Jökull - 01.12.1980, Side 62

neither necessary nor the most simple explan-
ation. Whether or not the model is the true
explanation, it is nevertheless interesting to
discuss the various objections that could be
offered against it. Some of these will be dis-
cussed below.
(1) It has often been stated that a condition
for the formation of an horizontal intrusion is
that the minimum principal stress must be
vertical. According to focal mechanism
solutions, the minimum principal stress is ex-
pected to be horizontal and not vertical in the
crust of the Reykjanes Peninsula (Klein et al.
1977). The answer to this objection is simple:
the statement is not true. In the first place, it
does not take into account the importance of
horizontal weakness in the strata. Such
weakness, with essentially no tensile stress
across it, is very important in deciding the
form of an intrusion (see e.g. Pollard 1973,
Mudge 1968). Secondly, the magma density
may be higher than that of the rock, and the
statement ignores the possible effects of vapor
pressure — both of these have been discussed
above. Thirdly, during downwarp and burial
of undeformed sediments the vertical stress is
higher than the horizontal stresses at all
depths (Price 1974). But sills are common in
sedimentary basins, and it is therefore likely
that some of them were intruded while the
maximum principal stress was vertical.
(2) In the model, the magma applies a
constant pressure perpendicular to the con-
tact, and the shear stress is assumed to be zero.
It could be pointed out, that we are dealing
with a flowing magrna, and the above
assumptions are therefore not true. Admit-
tedly, in this case shear stress will exist and the
normal pressure will decrease in direction of
flow; due to viscous drag along the contact.
However, Pollard (1973) has examined these
factors and concludes that they have negli-
gible effects on both form and stresses around
the intrusion.
(3) The throw is up to 20 m and the
maximum subsidence is 50 m in the Vogar
fissure swarm. This seems to be difficult to
explain by a 9.5 m thick intrusion. But the
model is only believed to have originated the
fractures, and not being responsible for the
subsequent subsidence. The subsidence,
which is a common feature in the neovolcanic
zone in Iceland, is probably due to various
factors. Some of these factors might be: (i)
Accumulation of high density lavas above the
low density magma source. The crust, and the
upper mantle above the source, would have a
tendency to sink into it; owing to their higher
density and also because of occasional press-
ure relief in the magma source — during and
after big eruptions and/or intrusions. (ii)
Continuous melting of the rock next to
(above) the magma source. In both the above
cases the subsidence would be easiest where
big fractures already existed; namely inside
the fissure swarms. (iii) It should be noted that
the measured subsidence is only relative, i.e.
only referred to the flanks of the fissure
swarms. But recent data indicate, that at least
part of the measured subsidence in such
swarms in Iceland is due to uplift of the flanks
and not an absolute sinkage of the area bet-
ween the flanks (Björnsson et al. 1978).
(4) As said before, the faults are usually
vertical or slightly inclined in the reverse
trend. This is difficult to explain by a single
sill. However, the process outlined by Bradley
(1965), in which normal faults, with the above
inclination, are formed ahead of the intruding
sill, could explain at least part of the observed
fracture pattern. But some of the fractures,
near the west end of the Vogar fissure swarm,
do cut rocks of different age, which makes a
single intrusion unlikely (although not im-
possible, as the sill could be younger than the
youngest lavas at the surface).
Without wishing to go into detail, I suggest
the real situation could be similar to that
outlined in Fig. 16. In this model each sill is
supposed to give rise to a horst; and the reverse
and vertical normal faults are initiated near
the ends of the sill and propagate upwards to
the surface. As before, the sills are not sup-
posed to be responsible for the subsequent
subsidence on the fissure swarm, but only to
originate the fractures. Although I do not at
60 JÖKULL 30. ÁR