Isle of Skye, Scotland: a proposed verification/calibration site

Marc Caffee (Purdue University), Lewis Owen (University of California), Gary Landis (USGS, Denver) and Douglas Benn (University of St. Andrews)

The Cuillin and adjacent mountains on the Isle of Skye (57^o15’N/6^o15’W) in Scotland contain the best and most intensely studied Younger Dryas moraine succession in the world (Walker et al 1988; Ballantyne, 1989; Benn, 1989; Walker and Lowe, 1990; Ballantyne et al. 1991; Benn et al. 1992). The moraine ages are well-defined on the basis of palynological and radiocarbon studies of peat bogs within and beyond the limit of the Loch Lomond Readvance (Younger Dryas) (Walker et al 1988; Walker and Lowe, 1990; Ballantyne et al. 1991; Ballantyne et al. 1991; Benn et al. 1992). The Loch Lomond Readvance cirque glaciers and a small ice cap that formed over the Cuillin Mountains produced a series of moraine ridges and ice scoured bedrock surfaces that are well preserved throughout the region (Ballantyne et al. 1991; Benn et al. 1992; Fig. 1). Prior to the Loch Lomond Readvance, during the Devensian (Last Glacial Maximum and probably up to ~ 15 ka), the Isle of Skye was essentially covered by the British Ice sheet (Ballantyne et al., 1998). Peaks higher than 750-870 m in the Cuillins, however, were not totally covered by the ice sheet and formed small nunataks that rise to 992 m asl on Sgurr Alasdair, but these are very small in extent (Dahl et al., 1996; Ballantyne et al., 1998). Recent work on chironomid assemblages at Whitrig Bog in southeast Scotland provides evidence that temperatures during the Lateglacial in Scotland agree closely, both in general trends and in detail, with the GRIP ice-core oxygen-isotope curve (Brooks and Birks, 2000: Fig. 2). Furthermore, similar chironomid work by Brooks et al. (in progress) at sites in Abernethy Forest (Cairngorms) and Loch Ashik (Isle of Skye) are confirming that the temperature changes documented at the Whitrig Bog site are of regional extent. This supports the view that temperature changes and glacier fluctuations during the Younger Dryas Stade in Scotland would have been synchronous with δ¹⁸O variations in the Greenland ice sheet. Therefore, we can directly correlate the ice-core oxygen-isotope stratigraphy with the glacial successions on the Isle of Skye.

During a field season in July 2002, we were able to examine the Younger Dryas moraines and associated landforms on the Isle of Skye. In particular, we examined potential calibration sites throughout southern Skye and collected a set of rock samples for terrestrial cosmogenic nuclides (TCN) surface exposure dating (Figs 3-5). As part of the CRONUS project, we wish to use these samples to help in calibration and verification of CRN production rates and to use the Younger Dryas landforms on the Isle of Skye as calibration/verification sites.

There are several reasons to believe that Younger Dryas glacial landforms on the Isle of Skye will provide excellent verification/calibration sites:

1. NW Scotland is very sensitive to changes in Northern Hemipshere/North Atlantic climate, and since the glaciers on Skye were small they would have responded rapidly (within a few decades) to climate change. Figure 2 shows that the times of temperature change in NW Scotland during the Lateglacial and earliest Holocene correlate with the ice-core data. As such, glacier fluctuations on Skye almost certainly were synchronous with variations in the δ¹⁸O values in the Greenland ice sheet. We can, therefore, correlate individual moraine ridges to distinct δ¹⁸O excursions in the Greenland ice-cores during the Younger Dryas Stade. In doing so, we can subdivide the Younger Dryas Stade from a simple 1300 year time span (11.6-12.9 ka) to centennial events during which distinct sets of moraines formed. Given that cirque glaciers have response times of decades, we can therefore assume that the maximum extent of small glaciers on Skye dates to the coldest part of the Stade, when equilibrium-line altitudes (ELAs) were lowest. As such, the outer most moraine ridges would have likely formed at ~12.5±0.2 ka. Similarly the inner most moraines and ice-scoured bedrock surfaces near the heads of glaciated valleys should date to the timing of the final deglaciation at ~11.6±0.1 ka (cf. Fig. 2). The new work being undertaken by Brooks et al. (in progress) on the chirominid record on sediments from Abernethy Forest and Loch Ashik will help substantiate and refine the timing of climate changes and glaciation on the Isle of Skye. The presence of the Vedde Ash in the deposits at Loch Ashik (Davies et al., 2001) can be used to reduce the uncertainty in radiocarbon dating undertaken on the sediment, further increasing the accuracy of the paleoclimatic interpretations.

2. TCN dating can be undertaken on a variety of lithologies, including granites, sandstones, basalts and gabbros, that contain suitable cosmogenic target minerals. This will allow us to study multiple TCNs on different rock types (Fig. 3). As the region is sufficiently small, asynchronous glaciation did not occur, , and so comparison of moraines and different lithologies between nearby valleys is possible.

3. Inheritance of TCNs should not be a major problem because the British ice sheet should have eroded sufficient amounts of rock during the Devensian to minimize exposure inheritance. We anticipate TCN nuclide inheritance in bedrock lithologies to be at most comprised of very minor deeper muonogenic and nucleogenic/radiogenic crustal nuclides typical of shielded rocks. The present precipitation on Skye is ~3000 mm a^-1, and estimates based on preliminary chironomid/glacier ELA work indicate that precipitation was similar during the Younger Dryas (Brooks et al., in progress). Thus, mass turnover was high and glaciers were clearly wet based, fast-flowing, and were likely to have been highly erosive. Where Loch Lomond readvance glaciers flowed at a high angle to the Devenisian ice flow direction (e.g. Coir' a Ghrunnda, Coire Lagan), there are no Devensian striae or roche moutonnee forms left. Therefore we know that the Younger Dryas glaciers eroded away the Devensian surface and inheritance should be insignificant. We will date Devensian moraines and ice scoured bedrock, however, to define the timing of Devensian deglaciation on Skye to be certain of the duration between deglaciation and the Loch Lomond Readvance. This will help assess the magnitude of inheritance. There is of course the problem that boulders from the Devensian glaciation could be incorporated into Younger Dryas moraines. However, we can easily assess the degree of likely inheritance by examining the lithologies of the moraines. Moraines with any significant inherited Devensian boulders would be apparent because they would contain numerous lithologies foreign to the local glacial catchments that were eroded during the Younger Dryas Stade. During our pilot study we did not find significant foreign boulders in any of the moraines that we studied. Furthermore careful sampling of boulders based on a knowledge of the catchment and glacial sedimentology can reduce this problem (cf. Benn, 1989).

4. Corrections for shielding due to snow cover are not necessary. Due to the maritime climate on Skye, snow does not persist for more than a few days a year, and it is unlikely that it was any more abundant during any time throughout the Holocene.

5. Holocene rates of weathering in this region are extremely low (cf. Ballantyne, 1994; Stone et al., 1998) and therefore corrections for weathering are small or negligible. Erosion rate scaling corrections (including moraine sediment shielding) will not seriously alter TCN age calculations.

6. The isostastic rebound is well defined for the region and therefore corrections for altitudinal changes can be be easily determined (Benn, 1991). A prominent rock platform just above sea level is present beyond the Younger Dryas glacier limits (ibid). This is thought to be equivalent to the Main Rock Platform that is present elsewhere in Scotland and is dated to the Younger Dryas Stade (ibid). The net sea-level change on Skye since the Younger Dryas is close to zero, which indicates 60±10 m of surface uplift during the last ~11 ka. Using the modeling of Lambeck (1993a and b) we will be able to determine the rate of surface uplift to make the appropriate altitude corrections. Furthermore, we can use the rock platform as another calibration site, yet the uncertainty of its age (11.6-12.9 ka) is larger than that of the moraines. However, dating this rock platform will provide confirmation that it formed during the Younger Dryas Stade and that we are certain about the uplift rates.

7. The region is extremely accessible and there are no restrictions on land access or scientific research.

On each of the samples already collected, we plan to measure ¹⁰Be, ²⁶Al, ³⁶Cl at Purdue University and ²¹Ne and ³He at the USGS. The TCN ages will be calculated and compared with the known ages on the moraines to refine the TCN production rates. The data we collect will also provide us with an intercalibration between the various TCNs.

Using the data from the pilot study we will return to the field area during the summer of 2004 and resample from sites that were particularly successful in the initial study. This will help us refine the results. Presently we have about five samples per moraine ridge within each of the study areas. Collecting up to 15 samples from selected moraine ridges will help reduce the statistical uncertainty of our results. We will also survey, using an electronic distance meter at a precision of 1-10cm, several of the moraine ridges to map their shape and we will collect boulders from different positions on the moraines to examine if there are significant variations in the CRN ages dependent upon sampling positions. This will help us test models for moraine degradation such as proposed by Hallet and Putkonen (1994). We should also like to sample some of the bedrock surfaces in more detail, for example, to compare areas of plucking from areas of abrasion to test for cosmogenic, nucleogenic, and radiogenic inheritance in bedrock and to quantify shielding or glacial erosion depths.

We will require support for 1) a graduate student (based at Purdue University, but who will also work at the USGS in Denver), 2) consumables incurred with mineral separation and mass spectrometer analysis, 3) costs of AMS measurements, 4) and travel for three weeks field work on Skye and to CRONUS meetings for each of the PIs.

REFERENCES

Ballantyne, C.K. (1989) The Loch Lomond Readvance on the Isle of Skye, Scotland: glacier reconstruction and paleoclimatic implications. Journal of Quaternary Science, 4, 95-108.

Ballantyne, C.K. (1994) Gibbsitic soils on former nunataks: implications for ice sheet reconctruction. Journal of Quaternary Science, 9, 73-80.

Ballantyne, C.K.(1998) Aeolian deposits on a Scottish mountain summit: characteristics, provenance, history and significance. Earth Surface Processes and Landforms, 23, 625-641.

Ballantyne, C.K., Benn, D.I., Lowe, J.J. and Walker, M.J.C. (1991) (eds) The Quaternary of the Isle of Skye: Field Guide. Cambridge: Quaternary Research Association.

Ballantyne, C.K., McCarroll, D., Nesje, A., Dahl, S.O. and Stone, J.O. (1998) The last ice sheet in North-West Scotland: reconstruction and implications. Quaternary Science Reviews, 17, 1149-1184.

Ballantyne, C.K., McCarroll, D., Nesje, A., Dahl, S.O., Stone, J.O. and Fifield, L.K. (1998) High-resolution reconstruction of the last ice sheet in NW Scotland. Terra Nova, 10, 63-67.

Ballantyne, C.K., Stone, J.O. and Fifield, L.K. (1998) Cosmogenic Cl-36 dating of postglacial landsliding at the Storr, Isle of Skye. The Holocene, 8, 347-351.

Benn, D.I. (1989) Debris transport by Loch Lomond Readvance glaciers in Northern Scotalnd: basin form and the within-valley asymmetry of lateral moraines. Journal of Quaternary Science, 4, 243-254.

Benn, D.I. (1991) Raised shorelines on Skye. In: Ballantyne, C.K., Benn, D.I., Lowe, J.J. and Walker, M.J.C. (eds) The Quaternary of the Isle of Skye: Field Guide. Cambridge: Quaternary Research Association, p. 90-97.

Benn, D.I. and Ballantyne, C.K. (2000) Classic Landforms of the Isle of Skye. Geographical Association, Sheffield, 56pp.

Benn, D.I., Lowe, J.J. and Walker, M.J.C. (1992) Glacier response to climatic change during the Loch Lomond Stadial and early Flandrian: geomorphological and palynological evidence from the Isel of Skye, Scotland. Journal of Quaternary Science, 7, 125-144.

Brooks, S.J. and Birks, H.J.B. (2000) Chironomid-inferred Late-glacial air temperatures at Whitrig Bog, southeast Scotland. Journal of Quaternary Science, 15, 759-764.

Brooks, S.J., Birks, J., Birks, H., Ballantyne, C.K. and Benn, D.I. (in progress) Reconstruction of climatic change and climatic gradients across the Scottish Highlands during the Loch Lomond (Younger Dryas) Stade.

Dahl, S-O., Ballantyne, C.K., McCarroll, D. and Nesje, A. (1996) Maximum altitude of Devensian glaciation on the Isle of Skye. Scottish Journal of Geology, 32, 107-15.

Davies, S.M., Turney, C.S.M. and Lowe, J.J. (2001) Identification and significance of a visible, basalt-rich Vedde Ash layer in a Late-glacial sequence on the Isle of Skye, Inner Hebrides, Scotland. Journal of Quaternary Science, 16, 99-104.

Hallet, B. and Putkonen, J. (1994) Surface dating of dynamic landforms: young boulders on aging moraines. Science, 265, 937-940.

Lambeck, K. (1993a) Glacial rebound of the British Isles. I: Preliminary model results. Geophysical Journal International, 115, 941-959.

Lambeck, K. (1993b) Glacial rebound of the British Isles. II: A high resolution, high precision model. Geophysical Journal International, 115, 941-959.

Stone, J.O., Ballantyne, C.K. and Fifield, K. (1998) Exposure dating and validation of periglacial weathering limits, N.W. Scotland. Geology, 26, 587-590.

Walker, M.J. and Lowe, J.J. (1990) Reconstructing the environmental history of the Last Glacial-Interglacial Transition: evidence from the Isle of Skye, Inner Hebrides, Scotland. Quaternary Science Reviews, 9, 15-49.

Walker, M.J., Ballantyne, C.K., Lowe, J.J. and Sutherland, D.G. (1988) A reinterpretation of the Lateglacial environmental history of the Isle of Skye, inner Hebrides, Scotland. Journal of Quaternary Science, 3, 135-146.

Figure 1: Location and reconstruction of the Younger Dryas glaciers in the Cuillin Mountains of the Isle of Skye (after Ballantyne, 1989).

Figure 2: a) Chironomid-inferred mean July air temperatures (^oC) at Whitrig Bog during the Lateglacial and earliest Holocene compared with b) the GRIP ice-core data (after Brooks and Birks, 2000). Both data sets show that the Younger Dryas Stade can be subdivided into distinct warmer and colder times. The Vedde Ash provides confidence in the radiocarbon dating at Whutrig Bog and can also bee found in Loch Ashik in the Isle of Skye (Davies et al., 2001)

Figure 3: The geology of southern Skye and the locations of the detailed study areas that we examined in our pilot study (after Ballantyne et al., 1991). Some of these study areas are shown in Figures 4 to 6. these illustrate the nature of the glacial geology and the locations of some of the samples that we have already collected.

Figure 3: CRN sampling sites on moraine boulders (red) and ice-scoured bedrock (blue) in Coire Fearchair on the granite hills of Beinn na Caillich (mapping after Benn et al., 1992). These moraines were formed by a small cirque glacier during the Younger Dryas Stadial. The outermost moraines, on which samples SC1 to SC6 were collected, probably formed during the early part of the stadial (12.5±0.2 ka). Samples SC7 to SC10 and SC68 to S70 were collected from the inner moraines and ice-scoured bedrock, respectively. These represent the final stage of deglaciation (11.6±0.1 ka).

Figure 4: the locations of CRN sampling sites on the gabbro hills of the Cuillins at Coir’ a’Ghrunnda. Samples SC28 to SC30 were collected from boulders on moraine crests that probably formed during the early part of the Younger Dryas Stadial at 12.5±0.2 ka. Samples from ice-scoured bedrock (SC58-SC60) should date to 11.6±0.1 ka when Younger Dryas glaciers finally retreated.

Figure 5: A) Sampling sites on ice-scoured sandstone an ice-divide in the Kyleakin Hills. These samples should represent the final retreat of the Younger Dryas glacier at 11.6±0.1 ka. The samples were collected from ice plucked zones of roche moutonnees and on ice-polished surfaces. B) Reconstruction of the former glaciers (after Ballantyne, 1989).