Jökull - 01.12.1969, Qupperneq 49

Fig. 1. Normal ice
limit in late winter,
by Nansen (1924).
is a weak cyclonic current (Stejánsson 1962).
These observations are however mainly made
in icefree years at Iceland, and one has to be
careful in concluding from them what is the
general current piclure in years with moderate
or heavy ice.
According to the reasoning above it seems
possible tliat the waters which bring the drift
ice to North-Iceland in April or May, were
situated in the vicinity of Jan Mayen in the
summer and the fa.ll, perliaps in August to
November. If tliis is true it might be useful
to investigate thoroughly the conditions in the
tvaters surrounding Jan Mayen in the summer
and autumn, in order to forecast the ice in the
coming ice season in Iceland.
But even if such observations are not avail-
able, there is some reason to think that weather
observations at Jan Mayen may be of some
value in this respect. We shall here in parti-
cular discuss the hypothesis that the air temp-
erature at Jan Mayen is a measure of the
hydrographic conditions of the sea surface in
the vicinity of the island, thus making possible
a long range ice forecast for Iceland.
The trajectories of all winds arriving at Jan
Mayen are lying over the sea or the sea ice
for a long distance, and thus the air tempera-
ture must be strongly influenced by the temp-
erature of the sea surface surrounding the is-
land. This will be even more true in summer
than winter. In summer the air is generally
stable and a little warrner than the sea, or
only slightly colder. It will therefore reflect
closely the temperature of the sea surface. In
winter the air shows much greater fluctuations
in temperature. Stable air has approximately
the sea temperature, but unstable air may be
much colder. Since the proportion of stable air
to unstable air is highly variable in winter, the
correlation of air and sea surface temperature
is then less than in sunmier. Even variations
JÖKULL. 19. ÁR 45