Jökull - 01.12.1972, Qupperneq 53

bulb temperatures in the screen. ltie radia-
tional error was defined as difference between
Assmann values and the screen values.
Many measurements were taken every day by
the ventilated psychrometer and the radiation
error in the screen was found to be appreciable.
No such error can be tolerated in the gradient
measurements. Therefore it was decided to use
the spot values from the Assmann measure-
ments in the calculations. Registrations in-
dicated that these values were fairly repre-
sentative for the mean temperature, less for
the relative humidity. Vapor pressures derived
from psychrometer measurements are taken to
be accurate to 0.1 mb assuming a possible error
of 0.1 °C in the temperature readings. For
periods when Assmann measurements were
lacking the registrations were calibrated by us-
ing an empirically derived relation for the
radiation error as a function of the global radia-
tion and the wind speed. The radiation error
was considered negligible during the night. This
procedure gave 3 hr. values during the whole
period. Daily means have been published by
Björnsson (1971, fig. 5 and 6). Vapor pressure
was found to increase with height above the
snow in nearly all cases as would be expected
with the stable conditions present. Application
of gradients of temperature in 1968 rather than
absolute values as in 1967 in the determina-
tion of the sensible heat term reduced the
effect of any radiational error to minima.
Short-wave radiation, albedo, cloudiness
The global radiation was registered by an
actinograph (Robitzch) mounted horizontally
at a height of 2.5 m above the glacier surface.
The calibration was made by comparation to
integrated values of a Kipp et Zonen solari-
meter at the Norwegian Meteorological Insti-
tute, Oslo. Values of total daily global radia-
tion were obtained by integrating the appro-
priate radiation curve on the strip chart. (See
fig. 5 and 6, Björnsson 1971). According to
procedures given by G. Wendler (1964) these
values have been used to calculate an areal
estimate for the incoming short wave radiation
on the glacier. For the period July to early
August the ratio between incoming radiation
and global radiation was calculated for various
directions of slope and slope gradients on the
glacier. Fig. 4 gives the slope gradients found
for the various slope directions. For the ob-
servation periods considered it was found that
for the glacier as a whole, the incoming short
wave radiation was roughly 84% of the global
radiation. Results are given in Fig. 5 and 6.
Measurements of albedo were mostly taken
with clear sky or totally overcast sky and no
measurements were made in varying cloudiness.
Readings were both done by a horizontal in-
stalled solarimeter and by holding it parallel
to the sloping glacier surface. It was not pos-
sible from the data material to determine in
details how the albedo changed with changes
in the snow surface layer and the spectral
distribution of tlie incoming radiation. Early
in the summer, when there was freezing during
the night and melting during the day, some 10—
15% lower values were found in the middle
of the day than in the early morning and the
late evening. On the other hand, later when
there was melting during the whole 24 hours,
no significant variation with solar altitude or
cloudiness could be traced from the spot values.
GRADIENT SLOPE
Fig. 4. The slope gradients of Bægisárjökull
for various directions of slope.
Mynd 4. Hallastefna Bœgisárjökuls og halla-
stigull.
JÖKULL 22. ÁR 51