Jökull - 01.12.1990, Qupperneq 144

April 1990) in the extreme low-precipitation area,
probably needs such an origin as explanation. In-
filtration from the braided Jökulsá in summer could
account for some part of this discharge, but that does
not change the glacial origin of the water.
Interpolation of the chloride content implies, that
the springs in Hólmatungur have a drainage area with
a centre not more than 20 km towards S, and that
only if the chloride in the springs is thought to be en-
riched by admixture of geothermal water. A centre
for the drainage towards lake Mývatn (analyses from
Ólafsson, 1979) would be N of Dyngjufjöll and for
Herðubreiðarlindir NE of the same mountains. Árna-
son (1976) presumes, on the base of his deuterium
mapping, that the springs at Mývatn have a similar
drainage basin as is here indicated. He also states
on the same basis, that the water in the springs of
Suðurárbotnar and Hólmatungur is of a glacial ori-
gin. This is possibly to a certain degree correct for the
water in Suðurárbotnar, but the chloride contents, the
hydrogeological conditions and the high groundwater
level in Dyngjufjöll present strong arguments against
such a solution for the spring-water in Hólmatungur.
Here again, a disregard of the altitudinal effect of the
mountains on the deuterium contents, together with
strong seasonal variations in the ”local“ springs and a
lack of proper springs for sampling may have led to
a confused mapping and too narrow an interpretation.
Most of the springs in the Kverkfjöll fissure zones,
the springs E and NE from Dyngjufjöll and those in
Suðurárbotnar show signs of a geothermal influence
(sulphate >5 ppm, sodium > 10 ppm). This is not the
case in the springs on upper Skjálfandafljót, despite
the presence of local thermes.
The groundwater stands probably high in the al-
tered and impermeable rocks of the central volcanoes
Dyngjufjöll and Kverkfjöll (in lake Öskjuvatn more
than 1,000 m a.s.l.), causing a wide distributionof the
flow of thermal water and at the same time blocking
off the flow of glacial groundwater along the fissure
zones. A glacial origin is not necessary for the spring
water in Suðurárbotnar on chemical grounds alone,
but the volume of the water (together with the river
Svartá, probably near to 20 mJ/s) requires an exten-
sive drainage area, at least reaching close to the margin
of the glacier. Thus the only groundwater in the re-
gion with a major probability of a glacial origin is in
the upper reaches of Skjálfandafljót and on Jökulsá
á Fjöllum, E and SE from Dyngjufjöll. It could be
to some degree present in Suðurárbotnar and in the
Kverkfjöll fissure zone, where geothermal influence
can be traced in the relatively deep circulating water
for a distance of up to 30-50 km away from the central
volcanoes Dyngjufjöll and Kverkfjöll, respectively.
GROUNDWATER HYDROLOGY OF THE
TUNGNAÁAREA
The area SW from western Vatnajökull is in this
discussion unique in that regard, that a comprehensive
hydrological model has been constructed and success-
fully operated for the area (Vatnaskil, 1987; 1988).
Although the basic meteorological data are for a good
deal based on estimates and calculations and the hy-
drogeological base was only a preliminary draft for
a hydrogeological map, the model fits pretty well
to check-values like the discharge of springs or the
groundwater level. The groundwater currents in the
model correspond, at least roughly, to what can be
concluded for their location from the chemical clas-
sification. The results of the model operations must
be given a considerable credibility, despite the still
incomplete basic data (Fig. 22). The previous maps
of the groundwater flow by Sigbjamarson (1972) and
Ámason (1976) differ from the one in the model, es-
pecially with regard to the Veiðivötn area proper. The
differences seem to be a matter of interpretation, based
on the scope of information, available at the respective
times. Yet it should be noted, that all the authors pre-
surne a considerable amount of glacial groundwater
entering the ice-free basin and passing through it.
The total outflow in great springs is estimated to be
near to 80 m3/s (Vatnaskil, 1987), but the total ground-
water flow is certainly somewhat more. A rough sum-
marization of the infiltration values used in the model
gives an area of the glacial drainage area of near to
600 km2 (Vatnaskil 1988), with a mean infiltration of
near to 2,000 mm/year, which yields near to 35 m3/s
total runoff. Similar estimates for the ice-free, highly
permeable part of the basin (2,000 km2, over 1,000
mm/year) give an runoff of near to 65 m3/s, or ap-
140 JÖKULL, No. 40, 1990