Jökull - 01.12.1990, Qupperneq 110

Figure 4. Core no. 1, showing the cov-
erage of the integrated samples 1.1 and
1.2 (Table I), the coverage of the down-
core samples 1.1, 1.2, 1.3... etc. (Ta-
ble II), the locations of ice lenses in
the core and some downcore chemical
changes. The ice lenses are shown by
the hatched lines but the length of each
core step is shown by the solid lines.
The 1986-1987 precipitation layer is
shaded. — Linurit fyrir 1. Kjarna.
Orvarnar sýna hvar heildarsýnin 1.1
og 1.2 (Tafla 1.) eru tekin úr kjarn-
anum en ferningarnir hvar sýnin í Töflu
2 eru tekin úr kjarnanum. Þverlínur
eru dregnar í kjarnann við hvern bor-
unaráfanga en 1. Kjarni var borað-
ur í 10 áföngum. Brotalínurnar sýna
hvar íslinsur eru í snjónum. Snjólag-
ið frá 1986-1987 er skyggt. Hœgri
hluti myndarinnar sýnir hvernig sýru-
stig, styrkur klóriðs og styrkur natríums
breytist niður eftir kjarnanum (Tafla2).
to its base. The spatial variation in ion concentrations
(Table I), that is the increase in ion concentrations
with increased altitude and downcore variation in pH
and ion concentrations, can be explained by partial
melting of snow resulting in a chemical fractionation.
During melting the ions are preferentially released to
the meltwater leaving behind purified snow. The ions
are washed with the meltwater, down to the frozen part
of the core where they are trapped. This is at about
190 cm depth in core 1 and 2. The 1986-1987 layer
is poorer in dissolved solids than the 1987-1988 layer
above, its concentration is more homogeneous and the
pH is higher than in the top layer. The average total
salt content (Na, K, Ca, Mg, Cl, S04) in the top layers
in core 1 and 2 is the same, 1.8 mg/kg but in the layer
below it is 0.44 and 0.69 mg/kg, respectively. The
average pH of the samples in the top layers in core 1
and 2 respectively is 5.57 and 5.59, but 5.85 and 5.73
in the layer below. It should be noted that the lower
layer in core 1 is more depleted in salts, and the pH is
higher than the one in core 2.
The chemical changes occuring at the boundary of
precipitation layers can aid in locating the boundary.
In the field, the definition is based on the frequency
of ice lenses, the size of ice and snow crystals (Fig. 3)
and sometimes the presence of dust at the boundary.
Ice lenses are numerous at the boundary, caused by the
summer melt, and ice crystals are bigger in the older
layer (Figs. 3,4, 5 and 6).
Some of the spatial changes in snow chemistry on
the western part of Vatnajökull (Table I, Fig. 2) might
be due to a ”rain shadow effect“. Björnsson (1988)
has measured a decrease in the winter precipitation on
a profile from Pálsfjall, north west towards Kerlingar
(Fig. 2), which he suggests might be a combination of
altitude and a rain shadow effect.
The pH values of unmelted 1987-1988 snow on
Vatnajökull, represented by the unmelted fraction of
106 JÖKULL, No. 40, 1990