SOURCES OF INFORMATION
The Central Appalachian Transect is anchored on both its northwestern and southeastern ends by the deep crustal seismic reflection and refraction profiles described above. The middle part of the transect contains the sites of the first commercial petroleum fields in North America, and continues to produce significant amounts of petroleum and natural gas. Consequently, seismic reflection data have been recorded along almost every road in the western half of the transect. These data all focus on the sedimentary upper crust, and provide no information about the middle and lower crust, but they nevertheless provide important insights into Applachian tectonics [Beardsly and Cable, 1983]. Most of the petroleum industry seismic data, such as the line PR-2 shown on the main display of the transect, remain proprietary. They are, however, rather widely available, and they have influenced the interpretations of faulting and folding of sedimentary rocks depicted on the transect. Published reflection profiles that were considered during the transect compilation may be found in Evans [1989], Wilson and Shumaker [1992], and Lampshire et al. [1994].
It is necessary to comment on the way the petroleum industry seismic reflection data and interpretations were used during the compilation of the transect. In the first place, a decision was taken that the line used for cross-sections would be straight. As none of the seismic profiles actually coincided with the line chosen for the cross-sections, this entailed projecting the published data and interpretations onto the line of the transect, with attendant uncertainties and possible changes in the structures depicted. Also, it was discovered, to no great surprise, that the published interpretations for adjoining parts of the transect could not be reconciled without major changes to one or both. For example, Evans' [1989] interpretation of the allochthonous platform carbonate rocks beneath the Blue Ridge was adopted in detail, while a modified version, to account for changes along-strike, of the Wilson and Shumaker [1992] interpretation was used for the Valley and Ridge. A problem occurred in the intervening Great Valley, where the quality of the seismic reflection data was poor, and the depth to basement in Evans' [1989] interpretation was too small in comparison to that required beneath the Valley and Ridge, if Wilson and Shumaker [1992] were correct. The problem was resolved by drawing a new interpretation of the Great Valley, which honored the same surface geologic constraints as Evans [1989], but the underlying blind structures were changed to accommodate a greater depth to basement.
Additional seismic data pertaining to crustal layering and seismic velocities are available from wide-angle reflection/refraction studies, and from earthquake studies. An interpretation of wide-angle reflection/refraction data along Edge line 802 by Holbrook et al. [1992] provided detailed information about P-wave velocities within the crust beneath the continental shelf. Low levels of earthquake activity occuring in the central Virginia and western Ohio portions of the transect have prompted studies to determine P- and S-wave velocity models for use in epicenter location programs. These are intraplate earthquakes, unrelated to tectonic activity, and not obviously related to ancient tectonic features. Chapman [1987] modeled the data available for onshore Virginia, including the data previously analyzed by James et al. [1968], data from the East Coast Onshore-Offshore Experiment (ECOOE), and travel time data from regional earthquakes, to produce the two-dimensional P- and S-wave velocity models summarized on the main display of the transect. Ruff et al. [1994] used P- and S-wave travel times from regional earthquakes to constrain crustal velocity models for the Anna, Ohio, region, at the northeastern end of the Central Applachian Transect. Considerable uncertainty exists, based only the earthquake data, about whether a distinct lower crustal layer exists there, and what its properties are. The presence of a distinct lower crust is more clear from the COCORP Ohio seismic reflection data [Pratt et al., 1989]. Thus, the crustal velocities summarized on the main display of the transect combine the earthquake and COCORP reflection data interpretations [Ruff et al., 1994].
There is a paucity of information about the structures and seismic velocities of the middle and lower crust beneath the Appalachian Basin and the fold-and-thrust belt, in the central portion of the transect. The Rome Trough, which is comprised of Eocambrian normal faults between the 300 - 400 km position along the transect, is well-known from petroleum industry reflection profiles. Visible as offsets in the top of basement and the overlying strata, the penetration of these normal faults into basement and the mechanics of their origin remains enigmatic. Single-fold, dynamite reflection profiles from the region west or northwest of the Rome Trough show overthrust structures in uppermost part of crystalline basement [Beardsly and Cable, 1983], which presumably date from the Grenville orogeny. The only information about deep crustal structure for this portion of the transect comes from studies by O'Neill [1983] of seismograms from a Long Range Seismic Monitoring (LSRM) observatory, that was temporarily in operation in the Valley and Ridge near Franklin, West Virginia. Using seismograms from teleseismic earthquakes and nuclear explosions at the Nevada Test Site, O'Neill identified a boundary at approximately 31 km depth where prominant P- to S-wave conversions were observed, implying a major change in the seismic properties of the crust. O'Neill interpreted the boundary to be the Mohorovicic discontinuity. However, the Bouguer gravity anomaly in this region is approximately -70 mgals, suggesting that the crustal thickness is roughly 45 km. For this reason, the boundary identified by O'Neill [1983] is interpreted as the top of the lower crust beneath the seismic station (near 475 km along the transect).
Gravity and aeromagnetic data are available for the entire area of the transect. These data were gridded during preparation of the Geophysics of North America CD-ROM [Hittleman et al., 1990], and those grids were adopted for the transect. The major difficulty encountered during the preparation of the gravity and magnetic maps was that different map projections were used for the transect and the CD-ROM. The Central Appalachian Transect is based on a Universal Transverse Mercator (UTM) projection for zone 17, using the Clarke 1866 ellipsoid [Williams, 1994]. The gravity and magnetic data grids from the CD-ROM used two other map projections. Also, the maps were rotated 32[[ordmasculine]] counterclockwise to the horizontal, as the real direction of the transect is northwest to southeast. It was necessary, therefore, to regrid the data from the CD-ROM, using grid rows and columns parallel to the edges of the transect. The gravity and magnetic anomaly values from the CD-ROM were assigned coordinates in their respective map projections, then these projection coordinates were inverted to obtain the corresponding latitudes and longitudes from which UTM coordinates were calculated, and the 32[[ordmasculine]] rotation was applied. Finally, the data were reinterpolated for the transect grid. This process, although cumbersome, does not alter the gravity and magnetic anomalies.
The offshore gravity anomalies, for the portion of the transect that was derived from the Edge seismic data, have been modeled by [Holbrook et al., 1994]. Onshore in Virginia, James et al. [1968] correlated the regional-scale gravity variations with the depth of the Mohorovicic discontinuity based on seismic refraction measurements, and Pratt et al. [1988] modeled the gravity anomalies along the route of the USGS I-64 seismic profile. In essence, these studies have shown that it is possible to find density distributions in the crust which are both geologically reasonable, and consistent with interpretations of the seismic data. However, the gravity data remain ambiguous, in the sense that it is not possible to distinguish between changes in crustal thickness and density variations within the crust, based on gravity anomalies alone [Hutchinson et al., 1983]. Thus, the seismic interpretations remain the principal source of detailed information about crustal structure for the eastern half of the transect. Hammer and Heck [1941] reported a gravity survey crossing the Valley and Ridge and the Great Valley in the region of the transect. This was possibly the first published gravity study were Hammer's terrain correction was applied to the data. Although Hammer and Heck [1941] did not interpret their data correctly in terms of geologic structures, their study showed that gravity anomalies amounting to 1-2 mgals are associated with the folding and faulting upper crustal strata in the central portion of the transect. The complete (terrain corrected) Bouguer gravity anomaly values for the western half of the transect vary slowly, mostly in the -50 to -70 mgal range. These values suggest that the Mohorovicic discontinuity remains relatively deep, 40 to 45 km, beneath western half of the transect. Although the ambiguity between crustal thickness and regional-scale density variations applies here, too, 40 to 45 km is consistent with estimates [Ruff et al., 1994, p. 57] from seismic refraction studies in the Anna, Ohio, earthquake zone. In the central portion of the transect, gravity is the principal source of information about the thickness of the crust, and is essential, as previously noted, to interpreting the seismic data [O'Neill, 1983] from the LSRM station near Franklin, West Virginia.
The magnetic anomalies have received less attention, during studies of the structure and tectonic history of the crust in the region of the transect, than the seismic data and gravity anomalies. The sources of individual magnetic anomalies generally lie within a few kilometers of the surface, an exception being the East Coast magnetic anomaly (near 975 km along the transect), which is evidently due to intrusive rocks at mid-to-lower crustal depths along the Atlantic Margin [Holbrook et al., 1994]. Of particular interest from the point-of-view of this transect is the Salisbury anomaly (near 700 km along the transect), which is expressed in both the magnetic and gravity data. Nowroozi and Davison [1990] modeled both kinds of anomalies in terms of east-dipping contacts, possibly faults, between metamafic and granitic rock. They suggested, as have others [Glover et al., 1989], that a Taconic suture is hidden beneath the Coastal Plain sediments near this position.
The geologic map for the transect combines information from a number of different sources, each at different scales and map projections. For Ohio, the portion of the state geologic map [Bownocker, 1920] within the transect corridor was digitized, except for the Carboniferous-Permian contact in the eastern portion of the state. The exposure of this particular contact is solely due to erosion, and the shape was judged to be too complex to render in detail. For West Virginia, the portion of the state geologic map [Cardwell et al., 1968] west of the Elkins Valley anticline was used. The map from Kulander and Dean [1986] was used for the fold-and-thrust belt, including the portion of the transect in West Virginia east of the Elkins Valley anticline and the Great Valley in Virginia, with modifications to incorporate the detailed map by Evans [1989] for part of the Great Valley and the western Blue Ridge. The geologic map by Glover et al. [1989] was used for the portion of the transect in Virginia, from the eastern Blue Ridge to the Coastal Plain.