Jökull - 01.12.1984, Blaðsíða 92

Fig. 13. Schematic illustration of the upper end of
one dyke in the research area. The dyke is 6 m
thick, made of tholeiite and contains many vesi-
cles, commonly a few millimeters in diameter and
empty. At its upper end the dyke gradates into
breccia and scoria, with at least three pillows
(from 0.2m to 0.6m in diameter). Ahead of the
dyke’s upper end occur four normal faults,
A,B,C and D. Fault A: strike 24°, dip 70°E,
throw 3m. Fault B: strike 76°, dip 65°W, throw
2m. Fault C: strike 60°, dip 70°W, throw lm.
Fault D: strike 70°, dip 68°W, throw 4m. All the
faults are closed, and only fault D has fault
breccia, up to 1—2m thick. The average inclina-
tion of the section is several tens of degrees to the
south.
Mynd 13. Einfölduð mynd af gangi sem endar
upp á við í fjórum siggengjum. Sniðið hallar
nokkra tugi gráða til suðurs.
up in the lava pile. However, because so few
faults were followed upwards, and also because
the lava flows used to measure the throws are
variable in thickness, the possibility still remains
that some of the faults in the research areas were
active over long periods of time, say for 105 years.
The displacement in most single earthquakes is
only of the order of a few centimeters (Jeffreys
1976, p. 477; Bullen 1963, p. 283) so a large
succession of movements may be needed to pro-
duce the observed throws.
Slickensides and fault breccia are common.
Measurements on the slickensides show that the
displacement was vertical, that is, no strike-slip
component is apparent. The fault breccia is com-
posed of fine-grained grey to brown matrix and
angular fragments of the country rock of various
sizes. The thickness of the breccia varies from
zero to, at most, a couple of meters. Most of the
faults are closed and they are inconspicuous on
the air photographs.
Seen from a distance, the fault walls appear to
be nearly straight, but on closer examination it is
seen that they are sinuous. Only one fault was
observed to dissect dykes, so the relative age of
the faults and dykes is unknown.
Apart from the faults observed in the profiles
and described above, a few faults are apparent on
the air photographs in the highlands to the west
of Vatnsfjördur (Fig. 10). These are also normal
faults, and the conspicuous fault scarps indicate
that these faults have been active in Postglacial
time (otherwise the scarps would have been
eroded by the Pleistocene glaciers). The strike of
the faults is NNW. The throw is variable, the
maximum measured throw being at least 12 m.
The fault scarps are mostly subvertical, but where
the maximum throw was measured the dip was
70°. Together the main faults form a narrow
graben (Fig. 10).
Joints
Joints are numerous in the lava flows. The
most conspicuous are the almost vertical colum-
nar joints. These joints are commonly 0.1-1 cm
wide and empty. They occur at the interval of 1 m
or so. In some places, mainly near sea level in the
southern part of the area, there are occasional
joints filled with minerals, that is, veins. Some of
the vertical joints are irregular in shape, but in a
horizontal plane most of them are nearly straight.
Apart from the vertical joints, there are hori-
zontal joints in the lava flows. Presumably, these
joints result either from shear stresses in the lava
flows while mobile, or from remnant stresses
caused by burial and subsequent erosion and
uplift of the lava pile.
Few non-columnar joints occur in the research
areas. Only in one place, in a lava flow at the east
coast of Kjálkafjördur, did I observe well-defined
tectonic (i.e. non-cooling) joint system. All the
joints in this system are subvertical and most of
them strike perpendicular to the dykes in the
area. All these joints are empty. A search was
90 JÖKULL 34. ÁR