the National Center for Atmospheric Research
Siberian, and Chukchi Seas (RSMOT, in prep.). To
(NCAR 1990). The results of the calculations were
avoid icing, ships must reduce their speed, change
used for statistically estimating the probability of
course, or seek shelter, thus increasing the time of
wind direction and wind speed at each point on
transit.
our grid. We compared these simulated wind sta-
The probability values for icing used in the
tistics with observed data at meteorological sta-
model were obtained from RSMOT (in prep.). The
tions located along the Northern Sea Route, and
icing data are read from ICEFOG**.DAT files
found reasonable agreement between observed
(Table B.4) using subroutine GETDAT.
and simulated data.
The model reads the tables of wind speed and
Snowstorms
direction observations from WINDS**.DAT files
Lack of visibility is cause for slowing a vessel
(Tables B.10 and B.11) using subroutine GETDAT.
when operating in ice concentrations of 30% and
greater (Gordienko et al. 1967, Gordienko 1977,
Icing
Himich 1977). Diminished horizontal visibility is
Strong winds, cold air, and water contribute to
very important, especially in the autumnwinter
accumulations of topside ice on vessels. Icing along
period when limited visibility due to fog and snow-
the Northern Sea Route is not a serious problem
storms is combined with darkness. In conditions
for large cargo ships, but along some routes (e.g.,
of limited visibility, ships can lose a channel or
Murmansk to Igarka) icing can be very danger-
become icebound in the channel, interrupting con-
ous, especially at the end of autumn, when air
voy motion. Work by icebreakers to free icebound
temperatures are below zero and there is no ice
ships and to reorganize the convoy adds to the
cover on the sea surface. In the Arctic seas, icing of
total transit time and decreases the efficiency of
vessels may occur throughout the year from either
commercial navigation.
atmospheric (rime icing, freezing rain, and the like)
We regarded the occurrence of snowstorms
or marine sources (freezing sea spray). From De-
along the route as one of three visibility factors to
cember through June, only atmospheric icing is
affect the speed of ship transit. Fog, another me-
possible due to the sea-ice cover. From July through
teorological factor, and darkness were considered
October marine icing accounts for 50% of all cases
similarly. Since probability of occurrence data were
of icing, mixed icing for 45%, and atmospheric
available for snowstorms and fog, and a simple
icing for 5%.
algorithm could simulate the occurrence of dark-
Duration of an icing event is 12 hours in 74% of
ness, these three slowing factors were integrated
cases, and the maximum duration is 7 days. In
into the model. Snowstorms occur only rarely in
September, slow icing occurs 2040% of the time
summer, so for modeling purposes we assumed
in the coastal areas, and 5070% of the time in the
that they would not occur in the months of June
central parts of the Arctic seas. Slow icing, for a
and August.
300- to 500-t displacement ship, is defined by
The snowstorm probability for April and Octo-
RSMOT (in prep.) as less than 1.5 t/hr mass rate
ber was digitized from maps presented in
of accumulation or less than 1 cm/hr thickness
Proshutinsky et al. (1994) that were derived from
rate of accumulation. The occurrence of fast icing,
the data of Mozalevskaia and Chukanin (1977),
defined as a 1.5 to 4 t/hr (or 13 cm) rate of accu-
The Soviet Arctic (1970), Polkhova (1980), and
mulation, ranges from 15% of time in the south-
Sergeeva (1983). In the model, snowstorm prob-
ern parts to up to 10% of time in the northern
ability is read from ICEFOG**.DAT files (Table C.4)
regions of the Arctic seas. These values increase
using subroutine GETDAT.
by about 10% in October. In the Barents Sea, the
frequency of marine icing varies from the begin-
Fog
ning of January to mid-March, and the maximum
The frequency data for the occurrence of fog is
frequency of occurrence of 78% is observed in Feb-
from Proshutinsky et al. (1994). These data were
ruary. Atmospheric icing is possible in the Arctic
digitized from maps appearing in The Soviet Arc-
seas throughout the year because negative air tem-
tic (1970) and Brower et al. (1988). The probability
peratures are possible at any time. Atmospheric
of fog occurring at each point is read from
icing has been observed 3050 times per year in
ICEFOG**.DAT files (Table B.4) using subroutine
the Kara Sea and 8090 times in the Laptev, East
GETDAT.
22
Previous Page
