Unconformity-bounded seismic reflection sequences define Grenville-age rift system and foreland basins beneath the Phanerozoic in Ohio

Currently, coarse geophysical data (seismic reflection and potential fields) and sparse drill holes are the main tools used to interpret the deep subsurface rocks of the Ohio region. We present a controversial reinterpretation of the Consortium for Continental Reflection Profiling (COCORP) OH-1 seismic reflection data, based on industry reprocessed output to 3 s traveltime. The basic research problem is a lack of drill holes deep into the Precambrian directly on the seismic line. We present this position paper as a request to collect refined, higher caliber geophysical data and verify them with deeper drill holes. A clear picture of the underpinning Precambrian of the midcontinent of North America will increase our understanding of how crustal stresses resulted in recent earthquake activity, future energy resources, and deep saline reservoirs for CO₂ sequestration. The need for exploration and deep drilling to test geologic models will provide answers to numerous questions and problems that should be solved or avoided.

The Precambrian of Ohio is covered by 750–4500 m of Phanerozoic sedimentary rocks and glacial deposits. Using sparse deep well control and regional gravity and magnetic data, Bass (1960) and numerous subsequent authors interpreted a Precambrian boundary in Ohio, referred to as the Grenville Front or Grenville tectonic zone, that separates Grenville Province metamorphic and/or igneous rocks on the east from the older Granite-Rhyolite Province on the west (Fig. 1). Previously scientists used regional trending magnetic anomalies in an attempt to extend the well-studied Canadian Grenville Province domains that crop out east of the Grenville Front tectonic boundary beneath Phanerozoic cover into Ohio (Hoehn, 1991; Lidiak and Hinze, 1993). It is not known if Archean and Mesoproterozoic rocks similar to northern Canadian counterparts are present in Ohio.

In the absence of drill core data in Ohio, we propose that use of the term Grenville front magnetic lineament be adopted until a more concise relationship can be established with the Canadian namesake. In this paper we show that the Ohio Grenville Province rocks are partly allochthonous to the Grenville front magnetic lineament in contrast to the Canadian parauthochthonous Grenville Front of Davidson (1986). The Grenville collision consists of three major episodes of crustal shortening, each punctuated by extensional events (Rivers, 1997, Fig. 10, p. 143). See Rivers (1997) and Carr et al. (2000) for more detailed discussion of the evolution of the Laurentian continent as it relates to the Grenville Province in Canada.

Although the presence of a Precambrian sedimentary basin in western Ohio has been suspected since the 1930s on the basis of well samples (well 2, Fig. 2) (Wasson, 1932), only gravity and magnetic data were available to support this interpretation (Rudman et al., 1965; Lidiak and Zietz, 1976; Keller et al., 1982; McPhee, 1983; Denison et al., 1984). The Fort Wayne Rift and East Continent Rift System are Proterozoic tectonic elements interpreted as sedimentary basins west of the Grenville front in Ohio, using drill hole (Fig. 2) and seismic reflection data (Fig. 1). Apart from gravity and magnetic data and disputable well drilling samples, the COCORP OH-1 seismic reflection profile was the first evidence to support the hypothesis that Precambrian sediment-filled rift basins are present in western Ohio. Pratt et al. (1989) depicted a faulted basin in western Ohio on a small-scale figure of COCORP OH-1. Sedimentary rocks of the Middle Run Formation were discovered during a routine deep stratigraphic test in Warren County, Ohio, and correlated to eastward-dipping layered seismic reflectors beneath Paleozoic rocks (well 1, Fig. 2) (Shrake et al., 1990; Shrake, 1991). This discovery confirmed the suspected presence of a Precambrian sedimentary basin in western Ohio, and supported the rift basin interpretation (Pratt et al., 1989; Culotta et al., 1990). Following the discovery of the Middle Run Formation, a reevaluation of well cuttings and sparse drill core from Precambrian tests in western Ohio and adjacent Indiana and Kentucky indicated that the East Continent Rift System consists of more than 6000 m of conglomeratic lithic arenites, shale, volcaniclastics, andesite, basalt, and tuffs (Drahovzal et al., 1992). Numerous seismic profiles throughout the United States midcontinent have shown layered Precambrian reflectors (Brown et al., 1982; Cannon et al., 1989; Pratt et al., 1989, 1992; Culotta et al., 1990; Shrake et al., 1990; Hoehn, 1991; Drahovzal et al., 1992; Milkereit et al., 1992; Hauser, 1993; Drahovzal, 1997; Potter et al., 1997; Stark, 1997; Dean et al., 1998; McBride et al., 2003).

Although U/Pb data are lacking for Ohio's Precambrian rocks, we assume that the Eastern Granite-Rhyolite Province rocks (Bass, 1960) are the oldest, and conclude that a major regional unconformity existed on the rocks at the eastern continental margin of the Laurentian continent prior to the Grenville orogeny. The unequivocal presence of rocks representing accretionary terrain in Ohio older than Grenville age (e.g., Elzevirian of Rivers, 1997) has yet to be confirmed. Therefore, it is generally assumed that midcontinent Precambrian rocks west of the Grenville front are an age similar to those rocks of the St. Francois Mountains of Missouri northeastward through Ohio to Ontario (Easton, 1986; Van Schmus et al., 1996; Culshaw and Dostal, 2002).

The Eastern Granite-Rhyolite Province surface in this paper has multiple contacts with overlying sequences, beginning with the Grenville metamorphic and/or igneous sequence and culminating with the base of the Paleozoic Sauk sequence. East of the Grenville front magnetic trend, it appears that Ohio Grenville metamorphic and/or igneous thrust sheets likely consist of westward-thrusted Laurentian continental-margin rocks. The westward-thrusted imbricate sheets in the subsurface of the Appalachian basin are well documented (Beardsley and Cable, 1983; Pratt et al., 1989; Culotta et al., 1990; Hoehn, 1991; Riley et al., 1993; Asgharzadeh, 2002; Paramo, 2002). Unfortunately, geochemical and age dating relationships do not exist to delimit and verify interpretations of Grenville or older domains in Ohio, similar to the Canadian Central Gneiss Belt or composite arc belts (Easton, 2000; Carr et al., 2000). From an Ohio perspective, this hypothesis has been based on regional magnetic and gravity maps, and very limited well data.

The U/Pb age dates for nonsedimentary rocks have not been determined for the Precambrian of Ohio. Calculated whole-rock Rb/Sr dates for Ohio range from ca. 0.9 to 1.3 Ga old, and were likely reset by later thermal or geochemical events (Lucius and von Frese, 1988). Diagenesis of the weathered Precambrian surface prior to, and authigenesis during, the Paleozoic deposition added to the complexity of accurately defining age relationships (Lidiak and Ceci, 1991). Consequently, Ohio Rb/Sr dates only indicate the presence of Precambrian-age rocks and do not show relationships to currently accepted provinces or boundaries.

Regional geochronological investigations beyond Ohio and one recent study of detrital zircons from within Ohio increase our understanding of the Precambrian of Ohio. Using regional analyses beyond Ohio, we assume that the Eastern Granite-Rhyolite Province is ca. 1.3–1.4 Ga old (Van Schmus et al., 1996). The Grenville Province (which is a complex of Archean and Mesoproterozoic domains in Canada) underwent tectonothermal changes between 1.0 and 1.2 Ga ago (Culshaw and Dostal, 2002). The Midcontinent Rift System (ca. 1.1 Ga old; Davis and Paces, 1990) and East Continent Rift System (Fort Wayne Rift is not dated) are located west of the Grenville front. Based on structural relationships from seismic data, Drahovzal et al. (1992) interpreted both rifts to have preceded the Grenville orogeny. These western Ohio basins have also been interpreted as partly or completely Grenville-age foreland basins (Hauser, 1993; Dean et al., 1998, 1999, 2000; Baranoski et al., 2001, 2002; Dean and Baranoski, 2002a, 2002b; Santos et al., 2002) and Cambrian basins (Wolfe et al., 1993). Santos et al. (2002) used sensitive high-resolution ion microprobe (SHRIMP) to date detrital zircon grains (lacking zenotine overgrowths) from the Middle Run Formation of the East Continent Rift System of Ohio (ca. 1 Ga old). Their work supports deposition of Middle Run sediments toward the end of the Grenville orogeny. Relative age relationships and structural and depositional history between the two rift systems and the Grenville Province will continue to be problematic until adequate drill cores are dated.

Despite ~200 wells drilled into Precambrian crystalline and sedimentary rocks of Ohio and ~805 km of seismic data available (Baranoski, 2002), there has been little advancement since the work of Bass (1960) to further refine and characterize the Eastern Granite-Rhyolite and Grenville Provinces, or the Grenville front boundary separating the provinces.

COCORP reflection data were originally acquired in Ohio in 1987 as part of deep crustal studies of the eastern midcontinent of North America (see Pratt et al., 1989). Two west-east lines totaling ~382 km were acquired with vibrators to obtain 120 channel data with geophone array spacing at 100 m, with 28 s sweep length and 12–50.5 sweep frequencies. This original data set was reprocessed to 3 s using standard industry techniques for hydrocarbon prospect evaluation by Lauren Geophysical of Denver, Colorado (Table 1). The shallow reflectors of this reprocessed data set are greatly enhanced in comparison to the earlier processing (see Pratt et al., 1989) for deep crustal features. A paper copy migration section of 3 s at 0.25 m/s was interpreted during the early stages of the work for this paper (Dean et al., 1998, 1999, 2000; Dean and Baranoski, 2002a, 2002b). A simplified interpretation from Dean and Baranoski (2002a, 2002b) was transferred to the exaggerated variable density display for this paper (Fig. 3). As a result, thrust faulting appears much steeper in Figure 3 herein than in Dean and Baranoski (2002a, 2002b). Color amplitude and contoured interval velocity (from root mean squared, RMS) displays from Dean and Baranoski (2002a, 2002b) were utilized to verify the characteristics of variable density display attributed to sequence boundaries of Paleozoic and Precambrian rocks. See Dean and Baranoski (2002a, 2002b) for a detailed view of the original interpretation utilized and expanded upon for this paper.

Within 3.2 km of the Ohio COCORP OH-1 line, 13 wells were drilled into the Precambrian (Table 2). Penetration into the Precambrian for these wells is typically <50 m. In the absence of sufficient drilling depths and/or core beneath the Cambrian Mount Simon Sandstone, penetration of Precambrian sedimentary rocks near the COCORP OH-1 line is at least partially inconclusive.

Adequate sampling of Precambrian rocks is generally problematic for a number of reasons. Most important, very few cores have been taken from the Precambrian in Ohio, especially in proximity to COCORP lines for this paper. The majority of deep test locations in Ohio were drilled on seismically defined monadnocks (Baranoski, 2002), thus biasing sampling of the Precambrian. The areal extent of the monadnocks is generally poorly defined, based on the drilling of one well. However, available seismic data across some of these features and multiple well control suggest that sizes range from <0.5 km to >10 km (Riley et al., 1993; Wicks, 1996; Baranoski, 2002; Dean and Baranoski, 2002a, 2002b). Some of the monadnocks are fault controlled.

Wells drilled into the Precambrian typically only penetrate a few tens of meters of rock at best. Low volume and subsilt-size rock sample recovery and mixing with overlying Paleozoic sedimentary rocks can confound lithologic interpretation. Further, more modern geochemical, petrographic, and mineralogic analyses (e.g., Lidiak and Ceci, 1991) have not been performed on many Precambrian samples, especially along the Ohio COCORP lines.

Of the 13 wells along the Ohio COCORP 1 seismic line (Table 2), 1 well recovered core and 8 wells recovered drill cutting samples from the Precambrian. Five of the Precambrian tests along the line do not have well sample cuttings available. Sample cuttings from wells are typically small pieces of rock <1 mm in diameter that were recovered during well drilling. When samples are examined under binocular microscope (10× to 30× magnification), lithologies generally range from fresh to weathered rhyolite, granite, gneiss, or indeterminate lithologies. Petrographic microscope examination of grain-mount thin sections from sample cuttings from five of these wells indicates granite, granodiorite, gneiss, syenite, amphibolite, volcanics, and arkose lithologies. Most of the samples examined appear altered to sericite, chlorite, and adularia (feldspar). Samples examined from the Prinkey well in Logan County, and Case and Jones wells in Delaware County, suggest that sedimentary units were drilled beneath the Paleozoic Cambrian Mount Simon Sandstone. The Johns well of Logan County, Ohio, reached a total depth in Eastern Granite-Rhyolite rocks, based on core and well sample cuttings. Samples from the remaining wells near the COCORP line were reported as altered granite, syenite, gneiss, or amphibolite.

Exceptions to generally poor data from sample cuttings are the five core tests that penetrated Precambrian sedimentary rocks in Ohio and the surrounding area (Fig. 2). Cores from four key locations verify that layered seismic reflectors beneath the Paleozoic reflectors in western Ohio, northern Kentucky, and central Michigan are of Proterozoic age (Fig. 2, wells 1, 4; Fig. 1, Texaco-Sherrer well in Kentucky and McClure-Sparks well in Michigan). Dean and Baranoski (2002a, 2002b) used a contoured RMS interval velocity profile along the Ohio COCORP lines and contoured second derivative magnetic intensity data (Dean et al., 1998, 1999, 2000) to speculate on the variable sedimentary lithologies represented by layered seismic reflectors.

Another significant drill hole is the Friend well south of the COCORP line in Clark County, Ohio (Fig. 2, well 2), that encountered an ~400-m-thick sequence of arkose, black carbonates, and volcanics beneath Paleozoic rocks (Wasson, 1932). The stratigraphic relationships of the complex rock sequences in the Friend well to those of the Middle Run Formation of western Ohio are unknown. Lithologies encountered by the Friend well differ greatly from the Middle Run Formation and suggest that extensive Proterozoic volcanic, carbonate, and clastic sedimentary sequences can be interpreted from seismic data. Dean and Baranoski (2002a, 2002b) hypothesized that deep layered seismic reflectors along the western Ohio COCORP line represent lower velocity sedimentary and volcaniclastic rocks. The use of these core locations and the anomalous Friend well support correlation to layered seismic sequences throughout western Ohio beneath the Paleozoic.

Seismic reflection and geopotential data are the accepted exploration tools for geologic modeling prior to deep well drilling. We use seismic stratigraphic techniques in an attempt to add clarity and unravel the episodes of late Precambrian geologic history in Ohio. The value of sequence stratigraphic analysis is well demonstrated by numerous oil and gas fields discovered worldwide, where this method is a primary tool used during exploration (e.g., Gregory, 1977; Mitchum et al., 1977; Abbot, 1979; Hubbard et al., 1985). It is the recognition of sequence boundaries and the lateral progression of sequences within the seismic profile that allow a better understanding of the rock record in the absence of well control. As a basin analysis tool, sequence geometries are routinely mapped to determine the extent, structural history, and evolution of sedimentary basins. Suspected Precambrian sedimentary basins are well known from layered seismic reflection data in the midcontinent (Brown et al., 1982; Cannon et al., 1989; Pratt et al., 1989, 1992; Culotta et al., 1990; Shrake et al., 1990; Hoehn, 1991; Drahovzal et al., 1992; Milkereit et al., 1992; Hauser, 1993; Drahovzal, 1997; Potter et al., 1997; Stark, 1997; Dean and Baranoski, 2002a, 2002b). Structurally complex seismic sequences representative of Precambrian sedimentary basins have been interpreted in the Kentucky region (Drahovzal, 1997). Similar layered sequences have also been isolated on seismic reflection data from the Illinois basin in the absence of conclusive drill hole data (McBride et al., 2003).

The COCORP OH-1 seismic reflection sequences A–D in this paper consist of packages of layered reflectors, which terminate internally against sequence boundaries, and in some instances against faults. In other cases, reflectors pinch out away from areas of sequence thickening along with apparent onlap or downlap. The seismic sequences are further characterized by low impedance contrast between reflectors, and changes in variable density display, in comparison to typically less coherent reflectors of the underlying crystalline complex (Dean and Baranoski, 2002a, 2002b). The sequences are labeled A, B, C, and D, with A being the oldest and D the youngest (Figs. 3 and 4). The interpretation drawn in Figures 3 and 4 was registered and transcribed digitally from an original interpretation on the 3 s output paper sections across the entire COCORP OH-1 and OH-2 (from Dean and Baranoski, 2002a, 2002b). Internal seismic reflection characteristics and parameters used to separate the sequences are listed in Table 3. Present-day ordinal directions are used for relative geographic references.

The sequences are bounded by major regional unconformities on the top of the Eastern Granite-Rhyolite and/or Grenville Provinces and the base of the Paleozoic. The estimated combined thickness of the sequences (based on an assumed shallow crustal velocity of 6 km/s) ranges from 0 at pinchouts to 2700 m in tectonically thick areas. The maximum thickness of combined sequences is ~4000 m in the East Continent Rift System.

Sequences A and B are oldest and are located within the thickest package of well-developed, highly deformed, westward-dipping reflectors (Fig. 3, vibrator point, v.p. 700–900). Sequence A reflectors are generally less continuous than B reflectors. The base of sequence A marks a major unconformity on Eastern Granite- Rhyolite rocks. Normal faulting (down to the west) between v.p. 700 and v.p. 900 on the east side of locally thick packages of highly deformed reflectors is prominent where reflectors onlap against normal faults and adjacent to stacked, lens-shaped reflectors (v.p. 700–800). Both A and B are also overthickened due to westward thrusting east of the normal faults.

Sequence B consists of higher continuity reflectors, which are thickest between v.p. 225 and v.p. 700, and thin westward beneath sequence C. Sequence B is overthickened where reflectors are offset by thrust faults at v.p. 500, just above 1.5 s (Fig. 4). Both A and B are truncated eastward by Eastern Granite-Rhyolite reflectors between v.p. 950 and v.p. 1050. Higher continuity reflectors of sequence C pinch out at v.p. 800 beneath sequence D and thicken between v.p. 300 and 500 in a normal fault-bounded basin (Fig. 4). Sequence C is overthickened and deformed from thrust stacking at about v.p. 500 above highly deformed sequence B (Fig. 4). Sequence C becomes thinner west of v.p. 225 (Fig. 4).

Sequence D is the uppermost sequence of Precambrian reflectors covering a broad area eastward and westward from the Grenville front magnetic lineament at v.p. 1000 and v.p. 1100. Lower continuity reflectors of sequence D are in unconformable contact with underlying Eastern Granite-Rhyolite rocks, the Grenville complex, and sequences A, B, and C. This uppermost Precambrian sequence has a well-defined angular unconformity with the overlying Paleozoic reflectors, where a thrust slice of Eastern Granite-Rhyolite occurs at v.p. 950. A broad paleotopographic high surface occurs where uppermost D reflectors are truncated between v.p. 750 and 950 beneath Paleozoic reflectors. Sequence D is thickest from v.p. 1000 to v.p. 1300, where the sequence pinches out and is overthickened from westward thrusting onto Eastern Granite-Rhyolite. Sequence D is also thick between v.p. 600 to v.p. 800, pinching out to the west over sequence C further west of v.p. 225 as well as to the east onto sequence B at v.p. 900.

In the absence of U/Pb dating and updated geochemical analyses of Ohio's Proterozoic rocks, a major paradigm shift is required to embrace the presence of Grenville-age sedimentary basins using seismic sequence stratigraphy. We utilize COCORP OH-1 seismic reflection data and limited well data to interpret continuous layered reflectors as angular unconformity-bounded sedimentary sequences. The stratigraphic and structural arrangement of the sequences clarify the relative timing of post–Eastern Granite-Rhyolite Province tectonic elements leading to formation of the Fort Wayne Rift, East Continent Rift System, and Grenville Province. These seismic sequences are further characterized by anomalous, relatively continuous reflectors, which may onlap or downlap. These characteristics are in significant contrast to typically less coherent, more chaotic reflectors representing crystalline rocks. The sequences record episodic deposition of sediments during the latter part of the Grenville orogeny into an expanding system of normal fault-bounded basins. Rifting and a westward shift of the major axis of depositional thickening began as a series of half-grabens, bounded on the west by step-faulted scarps and culminated with a highly deformed foreland basin region, thus marking the end of the Grenville orogeny. Similar packages of reflectors are observed in higher-quality seismic data elsewhere in Ohio (e.g., Shrake et al., 1990; Hoehn, 1991; Drahovzal et al., 1992; Paramo, 2002) and the surrounding region (e.g., Hoehn, 1991; Pratt et al., 1992; Riley et al., 1993; Wicks, 1996; Drahovzal, 1997; McBride et al., 2003). Grenville thrust sheets, which have been adequately defined on seismic lines in the Ohio region, help delimit our model (Beardsley and Cable, 1983; Pratt et al., 1989; Culotta et al., 1990; Hoehn, 1991; Riley et al., 1993; Asgharzadeh, 2002).

These sequences, from oldest to youngest A, B, C, and D, are tectonically thickened to 4000 m in western Ohio (Fig. 5). Sequences A, B, and C represent rifting phases of distal crustal extension, phases that occurred sequentially and punctuated the Grenville orogeny. Sequences A and B are the most localized, representing volcaniclastic and alluvial sediments deposited above a northwest-trending gravity low in a rapidly subsiding fault-bounded Fort Wayne Rift (Fig. 1), which may be coeval with the southeastern arm of the Midcontinent Rift in Michigan (Drahovzal et al., 1992). It is possible that rocks of the undrilled Fort Wayne Rift are similar to Keweenawan lithologies found in the Michigan McClure-Spark's well and represented as seismic sequences of the Midcontinent Rift System (Brown et al., 1982). However, relative timing relationships between the Fort Wayne Rift and Midcontinent Rift System have not been established with well data. Lens-shaped reflector geometries in sequence B suggest carbonate reefs(?), which could be stromatolite buildups. Alternate lithologic interpretations of the lens-shaped B reflectors include pods of evaporites or volcanics. Sequences A and B are also correlated to the Illinois-Indiana Centralia sequence reflectors of Pratt et al. (1992).

The reflectors of sequences C and D are correlated to the Middle Run Formation of western Ohio, which was deposited in the extensive East Continent Rift System throughout western Ohio and the surrounding area. The deposition of sequence C marked the end of subsiding rift basins. The depositional axis of the East Continent Rift System remained toward the western side of the basin throughout deposition of the Middle Run Formation, which we interpret as fluvial to alluvial volcaniclastics and shales deposited in highly faulted rift basins. Detrital zircons from the Middle Run Formation in southwestern Ohio, dated as 1.03 Ga old by Santos et al. (2002), and nearby seismic lines showing similar Precambrian sequences (e.g. Shrake et al., 1990) lend support to the regional extent of sequence C. However, the eastern depositional limit of the East Continent Rift System sediments is not known, and is generally assumed to be west of the Grenville front magnetic trend. Extensive crustal shortening and subsequent erosion during and following the Grenville orogeny could have resulted in removal of an eastward continuation of the East Continent Rift System beneath present-day stacked Grenville metamorphic and/or igneous thrust sheets, as suggested by Drahovzal et al. (1992). A westward shift of the East Continent Rift System depositional axis suggests that crustal extension migrated westward in response to the late Grenville continental collision, culminating with deposition of sequence D into a widespread regional foreland basin.

During deposition of sequence D, western Ohio became an extensive foreland basin region as mountain building continued, thus marking the end of the Grenville orogeny. Crustal shortening accelerated uplift of the craton, followed by progressive folding and thrusting. Foreland basins were segmented during late folding and thrusting events across the Ohio region. The lack of reflector continuity in sequence D is interpreted to represent discontinuous, inter-tonguing clastic sediments, derived from nearby or moderately distant highland areas. It is likely that a range of volcaniclastics, fluvial clastics, marginal marine to marine organic-rich shales and carbonates were deposited coeval with the later part of the East Continent Rift System, followed by the foreland basin development. The undated rhyolites, tuffs, and black carbonates of the Friend well (Fig. 2) support this interpretation. Lower reflector continuity in the foreland basin fill may also indicate numerous unconformities and thrust faults into basin-fill deposits.

We submit that the seismic sequences on COCORP OH1 represent highly deformed North American midcontinent rocks, and thus are the detailed records of episodic extension and compression during later Grenville orogenic events. Overthickening within all the sequences is related to normal faulting during depositional and postdepositional thrust stacking. The unconformity at the top of each sequence marks the demise of an episode of crustal extension in the west and rejuvenation of crustal shortening to the east, thus recording a westward progression of thrusting and crustal loading during collision. Each sequence, starting with the Fort Wayne Rift, became more widespread and preserved in the western basins due to continued subsidence, unlike depositionally coeval Canadian counterparts, which we propose were eroded from the Laurentian craton to the north.

As Grenville thrusting migrated westward, the depositional basin axis shifted eastward toward the Grenville front and foreland basins developed in response to crustal loading from rapidly eroding thrust massifs of the Grenville orogeny. Foreland basin sediments were deposited on eroded East Continent Rift System sediments and Grenville metamorphic and/or igneous thrust sheets to the east. Grenville thrusting and foreland basin deposition and thrusting extended west of Ohio into Indiana. Neoproterozoic–Middle Cambrian erosion removed related sedimentary rocks north of the Georgian Bay area. The present configuration of the Precambrian surface as it sharply descends from Canada beneath the Paleozoic section of Ohio shows that erosion in Ohio did not remove the large volume of rock as in Canada, as suggested by Hauser (1993). The apparent allochthonous nature of the Precambrian rocks at and near the Ohio Grenville front contrasts with the exposed Canadian parautochthonous counterpart. This appears to indicate crustal variations along the Grenville front magnetic lineament during and following the Grenville orogeny southwest of the exposed Canadian shield. The implication is that crustal heterogeneities affected paleotopography and resulting deposition and preservation of Precambrian and Paleozoic sedimentary sequences in the Ohio region. Erosion of foreland basin fill marked the end of the Proterozoic, which set the tectonic stage for deposition of Paleozoic sediments and stabilization of the interior Laurentian craton.

Transposition of foreland basin sediments is commonplace in such basins worldwide. For example, in Precambrian sedimentary strata in northwestern Canada, early-formed basin deposits were transported in piggyback fashion by later thrust faulting (Cook and Clark, 1990). White et al. (2000) used the eastern Canadian Cordillera with a preexisting Proterozoic rift margin as a possible analogy to resolve issues of Grenville extensional episodes possibly related to crustal thinning. Sequence D reflectors therefore represent foreland basin fill, which was deposited last into major depocenters throughout Ohio in response to thrust loading of the advancing Grenville thrust sheets.

Seismic signatures similar to those of the Precambrian sequences described for this paper also occur beneath Phanerozoic reflectors and above east-dipping reflectors (Grenville thrust sheets) on the Great Lakes International Multidisciplinary Program on Crustal Evolution (GLIMPSE), line J of the Lake Huron– Georgian Bay area (Geological Society of America, 1988), Lake Erie (Hoehn, 1991), Sandusky County, Ohio (Asgharzadeh, 2002), Warren County, Ohio (Shrake et al., 1990), and other industry seismic lines in Ohio (Figs. 1 and 2). Some of these published seismic data show smaller, localized rift basins, which most likely developed after the Grenville orogeny, during Neoproterozoic and/or Cambrian time. In a regional context with other seismic data, our model suggests that the Precambrian unconformity surface plunges southwest at a steep angle beneath Phanerozoic cover below Georgian Bay and extends into the midcontinent, where Grenville-age sedimentary rocks are preserved. We submit that these highly deformed midcontinent sequences are the detailed record of distal extension and compressional episodes of the latest Grenville orogeny.

The sequences described in this paper define cycles of rifting followed by crustal shortening during the Grenville orogeny in Ohio and the surrounding region (Fig. 5). A westward shift of depositional axes indicated by the thickest reflector packages suggests that crustal extension migrated westward in response to the late Grenville continental collision. Angular unconformities bound the sequences, the three most apparent being (1) the top of the Eastern Granite-Rhyolite Province; (2) the top of the Grenville metamorphic and/or igneous complex; and (3) the top of the Grenville-age sedimentary basins. It is very appealing that these sequences occur within the broader Grenville orogenic cycle and not prior to or following it. Thus, we interpret these seismic sequences as the preserved sedimentary record of the latter part of the Grenville orogeny (Ottawa and Ringolet orogenic pulses of Rivers, 1997). However, the original configuration of these basins and accompanying deposits cannot be determined in the overall context of the Grenville orogeny because of pronounced erosion at the Precambrian-Cambrian unconformity and a lack of deep well control. Similar sedimentary rock sequences were most likely deposited to the north on the Canadian shield, but have long since been eroded away, as suggested by Hauser (1993).

It is well accepted from work on the Canadian shield that the Grenville orogeny lasted several hundred million years, with long periods of crustal extension followed by episodes of continental collision in Canada (Rivers, 1997; White et al., 2000). The combined scenarios of these sequences indicate an overall westward progression of structural and tectonic events during the later part of the Grenville orogeny. Aside from vertical exaggeration of reflector packages in Figure 3, overall structural geometries associated with our interpreted thrust sheets (Dean and Baranoski, 2002a, 2002b) are consistent with foreland basins filled with nonmarine clastic and volcaniclastic sediments and marine carbonates (Cook and Clark, 1990).

Our proposed scenario is as follows.

1. A major regional unconformity developed on the Eastern Granite-Rhyolite Province. Crustal shortening began at the eastern continental margin of Laurentia during the Grenville orogeny.

2. Continued shortening and another major regional unconformity developed on the Eastern Granite-Rhyolite Province and accreted Grenville terranes, exposing the rocks defining the Grenville front magnetic lineament.

3. As continental collision entered eastern Ohio, western Ohio became the site of an extensive fault-bounded rift basin complex, starting with the Fort Wayne Rift followed by the extensive East Continent Rift System (e.g., Ottawa orogenic pulse of Rivers, 1997).

4. Continued continental collision resulted in westward-advancing thrust sheets and deposition of foreland basin sediments over the rift basins (Ringolet pulse of Rivers, 1997).

5. Continued Grenville thrusting resulted in segmented foreland basins, which were thrusted westward along with earlier formed rift basins.

6. A period of Neoproterozoic and Cambrian rifting and sedimentation and erosion removed continental-collision–related sedimentary rocks prior to deposition of younger Paleozoic sediments in portions of the Ohio region.

Useful analogies can be drawn from the extensive preserved Western Cordillera rocks, where the distal effects of variations in crustal stresses and the transmission of stresses are well known today, but were controversial in the 1980s. In a regional context our model suggests that the deeply eroded Precambrian unconformity surface exposed in Canada plunges southwest at a steep angle beneath Phanerozoic cover beneath Georgian Bay and extends into the midcontinent, where Grenville-age sedimentary rocks are preserved.

We thank Thomas McGovern of Lauren Geophysical, Denver, Colorado, for providing industry reprocessed COCORP OH-1 and OH-2 lines for this paper. Lisa Van Doren (Ohio Division of Geological Survey) performed the computer drafting artwork. We also thank Dennis Hull, Lawrence Wickstrom, Merrianne Hackathorn (present or retired staff of the Ohio Division of Geological Survey), and Joseph Eby for reviews of this paper. John McBride (Brigham Young University), Michael Hansen (retired Ohio Division of Geological Survey), and two anonymous reviewers provided critical reviews, which greatly improved this paper.