Jökull - 01.12.1982, Síða 122

Fig. 6. Typical arrangement of faults and mounds
in the South Iceland Seismic Zone, interpreted as
resulting from strike-slip movement underneath.
The axis of inferred maximum (a,) compressive
stress and least (a3) compressive stress are shown.
Mynd 6. Dœmigerb afstaða gjáa og sprunguhóla á
skjálftasvæði Suðurlands, sem má túlka sem afleiðingu af
sniðgengishreyfingu undir niðri. Sýndir eru ásar mestu
(°i) °g minnstu (o3) spennu í 'berggrunninum.
This is, in our view, not justified. The structure and
appearance of the South Iœland fracture systems is
quite different from that ofthe fissure swarms of the
volcanic rift zones. En echelon arrangements can be
found within the fissure swarms, but the sense is not
always the same and often depends on the strike of
the fissure with respect to that of the swarm. Each
swarm has the structure of a shallow graben, and
vertical displacements across the fissures are comm-
on. The swarms are active mostly during rifting
episodes similar to the current events in the Krafla
fissure swarm in NE-Iceland. There the displace-
ments have been shown to be in the form of widen-
ing ofthe fissures and subsidence of thegraben floor
(Gerke et al. 1978, Bjomsson et al. 1979, Möller and
Ritter 1980, Sigurdsson 1980).
The age of many of the faults is uncertain. The
1630, 1896, and 1912 faults are the only ones with a
known age. Yet it would be of great importance to
date the faults. One may hope to obtain a seismic
history for a good part of the postglacial time, and
thus improve previous estimates of the recurrence
time of large earthquakes in this zone. So far, no
evidence has been found for repeated activity along
any of the faults. Each fault seems to have been
active during one earthquake only. One should bear
in mind, however, that all the known faults cut very
young rocks and deposits that may conceal older
movements. But it is noteworthy, for example, that
the 1896 fault was formed 2.5 km W of and parallel
to a fault active 266 years earlier (Fig. 5). In this
case it appears to have been easier to form a new
fault than to reactivate an old one.
The main conclusion of this study seems to con-
firm the picture suggested by the destruction zones
of the earthquakes, namely that individual earth-
quakes are related to slip along faults perpendicular
to the main seismic zone. The earthquakes seem to
be associated with brittle deformation ofa 10-20 km
wide zone overlying an E-W trending zone of aseis-
mic deformation in the lower crust or upper mantle.
The net displacement along the zone is left-lateral
transform motion to accommodate the spreading in
the Eastern Volcanic Zone, but the brittle crust
responds by right-lateral slip along the conjugate
fault planes. The reason for this behaviour is not
clear, but it might be related to the apparent youth
of this zone. It is possible that the finite displace-
ment along the zone is not sufficiently large to break
the whole crust and produce a major fault. Perhaps
one has to view the South Iceland Seismic Zone as a
propagating fracture front instead of a steady state
feature. Such a front may be expected to propagate
southwards in response to increased activity and
southwards propagation of the Eastem Volcanic
Zone, and would leave behind it a block of thor-
oughly fractured crust. This model might explain
the extensive faulting of the Early to Late Quater-
nary Hreppar rock sequence exposed north of the
Seismic zone. Continued studies of the South Ice-
land earthquake fracturesand their relation to the
Hrepparsequence may further clarify these ideas.
REFERENCES.
Bjömsson, A., G.Johnsen, S. Sigurdsson, G. Thorbergsson
and E. Tyggvason 1979: Rifting of the plate
118 JÖKULL 32. ÁR