Jökull - 01.12.1990, Side 20

Diamictite units 11 and 12 in the Hvalfjörður area
are separated from the underlying tillites by approxi-
mately 100 m of olivine tholeiite lavas and appear in
sections in the Litla Botnsá gully and in the Brunná
gully (Figs. 3 & 7d). These diamictite units have the
same lateral extent as the previous two glacial de-
posits (diamictite units 9 and 10) and display almost
identical facies associations, i.e. basal tillite asso-
ciated with both slump deposits and thick lacustrine
sediments, beside thick and laterally extensive hyalo-
clastite sheets.
The uppermost diamictite unit under discussion
here crops out in the form of a brecciated hyaloclastite
and tuffaceous sandstone and appears approximately
100 m higher within the stratigraphic section.
IMPLICATION OF THE GLACIAL
DEPOSITS OF THE UPPER BORGAR-
FJÖRÐUR AND HVALFJÖRÐUR AREA
Lithofacies analysis of interbedded diamictites in
the upper Borgarfjörður area and the Hvalfjörður area
reveals repeated glacial advance and retreat. Five
major glacial advances are recognized in the upper
Borgarfjörður section and at least six glacial advances
are recognized in the Hvalfjörður area within the time-
span under discussion (3.1 Ma-1.8 Ma). The impli-
cations of the glacial lithofacies as climate indicators
are however severly limited. A glacial deposit, such
as melt-out or lodgement tillite, in a given sequence,
is an evidence of a glacial activity in that area, but it
offers very little information about the duration and
gives only the minimum extent of the glacier advance.
A chronological context and a meaningful compari-
son between areas is necessary in order to evaluate the
regional significance of these glacial events.
A correlation between sections of the upper Borg-
arfjörður area and the geopolarity time scale is based
on several K/Ar dates (McDougall et al., 1977). The
combined data indicate that the sequence is virtually
continuous from the early Matuyama back to the base
of epoch 6 and into epoch 7 (McDougall et al., 1977).
Correlation between sections around the Hvalfjörður
area is based on the results from magnetic polarity
measurements and supported by few K/Ar dates, from
the mountain Akrafjall (3.12-2.87 Ma, this study) and
from the top of the mountain Esja (1.8 Ma, Kristjáns-
son et al., 1980). Composite stratigraphic columns
of the two areas show at least seven reversals of the
Earths magnetic field during the time the volcanic
rocks of the region were erupted. The base of both
stratigraphic successions has been correlated to the
base of the Mammoth reversed polarity event of the
Gauss epoch the youngest strata discussed here are
thought to precede the Olduvai normal polarity event
(Kristjánsson et al., 1980). Based on these data, a
tentative correlation and speculations of the inferred
glaciations within the areas can be made.
The published K/Ar dates from the Borgarfjörður
area in conjunction with the palaeomagnetic polar-
ity data from the lavas show that the volcanism in
the region was virtually continuous between about 7.0
and 2.0 Ma. A linear regression analysis based on
22 K/Ar dates was used to assign dates to the strati-
graphic column (McDougall et al., 1977). By adopt-
ing the regression equation, it is possible to infer the
approximate age of the first glacial horizon identified
within the area (Fig. 8). In constructing the regression
equation, the accumulation rate was assumed constant.
Although this approach is unlikely to be completely
correct as pointed out by McDougall et al. (1977), it
can be used to obtain age estimates. The regression
equation McDougall et al. (1977) used is:
y = 6.72-0.001348®
where y is the age in million years and x is the height
in metres above the base of the section. Using this
equation a calculated age of 2.6 Ma for the first identi-
fied glacial deposit was obtained. If the same method
is applied to the fourth glacial deposit (diamictite unit
9) in Borgarfjörður, a calculated age of 2.07 Ma is
obtained. This suggests that the four diamictites iden-
tified as glacial deposits are distributed over approx-
imately 500,000 years. The fifth and last glacial unit
recognized in the area is not included in the regres-
sion analysis because of the unconformity between
the fourth and the fifth glacial deposit.
If the linear regression analysis derived from the
K/Ar dates in Borgarfjörður is valid, and the calculated
age for glacial deposit 4 (diamictite unit 9) is reason-
18 JÖKULL,No. 40, 1990