Metamorphic Formations and
Stratigraphy of Piedmont Georgia

Mapping and Stratigraphy in Georgia’s Piedmont

Thrust faults delimit large areas of the Piedmont. Within regions bounded by thrust faults -- and sometimes across those faults -- stratigraphy seems elusive. Locally named formations seem lithically distinct; however, they are difficult to connect across a region to construct a coherent larger formation. For instance:

• Where thrust faults -- such as the Carters Dam Fault -- delimit major regions, thrust contact often is interpreted as a boundary between two thrust sheets. Scattered throughout the Piedmont we see ultramafic bodies apparently derived from lower oceanic crust and mantle. Many of these bodies show thrust contact with underlying rock. In USGS Professional Paper 1475, Mike Higgins -- who is better-trained and more experienced than I ever will be -- proposed that many such bodies are erosional remnants of a thrust sheet comprising crust and mantle; however, in more recent work he proposes that some of these ultramafic bodies are discontinuous lumps in a matrix of metamorphosed marine mud. In this interpretation, although these bodies may show thrust contact, they do not represent a formerly contiguous sheet. Evidently, in an environment of widespread thrusting and deformation, any coherent body may be in thrust contact with its matrix.
  -
• Numerous granitic bodies outcrop in Piedmont Georgia. Although some of these bodies outcrop as exfoliation domes like Atlanta’s Stone Mountain, they do not form a narrow, contiguous, deep-rooted plutonic mountain belt like California’s Sierra Nevada; instead, they occur as discontinuous shallow-rooted bodies throughout the Piedmont. While these bodies are granitic (felsic), they are not all granite; Mount Arabia is migmatitic, and felsic gneiss outcrops abound. This -- along with geochemical evidence -- suggests that Georgia’s granites were not created as roots of stratovolcanoes over a subduction zone, but formed as friction-melted products of marine mud during closure of the Iapetus basin.

Based on recent mapping and reinterpretation south of the Brevard zone, it appears that:

• Atlanta's Big Cotton Indian, Ola, Appalachee, Kalves Creek, and Clairmont formations are phases or facies of Stonewall gneiss.
  -
• Stone Mountain granite and Mount Arabia migmatite are likely to be more-thoroughly-melted phases of Lithonia gneiss; also, Lithonia gneiss grades into Promised Land formation, and may be a near-melt product of Promised Land.
  -
• Stonewall and Lithonia gneisses occupy contiguous areas, appear similar, and may be identical.

We seem to be moving toward a concept that large areas of Piedmont Georgia show little stratigraphy in the conventional sense of orderly sequential layering. Not only is much of the region a lithologic jumble, but many rock types seem to derive from a generally similar parent material; thus, their diversity may be a product not of varying protolith but of locally varying metamorphic/tectonic conditions. Rock that in one location is a massive meta-sediment may in other locations be sheared to a schist or melted to a granite. The problem of finding stratigraphy in some metamorphic terranes may be a larger-scale version of the problem of finding stratigraphy in a tectonic melange. It seems not far wrong to consider the Piedmont largely as a collection of melanges.

My guess is that Georgia’s Piedmont from the Brevard zone south to the vicinity of Pine Mountain has a relatively undifferentiated progenitor: lithified marine mud, with intermixed fragments of oceanic crust and mantle. Composition of that mud varies; in some areas, it comprises water-laid volcanic tuffs; in other areas it looks like ubiquitous sedimentary crud -- the primordial ooze of ocean basins -- which mainly may be airborne continental dust. Stressed by irresistible tectonics, this mud is variously metamorphosed. Where metamorphic temperature and pressure rose sufficiently, this mud melted to form today’s granite of Stone Mountain, Palmetto, Ben Hill, Elberton, and Sparta; where a little less stressed, it formed gneisses like Lithonia, Stonewall, and Big Cotton Indian; where shearing had more effect than temperature, today’s rock is a schist like the Brevard zone’s garnet-mica schists. I suspect that many currently named "formations" describe local metamorphic effects rather than differences in parent material or depositional environment.

What current rock most closely resembles those original sediments? In the Atlanta area, the Promised Land formation seems a likely candidate. The Promised Land's mafic and felsic meta-tuffs preserve apparent original layering; they were "gently" intruded by Stone Mountain granite; and they can be mapped gradationally into Lithonia gneiss.

South of the Pine Mountain area, today’s rock derives also from marine mud; however, that mud may have traveled from eastern regions of the Iapetus basin. Southeast from the Elberton area, rock of Georgia’s Piedmont appears more closely related to Africa than to North America.

The Brevard Zone and the Clairmont Formation

The Brevard zone and the Clairmont formation -- and the relationship between them -- are not well understood.

Where the Brevard Zone is a near-linear zone of concentrated brittle and ductile deformation, the Clairmont formation -- lying near southeast of the Brevard Zone -- is a diffuse zone showing mainly ductile deformation, with diverse rock types embedded in a gneissic matrix that appears to flow around its distorted inclusions in a semi-fluid manner. The Clairmont formation’s internal structure leads some geologists to conclude that it is a metamorphosed tectonic melange. Approaching the Brevard zone from the southeast, relatively homogeneous Lithonia/Stonewall gneiss grades into the Clairmont formation’s tectonized jumble.

On seismic evidence, the Brevard Zone in its current form appears as a southeast-dipping thrust fault; however, other lines of evidence suggest that during its long and complex history, it has also moved in other directions. Regionally, the Brevard Zone is a northeast-trending strip of sheared, fractured rock flanked to the northwest by (a) volcanic belt(s) and to the southeast by a tectonic melange belt (the Clairmont Formation). On the face of that evidence, one might leap to the speculation that -- whatever its later history -- the Brevard Zone began as an oceanic subduction zone during closure of Iapteus and assembly of Pangea. Detailed evidence may not support this speculation. For one thing, the Brevard Zone dips in the wrong direction to produce a volcanic belt to its northwest.

The Murphy Belt

Historically, geologists have perceived the Murphy Marble Belt as a distorted syncline exposing continuous marble deposits -- a folded, metamorphosed carbonate layer -- along its flanks. On contour and shaded-relief maps, much of the Murphy Belt clearly shows a pair of parallel valleys separated by a central ridge (often labeled "Dividing Ridge"). In the classic model, these parallel valleys are thought to be produced by dissolution of folded carbonate strata. Closer inspection suggests that this central ridge is flanked by multiple valleys and ridges, now dissected by cross-cutting streams. If Murphy Belt structure is synclinal, then flanking valleys may reflect carbonate strata in multiple overturned folds. Alternately -- and perhaps more simply -- parallel valleys may be the product of erosion along imbricate thrust faults.

If Murphy Belt structure comprises multiple thrust faults, that suggests some structural similarity to the Brevard Zone, as well as the Carters Dam fault. Murphy marbles and carbonates in the Brevard Zone both may represent Ridge and Valley limestones and dolostones transported upward along thrust faults.

Other arguments question the conventional synclinal interpretation of Murphy structure:

• If the Murphy Belt is a syncline, it is an amazingly long, narrow one. Do we know of any other place in the Appalachians (or elsewhere) where we find a single elongate overturned syncline -- a neat pleat in the rocks -- 100 miles long? Can we think of a mechanism likely to produce such a structure?
  -
• Murphy ridges and valleys are remarkably uniform and well-aligned. Is this likely to be the product of erosion alone?
  -
• Is carbonate dissolution a plausible mechanism to produce Murphy Belt valleys? Granted, dissolution can produce topographic depressions, e.g. sinkholes and the Okefenokee swamp, as well as karst features in general; however, these are cases of dissolution by groundwater, not surface water. In the Ridge and Valley province or the Appalachian Plateau, unfractured carbonates do not erode to produce valleys or even common depressions. On the other hand, thrust-fractured carbonates, sandstone, and shale do erode into valleys. Also, marbles are present only intermittently in Murphy valleys, so their dissolution is not likely to produce such regular landforms
  -
• In the Great Valley, the Brevard Zone, and the Gold Belt, where we see rhythmically-spaced parallel ridges and long narrow valleys oriented along strike with strata dipping across strike, we think those landforms were generated by the underlying mechanism of thrusting. However, where we see similar parallel ridges and long narrow valleys with cross-dipping strata in the Murphy Belt, there we think we're looking at an overturned syncline. What's different about the Murphy Belt?
  -
• In the Ridge and Valley province, synclines make mountains; breached anticlines make valleys. In contrast, the Murphy synclinal model proposes that the Murphy valley is a syncline while the mountains east and west of that valley are anticlines. Again, why would Murphy structure be different?

The above objections to the Murphy synclinal model are products of reason and philosophy, and therefore are subject to well-known human failings of those enterprises. Aristotle notwithstanding, ground truth ultimately overrules reason; field evidence rectifies hypotheses. If observation on the ground shows conclusively that Murphy belt structure is synclinal, then that's what it is, regardless of what I think it reasonably should be. However, as I understand it, this field evidence is not obvious and certain but obscure and ambiguous. (In the Appalachians, what a surprise!) While a syncline has been the favored model for about 100 years, many competent geologists have disagreed with that interpretation.

After way too much time staring at Google Maps -- and not enough time looking at rocks -- the presence of similar ridges in all three regions has persuaded me that the Murphy Belt, the Brevard Zone, and the Gold Belt have similar structures and are the products of similar processes. But what are the structures and process?