Jökull - 01.12.1957, Blaðsíða 24

Fig. 1. The bigger one of the
two ice cauldrons formed in
Kötlukvos in connection with
the jökulhlaup in June 1955.
Diam (acc. to Rist’s measure-
ments) ab. 1000 m. Depth 80 m.
Stœrra sigið, er myndaðist i
Kötlukvosinni i sambandi við
hlaupið 25. júní 1955. Þvermál
um 1000 m. Dýpi 80 m.
Photo S. Thorarinsson
Aug. 27th 1955.
done since the late 19th century (Thorarinsson
1939. The behaviour of these lakes is in my
opinion best explained by assuming that a
lifting of the damming barrier has started the
tapping of these Iakes.
As to the Grímsvötn area its topography makes
it easily understandable that many km3 of water
can be stored there. But the topography of the
Katla area, within the intake area of Höfda-
brekkujökull, is quite another.
Even if we take into consideration the pos-
sibility that some amount — possibly a consider-
able amount —of the water discharged by aKötlu-
hlaup is magmatic water and meteoric ground-
water released at the beginning of the eruption,
we can hardly explain the tremendous hlaups
without a storage of water under the ice in the
Katla area. A proof of such storage was the small
jökulhlaup that occurred on June 25th 1955
and the subsequent formation of ice cauldrons
(Fig. 1) in the Katla area (Rist and Thorarins-
son 1955). This jökulhlaup was the first Kötlu-
hlaup the discharge of which was measured (by
S. Rist) so that a rather reliable discharge graph
could be drawn. This discharge graph has not
yet been published, but I have reconstructed it
on a logarithmic paper together with the dis-
charge graphs of some other jökulhlaups (Fig.
22
2). Two of these jökulhlaups are limno-glacial
(i.e. due to tapping of ice dammed lakes): the
Grænalónshlaup 1939 and the miniature hlaup
from the ice dammed marginal lake Gjánúps-
vatn at Hoffellsjökull on June 15th to 17th
1951. This miniature hlaup is the first one in
Iceland that has been registered on a self-
recording gauge (Hjulström 1953, Arnborg
1955) and it is of interest to note the close
resemblance between the discharge pattern of
this hlaup and that of the nearly 1000 times
bigger Grænalónshlaup in July 1939, the dis-
charge graph of which (Thorarinsson 1939) was
based mainly on a description of the hlaup by
an attentive observer, the farmer Hannes Jóns-
son of Núpsstadur. The discharge graph of the
1954-Skeidarárhlaup is based on Rists measure-
ments (Rist 1955) that may be regarded as satis-
factorily accurate. The discharge graph of the
1934-hlaup from Grímsvötn is reconstructed
from a graph in my above mentioned Gríms-
vötn paper. The similarity in the types of the
discharge graphs of the volcano-glacial Gríms-
vötn-(Skeidarár)hlaups and the limno-glacial
lilaups has already been stressed. At last I have
on my diagram included a graph of the jökul-
hlaup that occurred in the river Skaftá in early
Sept. 1955. The origin of this hlaup was in