Greinar (Vísindafélag Íslendinga) - 01.01.1977, Side 24

22
upon the great cooling about 2500 years ago, such tongues
from Bárðarbunga and Kverkfjallahryggur met in the valley
under the present Dyngjujökull, to form a barrier which in
the end caused the catastrophic flood of that time in the river
Jökulsá á Fjöllum, as recognized by H. Tómasson (14) and
verified upon further study by this co-workers.
This sudden drop in temperature may have led to the forma-
tion of some local glacier caps on high ground which is now
ice-free, and it would be of climatological and glaciological
interest to test this possibility in likely places. In that way,
we might get insight into the early stages of the last Plei-
stocene glaciation of Iceland. From which centers did the gla-
ciers spread? To what extent did they merge together into a
large ice cap? And is it a realistic picture which Þorvaldur
Thoroddsen inferred in the last century, that because striae
do everywhere in the lower ground diverge from the central
area of the country, there must have been a general ice cap,
covering the whole of Iceland? Or were there instead many
separate iceflows, which possibly could be distinguised in the
field by combination of theoretical inference and observations
of striae? These seem to be important problems which would
hardly have been suggested without a theoretical treatment
of the known data.
The result of this chapter is then: 1) There were many se-
parate glaciated areas in Iceland during the Older Dryas Stage.
2) As the maximum extension of the last Scandinavian glacia-
tion was not very much larger than the Older Dryas glacier,
it seems quite possible that rather large areas in Iceland were
ice-free during the maximum of the last glaciation. 3) Blowing
of loose snow by east wind played a significant role in a west-
ward extension of each local glacier. 4) From a volcanological
point of view, it is important to recognize ice-free areas in
Iceland during the last glaciation. 5) The recognition of domi-
nant strong east wind is also of particular volcanological im-
portance because, in accordance with the treatment in Part I,
such wind stress causes an increase of the greater compressive