Jökull - 01.12.2007, Side 26

Maciej Dąbski
RESULTS
Lichen populations (Rhizocarpon agg.) growing on
moraines I, II, III and IV are composite and are cha-
racterised by the presence of some exceptionally large
thalli which make the frequency gradient as well as
the correlation (R2) quite low (Figure 3, solid lines,
Table 1 A). Exceptionally large thallus on moraine
III with a 150 mm diameter, suggests that this sur-
face is older than the surface on moraine II, which is
obviously impossible (Dąbski, 2002). Also, the sin-
gle largest thallus (almost two times larger then the
second largest thallus) on moraine I suggests that it
began growing prior to the moraine ridge formation.
The extra large thallus found on ridge I may be inhe-
rited from a supraglacial period of growth, although
Fláajökull does not have a distinct medial moraine,
which would allow for relevant substrate stabilisation
(Griffey, 1978). This lichen may have been incorpora-
ted into the end moraine from an older deposit, distal
to the moraine. The extra large size of a single thallus
found on ridge III is difficult to explain.
Table 1. Regression lines parameters: A) whole li-
chen populations, B) populations without exceptio-
nally large lichens. – Stærðarhlutföll mældra fléttna
á jökulgörðum I–V.
A
moraine grad. (-1) R2 n
I 0.0219 0.5264 497
II 0.0502 0.9400 942
III 0.0195 0.4909 702
IV 0.0618 0.8918 846
V 0.0950 0.9856 890
B
moraine grad. (-1) R2 n
I 0.0505 0.9030 496
II 0.0554 0.9441 940
III 0.0748 0.9487 701
IV 0.0871 0.9612 842
V 0.0950 0.9856 890
It was concluded that the largest lichens should be
omitted in the following way: if there was a gap in
the frequency distribution of at least 2 size-classes (6
mm), all thalli falling to the right of such gaps were
excluded. After making this correction, subsequent
regression lines were constructed (Figure 3, dashed li-
nes). Lichen population frequency gradients calcula-
ted for the moraines of Fláajökull show a logical sequ-
ence (Figure 3, Table 1 B) only after implementation
of the above described correction. It is worth empha-
sising that the correlation parameter (R2) also gradu-
ally increases from the oldest to the youngest moraine
(Table 1B). The lichen population measured on mora-
ine ridge V appears to be single, does not exhibit any
extraordinary large thalli, and therefore, it is not ne-
cessary to implement the above described correction.
The frequency gradients (after corrections) were
plotted on the Bradwell’s (2004) dating curve for SE
Iceland and the age of moraine surfaces was obtained
(Figure 4). The new dates match quite closely the da-
tes previously derived by Dąbski (2002), see Figure
1, for older moraines, but there are some discrepan-
cies concerning younger ridges (Table 2).
Results of the frequency gradient analysis for li-
chen populations with a distinction between proximal,
distal and upper surfaces of stones are shown on Figu-
res 5–9. The following patterns are seen:
1. Lichens growing on the proximal (northern) si-
des of stones constitute the smallest measured
population on all moraines. However, their
share of the total measured population of li-
chens seems to gradually increase from 16–
19% to 26% as moraine age increases. Lichen
populations on the upper surfaces of stones are
usually the largest.
2. With only one exception from the oldest mora-
ine ridge, the largest thalli grow on the upper or
distal (southern) stone surfaces.
3. Proximal populations exhibit the smallest fre-
quency gradient on young moraines and the hi-
ghest on old ones, but there is a gradual dif-
ferentiation between the moraines (Figure 10),
which suggest that it is time dependent.
4. Proximal populations on each moraine are cha-
racterised by the least value of R2.
24 JÖKULL No. 57