Jökull - 01.12.1953, Qupperneq 18

Fig. 7. The south-western
most part of the Gríms-
vötn depression on 28
Aug. 1950.
Suðvestcisti hluti Grims-
vatnalœgðarinnar séður úr
lofti 28. ágúst 1950.
Aerial photograph
by S. Þórarinsson.
July 1953 or in 27 nronths the water level in
Grímsvötn has risen ab. 48 m (possible error
hardly more than +_ 5 m), and so has also the ice
surface in the SW part of the depression.
According to Sigurðsson’s map of 1942, the
area of the Grímsvötn depression at 1410 m
level would be nearly 40 km2 and ab. 26 km2
at the 1360 m level. Thus the established
raising of the water level in 27 months means
an increase in volume of water and ice in the
depression amounting to ab. 1.6 km3. A great
part of this increase seems to be due to the
increase in volume of water in the depression.
The volume increase of 1.6 km3 in nearly 27
months corresponds to an annual increase of
ab .0.7 km3, a figure near to the one we found
to be the average annual precipitation within
the Grímsvötn intake area.
In a previous paper (Thorarinsson, 1953) I
have expressed the opinion that the mechanism
of the Grímsvötn glacier bursts is most easily
understood by comparing it with the cata-
strophic drainage of ice-dammed lakes such
as Grænalón, where melt-water accumulates
until the water behind the ice barrier has
risen so high (roughly nine tenths of the height
of the barrier) that it can raise the barrier
and force its way underneath so that the lake
drains catastrophically. During the last two
centuries the average interval between the glaci-
er bursts from Grímsvötn has been ab. 10 years,
and before 1938 the intervals were strikingly
regular (glacier bursts in 1934, 1922, 1913, 1903,
1897, 1892, 1883, 1873, 1861, and so on). This
regularity is, in my opinion, most plausibly ex-
plained by assuming that 10 years ablation and
subglacial melting within the Grímsvötn intake
area is capable of filling the depression with
enough water to raise the damming ice barrier.
Furthermore, we may assume that during the
last centuries the accumulation in the Gríms-
vötn intake area has roughly balanced the drain-
age from the Grímsvötn depression as other-
wise this depression would either have been fill-
ed or emptied. The conclusion is that the total
discharge of a Skeiðarárhlaup (i. e. glacier burst
from Grímsvötn) during the last centuries has
on average been 7.0 to 7.5 km3 Graphs of the
drainage discharge pattern of the normal Skeið-
arárhlaup (Thorarinsson 1953, Figs. 9 and 10)
show, that a total discharge of 7.0 to 7.5 km3 cor-
responds to a maximum discharge of ab 50.000
m3/sec, which seems to be a plausible figure.
The question then arises: When may we ex-
pect the next Skeiðarárhlaup? This question is
difficult to answer. The Grímsvötn problem is
far from definitely solved yet. We do not know
with certainty the depth and volume of the de-
pression nor do we know to what extent it is
occupied by ice. We do not know exactly the
16