William and Mary

2001: Origin and evolution of high elevation Southern Appalachian plateaus

Matthew L. Kerwan (Geology)


Many high elevation summits in the southern and central Appalachians are broad, flat plateaus. These broad uplands contrast sharply with the rugged topography of their side slopes and the sharper peaks in other regions. The summits are generally covered with large fields of angular rock fragments, often sorted into distinctive shapes including circles and stripes. Because formation of these features is usually limited to exceptionally cold environments, these plateaus have been hypothesized by Clark and Hedges (1992) to reflect cryoplanation during the periglacial climate of the Late Pleistocene. 

Because of the widespread distribution of plateaus throughout the southern and central Appalachians and the remains of sorted patterned ground, several authors have speculated that glacial period geomorphic efficiency is much higher than at present. Braun (1988) estimated that glacial period erosion rates in the Appalachians were greater than 100m per million years, at least 10 times greater than modern rates.  

Our research proposed to answer questions regarding the origin and development of the high elevation plateaus, specifically testing the paleoperiglacial hypothesis of Clark and Hedges, speculating on the relative importance of glacial cycles on shaping the Appalachian landscape. 

Study Area:

The study area includes the many high elevation summits just west of the Allegheny Front in the Monongahela National Forest of Tucker and Grant counties, West Virginia. The summit adjacent to Bear Rocks (4091 ft) is one of many high elevation knobs characterized by broad flat summits composed of gently dipping coarse grained orthoquartzite sandstones and conglomerates of the Pennsylvanian Conemaugh Group (Price, 1968). 

A central valley extends through the interior of the study area, following a long NE-SW trending syncline. At least twelve high elevation plateaus rise above the valley. The high and flat interior of the study area is bounded to the east and west by steep topography underlain by moderately dipping Pennsylvanian sandstones and shales of the Allegheny Formation and Pottsville Group. 


The limited observations of Clark and Hedges (1992) were expanded to include detailed surficial mapping of surficial deposits onto a 1:24000 scale topographic map of the Blackbird Knob quadrangle. Features normally associated with periglacial climate were mapped including sorted stripes, sorted circles, and boulder streams.

Detailed measurements of block orientations were made to determine the origin of stripe and boulder stream deposits. At least 25 individual blocks were measured along two transects perpendicular to the deposit. Measurements included the dimensions of all three axis, strike of the long axis, and the dip angle and direction of the bedding plane. Block orientations in several environments were normalized to slope direction and plotted on rose diagrams. 

Topographic surveying was conducted on the western side slope of Cabin Mountain to document the occurrence of a series of terraces. Detailed structural measurements and lithology descriptions were taken. Profiles were plotted to compare terrace position with structure and lithology. 

To address the potential impact of modern cold-climate weathering processes on the landscape, two thermometers have been installed in the vicinity of the Bear Rocks summit. Thermometers were placed in natural fractures on both the north and south sides of a rock face. The thermometers are connected to data loggers and will be checked periodically throughout the year. Historical climate data has been collected and will be used to model paleo-temperature regimes of central Appalachian summits.  

To determine long term rates of bedrock erosion, samples were taken from eleven localities throughout the study area. Samples were processed in preparation for cosmosgenic radionuclide (CRN) analysis. As CRNs are input from a cosmic source, they accumulate only in nearly exposed rock. Their concentration can estimate the time since active plateau erosion, and their erosion rate.

Results and Discussion:

There was a notable absence of patterned ground throughout the study area. No sorted circles or polygons and only a few questionable sorted stripes were observed. The absence of patterned ground was surprising given the study area’s high elevation, lack of vegetation, and coarse grained massive sandstone lithology, conditions favorable for patterned ground formation and preservation (Clark, pers. comm.). Observations of patterned ground south of the glacial border were first documented by Clark (1968) and were of critical importance to his paleoperiglacial hypothesis (Clark and Hedges, 1992) for the origin of this broad plateau and others throughout the southern and central Appalachians.

Mapping of side slope boulder streams revealed their concentration in dry to slightly wet hollows. Steep slopes and hollows with running water generally lack boulders. The concentration of boulder deposits in hollows suggests transport by water, perhaps during debris flows following catastrophic rainfall. During high rainfall events, otherwise dry hollows probably fill with just enough water necessary for transport. Conversely, during high rainfall events, wet hollows likely have flow velocities capable of completely removing all boulders. Geobotanical evidence, ie. tree scars and leaning trees, also supports this scenario. Scars are abundant in steep hollows with running water suggesting active transport during the life of the tree, while scars are absent in dry hollows. Although not conclusive, these scenarios suggest boulder transport by processes independent of cold climate.     

Orientations of boulders in stripes and streams were measured and plotted to establish methods of transport (Fig 1. a-f). Frost related transport mechanisms typically result in nearly vertical dip surfaces with strong downslope orientations. Debris flows generally result in a more scattered arrangement with a significant orientation perpendicular to flow (Mills, 1990). Boulder streams located in side slope hollows (Fig 1. e,f) had a significant orientation perpendicular to slope direction, highly indicative of debris flow transport. Boulders in one gentle side-slope boulder field (1c) had a very strong downslope orientation while the other (1a) had a more scattered orientation. While the orientation of 1c may suggest frost related transport, dip surfaces were very slight and would argue against any type of transport. The boulder field on the terrace below Bear Rocks did have a significant downslope orientation (1b), but again there was also a significant orientation perpendicular to slope (1d). Dip surfaces on these boulders was generally very high which may suggest frost-related, periglacial transport.     

The western side slope of Cabin Mountain contains a series of terraces and risers, coinciding with changes in lithology (Fig 2). Terraces occur on an easily erodeable siltstone unit while intermittent risers are located on the more resistant massive sandstone units. Structural measurements indicate that the eastern side slope, which has no well developed terraces, is at or near the dip of the sandstone unit. A similar pattern was observed on the eastern edge of the study area, beneath Bear Rocks, that also coincides with lithologic and structural changes. As cryoplanation terraces form indiscriminately across structure and lithology, the dependence on lithology of terraces in this area strongly questions their restriction to a cold climate origin.      


The high elevation plateaus surrounding Bear Rocks, WV have formerly been interpreted to be formed by frost related processes, relict features of extraordinary glacial period weathering. The absence of patterned ground, dependence of terrace formation on structure and lithology, and boulder field characteristics strongly argue against this interpretation. When completed, long term erosion rates from CRN dating may offer further support. The failure of these plateaus to be restricted to a cold-climate origin causes the authors to question the effectiveness of glacial period weathering. 


Braun, D.D., 1989, Glacial and Periglacial Erosion of the Appalachians, in Gardner, T.W. and Sevon, W.D. eds., Appalchian Geomorphology, pp. 233-256, Elsevier, Amsterdam.

Clark, G.M., 1968, Sorted patterned ground: New Appalachian localities south of the glacial border: Science, vol.161, p.355-356.

Clark, G.M., and Hedges, J., 1992, Origin of certain high-elevation local broad uplands in the Central Appalachians south of the glacial border, U.S.A.-A paleoperiglacial hypothesis, in Dixon, J., and Abrahams, A., eds., Periglacial Geomorphology: John Wiley and Sons Ltd., Chichester, pp. 31-61.

Price, P.H., 1968, Geologic Map of West Virginia: West Virginia Geological and Economic Survey.