Jökull - 01.12.1990, Qupperneq 159

wards Jökulsá á Fjöllum. A narrow valley runs to
the E from this pass. This valley may have been cre-
ated erosionally by floods rushing down the slopes of
Bárðarbunga.
HAMARINN, THE LOKIRIDGE AND THE
fögrufjöll fissure swarm
Hamarinn is a steep-sided mountain, about 60 km2
in area. Its nearly circular rim ranges in elevation from
about 1200 m to 1570 m. A slight depression is found
inside the rim. A mountain ridge strikes SW from the
SE part of Hamarinn beneath Tungnaárjökull toward
Fögrufjöll. This will be referred to henceforth in the
text as the Fögrufjöll Ridge. Presumably it consists
of hyaloclastite built up on volcanic fissures. At the
glacier edge, the ridge joins the fissure swarm that
continues towards the Torfajökull volcanic complex.
This swarm has not been active in postglacial times
and no seismic activity has been found there in recent
years.
A 1100-1250 m high ridge extends eastward from
the NE comer of Hamarinn and joins the Grímsvötn
mountain massif. This ridge has at least three active
geothermal areas as well as being seismically quite
active. Here it will be called the Loki Ridge, after the
legend of Loki in Nordic mythology. 1 Epicentres
define an E-W trending belt of seismicity that follows
the Loki Ridge from Hamarinn towards Grímsvötn.
In light of the geothermal evidence we are inclined
to interpret this belt as a row of clusters rather than
an expression of an active fault. The seismic, topo-
graphic and geothermal evidence taken together then
suggest that Hamarinn and the Loki Ridge represent
a row of central volcanoes. Volcanic eruptions on the
Loki Ridge may cause jökulhlaups in Skaftá, or in
Kaldakvísl (Bjömsson, 1988).
The arrangement of these structures into volcanic
systems is not straight forward. It seems clear that
there is a structural connection between the central
1 The giant Loki was tied in a netherworld cave with venomous
water dripping from above. Nordic mythology explained earth-
quakes as the sudden writhings made by the luckless Loki when
the venom hit his face. On the Loki Ridge, water-filled vaults are
located beneath the ice cauldrons and water propagates into hot
boundaries of magma.
volcanoes on the Hamarinn-Loki Ridge and the Fögru-
fjöll fissure swarm outside the glacier edge. This
may be defined as a volcanic system, here called the
Loki-Fögrufjöll volcanic system. However, there also
seems to be links between this system and adjacent
systems, both to the west and east. A ridge connects
Bárðarbunga and Hamarinn, possibly indicating that
both volcanoes are within the same system. The high
seismic activity that began in Bárðarbunga in 1974
also affected the Hamarinn and Loki volcanoes, that
were unusually active during the same time. In addi-
tion, Hamarinn is located near the eastern border of
the Veiðivötn fissure swarm. Finally, the geothermal
activity on the Loki Ridge increased in the 1950’s,
apparently coincident with a decline in the activity
of Grímsvötn. It is therefore not inconceivable that
these systems influence each other, mechanically if
not chemically.
GRÍMSVÖTN, HÁABUNGA, ÞÓRÐARHYRNA,
PÁLSFJALL, AND THE LAKIFISSURE SWARM
The Grímsvötn area contains a huge mountain
massif, exceeding 1100 m in elevation for 18-20 km,
as measured from W to E, and about 10-15 km from
N to S. Mountains which rise above 1400-1700 m
encircle a depression with a low of about 1050 m
(Bjömsson, 1988; Guðmundsson, 1989). The depres-
sion is interpreted as a composite caldera. The area
of the caldera girded by the highest points on the rim
is 62 km2. The irregularly shaped caldera is divided
into two nearly equal parts by a median ridge that rises
to 1200 m elevation. The deepest part of the caldera,
with elevations down to 1050 m, is located W of the
ridge. The sites of the 1934 and 1983 eruptions are on
the inner side of the southem rim.
In the northernmost part of the central ridge in the
Grímsvötn depression, a large area of the ice surface
subsided in 1938, presumably due to an eruption that
did not penetrate the ice surface (Þórarinsson, 1974,
and Björnsson, 1983, 1988). The eastem rim of the
main depression comprises another ridge, almost par-
allel to the first, trending NNW towards Bárðarbunga.
This ridge fades out and does not reach all the way
to Bárðarbunga. The northernmost eruption site in
1867 may have been situated on this ridge (Bjöms-
JÖKULL, No. 40, 1990 155