Jökull - 01.12.1986, Síða 54

Table 3:
Sample groups 1 and 2 from Ketilsstadir (for discus-
sion, see text).
Ward’s method/
Euclidean distance
groupings Sample number
3
2
3
3
1
3
2
3
2
1
2
1
1
3
3
1
2
2
1
2
2
KEl/1
1/2
3/1
4/1
BRl/2
KEl/5
2/2
3/2
4/2
KE/SUPP
BRl/3
1/4
1/5
Post Land- 1/6
nám 1/7
1/8
1/9
1/10
KEl/3
BRl/11
1/12
Group average/
Jaccard
groupings
3
2
3
3
1
3
2
3
2
1
2
1
1
3
1
1
2
2
1
2
2
3 1/13 3
3 KEl/6 3
3 4/3 3
3 BRl/14 3
3 1/15 3
3 KEl/4 3
3 BRl/17 3
3 PreLand- 1/18 3
2 nam 1/19 3
3 1/20 3
3 1/21 3
3 1/22 3
3 KEl/7 3
3 1/8 3
3 1/9 3
3 1/10 3
3 1/12 3
2 1/13 2
3 1/14 3
3 1/15 3
Group 1 Group 2 Group 3
Post Landnám 7 8 6
Pre Landnám 0 1 19
52
the tephra layers. The use of several contemporaneous
samples, facilitated by tracing tephra horizons across
the former surface of the bog (Buckland et al., 1981),
can help to clarify spatial variations in the palaeo-
environment, although the low diversity of the Ice-
landic fauna and the relatively small numbers of
individuals in the Ketilsstadir samples raises doubts as
to the statistical validity of any comparisons. In such
situations, Perry et al. (1985) have recently advocated
the use of cluster analysis and two of the hierarchical
techniques used therein were applied to the Ketils-
stadir data (table 3) using the CLUSTAN package
(Wishart, 1978; for a further example, see Perry,
1981). Three broad groupings are apparent4. Group 1
is exclusively post-Landnám and group 2 is pre-
dominantly so. Group 3 is represented throughout the
whole sequence. Taken together groups 1 and 2 are
clearly indicative of a pronounced environmental
change in the post-Landnám period (fig. 7). Taking
group 1 to be the most extreme, the respective faunas
clearly show the environmental impact of man’s pre-
sence and the greater habitat diversity created by his
activities. The analyses support the suggestion that the
bog became intermittently less wet after Settlement,
with an increased minerogenic input and a more
diverse flora. Such effects can be explained by over-
grazing on the surrounding slopes, medieval peat
exploitation and introductions by man. That the
Group 1 type faunas are not in the majority in the
post-Landnám period is probably due in part to
sampling effects, but, most importantly, seems to be a
consequence of the —1357 eruption of Katla. The
post —1357 faunas are markedly impoverished, sug-
gesting that this ashfall had a major environmental
impact, which is still evident in the highest samples.
Although the modern faunas grouped together with
the post-Landnám samples, they were not in Group 1.
There would seem to be two reasons for this: they re-
present a biased snapshot of the actual living faunas,
and are necessarily less diverse because of this, and
the present day environment at Ketilsstadir is un-
doubtedly a further development from the one under
consideration.
Species groupings were also detected (table 4) and
provide a more detailed indication of the impact of
man and his livestock. Group 1 contains species
which are the characteristic ‘core assemblage’ of the
bog environment, occurring right through the succes-
sion. They emphasise the damp, vegetated nature of
the former bog surface and the presence of ephemer-
al pools. This is not an association which can be re-
garded as typical at the present day, which presum-