Jökull - 01.12.1974, Blaðsíða 29

Grímsvatnahlaup 1972,
Mechanism and Sediment Discharge
HAUKUR TÓMASSON,
NATIONAL ENERGY AUTHORITY, REYKJAVIK, ICELAND
ABSTRACT
The discharge of the Grimsvatnahlaup in
1972 is estimated as 2 kms in volume. About
29.5 • 10G tons of sediment load were carried
from the glacier. This sediment load is 50 per
cent coarse silt. Two thirds of the sediment
load are deposited on Skeidarársandur but one
thircl reaches the ocean and is deposited near
the shore. The mechanism of Grímsvatnahlaup
is assumed to be a complex one tuith mainly
Glens theory valid for the beginning of the
jökulhlaup and a mechanism first described by
Liestöl responsible for the continuation.
CALCULATIONS OF VOLUME
The last Grímsvatnahlaup of March 1972 was
observed and studied with greater effort than
any previous jökulhlaup from Grímsvötn. The
author was in charge of the sediment discharge
measurements and this paper presents the re-
sults of his findings, which are not limited to
the sediment discharge, but are also centered
on the mechanism and volume of the water re-
leased in the jökulhlaup.
The volume of flood water from Grímsvötn
is usually referred to as having been 3.2—3.5
km3. These figures are based on estimates of
discharge made on Skeidarársandur during the
jökulhlaup. This value also fits well to the
estimated water balance of the Grimsvötn basin.
An alternative approach to measuring the
jökulhlaup would be to measure the volume
changes in Grímsvötn with reference to Gríms-
vatnahlaup. Actually this has been done al-
teady. After the flood of January 1960 the
Grimsvötn basin was mapped (in June 1960)
and a storage diagram constructed according to
which the maximum storage in Grímsvötn is
1.5 km3 (Rist, 1961).
Although jökulhlaups from Grímsvötn last
for a few weeks this kind of flood has a very
sharp peak and approximately half of the flood
water is discharged in 3 days (Rist, 1973). Tlie
quantity of fresh water in the sea was therefore
measured by the Marine Research Institute on
the seconcl and thircl days after the peak flow.
The result is shown in Fig. 1. A well establish-
ed fresh water layer is found, the volume of
which is calculated as 0.7 km3 of fresh water.
The fresh water found in the sea corresponds
reasonably well to the volume of Grímsvötn if
the measured volume represents the bulk of 3
days peak flow, which, from knowledge of speed
of sea currents in this area, is reasonable.
Now we have to consider another aspect of
this problem, i. e. the water budget of the
Grímsvötn basin. Thorarinsson (1953) has esti-
mated the inflow into Grímsvötn in the form
of ice ancl water to be 0.7 km3 per annum
which makes 3.5 km3 in the 5 year interval be-
tween jökulhlaups. This is equal to Rist’s esti-
mate of the discharge volume. In his model
Thorarinsson does assume that there is no
leakage from Grímsvötn through ice or under-
lying bedrock.
The model of a watertight Grímsvötn basin
is not necessarily true (cf. also Tliorarinsson,
1974). Similar lakes outside glaciers often have
substantial leakage. Substantial leakage can
therefore take place through rock into some of
the branches of Skeidará. Leakage through the
glacier is also a possibility.
Water can also occasionally escape from
Grímsvötn through sub-glacial tunnels. The
Grímsvötn water is well characterized by its
high content of dissolved solids, which design-
JÖKULL 24. ÁR 27