Jökull - 01.12.1985, Qupperneq 65

of 50 boulders was used. Three readings were taken on
each boulder, these were then averaged and an overall
value for the site calculated using the 50 average read-
ings. As expected there is no significant difference in
weathering between sites on the foreland, a maximum
of 90-100 years usually being insufficient for noticeable
disparities in weathering to be recognised but outside
the foreland on surfaces probably deglaciated around
9000—10000 years ago the amount of weathering
increases. Whilst the difference in weathering is not
quite of the same order as that identified in southern
Norway (Matthews and Shakesby 1984), possibly due to
the small sample of sites and differences in geology, the
overall pattern of weathering is similar, with lower
mean values with increasing distance from the glacier,
but some degree of overlap taking into account values
for ±1 standard deviation from the mean.
During the course of the 1981 survey a further ridge
was identified on the western side of the valley lying
immediately outside the very extensive and complete
ridge system identified as the Little Ice Age, late 19th
century, terminus. It also lay well inside the peats
examined tephrochronologically by Miiller (see above)
and could therefore only date from sometime within the
Neoglacial period. This outermost feature was cut by
debris flows, unlike the other extensive ridge system,
and had no obvious counterpart on the eastern side of
the valley. Morphologically however it was no different
from the ridges mapped as moraines and it had a similar
vegetation and soil cover to the outermost mapped
moraine. In terms of the weathering of boulders in the
ridge the results were not statistically separable from
those of the outer ridge system, although the mean was
slightly lower. Measurement of lichens on the ridge
following the procedures of Caseldine (1983), utilising a
search area of 200 sq. m. on its proximal slope, pro-
duced a date of between 1898-1901, again virtually
indistinguishable from the other ridge. There is there-
fore no direct evidence to place this ridge any earlier
than the end of the Little Ice Age. It presumably marks
a slightly more extensive position than the better
developed feature previously identified as marking the
maximum, either a position at which the glaciers did not
remain sufficiently long to deposit a more extensive
moraine system, or the remnant of a more extensive
ridge subsequently destroyed by meltwater or debris
flow activity.
LICHENOMETRY
The growth rate for the lichen Rhizocarpon geo-
graphicum agg. used in all the previous lichenometric
'Schmidt' readings - Gljúfurárdalur
60-
55-
50-
45- 1
40-
35-
L1 R8 R10 R36 Outer Outside
Fig. 2. ”Schmidt“ hammer readings for Gljúfurárdalur
(the scale is related to, but not equivalent to, the
coefficient of restitution R determined by the distance
the hammer rebounds from the rock surface).
— 2. mynd. Lausleg könnun á veðrun bergs í Gljúfurár-
dal.
work had been established from a small number of
dated surfaces in Skíðadalur and caution was expressed
over the use of the growth curve and its applicability in
Gljúfurárdalur. In the absence of other dated surfaces a
long term evaluation of the accuracy of the growth rate
was begun in 1983 by setting up a number of permanent
direct measurement lichen sites. On three ridges (Fig.
1) thalli were accurately measured along defined longest
axes and also at 90° to the longest axis. Each lichen was
given a number painted on the boulder together with a
line marking the axes of measurement. In all, 25 thalli
were measured, 15 on the outer moraine (Gl-15), 5
inside the moraine (G16—20) and a further 5 near EUB
(G21— 25), selecting a range of sizes at each site but
concentrating on the largest lichens. These measure-
ments, as made in July 1983, are given in Table 1. In an
environment such as Northern Iceland with suggested
growth rates for largest lichens of 57 mm/100 yrs (shortest
axis i.e. diameter of largest inscribed circle) (Caseldine,
1983) it will be necessary for continued observation up
to at least 5 years and preferably longer to establish the
accuracy of the previously derived growth rates, as
annual observations will be very susceptible to inaccur-
acies in the measurement technique.
JÖKULL 35. ÁR 63