Jökull - 01.12.1966, Qupperneq 10

Taking either curve 3 or 4 to correspond in
the main to reality, it emerges that after about
8800 B.P. sea-level at Reykjavík fell below the
present level and remained so for nearly 3 mil-
lennia. This result is retained if S2 is used in-
stead of Si but the drop in sea-level is less in
that case.
On account of the synchronous rise of the
country (which must at least be true for South-
ern Iceland) we can find the rise-curves for
other localities by multiplying the Reykjavík-
curve by a locality factor. For Stokkseyri, where
the maximum raised beach is assumed to be
55 m (by comparison with Hjalli in Ölfus), the
factor is (32 + 55)/(32 + 43) = 1.16, the figure
32 being the ordinate of the Si curve at 11.000
B.P. The curve for Stokkseyri lies insignificant-
ly below the Reykjavík-curve after 9000 B.P.,
and sea-level must have dropped in the fol-
lowing time by nearly the same amount. Using
curve 3 for Reykjavík, I have drawn, Fig. 7,
the corresponding curves for Stokkseyri, Hrepp-
ar (125 m beach), Melgerdismelar, F.yjafjord
(35 m), Dalvík, Eyjafjord (12 m), Siglunes,
mouth of Eyjafjord (0 m), and for the assumed
centre of the depression where the (unobserv-
Fig. 8. Approximate position of strandline near
Grímsey 18000 B.P.
8. myncl. Aœtluð lega strandlinu kringum
Grimsey 18000 árum fyrir nútimann.
able) beach height is taken as 160 m. The
corresponding maximum depression at this
centre was 346 m.
In this way theoretical data may be obtained
for various localities and tested by local find-
ings. We see for instance that at about 18.000
B.P. sea-level should have fallen about 40 m
below the present one at the coast near Siglu-
nes, and a rim of low land would have been
added that may in part have been unglaciated.
Grímsey must have been much larger than
now, cf Fig. 8, ancl at least nearly connected
with the mainland. But local data for testing
are scarse and on the whole it would carry
us too far to enter upon such cliscussions liere.
For the time being we shall restrict us to the
material already given and the rise-curves built
on this as shown above.
We now proceed to obtain the relative vis-
cosity. In the theory of Haskell for the rise
of Scandinavia (after Scheidegger 1958) is con-
sidered an uncompensated depression of deptli
d and radius R (R being the distance from the
centre to a point where the depth has decreas-
ed to a certain small part of the maximum
depth cl). Viscosity of the underlying material
is t). Then the velocity of rise at the centre of
the depression is V0 = q • d • R/r], where q is a
constant wich we can take as unity. This de-
pression ancl its progressive shallowing corre-
spond to the stage after the ice load has dis-
appeared.
Gomparing two such depressions of radii Ri
and R2 ancl depths di and d2 respectively, the
central velocities of rise are Vi = diRi/rn and
V2 = d2R2/r]2- If the form were the same, i. e.
(d/R)i = (d/R)2, ancl the viscosity the same in
both cases then we fincl that the ratio of the
velocities is the same as the ratio of the squares
of the raclii. If we apply this to our case ancl
take for Scandinavia R = 900 km and for Ice-
land R = 240 km, we find that Scandinavia
would rise with a 14 times greater velocity
than Iceland, and a filling up of the depres-
sions would take 3.75 times longer time in
Iceland than in Scandinavia. As the actual
times for filling up are in reality about reversed,
the main reason must be much less viscosity
for Iceland than for Scandinavia.
For a final comparison of the two depressions
we need not use the centres, as the synchronous
movement enables us to use anv point in the
164 JÖKULL