Jökull - 01.12.1991, Qupperneq 30

Fig. 8.
a) Grímsvötn viewed from the northwest on June 23, 1938. The subsided area is in the centre of the picture (photo
Danish Geodetic Survey. © Copyright Iceland Geodetic Survey). Grímsvötn og sigdæld norðan þeirra 23. júní
1938. Horft úr norðvestri.
b) Grímsvötn and the subsided area seen from the northeast on August 30, 1938. The arrow points to the dark
layer that covers the northern part of the area (photo Danish Geodetic Survey. © Copyright Iceland Geodetic
Survey). Grímsvötn og sigdældin úr norðaustri 30. ágúst 1938. Örin bendir á dökkt lag sem þekur hluta
sigdœldarinnar.
ried out radio-echo soundings in the area in 1987 and
1991 and found that the location of a north trending
subglacial ridge coincides with that of the subsided
area of 1938. Further, he suggested that the ridge
may have been formed in an eruption in 1938.
The dark ash layer observed in the northernmost
part of the subsided area in July and August 1938,
suggests that an eruption broke through the surface
after May 28. Hence, volcanic activity may have per-
sisted for some weeks. However, its vigour was
probably not great after May 28, as the extent of the
subsided area does not seem to have increased sig-
nificantly during the summer.
The course of events in 1938 may have been as
follows: An eruption started on a several km long N-
S trending fissure to the north of the northwest corner
of Grímsvötn, probably several days prior to the
onset of the jökulhlaup. By May 28 the eruption had
melted at least 2 km3 of ice. The meltwater drained to
Grímsvötn and caused a sudden rise in water level
and rapid opening and draining of the lake. Before
May 28 anyway, the eruption was entirely subglacial
but later renewed or continued activity broke the sur-
face of the glacier and a small ash layer was spread
over the northem part of the subsided area.
The bedrock in the area of the subsidence is fairly
28 JÖKULL, No. 41, 1991