Chapter 5: Results
Data was collected for thirteen moraines on the forelands of four glaciers. In total, twenty-seven stations were examined (although measurements were abandoned on station 17 due to instability).
Summary statistics for each station are presented in appendix A. Table 4 presents a summary of the most important data for each station and moraine. The overall 'mean 5' values for each moraine are a mean of the five largest lichens from all the stations (not an average of the 'mean 5' values for each station).
A Spearman rank correlation of subgenus Rhizocarpon measurements shows a very high degree of correlation between the parameters measured (table 5). All parameters have a correlation coefficient greater than +0.9 and were significant at the 0.0005 level. This strong correlation is graphically illustrated in figure 7 where the different parameters are plotted against one another in scatter graphs. Whilst the single largest 'largest inscribed circle' is used for dating in this study (as this is what the dating curve is based on), this analysis suggests that any of the parameters would be appropriate were a dating curve available.
Few Alectoria measurements were made in the field due to coalescence. The small sample size makes statistical testing of the degree of correlation between subgenus Rhizocarpon and Alectoria unreliable. Subjectively, however, the results appear generally consistent. The 'largest inscribed circle' measurements appear to be the most consistent. However, Alectoria consistently gives lower age estimates than subgenus Rhizocarpon. This is unsurprising because although the subgenus Rhizocarpon curve has been refined as more data has become available, this is not true of the Alectoria curve. The Alectoria measurements are, therefore, not considered to be reliable as anything other than an indicator of relative-age. When used in this way they support the subgenus Rhizocarpon data. If recalibrated, the Alectoria curve may provide better support for the subgenus Rhizocarpon derived ages and ages for surfaces where the largest subgenus Rhizocarpon thalli are smaller than 10 mm (and percentage measurement errors are high).
To test the reliability of this subjective technique a statistical test of its reproducibility was applied. The distributions of these data are not normal (see histograms in appendix B) so parametric statistical techniques are inappropriate.
The Wilcoxon test was used to assess whether the operator's estimates were consistent. This tests the null hypothesis that two paired variables have the same distribution by analysing the magnitude of the differences within pairs. Table 6 shows the results of this analysis, with cases where the null hypothesis can be rejected at the 0.05 level in emboldened italics. If the operators were consistent, the null hypothesis should be accepted, demonstrating that the data is likely to come from the same population.
It is apparent from table 6 that the operators gave inconsistent estimates of overall lichen cover much of the time. It is concluded that the overall lichen cover should not be used as an indicator of relative age of a surface. The consistency for the most covered face is higher, with the null hypothesis being rejected on only two occasions, but still cannot be considered reliable.
Table 7 shows the results of Spearman rank analysis of the correlation between the single largest 'largest inscribed circle' of subgenus Rhizocarpon and percentage cover of the most covered face of boulders (average of both operators). It is known that the lichen diameter provides a reliable indicator of surface age so a significant result would suggest that lichen cover is also a useful indicator of age. The test gives a correlation coefficient of +0.58 which, whilst significant at the 0.05 level, is not considered sufficiently high to be used other than to support other evidence.
5.4 Pebble roundness and boulder weathering
The Spearman rank correlation coefficient between the single largest 'largest inscribed circle' for subgenus Rhizocarpon and the pebble roundness index (table 7) was +0.44. The coefficient between this lichen parameter and the percentage of weathered boulders (table 7) was +0.50. These values are not significant at the 0.05 level. It is, therefore, concluded that neither pebble roundness or boulder weathering are able to differentiate between moraines of the ages studied. Previous work (Carrara and Andrews, 1972; Dugdale, 1972) has found these techniques useful, but they were only used on surfaces several thousand years older than those examined in this study.
Caribou glacier (figure 8) is an outlet glacier from the Penny Ice Cap. Its snout is located at approximately 66°38'N 65°15'W and has a south-easterly aspect. The foreland has massive ice cored lateral moraines. The proximal slopes have slope angles of up to 40° and are highly unstable with ice exposures on the right lateral. The distal slopes and crests are lichen covered and appear stable. The right lateral (see plate 3) has two distinct ridges, the outer of which (moraine 3) has very dense lichen cover and is clearly considerably older than the inner (moraine 4). Moraine 2 represents the maximum recognisable Holocene advance of the glacier. It has a relatively subdued relief suggesting it may have lost its ice core. It contains a kettle lake. Moraines 1 and 5 represent ice marginal positions intermediate between moraine 2 and the current position of the snout.
Table 8 gives estimated ages of the moraines based on the lichen dating curves.
Although the snout position was not accurately measured in 1995, comparison with an aerial photograph from 1975 (Natural Resources Canada) (plate 4) indicates it has certainly retreated since then, probably by about 500 metres. Caribou was the only glacier studied to show evidence of significant retreat over this period.
Nerutusôq glacier (figure 9) is a valley glacier with its origin in the fretted mountains to the south-east of Akshayuk Pass. Its snout is located at 66°36'N 65°13'W with a north-westerly aspect. There is evidence of several advances of greater or lesser extent, with the ice cored lateral moraines consisting of two distinct ridges in places. Two of these ridges on the left lateral were studied (moraines 7 and 8) but instability meant that only one station could be studied on moraine 7. The maximum extent of Holocene advance is marked by a clear ice cored terminal moraine (moraine 6) which is continuous with the lateral moraines. There are also many small ridges which are interpreted as ablation ridges. Moraine 9 is theablation ridge closest to the terminal moraine. The foreland has two large kettle holes and abundant evidence of fluvioglacial activity such as kames.
Lichenometrically derived ages for the moraines are presented in table 8.
Turnweather glacier (figure 10) is a valley glacier with its origin in the fretted mountains to the south-east of Akshayuk Pass. Its snout, located at 66°24'N 65°23'W, has a westerly aspect. The glacier is unusual amongst those studied in that the snout is located 2 km from the terminal moraine whilst elsewhere the distance was a few hundred metres.
The foreland has no distinct terminal moraine, this apparently having been reworked and partially removed by glacial meltwater. There are large ice cored lateral moraines on both sides of the foreland (moraines 10 and 13) and a low ridge within these on the left side assumed to be a more recent lateral moraine (not studied). Outside these ridges on the left side is an area of older moraine material (moraine 11) composed of ridges and disorganised piles of till. The distal slope of this feature is largely lichen free and reaches slope angles of up 40°. This is currently, or has been until recently, unstable.
Lichenometrically derived ages for the studied features are presented in table 8.
The snout of Tete des Cirques glacier (figure 11) is located at 66°27'N 65°27W and has a westerly aspect. It is a valley glacier with its origin in the fretted mountains to the south-east of Akshayuk Pass. The foreland is almost circular, and defined by a continuous ice cored latero-frontal moraine. The morphology within this moraine is complex with a mixture of fluvially and glacially formed features. A lake is dammed within the foreland, the size of which was observed to vary depending on the volume of glacial meltwater input, and which probably disappears entirely in the winter.
Due to time constraints, only the outer moraine was studied (moraine 12) and its lichenometrically determined age is presented in table 8.
