Jökull - 01.01.2009, Blaðsíða 99

Slope creep in East Iceland observed by InSAR
fjörður (Jónsson, 2007). Here I briefly describe two of
these sites, more completely than in Jónsson (2007).
Loðmundarfjörður is one of the Eastern Fjords
and used to be a small farming community, but no
permanent settlement has been in this fjord since
1973. Jónsson (1976) describes large deposits in Loð-
mundarfjörður in detail that are called Loðmundarskr-
iður or Stakkahlíðarhraun. He states that the deposits
originate from the Flatafjall and Bungufell mountains
and extend southwards to the fjord bottom, bounded
to the west by the slopes of Karlfell mountain.
Clear signs of displacement are detected in Loð-
mundarfjörður in one-year interferograms spanning
2004–2005. The area that is moving is extensive,
over 1 km wide, and during this year the maximum
LOS displacement is about 3 cm. The detected move-
ment, however, does not take place on the large de-
posits described by Jónsson (1976), but on the east-
ern slopes of Karlfell mountain, under its high cliffs,
somewhat above the Hrauná river, and appears to co-
incide with a location of older deposits (Hjartarson,
1997). The well known Loðmundarskriður, on the
other hand, show no signs of displacement during this
time period. It is not possible resolve precisely when
the movement under Karlfell mountain took place. No
displacement is detected in 1-month interferograms
from summers of 2004 and 2005, but it is seen in
all interferograms that span the time from summer
2004 to summer 2005 (Jónsson, 2007). Therefore, the
movement was either slow and steady, i.e. too slow to
be detected in the 1-month interferograms, or it was
sudden and took place sometime between 12 Septem-
ber 2004 and 15 May 2005.
Eyvindarárdalur is a valley about 10 km southeast
of the town of Egilsstaðir. The road to Mjóifjörður
runs along this valley. The valley floor is 1–200 m
above sea level and it is surrounded by mountains that
are 800–1000 m high. Several interferograms show
deformation in a small area on the southwest facing
slope of the valley, between the rivers Ytri-Grjótá and
Innri-Grjótá. The observed deformation amounts to
about 1–2 cm in these interferograms, consistently
indicating downslope movement and it is always at
the same location. Deformation in Eyvindarárdalur is
seen in several interferograms that span various dif-
ferent time periods during 1995–1999 but no defor-
mation is observed at this location in 2004–2005.
One of the few sites that were known to be un-
stable prior to the start of this project is Urðarbotn,
located 600–700 m above the town of Neskaupstaður.
During the past decades, several debris flows follow-
ing heavy rainfall have originated in Urðarbotn, caus-
ing damage in the town (Jensen and Hjartarson, 2001).
GPS measurements during 1991–2001 indicated non-
steady motion of the Urðarbotn deposits from year to
year with one benchmark showing displacement of
138 cm during 1992–1993, but only 11 cm during
1994–1996 (Jensen and Hjartarson, 2001). Urðarbotn
is located on a SSE-facing slope and can thus only be
imaged from ascending orbits. No suitable ascending
ERS data exist in the ESA archives, so the only usable
data are the Envisat data from 2004–2005. Although
imaging from ascending orbits is possible, it is chal-
lenging due to the steep slopes and cliffs and the lim-
ited extent of the Urðarbotn site. No displacements
were observed in Urðarbotn in the processed ascend-
ing Envisat interferograms (Jónsson, 2007). Three
one-month interferograms exhibit a reasonable degree
of interferometric coherence high on the slopes above
the town of Neskaupstaður, but none of them show
any sign of displacement at the Urðarbotn site. These
interferograms span 4 July–8 August and 8 August–
12 September 2004, and 15 May–19 June, 2005. One
interferogram that spans about one year 2004–2005
shows similar results.
DISCUSSION
The detection of displacements in InSAR data in Þófi
up to two years before cracks were discovered in the
field demonstrates the monitoring potential of this
technique in East Iceland. Deformation on these de-
posits can be seen in InSAR data from 1998 and 1999,
i.e. well before fresh surface cracks were discovered
in 2000. While the Þófi results may demonstrate the
monitoring potential of InSAR, the Urðarbotn obser-
vations expose one of its limitations, i.e. that InSAR
cannot be used to image sites that are small in extent
(<200 m), which is probably the case for many un-
stable slopes in Iceland. However, the Þófi case and
the discovery of a number of other sites that are ac-
JÖKULL No. 59 99