hunter, 1970; Brett et al., 1990, 1998). The bathymetric high occupied by this bank (the proto–Cincinnati Arch) approximates the axis of the modern Cincinnati-Findlay-Algonquin arch system (Root and Onasch, 1999).

Chronostratigraphy

Conodonts are widely used for biostratigraphy in the calcareous lower Silurian rock units found in eastern North America (e.g., Rexroad et al., 1965; Rexroad, 1967; Rexroad and Nicoll, 1972; Cooper, 1975; Kleffner, 1987, 1994). However, most of this work was conducted prior to recent advances in the field of conodont biostratigraphy, most notably the development of highly refined conodont biozonations (Fig. 3; Jeppsson, 1997; Jeppsson et al., 2006; Männik, 2007a, 2007b). Earlier reports of Llandovery conodonts from the eastern flank of the Cincinnati Arch employed a comparatively low resolution zonation that was inconsistently applied (see discussion by Kleffner, inMcLaughlin et al., 2008b). Although these previous studies did much to constrain the age and correlations of Llandovery strata, new initiatives have demonstrated promise for a much greater refinement in conodont-based correlations (Loydell et al., 2007; Kleffner et al., 2012; Cramer et al., 2011).

These improvements coincide with the rise of carbonate carbon isotope (δ13Ccarb) chemostratigraphy as a tool for correlation of Silurian rocks (Saltzman, 2002; Cramer et al., 2010, 2011). Discrete and time-specific intervals characterized by high δ13Ccarb values (i.e., positive excursions) have recently been characterized in the well-constrained sections of the Baltic region, Anticosti Island, and the U.K. (e.g., Munnecke et al., 2003; Kaljo and Martma, 2006; Munnecke and Männik, 2009; Hughes et al., 2014). Such signals are known to be resistant to late diagenetic alteration (Saltzman et al., 2000) and recognizable in a wide variety of facies and depositional settings (McLaughlin et al., 2012). A wealth of new δ13Ccarb data has recently been summarized and integrated with biostratigraphic zonations to create a composite δ13Ccarb curve tied to the Silurian time scale, which provides a powerful tool for regional and global correlation (Fig. 3; Cramer et al., 2011).

Recognition of δ13Ccarb excursions in Silurian strata of eastern North America has greatly facilitated correlation, even when biostratigraphic data are limited (Saltzman, 2001; Cramer and Saltzman, 2005; Cramer et al., 2006; McLaughlin et al., 2012). At least one global δ13Ccarb excursion has been recognized in Llandovery strata of the Appalachian Basin; this is closely associated with the Valgu Event, a phase of biotic and climatic turnover recorded in the Pterospathodus eopennatus conodont Superzone (Männik, 2005, 2007a; Munnecke and Männik, 2009; McLaughlin et al., 2012). The Valgu excursion therefore provides a useful chronostratigraphic anchor for Llandovery strata in the western Appalachian Basin.

Sequence Stratigraphy

In Brett et al. (1990, 1998) a sequence stratigraphic framework was established for the Silurian of the Appalachian Basin that comprises six third-order sequences (in ascending order, S-I to S-VI), roughly equivalent to group-level stratigraphic units. Although the primary basis for this framework was the classic Niagaran Series of western and central New York, it was subsequently extended into east-central Kentucky and south-central Ohio (Brett and Ray, 2005; Cramer, 2009).

Comparatively thick successions of siltstone, shale, and marl have been interpreted as highstand and falling stage deposits in these sections (Brett et al., 1990; McLaughlin et al., 2008b). Transgressive intervals are often associated with authigenic minerals (e.g., glauconite, hematite, phosphate, or pyrite), abundant conodonts, hardgrounds and/or firmgrounds, and reworked clasts (see Brett et al., 1998; McLaughlin et al., 2008a). The link between these unusual facies and transgression has been attributed by some to condensation (i.e., Brett et al., 1990; McLaughlin et al., 2008a); however, factors relating to seawater chemistry and redox conditions likely play an important role as well (McLaughlin et al., 2012).

The rocks studied here have been assigned to sequences S-I, S-II, and S-IV; sequence S-III was presumed absent in Ohio and Kentucky due to a regional angular unconformity beneath S-IV in New York that progressively truncates underlying beds to the west (Fig. 4; Brett et al., 1990, 1998; Brett and Ray, 2005). New chronostratigraphic data suggest that stratigraphic units previously recognized as the basal transgressive systems tract of S-IV are of different ages in Ohio (Cramer, 2009; McLaughlin et al., 2008b, 2012; Kleffner et al., 2012), Kentucky (Sullivan et al., 2014a), and New York (Loydell et al., 2007; Sullivan et al., 2014b). Ongoing work (Sullivan et al., 2012; Ettensohn et al., 2013) attempts to reconcile these new results with the sequence stratigraphic interpretations in Brett et al. (1990).

STRATIGRAPHIC FRAMEWORK

Mixed Carbonate-Shale Facies of Kentucky and Southern Ohio

In east-central Kentucky and south-central Ohio, the names and definitions of Silurian stratigraphic units at group and formational rank have undergone many revisions that are inconsistently applied (Fig. 4; Foerste, 1935; Rexroad et al., 1965; Simmons, 1967; McDowell, 1983; Ettensohn et al., 2013). However, these stratigraphic divisions are invariably built around a succession of lithologically distinct subunits, the terminology of which has remained stable since the report by Foerste (1906) on the Silurian stratigraphy of the Cincinnati Arch. In ascending stratigraphic order, this succession comprises the Brassfield, Plum Creek, Oldham, Lulbegrud, Waco, and Estill units (Fig. 4; Foerste, 1906; Brett and Ray, 2005; Ettensohn et al., 2013).

The upper beds of the Brassfield, sometimes informally referred to as the Rose Run iron ore or upper massive member, are genetically distinct from the underlying strata (Fig. 5; Foerste, 1906; Gordon and Ettensohn, 1984). The subunit is recognizable by dark red, ferruginous dolograinstone bearing cogwheel-shaped crinoid columnals, commonly referred to as “beads” (Fig. 6; Foerste, 1906; Rexroad et al., 1965; McDowell, 1983), but more accurately identified as the morphogenus Floricolumnus (col.) sp. (Donovan and Clark, 1992). These show evidence of reworking out of the underlying shaly member of the Brassfield (Thomka et al., 2013).

The overlying Plum Creek Shale Member (of the Drowning Creek Formation) is a relatively thin (1–2 m) interval of blue-gray mudstone with silty interbeds; this in turn is overlain by 3–4 m of bedded ferruginous dolostone and shale called the Oldham Member (of the Drowning Creek Formation), which is characterized by abundant brachiopods, particularly the distinct large pentamerid Ehlersella norwoodi (Foerste, 1906; Rexroad et al., 1965). In the southern part of the study area, the Oldham and Waco are separated by the Lulbegrud Shale Member (of the Alger Shale Formation), a 3–5-m-thick, unfossiliferous blue-gray shale (Foerste, 1906).

In the vicinity of central Kentucky, the Waco consists of a 1-m-thick basal carbonate horizon overlain by 3–4 m of green-gray shales interbedded with fossiliferous dolostones and siltstones; however this is sometimes truncated by the sub-Devonian Wallbridge Unconformity (Fig. 7; Foerste, 1906, 1935; McDowell, 1983; Sullivan et al., 2012). The contact with the overlying Estill Shale Member of the Alger Shale Formation is a cryptic shale-shale boundary, recognizable by abundant, granular glauconite overlain by one or more bands of bright red mudstone (Rexroad et al., 1965; McDowell, 1983).

At localities in southern Ohio, the Waco manifests as stacked dolomitic carbonates that were historically labeled “Dayton” (Fig. 4; Rexroad et al., 1965; McDowell, 1983). Here it can be subdivided in to a lower light colored, fossiliferous subunit, informally termed white Waco, and an upper ferruginous, heavily bioturbated subunit, termed the orange Waco (Fig. 8; Sullivan et al., 2014a). In southern Ohio, the Estill unit is a thick succession of green, red, and maroon shales with zones of abundant glauconite and occasional brown, dolomitic calcareous siltstone beds (Rexroad et al., 1965; McLaughlin et al., 2008b, 2012).

Many meters of strata, comprising the Plum Creek, Oldham, and Lulbergrud units, separate the ferruginous upper massive member of the Brassfield Formation and the Waco Member of the Alger Shale Formation in central Kentucky, but these units are progressively truncated to the northwest by a sub-Waco unconformity (Fig. 9). At its northern terminal extent, the Waco may directly overlie the Brassfield (Figs. 5 and 8).

Age control for these strata is limited. Conodont samples from the Brassfield have yielded specimens of Distomodus kentuckyensis and Ozarkodina hassi, suggesting a Rhuddanian or possibly earliest Aeronian age for this unit (Cooper, 1975; Kleffner, inMcLaughlin et al., 2008b). The presence of the brachiopod Ehlersella in the Oldham also points to a middle Aeronian age (Rexroad et al., 1965; McLaughlin et al., 2008b).

Although conodonts are sparse in the Waco, McDowell (1983) assigned it to the Pterospathodus celloni Zone. The lower Estill was also assigned to the Pt. celloni Zone by McDowell (1983). The rest of the unit was assigned to the Pt. amorphognathoides Zone, a diagnosis that was later corroborated in Kleffner (1987). One of the co-authors on this study (Kleffner) identified zonal conodonts of the Pt. eopennatus Superzone in the Waco at Eagle Stone Quarry in Brown County, Ohio (Figs. 2 and 9; McLaughlin et al., 2008b). This is consistent with elevated δ13Ccarb values (+2‰ to +3‰) recorded in the upper Waco, interpreted as a local manifestation of the lower Telychian Valgu excursion (McLaughlin et al., 2012). Although the lower Estill has not been placed within the updated biozonation of Männik (2007a), the middle and upper parts of the unit have been assigned to the Pt. am. amorphognathoides Zone based on the occurrence of Oz. polinclinata and Pt. am. amorphognathoides (Kleffner, reported in McLaughlin et al., 2008b).

Carbonate-Dominated Facies of Western Ohio

In western Ohio, the sub-Waco unconformity caps a complex association of lower Silurian dolograinstone and shale containing abundant corals, echinoderms, and cephalopods (Fig. 10). This is commonly called the Brassfield Formation, but its precise relationships to strata of that name in central Kentucky are uncertain (McLaughlin et al., 2008b). It is informally subdivided into a lower white Brassfield, and an upper red Brassfield (Fig. 4; sensu McLaughlin et al., 2008b). Recognition of the conodont Icriodina stenolophata and the brachiopod Ehlersella in the red Brassfield (Rexroad et al., 1965; C.B. Rexroad, personal commun., 1983) have led some to suggest equivalence with the Oldham of Kentucky (McLaughlin et al., 2008b). Although Rexroad et al. (1965) reported Floricolumnus (col.) sp. in the red Brassfield near the city of Dayton, this has not been corroborated in subsequent field work, and other echinoderm faunal data do not suggest a correlation (Ausich et al., 2015).

Unconformably overlying the red Brassfield is the Dayton Formation, which consists of light gray dolomitic carbonates that often display a styolitic, nodular texture. Unlike the Waco, it is sparsely fossiliferous, though rare pentamerid brachiopods are found (Fig. 10). Mineralized hardgrounds with abundant pyrite and phosphate occur at several horizons (Fig. 10; McLaughlin et al., 2008b). The overlying Osgood Formation is a succession of blue-gray mudstones interbedded with dolomitic marls that has been correlated with the upper Estill Shale (McLaughlin et al., 2008b, 2012; Brett et al., 2012). The Dayton thins progressively to the west. In the vicinity of Preble and Miami Counties, Ohio, it is present only as a thin (10–20 cm) phosphatic carbonate horizon that overlies the red Brassfield (Kleffner, 1994; Kleffner et al., 2012).

Pt. am. amorphognathoides and Oz. polinclinata have been recovered from the Dayton, suggesting a late Telychian age; Pt. am. amorphognathoides is also found in the lower Osgood, but the presence of Kockelella ranuliformis in higher strata suggest this stratigraphic unit straddles the Llandovery-Wenlock boundary (Kleffner, 1990; McLaughlin et al., 2008b; Kleffner et al., 2012). Elevated or rising-upward δ13Ccarb values in the Osgood are interpreted as features of the early Sheinwoodian Ireviken positive excursion (Munnecke et al., 2003; Cramer, 2009; McLaughlin et al., 2008b, 2012).

METHODS

We measured and sampled 20 outcrops along a proximal-to-distal transect of the Cincinnati Arch between Madison County, Kentucky, and Greene County, Ohio (Fig. 3; Table 1). Material was collected at 10–30 cm intervals depending on availability of carbonate. Powdered rock samples were generated for δ13Ccarb analysis using a power drill with a tungsten carbide bit.

Samples were generated from minimally altered samples, with the goal of isolating primary micrite. Vugs, stylolites, weathering varnishes, and other features of clearly late diagenetic origin were avoided. Processed samples were sealed in plastic capsules and sent to the W. M. Keck Paleoenvironmental and Environmental Stable Isotope Laboratory (KPESIL) at the University of Kansas (Lawrence). Carbonate samples were mixed with phosphoric acid, and carbon dioxide produced by the reaction was analyzed by a ThermoFinnigan GasBench II in-line with a Finnigan MAT 253 isotope ratio mass spectrometer. The δ13Ccarb and δ18Ocarb values were analyzed with respect to internal standards and are presented here in per mil (‰) notation, normalized to the Vienna Peedee belemnite standard.

RESULTS

Physical Stratigraphy

The Waco Member of the Alger Shale Formation and upper massive member of the Brassfield vary little in their overall thicknesses and lithologic characteristics, which are recognizable at nearly every locality between Madison County, Kentucky, and Highland County, Ohio (Fig. 1). Both units overlie sharp lithologic contacts, and the intervening strata thin progressively to the north (Fig. 9). At all localities in Estill and Madison Counties, Kentucky, the base of the Waco is a thin (3 cm) bed containing mineralized surfaces (Fig. 7); above this, the calcareous basal Waco is a 40–50 cm interval of light gray dolostone containing favositid corals, glauconite, and pyrite (Fig. 7). This is overlain successively at all localities by ∼10 cm of shale capped by a carbonate bed with abundant hypichnial burrows referable to Teichichnus sp. (Fig. 7).

At several sites, the Waco is unconformably overlain by the Devonian-age Boyle Formation (Fig. 7). However, the Irvine North locality in Estill County, Kentucky, contains the typical Waco in its entirety (Fig. 11). Overlying the Teichichnus Bed at this locality is a 5 m interval of green-gray shale, the lower half of which contains richly fossiliferous orange dolostone interbeds. The upper shale is comparatively barren, with several light gray siltstone beds that are hummocky cross-stratified and heavily bioturbated with numerous discrete Planolites and Chondrites. The base of the Estill Shale is recognized at this locality by a band of bright red mudstone and abundant granular glauconite pellets (Fig. 11). These features have been recognized widely in central Kentucky and are used to identify the Waco-Estill contact (cf. Rexroad et al., 1965; McDowell, 1983).

In Ohio, fossiliferous Waco shales could not be identified, but the lower light colored, calcareous, coral-bearing horizon (white Waco sensuSullivan et al., 2014a) persists in outcrops as far north as Highland County (Fig. 1). A succession of unfossiliferous, dark orange, heavily burrowed dolomitic carbonates (the orange Waco sensuSullivan et al., 2014a) sharply overlies this (Fig. 7). At these northern localities, the Estill contains numerous bands of green, red, and maroon mudstone, interbedded with orange calcareous siltstone beds (Figs. 9 and 12).

In Ohio, through northern Adams and Clinton Counties, the Waco Member of the Alger Shale Formation and the ferruginous upper massive member of the Brassfield appear to be mutually exclusive in individual outcrops. At localities where the upper massive is present, the Waco is absent, save perhaps for a 10 cm ferruginous bed of the orange Waco (Figs. 5 and 12). Conversely, at localities where the entire Waco is present, the upper massive member of the Brassfield is absent. At some of these sections, large ferruginous rip-up clasts derived from this bed can be found in the lower white Waco (Fig. 8). North of Clinton County, Ohio, the sub-Waco unconformity caps the red Brassfield (Fig. 12; McLaughlin et al., 2008b).

The transition between Waco-bearing and Dayton-bearing successions can be observed in a procession of cores and outcrops found through Highland, Fayette, Franklin, and Greene Counties, Ohio (Fig. 12). Along this transect the lower Estill becomes increasingly calcareous, a phenomenon that was also documented in McLaughlin et al. (2012). This stratigraphic change is documented by the Pleasant and Paint Township cores, where phosphatic calcareous strata, transitional between Estill and Dayton facies, overlie the Waco. North and west of here the Dayton Formation (sensu Brett et al., 2012) directly overlies the red Brassfield (Fig. 12).

δ13Ccarb Chemostratigraphy

New δ13Ccarb data are plotted against stratigraphic height in Figures 9, 11, and 12 (supplemented with data from Cramer, 2009; McLaughlin et al., 2012). The most extensive section sampled below the Waco is at Drowning Creek West in Madison County, Kentucky (Fig. 9). Here, the carbon isotope values were fairly low, with a mean of –0.4‰. The highest values were recorded in the lower Brassfield, the upper Oldham, and the Waco. The upper massive Brassfield yields much lower values between 0.0‰ and –1.0‰. The lower carbonate bed of the Waco yields δ13Ccarb values that are between 0.5‰ and 1.0‰ (Fig. 9).

At Irvine North in Estill County, Kentucky, values are ∼1.0‰–2.0‰ lower, and the range of values is much higher (values range from –8.2‰ to –0.5‰, with a mean of –3.8‰). Nevertheless, the data show distinct trends and a low degree of scatter (Fig. 11). A negative shift in carbon isotope values coincides with a shale horizon between the lower coral-bearing zone and the Teichichnus bed. A similar pattern occurs at the upper contact of the basal Waco carbonate bed at all localities. A positive shift in values can be traced through the lower 1.5 m of fossiliferous Waco shales at Irvine North; values drop off again in the subsequent 1.5 m; the level at which they begin to decline coincides with the transition from orange fossiliferous dolomitic beds to bioturbated gray siltstones (Fig. 10). Farther north in Ohio, the localities in Highland County recorded δ13Ccarb values that are, on average, much higher than the Irvine North section.

The upper massive Brassfield was sampled at the Martin Marietta quarry and Concord Township core in Adams County, Ohio, as well as in the Melvin Township core in Clinton County, Ohio (Fig. 12). It consistently yields values ranging from –1.0‰ to 0.5‰, significantly lower than those found in overlying strata of the white Waco. In contrast, the red Brassfield yields high values, ranging from 1.0‰ to 2.0‰ (Fig. 12).

The orange Waco generally yields the highest δ13Ccarb values observed in a given section; these were typically 0.5‰–1.0‰ greater than the values recorded in the underlying white Waco (Fig. 11). The Estill, Osgood, and Dayton yield δ13Ccarb values that are lower than those recorded in the orange Waco, yet higher than those recorded in the white Waco. Generally, δ13Ccarb curves through Estill, Dayton, and lower Osgood are relatively flat, with a few exceptions. At the Irvine North locality, δ13Ccarb values generated from bands of red shale were negatively offset from surrounding values (Fig. 11). This is also the case in the Melvin core, where there is a ∼3.0‰ negative excursion coinciding with a band of red shale in the lower Waco. The Dayton Formation yields lower values that may steadily rise upward into overlying shales of the Osgood (Fig. 11).

DISCUSSION

Correlation

The progressive truncation of units between the Waco Member of the Alger Shale Formation and upper massive member of the Brassfield is documented as far north as the Martin Marietta quarry and Melvin core, where the sub-Waco unconformity may cap the Brassfield (Fig. 9). The precise relationships between the ferruginous upper massive and the red Brassfield are difficult to determine. Although Rexroad (Rexroad et al., 1965; C.B. Rexroad, personal commun., 1983) reported diagnostic faunal elements of the upper massive (Floricolumnus [col.] sp.) and the Oldham (Ehlersella and Icriodina stenolophata) in the red Brassfield, none of these observations could be independently corroborated in our own field studies. The crinoids may be an unrelated large discoidal columnal, which is found at Cemex quarry in Greene County, Ohio (Thomka, this study). Furthermore, the red Brassfield and upper massive ironstone yield different δ13Ccarb signatures (Fig. 12). Slightly elevated values in both the Oldham and red Brassfield are consistent with arguments for their lateral equivalence (Fig. 13; McLaughlin et al., 2008b).

The similarity of δ13Ccarb isotope profiles generated for the Waco at most localities confirms its isochroneity. A confident correlation can be made between the basal Waco carbonate bed in Kentucky and the lower white Waco of southern Ohio through conventional physical stratigraphic procedures. Although results from Kentucky have different ranges of absolute values than sections in Ohio, this may be due to local phenomena, such as the influence of isotopically light terrestrial or meteoric carbon (Hudson, 1977; Cowan et al., 2005; Algeo et al., 1992). The section at Irvine North in particular (Fig. 11) is characterized by exceptionally low values. However, systematic positive and negative shifts, with relatively few outliers, may indicate that the primary structure of the curve remains intact.

The fossiliferous lower interval of the shaly upper Waco in Kentucky likely correlates to the orange Waco of south Ohio. A positive arc of δ13Ccarb values coincident with the lower fossiliferous Waco shale (Fig. 11) could be an expression of the excursion that is well documented within the orange Waco (Fig. 12; McLaughlin et al., 2012). This zone of elevated δ13Ccarb represents the Valgu excursion (Fig. 13; McLaughlin et al., 2008b, 2012). Flat-lying values are characteristic of the lower Estill where data range from <0.2‰ of the mean (Fig. 12); this is always lower than peak values recorded in the underlying orange Waco. The same sequence of patterns (flat-lying values over a positive excursion) is found above the Waco in the Pleasant Township core of south-central Ohio, where argillaceous, phosphatic carbonates overlying the Waco show characteristics that are transitional between the Dayton and Estill units (Fig. 12).

At least two distinct dolomitic beds with different δ13Ccarb signatures can be identified in the Paint Township core from Ohio: a lower (∼1 m) bed with moderate values (2.0‰) assigned to the Waco, and an upper (∼2 m) light colored interval with low δ13Ccarb values (1.3‰) at its base that rise steadily upward to 2.4‰, which represents the Dayton. These are separated by an ∼5-cm-thick glauconitic bed, similar to that found in basal Estill strata. This rising trend of δ13Ccarb values is therefore interpreted as the rising limb preceding the Ireviken excursion (Munnecke et al., 2003; Cramer, 2009; McLaughlin et al., 2012). This same pattern has been documented in the Cedarville core, which bears strata unambiguously assigned to the Dayton Formation (Cramer, 2009).

Sequence Stratigraphic Framework

Three regionally consistent unconformity-bound packages of strata can be identified in the Aeronian–Telychian interval of the Cincinnati Arch (Fig. 14). These were recognized previously, but categorized as two third-order stratigraphic sequences in Brett and Ray (2005), and have been slightly modified in light of the new data presented here. The lower sequence is bound by the base of the upper massive member of the Brassfield and at its upper contact by the sub-Waco unconformity. This was recognized as S-II in Brett et al. (1990; also see Brett and Ray, 2005). Given the lateral persistence of the unit, its assemblage of abraded and reworked fossils, rip-up clasts, and the high concentration of authigenic minerals, we suggest that it is a time-averaged basal transgressive deposit (Fig. 14). An ensuing episode of a highstand is recorded by the Plum Creek Shale, which reflects increased siliciclastic input as sedimentation rates began to outpace the rate of sea-level rise. The overlying Oldham, dominated by ferruginous, dolomitic packstones, is interpreted as a fourth-order transgressive episode overlain by its highstand counterpart, the Lulbegrud Shale Member. Units of this sequence are progressively truncated to the northwest by the sub-Waco unconformity (Fig. 14).

The distribution of the upper massive member of the Brassfield, the Waco Member of the Alger Shale Formation, and the Dayton Formation can be used to understand the topography of the sub-Waco unconformity surface. Concentrations of authigenic minerals (e.g., carbonate, glauconite, and pyrite) in the basal beds of the Waco, coupled with the abundance of colonial frame-building organisms, may indicate sediment starvation during transgression, which was covered by highstand shales of the upper Waco (Brett et al., 1998; McLaughlin et al., 2008b). The clastic mudstone and argillaceous carbonates of the Estill may represent a subsequent third-order highstand, likely enhanced by the influx of sediment with the beginning of the Salinic disturbance (Goodman and Brett, 1994). The transgressive systems tract for this highstand would not be the Waco, but rather the highly glauconitic shales of the basal Estill.

The Waco was deposited over an irregular topography along the proto–Cincinnati Arch. In some areas, deposition was restricted to local topographic lows (such as southwest wall of the Martin Marietta quarry and the Paint Township core) or it may have onlapped against minor topographic highs and mounds (northeast side of Martin Marietta quarry and Melvin Township core). This is also consistent with observations from eastern Indiana, where localized, lower Telychian, glauconitic dolostones referred to as the Lee Creek Member of the Brassfield Formation represent outliers of Waco deposition (Fig. 13; Kleffner et al., 2012; Brett et al., 2012).

The Waco is assigned to the Pt. eopennatus Zone and is thus older than any strata assigned to sequence S-IV in the type region of New York State (Brett et al., 1990, 1998; Loydell et al., 2007; McLaughlin et al., 2012; Sullivan et al., 2014b). Therefore, the regionally angular sub-Waco unconformity more likely represents that base of the slightly older S-III (Gillette, 1947; Brett et al., 1990). This is supported by the findings of Hunter (1960, 1970), who correlated the Waco Member with the Wolcott Limestone, a pentamerid-rich limestone found in east-central New York. However, this was placed within the sequence S-II by Brett et al. (1990, 1998) as a distinct upper fourth-order sequence. Although very few biostratigraphic data are available from the Wolcott Limestone, conodonts from the Wolcott Furnace Hematite, which caps the limestone, suggest assignment to the Pt. eopennatus Zone (Sullivan et al., 2014b).

Current data favor correlation of the Wolcott Limestone and Wolcott Furnace Hematite to the Waco Member of the Alger Shale Formation, and suggest that they are basal transgressive units of sequence S-III (as argued by Sullivan et al., 2012; Ettensohn et al., 2013). However, this is also an imperfect solution. The extensive paleontological data compiled by Gillette (1947) suggest that the Wolcott Limestone shares more genetic and faunal affinities with underlying strata than with overlying units. Furthermore, the Wolcott Limestone has a gradational lower contact with the underlying Sodus Shale, both of which are progressively truncated to the west by a regionally angular unconformity at the base of S-IV (Gillette, 1947; Brett et al., 1990, 1998).

The base of the Dayton Formation is a younger surface that has merged with the sub-Waco unconformity. This contact is cryptic in the southeast, but it may be coextensive with the glauconite granule zone that has long been used to delineate the base of the Estill Shale Member in Kentucky (Fig. 10; Rexroad et al., 1965) and may correlate to the basal S-IV boundary in New York.

The unconformity surfaces and facies traced throughout the study area highlight the complex interplay between sea-level change, paleoceanographic events, and clastic deposition associated with the distant effects of Salinic orogeny. By establishing the interval of recorded time and the duration of stratigraphic gaps, the results presented here may provide a firmer foundation for understanding early Silurian bioevents and the aftermath of the Late Ordovician extinction in the Appalachian foreland basin and Cincinnati Arch.

CONCLUSIONS

Sequence S-II is bound at its base by the ferruginous upper massive member of the Brassfield of east-central Kentucky and southern Ohio; δ13Ccarb values recorded in this unit are significantly different from those of the red Brassfield, which occurs in western Ohio, casting doubt on a possible correlation of these units. New isotope results from the Oldham may corroborate arguments for its equivalence with the red Brassfield.

A positive δ13Ccarb excursion associated with the Valgu Event was identified in the upper Waco Member of the Alger Shale Formation. The slightly younger Dayton Formation is lithologically and isotopically distinct; it overlies the Waco in cores from Franklin and Fayette Counties, Ohio. Sequence S-III is bound at its base by the Waco Member, which overlies a regionally angular unconformity that progressively truncates S-II to the northwest. The overlying S-IV is bound at its base by the Estill Shale Member and Dayton Formation, which onlap and partially truncate older sequences.

Detailed criticism from Alyssa Bancroft and Wojciech Kozłowski greatly strengthened the final report. We thank Arnold I. Miller and David L. Meyer, who provided helpful feedback on an earlier draft of this manuscript. Financial support for this study was provided in part by a grant from the U.S. Geological Survey STATEMAP project, a Graduate Student Research Grant from the Geological Society of America, a Graduate Student Assistance Grant from the SEPM (Society for Sedimentary Geology), and the Department of Geology at the University of Cincinnati (Caster Fund and Sed Fund). We also thank the Ohio Geological Survey for providing drill core for analysis and sampling. The research presented here is a component of the master’s thesis of Sullivan, completed at the University of Cincinnati. This paper is a contribution to the International Geoscience Programme (IGCP) 591, The Early to Middle Paleozoic Revolution.