Jökull - 01.06.2000, Qupperneq 42

Surface and bedrock topography of Mýrdalsjökull
p. 255 and 262) described the 1823 eruption site in
a detailed diary from the district Álftaver (Figure 1).
He also climbed Austmannsbunga from where he
portrayed the eruption site at the northern slopes of
Háabunga. He described a horseshoe shaped glacier
fissure with a SW-NE direction closest to Austmanns-
bunga but a SE-NW direction close to Háabunga. The
bearing from Vík was by the western slopes of the
mountain Hatta (p. 268). We suggest that this eruption
may have taken place on the 2-3 km long ridge strik-
ing NNW from the eastern rim of Háabunga (Figure
12).
In 1918, Jóhannsson (1919, p. 12 and 47) viewed
two eruption vents from the south but the location was
reported on the northern slopes of Háabunga, 1500
to 200 m north of its highest crest (Sveinsson, 1919,
p. 15 and 57). Rist (1967a), however, suggested that
the 1919 eruption site was farther north, where two
cauldrons suddenly subsided in June 1955 about 3 km
southwest of Kötlukollar (Figures 3 and 4), followed
by a jökulhlaup under Kötlujökull (Rist, 1967b).
Tryggvason (1960) presented seismic data support-
ing the hypothesis that a small subglacial eruption had
taken place in June 1955 where the cauldrons formed.
Our radio echo soundings show a 60 m high mound
with a diameter of 300 m beneath the southernmost
cauldron. The northern cauldron is situated above a
depression in the base between two 100-150 m high
mounds, that may be the remnants of a crater with an
inner diameter of 400-500 m, or a ridge formed during
an eruption. The ice there is 400 m thick.
Björnsson (1970) presented bearings of the 1918
eruption site taken from a coastal vessel located to the
east of the ice cap that combined with a bearing from
Vík gave a location slightly south of the cauldrons
(63 / 37’5”N, 19 / 3’W). Photographs taken by Kjart-
an Guðmundsson 23. June 1919 on Háabunga toward
Kötlukollar (Þórarinsson, 1959, p. 16, figure 8) show
that the crater was neither situated at this position
nor where the cauldrons subsided in 1955. Therefore,
we suggest that the description of Sveinsson (1919)
should be taken literally and that the position of the
crater was as shown in Figure 12.
No observations locate sites of eruptions trigger-
ing jökulhlaups draining toward Sólheimasandur.
During the 1860 eruption a small jökulhlaup drained
this way (Hákonarson, 1860) but the main flood went
down to Mýrdalssandur. Only one eruption site was
reported. However, some melting must have taken
place in the western part of the caldera.
0 5 10 km
? ?
1955
1823
1918
1755
Goða-
bunga
Austm anns-
bunga
Enta
Háabunga
Legend:
E rup tion s ites P resent geo therm a l a rea
? ? suspected fissu re
in 1755 e ruption
C a ldera rim
P resent w ate r d iv ides
Kötlu-
kollar
Figure 12. Likely location of recent eruptions in Mýr-
dalsjökull (1755, 1823, 1918, 1955? A.D.). – Líkleg
lega nýlegra gosstöðva (1755, 1823, 1918 og 1955?).
Prehistoric eruptions have also drained down
Entujökull (Haraldsson, 1981; Sigurðsson, 1988).
Extensive flood deposits, found between Öldufell and
Kötlujökull, may also be produced by volcanic acti-
vity outside the caldera. In 934 A.D. a subglacial part
of the Eldgjá fissure erupted and jökulhlaups drained
down to Mælifellssandur and south of Öldufell to the
river Hólmsá (Larsen, 2000).
JÖKULL No. 49 41