Jökull - 01.12.1963, Page 21

tions may play some part in ion supply, but it
is unlikely that conditions under the glacier
would favour any significant buiid up of sul-
furic acid concentration as observed in thermal
areas elsewhere i'Sigvaldason 1959).
SUMMARY AND CONCLUSIONS.
The study of the chemistry of three glacier
rivers in Soutli Iceland over a period of one
year indicates the following major sources of
ion supply:
1. Atmospheric origin of major part of the
chloride, and some of the sodium and
sulfate. Bicarbonate in two of the rivers,
Skdlm and Múlakvisl, may partly be of
atmospheric origin.
2. Leaching of the country rock provides all
of the silica and a large part of the alka-
lies and earth alkalies.
3. Effluents from thermal areas beneath the
glacier provide hydrogen sulfide founcl in
one river, Fulilœkur. Bicarbonate in this
river is also of sarne origin. It is believed,
that higher concentrations of sulfate in
the river Skábn is caused by englacial or
subglacial oxydation of hydrogensulfide or
sulfurdioxyde from an active thermal area.
Variations in the concentration of each com-
ponent and total dissolved load of the rivers
are causecl by (1) difference in length of flow
before sampling and (2) different content of
dissolved carbon dioxyde, the water containing
the higher amounts being more effective in
leaching the rock particles during the winter
months.
Establishment of mean concentration levels
of chemical elements in these rivers and normal
yearly variations provides a basis for detecting
unusual fluctuations, which might occur in the
thermal activity beneath the glacier, providecl
that the variations are of sufficient magnitude.
Knowledge of the chemistry of the water over
a longer period will of course help towards a
better definition of the normal state and thus
lower the limits of detection of what might be
considered anomal.
The value of continuous observation of the
chemistry of the rivers as a means of predicting
a volcanic eruption in Katla depends entirely
on the assumption, that a similar increase in
the thermal activity will occur as before the
Askja eruptiön in 1961. Should this be the case,
JÖKULL 1963
there appear to be fair chances that this in-
crease coulcl be detected by chemical analysis
of the river waters. The compounds, which
should be looked for are especially HoS, SO4
and HCO3, or rather the ratios of these com-
pounds to chloride, since chloride is not likely
to be emitted by the thermal activity.
ACKNOWLEDGEMENTS.
The Icelandic National Science Fund (Vís-
indasjóður) has provided funds for this study,
and I ivould like to express rny thanks to the
Fund for its support. Furthermore I would
like to express my thanks to Miss Á. Trausta-
dóttir and Mrs. U. Sveinsdóttir for valuable
assistance during the analytical work. Mr. S.
Rist of the State Electricity Authority provided
information on the discharge of Skálm.
REFERENCES.
Gorham, E., Factors mfluencing supply of major
ions to inland waters with. special reference
t.o the atmosphere. Geol. Soc. of Am. Bull.
V. 72, pp. 795-840 (1961).
Keller, W. D. and A. L. Reesman, Glacial milks
and their laboratory — simulated counter-
paris. Geol. Soc. of Am. Bull. V. 74, pp.
61-76 (1963).
Rainwater, F. H. and L. L. Thatcher, Methods
for collection and analysis of water samples.
U. S. Geol. Survey Water-Supply Paper
1454 (1960).
Sigvaldason, G. E., Mineralogische Untersuch-
ungen iiber Gesteinszersetzung durch post-
vulkanische Aktivitat in Island. Beitr. z.
Mineral. u. Petr. V. 6, pp. 405—426 (1959).
— Some geochemical and hydrothermal aspects
of the 1961 Askja eruption. Beitr. z. Min-
eralogie u. Petrographie, V. 10, Correns-
Festband, pp. 263—274 (1964).
— ancl D. E. White, Hydrothermal alteration
of roclis in two drill holes at Steamboat
Springs, Washoe County, Nevada. U. S.
Geol. Survey Prof. Paper 424-D pp. D-116
to D-122, Art. 331 (1961).
Thorarinsson, S.. On the possibilities of pre-
dicting the next eruption of Katla. Jökull,
9, pp. 6—18 (1959). (In Icelandic with an
English summary).
Veðráttan. The monthly of the Icelandic Mete-
orological Office (1962).
17