Abstract In 1975 the Consortium for Continental Reflection Profiling (COCORP) acquired 37 km of 12-fold common-midpoint (CMP) stacked seismic reflection profiles in Hardeman County, Texas ( Oliver et al. 1976 ). The entire crustal section was the object of study, so a record length of 15 sec (two-way traveltime) was recorded. Three separate profiles were arranged in an H-shaped pattern to provide two-dimensional (2-D) surface coverage. In addition to numerous reflections, the seismic sections exhibit pronounced apparent diffractions. These deep diffractions are more numerous and pronounced here than on any other CO-CORP survey to date. In many cases diffractions on different profiles intersect at the profile junctions. If these intersecting diffractions represent the same structural feature observed along different profiles, the diffraction time-surface provides an estimate of the location of the diffractor and an effective velocity (analogous to a stacking velocity) to the depth of the diffractor. Arrival times of these diffraction events were read, and a least-squares technique was developed to model intersecting diffractions simultaneously and derive location and velocity parameters, assuming point or line diffractors. Nine apparent diffraction events were identified, occurring mostly in the 5-9 sec range (~15–30 km). Point diffractor models generally give a closer fit than line models, with most of the resulting velocities between 6.0 and 7.5 km/sec. However, the introduction of a radial dimension (such as a dome instead of a point) significantly lowers these velocities without affecting the quality of fit. The locations of the point models, when projected to the surface, are distributed roughly along a northwest-southeast trend. The trend is subparallel to the Wichita Mountains and may indicate deep intrusive and/or tectonic structures related to their evolution. These results are important evidence that strong heterogeneities on the scale of a few kilometers are present at midcrustal depths.

Regional seismic reflection studies in the New England and southern Appalachians by COCORP and in Québec by the Ministére des Richesses Naturelles have provided critical subsurface geological information. The data clearly show considerable horizontal transport of off-shelf metasediments over coeval, relatively undeformed, lower Paleozoic shelf and miogeoclinal rocks. In the southern Appalachians, long distance (>200 km) transport of thin crystalline thrust sheets can be shown as well. The COCORP data from the Green Mountains of Vermont and a USGS seismic study in the Grandfather Mountain window of North Carolina and Tennessee appear to indicate that Precambrian (ca. 1.0 b.y.) Grenville basement in those areas is allochthonous and underlain either by shelf sediments or detachment horizons. In Québec, allochthonous basinal facies clastics are still preserved over a major anticlinorial structure, and extensive exposures of Precambrian basement are not found in an internal position in this part of the Appalachian mountain belt. The Bouguer gravity gradient in the central and southern Appalachians and the gravity high in New England and Québec are interpreted to mark a fundamental crustal density change at depth along the mountain chain, perhaps representing a preserved transition from continent to ocean. We infer, in part from the distribution of surface rock units with respect to the locus of the gravity gradient, that allochthonous off-shelf rocks may have been transported farther in the southern than in the northern Appalachians. Perhaps this is true for allochthonous Grenville basement as well, although the question cannot be unequivocally answered at this time. The seismic data suggest that highly deformed rocks exposed in the Appalachian chain are part of a relatively thin, composite allochthon presently confined to high structural levels and that the deeper part of the crust may constitute a largely undeformed ancient continental margin, perhaps including a transition from continental to rift stage or oceanic crust. Surface geologic relationships similar to those described for the Appalachians exist in a number of other mountain belts, and a modern analog for the subsurface structure of the frontal part of the Appalachians is present in the Banda arc of Indonesia. Regional deep crustal seismic surveys are clearly needed in other ancient deformed mountain belts and their active modern analogs.

Abstract This paper relates seismic reflection observations of subsurface structure in the southern Appalachians. These data provide significant new information on the existence, extent, and nature of thin, crystalline thrust sheets which appear to dominate at least part of this orogen. Consistent with the theory of plate tectonics, which maintains that many orogenic belts result from the lateral interaction of lithospheric plates, the evolution of the southern Appalachians was apparently characterized by the relative westward (in present coordinates) translation of horizontally moving thrust plates. Such an interpretation severely limits the role that vertical tectonism has played in the development of this orogen. Analogies with other mountain belts, both Precambrian and Phanerozoic, which exhibit similar surface structures and geophysical anomalies to the Appalachians, suggest that horizontal transport of extensive, thin thrust sheets may be more common than was thought previously. The evolution of continental masses may very well be characterized by the emplacement of large thrust plates along subhorizontal boundaries. The existence of thrust sheets has been recognized in the sedimentary foreland of the Appalachians (Valley and Ridge) for many years (Rich, 1934). However, although the idea that the crystalline metamorphic and igneous “core” as well as the Valley and Ridge may also be allochthonous was proposed as early as 1929 (Jonas, 1929), this idea has not been widely entertained. COCORP seismic data clearly extend the thin thrusting concept to the Blue Ridge and Piedmont and probably eastward to the crystalline rocks beneath the Coastal Plain. The data obtained by the Consortium

Abstract The COCORP southern Appalachian data were recorded in two major phases of field operation. Initial profiling in 1978 and 1979 resulted in the acquisition of 348 kilometers of data extending from the Valley and Ridge near Madisonville, Tennessee to the east edge of the Carolina slate belt near the Clark Hill reservoir in northeast Georgia. On Figure 2 these lines are denoted as lines 1 and 2 in Tennessee, and lines 1 through 4 in Georgia. During the second phase of operations, in 1980, 349 kilometers of profile were obtained from Lexington, Georgia to about 30 kilometers northwest of Savannah. These are denoted as lines 5 through 9 on Figure 2. At the beginning of the second phase, line 4 was extended southwestward to tie with line 5 and was relabeled line 4A. Table 1 lists the significant field parameters for both phases of the data acquisition. On all of the lines, the geophones were arranged in a center-weighted linear array at the station locations. The vibrator sweep frequencies were 8–32 Hz for the first phase, and 8–40 Hz for the second phase. The sweep length was 30 seconds and the listening time was 50 seconds, thus producing a nominal record length of 20 seconds. In general the processing of the data was carried out in a standard sequence. These include, in order (Fig. 3), demultiplexing of the field tapes, VIBROSEIS correlation with the appropriate sweep frequencies, editing of the unusable traces, sorting of the traces into common depth point

Abstract The surface rocks of the southern Appalachians are arranged in provinces and belts which generally strike northeast-southwest. In the vicinity of the COCORP traverse (Fig. 2), these include (from northwest to southeast): the Valley and Ridge province, the Blue Ridge province, the Brevard zone/Chauga belt, the Inner Piedmont province, the Charlotte belt, the Carolina slate belt, the Kiokee belt, the Belair belt, and the Coastal Plain. The traverse also crosses two major granitic bodies in the Piedmont—the Elberton and Sparta granites. The Valley and Ridge is perhaps the most intensively studied area of the Appalachian orogen. The reason for this is clear: exploitation of hydrocarbon deposits has produced abundant surface and subsurface structural and stratigraphic information. The acquisition of these data has provided a solution for one of the most intense debates concerning structural interpretation: the degree of basement involvement in the Valley and Ridge sedimentary structures. Ever since Rich (1934) proposed that the Valley and Ridge faults and folds developed above one (or more) subhorizontal detachments (decollements) with no significant involvement of the underlying Precambrian basement (“thin-skinned tectonics” as coined by Rodgers in 1949), controversy developed over the extent of Precambrian basement control of the sedimentary structures. Significantly, the experimental models proposed by Willis (1893, Plate LXVI) to explain Appalachian structures include both horizontal compression and intense deformation above a detachment zone (the contact between the board and the clay in his models). The “compression” machine of Willis (1893) thus seems to be the first suggestion of “thin-skinned tectonics”

Abstract Until recently, the COCORP traverse in Georgia was the only public seismic reflection profile which crossed the metamorphic “core” of the Appalachians with sufficient length of recording time to observe reflections which are returned from lower crustal depths. Clearly, any tectonic evolutionary model which uses these data as constraints will strongly depend on the nature of the data interpretation. For example, if the layers observed at depth beneath the Coastal Plain are interpreted to correlate with the (meta?) sediment layers beneath the Inner Piedmont, then the detachment(s) which must separate them from the overlying crystalline rocks would root somewhere to the east. On the other hand if such a correlation is not made, the dipping reflectors beneath the Charlotte belt may be a root zone for the thrusts of the Inner Piedmont and Blue Ridge. New COCORP data from the northern Appalachians display remarkably similar reflection events when compared with the southern Appalachian data (Ando et al, in press). In particular, Paleozoic sediment reflectors beneath the Taconic mountains may extend eastward beneath the Green Mountains. On the east side of the Green Mountains, east-dipping, layered reflections appear similar to the east-dipping layers beneath the Charlotte belt on line 1 and line 5 in the south, but in the north, they correlate with known metasedimentary rocks at the surface. Figure 8 thus shows two “end-member” interpretations of the southeastern United States based on interpretations of the COCORP data. In both of these models, the layers beneath the Blue Ridge and Inner

Abstract Although the subsurface information provided by the COCORP reflection data place strong constraints on the crustal structure and evolution of the southern Appalachians, numerous questions remain. It is the purpose of this section to outline some of the fundamental problems which may help to further enhance the understanding of the structural development in the Appalachians. Some examples include: What is the nature of the reflecting boundaries in the middle crust east of the Inner Piedmont? Are they basinal facies (meta-) sediments which stratigraphically correlate with the layers beneath the Inner Piedmont, or are they unrelated? This question is critical to determining whether the thrusts extend east of the Inner Piedmont, and thus to beneath the present Atlantic shelf. How do the reflectors beneath the Blue Ridge, Inner Piedmont, Eastern Piedmont, and Coastal Plain vary along strike? Do they vary in thickness and metamorphic grade significantly? Do the faults such as the Brevard and Augusta faults vary in dip along strike? Are the Precambrian (Grenville) units in the Blue Ridge and Piedmont all underlain by layered reflectors (presumably sediments), and thus allochthonous; or, are some of them autochthonous and thus true fensters through an overlying thrust sheet? Seismic data across the Grandfather Mountain window (Harris and Bayer, 1979a; Harris et al, 1981) and the Green Mountains (Ando et al, in press) can be interpreted to indicate emplacement of the Precambrian (Grenville) rocks along thrust faults, and thus imply they are allochthonous. A particularly important area is the Pine Mountain belt

Abstract Before discussing the implications of the data and their interpretation, we briefly recapitulate the major findings so far. Clearly, as further processing and future lines are completed, modifications and/or enhancement of the interpretations presented here will take place.

Abstract This publication consists of 5 chapters and resulted from the need to incorporate most of the COCORP southern Appalachian seismic reflection data and interpretations into a single work. Presented are ideas and models which are not only internally consistent, but also are consistent with what is known about the major geological and geophysical features.