Náttúrufræðingurinn - 1989, Qupperneq 59

Walker, G.P.L. 1974b. Eruptive mechan-
isms in Iceland. I Geodynamics of Ice-
land and the North Atlantic area
(ritstj. Leó Kristjánsson). Reidel, Dor-
dreclit. Bls. 189-201.
Wickman, F.E. 1976. Markov models of
repose-period patterns of volcanoes. I
Random processes in geology (ritstj.
D.F. Merriam). Springer-Verlag, New
York. Bls. 135-161.
SUMMARY
Intrusion frequencies of magma
chambers and extrusion frequen-
cies of volcanic systems
by
Ágúst Guðmundsson
Nordic Volcanological lnstitute
University of Iceland
IS-101 REYKJAVIK
ICELAND
This paper reviews and extends the au-
thor’s earlier model on the intrusion and
extrusion frequencies associated with vol-
canoes at divergent plate boundaries
(Ágúst Guðmundsson 1986a, 1988). Here
the model is applied to two specific re-
gions in Iceland. The long-term dike in-
trusion frequency (i.e., of the order of 104
years or more), as predicted bv the model,
is applied to the Álftafjörður dike swarm,
of Tertiary age, in Eastern Iceland (Ágúst
Guðmundsson 1983). The short-term ex-
trusion frequency (i.e., of the order of 103
years or less), as predicted by the model,
is applied to Holocene volcanism on the
Reykjanes Peninsula in Southwest Iceland
(Ágúst Guðmundsson 1986d). The main
conclusions of the paper may be summa-
rized as follows.
1) At divergent plate boundaries, the
dike intrusion frequency of a deep-seated
magma reservoir, i.e., the frequency of
magma flow from the reservoir, is largely
determined by the associated spreading
rate as well as the aspect (height/width)
ratio of the reservoir. High spreading
rates and/or aspect ratios increase the as-
sociated dike intrusion frequency.
2) The dike intrusion frequency of a
shallow magma chamber, i.e., the fre-
quency of flow of magma out of the cham-
ber, is normally controlled by that of the
magma reservoir that feeds the chamber.
Because the magma content of a typical
reservoir is from five to one hundred
times the magma content of a typical
chamber (Fig. 2), a single magma flow
from a reservoir may lead to the injection
of up to several dozen dikes and inclined
(cone) sheets from the associated cham-
ber. This may explain the enormous num-
ber of sheets that occur near to some large
plutons (extinct magma chambers) in Ice-
Iand (Figs. 2 and 4).
3) According to the model, the short-
term extrusion frequency of a volcanic
system may be similar to the dike intru-
sion frequency of the source reservoir, in
which case most or all of the dikes would
be feeders. However, the long-term extru-
sion frequency of a volcanic system is nor-
mally only about 10% of the associated
dike intrusion frequency.
4) The Grímsvötn volcano beneath the
Vatnajökull ice sheet has erupted 7-11
times every century during the past 400
years. The Grímsvötn volcano is fed by a
double chamber, i.e., a chamber associ-
ated with a reservoir (Fig. 2). The model
explains such a high extrusion frequency
by two factors. First, it is proposed that
the aspect ratio of the source reservoir is
high (Fig. 3) compared with other reser-
voirs at the divergent plate boundary in
Iceland, which leads to a high dike intru-
sion frequency. Second, because the
chamber is much smaller than the source
reservoir, each magma flow from the res-
ervoir may trigger injection of up to sever-
al dozen dikes (or inclined sheets) from
the associated chamber. In contrast to the
current activity in the Krafla volcano in
north Iceland, where less than 50% of
dikes from the chamber liave reached to
the surface (led to an extrusion), most or
all the dikes or sheets in Grímsvötn have,
53