Sequential Development of Platform to Off-platform ...

15

Brezinski, David K., John F. Taylor, and John E. Repetski, 2012, Sequential development of platform to off-platform facies of the great American carbonate bank in the central Appalachians, in J. R. Derby, R. D. Fritz, S. A. Longacre, W. A. Morgan, and C. A. Sternbach, eds., The great American carbonate bank: The geology and economic resources of the Cambrian – Ordovician Sauk megasequence of Laurentia: AAPG Memoir 98, p. 383 – 420.

Sequential Development of Platform to Off-platform Facies of the Great American Carbonate Bank in the Central Appalachians David K. Brezinski Maryland Geological Survey, Baltimore, Maryland, U.S.A.

John F. Taylor Geoscience Department, Indiana University of Pennsylvania, Indiana, Pennsylvania, U.S.A.

John E. Repetski U.S. Geological Survey, Reston, Virginia, U.S.A.

ABSTRACT In the central Appalachians, carbonate deposition of the great American carbonate bank began during the Early Cambrian with the creation of initial ramp facies of the Vintage Formation and lower members of the Tomstown Formation. Vertical stacking of bioturbated subtidal ramp deposits (Bolivar Heights Member) and dolomitized microbial boundstone (Fort Duncan Member) preceded the initiation of platform sedimentation and creation of a sand shoal facies (Benevola Member) that was followed by the development of peritidal cyclicity (Dargan Member). Initiation of peritidal deposition coincided with the development of a rimmed platform that would persist throughout much of the Cambrian and Early Ordovician. At the end of deposition of the Waynesboro Formation, the platform became subaerially exposed because of the Hawke Bay regression, bringing the Sauk I supersequence to an end. In the Conestoga Valley of eastern Pennsylvania, Early Cambrian ramp deposition was succeeded by deposition of platform-margin and periplatform facies of the Kinzers Formation. The basal Sauk II transgression during the early Middle Cambrian submerged the platform and reinitiated the peritidal cyclicity that had characterized the pre-Hawke Bay deposition. This thick stack of meter-scale cycles is preserved as the Pleasant Hill and Warrior Formations of the Nittany arch, the Elbrook Formation of the Great Valley, and the Zooks Corner Formation of the Conestoga Valley. Deposition of peritidal cycles was interrupted during deposition of the Glossopleura and Bathyriscus-Elrathina Biozones by third-order deepening episodes that submerged the platform with subtidal facies. Regressive facies of the Sauk

Copyright n2012 by The American Association of Petroleum Geologists. DOI:10.1306/13331500M983500

383

384

Brezinski et al.

II supersequence produced platform-wide restrictions and the deposition of the lower sandy member of the Gatesburg Formation, the Big Spring Station Member of the Conococheague Formation, and the Snitz Creek Formation. Resubmergence of the platform was initiated during the late Steptoean (Elvinia Zone) with the expansion of extensive subtidal thrombolitic boundstone facies. Vertical stacking of no fewer than four of these thrombolite-dominated intervals records third-order deepening episodes separated by intervening shallowing episodes that produced peritidal ribbony and laminated mudcracked dolostone. The maximum deepening of the Sauk III transgression produced the Stonehenge Formation in two separate and distinct third-order submergences. Circulation restriction during the Sauk III regression produced a thick stack of meter-scale cycles of the Rockdale Run Formation (northern Virginia to southern Pennsylvania), the upper Nittany Dolomite, the Epler Formation, and the lower Bellefonte Dolomite of the Nittany arch (central Pennsylvania). This regressive phase was interrupted by a third-order deepening event that produced the oolitic member of the lower Rockdale Run and the Woodsboro Member of the Grove Formation in the Frederick Valley. Restricted circulation continued into the Whiterockian, with deposition of the upper Rockdale Run and the Pinesburg Station Dolomite in the Great Valley and the middle and upper parts of the Bellefonte Dolomite in the Nittany Arch region. This deposition was continuous from the Ibexian into the Whiterockian; the succession lacks significant unconformities and there are no missing biozones through this interval, the top of which marks the end of the Sauk megasequence. During deposition of the Tippecanoe megasequence, the peritidal shelf cycles were reestablished during deposition of the St. Paul Group. The vertical stacking of lithologies in the Row Park and New Market Limestones represents transgressive and regressive facies of a third-order deepening event. This submergence reached its maximum deepening within the lower Row Park Limestone and extended into the Nittany arch region with deposition of the equivalent Loysburg Formation. Shallow tidal-flat deposits were bordered to the south and east by deep-water ramp deposits of the Lincolnshire Formation. The St. Paul Group is succeeded upsection by ramp facies of the Chambersburg and the Edinburg Formations in the Great Valley, whereas shallow-shelf sedimentation continued in the Nittany arch area with the deposition of the Hatter Limestone and the Snyder and Linden Hall Formations. Carbonate deposition on the great American carbonate bank was brought to an end when it was buried beneath clastic flysch deposits of the Martinsburg Formation. Foundering of the bank was diachronous, as the flysch sediments prograded from east to west.

INTRODUCTION The outcrop belts that expose Cambrian and Ordovician strata in the central Appalachians (Figure 1) provide an excellent opportunity to examine the transition from platform to off-platform facies of the great American carbonate bank (GACB) of Ginsburg (1982). Platform facies are exposed in the Nittany arch region of central Pennsylvania and the Great Valley, which stretches from northern Virginia (Shenandoah Valley) northeastward to eastern Pennsylvania (Lebanon Valley) and adjacent New Jersey (Paulinskill Valley). Rapid subsidence in the Pennsylvania depocenter (Read, 1989a, b) resulted in the accumulation of more than 4000 m (>13,120 ft) of carbonate-dominated strata in these outcrop belts. Comparably thick wedges of periplatform strata that accumulated in continental slope and continental rise environments are preserved in the western parts

of the Conestoga Valley of Pennsylvania and the Frederick Valley of Maryland (Figure 1). In the easternmost exposures of these valleys, much of the sequence is condensed into comparatively thin packages of black shale and shaly limestone that were deposited in sedimentstarved basinal environments. The goal of this chapter is to summarize the results of research conducted during the last two decades on the history of sea level change and derivative sequence stratigraphy and cyclostratigraphy in the deposits of the GACB in the central Appalachian region (i.e., the Pennsylvania depocenter of Read, 1989a, b). We have continued to subdivide this immense stack of Cambrian and Ordovician platform and periplatform carbonates on an increasingly finer scale through an integrated approach best described as a biostratigraphically constrained event stratigraphy. Recently recovered trilobite and conodont faunas from all of the outcrop belts have

Sequential Development of Platform to Off-platform Facies

385

Figure 1. Outcrop belts (shaded) and specific areas of exposure (white lettering) of Cambrian and Ordovician strata of the great American carbonate bank in the central Appalachians. 50 km (31 mi).

expedited recognition and correlation of numerous third-order (grand cycle-scale) transgressive-regressive cycles across the entire platform. Some of these thirdorder events also affected the style of deposition in off-platform environments and/or in other shallowmarine depositional basins, facilitating correlation into coeval successions in other regions. In many poorly studied regions, however, the biostratigraphic control is sparse and the correlations are offered only as hypotheses to be tested as new information becomes available. The numerical orders of cycles and sequences used in this chapter are intended only as categories of convenience (see Schlager, 2004) based entirely on scale (thickness and implied duration), with no implication of a specific forcing mechanism or presumption of eustatic origin. Many of the third-order sequence boundaries are mappable because of the contrast in the underlying and overlying lithofacies, typical of flooding surfaces and unconformities. Consequently, many of these horizons have been used as formation or member boundaries to produce more highly refined lithostratigraphy both for surface mapping and for correlation in the subsurface. Many of the figures provided here are updated cross sections that show the distribution of lithofacies and correlation of the sequences that they define. Most depict the distribution of sequences de-

lineated in the Sauk megasequence, from which most of our new information were collected. A brief and more tentative treatment of the cycle and sequence stratigraphy of the Middle Ordovician (Tippecanoe megasequence) carbonates is provided at the end of the chapter to carry the history of the GACB through to completion with its destruction during Taconic orogenesis.

THE SAUK MEGASEQUENCE As in other areas of Laurentian North America, the carbonates deposited on and adjacent to the GACB in the central Appalachians constitute much of the Sauk and Tippecanoe sequences of Sloss (1963). In recent studies (Golonka and Kiessling, 2002; Miller et al., 2004), these large-scale first-order sequences or cycles are referred to as megasequences. Palmer (1981) demonstrated that the Sauk megasequence is divisible into three subsequences (now supersequences) that he termed Sauk I, Sauk II, and Sauk III, in ascending order (Figure 2). We interpret each of these Sauk supersequences as second-order transgressive-regressive cycles. Read (1989) further subdivided Sauk I and Sauk III into two parts, thereby delineating five transgressive-regressive cycles or sequences within the Sauk megasequence

386

Brezinski et al.

Figure 2. The correlation of groups and formations in the Sauk and Tippecanoe megasequences in the central Appalachians. Dol = Dolomite; Fm = Formation; Gp = Group.

in this region. The basal sequence (sequence 1) of Read (1989), termed the preplatform shelf, is composed of siliciclastic deposits that accumulated in rift-to-drift shelf and basinal environments on the southern margin of Laurentia before initiation of carbonate deposition during passive margin creation. Hence, the oldest carbonate strata at the base of sequence 2 (in the middle of the Sauk I supersequence of Palmer, 1981) record the birth of the GACB in the central Appalachians (Cecil et al., 2004).

Sauk I Ramp to Shelf Facies Transition Recent detailed studies of the Cambrian stratigraphy of the eastern Great Valley in Maryland (Brezinski, 1992) and the Conestoga Valley in southeastern Pennsylvania (Taylor and Durika, 1990; Taylor et al., 1997) sup-

port and clarify the major elements of the depositional history in the Early Cambrian reconstructed by Read (1989a) for this region. They bear out the initial development of a carbonate ramp, followed by evolution of the margin into a high-relief shelf with widespread shale deposition landward of a narrow rim of microbial reefs and ooid shoals. However, new biostratigraphic data, in conjunction with a refined member-level lithostratigraphy, reveal inaccuracies in some previous interpretations and miscorrelation of some units within Pennsylvania and Maryland. Detailed mapping facilitated by subdivision of the Tomstown Formation into four members in the Great Valley (Brezinski, 1992) resulted in the discovery of complex structures that had been overlooked in previous studies (Figure 3). The lowest Tomstown strata are made up of finely laminated marble within the basal Bolivar Heights Member

Sequential Development of Platform to Off-platform Facies

387

Figure 3. Lateral variations of lithologies and facies and members of the Tomstown Formation along depositional strike in the eastern Great Valley from Virginia to southern Pennsylvania, with derivative relative sea level curve. 10 m (33 ft); 1 km (0.6 mi).

(Figure 4A). This marble is traceable along the western margin of the Blue Ridge from Pennsylvania to Virginia. This unit, named the Keedysville marble bed, is a mylonite that marks a detachment zone along which the entire Cambrian and Ordovician carbonate stack appears to have been decoupled from the underlying Chilhowee Group clastics during later Paleozoic orogenesis (Brezinski, 1992; Brezinski et al., 1996; Campbell and Anderson, 1996). Jonas and Stose (1930, 1944) described what appears to be an analogous tectonite within the correlative lower Vintage Formation of the Conestoga Valley. Some formation contacts in or near this detachment zone that had been interpreted as depositional in nature (Reinhardt and Wall, 1975; Read, 1989a) are now known to have resulted from juxtaposition of units by faulting. This is significant because cyclic peritidal carbonates reported from the base of the Tomstown Formation by Reinhardt and Wall (1975) led Read (1989) to conclude that shallow peritidal deposition was initiated very early in the deposition of sequence 2 in Maryland, in contrast to more protracted deposition of deep subtidal ramp facies at the base of this sequence in Virginia. The discovery that the cyclic facies directly overlying the Chilhowee clastics in Mary-

land represents the highest member (Dargan Member) of the Tomstown Formation, emplaced along a fault nappe (Brezinski, 1992), eliminates the evidence for a contrast in depositional conditions between Maryland and Virginia, as well as the interpreted facies replacement of nearshore clastics (Antietam Formation) by peritidal cycles without an intervening phase of subtidal ramp deposition in Maryland. In fact, where nondeformed, the Bolivar Heights and Fort Duncan Members of the Tomstown Formation in Maryland consist mostly of burrow-mottled, noncyclic, subtidal ramp carbonates similar to those in the lower part of the Sauk I supersequence (sequence 2 of Read, 1989) in Virginia (Patterson Creek and Austinville Members of the Shady Dolomite) and Pennsylvania (Vintage Formation). The mottled fabric of the Bolivar Heights Member is unquestionably the product of bioturbation (Figure 4B); the origin of fabrics in the dolomite of the overlying Fort Duncan Member is less certain. At least some parts of the Fort Duncan Member display relict fabrics that resemble stromatactoids and fenestrate thrombolitic boundstone (Figure 4C). This suggests that the Fort Duncan Member records the early stages of development of the microbial reefs that eventually created a narrow carbonate rim at the seaward margin of the

388

Brezinski et al.

Figure 4. The Sauk I ramp and platform lithologies. (A) The laminated tectonite Keedysville marble bed at the base of the Tomstown Formation. (B) The typical burrow-mottled limestone of the of the Bolivar Heights Member ramp facies of the Tomstown Formation. (C) The clotted microbial fabric of the Fort Duncan Member of the Tomstown Formation. (D) The massive, fractured, light-gray dolomitized sand shoal facies of the Benevola Member of the Tomstown Formation. (E) The meter-scale limestone-dolomite cycles (arrows) within the Dargan Member of the Tomstown Formation. (F) The shaly limestonedolomite cycles of the Red Run Member of the Waynesboro Formation. (G) The massive dolomitized mud biostrome of the upper Cavetown Member of the Waynesboro Formation. (H) The transition from subtidal limestone to peritidal cycles at the contact between the Cavetown (Cwak) and Chewsville (Cwac) Members of the Waynesboro Formation.

shale-dominated shelf during deposition of the uppermost part of Sauk I. The light-colored dolomite of the overlying Benevola Member (Figure 4D) strongly resembles the Ledger Formation in southeastern Pennsylvania and the Austinville Member of the Shady Dolomite of Virginia, which accumulated in ooid sand shoals at the very edge of the shelf (Figure 3). Both the Benevola Member and lower Ledger display some relict cross-stratification and are quarried extensively because of their exceptional purity. As previously noted, the overlying Dargan Member is characterized by well-

developed meter-scale peritidal cycles (Figure 4E) similar to those that are ubiquitous in the Middle and Upper Cambrian units in this region. It is likely that the cycles of the Dargan Member formed through periodic shoreward progradation of broad outer-shelf banks similar to those that produced its slightly younger counterparts (Figure 3). A slightly different style of deposition prevailed on the outer shelf in the latest Early Cambrian when the pure carbonate belt shrank to form a narrow rim only 10 to 15 km (6.2 –9.3 mi) wide during deposition of the

Sequential Development of Platform to Off-platform Facies

389

shale-rich Waynesboro Formation (Read, 1989a). The Waynesboro Formation is a succession of siliciclasticrich carbonate strata that was subdivided into three members in the eastern Great Valley by Brezinski (1992): the Red Run, Cavetown, and Chewsville Members, in ascending order (Figure 5). The basal and upper members (Red Run and Chewsville) are dominated by peritidal cycles that consist of carbonates interbedded with red and green siliciclastics. A typical mixed clasticcarbonate meter-scale cycle comprises reddish to lightgray Skolithos-bearing sandstone that grades upward into tan laminated dolomite and red-brown mudcracked siltstone and shale. The intervening Cavetown Member lacks the clastic components of the other members. This middle member contains massive dolomitic lime mudstone, in packages as much as 20 m (66 ft) thick at the base and near the top of the unit. The remainder of the member consists of peritidal cycles in which gray, thick-bedded, burrow-mottled dolomitic limestones grade upward into tan, laminated, mudcracked dolomites (Figure 5). Within the Great Valley, Lower Cambrian strata of the Waynesboro Formation (Bonnia-Olenellus Biozone) are directly overlain by middle Middle Cambrian (Glossopleura Zone) strata of the Elbrook Formation (Brezinski, 1996a). The absence of the Poliella and Albertella Zones is attributable to the Hawke Bay event (Palmer and James, 1979), a major craton-wide regressive episode that resulted in a sizable unconformity between Sauk I and Sauk II in all but the most rapidly subsiding depocenters (Palmer, 1981) (Figure 5). Within the Nittany arch of central Pennsylvania, only thin slices of the Lower Cambrian section are exposed in the hanging walls of several large thrust faults. The red siltstones and shales resembling the Chewsville Member of the Waynesboro Formation exposed in that area are the only Lower Cambrian strata exposed west of the Great Valley in the central Appalachian transect. Figure 5. The stratigraphic column of the Waynesboro

Sauk I Platform-margin and Off-platform Facies Although some uncertainty remains regarding a shelfbreak origin for the members of the Tomstown Formation, the formations in the upper part of Sauk I and at the base of Sauk II in the Conestoga Valley undoubtedly represent shelf-marginal and off-platform environments (Rodgers, 1968; Gohn, 1976; Reinhardt, 1977; Taylor and Durika, 1990; Taylor et al., 1996; De Wet et al., 2004). Figure 6 shows the lithostratigraphic and biostratigraphic units recognized in the Lower and Middle Cambrian of the eastern Great Valley and Conestoga Valley, along with the biozones established for this interval. Figure 7 illustrates the lateral facies relationships

Formation with relative sea level curve based on stacking of lithologic components. The amplitude and frequency of fourth-order and smaller cycles were approximately portrayed. 50 m (164 ft).

between the carbonate-rich units of the eastern Great Valley and the periplatform and off-platform facies of the Conestoga Valley. The Vintage Formation, which directly overlies the Chilhowee clastics in the Conestoga Valley, consists of burrow-mottled carbonates similar to those within the Bolivar Heights and Fort Duncan Members of the Tomstown Formation (Figure 8A). Like the lower members

390

Brezinski et al.

Figure 6. The revised lithostratigraphy of carbonate units of the Sauk I and Sauk II in the Conestoga Valley and Great Valley and relationship to trilobite biozonation.

of the Tomstown Formation, the Vintage Formation apparently formed through deposition on a subtidal ramp. The overlying Kinzers Formation, which has no counterpart in the Great Valley succession, has been interpreted as a wedge of periplatform sediment that accumulated in slope-to-rise environments (Gohn, 1976; Taylor and Durika, 1990), seaward of the shelf-margin ooid shoals on which the pure dolomite of the overlying Ledger Formation was deposited. Marked changes in lithofacies and profound thickening of the Kinzers Formation from the eastern to western Conestoga Valley indicate that the platform had evolved into a high-relief constructional rimmed shelf by the time of Kinzers deposition. In the eastern Conestoga Valley, the entire Kinzers is approximately 70 m (~230 ft) thick and comprises a lower member dominated by darkgray shale (Emigsville Member) (Figure 8B), an unnamed middle member consisting of bluish gray mottled limestone with shaly interbeds (Jonas and Stose, 1930; Campbell, 1969) (Figure 8C), and a shale-rich upper member (Longs Park member). This relatively thin eastern Kinzers accumulated in a sediment-starved basinal setting. In the western Conestoga Valley, the Emigsville Member triples in thickness to approximately 60 m (~197 ft) and the overlying carbonate part of the Kinzers, referred to here as the upper member, attains a thickness of more than 500 m (>1640 ft) (Ganis and Hopkins, 1990). A deep-water periplatform origin for the greatly thickened upper Kinzers carbonates is reflected in the

local occurrence of limestone cobble to boulder conglomerates, which are interpreted as proximal debrisflow or bypass-channel deposits (Figure 8D). In addition, the upper member contains several thin (10–15 m [33–49 ft]) intervals of dark impure limestone with cosmopolitan trilobite taxa found elsewhere only in the toe-of-slope limestone conglomerates in the northern Appalachians (Taylor and Durika, 1990; Taylor et al., 1997). In previous studies (Jonas and Stose, 1930; Ganis and Hopkins, 1990; Taylor and Durika, 1990), attempts to correlate various Kinzers sections throughout the Conestoga Valley via key bed stratigraphy suffered from the false assumption that only a single interval of dark impure carbonate (Upper Kinzers Sandstone of Jonas and Stose, 1930, i.e., Greenmount Member of Ganis and Hopkins, 1990, and Taylor and Durika, 1990) exists at the top of the formation (Figure 8F). Subsequent discovery (Taylor et al., 1997) that several such intervals are present, each with a different Lower or Middle Cambrian fauna, required substantial revision of the previous correlation schemes. The current model, showing multiple tongues of deeper water carbonate in the upper member of the Kinzers, is provided as Figure 7. So far, none of the faunas have been recovered from more than one locality, so more such intervals might await discovery. None of the tongues identified so far in the upper member of the western Conestoga Valley has yielded the Middle Cambrian fauna of the Longs Park member in the condensed eastern Kinzers to tie the sections together across the valley.

Sequential Development of Platform to Off-platform Facies

391

Figure 7. The stratigraphic cross section showing distribution of lithofacies in Sauk I and II in the eastern Great Valley (column A) and the western (column B) and eastern (column C) Conestoga Valley of Pennsylvania. 50 m (164 ft); 40 km (25 mi); m–C = Middle Cambrian; L–C = Lower Cambrian.

Only the Emigsville Member at the base of the Kinzers can be confidently correlated across the Conestoga Valley, where it plays a significant role as a major aquiclude in the hydrogeology of the area (Meisler and Becher, 1971). As a dominantly siliciclastic unit, it also represents a significant departure from the primarily carbonate deposition that prevailed in this area from the end of Chilhowee deposition in the Early Cambrian well into the Late Ordovician. The unit remains enigmatic in many respects. Despite its exceptionally preserved Lower Cambrian fauna in the eastern part of the valley (Dunbar, 1925; Jonas and Stose, 1930; Resser and Howell, 1938; Campbell and Kauffman, 1969; Conway Morris, 1985; Skinner, 2004), its age has not been constrained beyond some part of the very thick Bonnia-Olenellus Zone, which encompasses all the fossilbearing Lower Cambrian formations in the central Appalachians. Lacking any more precise age information, it is not even clear what horizon or interval in the relatively nearby eastern Great Valley succession (left column of Figure 7) formed at the time that carbonate deposition was suspended in the Conestoga Valley

to create the Emigsville Member. Figure 7 suggests equivalence with the middle of the Tomstown Formation (Benevola Member) and, hence, the initial development of the rimmed shelf. But this is not well constrained and any horizon or unit within Sauk I in the Great Valley could be correlated with the Emigsville Member without conflicting with the available biostratigraphic data. Nonetheless, what makes this correlation particularly inviting is a strikingly similar stratigraphic succession documented in coeval Lower and Middle Cambrian platform-margin deposits in southern Virginia (Barnaby and Read, 1990; Betzner and Read, 2009). In that succession, a dark quartzose shaly unit known as the Taylor marker separates the ramp carbonates of the underlying Patterson Member of the Shady Dolomite from the overlying periplatform deposits of the upper Shady Dolomite, which formed in front of a high-relief rimmed shelf in the latest Early and earliest Middle Cambrian. Although more rigorous evaluation of the posited equivalence of the Emigsville Member and the Taylor marker (and many other plausible

392

Brezinski et al.

Figure 8. The lithologies within the Sauk I platform edge and periplatform deposits in the Conestoga Valley. (A) The Keedysville marblelike tectonite at the base of the Vintage Formation. (B) The burrow-mottled fabric characteristic of the Vintage Formation ramp deposits. (C) The dolomitized void fillings in microbrialite(?) Vintagetype section. (D) The typical phyllitic character of the Emigsville Member of the Kinzers Formation. (E) Thin intervals of off-shelf black limestone (bottom black facies [BB]), interfingering with purer periplatform limestone at Thomasville, York County, Pennsylvania. (F) The debris-flow cobble conglomerate or breccia in periplatform carbonates of the upper member of the Kinzers (from location of arrow in photograph E). (G) The burrow-mottled deepramp limestone of Kinzers Formation (upper member) at type section. (H) The thinly bedded silty dolomites of the upper Kinzers Formation, Thomasville, York County, Pennsylvania.

correlations within Sauk I) awaits subdivision of the Bonnia-Olenellus Zone, the virtually identical lithofacies succession within the uppermost part of that zone in the two areas is fairly compelling in itself. Interpretations of the depositional environment for the Emigsville Member have also varied considerably. In most studies, a base-of-slope or basinal setting was envisioned, invoked by the significant reduction in carbonate and, perhaps, by association with other occurrences of exceptional Burgess Shale-type preservation in deep-marine facies. Recently, however, evidence for an alternate interpretation of the Emigsville Member as

a deep-shelf to upper-slope facies, within reach of storm-wave base, has been proposed (Skinner, 2004). The interpretation of the probably coeval Taylor marker as the product of a sea level fall, followed by a shortlived drowning event and backstepping of the margin with inception of the rimmed shelf (Barnaby and Read, 1990), suggests that the Emigsville Member might bear the imprint of both shallow- and deep-water conditions. In contrast to the break at the Tomstown-Waynesboro contact in the platform carbonates of the Great Valley, no stratigraphic gap attributable to the Hawke Bay event has yet been documented in the western Conestoga

Sequential Development of Platform to Off-platform Facies

Valley. In fact, the discovery in that thick periplatform succession of the Poliella trilobite Zone, an interval generally not present because of Hawke Bay erosion, suggests that little or no interruption in sedimentation occurred during deposition of the uppermost Kinzers in the western Conestoga Valley (Taylor et al., 1997). The biostratigraphic data published for the eastern Conestoga Valley succession (Campbell 1969, 1971) place the Sauk I-Sauk II boundary somewhere within the greatly condensed Longs Park member at the top of the Kinzers. Barnaby and Read (1990) reported a similar situation in the Shady Dolomite in Virginia. The Kinzers can be interpreted as a western autocthonous facies of the mostly resedimented Conestoga Formation. The latter unit consists of limestone-shale rhythmites and limestone conglomerates. These lithologies formed in off-platform environments seaward of the peritidal bank that rimmed the shelf in the Middle to Late Cambrian during deposition of Sauk II and Sauk III. The conglomerates that characterize the Conestoga Formation are interpreted as proximal to distal debrisflow deposits that formed in a toe-of-slope setting and became interstratified with deep-water hemipelagic rhythmites (Figure 8G). In the Frederick Valley of Maryland, off-shelf facies assignable to Sauk I are found in the Araby and the lowest Frederick Formations. The Araby Formation (Reinhardt, 1974) crops out along the eastern border of the Frederick Valley and is a sequence of fine-grained bioturbated siltstone and fine-grained sandstone that Reinhardt (1974) correlated with the upper Chilhowee Group of the Great Valley. The Araby Formation is conformably overlain by the basal member of the Frederick, the Monocacy Member, which consists of a succession of approximately 70 m (~230 ft) of black to dark-gray shale interbedded with rhythmic and brecciated limestones and shale (Brezinski, 2004). A Lower Cambrian fauna that includes Olenellus and the enigmatic conical fossil Salterella has been recovered from the lower part of the Monocacy Member (Reinhardt, 1974; Brezinski, 2004). This fauna indicates equivalence to some parts of the Antietam, Tomstown, and/or Waynesboro Formations of the Great Valley. However, a much younger Middle Cambrian (upper Sauk II) fauna occurs in the highest beds of the Monocacy Member. Consequently, Brezinski (2004) postulated that the Araby Formation and Monocacy Member of the Frederick Formation represent sediment-starved basinal facies deposited in a deep-ocean floor setting. As in the eastern Conestoga Valley, the Sauk I-Sauk II boundary in the distal off-platform succession in the eastern part of the Frederick Formation lies somewhere within a greatly condensed interval of dark shale and basinal limestone.

393

Sauk II Platform Facies The reexpansion of the carbonate platform after the Hawke Bay regression resulted in deposition of hundreds of meters of peritidal carbonates on the central Appalachian platform through the Middle Cambrian, most of them packaged in meter-scale cycles. These carbonates constitute sequence 3 of Read (1989), which is equivalent to the Sauk II supersequence of Palmer (1981). The entire succession has been mapped as the Elbrook Formation in the Great Valley and as the Zooks Corner Formation in the Conestoga Valley (Meisler and Becher, 1971; Brezinski, 1996a). In the Nittany arch, Sauk II is divided into two units, the Pleasant Hill and Warrior Formations (Figure 2). Brezinski (1996a) divided the Elbrook Formation into three informal members. The lower member consists of 200 m (656 ft) of tan shaly dolomite, interbedded with thin (5 m [16 ft]), gray, bioturbated lime mudstone intervals. This member exhibits peritidal cycles that, in most cases, are totally dolomitic (Figure 9A). The middle member consists of as much as 70 m (230 ft) of mostly noncyclic bioturbated lime mudstone with only a few thin dolomitic laminites (Figure 9B). The upper member is 500 m (1640 ft) thick and consists entirely of meter-scale peritidal cycles. A typical cycle in the upper member consists of a thin (