Jökull - 01.12.1990, Qupperneq 23

three oldest glacial deposits in Hvalfjörður.
Previous work correlated the first diamictite unit
111 the upper Borgarfjörður area (Giljafoss) with the
first diamictite unit in the Hvalfjörður area (Akra-
fjall), both deposits being interpreted as tillites and
ascribed to a major glaciation in both areas (Kristjáns-
son et al. 1980). This correlation is not supported by
the sedimentary analyses of the diamictites. Although
these diamictites occur within the same stratigraphic
position (based on K/Ar dates and paleomagnetic re-
sults from the underlying lava flows), they can not
have been formed by the same sedimentary processes.
The Hvalfjörður diamictite indicates that sedimenta-
tion occurred periodically in a fluvial environment and
not as a continuous process. It is located between lava
flows, 3.12 to 2.87 Ma of age, and could thus have
been formed during the 250,000 years between lava
formation. The Giljafoss diamictite in Borgarfjörður,
however, preserves features indicative of a very sud-
den deposition as a surge or a volcanic debris flow.
CONCLUSION
This study reveals that the lowermost three dia-
mictites at Borgarfjörður and at least the two low-
ermost diamictites in the Hvalfjörður area are non-
glacial. Volcanic activity has prevailed in both areas
during the formation of these sediments. Most of them
show the effect of water during deposition. That,
however, is not a direct evidence of a major glacial
advance. Such advance is first identified with the
fourth diamictite unit in Borgarfjörður and the fifth
diamictite unit in the Hvalfjörður area. Five major
glacial units are identified in the Borgarfjörður area
and at least six in the Hvalfjörður area within the late
Pliocene time. Their stratigraphic position does not
allow a direct correlation, which is mainly hampered
by a poorer stratigraphic control of the Hvalfjörður
sequence. However, the magnetic polarity data and
the potassium-argon data from the Borgarfjörður area
Permit a tentative correlation between some of the
'dentified glacial horizons.
In order to assign a regional glacial status to de-
Posits generally described as diamictites, a multicri-
teria approach has to be advocated. These deposits
must conform to particular lithofacies characteristics
and meet other requirements such as mappability, spe-
cific contact relations, and regional extent. The sig-
nificance of the glacial units in westem Iceland can
not be evaluated unless they can be correlated with
glacial deposits in other parts of Iceland. Opposite to
lava formation, the time involved during deposition of
sedimentary units may vary greatly. They may form
in the course of several days, weeks or months, but
they may also form in decades or hundreds of years.
To accomplish valid correlation between sedimentary
units, two problems have to be addressed. One prob-
lem is the correct genetic identificationof the deposits.
This should also include more work on the formation
and deposition of hyaloclastites. The second problem
has to do with the dating of the sediments. Continuing
studies of diamictites interbedded within basaltic lava
suites of Cainozoic age in Iceland, deal specifically
with these two problems (Eiríksson and Geirsdóttir in
press). Furthermore, it is anticipated that continuing
work on diamictites in Iceland can further define the
trends of Pliocene and Pleistocene glaciation in the
North Atlantic and be tested through comparison with
the latest deep sea record of glacial-interglacial cycles.
ACKNOWLEDGEMENTS
This paper forms a part of a Ph.D. thesis supervised
by Dr. J.T. Andrews at the University of Colorado,
Boulder. The work was funded by NSF dissertation
grant no. EAR-8508997, by a grant in aid of research
from Sigma Xi and by grants from the Graduate School
and the Department of Geological Sciences of the Uni-
versity of Colorado, Boulder. Thanks are due to Dr.
Jón Eiríksson for suggesting the topic for the disser-
tation. Special thanks are due to the family at Auga-
staðir, Hálsasveit in Borgarfjörður for their hospitality
during fieldwork. Thanks are also due to Andrés I.
Guðmundsson, Björg Pétursdóttir, Már Vilhjálmsson
and Rainer Zahn, who all gave generously of their
time to accompany and assist me in the field. I would
also like to thank Dr. Robert C. Walter, for making
the K/Ar age measurements from Akrafjall available
and Dr. Leó Kristjánsson for providing the samples.
The manuscript has been improved by the comments
of Dr. Jón Eiríksson, Dr. Kristján Sæmundsson, Dr.
Hreggviður Norðdahl and Andrés Guðmundsson.
JÖKULL, No. 40, 1990 21