Quaternary Sciences Institute - University of Waterloo, Waterloo, Ontario, N2L 361
by: Alan Morgan
Introduction
In October 2001 I had the opportunity of attending the Ontario Association of Geographic and Environment Education Conference at Niagara Falls. As a Quaternary geologist I was particularly interested in attending a session by Paul Hackl and Ling Wong of Riverdale Collegiate Institute, Toronto, with the fascinating title of "Rockwood Conservation Area Field Trip: A Search for Pattern and Process in Landscape".
This was an excellent presentation that brought out the karst and glacial geomorphologic aspects of the Rockwood area in some detail. Amongst the items mentioned were the potholes that lie within and adjacent to the Rockwood Conservation area. In the question and answer session that followed the presentation there appeared to be some interest in finding out more about the origins of these features from several teachers that had been involved in other fieldtrips to this area along the Eramosa River. Potholes are quite numerous within and outside the conservation area at Rockwood and also in the area immediately south of Everton, about 5 km north of Rockwood.
When I was a tyro geologist just completing my doctoral degree at Birmingham I was fortunate in being the first person to see three glacial potholes that had just been exhumed as a result of gas-pipeline laying in the English Midlands. Because these features are quite rare in Britain and because they were in intimate association with deposits of the last ice advance, I ended up describing them in the Journal of Glaciology (Morgan 1970). I was also aware that students at Waterloo had completed two excellent studies (both unpublished) on the Rockwood area (Kershaw 1973, Kunert 1997). Accordingly I promised to provide a description of the origins of potholes and more details on the Rockwood features in this issue of WAT ON EARTH.
Glacial potholes
The earliest ideas on the creation of potholes are that they were associated with "moulins de glacier" (glacier mills) formed where surface streams on glaciers and ice sheets fall into holes in the ice. Water entering these surficial holes was believed to impact on the bedrock beneath creating a large pothole. The "Moulin Hypothesis", first suggested by Bršgger and Reusch (1874) continued to be accepted by many authors until the 1950s (Faegri 1952). However, commencing in the 1930s, other authors have suggested dissatisfaction with the moulin hypothesis, largely on the grounds that it failed to explain how ice could remain stable long enough for the "giant" potholes to form and why many potholes (like those at Rockwood) were present in large numbers. Objections to the moulin hypothesis and alternative explanations the "Glacio-fluvial Hypothesis" are provided by Alexander (1932) and subsequent authors. If you are interested in details I would suggest looking at the citations provided in Charlesworth (1957), Embleton and King (1970), and Morgan (1970).
In summary "glacial" potholes are similar to more normal (or at least, more frequently encountered) "fluvial" potholes. A walk along a dry riverbed, particularly one that runs over bare expanses of soft or soluble or well-jointed bedrock will reveal potholes. In 1997 I had the opportunity of walking many dry riverbeds in the Bungle Bungles in Western Australia. The floors of the stream and riverbeds exhibited numerous excellent examples of potholes, many of which are remarkably similar to those found in the Rockwood area. I hasten to add that the potholes in the Bungle Bungles are formed under torrential flood conditions during the early months of the year when northwestern Australia is affected by cyclonic rains.
Certainly the Rockwood potholes are glacio-fluvial in origin, probably formed by subglacial or latero-glacial (ice-marginal) streams as explained in the next section. The similarity between the features is shared in the explanation that both were formed by torrential stream flow.
The Rockwood potholes
There seems to be general agreement amongst Quaternary workers that the Eramosa Valley is part of a complex series of meltwater channels that drained the ice margins retreating both into the Huron and Ontario Basins (Chapman and Putnam 1984) and Karrow (personal communication). The potholes that are found at Everton (not described herein) and at Rockwood appear to be associated with the floods of meltwater associated with the retreating Wisconsinan ice about 13,000 to 12,000 years ago. Potholes are found at many different elevations within both the Grand River Conservation Authority area (GRCA), and in the adjacent University of Waterloo-owned area. Given the altitude of the tops of the highest potholes they are obviously not associated with modern (or even near-modern) fluvial activity. Even those close to the level of the present day Eramosa River are likely the products of glacial meltwater.
Since well over 300 potholes have been recorded in the GRCA and adjacent areas at Rockwood (Kershaw 1973) it seems appropriate just to concentrate remarks on the most spectacular of these, the pothole known as the Devil's Well and the adjacent valley system.
The Devil's Well is located on a rocky bluff over 25m above Richardson Creek in the southeastern part of the University of Waterloo property. It is not signposted and is difficult to find, however, it is spectacular (as potholes go!) and is believed to be one of the largest described potholes in the world. It is just possible that it is the largest pothole by volume, and claims have been made in the Guinness "Book of Records" that this is the case. The pothole lies just beyond a small blind valley that runs from the cliff face which drops 10m into Richardson Creek. You can enter this small valley by following the pathway to the "15m drop" and descending the steep rock face to the valley floor at the word "CAUTION!".
It has been suggested that the valley represents the former position of a sequence of coalesced potholes aligned along a master joint system in the bedrock. The Devil's Well would represent the last in the sequence of potholes, fortunately created (and abandoned) before the thin rock wall that separates it from the blind valley could collapse. Traces of smoothed and circular areas along the walls of the small valley suggest that this might have been the case.
The Devil's Well is a volumetrically huge pothole. It measures 13.1m deep by 6.4m wide at the top and 4.9m wide at the base. Put in a somewhat different fashion, it is large enough to take two city busses side by side (admittedly a bit "squeezed") but it would have room to spare in the vertical dimension.
So, how were the potholes formed? The generally accepted belief is that high volumes of rapidly-flowing water, probably in an ice-marginal, or possibly sub-glacial river enlarged weak points in the bedrock and established the site for a pothole. A number of authors have indicated that massive bedrock (of virtually any lithology) is more suitable for the creation and retention of pothole forms. In the Rockwood potholes virtually all are formed in the Amabel Formation. Detritus carried in the water as bedload, created the "cutting tools" for a downward attrition of the rock surface. Large boulders that are normally assumed as the principal cutting tool, probably played a lesser role than smaller (perhaps fist size and smaller) clasts as cutting tools. In the excavation of a small pothole at Rockwood, Kunert (1997) illustrated highly-rounded and polished Precambrian clasts, that appear to be mostly in the 8 to 18cm range. This is about the same size as the largest clasts recovered from the potholes that I excavated in Wolverhampton. Certainly by far the largest bulk of sediment in a sealed pothole at Wolverhampton consisted of a package of gravel sized (1 to 3cm) clasts, in a sand matrix. However, one very large (800kg) granite was found perched on the side of the infilled pothole. One thing that several authors (including myself at Wolverhampton) observed, were helical- (spiral-) shaped grooves on the sides of the potholes. These indicate that water circulated in a spiral fashion within the pothole exiting up the centre of the pothole. Such features have not been reported from Rockwood.
A classroom experiment
Alexander (1932) created an experiment to simulate water flow in a pothole. This might make a useful class project. The apparatus is sketched below (Figure 8a) and I have modified it to show that it really does illustrate what I was able to observe in one of the three potholes excavated in the English Midlands almost 40 years later. As you can see there is a remarkable similarity between the two diagrams. However, from my field sketches (Figure 8b) I felt that water entered from the right, flowing down the helical flutings and then circulating several times around the central bedrock high, before exiting up the centre of the potholes (white arrows). Alexander's tube enters from the left, he has the water crossing the cylinder, but thereafter following the same flow pattern as in the diagram on the right, and also exiting through a central vortex.
Obviously an experiment in a glass cylinder cannot replicate the morphologies observed in relatively soft sandstone. However, there are remarkable similarities. It should be easy enough to copy Alexander's experiment using a large glass beaker with some marbles (or lentils) acting as the "grindstones" and to observe water flow. Perhaps some food colouring added to the water might help to clarify flow patterns within the artificial "pothole". I would encourage feedback on this matter.
Figure 8: (a) - left: Alexander's experiment, using a glass beaker, with water entry inside a tube. Circulation pattern is illustrated with black arrows. (b) - right: Field sketch (Morgan unpublished, 1969) of pothole morphology and postulated water circulation.
Acknowledgments
My first visit to Rockwood was made in 1971 and I have visited the Eramosa Valley, including the potholes in the Everton area several times in the intervening period. I was at the Rockwood Conservation area and the adjacent university of Waterloo property twice in the Fall of 2001 to gain insight into and to photograph the potholes and surrounding topography. However, much of the information on the Devil's Well is derived from Kershaw (1973) and Kunert (1997). I must claim responsibility for the observations on the potholes at Wolverhampton, and for the photographs within this article. Figure 8(a) is modified from Kershaw (1973) and from Alexander (1932). My thanks to former U of W Geography professor Sandy McLellan, who assisted me by providing a copy of Kershaw's report, and information on the "Devil's Well". Special thanks to Paul Hackl and Ling Wong of Riverdale Collegiate, Toronto. Without the intellectual stimulus of their presentation at OAGEE, this article likely would not have been written.
References
Alexander, H. S. 1932. Pothole erosion. Journal of Geology, 40: (4), 305 - 337.
Brogger, W. C., and Reusch, H. H. 1874. Giants' kettles at Christiania. Quarterly Journal of the Geological Society of London, 30: (120), 750 - 771.
Charlesworth, J. K. 1957. The Quaternary era, with special reference to its glaciation. London, Edward Arnold. 2 vols.
Embleton, C., and King, C.A.M. 1971. Glacial and Periglacial Geomorphology. McMillan of Canada, Toronto.
Faegri, K.1952. On the origin of pot-holes. Journal of Glaciology, 2: (11), 24 - 25.
Hackl, P. and Wong, L. 2001. Rockwood Conservation Area Field Trip: A search for pattern and process in Landscape. Presentation at OAGEE, Niagara Falls. 9pp. and appendices.
Kershaw, G. P. 1973. Rockwood: Resource Analysis and A Development Proposal. Unpublished B.E.S. thesis, Department of Geography, University of Waterloo, 70 pp. and appendices.
Kunert, M. 1997. Solution enhancement, erosional features and lithofacies relationships in the Middle Silurian Amabel and Guelph Formations, north of Guelph, Ontario. Unpublished M.Sc. thesis, Department of Earth Sciences, University of Waterloo, 135 pp., including appendices.
Morgan, A.V. 1970. Late Weichselian Potholes near Wolverhampton, England. Journal of Glaciology, 9: (55), 125 - 133.
Please note that photographs are copyright of the author. The web address is given below.
The website also contains an Adobe.pdf file available for downloads. Permission is given to duplicate this material for classroom use. Please cite the source.