Jökull - 01.01.2001, Blaðsíða 62

Ármann Höskuldsson
The opening phase of the eruption is marked by la-
haric and ash–flow deposits. Summit eruptions at ice–
capped volcanoes have been known to generate small
mud–flows or lahars (Major and Newhall, 1989). La-
hars of this type occurred during the beginning of the
1947 eruption of Hekla, where water–rich low temper-
ature currents confined to gorges and valleys (Kjar-
tansson, 1951). Ash–flow and lahar deposits from
Cerro las Cumbres are by far the finest grained de-
posits in the succession, indicating that they were
formed by intense magma fragmentation. As demon-
strated by the SEM study on particles from the ash–
flow, the quenching structure indicates fragmentation
by magma–water interaction (oval shaped vesicles
and dust welding) as well as volatile saturation (tab-
ular grains and vesicles). The co–existence of tabu-
lar and oval vesicles in the clast population suggests
that fragmentation of the magma began at depth in the
conduit (magmatic), generating rod and chip–like par-
ticles with tabular vesicles. This generated an open
vent situation where the freely expanding magma was
fragmented as it came into contact with abundant sur-
face water (Figure 11) to form Y–shaped platey and
blocky particles with oval vesicles (Heiken and Who-
letz, 1985). However, occurrence of pumice lapilli
in the deposits shows that the magma was not totally
fragmented, suggesting that the magma/water ratio
was less than one (Self and Sparks, 1978; Fisher and
Schmincke, 1984).
The initial phreatomagmatic phase was generated
by an explosive interaction between the glacial melt-
water and the magma. The early lahars that descended
down the flanks of the volcano were formed by an
eruptive mixture of meltwater, ash and coarse lithic
fragments. The accompanying eruption column con-
sisted of ash and superheated steam. The steam cooled
and condensed during ascent reducing the buoyancy
of the column and promoting its collapse to gener-
ate the ash flow that followed the floods and lahars.
The ash flow blanketed the whole eastern side of the
volcano and destroyed vegetation in its path (Figure
11–2). As the eruption reached its climax the eruption
rate (hence the magmatic pressure) increased enough
to prevent external water from coming into contact
with the magma, and the eruption style changed from
phreatomagmatic to plinian. This is indicated by the
lower plinian pumice–fall (P1 on Figures 3, and 11–
3). An estimate for the volume of magma erupted
by the first plinian phase is about 4 km


(2.1 km


DRE, Table I). Evacuation of this magma volume
from the holding chamber is interpreted to have ini-
tiated caldera collapse of the central central block.
The sharp transition from plinian fall (P1) to vul-
canian fall and surge (SB1) deposits in outcrops clos-
est to the caldera (e.g. Figure 3) suggests that exter-
nal water regained access to the vents to initiate the
second phreatomagmatic phase of the eruption. It
is likely that the subsidence of the central block im-
posed restrictions on the flow of magma within the
conduit feeding the eruption, consequently reducing
the magma effusion rates from the magmastatic pres-
sure in the conduits and thus allowing external water
(melt water and/or groundwater) to flow freely into
the vent system and re–initiate phreatomagmatic ac-
tivity (Figure 3 Sb1 and Figure 11–4).
The second plinian layer is interpreted to have
formed as vent clearing allowed a second increase in
effusion rate and external water was again isolated
from the vent (Figure 3, P2 and Figure 11–3). As
the eruption declined, the effusion rate decreased, wa-
ter regained access to the magma and vulcanian de-
posits were produced (Figure 3, Tsb2 and Figure 11–
4). Such a transition, from plinian to vulcanian, has
been reported for other eruptions. Bond and Sparks
(1976) showed that such a transition took place dur-
ing the Minoan eruption of Santorini, as a result of
seawater gaining access to the vents. At Cerro las
Cumbres, glacial meltwater and groundwater played
the same role. At present there is a region in the upper
reaches of the volcano and circumscribing the caldera
that is totally devoid of deposits from this eruption .
This area represents the extent of the glacier on the
volcano at the time of the eruption (Figure 2a). All
material that fell upon the glacier was washed away as
the glacier melted, leaving a deposit–free zone around
the caldera. This material is found as resedimented
epiclastic tephra units on alluvial plains to the east of
the volcano. The activity of Cerro las Cumbres ended
with extrusion of a dacitic dome complex (Table I and
Figure 10) on the caldera floor (Figure 11–5).
62 JÖKULL No. 50