Jökull - 01.12.1977, Qupperneq 73

The Cause of Jökulhlaups in the Skaftá River, Vatnajökull
HELGI BJÖRNSSON,
SCIENCE INSTITUTE, UNIVERSITY OF ICELAND
REYKJAVÍK, ICELAND
ABSTRACT
For many decades frequent jökulhlaups have
occurred in the river Skaftá. Since 1955 thirteen
jökulhlaups in Sltaftá have been studied. All
the jökulhlaups since 1955 liave been accom-
panied by sinking of an ice cauldron 10 km
north-west of Grímsvötn. The interual between
the jökulhlaups has been from 1 to Ji/4 years,
and tlie volume has varied from 50 ■ 108 m3 to
250 ■ 10e m3.
One concludes that tlie jökulhlaups since
1955 originate from a water storage which is
accumulated at a subglacial geothermal area,
releasing lieat equivalent to 700—800 MW. Pre-
sumably tlie geothermal area is a part of a geo-
thermal field which includes the Grimsvötn
area — the Grimsvötn geothermal field.
All tlie jökulhlaups in Skaftá may originate
from tlie same geothermal field. Before 1955 the
jökulhlaups were much smaller in volume than
most of those which have occurred later. The
Grimsvötn water basin may have diminished in
area and gradually left a larger part of the
Grimsvötn geothermal field outside the water
basin. Alternatively, geothermal activity may
have migrated across the water divide of tlie
Grimsvötn water basin. However, there is no
indication that the total strength of geothermal
activity has mcreased recently in the central
parl of Vatnajökull.
INTRODUCTION
Thirteen jökulhlaups have occurrecl in the
river Skaftá during the last 23 years. The first
jökulhlaup in 1955 was accompanied by the
formation of an ice cauldron about 15 km
north-west of Grímsfjall (Figs. 1 and 2). The
diameter of the cauldron was about 2.5 km and
its maximum depth was estimated 100—150 m
(Thorarinsson and Rist 1955). After the jökul-
hlaup the cauldron became gradually shallower
due to inflow of ice. But a new jökulhlaup
occurrecl before the cauldron had quite dis-
appeared and the ice sank down again. This
has been a characteristic sequence for the jökul-
hlaups from the cauldron.
No ice-dammed lake similar to Grímsvötn has
been observed at the ice cauldron. As far as
one knows, the area is covered with a 500—600 m
thick glacier. However, no detailed data are
available on the bottom topography. A strong
sulphurous smell during the jökttlhlaups proves
that a substantial part of the burst water is
melted at the glacicr bed. But volcanic activity
has not been observed in Vatnajökull since the
eruption in Grímsvötn in 1934.
During the jökulhlaup in 1971 another caul-
dron was reported 3—4 km west of the former
one. According to Haukur Tómasson (pers.
comnt.), the western cauldron sank down during
this jökulhlaup. Since then both cauldrons have
been observed but only the eastern one has
subsided during the jökulhlaups.
The formation of the western cauldron lias
not been fixed in time. The cauklron was not
observed immediately after the jökulhlaup in
1966 (Sigurdur Thorarinsson, pers. comrn.). Ex-
peditions travelling in the area in May 1968
and May 1969 did not observe the western
cauldron. This may be ascribed to poor visibility
(Páll Theodórsson, pers. comm.).
The cause of the recent jökulhlaups in Skaftá
has been discussed by several authors (Thorar-
insson and Rist 1955, FJnarsson 1966). The
main problem is whether the jökulhlaups are
caused by volcanic or permanent geothermal
JÖKULL 27. ÁR 71