Jökull - 01.12.1969, Qupperneq 48

Forecasting Drift Ice at Iceland by Means
of Jan Mayen Air Temperature
PÁLL BERGTHÓRSSON,
THE ICELANDIC METEOROLOGICAL OFFICE, REYKJAVIK, ICELAND
ABSTRACT
In the last years the drift ice off the lce-
landic coasts has increased markedly, causing
various implications in shipping and other
activities. Ice forecasts would therefore be of
great importance for the economy of Iceland.
Considering the ocean currents north of Ice-
land the air temperature of the islancl Jan
Mayen is thought to be of prognostic value for
the ice conditions at Iceland approximately
half a year in advance. This hypothesis is to
some extent verified by an investigation shoxu-
ing that the summer and fall temperature at
Jan Mayen is rather closely correlated with
drift ice incidence at Iceland in the folloxving
xvinter, spring and sumrner. In this xvay it is
possible to issue at the end of November an
ice forecast for Iceland, valid until the end of
September in the next year. During this period
the forecast for the next month may at any
time be amended according lo the east xuind
componeht north of Northxvest-Icelancl during
the last xveeks.
INTRODUCTION
It is well known that the drift ice arriving
to Iceland is generally flowing southward along
the East-Greenland coast. It does not change
this fact that some rotation and irregular fluc-
tuations are often observed in a broad area
near the ice edge, sometimes expanding the ice
area, sometimes making it narrower. In the
author’s opinion the ice edge may, in spite of
these vortices, be considered as a quasistation-
ary streamline in the sea surface, made visible
bv the transition from open to frozen sea. At
least it can be assumed that the currents across
this boundary must be rather insignificant in
the long run. In Fig. 1 we show the ice bound-
44 JÖKULL 19. ÁR
ary in late winter as observed by Nansen in
the latter part of the 19th century (Nansen
1924). The “streamline” is here lying just
southeast of the island Jan Mayen. From there
it meanders irregularly towards Iceland. In the
drawing Nansen shows some weak vortices in
the currents of this area. On monthly pressure
maps the isobars are generally parallel to the
ice edge as shown in Fig 1, with dominating
northeasterly winds. That is a further indica-
tion of the usual track of the pack ice visiting
North-Iceland.
This current picture is by ancl large veri-
fied by drift bottle observations. In Fig. 2 we
have plotted the positions where 12 drift
bottles recovered in Iceland have been launched
in three diíferent years, 1896, 1949 and 1951.
The drift period median of these bottles was
164 days (5.5 months), ranging from 110 to 438
days (3.7 to 14.6 months). This goes to show
that generally the currents towards southwest
are slow in this area. Even if we subtract 10
to 20 days from these drift periocls, allowing
for some delay in finding the bottles ashore,
we get 5 months in the median. Furthermore,
most of the bottles have only drifted some 50
to 90% of the whole distance from Jan Mayen
to Iceland. The drift time from there to Ice-
land is therefore hardly less than half a year
on the average, i.e. some 2 nautical niiles a day.
This is quite different from the swift East-
Greenland current further west (Fig. 1), ancl
many erroneous conclusions have been drawn
because people have assumed that all the East-
Greenland ice was moving with that great velo-
city.
In this connection it should be mentionecl
that from hydrographic conclitions and other
observations Stefánsson has concluded that in
the area between Iceland and Jan Mayen there