- Email: [email protected]
Insights into the mode of the South Georgia rift extension in eastern Georgia, USA
Tectonophysics 608 (2013) 613–621
Contents lists available at ScienceDirect
Tectonophysics journal homepage: www.elsevier.com/locate/tecto
Insights into the mode of the South Georgia rift extension in eastern Georgia, USA C.W. Clendenin Jr. ⁎ South Carolina Department of Natural Resources—Earth Science Group, 5 Geology Road, Columbia, SC 29212, USA
a r t i c l e
i n f o
Article history: Received 10 April 2013 Received in revised form 9 August 2013 Accepted 17 August 2013 Available online 28 August 2013 Keywords: Triassic extension Mode of rifting Core complex Lower crustal flow Flat Moho
a b s t r a c t The South Georgia rift (SGR) lies oblique to the east coast margin of North America and across the Alleghenian suture between Laurentia and Africa in southern Georgia. Regionally, the SGR can be divided into a southwest compartment and a northeast compartment across the Jacksonville structure that is located in the vicinity of that suture. Analytical and numerical models are used to characterize the mode of rifting in the northeast compartment. Borehole, COCORP seismic, and regional geophysical information from the compartment, that were used previously to infer the geometry of the basin, are reassessed with the use of those models to analyze the lithospheric conditions influencing Triassic extension. This approach led to the interpretation of core complex mode extension and to the proposal of a model of progressive rifting. The model shows how the Riddleville and Main SGR basins are associated and how changes in structural style of those two basins resulted from changing lithospheric conditions during extension. The core complex model also indicated that extension was influenced by distributed deformation of a younger, warmer, and less stable lithosphere adjacent to the Permian suture; whereas extension in other east coast rifts that lie subparallel to structural fabric was probably localized by preexisting zones of weakness. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Continental extension is characterized by differences in structural styles, width, and magmatic processes. These differences are the result of lithospheric conditions prior to and during extension (ten Brink et al., 2000). Buck (1991) emphasized that geothermal state and crustal thickness are the dominant lithospheric conditions influencing continental extension and described three modes of rifting: narrow, wide, and core complex. Modeling also shows that old, thick, and cold crusts tend to yield narrow rifts; whereas extension of younger, thinner, and warmer lithospheres give wider rifts, in which core complexes may develop (Corti et al., 2003). Keranen et al. (2009), however, demonstrated that preexisting zones of weakness influence rift mode under different conditions, if the prevailing stress field and preexisting weaknesses are properly oriented. In nature, the resulting structural style, therefore, should be considered a function of heat flow, crustal age, crustal thickness, and orientation of preexisting weakness (Buck, 1991; Corti et al., 2003; Keranen et al., 2009). Van Wijk and Blackman (2005) also point out that structural style varies along the rift axis during extension. Appreciation of the described relations is important when one tries to characterize the tectonic development of the South Georgia rift (SGR). The SGR is the southernmost Triassic basin along the east coast of North America and is buried under Atlantic Coastal Plain sediments (Fig. 1). Being buried has hampered the understanding of the SGR; and what little is known has come from borehole, regional geophysical ⁎ Tel.: +1 803 896 7702. E-mail address: [email protected]. 0040-1951/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.tecto.2013.08.019
information, and a few seismic images. Interest was reignited in 2010 when the U.S. Department of Energy funded a characterization study of the basin. However, borehole samples that could have provided a better understanding were lost with the closure of the Georgia State Geological Survey. Loss of those samples forced this part of the characterization study to rely on regional relations, published data, and a priori evidence. In this paper, subcrop maps, seismic images, and regional geophysical patterns of the SGR in eastern Georgia are reexamined, and the information is compared to analytical and numerical models to decipher structural style. This information is used to interpret mode of rifting, and then a progression rifting is proposed for the tectonic development of the SGR in eastern Georgia. The regional implications of that model are discussed to help characterize what may have influenced the mode of rifting. 2. Triassic precursor basins Regional relations show that two series of Triassic rift basins formed in western Pangea prior to the opening of the central Atlantic Ocean and that these rift series are time equivalent (Traverse, 1987). Basins found along the east coast of North America commonly are exposed on the surface, are subparallel to structural fabric, and are referred to here as the Newark series (Fig. 1). The Newark series basins are characterized by elongate half-grabens, bounded by listric-like faults lying subparallel to the coast (Withjack et al., 1998). Although the SGR is oblique to the coast, published cross sections also tend to depict a half-graben structural style (McBride et al., 1987; Withjack et al., 1998).
614
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
Fig. 1. Reference map showing seismic lines L5, L8, and SCCO21; Fall Line, and other features. Newark series basins are shown to north-northeast on reference map. NB: Newark basin; FVT: Felsic Volcanic Terrane; JS: Jacksonville structure (modified after Chowns, 2009).
Less attention has been given to the second series of Triassic basins that is buried in the Gulf Coast area and referred to here as the Eagle Mills series (Fig. 2). This series of precursor basins begins offshore of Florida immediately southwest of the SGR and extends in the subsurface northwest across Alabama–Mississippi, west across southern Arkansas– northern Louisiana, and then southwest into eastern Texas (Moy and Traverse, 1986; Pindell and Kennan, 2009; Tew et al., 1991). Pindell and Kennan (2009) also have correlated another set of basins off the northeast coast of Yucatan with this series. Structural style is not known, but regional distribution and limited seismic information suggest a horst-and-graben pattern (McBride, 1991; Pindell and Kennan, 2009; Sartain and See, 1997). 3. SGR in eastern Georgia 3.1. Borehole information Chowns and Williams (1983) produced a subcrop map of the SGR over parts of Florida, Alabama, Georgia, and South Carolina (Fig. 1).
The map shows that the SGR is divided into northeast and southwest compartments by the Jacksonville transfer fault system in southern Georgia (Tauvers and Muehlberger, 1987). The existence of the Jacksonville transfer system is considered controversial, but the polarity of extension shifts from down-to-the-northwest in the southwest compartment (McBride, 1991) to down-to-the-southeast in the northeast compartment (Cook et al., 1981) across the mapped faulting. A change in polarity implies transfer faulting or an accommodation zone. An investigation of the structure is beyond the scope of this study; and for the time being, it simply is referred here as the Jacksonville structure. In southern Georgia, the southwest compartment is characterized by horsts and grabens (McBride, 1991; Sartain and See, 1997; Fig. 1). In eastern Georgia, the northeast compartment consists of two basins separated by a horst block. The northern basin, identified from aeromagnetic information, is known as the Riddleville basin (Daniels et al., 1983). A borehole near its northwest margin penetrated 2522 m of Triassic sediments (Chowns, 2009). Depth-to-basement rocks estimated from magnetic data suggest a down-to-the-southeast asymmetric geometry (Daniels et al., 1983). The southern basin simply is referred
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
615
Fig. 2. Buried, Triassic Eagle Mills series basins south and west of SGR. JS: Jacksonville structure; BFZ: Bahamas fracture zone; DS: De Soto basin; EM: Eagle Mills basin; YB: Yucatan basins; A–O–M: Alleghenian–Ouachita–Marathon suture (modified after Chowns and Williams, 1983; Moy and Traverse, 1986; Pindell and Kennan, 2009).
to as Main SGR, and analysis of magnetic data indicates that ~3500 m of Triassic sediments are preserved (Daniels et al., 1983). Heck (1989) identified the southeast margin to be a high-angle fault, and spatial patterns of boreholes suggest that Main SGR is a full graben. The intervening horst consists of Piedmont rocks; boreholes in Treutlen County, Georgia, bottom in either biotite gneiss or cataclastic garnetiferous quartz-feldspathic gneiss (Chowns and Williams, 1983). Although the subcrop map shows closure to the northeast, Daniels et al. (1983) extended the horst northeast into South Carolina. Immediately south of Main SGR is a felsic volcanic terrane (Fig. 1). Boreholes show that principal lithologies are vitric crystal tuff, tuffaceous arkose, pyroclastic porporphyric rhyolite, and granite (Chowns and Williams, 1983). Radiometric ages of the volcanics cover an extended period of time, with two main groups from 500 to 350 Ma and 300 to 150 Ma (see Chowns and Williams, 1983). Basalt and diabase from boreholes also show another group of dates ranging from 300 to 50 Ma.
3.2. Seismic reflection data Recently, the U.S. Department of Energy sponsored 240 km of new seismic reflection data over the South Carolina portion of the SGR. SCCO2-1 clearly images the northwest margin of the rift (Clendenin et al., 2011; Fig. 1). Faults delineating that margin are inverted as harpoon-style reverse faults (see McClay and Buchanan, 1992) that look like faulted anticlines (SP 360, Fig. 3). The largest of the inverted structures lies downdip along the margin just below 3 s. The precursor extension faults are interpreted to have been steep, planar features. After review of the other project seismic lines, it was discovered that, unless the line was oriented nearly orthogonal to the margin, reactivation tended to mask structures on the image. The SGR's inverted margin in South Carolina is northeast along strike of the northwest margin of Main SGR in eastern Georgia. Recognition of inversion initiated a reexamination of COCORP Line 5 (L5) and L8 (Figs. 1, 4) that are orthogonal to the margin. A faulted anticline is clearly visible on COCORP Line 8 (L8) (Fig. 4), and the structure imaged was described in those terms by Cook et al. (1981). A reexamination also was needed because, when the available information was initially interpreted in the early 1980's, contemporary thinking tended to be focused on only the upper part of the crust and did not always consider to
what degree the entire crust was affected by extension (Rey et al., 2001). The southeastern half of L5 and all of L8 lines extend southeast from the Fall Line, the northwest edge of the Coastal Plain, to within 30 km of Savannah, Georgia. L8 is nearly parallel to the southeast end of L5 and is offset 24 km to the west (Petersen et al., 1984). Interpretations of L5 and L8 have been published in Cook et al. (1981), Petersen et al. (1984), McBride et al. (1987), Heck (1989), and others. Shown here is the original interpretive section of Cook et al. (1981). The reflectors (R) of interest are labeled A through F in Fig. 4 and are referred to in the text as RA through RF. RA begins near SP 1000 on L5, dips to the southeast, and extends to ~5 s at SP 1600. Cook et al. (1981) correlated RA with the Augusta fault. RB begins at SP 1500, dips southeast, and projects to ~5 s at SP 1600 on L5. This reflector is below the Triassic Riddleville basin and bounds the basin on the northwest. Ramping reflectors that join RB between SPs 1500 and 1800 suggest a faulted anticline. The down-dip trace of RB appears to cut RA; however, in Petersen et al. (1984) Fig. 3, RA actually appears to be cut or to merge with their reflector F that lies between RB and RA. Both RB and F project toward the northwest margin of the Riddleville basin, have a quasi-listric shape, and are interpreted to be Triassic faults (Petersen et al., 1984). RC underlies both RA and RB. The three reflectors appear to merge near SP 1600, with RC continuing onto L8. RC consists of fairly flat, discontinuous, layered reflectors that can be traced at ~5 s across L5. Examination of L5 (see Cook et al., 1981) shows that RC dips gently to the southeast from ~4 s on L5 to ~6 s on L8. Cook et al. (1981) suggested that RC represents the extension of a decollement to the northeast under the Piedmont Charlotte Belt. Heck (1989), however, proposed RC to be the brittle–ductile transition (BDT). RD begins at about SP 100 on L8 and is marked by northwestdipping reflectors beginning at ~7 s. The reflectors warp up to ~4 s at about SP 400 before reversing dip to the southeast. After reversing dip, the discontinuous reflectors of RD can be traced southeast to ~7 s at about SP 700. Reversal produces an antiformal geometry that may mark a major crustal boundary (Cook et al., 1981). It is interesting that, in different presentations of L8, no reflectors are shown above RD (see Cook et al., 1981; Heck, 1989; Petersen et al., 1984). Cook et al. (1981), however, show that fine-scale layering overlies RD in a seismic line shot perpendicular to L8.
616
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
Fig. 3. Northwest margin of SGR in South Carolina as imaged by SCCO2-1. Dotted line marks rift margin. View is to the northeast.
RE lies above 5 s between SPs 600 and 800 and is a distinct set of antiformal-looking reflectors with a recognizable southeast-dipping limb. RE is juxtaposed over the lower southeast portion of RD. Cook et al. (1981) interpreted a steeper, southeast-dipping reflection at SP 600 as a fault plane and proposed that RE represents a faulted anticline of Paleozoic age. RF, interpreted by previous workers as the Moho, can be traced across L5 onto L8 at ~10 s. The discontinuous reflector has a slightly undulating appearance. Cook et al. (1981) show some minor relief on the
Moho, whereas others interpret it to be flat (Heck, 1989; McBride et al., 1987; Petersen et al., 1984). 3.3. Aeromagnetic and Bouguer-gravity anomalies In their Fig. 6, Daniels et al. (1983) show an aeromagnetic profile and interpreted geologic cross section that depicts the deeper Main SGR separated from the Riddleville basin by an intervening high. The northwest half of Fig. 5 illustrates that aeromagnetic profile. The northwest margin
Fig. 4. Interpretive cross section of Lines 5 and 8. View is to the northeast. Reflectors abbreviated RA–RF are discussed in text (modified after Cook et al., 1981).
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
of Main SGR is easily recognized as a strong northeast-striking lineament on aeromagnetic maps of Georgia (Daniels, 2001). This lineament correlates with marginal structures seen on L8 (RE) and SCCO2-1 (Fig. 1) and with the southeast margin of a 15–20 km wide, northeast-trending, small ridge-like Bouguer-gravity high (Fig. 6). Although overall patterns infer an east–west orientation, magnetic highs can be aligned with the northeast-trending Bouguer-gravity ridge. Both aeromagnetic and Bouguer-gravity maps show a broad, and somewhat featureless low southeast of the lineament, that marks Main SGR. Northwest of the Bouguer-gravity ridge is a Bouguer-gravity low that is bounded on the north by a major east–west regional lineament (Daniels et al., 1983; Fig. 6). The lineament corresponds to the Magruder fault. The Bouguer-gravity low is interpreted to be the Riddleville basin. The basin, however, is vague on aeromagnetic maps and is marked by a featureless magnetic low that extends east from Riddleville, Georgia (Daniels et al., 1983), along the southern margin of the Magruder fault. The aeromagnetic low also is northwest of the Bouguer-gravity ridge. The exact orientation of the Riddleville basin is not well constrained by the aeromagnetic or Bouguer-gravity patterns. The juxtaposition to the east–west Magruder fault gives the impression of an east–west basin (Fig. 6). Geophysical lows on the south side of that fault, however, may be overprinting the Riddleville footprint. A pronounced northeasttrending Bouguer-gravity low that lies adjacent to the northwest margin of the Bouguer-gravity ridge may actually delineate the basin. This Bouguer-gravity low correlates with aeromagnetic signals that Daniels et al. (1983) interpret as the basin. It should be noted that circular and elliptical aeromagnetic and Bouguer-gravity highs lie to the northeast and southwest of the northeast compartment of the SGR (Fig. 6). The different geophysical highs have equally large amplitudes, and Daniels et al. (1983) have interpreted these highs as unmetamorphosed mafic–ultramafic intrusive complexes. Although direct evidence of age of emplacement is unavailable, a priori evidence led Daniels et al. (1983) to infer that the circular magnetic-Bouguer gravity anomalies may have intruded in the lower Mesozoic. 4. Discussion In overview, the Riddleville and Main SGR basins are depicted as about the same size northeast of the Jacksonville structure, with Main SGR expanding in width to the northeast (Fig. 1). The bore-hole data (see Chowns, 2009), however, show that the width of Main SGR would increase if drawn using the half-way method of unknowns. Geophysical evidence also clearly shows that Main SGR is the larger of the two basins (see Daniels et al., 1983). The Piedmont horst also would be narrower using the described method. Combining Daniels et al.'s (1983) aeromagnetic profile with Heck's (1989) interpretation of the southern margin of Main SGR produces the cross section of the two basins shown in Fig. 5. The COCORP seismic information can be interpreted differently as well (Fig. 4). RA may have had little or no influence on rifting and is not discussed further. RB and the warped reflectors are interpreted to be a harpoon-style inversion fault. Shape and location of inversion could have easily been interpreted as reverse drag on a normal fault. As stated in Section 1, older conventional thinking tended to focused
617
on the upper crust, and Cook et al. (1981) point out that data were discussed in reference to surface geological nomenclature. Petersen et al. (1984) suggested that normal fault motion on RB must have been taken up along a portion of underlying RA. Instead, models by Reynolds and Lister (1990) suggest that RA was captured and displaced by RB. Capture by RB is probable because little interaction occurs between normal faults and shallow-dipping, preexisting zones of weakness (Mattioni et al., 2006). RB bounds the Riddleville basin's northwest margin and is interpreted to be an inverted, Triassic-age fault (Fig. 4). Heck (1989) interpreted RC to be the BDT (Fig. 4). Lister and Davis (1989) have argued against such interpretations because the BDT would not remain fixed to a particular plane during progressive extension. If RC is a subhorizontal decollement, it represents an inherited heterogeneity in the crust, that traces its origins to a Paleozoic orogeny (Cook et al., 1981). Jolivet et al. (2010) envision that inherited heterogeneities, like RC, should lie above and in the BDT, with cataclastic flow at the top and ductile flow at the base. The horst consists of Piedmont rocks. In overview, the Piedmont is a polyphase-deformed terrane, characterized by thin-skinned thrusting (Cook et al., 1981). A shallow-dipping crustal heterogeneity may have either a high- or low-competency contrast with an overlying thrust package (Le Pourheit et al., 2004). If this type of weakness is involved in active extension, it may strongly localize faulting into a narrow zone and influence mode of rifting (Le Pourheit et al., 2004; Mattioni et al., 2006). Faults developed in an overlying thrust package root in such crustal heterogeneities, and some of the faults may intersect each other (Mattioni et al., 2006). Such relations can be seen in Fig. 4. RF has been interpreted as a flat, relatively shallow Moho. If true, the flat Moho indicates that the crust was sufficiently hot and thick enough to allow lower crustal flow (Block and Royden, 1990). Flow is so efficient that long-wave topography on the Moho is effectively removed (McKenzie et al., 2000). Flow also keeps crustal thickness roughly constant and distributes deformation (Block and Royden, 1990). Models show that flow starts before faulting in the upper crust (Tirel et al., 2006) and that extension forms a narrow graben. With increased extension, flow moves into the locally thinned upper crust and restricts migration of deformation (Le Pourheit et al., 2004). Flow is always toward and into the thinned area (Hopper and Buck, 1996; Rey et al., 2009). In flow mode, thinning is significantly lower than that of extension (Bertotti et al., 2000). Although some of the felsic volcanics may be related to the younger, mafic Central Atlantic Magmatic Province in eastern Georgia (Chowns and Williams, 1983), the presence of felsic volcanics immediately to the south (Fig. 1) is considered a priori evidence of an increased local geotherm. Chowns and Williams (1983) have offered a number of interpretations for the felsic volcanic rocks to be either early Paleozoic or Proterozoic in age. Age of the rocks is not that important because Ziegler and Cloetingh (2004) point out that heat flow even in older orogenic belts is still elevated because of the younger age of their mantle– lithosphere and possibly to higher radiogenic heat generation potential of their crust. Heat flow is important in extension, and part of that heat could have come from contemporaneous volcanism. Evidence of this volcanism could be clasts of felsic volcanic debris that are common in inferred Triassic-age litharenites (Chowns and Williams, 1983). Mixing of
Fig. 5. Cross section of Riddleville and Main SGR basins constructed from Daniels et al. (1983) and Heck (1989). View is to the northeast. RB — Riddleville basin; CP — Coastal Plain.
618
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
Fig. 6. Bouguer gravity anomalies related to the SGR. Small Bouguer ridge is bounded by magnetic lineaments that define northwest margins of Riddleville and Main SGR basins. Seismic lines also are shown for reference. MF — Magruder Fault (modified after Daniels, 2001).
sediment and volcanic debris is not surprising in an extensional environment. In the Basin and Range province, such explosive volcanics have erupted contemporaneously in areas that were undergoing localized extension (McKenzie et al., 2000). Some combination of possible magmatic activity, presence of underplating mafics, and thickened crust in the vicinity of the suture with extension starting soon after collision implies that the general area had an increased geotherm. As a result, a partially molten layer may have been present in the lower crust (see Rey et al., 2009). The presence of heat would have modified the thermal field and promoted partial melting of warm, thickened crust. Liu (2001) also noted that, in the Great Basin, extension is near margins of the volcanic field where the crust was weakest. The SGR is juxtaposed to the FVT (Fig. 1); and rheologic contrast between the two areas probably had some influence on the spatial location of Triassic extension. Slow extension with partial melt enhances upward exhumation of hot deep rocks (Rey et al., 2009). Crystallization of rising partial melt in the subsurface produces an antiformal or domal geometry that represents preferential exhumation (see Rey et al., 2009). RD fits that geometry and is interpreted to be the upper limit of exhumation of partially molten lower crust. If correct, this implies that the Piedmont horst has a migmatitic core. The northeast-striking, Bouguer-gravity ridge (Fig. 6) and the aeromagnetic high separating the Riddleville basin and Main SGR are thought to be imaging the suspected migmatites. Thompson and McCarthy (1990) point out that gravity anomalies associated with highly extended terranes, including core complexes, show no prominent, characteristic gravity signature and any anomaly would be small. The likely explanation for those observations is that compensation has taken place by emplacement of material of crustal density (Thompson and McCarthy, 1990). Emplacement would further suggest that products of crustal melting and mixing with basaltic magma rose to intermediate or high levels in the crust, and this relation is supported by flat Moho (Thompson and McCarthy, 1990). Cataclasites mark the Piedmont horst (Chowns and Williams, 1983). The presence of garnet in the described cataclasite clearly differentiates it from the silicified cataclasites commonly mapped in the proximity of
faults in the Piedmont (Garihan, 2013, pers. commun.). The gneissic character of the rock also implies mylonitization, whereas the Piedmont silicified cataclasites are brecciated quartz bodies. Cataclasites form in a special sequence close to and directly above the BDT (Lister and Davis, 1989). A simple interpretation is that the garnet cataclasite is part of a preexisting shear zone captured and exhumed from deeper levels (Reynolds and Lister, 1990). A second interpretation is that the garnet cataclasite is part of a detachment fault that had been warped over a growing dome. The second interpretation is preferred here because isostatic warping of a detachment fault is common in exhumation models (Brun et al., 1994). The lack of reflectors above RD and the fine-scale layering imaged by Cook et al. (1981) on another seismic line support the idea of a core complex and doming. McCarthy and Parsons (1994) describe the seismic signature of a core complex as characterized by a pervasive, weak, laterally discontinuous reflective fabric represented by abundant fine-scale layering. RE was interpreted by Cook et al. (1981) as related to Paleozoic compression. Heck (1989) disagreed and pointed out that younger tectonic events should be considered in interpretations of seismic images from this area. If Paleozoic in age, RE should have been reactivated as a normal or normal-oblique fault by the Triassic extensional stress field. The shape of RE, however, clearly fits the characteristics of a harpoonstyle structure. Inversion could have resulted from an even younger tectonic event, e.g. Atlantic tectonics or Late Mesozoic–Early Cenozoic collision of Cuba and the Bahamas. If true, any younger, properly oriented compressive stress field would have inverted both Main SGR and the Riddleville basin. Accepting RE as a harpoon-style structure, its predecessor had to have been a younger, down-to-the-southeast normal fault. The location of RE over the southeast flank of RD further implies that the youngerplanar fault formed during progressive extension (Fig. 4). The progression began as an older detachment fault was rotated, or warped, to a lower dip; and as the dip forced abandonment, the younger highangle, planar fault formed in the hanging wall (see Jolivet et al., 2010). During those related processes, thinning and progressive cooling would have resulted in a downward migration of BDT (Tirel et al., 2006). If RE is an arched continuation of RC, younger-normal faulting
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
cut through the heterogeneity and rooted on a deeper BDT. This interpretation is consistent with Cook et al.'s (1981) statement that decollements could extend eastward as far as the present continental shelf. The width and thickness of sediments also imply that younger faulting localized in Main SGR, as proposed by Jolivet et al. (2010) for the Corinth Rift. Footwall rebound also would have helped exhume the Piedmont horst. With continuing extension, detachment faults that characterize core complex extension are deactivated, and extension becomes characterized by steep-planar faults that cut the older features (Bertotti et al., 2000). This change in structural style has been proposed as the result of the exhaustion of crustal material that can flow and of the increasing distances crustal material had to flow (Bertotti et al., 2000). A complementary explanation may be that crustal flow occurred only until the hot rocks reached shallower depths in the rising dome and cooled (Whitney et al., 2013). Either explanation would result in a change from lower crustal flow to no-flow mode lithospheric conditions. With a shift to no-flow mode, thinning becomes directly proportional to extension that causes subsidence (Bertotti et al., 2000); subsidence was expressed as a deepening of Main SGR. Faulting of older detachment structures with continuing extension produces a relevant modification to structural style with subsequent normal faults steep, planar structures extending down to the relocated BDT (Bertotti et al., 2000; Jolivet et al., 2010). Differences in structure style also may be the reason that the Riddleville basin (half-graben) and Main SGR (McKenzie-type full graben; Fig. 4) have not been structurally related in the past. On the basis of the described subcrop and geophysical relations, core complex mode rifting is interpreted to have occurred between the Jacksonville structure and the Savannah River in eastern Georgia (Fig. 1). The northeast-striking horst would define the subcrop of the domed, higher-grade metamorphic rocks. The structure is referred to here as the Treutlen County core complex, after the location where the garnet cataclasite was recovered.
619
lower crustal strength ratio (Wijns et al., 2005). Localized thinning of the brittle crust was associated with lower crustal flow that kept the Moho rather flat (Corti et al., 2003). Localization of early extension was expressed as only a few, major normal faults (Brun et al., 1994; Whitney et al., 2013; Wijns et al., 2005). A slow extension rate and a higher geotherm were the determining factors in the number of major faults developed (Tirel et al., 2006), which are shown here as a few southeast-dipping and northwestdipping conjugate-fault pairs rooted initially in RC (Fig. 7a). Three faults are designated: the farthest northwest fault (RB); the most interior southeast-dipping fault (SDF); and the farthest southeast, northwestdipping fault (NDF). It should be noted that, throughout the description of progression of rifting, two-letter abbreviations refer to features in Fig. 4. Rebound of SDF's footwall provided space for lower crustal flow that was moving toward the extended area; described flow and decoupling from the mantle kept the Moho flat (RF, Fig. 7b). Movement on RB and flow produced counterclockwise block rotation between RB and SDF that initiated the Riddleville basin and assisted exhumation by increasing footwall space. As extension continued, exhumation was determined by a feedback system between upper crustal extension and necessity for flow to fill the thinned zone (Rey et al., 2009; Fig. 7c). Continued movement on RB and rigid-body rotation of the rising dome deepened the Riddleville basin. During this time, RC continued to act as a decollement as the hanging-wall block bound by NDS shifted slightly down the shallow-dipping heterogeneity RC (Mattioni et al., 2006). Shifting resulted in asymmetric widening of Main SGR. Minor faults
5. Progression of rifting Interpretation of the Treutlen County core complex, presence of horsts and grabens in the southwest compartment, and a flat Moho show that distributed deformation influenced kinematics and controls on the progression of rifting. Tirel et al. (2006) point out that high extension rates driven by free-gravity spreading, i.e. lithospheric extension under its own weight without any boundary limitation, produce a series of tilted blocks, whereas lower extension rates influenced by boundary controlled gravity spreading favor conjugate patterns and the development of horsts and grabens. Core complexes occur when a low viscosity, thick lower crust and a higher syntectonic geotherm are present (Corti et al., 2003; Whitney et al., 2013). Only a small number of faults are developed under lower extension rates (Tirel et al., 2006; Wijns et al., 2005). Under a higher geotherm, strongly localized extension in the upper crust also only produces a limited number of faults (Whitney et al., 2013). The lower crust will flow into and replace the thinned brittle crust and remove the effect of gravity forces, which restricts migration of deformation (Corti et al., 2003). Any proposed progression of rifting needs to satisfy those relations and to account for two basins. The geometry of those two basins also has to compare to that shown in Fig. 5 — a half-graben juxtaposed to a full graben. Brun et al.'s (1994) model that was expanded by Tirel et al. (2006), with a few other additions, is applied here. After Late Paleozoic collision and subsequent thermal relaxation of the thickened crust, Triassic strain was localized in weak crust adjacent to the felsic volcanic terrane (Fig. 1). Heat had modified or was modifying the thermal field; advection of heat promoted partial melting that further enhanced localization (Corti et al., 2003). Localization also may have resulted from a low-competency contrast between the overlying crust and RC (Le Pourheit et al., 2004) and an assumed large upper to
Fig. 7. Progression of rifting; a–c formed under lower crust flow mode and d under noflow mode. Arrows and X indicate either movement directions or deactivation. Views are to the northeast. NDF — north dipping fault; SDF — south dipping fault; RB — Riddleville basin; BDT — brittle ductile transitions; RA, RB, RD, RE, and RF — seismic features described in text.
620
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
that developed between NDS and SDS rooted in RC and effectively thinned the overlying crust (Mattioni et al., 2006). As exhumation continued, SDF was progressively warped and folded over the rising dome (Fig. 7c). Partial melt flowed upward (Rey et al., 2009), and the relative rate of cooling vs. exhumation influenced the location of RD (Wijns et al., 2005). Cooling and downward migration of the BDT accompanied that deformation. A deeper BDT and exhaustion/cooling of crustal material able to flow forced faulting on RE to cut deeper into the crust as movement on SDF was abandoned (Fig. 7d). Deep, planar faulting deactivated RC as faulting rooted in the deeper BDT. The exhumed dome also enhanced lateral partitioning of deformation (Tirel et al., 2006). Deformation was expressed by larger finite extension that deepened Main SGR as deformation changed to the no-flow mode (Bertotti et al., 2000; Jolivet et al., 2010). 6. Regional implications Distributed deformation influenced by boundary controlled gravity spreading also has regional implications. A flat Moho below domains of distributed deformation has been described as a fundamental difference between wide rifts and narrow rifts (Tirel et al., 2006). The structural style of Newark-series basins clearly fits narrow mode rift characteristics of localized deformation. The narrow mode for the Newark-series basins would imply that the lithosphere extended along the east coast of Laurentia in Triassic was old, cool, and stable as compared to the lithosphere in eastern Georgia near the Permian suture. An alternative explanation is that reactivation of zones of preexisting weakness may have had a greater influence on the Newark-series basins than rheological stratification (see Keranen et al., 2009). The east coast of Laurentia was deformed through the Paleozoic by the Taconic, Acadian, and Alleghenian orogenies, and Newark-series basins are subparallel to regional structural fabric. As pointed out, heat flow even in older orogenic belts is still elevated (Ziegler and Cloetingh, 2004); but narrow rifts, e.g. the Main Ethiopian Rift, can form in hot, weak lithosphere if the primary control on mode of extension is preexisting lithospheric weakness (Keranen et al., 2009). Where rifting crosses regional structural fabric, however, preexisting lithospheric weakness cannot be reactivated by the prevailing stress field, and new faults develop (Ziegler and Cloetingh, 2004). The regional distribution of basins in the Eagle Mills series gives the impression that horsts and grabens characterize structural style (Fig. 2). Horst and grabens have been recognized in the southwest compartment across the Jacksonville structure (Sartain and See, 1997). Transfer faults form when rift propagation is blocked by the boundary of two geologic provinces (Van Wijk and Blackman, 2005); in this case, such a boundary would be the Permian Laurentia/Africa suture. Stalling of rift propagation results in a change from localized to distributed deformation, and the two deformation styles are balanced by the development of a transfer fault (Van Wijk and Blackman, 2005). If this is true, the location of the core complex in the northeast compartment also substantiates the Van Wijk and Blackman's (2005) interpretation that structural style varies along the rift axis during extension. Differences in deformation style across the Jacksonville structure help to further establish how structural style varies during extension. Corti et al. (2003) point out that, in the Basin and Range Province, rifting evolved in two stages: initial core complex development and later horse-and-graben structures. Bertotti et al. (2000) also describe that initial core complexes were overprinted by horst-and-graben structures by progressive extension in the North Tyrrhenian area between North Corsica and Tuscany. Lower crustal flow during these stages may have been efficient enough to prevent localized thinning that would lead to crustal separation (Bertotti et al., 2000) and would explain why the SGR is a failed rift. Accepting these relations along with the interpretation of the Treutlen County core complex indicate that the SGR is probably part of the Eagle Mills series as suggested by Gulf Coast geologists (Tew et al., 1991; Sartain and See, 1997; Fig. 2). The juxtaposition of the Eagle
Mills series basins to the Alleghenian suture further shows that extension was controlled by a younger, warmer, and less stable lithosphere thickened by the Permian contractional episode. The location of the Treutlen County core complex also may indicate direction of lower crustal flow. Northeastward flow may have equilibrated crustal thicknesses across the Permian suture that separated possibly thicker African crust from Laurentian crust. Directional flow initially was suggested as a response to greater extension to the north in the East Humboldt Range, Nevada, as well as a way to equilibrate crustal thicknesses across a Proterozoic crustal boundary (MacCready et al., 1997). Lower crustal flow would constrain the lithosphere's thermal state as Triassic extension was imposed. Boundary driven gravity spreading requires that plate boundary kinematics control extension of the hot, weak lithosphere (Tirel et al., 2006). Recently, Clendenin et al. (2012) proposed that Permo-Triassic, circum-Pacific subduction-related tectonics along the western margin of Pangea induced that boundary drive. Acknowledgments This study was supported by the U. S. Department of Energy under Award Number DE-FE0001965 to characterize the South Georgia rift. Jack Garihan, David Heffner, and David Prowell are thanked for interesting discussions. Kerry Castle, Jack Garihan, P.F. Rey, Christian Tyssier, R.B. Hawman, and an anomalous reviewer also are thanked for thoughtful and constructive comments. Matthew Henderson is thanked for graphics. References Bertotti, G., Podladchikov, Y., Daehler, A., 2000. Dynamic link between the level of ductile crustal flow and style of normal faulting of the brittle crust. Tectonophysics 320, 195–218. Block, L., Royden, L.H., 1990. Core complex geometries and regional scale flow in the lower crust. Tectonics 9, 557–567. Brun, J.-P., Sokoutis, D., Van Den Driessche, J., 1994. Analogue modeling of detachment fault systems and core complexes. Geology 22, 319–322. Buck, R., 1991. Modes of continental lithospheric extension. J. Geophys. Res. 96, 161–178. Chowns, T.M., 2009. The Riddleville basin, Mesozoic rifting and Suwannee–Wiggins suture. Georgia Geol. Soc. Guideb. 29, 71–78. Chowns, T.M., Williams, C.T., 1983. Pre-Cretaceous rocks beneath the Georgia Coastal Plain — regional implications. In: Gohn, G.S. (Ed.), Studies Related to the Charleston, South Carolina, Earthquake of 1886 — Tectonics and Seismicity. U.S. Geol. Surv. Prof. Paper, 1313, pp. L1–L42. Clendenin, C.W., Waddell, M.G., Addison, A.D., 2011. Reactivation and overprinting of South Georgia rift extension. Geol. Soc. Am. Abstr. Programs 43 (5), 551. Clendenin, C.W., Waddell, M.G., Addison, A.D., 2012. Triassic rifting in western Pangea. Geol. Soc. Am. Abstr. Programs 44 (7), 286. Cook, F.A., Brown, L.D., Kaufman, S., Oliver, J.E., Petersen, T.A., 1981. COCORP seismic profiling of the Appalachian orogen beneath the Coastal Plain of Georgia. Geol. Soc. Am. Bull. 92, 738–748. Corti, G., Bonini, M., Conticelli, S., Innocenti, F., Manetti, P., Sokoutis, D., 2003. Analogue modeling of continental extension: a review focused on the relations between the patterns of deformation and the presence of magma. Earth Sci. Rev. 63, 169–247. Daniels, D.L., 2001. Georgia aeromagnetic and gravity maps and data: a web site for distribution of data. USGS Open-file Report 01-0106 (http://pubs.usgs.gov/openfile/ of01-106/). Daniels, D.L., Zietz, I., Popenoe, P., 1983. Distribution of subsurface Lower Mesozoic rocks in the southeastern United States as interpreted from regional aeromagnetic and gravity. In: Gohn, G.S. (Ed.), Studies Related to the Charleston, South Carolina, Earthquake of 1886 Tectonics and Seismicity. U.S. Geol. Surv. Prof. Pap., 1313, pp. K1–K24. Heck, F.A., 1989. Mesozoic extension of the southern Appalachians. Geology 17, 711–714. Hopper, J.R., Buck, W.R., 1996. The effect of lower crustal flow on continental extension and passive margin formation. J. Geophys. Res. 101, 20,175–20,194. Jolivet, L., Labrousse, L., Agard, P., Locombe, O., Bailly, V., Lomonte, E., Mouthereau, F., Mehl, C., 2010. Rifting and shallow-dipping detachments, clues from the Corinth Rift and the Aegean. Tectonophysics 483, 287–304. Keranen, K.M., Kleperer, S.L., Julia, J., Lawrence, J.F., Nyblade, A.A., 2009. Low lower crustal velocity across Ethiopia: is the Main Ethiopian Rift a narrow rift in a hot craton? Geochem. Geophys. Geosyst. 10. http://dx.doi.org/10.1029/2008GC002293. Le Pourheit, L., Burov, E., Moretti, 2004. Rifting through a stack of inhomogeneous thrusts (the dipping pie concept). Tectonics 23, TC4005. Lister, G.S., Davis, G.A., 1989. The origin of metamorphic core complexes and detachment faults formed during Tertiary continental extension in the northern Colorado River region, U.S.A. J. Struct. Geol. 11, 65–94. Liu, M., 2001. Cenozoic extension and magmatism in the North American Cordillera: a role of gravitational collapse. Tectonophysics 342, 407–433.
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621 MacCready, T., Snoke, A.W., Wright, J.E., Howard, K.A., 1997. Mid-crustal flow during Tertiary extension in the Ruby Mountain core, complex, Nevada. Geol. Soc. Am. Bull. 109, 1576–1594. Mattioni, L., Le Pourhiet, L., Moretti, I., 2006. Rifting through a heterogeneous crust: Insights from analogue models and application to the Gulf of Corinth. In: Butler, S.J.H., Schreurs, G. (Eds.), Analogue and Numerical Modelling of Crustal-scale Processes. Geol. Soc. London Spec. Pub, 253, pp. 213–231. McBride, J.H., 1991. Constraints on the structure and tectonic development of the early Mesozoic South Georgia Rift, southeastern United States; seismic reflection data processing and interpretation. Tectonics 10, 1065–1083. McBride, J.H., Nelson, K.D., Brown, L.D., 1987. Early Mesozoic basin structure and tectonics of the southeastern United States as revealed from COCORP reflection data and the relation to Atlantic rifting. In: Beaumont, C., Tankard, A.J. (Eds.), Sedimentary Basins and Basin-forming Mechanisms. Can. Soc. Petrol. Geo. Mem, 12, pp. 173–184. McCarthy, J., Parsons, T., 1994. Insights into the kinematic evolution of the Basin and Range—Colorado Plateau transition from coincident seismic refraction and reflection data. Geol. Soc. Am. Bull. 106, 747–759. McClay, K.R., Buchanan, P.G., 1992. Thrust faults in inverted extensional basins. In: McClay, K.R. (Ed.), Thrust Tectonics. Chapman & Hill, London, pp. 93–104. McKenzie, D., Nimmo, F., Jackson, J.A., Gans, P.B., Miller, E.L., 2000. Characteristics and consequences of flow in the lower crust. J. Geophys. Res. 105, 11,029–11,046. Moy, C., Traverse, A., 1986. Palynostratigraphy of the subsurface Eagle Mills Formation from a well in east-central Texas. Palynology 10, 225–234. Petersen, T.A., Brown, L.D., Cook, F.A., Kaufman, S., Oliver, J.E., 1984. Structure of the Riddleville basin from COCORP data and implications for reactivation tectonics. J. Geol. 92, 261–271. Pindell, J.L., Kennan, L., 2009. Tectonic evolution of the Gulf of Mexico, Caribbean and northern South America in the mantle reference frame: an update. In: James, K.H., Lorente, M.A., Pindell, J.L. (Eds.), The Origin and Evolution of the Caribbean Plate. Geol. Soc. London Spec. Pub, 328, pp. 1–55. Rey, P., Vanderhaeghe, O., Teyssier, C., 2001. Gravitational collapse of continental crust: definition, regimes, and modes. Tectonophysics 342, 435–449.
621
Rey, P.F., Teyssier, C., Whitney, D.L., 2009. Extension rates, crustal melting, and core complex dynamics. Geology 37, 391–394. Reynolds, S.J., Lister, G.S., 1990. Folding of mylonitic zones in Cordilleran metamorphic core complexes: evidence from near the mylonitic front. Geology 18, 216–219. Sartain, S.M., See, B.E., 1997. The South Georgia basin: an integration of landsat, gravity, magnetic and seismic data to delineate basement structure and rift basin geometry. Gulf Coast Assoc. Geol. Soc. Trans. 47, 493–498. Tauvers, P.R., Muehlberger, W.R., 1987. Is the Brunswick magnetic anomaly really the Alleghanian suture. Tectonics 6, 331–342. ten Brink, U.S., Zhang, J., Brocher, T.M., Okaya, D.A., Klitgord, K.D., Fuis, G.S., 2000. Geophysical evidence for the evolution of the California Inner Continental Borderland as a metamorphic core complex. J. Geophys. Res. 105, 5835–5857. Tew, B.H., Mink, R.M., Mann, S.D., Bearden, B.L., Mancini, E.A., 1991. Geologic framework of Norphlet and pre-Norphlet strata of the onshore and offshore eastern Gulf of Mexico area. Gulf Coast Assoc. Geol. Soc. Trans. 41, 590–600. Thompson, G.A., McCarthy, J., 1990. A gravity constraint on the origin of highly extended terranes. Tectonophysics 174, 197–206. Tirel, C., Brun, J.-P., Sokoutis, D., 2006. Extension of thickened and hot lithospheres: inferences from laboratory modeling. Tectonics 25, TC1005. Traverse, A., 1987. Pollen and spores date origin of rift basins from Texas to Nova Scotia as early Late Triassic. Science 236, 1469–1472. Van Wijk, J.W., Blackman, D.K., 2005. Dynamics of continental rift propagation: the end member modes. Earth Planet. Sci. Lett. 229, 247–258. Whitney, D.L., Teyssier, C., Rey, P., Buck, W.R., 2013. Continental and oceanic core complexes. Geol. Soc. Am. Bull. 125, 273–298. Wijns, C., Weinberg, R., Gessner, K., Moresi, L., 2005. Mode of crustal extension determined by rheological layering. Earth Planet. Sci. Lett. 236, 120–134. Withjack, M.O., Schlische, R.W., Olsen, P.E., 1998. Diachronous rifting, drifting, and inversion on the passive margin of eastern North America: an analogue of other passive margins. Am. Assoc. Petrol. Geol. Bull. 82, 817–835. Ziegler, P.A., Cloetingh, S., 2004. Dynamic processes controlling evolution of rifted basins. Earth Sci. Rev. 64, 1–50.
Contents lists available at ScienceDirect
Tectonophysics journal homepage: www.elsevier.com/locate/tecto
Insights into the mode of the South Georgia rift extension in eastern Georgia, USA C.W. Clendenin Jr. ⁎ South Carolina Department of Natural Resources—Earth Science Group, 5 Geology Road, Columbia, SC 29212, USA
a r t i c l e
i n f o
Article history: Received 10 April 2013 Received in revised form 9 August 2013 Accepted 17 August 2013 Available online 28 August 2013 Keywords: Triassic extension Mode of rifting Core complex Lower crustal flow Flat Moho
a b s t r a c t The South Georgia rift (SGR) lies oblique to the east coast margin of North America and across the Alleghenian suture between Laurentia and Africa in southern Georgia. Regionally, the SGR can be divided into a southwest compartment and a northeast compartment across the Jacksonville structure that is located in the vicinity of that suture. Analytical and numerical models are used to characterize the mode of rifting in the northeast compartment. Borehole, COCORP seismic, and regional geophysical information from the compartment, that were used previously to infer the geometry of the basin, are reassessed with the use of those models to analyze the lithospheric conditions influencing Triassic extension. This approach led to the interpretation of core complex mode extension and to the proposal of a model of progressive rifting. The model shows how the Riddleville and Main SGR basins are associated and how changes in structural style of those two basins resulted from changing lithospheric conditions during extension. The core complex model also indicated that extension was influenced by distributed deformation of a younger, warmer, and less stable lithosphere adjacent to the Permian suture; whereas extension in other east coast rifts that lie subparallel to structural fabric was probably localized by preexisting zones of weakness. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Continental extension is characterized by differences in structural styles, width, and magmatic processes. These differences are the result of lithospheric conditions prior to and during extension (ten Brink et al., 2000). Buck (1991) emphasized that geothermal state and crustal thickness are the dominant lithospheric conditions influencing continental extension and described three modes of rifting: narrow, wide, and core complex. Modeling also shows that old, thick, and cold crusts tend to yield narrow rifts; whereas extension of younger, thinner, and warmer lithospheres give wider rifts, in which core complexes may develop (Corti et al., 2003). Keranen et al. (2009), however, demonstrated that preexisting zones of weakness influence rift mode under different conditions, if the prevailing stress field and preexisting weaknesses are properly oriented. In nature, the resulting structural style, therefore, should be considered a function of heat flow, crustal age, crustal thickness, and orientation of preexisting weakness (Buck, 1991; Corti et al., 2003; Keranen et al., 2009). Van Wijk and Blackman (2005) also point out that structural style varies along the rift axis during extension. Appreciation of the described relations is important when one tries to characterize the tectonic development of the South Georgia rift (SGR). The SGR is the southernmost Triassic basin along the east coast of North America and is buried under Atlantic Coastal Plain sediments (Fig. 1). Being buried has hampered the understanding of the SGR; and what little is known has come from borehole, regional geophysical ⁎ Tel.: +1 803 896 7702. E-mail address: [email protected]. 0040-1951/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.tecto.2013.08.019
information, and a few seismic images. Interest was reignited in 2010 when the U.S. Department of Energy funded a characterization study of the basin. However, borehole samples that could have provided a better understanding were lost with the closure of the Georgia State Geological Survey. Loss of those samples forced this part of the characterization study to rely on regional relations, published data, and a priori evidence. In this paper, subcrop maps, seismic images, and regional geophysical patterns of the SGR in eastern Georgia are reexamined, and the information is compared to analytical and numerical models to decipher structural style. This information is used to interpret mode of rifting, and then a progression rifting is proposed for the tectonic development of the SGR in eastern Georgia. The regional implications of that model are discussed to help characterize what may have influenced the mode of rifting. 2. Triassic precursor basins Regional relations show that two series of Triassic rift basins formed in western Pangea prior to the opening of the central Atlantic Ocean and that these rift series are time equivalent (Traverse, 1987). Basins found along the east coast of North America commonly are exposed on the surface, are subparallel to structural fabric, and are referred to here as the Newark series (Fig. 1). The Newark series basins are characterized by elongate half-grabens, bounded by listric-like faults lying subparallel to the coast (Withjack et al., 1998). Although the SGR is oblique to the coast, published cross sections also tend to depict a half-graben structural style (McBride et al., 1987; Withjack et al., 1998).
614
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
Fig. 1. Reference map showing seismic lines L5, L8, and SCCO21; Fall Line, and other features. Newark series basins are shown to north-northeast on reference map. NB: Newark basin; FVT: Felsic Volcanic Terrane; JS: Jacksonville structure (modified after Chowns, 2009).
Less attention has been given to the second series of Triassic basins that is buried in the Gulf Coast area and referred to here as the Eagle Mills series (Fig. 2). This series of precursor basins begins offshore of Florida immediately southwest of the SGR and extends in the subsurface northwest across Alabama–Mississippi, west across southern Arkansas– northern Louisiana, and then southwest into eastern Texas (Moy and Traverse, 1986; Pindell and Kennan, 2009; Tew et al., 1991). Pindell and Kennan (2009) also have correlated another set of basins off the northeast coast of Yucatan with this series. Structural style is not known, but regional distribution and limited seismic information suggest a horst-and-graben pattern (McBride, 1991; Pindell and Kennan, 2009; Sartain and See, 1997). 3. SGR in eastern Georgia 3.1. Borehole information Chowns and Williams (1983) produced a subcrop map of the SGR over parts of Florida, Alabama, Georgia, and South Carolina (Fig. 1).
The map shows that the SGR is divided into northeast and southwest compartments by the Jacksonville transfer fault system in southern Georgia (Tauvers and Muehlberger, 1987). The existence of the Jacksonville transfer system is considered controversial, but the polarity of extension shifts from down-to-the-northwest in the southwest compartment (McBride, 1991) to down-to-the-southeast in the northeast compartment (Cook et al., 1981) across the mapped faulting. A change in polarity implies transfer faulting or an accommodation zone. An investigation of the structure is beyond the scope of this study; and for the time being, it simply is referred here as the Jacksonville structure. In southern Georgia, the southwest compartment is characterized by horsts and grabens (McBride, 1991; Sartain and See, 1997; Fig. 1). In eastern Georgia, the northeast compartment consists of two basins separated by a horst block. The northern basin, identified from aeromagnetic information, is known as the Riddleville basin (Daniels et al., 1983). A borehole near its northwest margin penetrated 2522 m of Triassic sediments (Chowns, 2009). Depth-to-basement rocks estimated from magnetic data suggest a down-to-the-southeast asymmetric geometry (Daniels et al., 1983). The southern basin simply is referred
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
615
Fig. 2. Buried, Triassic Eagle Mills series basins south and west of SGR. JS: Jacksonville structure; BFZ: Bahamas fracture zone; DS: De Soto basin; EM: Eagle Mills basin; YB: Yucatan basins; A–O–M: Alleghenian–Ouachita–Marathon suture (modified after Chowns and Williams, 1983; Moy and Traverse, 1986; Pindell and Kennan, 2009).
to as Main SGR, and analysis of magnetic data indicates that ~3500 m of Triassic sediments are preserved (Daniels et al., 1983). Heck (1989) identified the southeast margin to be a high-angle fault, and spatial patterns of boreholes suggest that Main SGR is a full graben. The intervening horst consists of Piedmont rocks; boreholes in Treutlen County, Georgia, bottom in either biotite gneiss or cataclastic garnetiferous quartz-feldspathic gneiss (Chowns and Williams, 1983). Although the subcrop map shows closure to the northeast, Daniels et al. (1983) extended the horst northeast into South Carolina. Immediately south of Main SGR is a felsic volcanic terrane (Fig. 1). Boreholes show that principal lithologies are vitric crystal tuff, tuffaceous arkose, pyroclastic porporphyric rhyolite, and granite (Chowns and Williams, 1983). Radiometric ages of the volcanics cover an extended period of time, with two main groups from 500 to 350 Ma and 300 to 150 Ma (see Chowns and Williams, 1983). Basalt and diabase from boreholes also show another group of dates ranging from 300 to 50 Ma.
3.2. Seismic reflection data Recently, the U.S. Department of Energy sponsored 240 km of new seismic reflection data over the South Carolina portion of the SGR. SCCO2-1 clearly images the northwest margin of the rift (Clendenin et al., 2011; Fig. 1). Faults delineating that margin are inverted as harpoon-style reverse faults (see McClay and Buchanan, 1992) that look like faulted anticlines (SP 360, Fig. 3). The largest of the inverted structures lies downdip along the margin just below 3 s. The precursor extension faults are interpreted to have been steep, planar features. After review of the other project seismic lines, it was discovered that, unless the line was oriented nearly orthogonal to the margin, reactivation tended to mask structures on the image. The SGR's inverted margin in South Carolina is northeast along strike of the northwest margin of Main SGR in eastern Georgia. Recognition of inversion initiated a reexamination of COCORP Line 5 (L5) and L8 (Figs. 1, 4) that are orthogonal to the margin. A faulted anticline is clearly visible on COCORP Line 8 (L8) (Fig. 4), and the structure imaged was described in those terms by Cook et al. (1981). A reexamination also was needed because, when the available information was initially interpreted in the early 1980's, contemporary thinking tended to be focused on only the upper part of the crust and did not always consider to
what degree the entire crust was affected by extension (Rey et al., 2001). The southeastern half of L5 and all of L8 lines extend southeast from the Fall Line, the northwest edge of the Coastal Plain, to within 30 km of Savannah, Georgia. L8 is nearly parallel to the southeast end of L5 and is offset 24 km to the west (Petersen et al., 1984). Interpretations of L5 and L8 have been published in Cook et al. (1981), Petersen et al. (1984), McBride et al. (1987), Heck (1989), and others. Shown here is the original interpretive section of Cook et al. (1981). The reflectors (R) of interest are labeled A through F in Fig. 4 and are referred to in the text as RA through RF. RA begins near SP 1000 on L5, dips to the southeast, and extends to ~5 s at SP 1600. Cook et al. (1981) correlated RA with the Augusta fault. RB begins at SP 1500, dips southeast, and projects to ~5 s at SP 1600 on L5. This reflector is below the Triassic Riddleville basin and bounds the basin on the northwest. Ramping reflectors that join RB between SPs 1500 and 1800 suggest a faulted anticline. The down-dip trace of RB appears to cut RA; however, in Petersen et al. (1984) Fig. 3, RA actually appears to be cut or to merge with their reflector F that lies between RB and RA. Both RB and F project toward the northwest margin of the Riddleville basin, have a quasi-listric shape, and are interpreted to be Triassic faults (Petersen et al., 1984). RC underlies both RA and RB. The three reflectors appear to merge near SP 1600, with RC continuing onto L8. RC consists of fairly flat, discontinuous, layered reflectors that can be traced at ~5 s across L5. Examination of L5 (see Cook et al., 1981) shows that RC dips gently to the southeast from ~4 s on L5 to ~6 s on L8. Cook et al. (1981) suggested that RC represents the extension of a decollement to the northeast under the Piedmont Charlotte Belt. Heck (1989), however, proposed RC to be the brittle–ductile transition (BDT). RD begins at about SP 100 on L8 and is marked by northwestdipping reflectors beginning at ~7 s. The reflectors warp up to ~4 s at about SP 400 before reversing dip to the southeast. After reversing dip, the discontinuous reflectors of RD can be traced southeast to ~7 s at about SP 700. Reversal produces an antiformal geometry that may mark a major crustal boundary (Cook et al., 1981). It is interesting that, in different presentations of L8, no reflectors are shown above RD (see Cook et al., 1981; Heck, 1989; Petersen et al., 1984). Cook et al. (1981), however, show that fine-scale layering overlies RD in a seismic line shot perpendicular to L8.
616
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
Fig. 3. Northwest margin of SGR in South Carolina as imaged by SCCO2-1. Dotted line marks rift margin. View is to the northeast.
RE lies above 5 s between SPs 600 and 800 and is a distinct set of antiformal-looking reflectors with a recognizable southeast-dipping limb. RE is juxtaposed over the lower southeast portion of RD. Cook et al. (1981) interpreted a steeper, southeast-dipping reflection at SP 600 as a fault plane and proposed that RE represents a faulted anticline of Paleozoic age. RF, interpreted by previous workers as the Moho, can be traced across L5 onto L8 at ~10 s. The discontinuous reflector has a slightly undulating appearance. Cook et al. (1981) show some minor relief on the
Moho, whereas others interpret it to be flat (Heck, 1989; McBride et al., 1987; Petersen et al., 1984). 3.3. Aeromagnetic and Bouguer-gravity anomalies In their Fig. 6, Daniels et al. (1983) show an aeromagnetic profile and interpreted geologic cross section that depicts the deeper Main SGR separated from the Riddleville basin by an intervening high. The northwest half of Fig. 5 illustrates that aeromagnetic profile. The northwest margin
Fig. 4. Interpretive cross section of Lines 5 and 8. View is to the northeast. Reflectors abbreviated RA–RF are discussed in text (modified after Cook et al., 1981).
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
of Main SGR is easily recognized as a strong northeast-striking lineament on aeromagnetic maps of Georgia (Daniels, 2001). This lineament correlates with marginal structures seen on L8 (RE) and SCCO2-1 (Fig. 1) and with the southeast margin of a 15–20 km wide, northeast-trending, small ridge-like Bouguer-gravity high (Fig. 6). Although overall patterns infer an east–west orientation, magnetic highs can be aligned with the northeast-trending Bouguer-gravity ridge. Both aeromagnetic and Bouguer-gravity maps show a broad, and somewhat featureless low southeast of the lineament, that marks Main SGR. Northwest of the Bouguer-gravity ridge is a Bouguer-gravity low that is bounded on the north by a major east–west regional lineament (Daniels et al., 1983; Fig. 6). The lineament corresponds to the Magruder fault. The Bouguer-gravity low is interpreted to be the Riddleville basin. The basin, however, is vague on aeromagnetic maps and is marked by a featureless magnetic low that extends east from Riddleville, Georgia (Daniels et al., 1983), along the southern margin of the Magruder fault. The aeromagnetic low also is northwest of the Bouguer-gravity ridge. The exact orientation of the Riddleville basin is not well constrained by the aeromagnetic or Bouguer-gravity patterns. The juxtaposition to the east–west Magruder fault gives the impression of an east–west basin (Fig. 6). Geophysical lows on the south side of that fault, however, may be overprinting the Riddleville footprint. A pronounced northeasttrending Bouguer-gravity low that lies adjacent to the northwest margin of the Bouguer-gravity ridge may actually delineate the basin. This Bouguer-gravity low correlates with aeromagnetic signals that Daniels et al. (1983) interpret as the basin. It should be noted that circular and elliptical aeromagnetic and Bouguer-gravity highs lie to the northeast and southwest of the northeast compartment of the SGR (Fig. 6). The different geophysical highs have equally large amplitudes, and Daniels et al. (1983) have interpreted these highs as unmetamorphosed mafic–ultramafic intrusive complexes. Although direct evidence of age of emplacement is unavailable, a priori evidence led Daniels et al. (1983) to infer that the circular magnetic-Bouguer gravity anomalies may have intruded in the lower Mesozoic. 4. Discussion In overview, the Riddleville and Main SGR basins are depicted as about the same size northeast of the Jacksonville structure, with Main SGR expanding in width to the northeast (Fig. 1). The bore-hole data (see Chowns, 2009), however, show that the width of Main SGR would increase if drawn using the half-way method of unknowns. Geophysical evidence also clearly shows that Main SGR is the larger of the two basins (see Daniels et al., 1983). The Piedmont horst also would be narrower using the described method. Combining Daniels et al.'s (1983) aeromagnetic profile with Heck's (1989) interpretation of the southern margin of Main SGR produces the cross section of the two basins shown in Fig. 5. The COCORP seismic information can be interpreted differently as well (Fig. 4). RA may have had little or no influence on rifting and is not discussed further. RB and the warped reflectors are interpreted to be a harpoon-style inversion fault. Shape and location of inversion could have easily been interpreted as reverse drag on a normal fault. As stated in Section 1, older conventional thinking tended to focused
617
on the upper crust, and Cook et al. (1981) point out that data were discussed in reference to surface geological nomenclature. Petersen et al. (1984) suggested that normal fault motion on RB must have been taken up along a portion of underlying RA. Instead, models by Reynolds and Lister (1990) suggest that RA was captured and displaced by RB. Capture by RB is probable because little interaction occurs between normal faults and shallow-dipping, preexisting zones of weakness (Mattioni et al., 2006). RB bounds the Riddleville basin's northwest margin and is interpreted to be an inverted, Triassic-age fault (Fig. 4). Heck (1989) interpreted RC to be the BDT (Fig. 4). Lister and Davis (1989) have argued against such interpretations because the BDT would not remain fixed to a particular plane during progressive extension. If RC is a subhorizontal decollement, it represents an inherited heterogeneity in the crust, that traces its origins to a Paleozoic orogeny (Cook et al., 1981). Jolivet et al. (2010) envision that inherited heterogeneities, like RC, should lie above and in the BDT, with cataclastic flow at the top and ductile flow at the base. The horst consists of Piedmont rocks. In overview, the Piedmont is a polyphase-deformed terrane, characterized by thin-skinned thrusting (Cook et al., 1981). A shallow-dipping crustal heterogeneity may have either a high- or low-competency contrast with an overlying thrust package (Le Pourheit et al., 2004). If this type of weakness is involved in active extension, it may strongly localize faulting into a narrow zone and influence mode of rifting (Le Pourheit et al., 2004; Mattioni et al., 2006). Faults developed in an overlying thrust package root in such crustal heterogeneities, and some of the faults may intersect each other (Mattioni et al., 2006). Such relations can be seen in Fig. 4. RF has been interpreted as a flat, relatively shallow Moho. If true, the flat Moho indicates that the crust was sufficiently hot and thick enough to allow lower crustal flow (Block and Royden, 1990). Flow is so efficient that long-wave topography on the Moho is effectively removed (McKenzie et al., 2000). Flow also keeps crustal thickness roughly constant and distributes deformation (Block and Royden, 1990). Models show that flow starts before faulting in the upper crust (Tirel et al., 2006) and that extension forms a narrow graben. With increased extension, flow moves into the locally thinned upper crust and restricts migration of deformation (Le Pourheit et al., 2004). Flow is always toward and into the thinned area (Hopper and Buck, 1996; Rey et al., 2009). In flow mode, thinning is significantly lower than that of extension (Bertotti et al., 2000). Although some of the felsic volcanics may be related to the younger, mafic Central Atlantic Magmatic Province in eastern Georgia (Chowns and Williams, 1983), the presence of felsic volcanics immediately to the south (Fig. 1) is considered a priori evidence of an increased local geotherm. Chowns and Williams (1983) have offered a number of interpretations for the felsic volcanic rocks to be either early Paleozoic or Proterozoic in age. Age of the rocks is not that important because Ziegler and Cloetingh (2004) point out that heat flow even in older orogenic belts is still elevated because of the younger age of their mantle– lithosphere and possibly to higher radiogenic heat generation potential of their crust. Heat flow is important in extension, and part of that heat could have come from contemporaneous volcanism. Evidence of this volcanism could be clasts of felsic volcanic debris that are common in inferred Triassic-age litharenites (Chowns and Williams, 1983). Mixing of
Fig. 5. Cross section of Riddleville and Main SGR basins constructed from Daniels et al. (1983) and Heck (1989). View is to the northeast. RB — Riddleville basin; CP — Coastal Plain.
618
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
Fig. 6. Bouguer gravity anomalies related to the SGR. Small Bouguer ridge is bounded by magnetic lineaments that define northwest margins of Riddleville and Main SGR basins. Seismic lines also are shown for reference. MF — Magruder Fault (modified after Daniels, 2001).
sediment and volcanic debris is not surprising in an extensional environment. In the Basin and Range province, such explosive volcanics have erupted contemporaneously in areas that were undergoing localized extension (McKenzie et al., 2000). Some combination of possible magmatic activity, presence of underplating mafics, and thickened crust in the vicinity of the suture with extension starting soon after collision implies that the general area had an increased geotherm. As a result, a partially molten layer may have been present in the lower crust (see Rey et al., 2009). The presence of heat would have modified the thermal field and promoted partial melting of warm, thickened crust. Liu (2001) also noted that, in the Great Basin, extension is near margins of the volcanic field where the crust was weakest. The SGR is juxtaposed to the FVT (Fig. 1); and rheologic contrast between the two areas probably had some influence on the spatial location of Triassic extension. Slow extension with partial melt enhances upward exhumation of hot deep rocks (Rey et al., 2009). Crystallization of rising partial melt in the subsurface produces an antiformal or domal geometry that represents preferential exhumation (see Rey et al., 2009). RD fits that geometry and is interpreted to be the upper limit of exhumation of partially molten lower crust. If correct, this implies that the Piedmont horst has a migmatitic core. The northeast-striking, Bouguer-gravity ridge (Fig. 6) and the aeromagnetic high separating the Riddleville basin and Main SGR are thought to be imaging the suspected migmatites. Thompson and McCarthy (1990) point out that gravity anomalies associated with highly extended terranes, including core complexes, show no prominent, characteristic gravity signature and any anomaly would be small. The likely explanation for those observations is that compensation has taken place by emplacement of material of crustal density (Thompson and McCarthy, 1990). Emplacement would further suggest that products of crustal melting and mixing with basaltic magma rose to intermediate or high levels in the crust, and this relation is supported by flat Moho (Thompson and McCarthy, 1990). Cataclasites mark the Piedmont horst (Chowns and Williams, 1983). The presence of garnet in the described cataclasite clearly differentiates it from the silicified cataclasites commonly mapped in the proximity of
faults in the Piedmont (Garihan, 2013, pers. commun.). The gneissic character of the rock also implies mylonitization, whereas the Piedmont silicified cataclasites are brecciated quartz bodies. Cataclasites form in a special sequence close to and directly above the BDT (Lister and Davis, 1989). A simple interpretation is that the garnet cataclasite is part of a preexisting shear zone captured and exhumed from deeper levels (Reynolds and Lister, 1990). A second interpretation is that the garnet cataclasite is part of a detachment fault that had been warped over a growing dome. The second interpretation is preferred here because isostatic warping of a detachment fault is common in exhumation models (Brun et al., 1994). The lack of reflectors above RD and the fine-scale layering imaged by Cook et al. (1981) on another seismic line support the idea of a core complex and doming. McCarthy and Parsons (1994) describe the seismic signature of a core complex as characterized by a pervasive, weak, laterally discontinuous reflective fabric represented by abundant fine-scale layering. RE was interpreted by Cook et al. (1981) as related to Paleozoic compression. Heck (1989) disagreed and pointed out that younger tectonic events should be considered in interpretations of seismic images from this area. If Paleozoic in age, RE should have been reactivated as a normal or normal-oblique fault by the Triassic extensional stress field. The shape of RE, however, clearly fits the characteristics of a harpoonstyle structure. Inversion could have resulted from an even younger tectonic event, e.g. Atlantic tectonics or Late Mesozoic–Early Cenozoic collision of Cuba and the Bahamas. If true, any younger, properly oriented compressive stress field would have inverted both Main SGR and the Riddleville basin. Accepting RE as a harpoon-style structure, its predecessor had to have been a younger, down-to-the-southeast normal fault. The location of RE over the southeast flank of RD further implies that the youngerplanar fault formed during progressive extension (Fig. 4). The progression began as an older detachment fault was rotated, or warped, to a lower dip; and as the dip forced abandonment, the younger highangle, planar fault formed in the hanging wall (see Jolivet et al., 2010). During those related processes, thinning and progressive cooling would have resulted in a downward migration of BDT (Tirel et al., 2006). If RE is an arched continuation of RC, younger-normal faulting
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
cut through the heterogeneity and rooted on a deeper BDT. This interpretation is consistent with Cook et al.'s (1981) statement that decollements could extend eastward as far as the present continental shelf. The width and thickness of sediments also imply that younger faulting localized in Main SGR, as proposed by Jolivet et al. (2010) for the Corinth Rift. Footwall rebound also would have helped exhume the Piedmont horst. With continuing extension, detachment faults that characterize core complex extension are deactivated, and extension becomes characterized by steep-planar faults that cut the older features (Bertotti et al., 2000). This change in structural style has been proposed as the result of the exhaustion of crustal material that can flow and of the increasing distances crustal material had to flow (Bertotti et al., 2000). A complementary explanation may be that crustal flow occurred only until the hot rocks reached shallower depths in the rising dome and cooled (Whitney et al., 2013). Either explanation would result in a change from lower crustal flow to no-flow mode lithospheric conditions. With a shift to no-flow mode, thinning becomes directly proportional to extension that causes subsidence (Bertotti et al., 2000); subsidence was expressed as a deepening of Main SGR. Faulting of older detachment structures with continuing extension produces a relevant modification to structural style with subsequent normal faults steep, planar structures extending down to the relocated BDT (Bertotti et al., 2000; Jolivet et al., 2010). Differences in structure style also may be the reason that the Riddleville basin (half-graben) and Main SGR (McKenzie-type full graben; Fig. 4) have not been structurally related in the past. On the basis of the described subcrop and geophysical relations, core complex mode rifting is interpreted to have occurred between the Jacksonville structure and the Savannah River in eastern Georgia (Fig. 1). The northeast-striking horst would define the subcrop of the domed, higher-grade metamorphic rocks. The structure is referred to here as the Treutlen County core complex, after the location where the garnet cataclasite was recovered.
619
lower crustal strength ratio (Wijns et al., 2005). Localized thinning of the brittle crust was associated with lower crustal flow that kept the Moho rather flat (Corti et al., 2003). Localization of early extension was expressed as only a few, major normal faults (Brun et al., 1994; Whitney et al., 2013; Wijns et al., 2005). A slow extension rate and a higher geotherm were the determining factors in the number of major faults developed (Tirel et al., 2006), which are shown here as a few southeast-dipping and northwestdipping conjugate-fault pairs rooted initially in RC (Fig. 7a). Three faults are designated: the farthest northwest fault (RB); the most interior southeast-dipping fault (SDF); and the farthest southeast, northwestdipping fault (NDF). It should be noted that, throughout the description of progression of rifting, two-letter abbreviations refer to features in Fig. 4. Rebound of SDF's footwall provided space for lower crustal flow that was moving toward the extended area; described flow and decoupling from the mantle kept the Moho flat (RF, Fig. 7b). Movement on RB and flow produced counterclockwise block rotation between RB and SDF that initiated the Riddleville basin and assisted exhumation by increasing footwall space. As extension continued, exhumation was determined by a feedback system between upper crustal extension and necessity for flow to fill the thinned zone (Rey et al., 2009; Fig. 7c). Continued movement on RB and rigid-body rotation of the rising dome deepened the Riddleville basin. During this time, RC continued to act as a decollement as the hanging-wall block bound by NDS shifted slightly down the shallow-dipping heterogeneity RC (Mattioni et al., 2006). Shifting resulted in asymmetric widening of Main SGR. Minor faults
5. Progression of rifting Interpretation of the Treutlen County core complex, presence of horsts and grabens in the southwest compartment, and a flat Moho show that distributed deformation influenced kinematics and controls on the progression of rifting. Tirel et al. (2006) point out that high extension rates driven by free-gravity spreading, i.e. lithospheric extension under its own weight without any boundary limitation, produce a series of tilted blocks, whereas lower extension rates influenced by boundary controlled gravity spreading favor conjugate patterns and the development of horsts and grabens. Core complexes occur when a low viscosity, thick lower crust and a higher syntectonic geotherm are present (Corti et al., 2003; Whitney et al., 2013). Only a small number of faults are developed under lower extension rates (Tirel et al., 2006; Wijns et al., 2005). Under a higher geotherm, strongly localized extension in the upper crust also only produces a limited number of faults (Whitney et al., 2013). The lower crust will flow into and replace the thinned brittle crust and remove the effect of gravity forces, which restricts migration of deformation (Corti et al., 2003). Any proposed progression of rifting needs to satisfy those relations and to account for two basins. The geometry of those two basins also has to compare to that shown in Fig. 5 — a half-graben juxtaposed to a full graben. Brun et al.'s (1994) model that was expanded by Tirel et al. (2006), with a few other additions, is applied here. After Late Paleozoic collision and subsequent thermal relaxation of the thickened crust, Triassic strain was localized in weak crust adjacent to the felsic volcanic terrane (Fig. 1). Heat had modified or was modifying the thermal field; advection of heat promoted partial melting that further enhanced localization (Corti et al., 2003). Localization also may have resulted from a low-competency contrast between the overlying crust and RC (Le Pourheit et al., 2004) and an assumed large upper to
Fig. 7. Progression of rifting; a–c formed under lower crust flow mode and d under noflow mode. Arrows and X indicate either movement directions or deactivation. Views are to the northeast. NDF — north dipping fault; SDF — south dipping fault; RB — Riddleville basin; BDT — brittle ductile transitions; RA, RB, RD, RE, and RF — seismic features described in text.
620
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621
that developed between NDS and SDS rooted in RC and effectively thinned the overlying crust (Mattioni et al., 2006). As exhumation continued, SDF was progressively warped and folded over the rising dome (Fig. 7c). Partial melt flowed upward (Rey et al., 2009), and the relative rate of cooling vs. exhumation influenced the location of RD (Wijns et al., 2005). Cooling and downward migration of the BDT accompanied that deformation. A deeper BDT and exhaustion/cooling of crustal material able to flow forced faulting on RE to cut deeper into the crust as movement on SDF was abandoned (Fig. 7d). Deep, planar faulting deactivated RC as faulting rooted in the deeper BDT. The exhumed dome also enhanced lateral partitioning of deformation (Tirel et al., 2006). Deformation was expressed by larger finite extension that deepened Main SGR as deformation changed to the no-flow mode (Bertotti et al., 2000; Jolivet et al., 2010). 6. Regional implications Distributed deformation influenced by boundary controlled gravity spreading also has regional implications. A flat Moho below domains of distributed deformation has been described as a fundamental difference between wide rifts and narrow rifts (Tirel et al., 2006). The structural style of Newark-series basins clearly fits narrow mode rift characteristics of localized deformation. The narrow mode for the Newark-series basins would imply that the lithosphere extended along the east coast of Laurentia in Triassic was old, cool, and stable as compared to the lithosphere in eastern Georgia near the Permian suture. An alternative explanation is that reactivation of zones of preexisting weakness may have had a greater influence on the Newark-series basins than rheological stratification (see Keranen et al., 2009). The east coast of Laurentia was deformed through the Paleozoic by the Taconic, Acadian, and Alleghenian orogenies, and Newark-series basins are subparallel to regional structural fabric. As pointed out, heat flow even in older orogenic belts is still elevated (Ziegler and Cloetingh, 2004); but narrow rifts, e.g. the Main Ethiopian Rift, can form in hot, weak lithosphere if the primary control on mode of extension is preexisting lithospheric weakness (Keranen et al., 2009). Where rifting crosses regional structural fabric, however, preexisting lithospheric weakness cannot be reactivated by the prevailing stress field, and new faults develop (Ziegler and Cloetingh, 2004). The regional distribution of basins in the Eagle Mills series gives the impression that horsts and grabens characterize structural style (Fig. 2). Horst and grabens have been recognized in the southwest compartment across the Jacksonville structure (Sartain and See, 1997). Transfer faults form when rift propagation is blocked by the boundary of two geologic provinces (Van Wijk and Blackman, 2005); in this case, such a boundary would be the Permian Laurentia/Africa suture. Stalling of rift propagation results in a change from localized to distributed deformation, and the two deformation styles are balanced by the development of a transfer fault (Van Wijk and Blackman, 2005). If this is true, the location of the core complex in the northeast compartment also substantiates the Van Wijk and Blackman's (2005) interpretation that structural style varies along the rift axis during extension. Differences in deformation style across the Jacksonville structure help to further establish how structural style varies during extension. Corti et al. (2003) point out that, in the Basin and Range Province, rifting evolved in two stages: initial core complex development and later horse-and-graben structures. Bertotti et al. (2000) also describe that initial core complexes were overprinted by horst-and-graben structures by progressive extension in the North Tyrrhenian area between North Corsica and Tuscany. Lower crustal flow during these stages may have been efficient enough to prevent localized thinning that would lead to crustal separation (Bertotti et al., 2000) and would explain why the SGR is a failed rift. Accepting these relations along with the interpretation of the Treutlen County core complex indicate that the SGR is probably part of the Eagle Mills series as suggested by Gulf Coast geologists (Tew et al., 1991; Sartain and See, 1997; Fig. 2). The juxtaposition of the Eagle
Mills series basins to the Alleghenian suture further shows that extension was controlled by a younger, warmer, and less stable lithosphere thickened by the Permian contractional episode. The location of the Treutlen County core complex also may indicate direction of lower crustal flow. Northeastward flow may have equilibrated crustal thicknesses across the Permian suture that separated possibly thicker African crust from Laurentian crust. Directional flow initially was suggested as a response to greater extension to the north in the East Humboldt Range, Nevada, as well as a way to equilibrate crustal thicknesses across a Proterozoic crustal boundary (MacCready et al., 1997). Lower crustal flow would constrain the lithosphere's thermal state as Triassic extension was imposed. Boundary driven gravity spreading requires that plate boundary kinematics control extension of the hot, weak lithosphere (Tirel et al., 2006). Recently, Clendenin et al. (2012) proposed that Permo-Triassic, circum-Pacific subduction-related tectonics along the western margin of Pangea induced that boundary drive. Acknowledgments This study was supported by the U. S. Department of Energy under Award Number DE-FE0001965 to characterize the South Georgia rift. Jack Garihan, David Heffner, and David Prowell are thanked for interesting discussions. Kerry Castle, Jack Garihan, P.F. Rey, Christian Tyssier, R.B. Hawman, and an anomalous reviewer also are thanked for thoughtful and constructive comments. Matthew Henderson is thanked for graphics. References Bertotti, G., Podladchikov, Y., Daehler, A., 2000. Dynamic link between the level of ductile crustal flow and style of normal faulting of the brittle crust. Tectonophysics 320, 195–218. Block, L., Royden, L.H., 1990. Core complex geometries and regional scale flow in the lower crust. Tectonics 9, 557–567. Brun, J.-P., Sokoutis, D., Van Den Driessche, J., 1994. Analogue modeling of detachment fault systems and core complexes. Geology 22, 319–322. Buck, R., 1991. Modes of continental lithospheric extension. J. Geophys. Res. 96, 161–178. Chowns, T.M., 2009. The Riddleville basin, Mesozoic rifting and Suwannee–Wiggins suture. Georgia Geol. Soc. Guideb. 29, 71–78. Chowns, T.M., Williams, C.T., 1983. Pre-Cretaceous rocks beneath the Georgia Coastal Plain — regional implications. In: Gohn, G.S. (Ed.), Studies Related to the Charleston, South Carolina, Earthquake of 1886 — Tectonics and Seismicity. U.S. Geol. Surv. Prof. Paper, 1313, pp. L1–L42. Clendenin, C.W., Waddell, M.G., Addison, A.D., 2011. Reactivation and overprinting of South Georgia rift extension. Geol. Soc. Am. Abstr. Programs 43 (5), 551. Clendenin, C.W., Waddell, M.G., Addison, A.D., 2012. Triassic rifting in western Pangea. Geol. Soc. Am. Abstr. Programs 44 (7), 286. Cook, F.A., Brown, L.D., Kaufman, S., Oliver, J.E., Petersen, T.A., 1981. COCORP seismic profiling of the Appalachian orogen beneath the Coastal Plain of Georgia. Geol. Soc. Am. Bull. 92, 738–748. Corti, G., Bonini, M., Conticelli, S., Innocenti, F., Manetti, P., Sokoutis, D., 2003. Analogue modeling of continental extension: a review focused on the relations between the patterns of deformation and the presence of magma. Earth Sci. Rev. 63, 169–247. Daniels, D.L., 2001. Georgia aeromagnetic and gravity maps and data: a web site for distribution of data. USGS Open-file Report 01-0106 (http://pubs.usgs.gov/openfile/ of01-106/). Daniels, D.L., Zietz, I., Popenoe, P., 1983. Distribution of subsurface Lower Mesozoic rocks in the southeastern United States as interpreted from regional aeromagnetic and gravity. In: Gohn, G.S. (Ed.), Studies Related to the Charleston, South Carolina, Earthquake of 1886 Tectonics and Seismicity. U.S. Geol. Surv. Prof. Pap., 1313, pp. K1–K24. Heck, F.A., 1989. Mesozoic extension of the southern Appalachians. Geology 17, 711–714. Hopper, J.R., Buck, W.R., 1996. The effect of lower crustal flow on continental extension and passive margin formation. J. Geophys. Res. 101, 20,175–20,194. Jolivet, L., Labrousse, L., Agard, P., Locombe, O., Bailly, V., Lomonte, E., Mouthereau, F., Mehl, C., 2010. Rifting and shallow-dipping detachments, clues from the Corinth Rift and the Aegean. Tectonophysics 483, 287–304. Keranen, K.M., Kleperer, S.L., Julia, J., Lawrence, J.F., Nyblade, A.A., 2009. Low lower crustal velocity across Ethiopia: is the Main Ethiopian Rift a narrow rift in a hot craton? Geochem. Geophys. Geosyst. 10. http://dx.doi.org/10.1029/2008GC002293. Le Pourheit, L., Burov, E., Moretti, 2004. Rifting through a stack of inhomogeneous thrusts (the dipping pie concept). Tectonics 23, TC4005. Lister, G.S., Davis, G.A., 1989. The origin of metamorphic core complexes and detachment faults formed during Tertiary continental extension in the northern Colorado River region, U.S.A. J. Struct. Geol. 11, 65–94. Liu, M., 2001. Cenozoic extension and magmatism in the North American Cordillera: a role of gravitational collapse. Tectonophysics 342, 407–433.
C.W. Clendenin Jr. / Tectonophysics 608 (2013) 613–621 MacCready, T., Snoke, A.W., Wright, J.E., Howard, K.A., 1997. Mid-crustal flow during Tertiary extension in the Ruby Mountain core, complex, Nevada. Geol. Soc. Am. Bull. 109, 1576–1594. Mattioni, L., Le Pourhiet, L., Moretti, I., 2006. Rifting through a heterogeneous crust: Insights from analogue models and application to the Gulf of Corinth. In: Butler, S.J.H., Schreurs, G. (Eds.), Analogue and Numerical Modelling of Crustal-scale Processes. Geol. Soc. London Spec. Pub, 253, pp. 213–231. McBride, J.H., 1991. Constraints on the structure and tectonic development of the early Mesozoic South Georgia Rift, southeastern United States; seismic reflection data processing and interpretation. Tectonics 10, 1065–1083. McBride, J.H., Nelson, K.D., Brown, L.D., 1987. Early Mesozoic basin structure and tectonics of the southeastern United States as revealed from COCORP reflection data and the relation to Atlantic rifting. In: Beaumont, C., Tankard, A.J. (Eds.), Sedimentary Basins and Basin-forming Mechanisms. Can. Soc. Petrol. Geo. Mem, 12, pp. 173–184. McCarthy, J., Parsons, T., 1994. Insights into the kinematic evolution of the Basin and Range—Colorado Plateau transition from coincident seismic refraction and reflection data. Geol. Soc. Am. Bull. 106, 747–759. McClay, K.R., Buchanan, P.G., 1992. Thrust faults in inverted extensional basins. In: McClay, K.R. (Ed.), Thrust Tectonics. Chapman & Hill, London, pp. 93–104. McKenzie, D., Nimmo, F., Jackson, J.A., Gans, P.B., Miller, E.L., 2000. Characteristics and consequences of flow in the lower crust. J. Geophys. Res. 105, 11,029–11,046. Moy, C., Traverse, A., 1986. Palynostratigraphy of the subsurface Eagle Mills Formation from a well in east-central Texas. Palynology 10, 225–234. Petersen, T.A., Brown, L.D., Cook, F.A., Kaufman, S., Oliver, J.E., 1984. Structure of the Riddleville basin from COCORP data and implications for reactivation tectonics. J. Geol. 92, 261–271. Pindell, J.L., Kennan, L., 2009. Tectonic evolution of the Gulf of Mexico, Caribbean and northern South America in the mantle reference frame: an update. In: James, K.H., Lorente, M.A., Pindell, J.L. (Eds.), The Origin and Evolution of the Caribbean Plate. Geol. Soc. London Spec. Pub, 328, pp. 1–55. Rey, P., Vanderhaeghe, O., Teyssier, C., 2001. Gravitational collapse of continental crust: definition, regimes, and modes. Tectonophysics 342, 435–449.
621
Rey, P.F., Teyssier, C., Whitney, D.L., 2009. Extension rates, crustal melting, and core complex dynamics. Geology 37, 391–394. Reynolds, S.J., Lister, G.S., 1990. Folding of mylonitic zones in Cordilleran metamorphic core complexes: evidence from near the mylonitic front. Geology 18, 216–219. Sartain, S.M., See, B.E., 1997. The South Georgia basin: an integration of landsat, gravity, magnetic and seismic data to delineate basement structure and rift basin geometry. Gulf Coast Assoc. Geol. Soc. Trans. 47, 493–498. Tauvers, P.R., Muehlberger, W.R., 1987. Is the Brunswick magnetic anomaly really the Alleghanian suture. Tectonics 6, 331–342. ten Brink, U.S., Zhang, J., Brocher, T.M., Okaya, D.A., Klitgord, K.D., Fuis, G.S., 2000. Geophysical evidence for the evolution of the California Inner Continental Borderland as a metamorphic core complex. J. Geophys. Res. 105, 5835–5857. Tew, B.H., Mink, R.M., Mann, S.D., Bearden, B.L., Mancini, E.A., 1991. Geologic framework of Norphlet and pre-Norphlet strata of the onshore and offshore eastern Gulf of Mexico area. Gulf Coast Assoc. Geol. Soc. Trans. 41, 590–600. Thompson, G.A., McCarthy, J., 1990. A gravity constraint on the origin of highly extended terranes. Tectonophysics 174, 197–206. Tirel, C., Brun, J.-P., Sokoutis, D., 2006. Extension of thickened and hot lithospheres: inferences from laboratory modeling. Tectonics 25, TC1005. Traverse, A., 1987. Pollen and spores date origin of rift basins from Texas to Nova Scotia as early Late Triassic. Science 236, 1469–1472. Van Wijk, J.W., Blackman, D.K., 2005. Dynamics of continental rift propagation: the end member modes. Earth Planet. Sci. Lett. 229, 247–258. Whitney, D.L., Teyssier, C., Rey, P., Buck, W.R., 2013. Continental and oceanic core complexes. Geol. Soc. Am. Bull. 125, 273–298. Wijns, C., Weinberg, R., Gessner, K., Moresi, L., 2005. Mode of crustal extension determined by rheological layering. Earth Planet. Sci. Lett. 236, 120–134. Withjack, M.O., Schlische, R.W., Olsen, P.E., 1998. Diachronous rifting, drifting, and inversion on the passive margin of eastern North America: an analogue of other passive margins. Am. Assoc. Petrol. Geol. Bull. 82, 817–835. Ziegler, P.A., Cloetingh, S., 2004. Dynamic processes controlling evolution of rifted basins. Earth Sci. Rev. 64, 1–50.