“Oreo cookie” stratigraphy and the geologic setting of the Frog Legs Gorge

by Philip S. Prince, Virginia Tech Active Tectonics and Geomorphology Lab

While the “frog legs gorge” post was supposed to be a bit funny, the real purpose of it was to link outcrop patterns with geologic structure. This follow-up post tries to put the frog legs’ underlying structure into a broader context within the Appalachian Valley and Ridge, a deeply eroded fold-thrust belt with locally distinct structural styles. The frog legs gorge is located in a part of the Valley and Ridge in which a ~2,000 ft (600 m)-thick section of sedimentary rock is able to fault and fold relatively independently of the much thicker sedimentary sequences above and below it.

PenCo leg location
From Kulander and Dean (1986), one of my all-time favorites. The dark blue carbonates and light blue clastics are strong and brittle; the yellow layer also contains brittle sandstones. Above and below the yellow, however, are weak shale horizons that represent the Oreo filling and disconnect all three brittle zones from each other. As a result, they can fault and fold independently from one another.

The cross section above from Kulander and Dean (1986) in intended to communicate the general idea of independent faulting and folding within different parts of the overall layer sequence. The frog legs occur in an exposure of the thin yellow zone, which contains lots of sandstone, along with some limestone and shale. The black lines show closely spaced faults within this zone. These faults and their associated folds don’t extend upwards or downwards into the layer sequences above and below, so their wavelength is short (3,300 ft/ 1 km or less). This pattern can be reproduced in a sand model using granular materials of varying strengths, as seen in the Aguaragüe anticline post. 

model clean
Red circle shows the conceptual position of the folds that would produce a frog legs gorge-type feature when exposed at the surface by erosion. White layers above and below it are glass microbeads. They are the “soft filling” sandwiched between stiffer, stronger horizons.
model with faults.jpg
Black lines are faults.

All of the layers in the model are faulted and folded, but if you look closely, it is not possible to link faults from the deep dark blue layers to faults in the uppermost light blue layers. The white stuff in the middle is weak microbeads, and these effectively disconnect the uppermost sequence from the lowermost sequence. The thin yellow layer in the middle is quite brittle, but it is most along for the ride within the weak zone. This model would need to be deeply eroded to expose its inner portions and produce an outcrop pattern like what is seen in today’s Valley and Ridge.

frog leg summary
Conceptual comparison of the model to the Kulander and Dean (1986) section. The second model shown further down is probably a better match.  The real fold belt in the top cross section is about 4 miles (6.4 km) thick beneath the red circle.

This style of faulting can be seen on the model’s surface while it is deforming. Because the uppermost light blue layer can slide somewhat freely from layers below, it produces small and closely spaced surface folds due to its thickness. When the entire layer pack faults at once, the surface fold is much broader. This video shows deformation of the model above, as well as an additional experiment:

I call this model setup the “Oreo Cookie” because it involves a middle weak zone sandwiched between  stiffer, more brittle layers above and below. In my experience, the results most closely resemble the reality of this part of the Valley and Ridge when the uppermost layer sequence is both the strongest and the thickest part of the layer pack. Some of the stylistic features that form in this model setup can be compared to real structural features and the landforms they produce.


1 is the outermost folding near the deformation front. 2 is the frontal structural high, where faulting that affects the entire section brings the deepest layers to their highest point within the model. 3 is a broad syncline that preserves shallower layers after they have been eroded away elsewhere. 4 is a structural high at the rear of the thrust belt that still preserves a few folds of the brittle yellow layer, which make a complicated pattern of ridges. 1 and 4 are about 45 miles (75 km) apart.

The second experiment in the video link above produced slightly different results because it used slightly different sand (my materials are color coded by their relative strength). In many ways, the second model matches the structural surroundings of the frog legs gorge more closely because it developed a huge, broad syncline on its hinterland side (see the light blue syncline in the real Kulander and Dean section). This is the broad, downward-flexed zone of pink and red layers on the right side of the model below.

Elliott Knob clean
The black circle shows the conceptual position of the frog legs gorge at the end of deformation.

Significant erosion of this model would create an outcrop pattern that is dominated by the lower portions of the pink/red layers where they are flexed downward into the syncline. A similar feature in the real Valley and Ridge creates a broad zone of high and rugged mountain topography today.

elliott landscape
1 is the frontal structural high, and 2 is the massive, broad syncline that preserves shallow layers and creates a structural high. Tiny red circle and black arrow show frog legs gorge location.
DEM Elliott
1 is the frontal structural high, the red dot is the frog legs gorge actual location, and 2 is the rugged mountain range created by the large syncline at the rear of the thrust belt. 1 and 2 are about 25 miles (40 km) apart.

Another section from this model produced a higher displacement frontal thrust fault. Again, this thrust does not cut straight through the entire layer pack. Instead, its displacement is distributed onto several smaller faults within and above the weak white layers. I don’t even try to draw them here, but you can see that the yellow/green layer is not offset like the dark blue below it. Changes in thickness of the weak white layer is the result of displacement of the frontal thrust fault being redistributed throughout the section.

Elliott model 2
The structural high in the pink and red layers (underneath “anticline”) actually overlies a deep syncline in the lower brittle section (dark blue). I suppose the scaled thickness of this model fold belt would be about 10-12 km on the right side; it should be eroded down to expose the yellow/green layer at the surface to reflect modern erosional depth in the central Appalachian Valley and Ridge.
2 thrusts
LiDAR-derived hillshade imagery shows minor thrust faults in the brittle sandstone of the real frontal structural high near the frog legs. Both faults, which are backthrusts, offset the sandstone layers about 800 ft (254 m).  The overall anticline is about 3,000 ft across here (just less than 1 km).

Like many other models I have posted on this blog, these experiments produce folds and fault patterns that are limited to specific sections within the overall stratigraphy. If these models were made with uniform sand, they would produce 4 or 5 thrust sheets that each carry the entire layer pack and bring deep layers to shallow depths. With the addition of microbeads, the result is an upper and lower sheet, each of which is internally faulted and folded. Each of the disconnected horizons contains its own set of faults and folds, and they often don’t line up vertically in the section. You could stand on a surface anticline that is underlain by a syncline at depth, and vice versa. The tendency of the microbeads to “flow” to a certain degree allows them to fill in the space to make the fold mismatches possible.

fold projection

PenCo leg location
The left arrow points to a small anticline in the yellow/green layer with a syncline in the dark blue layer beneath it. The right arrow points to an anticline in the dark blue layer that lies almost beneath the axis of the big syncline in the upper sheet. Compare to features in the section below, but keep in mind lots of the model would need to be eroded away to reflect the present-day valley and ridge. Again, the fold belt in the real section is about 3.5-4 miles (6-6.4 km) thick beneath the red circle. The Whip Cove Syncline is the broad, light blue syncline at the right.

I really like the monster syncline under the words “Upper Sheet” above because it resembles the real Whip Cove Syncline (landform 2 above) of Virginia and West Virginia. Today, this broad expanse of Devonian- and Mississippian-aged sedimentary rocks produces a significant mountain range due to its size and erosional resistance. The area is one of surprisingly unbroken wilderness because the topography is extremely rugged and the soil is not good for agriculture. The huge dark swath running through the middle of the image below is about 10 miles (16 km) across, and stretches for many 10’s of miles along the trend of the Valley and Ridge.

Shenandoah mtn
Lots of cool spots in this part of the world. Also lots of reddish-brown, flaggy Devonian sandstone. Lots…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s