Rit (Vísindafélag Íslendinga) - 01.06.1976, Page 53

to that of the winter precipitation. The profiles of V—1, V—10 and
V—18 are located at 1300, 1400 and 2000 m above sea-level respec-
tively and therefore give good examples of how the extent of the
homogenization depends on the runoff ratio, i.e., the fraction of
water that percolates through the snowpack.
At V—1 approximately 50% of the precipitation escapes as runoff
during the summer thaw. At V—10 possibly 30% escapes as runoff
and at V—18 the amount of meltwater is so small that it does not
markedly affect the initial ð-variations of the snowpack. At V-18
the glacier can, therefore, obviously be classified as a percolation
zone.
The 8-variations in the last winter layer in V—1 are of the same
magnitude as those in the precipitation (Árnason et al. 1967). The
same is also the case for V—10 and V—18. In the previous annual
layer, below the summer surface of 1967, extensive homogeniza-
tion has apparently occurred in V—1. At the same time, the remain-
ing firn has apparently been enriched in such a way that it becomes
significantly richer in deuterium than even the summer precipi-
tation (see Table 6). Thus the only way to explain the enrichment
of the heavier isotopes in the firn at V—1, is to assume that isotopic
exchange occurs as the snowpack recrystallizes and the ice-water
system tends to reach equilibrium conditions. Ice in equilibrium
with water will contain approximately 20%0 and 3%0 more deu-
terium and O18 respectively than the water (O’Neil 1968, Ámason
1969 a,)
At V—18 the amount of meltwater is so small that it does not
markedly affect the deuterium variations in the snowpack. In this
case, the ö-variations are apparently preserved in the firn down to
a depth of approximately 20 m, which corresponds to 6 years of
snow accumulation. At that depth, the 8-variations are rather rapid-
ly smoothed out, yet without noticeable changes in the mean isotopic
content of the firn.
In order to study the isotopic exchange during fimification on
temperate glaciers more quantitatively, a model experiment was
carried out in the laboratory (Árnason 1969). In the model ex-
periment a fine-grained and isotopically homogenous snow column
gradually melted from the top. The drain water was collected at
the bottom of the column and its deuterium content measured. An
L