Jökull - 01.01.2016, Síða 88

Leó Kristjánsson
Figure 3. View across Ísafjörður, showing the regular stratigraphy in the area. However, many outcrops are
small and discontinuous, and some flows may be entirely hidden by soil. Profile DW wiggles from sea level up
to about 100 m altitude on the southwest side of the stream Álftagrófará seen at the left-hand edge of the photo;
profile DU is near its right-hand edge. A slight apparent downward tilt of the strata from right to left changes
to an upward tilt farther northeast. – Horft til sniðs DW í Álftagrófará á austurströnd Ísafjarðar, nálægt vinstri
kanti myndarinnar. Hraunastaflinn hefur hér lítilsháttar sýndarhalla út eftir (frá sniði DU hægra megin), en
utar í firðinum snýst hann við. Ljósm./Photo. L.Kr. 2014.
3 kilometers north of DF, see Figure 2b. It was de-
cided to collect at least 5 oriented 25-mm core sam-
ples from each lava in the new profiles instead of 4
in the 1980s survey, in anticipation of relatively low
remanence intensities which turned out to be the case.
The lavas were also somewhat altered by hydrother-
mal activity, and many dike intrusions were noted in
Ísafjörður. This was unexpected, in view of the fact
that the nearest volcanic center (Hjartarson and Sæ-
mundsson, 2014) lies more than 15 kilometers away,
to the south-southeast.
Laboratory methods, and 2012–2014 field work
Both in the study of Kristjánsson (2015) and in its
continuation reported here, AF demagnetization was
routinely carried out on all specimens at 10, 15, 20,
25 and 30 mT peak fields in a tumbler device. Ad-
ditional treatment steps of 35 and 40 mT were often
required, in order to confirm the elimination of a vis-
cous component of unusually high coercivity. In some
cases, duplicate treatment was applied at the 30 mT or
higher steps, to compensate for minor amounts of ro-
tational remanence acquired. A number of samples
had to be rejected due to unstable or aberrant rema-
nence directions. The units thus affected were gener-
ally resampled in a new location, removed meters or
tens of meters from the initial one. The demagnetiz-
ing step yielding the smallest α95-angle was selected
as the best mean for each site.
The results presented in Table 1 of Kristjánsson
(2015) showed that the within-flow agreement is gen-
erally excellent. The estimates k of the precision
parameter usually are greater than 200 and the α95-
values are smaller than 6◦. This accuracy permits the
use of remanence directions for correlation of indi-
vidual lava flows in different sampling profiles. Some
of these correlations are shown in Figure 4 (modified
from Kristjánsson, 2015). The method is in particu-
lar justified during periods of irregularly varying field.
The field direction is then unlikely to precisely repeat
itself at later times, so that the occurrence of closely
similar remanence directions at two sites will indicate
that they represent either the same flow or two flow
units erupted a short time (centuries or less) apart.
Aided also by the overall apparent dip and superficial
lithological observations on outcrops and cores, Krist-
jánsson (2015) merged the results of the new profiles
into a single visualization of VGP movement. For this
pole path the 25 successive excursion flows DX 1–3
and 6, DV 1–7 and DT 2–14 were selected. These
are shown as red dots in Figure 5, connected by ar-
bitrary curves (segments of great circles, or modifi-
cations thereof to avoid overlaps). The three normal-
polarity lavas sampled at the top of profile DU are also
included. Some VGPs from the profiles of Kristjáns-
son and Jóhannesson (1996) were identical to poles
in Figure 5, while in other cases (partly due to the
presence of unexposed and inaccessible flows in the
88 JÖKULL No. 66, 2016