Jökull - 01.07.2003, Side 31

Paleomagnetic observations on Late Quaternary basalts, SW-Iceland
in three hills in the southern central part of the penin-
sula. The geomagnetic excursion generating these
was named the “Skálamælifell excursion” after one of
the three hills (see photo in Kristjánsson, 1992), the
others being Siglubergsháls and Austara Hraunssels-
Vatnsfell (Figure 4).
Additional sampling in the 1980’s revealed other
sites carrying similar transitional remanence direc-
tions (Levi et al., 1990). The lavas in question tend
to be light-gray, olivine-rich but not picrites. Samples
of these rocks were K-Ar dated at two laboratories. A
mean age of 43 8 kyr was indicated, but it must be
kept in mind that these measurements are difficult to
carry out on such young lavas with very low K con-
tent. Extensive paleointensity measurements (Mar-
shall et al., 1988; Levi et al., 1990) showed that the
field strength was less than one-tenth of the present
Earth’s field. It is most likely that the Skálamælifell
excursion dates from the same period of instability of
the geomagnetic field as the Laschamp lavas. Relative
paleointensity estimates on late Quaternary sediments
(e.g. Laj et al., 2000) confirm a drastic reduction in
the field intensity at around this time.
The Skálamælifell excursion lavas still represent
the only group of pre- Holocene basalts in the penin-
sula which has been dated directly by radiometric
methods. These lavas have been found to be a valu-
able reference horizon in research on glaciations in
the peninsula (Guðmundsson, 1995) and will no doubt
also be of value in future work on the history of vol-
canism and tectonics in the area.
New results – Skálamælifell excursion
Since 1990, a number of sampling trips to the area
around Skálamælifell have been undertaken by the au-
thor, to search for further evidence of the geomagnetic
excursion. Only crystalline rocks have been sampled
(mostly lavas but also a couple of dikes), and only
one sample has been collected at most of the outcrops
visited. One new separate location carrying the dis-
tinctive Skálamælifell direction has been discovered,
namely the small hill Einbúi southwest of Fagradals-
fjall (Kristjánsson, 1992). Two previous outcrop areas
have been extended: to the northeast from Skálamæli-
fell (eastern slopes of Einihlíðar ridge), and around
Siglubergsháls. From Figure 4 which is a revision
of Figure 1 in Levi et al. (1990) it is fairly evident
that the emplacement of the Skálamælifell lavas took
place on a SW-NE trending fissure swarm. As before,
lavas carrying the excursion direction tend to be light-
gray in color, and contain olivine phenocrysts. Plagio-
clase feldspar phenocrysts are also seen occasionally
in these.
Levi et al. (1990) calculated average remanence
directions for lavas in three groups from the follow-
ing areas: A, Siglubergsháls; B, Both of the two
Hraunssels-Vatnsfell hills, and Höfði; C, Skálamæli-
fell and outcrops north of there.
20 mT peak field is usually sufficient to give a
stable direction (very little change occurring from 15
mT). With the additional material from the Siglu-
bergsháls area and Einbúi, there are now 80 reliable
samples from area A. Their mean direction has D =
264Æ, I = -19Æ, with vector sum R = 78.46, a.s.d. =
11Æ. The correspondingV.G.P. is at 11ÆS, 249ÆE (Fig-
ure 2b).
In area C, I have also added 13 new samples from
Skálamælifell and Einihlíðar. No stable samples were
collected in area B after 1990. As the areas B and
C gave very similar directions, they have now been
combined, giving a mean direction with D = 257Æ, I
= -11Æ, R = 64.0 from 66 samples, a.s.d. = 14Æ. The
V.G.P. is at 10ÆS, 257ÆE (Figure 2b). Altogether 18
samples were discarded because their remanence was
very weak or unstable, and a few other samples were
discarded because of lightning-strike problems.
Remanence inclinations in the two groups are
somewhat scattered, mostly lying between +10Æ and
-30Æ. This scatter may have a number of causes other
than a change in the main geomagnetic field between
eruptions. These include: residual secondary (vis-
cous) magnetization; instability of the primary re-
manence; unstable components acquired during de-
magnetization and measurements; and local anoma-
lies present during emplacement. Such anomalies are
potentially more disturbing for accurate measurement
of the primary remanence in the present case than in
Iceland in general, as the primary remanence is very
weak and has a low inclination. Furthermore, the
problem of outcrop movement may be more serious
here than in, say, the Tertiary areas of Iceland, be-
JÖKULL No. 52, 2003 29