Jökull - 01.12.1963, Blaðsíða 20

ponents. The concentration levels for individual
compounds are quite similar for all three rivers,
with the exception of bicarbonate, silica and
sulfate.
Chloride is probably for the most part derived
from the precipitation. Rainwater at Rjúpna-
hæð near Reykjavík (Table 1) contains from
2 to 13 ppm C1 and rain from the weather
station Vegatunga farther inland contains 1 to
9 ppm Gl. The range of chloride content in
the riverwaters is from 4 to 16 ppm C1 the main
bulk of analysis lying below 8 ppm Cl.
Sulfate in rain water varies within narrow
limits from 1.5 to 3.5 ppm SO4. The sulfate con-
tent in the rivers is always higher, the highest
value being 15.4 ppm SO4 in the March sample
from Skáhn. Gorham (1961) suggests that ”snow
stored as glacier ice may loose salts through se-
lective washout of intercrystalline brines during
summer melting periods”. If such a mechanism
is important a direct comparison of the chemist-
ry of precipitation and the melt water is of
little value. It would however be expected, that
the ratio SO4/CI in the meltwater should be the
same or slightly less, than in the precipltation.
This is not the case and some other source of
sulfate is to be expected. Comparing the sulfate
content in the three rivers, Skálm shows distinct-
ly higher values throughout. Furthermore the
SO4/CI ratio is heigher in Skálm than in the
other rivers, the average of the nine analysis
is 1.8 (range 1.5 to 2.3). The average SO4/CI
ratio for Múlakvisl is 1.3 (range 1.0 to 1.7) and
for Fúlilækur 1.1 (range 0.6 to 1.5). The minera-
logy of the suspended material does not provide
any evidence, which could throw light on these
variations, and the most likely explanation is
influence from thermal activity. Fúlilœkur,
which carries detectable amounts of hydrogen-
sulfide shows the lowest sulfate concentraton of
the three rivers. If all sulfide carried by Fúli-
leekur were oxydized the resulting sulfate con-
tent would however, approach the values found
in Skálm, A possible explanation is that the
thermal areas feeding the hydrogen sulfide into
the glacier melt water are at different distances
from the glacier snouts. Water from the drainage
area of Skálm would accordingly have received
its sulfur compounds at a considerable distance
from the snout and the sulfide woulcl oxydize
during flow in englacial or subglacial channels
where the yield of sulfate would probably be
higher than in an open stream. Fúlilcekur on
the other hands is probably fed by thermal ef-
fluents at a relatively short distance from the
glacier snout and sulfide is only partly oxidized
at the headspring. Running as a open stream
the water quicklv looses the HoS by airation.
Silica, the alkalies and earthalkalies all show
slightly higher values in Skálm during the sum-
mer months. The reason for this difference has
already been indicated in the discussion on to-
tal dissolved load. All silica and a major part
of the cations is derived from leaching of the
rock material. I.eaching by acid thermal solu-
JÖKULL 1963
16