Jökull - 01.12.1991, Qupperneq 25

Fig. 2. Grímsvötn, bedrock map based
on radio-echo soundings and seismic
reflection. Contour interval 25 m. The
shaded areas under the southern
caldera rim show the eruption sites of
1934 and 1983. The location of the
cross section on Fig. 5 is shown as
well as that of the sinkhole of 1954.
Grímsvötn, botnkort. Skyggðu svœðin
sýna legu gosstöðvannafrá 1934 og
1983 og sniðið á 5. mynd er merkt inn
og einnig staðsetning ketilsins sem
myndaðist 1954.
1974). Since 1953, annual expeditions have visited
the area and Grímsvötn has been inspected from the
air during every jökulhlaup since 1941 (Þórarinsson
and Sigurðsson, 1947; Þórarinsson, 1974; Björnsson,
1988).
In order to study further the volcanic history of
Grímsvötn in this century we have examined data
acquired within the Grímsvötn area since 1934.
These data are written accounts from expeditions, air
photos, old maps of ice surface topography and pho-
tographs. Further, the bedrock topography in the
area (Fig. 2) has now been defined in considerable
detail (Björnsson, 1988; Guðmundsson, 1989).
Potential eruption sites can therefore be inspected for
evidence of eruptive vents.
EFFECTS OF ERUPTIONS IN GRÍMS-
VÖTN; DIRECT OBSERVATIONS
THE ERUPTIONS IN1934 AND 1983
The eruption in March-April 1934 lasted for at least 2
weeks and three craters were active during the emption
(Áskelsson, 1936; Nielsen, 1937; Þórarinsson, 1974).
The craters were all located near the southern caldera
rim and the largest crater was under the southem
caldera wall to the north of the nunatak Svíahnúkur
Vestri (Figs. 2 and 3). The eruption formed an opening
in the ice shelf, 500-600 m in diameter with vertical
ice walls, a few tens of metres high. A semicircular
island had formed in the centre but fans of dark sand or
tephra could be seen under the ice walls to the north
and the west. The two other craters were much small-
er, located in the southwest comer of the caldera. A
JÖKULL, No. 41, 1991 23