Jökull - 01.12.1980, Side 61

about 1200°C, is about 2.6 g/cm3 (Williams
and McBirney 1979). All natural magma con-
tains water, but the water content of basalt
magma appears to be not over 1% by weight
(Carmichael et al. 1974). If so, the effect of the
water in lowering the density of the magma
would be negligible, and is therefore omitted
here. Putting the above density values into
formula (6) we get Po=900 bars, at depth of 2
km. Putting this value into formula (5) we
obtain the maximum uplift, wmax=9 m. This is
not much although, as we shall see later, it is
enough to explain the measured dilation at
the surface of the Vogar area. Sills in Iceland
are commonly a few meters thick, so the above
figure seems to be in good agreement to what
could be expected. However, much thicker
sills do occur (up to 200 m in Iceland, Frid-
leifsson 1977), and these must be explained by
a different mechanism of emplacement.
This different mechanism can be of two
types: Either the intruded rock behaves as
plastic during the intrusion, or the magma is
denser than the rock it intrudes so the roof
may float on the sill. The former mechanism
was suggested by Price (Fyfe et al. 1978); the
latter by Bradley (1965). Combination of both
these mechanisms is of course also possible.
Both these mechanisms are highly probable
during intrusion into sedimentary rocks of
high porosity and low density. Hyaloclastite is
essentially sedimentary rock, and these could
therefore easily operate during intrusion into
hyaloclastite. But inside the basalt lava pile,
where sills are in fact common, such mechan-
isms are improbable, and the elasticity theory
must be used to account for such intrusions
there. In the upper part of the crust beneath
the Reykjanes Peninsula these could, how-
ever, operate.
Another important factor that should be
considered is the vapor pressure ahead of the
intrusion. As said before, evidence indicates
that the water content of the magma (basalt)
is not over 1% by weight. A separate vapor
phase would therefore not form below about 1
km depth. However, ground water exists
down to 2—3 km depth, and its contact with
the hot magma leads to formation of a sepa-
rate vapor phase ahead of the intruding dyke.
The vapor pressure can easily exceed the
lithostatic pressure at depth of, say, 2 km and
cause horizontal fractures to open. Taking
also into account, that the uppermost 2 to 3
km of the crust beneath the Reykjanes Penin-
sula have the average density of only about 2.6
g/cm3, the tendency to sill formation must be
strong.
We now consider the dilation at the surface
as a result of the assumed horizontal intrusion.
As shown in Fig. 14, the uplifted area is larger,
and has different form from that of the sill.
Only in the middle part of the uplifted area do
we get dilation. If we take the cross section of
this middle part as part of a circle, i.e. we
assume pure bending in that part, the dilation
is given by:
D = L t/R (7)
where L is the original length of the dilated
area, t is half thickness of the bent layers, and
R is the radius of curvature of the arc. For the
maximum dilation we have D = 15 m, L = 5000
m, and assuming the intrusion to take place at
depth of 2 km we have t=1000 m. Putting
these values into formula (7) we get
R=3.3x 105 m.
To find the maximum uplift necessary to
cause the maximum dilation, we use:
Wmax=a2/2R (8)
where wmax is the maximum uplift, and a = ‘/2L
= 2500 m. Putting the values for a and R into
formula (8) we get wmax=9.5 m. This is not
much, and sills of this thickness are common
in Iceland. Although this thickness is a little
over the value we got from formula (5), it is
not a serious problem and could be accounted
for by e.g. greater depth of the magma source,
or greater depth of the intrusion. However, we
will now discuss some difficulties this model
has to face.
Criticism of the sill model. The above model is
in fact only intended to show how little is
needed to account for fissure swarms of this
size; that regional plate movements are
JÖKULL 30. ÁR 59