Jökull - 01.12.1989, Page 81

extrusive origin of the silicic
Rocks of the kækjuskörð
Rhyolitic volcano
The extension of the silicic lavas of the Kækju-
skörð volcano is shown in Fig. 4. Contrary to the
view of Dearnley (1954) who regarded most of the
silicic rocks of the Loðmundarfjörður area as
intrusive, there are several reasons to believe that
the silicic rocks have formed as subaerial lava flows:
a. The silicic flows show a zonal structure with an
upper, central and basal zone.
b- Autobrecciation on top and on the flow fronts is
common.
c- Updoming of the surrounding basaltic lava flows
is nowhere visible at the contact to the rhyolite.
Fig. 5a shows a schematic section through a rhy-
olitic lava flow, displaying its lithologies. Fig. 5b
shows a cross section through the Rocche Rosse
obsidian coulée on Lipari Island, southem Italy (Cas
andWright, 1987).
In the basal zone of a rhyolite flow the rock
becomes áutobrecciated under the slowly moving
and quickly cooling lava. The central zone is
characterized by mostly crystallized rhyolite, so-
called "stony rhyolite" (Cas and Wright, 1987). The
uPper zone of the rhyolite flow can be the one with
the most complicated structure. It is often divided
tnto different layers and is intensely folded and brec-
ciated. Due to emission of gas from the cooling lava
below it often has a pumiceous structure. Autobrec-
ciation is typical of the surface, base and front of the
highly viscous lava flows. Surface ridges on many
rhyolitic lava flows, so- called ogives, are most
likely surface manifestations of an intemal ramp
structure as shown in Fig. 5b. The layering has been
folded in the direction of the flow in concave forms
°f amplitudes of 10 to 50 metres (Loney, 1968;
MacDonald, 1972; Fink, 1980; Cas and Wright,
1987).
Many of the features named above occur in the
rhyolitic lavas of the Kækjuskörð volcano. Auto-
brecciation is common (Fig. 6), especially on top of
the flow and on its front. These upper parts of the
lavas are more prone to weathering and erosion than
the more solid lower parts and are thus not well
preserved. Outcrops of the basal layer are rarely
found, because it is usually covered by scree depo-
sits. The brecciated areas consist partly of fragments
of the stony rhyolite from undemeath and partly of
perlitic pieces, pumice and obsidian. The size of the
components in the autobrecciated rhyolites is com-
monly in the centimetre range. The components are
dispersed loosely in a finer matrix, but dense pack-
ing without any matrix also occurs. The matrix con-
sists of pumiceous puffed up glass, quartz and secon-
dary chalcedony. The central part of the lava flows,
being the most solid part of a rhyolite flow, often
forms prominent cliffs. At Orustukambur, on the
southem margin of the Kækjuskörð volcano, good
sections through rhyolite lavas are exposed. The sili-
cic rocks here form prominent cliffs and ramp struc-
tures are clearly visible. The layering of the flow
foliation is frequently folded. The size of the folds
ranges from several millimetres (Fig. 7) up to tens of
metres (Fig. 8) in amplitude.
STRUCTURE AND DEVELOPMENT OF
THE KÆKJUSKÖRÐ RHYOLITIC
VOLCANO
The structure of the silicic rocks, as shown above,
strongly suggests the existence of a small rhyolitic
volcano with its centre in the Kækjuskörð area. This
volcano could have had its own magma chamber,
but the possibility of its formation as a parasitic
cone of either the Dyrfjöll central volcano or the
Breiðavík central volcano should also be con-
sidered. The extrusive silicic rocks of the volcano
can be subdivided into at least four different lava
flows and two ignimbrites. The total thickness of the
exposed lava pile of the volcano, including a few
basaltic lava flows, measures at least 600 m and the
volcano covers an area of about 13 km2. Quartz
tholeiites and the Orustukambur Ignimbrite I form
the base of the Kækjuskörð volcano. In the south
this base lies about 350 m above sea-level, in the
north about 500 m. Regional variations in the under-
lying landscape are probably responsible for this
height difference, because there are no signs of a tilt-
ing of the complex as a whole to the southwest.
JÖKULL, No. 39, 1989 79