Jökull - 01.12.2006, Page 50

Kristjánsson et al.
40-44. A subjective estimate of the directional groups
indicates that the 250 lavas sampled by us represent
only about 150 independent spot readings of the ge-
omagnetic field. This in turn indicates that the build-
up of the lava pile took place episodically, often with
centuries or less passing between successive flows.
Average remanence directions in the collection.
The average direction of remanence in N = 247 lavas
had a declination D = 349◦, inclination I = 74.9◦.
The direction of a central axial dipole field in the area
would be D = 0◦, I = 77.1◦. The agreement between
these two is quite satisfactory considering uncertain-
ties (especially in the estimated tectonic tilt) already
indicated. The vector sum of the directions is R =
231.8, giving an angular standard deviation (a.s.d.) of
the directions as approximately arccos(R/N) = 20.3◦,
uncorrected for within-flow scatter. The mean of
our virtual geomagnetic pole positions has longitude
239◦E, latitude 85.9◦N, with R = 217.6, a.s.d. = 28.3◦
and 95%-confidence angle 3.3◦. If one assumes that
the lavas in this study have a mean age of 7-8 Ma, the
a.s.d. value fits well with the top two curves in Fig-
ure 4 of Kristjansson (2002) which show the observed
variation with age in this parameter in other Icelandic
lava collections.
POLARITY ZONES AND THEIR
RELATION TO THE STRATIGRAPHY
The results of our mapping and paleomagnetic results
are summarized in Figure 3. Beginning in the old-
est lavas in Norðurárdalur, we see that the entire pro-
file TD 2-40 as well as the flows TB 1-59 are re-
versely magnetized. We consider this polarity zone
to correspond more or less to that in flows PF 0a-
35 of Saemundsson et al. (1980), with some down-
dip thickening as is commonly observed in Iceland.
According to the stratigraphic division of Hjartarson
(2003), the thick reverse sequence in TD, TB and PH
represents the Sólheimar formation. Given a south-
southwesterly regional tilt it is not unlikely that the
groups TD 18-24 and TB 1-5, both of which contain
some olivine tholeiites and feldspar-porphyritic lavas,
are at the same stratigraphic level. A small geomag-
netic excursion in TD 26 (Table 1) may be contem-
poraneous with that in TB 9 (and possibly TB 10).
Major geomagnetic excursions occur in PH 12-15 and
in TB 33-36, but we have not indicated a correlation
between these lava groups in Fig. 3. This is because
the resulting apparent dip would be only around 2◦,
rather than several degrees as indicated in Figure 1b.
However, the large dip could be offset by faults in the
intervening area, similar to one seen above PH 39 (cf.
Appendix), and for the same reason the sediments be-
tween PH 44 and 45 may be the same as those above
TB 47. Given that the distance between the profiles
is of the order of 7 km here, it would be desirable to
conduct additional mapping in the intervening valley
of Tungudalur-Egilsdalur.
Not far above the R to N transition at TB 59/59A,
a sedimentary layer of approx. 10 m thickness occurs.
We suggest that this layer may correlate with any of
the sedimentary rocks occurring between flows PH 48
and 51, which also lie just above an R to N transition
at PH 47/48. Continuing along Austurdalur, we ex-
pect that the sediments between flows PG 14 and 18
are the same as those just mentioned.
Northeast of our site PH 60 in Mosgil (Figure 2a),
the gully becomes quite steep-sided and inaccessible
for sampling. We have therefore not been able to look
there for other stratigraphic ties with distinct features
in the uppermost part of TB, such as the single-flow
polarity zone TB 82. This zone seems to appear in the
PG-profile where it has thickened considerably while
the underlying normal-polarity zone is getting thin-
ner. The sedimentary rocks above TB 85 also have
—————————————————————————————————
Figure 3. Stratigraphy and paleomagnetic polarities of the profiles sampled in this study (cf. Figure 1c), also
including polarities in profile PG (Saemundsson et al. 1980). Elevations in meters above sea level are on the
left-hand side. Minor interbasaltic sediments are not shown. Lines indicate correlations between the profiles as
suggested in the text. – Jarðlög í sýnasöfnunarsniðum, með áætluðum tengingum milli þeirra.
48 JÖKULL No. 56