Jökull - 01.12.2006, Side 44

Kristjánsson et al.
that four central volcanoes were active successively in
the period c. 10–6 Ma ago the Flókadalur, Öxnadalur,
Ábær and Keldudalur centers in order of decreasing
age. He pointed out the presence of thick (up to 100–
200 m) coarse sedimentary rocks at a certain level in
the lava pile all across the southern part of the area.
Jóhannesson (1991) suggested that these sedimentary
rocks were deposited towards the end of activity in
the Öxnadalur central volcano, which according to a
diagram on p. 40 of his paper was occurring around
8.0 Ma ago. Jóhannesson did not indicate whether
they represent a major hiatus in the regional volcanic
build-up in North Iceland or an angular unconformity.
Kristjánsson et al. (2004) published a detailed
map and paleomagnetic study of a composite section
of 8 profiles in the valleys south and south-west of
Eyjafjörður (Figure 1a), referring also to preliminary
Ar-Ar dating results from these profiles presented by
Hardarson et al. (1999). The thick sedimentary rocks
mapped by Jóhannesson (1991) are present in their
composite section. As the preliminary dates are some-
what scattered and not always in stratigraphic order,
the age of the sediments based on them can hardly be
given with better accuracy than 8.5 ± 0.5 Ma. Pre-
liminary Ar-Ar dates (Hardarson et al. 1999; B. S.
Hardarson, pers. comm. 1999) indicate that the age
of the youngest rocks sampled by Kristjansson et al.
(2004), i.e. the upper part of profile GR of Figure 1a,
is 5–5.5 Ma.
In his Ph.D. thesis Hjartarson (2003) made a de-
tailed study of the bedrock geology of the Austur-
dalur and Vesturdalur valleys in Skagafjörður. He de-
scribed the stratigraphy of the valleys and divided it
into groups, formations and members. A geological
map (Hjartarson et al. 2003a) in this thesis includes
the southern part of Figure 1b. The Tinná central
volcano which is the most prominent geological fea-
ture of the area, was described in detail by Hjartarson
(2005) who considers this volcano to have been active
from 6–5 million years ago. It encompasses both of
the above-mentioned Keldudalur and Ábær centers of
Jóhannesson (1991).
To improve the knowledge of the geology of
the Skagafjörður-Eyjafjörður region, we have car-
ried out some additional stratigraphic mapping in the
Norðurárdalur and Austurdalur valleys, along with
paleomagnetic sampling in six profiles of Figures
1b,c. Brief notes on these profiles and profile PG are
given in the Appendix. Access to the lava pile in the
area is variable; our profiles lie mostly along streams,
which in general provide the most complete exposures
cf. Figures 2a,b.
PALEOMAGNETIC SAMPLING AND
MEASUREMENTS
Cores of 25 mm diameter and 4-8 cm length were
collected in 2002-2004 using water-cooled portable
drills, by Leó Kristjánsson and Haraldur Hallsteinsson
with assistance by Eyjólfur Magnússon. Of the pro-
files shown in Figures 1a,b,c, PG was sampled in 1976
at three cores/flow and results on its paleomagnetic di-
rections were published in detail by Saemundsson et
al. (1980). It should be noted that some lavas in this
profile were either not sampled (PG 16 and all above
60) or gave inadequate agreement between the sam-
ple directions (PG 2, 10, 11, 26, 27A). At least four
core samples were collected from most of the num-
bered lava flows in the profiles TD,TB, PH, PM, PN
and PO. These are all listed in Table 1 which also in-
cludes two flows (PH 47, 48) where two samples were
taken and some where only one sample was collected
(PH 25, TB 32, 86 and 86A, TD 24). Lavas PH 37,
PM 9, TD 1, 14 and TB 10, 15, 30 were not sampled.
The outcrops with 0-2 samples were either difficult
to reach, near a dike, crumbly, or thin units without
sedimentary interbeds above or below. The core sam-
ples were oriented in situ by sighting on the Sun or
on distant objects whose position was read from maps
or determined with a pocket GPS receiver. The total
uncertainty in orientation is of the order of 2-3◦.
One specimen of about 22 mm length was cut
from each core. Except for profile PG, remanence
measurements were made by L. K. using an “Institut
Dr. Förster” four-probe static fluxgate magnetometer.
In each specimen, the natural remanence was mea-
sured before alternating field treatment and then af-
ter treatment at 10, 15 and 20 mT peak fields in a
Molspin demagnetizer with a two-axis tumbler. The
10 mT treatment removes most or all of the viscous
42 JÖKULL No. 56