Jökull - 01.01.2013, Page 50

J. Helgason and R. Duncan
SR11. In contact with erosion surface SR11 are 11 vol-
canic formations (SV24 to SV34) of group S5, that
bank up against the mountains Skarðatindur, Ösku-
hnúta and Svínafell ridges (Figure 13c). The strata
in these massifs are continuous and assumed to have
extended far in an east–west direction. That is, the
formations do not peter out toward east or west but
are terminated abruptly by an erosion surface. At the
end of erosion stage SR11 a 300 to 400-m-deep valley,
here referred to as the "Svínafell valley", had formed
on the west side of Skarðatindur and Öskuhnúta where
volcanic group S5 later accumulated. Available dat-
ing helps narrow down the time of formation of the
valley beneath Svínafellsjökull. During SR11 glaciers
carved out the "Svínafell valley". After that erosion
has extended farther west, where Svínafellsjökull is
now located. During formation of volcanic group S5
the floor level of "Svínafell valley" must have been
over 300 m above the present sea level. Since de-
position of group S5 erosion has continued into the
"Svínafell valley" down to about 250 m above present
sea level where group S7 was deposited, about 215
ka. Subsequently, the base below Svínafellsjökull was
eroded down to below 100 m above the present sea
level.
SR12. This erosion surface follows the base of group
S7, or formation SV37 that is correlated with forma-
tion HF43 in Hafrafell and has a date of 215 ka (Hel-
gason and Duncan, 2001). Group S7 is stratigraph-
ically the youngest in Svínafell. Formation SV37,
highly porphyritic basalt lavas, occurs at the same
levels both in Svínafell and Hafrafell (Figure 13d).
The SV37 age correlates well with the third last in-
terglacial. A major valley had formed in the cur-
rent bed of Svínafellsjökull during SR12, as formation
SV37 is exposed on both sides of the outlet glacier.
It is interesting to trace SV37 southward in Svínafell
as it crosses Hrútagil at 250 m a.s.l. There, it di-
verts abruptly toward the SW, i.e., down to about 100
m.a.s.l., where it reaches the base of Bæjargil gully.
We assume that SV37 flowed along a glaciated valley
formed by Svínafellsjökull. The SV37 lava flows may
have covered the lower level of Svínafellsjökull or the
glacier may not have been present in the valley which
may have had a base at some 250 m above the present
sea level. Once out of the valley the lava could flow
down to much lower levels as seen by the outcrop in
Bæjargil at about 100 m a.s.l. Continued erosion up to
the present has further deepened the "Svínafell valley"
to this level (Figure 13d).
Main erosional stages in Svínafell
We now synthesize the temporal and erosion evolu-
tion of the Svínafell area and quantify some of the
associated geological factors. The activity responsi-
ble for erosion surfaces SR1 to SR12 grades from mi-
nor "smoothing" where glaciers have gently scraped
the substrata, to major angular unconformities where
several hundreds of meters of volcanic strata were re-
moved with, in cases, clear examples of valley forma-
tion. The erosion surfaces can broadly be divided into
four stages as presented in Table 4.
INTERPRETATION AND DISCUSSION
Age of the Svínafell lacustrine sediments
No direct dating exists for the Svínafell lacustrine
sediments. Therefore, their age can be constrained
only by information about the age of strata above
and below. Volcanic rock units above the Svína-
fell sediments (group S4) are normally magnetized
and sourced from Öræfajökull volcano. Accord-
ingly, these strata formed during Brunhes chron, i.e.,
<0.78 Ma. This assignment of Brunhes age for vol-
canic rocks above the sediments is confirmed by our
new 40Ar–39Ar age determinations that have provided
three independent ages of lower Brunhes lava flows.
Their mean age is 698 ka, for lava flows that are
stratigraphically some 300 m above the Svínafell sed-
iments. 40Ar–39Ar age determinations for lavas in
Skjólgil, below the Svínafell sediments, proved un-
successful but a dyke that cuts these lavas but which
is terminated beneath the Svínafell sediments gives
a plateau age of 1.67±0.15 Ma. It follows that the
Skjólgil lavas are somewhat older. Their magnetic
signature is, however, important in that they are re-
versely magnetized, except for three normally magne-
tized flows near the section base. This brief N-interval
is of importance for stratigraphic correlation.
It is most likely that the reversely magnetized
Skjólgil lavas can be correlated with the Matuyama
50 JÖKULL No. 63, 2013