Jökull - 01.12.1964, Qupperneq 7

above procedure for computing the thickness
of the ice. The largest uncertainty lies in the
assumption of a constant regional anomaly,
which has to be separated from the effect of
the ice-sheet. The use of an infinite sheet-
model in calculating the ice-thickness may in-
troduce minor errors and will smooth out sub-
glacial topographical features which are not
large compared with the ice-thickness. Altitudes
of some of the stations were determined by an
altimeter and may be in error by a few meters.
Errors due to this are, however, small in com-
parison with the uncertainties mentioned above.
A first estimate of the ice-thickness in the
Grímsvötn depression and the area east of it
may be obtained by assuming a regional ano-
maly of —36 mgal here also. This gives a thick-
ness of about 110 m near the centre of the
depression decreasing to a very srnall value at
the ridge at the eastern edge of the depression.
The bottom of the depression would then be
at an altitude of 1270 meters. Further to the
east of Grímsvötn the ice-thickness would, with
the same assumption, be 350—400 m with no
major variations in the area surveyed.
These estimates of absolute ice-thickness in
and around the Grímsvötn depression are prob-
ably too low, as it is likely from the general
picture of the Bouguer anomalies in lceland
that the regional anomaly in the Grímsvötn
area is smaller than —36 mgal, giving a larger
thickness to the ice. It is, however, likely that
the variations in ice-thickness in the Grímsvötn
area are of the right magnitude. Thus, for in-
stance, the height of the subglacial ridge at the
eastern edge of the Grímsvötn depression is
very likely 150—200 meters above the bottom
of the depression.
Seismic reflection determinations of ice-
thickness have been attempted in the Gríms-
vötn depression. Joset and Holtzscherer (1954)
give a thickness of 500—700 m at stations in the
depression. Unpublished results of J. Martin
(Thorarinsson, pers. comm.) indicate a thickness
of about 500 m at a station in the eastern part
of the depression, and a thickness of about
540 m at a station 3 km southeast of the hut
on Grímsfjall. If these thicknesses are correct
the regional Bouguer anomaly in the Gríms-
vötn area is —10 to —20 mgal instead of the
one of —36 mgal, which was deduced near the
centre of the profile Grímsfjall towards Kverk-
JÖKULL 1964
Fig. 4.
fjöll. This would then indicate a somewhat
greater ice-thickness near the southwestern end
of this profile than is given in Fig. 3.
It should be mentioned that the seismic
measurements in the Grímsvötn depression are
as yet hardly conclusive in regard to ice-thick-
ness. Several of these measurements did not
yield usable reflections, which is probably due
to the very rough topography, which may be
expected under the ice in this area.
The greatest ice-thickness, 1060 m, which was
found by Joset and Holtzscherer on Vatna-
jökull, was at a station about 6 km east south-
east of Grímsfjall. Such an ice-thickness would
contribute about —76 mgal to the total Bouguer
anomaly. Although this station is located out-
side the area here surveyed, a trend towards
larger gravity anomaly would be expected at
the gravity stations closest to this seismic sta-
tion. Such a trend is not visible on the Bouguer
map.
CONCLUSIONS
Ice-thickness determination by gravity mea-
surements depends on the knowledge of the
crustal Bouguer anomaly due to the underlying
rock formations. This anomaly may be estimat-
65