experiments. For a prescribed shortwave radiation flux (i.e., one that does not
depend on cloud conditions), the reduced variability in total cloud amount in sum-
mer caused by interpolating monthly averages to daily values results in greater
equilibrium ice thicknesses in experiments 7 and 8 than those reported for Makshtas
and Timachev's (1992) numerical experiments.
We thus believe that, for sea ice models driven by atmospheric data, reconstruct-
ing total cloud amount from air temperature provides sufficient accuracy for cal-
culating the surface heat budget in the winter and, especially, for studying its
monthly variability (Table 10). On the other hand, in summer, when the surface
temperature is nearly constant at the melting point and the feedback between the
main components of the surface heat budget is consequently weaker, models will
require better cloud data and a more accurate description of cloud effects on short-
wave and longwave radiation.
CONCLUSIONS
We summarize our results in the following conclusions:
Our analysis of observations made on the North Pole drifting stations shows
that the frequency distribution of total cloud amount in the Arctic Basin,
especially in winter, is U-shaped. We fitted these histograms with beta dis-
tributions; Table 4 lists the fitting parameters. Since the mean of a quantity
with a U-shaped distribution actually corresponds to the least likely value
of the quantity, representing Arctic cloud amounts with monthly averaged
values is a flawed approach.
The series of meteorological observations on drifting stations that worked
above 77N from 1955 through 1991 suggests reduced frequency of clear
skies in winter and of overcast skies in summer (see Fig. 6 and 7). The former
may be associated with increasing atmospheric pollution in the Arctic; and
the latter, with a shift in the general atmospheric circulation.
We developed a reliable method for estimating total cloud amount in win-
ter from atmospheric surface-layer temperature. This method yields a fre-
quency distribution for total cloud amount that has a U-shaped distribu-
tion--as it should--although the input air temperature values are normally
distributed. Because modeling results based on this method compare rea-
sonably well with observations, the method is an algorithm that could
improve the remote sensing of total cloud amount.
Meteorological data collected on Ice Station Weddell suggest that Knig-
Langlo and Augstein's (1994) method for handling total cloud amount in
parameterizations of incoming longwave radiation in polar regions is the
best among five popular candidates.
We have confirmed the value of the KL&A parameterization through numer-
ical experiments using a one-dimensional thermodynamic sea ice model and
an annual cycle (October 1982 to October 1983) of meteorological observa-
tions from North Pole 25. This modeling shows that, of the five param-
eterizations considered, the one from KL&A produces estimates of equilib-
rium sea ice thickness and its seasonal variability closest to existing notions
about those quantities (Table 9). Associated calculations suggest that in
winter the atmospheric surface-layer temperature is largely an integrated
parameter because of the essential three-way feedback among air tempera-
24
Previous Page
