Jökull - 01.12.1991, Qupperneq 41

TABLE 1. ERUPTIONS IN GRÍMSVÖTN 1934-1991
year ár duration tímalengd type and number of craters fjöldi gíga volume of mound/ridge rúmmál hryggs krn3 volume of tephra rúmmál gjósku km3 total volume heildarrúmmál km3
1934 >2 weeks 3 0.02 0.01-0.02 0.03-0.04
>2 vikur
1938 1-5 weeks 7-8 km 0.3-0.5 minute 0.3-0.5
1-5 vikur fissure örlítið
sprunga
1983 5-6 days i 0.01 minute 0.01
5-6 dagar örlítið
1984 1 hour? ? small? none small?
1 klst? lítið? ekkert lítið?
Grímsvötn or accumulated above the fissure in sub-
glacial vaults. The water filled vaults may have
drained periodically to Grímsvötn, increasing sud-
denly the lake level, perhaps by an order of a few
metres at a time. This meltwater was an addition to
the volumes melted within Grímsvötn, causing
exceptionally fast rise in lake level. Moreover, during
periods of increased geothermal heat flow, the tem-
perature of the lake water may be somewhat higher
than usual. That may lead to jökulhlaups at a lower
lake level than during periods of lower geothermal
heat flow. Thus, the frequent jökulhlaups from
Grímsvötn in the first ten years after 1938 may have
been a consequence of a single large fissure eruption
in 1938. There is no need to claim that several erup-
tions took place in the period.
DISCUSSION
The eruption of 1938 is one of largest in Iceland this
century, in terms of volume erupted. Only two erup-
tions are known to have been more voluminous, i.e.
Hekla in 1947 and Surtsey in 1963-1967 both of
which produced around 1 km3 (Þórarinsson, 1967).
This large fissure eruption in 1938 was almost entire-
ly subglacial while the two fairly small eruptions of
1934 and 1983 broke through the surface of the ice
soon after their start. This is probably due to the great
ice thickness at the site of the 1938 fissure, 400-700
m, requiring larger buildup of hyaloclastites, while
the water depth at the sites of the 1934 and 1983
eruptions was of the order of 100 m. This shows that
fairly large subglacial fissure eruptions may have
occurred beneath Vatnajökull in past centuries with-
out being observed from the lowlands surrounding
the glacier. Large jökulhlaups, however, would wit-
ness to the event.
Þórarinsson (1967) estimated the total volume
erupted in Grímsvötn in historical times (1100 years)
as 1.5 km3. We have shown here that the magma pro-
duction was about 0.35 km3 (magma density 2650 kg
m3) in the last 57 years, a period with relatively few
eruptions. Large eruptions, similar to the event in
1938, have occurred earlier, i.e. in 1867 and probably
also in 1892 (Björnsson, 1988). The production rate
observed in 1934-1991, of 6-106 m3 a1, may there-
fore be representative for the period after 1600 AD,
yielding a total volume of 2.3 km3. If this high pro-
duction rate has prevailed over the last 1100 years the
total volume erupted would be 6-7 km3. A total of 29
eruptions have been attributed to Grímsvötn in the
period after 1600 but only 7 in the period 900-1600.
JÖKULL, No. 41, 1991 39