On 21st January we (along with much of the media) published a story on research that showed, said its authors, the existence of a glacier in Scotland a mere 400 years ago. Just two days later came a strong rebuttal of the work from Dr Adam Watson, a leading expert on the Cairngorms. Clearly, the scientific community does not move at a glacial pace. Now, at the risk of turning the issue into an endless academic tit for tat, we bring you a response to his criticisms from the scientists behind the original research. And that's the last you'll read of it on UKH. Probably. If for no other reason than that we're running out of Coire an Lochain shots.
We would like to further test the patience of you all by replying to each of Adam Watson’s specific criticisms about the evidence of a recent glacier in Coire an Lochain. This debate should really be conducted in peer-reviewed science journals. However, we’ve been surprised and pleased by the interest shown in the wider media by this study, so we hope readers find this interesting.
Taking each point in turn....
1. “Moraines are ruled out because of their descriptions of the sharp-angled pink, lichen-free boulders in a matrix of coarse sand and grit”.
Any form of recent deposit is likely to contain unweathered boulders which will lack lichens and be the colour of the unweathered minerals in the rock (the pink of Cairngorm granite coming from the feldspar crystals, which weather to grey). The angularity of rock proves nothing in this case. Rockfall boulders have angular edges, and when they fall onto glacier surfaces they remain unmodified, because there is nothing happening to the rocks to round off the edges. They look no different from snow-avalanche transported boulders. Therefore, these observations are consistent with a recent age for the boulder ridges, but not in themselves proving that it was or wasn’t a glacier that built them.
2. “Debris flows are the most likely explanation, with intense rainstorms gouging gravel and boulders down the gullies.”
Yes, this certainly happens, but it is problematic as an explanation for the boulder ridges. These border the lateral edges of the Great Slab, and descend obliquely downslope (not straight down). One problem is with the landforms, the other with the seasonal timing of when these processes take place.
The ridges we interpret as moraines don’t lie directly beneath scree-filled gullies, which would be the only source of large volumes of debris-flow deposits. The main gullies in the corrie (The Vent, The Couloir, Y-Gully) are too steep to hold much scree anyway. The biggest ridge, on the east side of the Great Slab, lies below The Vent, which is mostly a rock chimney in summer, b ut the ridge doesn’t lead from the base of the gully as it would if it was a debris-flow deposit. The boulder ridge on the west side is protected by a small buttress from material coming out of Y Gully. The ridges do, however, correspond to the typical form of lateral moraines of small cirque glaciers that (between us) we have observed or studied in six continents.
The timing is problematic because debris flows would have to occur in summer, when the gullies are snow-free. But the Great Slab will also be snow free then, so any debris flows coming from the headwall of Coire an Lochain couldn’t pile debris up to the height of the ridges. They would run along the side of the ridges and down the grassy slope below (you can see where debris flows have eroded this slope in places). Furthermore, where the moraine ridges are located, there is no reduction in slope angle which would cause debris flows to slow down and deposit debris, unlike the locations of most debris cones and debris flow lobes.
3. “The papers are typical of geomorpholologists who fail to dig a single soil pit and ignore fundamental principles of soil science.”
My co-author Jez Everest of the British Geological Survey comments “but I spend half my life with my head in a soil pit!” Dr Watson argues that moraines have soils with well-developed horizons, therefore the lack of such horizons means that the Coire an Lochain ridges cannot be moraines. But, at 1000 m above sea level, it takes a very long time for such horizons to develop mainly by the leaching and precipitation of iron. He is right that we are not soil scientists, so perhaps he could put some numbers on this for us. Soils accumulating on 12,000 year-old moraines in the Cairngorms have well-developed horizons. Would Dr Watson expect to find such horizon development in a soil only a few centuries old at this altitude?
Also, we do not say that there is no organic material in the soil (as he states). There is organic staining to variable depths, but no podsolization.
4. “This failure includes Sugden, who made the original proposal of glaciers in several corries of the Cairngorms in the 1700s and one in Garbh Choire Mor in the 1800s.”
David Sugden is an eminent geomorphologist who follows the scientific principles of critical testing and falsification of hypotheses. Such scientists have to be brave, because this involves being prepared to be wrong, by exposing their theories to constructive attacks. This is actually how science progresses, rather than by adding more observations to fit a pre-conceived world view. Being wrong for the right reasons helps other people’s understanding to develop.
In the 1970s, Sugden put forward the hypothesis that corrie moraines in the Cairngorms might have been formed in the 16th-19th centuries and tested this by measuring lichen sizes on boulders. This wasn’t conclusive, but didn’t rule it out: he needed a better critical test. So he took on a PhD student (Sheila Rapson) who collected sediment cores from the bottom of several corrie lochans (no mean feat in itself) and radiocarbon dated the organic layers in the sediment. It turned out the sediments had been accumulating undisturbed for thousands of years, so they concluded that no glacier could have advanced into the lochans in the Little Ice Age.
To describe this as a failure is to misunderstand how science progresses: a hypothesis was tested and successfully falsified at those sites. This also illustrates how we have to seek different methods and different types of evidence to test our hypotheses, and not just rely on a single method.
5. “The claim in Kirkbride about moraines in Coire an Lochain of Cairngorm is particularly unlikely. A snow patch survives all winter in very few years in that corrie, whereas in Garbh Choire Mor the patches almost always survive all winter, and hence this is the most likely site for a glacier in Scotland”
This is why we (there are five authors on our paper) were surprised to find evidence of a former glacier in Coire an Lochain. We would have expected Garbh Coire Mor to be the most likely site based on Dr Watson’s snow data collected over many years, which we are familiar with. But our conclusion that a glacier existed in Coire an Lochain instead is based on the landforms we observed and dated, not on indirect inferences from present-day snow patches. Nevertheless Dr Watson’s point needs to be addressed.
One possible reason why Coire an Lochain lacks late-lying snowbeds is that the snow accumulates on impermeable rock slabs, but in Garbh Choire Mor the snowbeds rest on bouldery scree and debris-flow material. In the latter case, meltwater from the snow percolates into the debris beneath, but in Coire an Lochain it cannot do this, and water lubricates the base of the melting snowpack leading to the spectacular crevasses and full-depth avalanches which occur as the snowpack starts to slide down the slab. These avalanches also erode the ends of the moraine ridges, rather than adding debris to the ridges. The avalanche run-out zone is lower down and less shaded, so the snow melts more quickly. Snow remaining above the avalanche crown walls suffers secondary sloughs as it is unsupported on the downhill side. So the lack of summer snow is not due to lower snow accumulation, but to more efficient loss than in Garbh Choire Mor.
We thank Adam for his comments, and are happy to discuss these issues.
On behalf of our co-authors.