Document Actions

ijege-13_bs-kristensen-et-alii.pdf

background image
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
339
DOI: 10.4408/IJEGE.2013-06.B-32
GB INSAR MEASUREMENT AT THE ÅKNES ROCKSLIDE, NORWAY
L
ene
KRISTENSEN
(*)
, C
arLo
RIVOLTA
(**)
, J
ohn
DEHLS
(***)
& L
ars
h. BLIKRA
(*)
(*)
Åknes/Tafjord Early Warning Centre - Ødegårdsvegen 176 - 6200 Stranda, Norway
Email: lk@aknes.no. Telephone: +47 40040265
location on the opposite side of the fjord since 2006.
In 2012 these measurements were supplemented with
a short overlapping campaign from a closer location
immediately outside the unstable area.
InSAR (or Differential InSAR - DInSAR) is a re-
mote sensing technology that is increasingly being used
to detect displacements on the earth surface. It utilizes
the phase difference between two coherent images to
measure change in distance. A detailed description of
the InSAR method, in particular as measured by satel-
lite is given by F
arretti
et alii, (2007). The method
can detect small changes in distance to a satellite or a
ground based system in a way that can be independent
of installations on the ground. InSAR from satellites
have been used for regional landslide mapping or char-
acterization of particular landslides (t
aburini
et alii,
2011; r
eiden
& W
aLther
, 2007). Satellite based In-
SAR is becoming more applicable to detailed surface
instability studies, as images from new satellites have
improved temporal and spatial resolution, for example
Terrasar X and in future the ESA Sentinel satellites.
When monitoring a limited area, such as a single
landslide; ground based InSAR systems have some
important advantages over a satellite based ones.
They offer measurements of much shorter time inter-
vals (minutes rather than several days) and a better
spatial resolution without being controlled by satellite
geometry (C
asagi
et alii, 2010). Ground based InSAR
systems usually measure from a fixed position so the
baseline (difference in radar position) is zero.
ABSTRACT
Åknes is a 54 mil. m
3
rockslide at Storfjorden,
Western Norway. It has the potential to form a devastat-
ing tsunami in the fjord, and therefore it is thoroughly
investigated and continuously monitored. Five GB In-
SAR campaigns have been carried out since 2006 in
order to investigate the patterns of surface displace-
ment. Here we present the results of these measure-
ments, and compare them with in situ measurements of
surface displacement. The data provide detailed infor-
mation on the displacement fields in the rockslide. The
InSAR system is placed opposite the fjord in relation to
the rockslide and due to unusually strong atmospheric
disturbance, a statistical filter needs to be applied to
the signal when processing. Therefore the data is not
included in the early warning system, but in 2012 an-
other GB InSAR system was installed at a different
site, which focus on the upper and most active part of
the rockslide - and this will be used for early warning
in critical phases of the rockslide monitoring, such as
during accelerated movement or after a partial collapse.
K
ey
words
: Åknes, rockslide, landslide, monitoring, GB InSAR
INTRODUCTION
The Åknes rockslide poses a big risk to human
lives, and is thoroughly investigated and closely mon-
itored. As a part of the rockslide investigations, five
ground based interferometric synthetic aperture radar
(GB InSAR) campaigns have been carried out from a
background image
L. KRISTENSEN, C. RIVOLTA, J. DEHLS & L.H. BLIKRA
340
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
present activity of rockslides for risk classification
(K
ristensen
, 2011), K
ristensen
, 2012 (b), K
ristensen
,
2012 (c) and to identify unstable areas in along roads
(K
ristensen
& r
ivoLta
, 2011).
THE ÅKNES ROCKSLIDE
Åknes is a large rock instability in Storfjorden,
Western Norway, and is one of four mapped high-
risk rockslides in Norway subjected to permanent
monitoring. The maximum volume of the unstable
area is estimated to 54 million m
3
(b
LiKra
et alii,
2010). Due to its size, relatively high probability, and
Ground based InSAR systems are for exam-
ple used to monitor open pit mines (L
ingua
& r
in
-
audo
, 2008; F
arina
et alii, 2011), construction sites
(b
ozzano
et alii, 2011) and volcanos and landslides
(a
ntoneLLo
et alii, 2004; s
ChuLz
et alii, 2012). For
rockslide studies it has been used on Ruinon (t
arChi
et alii, 2003; a
gLiardi
et alii, 2011) and Cortenova
rockslide (t
arChi
et alii, 2005). In Norway, the GB
InSAR method has been used in the investigations
and early warning at the high risk rockslides Mannen
(K
ristensen
& b
LiKra
, 2011) and Jettan [K
ristensen
,
2012 (a)]. Also it has been used to characterize the
Fig. 1 - Map of the Åknes rockslide with its main features and the location of some of the instruments
background image
GB INSAR MEASUREMENT AT THE ÅKNES ROCKSLIDE, NORWAY
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
341
tem was placed at the opposite side of the fjord from
the rockslide (see Fig. 3). In 2005, the first attempt of
radar measurements was made with the radar placed
very close to the fjord; however the results were of
low quality due to atmosphere and tidal effects in the
images. From 2006, the radar was placed in a perma-
nent shed, with a plastic covered window, which is
close to transparent to the microwaves. Power and in-
ternet is available at the site.
In 2012, a similar system was placed at Fjel-
lvåken, immediately west of the fault at the western
limit of the landslide, looking into the most active
part of the rockslide (Fig. 2). The measurements last-
ed only for 4 weeks, compared with almost 4 months
at Oaldsbygda, and the radar was left without pro-
tection during the measurements. In Tab. 1 the radar
parameters from both sites are listed.
In general, atmospheric changes disturb the phase,
and this effect is normally corrected in the LiSALab
software by defining a stable area and assuming these
conditions are similar at the instable areas as well.
From Oaldsbygda, the atmospheric effects are
unusually strong, mainly because the radar is look-
ing across the fjord.
Weather and local atmospheric conditions here are
location above a fjord with the possibility to generate
large tsunamis, it is considered to be the rockslide
in Norway with the highest risk (g
anerød
et alii,
2008). The modeled tsunami run-up for the worst-
case scenario at the closest village Hellesylt is 85 m
(NGI, 2011). Since 2004 detailed investigations have
been carried out at Åknes. It includes geomorpholog-
ical and structural mapping, geophysical surveying,
borehole drilling and measurements of surface and
subsurface displacement.
The landslide is about 1000 x 500 m in size, with
the uppermost graben and upper tension fracture at
900 m and the toe at 100 m asl. (Fig. 1).
The bedrock consists of gneisses of Proterozoic
age, which were altered and reworked during the Cal-
edonian orogeny (g
anerød
et alii, 2008). The preex-
isting structures in the gneiss, such as foliation, faults,
joints and folds control the rockslide development and
movement. The upper back scarp is parallel to near
vertical foliation planes in the strongly folded gneiss.
The basal sliding plane is located along foliation,
subparallel to the surface and related to biotitic lay-
ers in the gneiss, probably with some subvertical steps
along some steep fractures (g
anerød
et alii, 2008; J
a
-
boyedoFF
et alii, 2011). This is supported by studies
of the prehistoric sliding surface of Rundefjellet, only
2.5 km from Åknes (o
ppiKoFer
et alii, 2011).
The continuous monitoring network at Åknes
consists of: nine GPS antennas, two lasers, three ex-
tensometers, a robotic total station measuring to 28
prisms, a micro seismic network and a meteorologi-
cal station (temperature, precipitation, snow depth and
windspeed). In three boreholes drilled to penetrate the
basal sliding surface, DMS columns of 100 and 120 m
are inserted, which measure inclination at every meter
along its length. The current displacement rate ranges
from 5-8 cm pr. year in the upper flank and 2-3 cm pr.
year in the middle sector. The lowermost part of the
mapped instability is inactive at the present. From the
DMS measurements and the logged boreholes we see
the basal sliding surface at 50-63 m depth at the upper
flank, at 33 m at the middle borehole, and is not seen
presently at the lower borehole.
SYSTEM DESCRIPTION AND METHODS
The rockslide has been scanned in five campaigns
from Oaldsbygda (2006, 2008, 2009, 2010, 2012) by a
GB InSAR system from LiSALab Ellegi Srl. The sys-
Fig. 2 - The LiSALab InSAR system used at both Oalds-
bygda and Fjellvåken. Photo from Fjellvåken,
looking toward the upper flank of the rockslide
Tab. 1 - Radar system parameters
background image
L. KRISTENSEN, C. RIVOLTA, J. DEHLS & L.H. BLIKRA
342
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
that was subtracted to the radar images. This proce-
dure led to radar images, where the individual pixels
represented the phase values that can be considered
free from the atmospheric effects and noise and of
good quality. After applying the algorithms, the re-
sidual atmospheric effect in the images was in the
order of 1 mm, which is an acceptable level for the
GB InSAR method.
At Fjellvåken the radar is located at similar al-
titude as the upper flank and graben area, and the
unique atmospheric disturbance experienced at
Oaldsbygda is entirely absent.
All radar images were processed with the LiSAL-
ab software. Daily images were cumulated, in order
to avoid phase wrapping when looking at longer time
intervals. Then the images were projected or subject to
coordinate transformation to display the data in Arc-
GIS, together with other types of monitoring data and
satellite based InSAR.
DISCUSSION OF LOS
An InSAR system will measure the difference
in distance to the target in the Line Of Sight (LOS),
which is usually less than the true displacement vec-
tor. It is possible to obtain the true displacement if
three (or, with some assumptions, two) radar systems
are measuring the same target area simultaneously
(s
everin
et alii, 2011). The very big advantage of
obtaining overlapping measurement is the possibility
to draw the displacement vectors and thus identify
sectors or blocks with different velocity or direction.
At Åknes we had a period of two systems measur-
ing at overlapping interval; however the view angle
and ground resolution was very different, making it
difficulties to calculate the true displacement vectors.
It would basically have required a third radar sys-
tem, and still the pixel coverage would be poor.
At Åknes, measurements from several GPS anten-
nas and from a total station provide the exact displace-
ment vectors in three coordinates at several points.
These were used to calculate the discrepancy between
the true displacement vector and the LOS (Fig. 3).
From Oaldsbygda to the central part of Åknes the
radar will measure around 70% of the true displace-
ment. A graben is developed in the uppermost part
of the landslide, and the vertical displacement here
is much greater, and furthermore there is a strong
western displacement component towards the western
characterised by sudden and very fast changes. Changes
in the environmental conditions that occur inside the
same acquisition disturb the image quality proportion-
ally to the magnitude of the change - the greater the
changes are, the more the image quality is reduced. This
means that not only do the SAR radar images differ
from each other, because of the atmospheric effect, but
even within single images this effect produces errors.
The problems affecting the quality of the radar im-
ages in this specific site can be summarised as follows:
1. The lower part of the rockslide is covered by ve-
getation which moves randomly in the wind, cau-
sing low coherence values in the SAR images.
2. The microwave signal travels through a strongly
stratified atmosphere, and in each layer the lo-
cal temperature and humidity are very different.
This stratification is caused firstly by the big dif-
ference in height of the target area (0 to 1.400 m
a.s.l.) and secondly probably caused by air stre-
ams flowing to and from the slopes caused by the
presence of the fjord.
3. The sea act as a mirror for the microwave signals
producing multi-path effects. The tide effects
create additional noise in the lower part of the
target area, which is already affected by decorre-
lation due to vegetation.
Due to these strong atmospheric effects it was
necessary to apply and develop appropriate process-
ing methods to “filter” the data. All the images ac-
quired for each campaign were statistically analysed
and processed in order to identifying the fluctuant
values of the phases linked to the atmospheric ef-
fects and noise, producing a bi-dimensional mask
Fig. 3 - The location of the two GB InSAR systems, and the
LOS. The black arrows show size and direction of
displacement from in situ instruments
background image
GB INSAR MEASUREMENT AT THE ÅKNES ROCKSLIDE, NORWAY
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
343
movement here is measured by the radar. However,
the main part of the flank area (Renne and Upper bore-
hole in Fig. 4), as well as towards the middle borehole
is seen very well from the radar. In fact, in this area the
data from the GB InSAR provide the most complete
picture of the displacement, and these data can be used
to improve the estimated failure scenarios.
The total displacement from the 2012 radar cam-
paign (12.07.2012 - 20.10.2012) at the point of five
prisms was compared with displacement measured by
the prisms in the same time interval in Tab. 2. The
percentage of the true 3D movement that is measured
in LOS is calculated for each prism, and a corrected
“true” movement measured by the radar.
fault, which differs from the LOS direction. It means
that the radar only see about 30% of the 3D displace-
ment at this area.
From Fjellvåken to the most active graben area,
the displacement in the LOS direction is about 42% of
the total movement. However, the area further down
towards the small canyon has a displacement vector
that is basically perpendicular to the LOS, and thus no
movement can be seen from the GB InSAR.
RESULTS
Figure 4 show the result of the 2012 measure-
ments from Oaldsbygda. The movement is compared
with the 3D displacement from the DGPS antennas
(blue arrows) and prisms as measured by an automatic
total station (black arrows).
The graben area where we measure the largest
velocities with GPS and total station (5-8 cm/yr) is
poorly covered with radar data, and since the verti-
cal depression in this area is larger than the horizon-
tal displacement (dip 50-60°) only 31% of the actual
Fig. 4 - The GB InSAR data from 2012, compared with 3D displacement of GPS antennas (blue arrows and prisms measured
by a total station (black arrows)
Tab. 2 - Displacement from the radar image (mm), com-
pared with displacement measured at prisms at
the same location
background image
L. KRISTENSEN, C. RIVOLTA, J. DEHLS & L.H. BLIKRA
344
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
We see that the movement measured by the radar
is very similar to that of the prisms in the same inter-
val, with the biggest errors of -13 and + 17%. This is
within the uncertainty of the instruments.
COMPARISON BETWEEN THE DIFFERENT YEARS
In Fig. 5, the measurements from 2006, 2008,
2010 & 2012 are shown. All campaigns started in July
and ended in October, but there are some slight differ-
ences in campaign length, as listed in the figure.
In general, we see a consistent pattern of displace-
ment from year to year, with some variations in cover-
age and magnitude of displacement. Some of the differ-
ence can be explained by different campaign lengths.
COMPARISON WITH SATELLITE INSAR
In Fig. 6 the mean velocity, as measured by Ra-
darsat-2 Ultrafine, average between 2008 and 2012 is
shown. We see a very similar pattern to the GB InSAR
measurements, but with much less coverage in the mid-
dle and lower parts of the slope. Coverage in the Graben
area is lacking in the satellite InSAR data as well.
CHANGE IN MOVEMENT IN 2012
On 22.09.2012., a block collapsed into the upper
tension fracture (Fig. 7). This was one of several visual
signs of increased activity in this part of the rockslide.
Extensometer two, which is located very near
the collapsed block, showed only a little unusual
movement before and after the collapse, and no
change in the general velocity was observed. The
dataseries which goes back to 2006 show displace-
ment of about 2 cm/yr.
However, as Fig. 8 shows, the movement meas-
ured by the GB InSAR in this particular area, was
much larger in 2012 compared to the previous years.
One of the prisms measured by total station (prism
14, location: Fig. 8) is located on this block. Fig. 9
shows one year of displacement measurements in
three directions. The movement is primarily vertical
(dip: 70º) and this is probably why not all the move-
ment is recorded by the extensometers. It appears that
the area is undergoing change and subsidence, though
unfortunately we do not have reliable 3D data before
September, 2011. The implication is probably that the
Fig. 5 - Total cumulated displacement in four of the five years of summer GB InSAR campaigns from Oaldsbygda. The black
lines are the mapped scenarios
background image
GB INSAR MEASUREMENT AT THE ÅKNES ROCKSLIDE, NORWAY
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
345
DATA FROM FJELLVÅKEN
In 2012, the uppermost part of Åknes was meas-
ured by GB InSAR placed at Fjellvåken. As shown
in Fig. 3 the LOS sees maximum 42% of the true
displacement, which is not ideal. However, these
measurements were of very good quality, and give
western graben structure is expanding and being de-
veloped towards east. The GB InSAR measurements
are, together with visual inspections of collapses and
crack developments, our primary source of informa-
tion on change in this section.
Fig. 6 - Radarsat-2 Ultrafine, average velocity from 2008 to
2012, mm/yr. LOS is 32° from vertical, direction 257
Fig. 7 - Collapse of block on 22.09.2012. Extensometer 1
can be seen just left of the collapsed block
Fig. 8 - Change in displacement in an area near the extensometers, from 2010 to 2012
Fig. 9 - One year of displacement from prism 14, located in the block of increased velocity
background image
L. KRISTENSEN, C. RIVOLTA, J. DEHLS & L.H. BLIKRA
346
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
Fig. 10 - Interferogram showing displacement at fjellvåken 27.08.2012 – 20.09.2012
slide, where many other sensors are vulnerable to ca-
ble damage etc. In the LOS from the radar at this site,
the rock is so steep that snow accumulation is less of
a problem and the atmospheric noise is limited. After
a major rockslide event, the radar will be immedi-
ately established in order to document displacements
and analyzing the stability conditions. This will be
critical data for the evaluation of prolonged evacua-
tion and for the risk related to reestablishing in-place
monitoring systems after an event.
SUMMARY AND CONCLUSIONS
The results of five years of summer GB InSAR
campaigns from Oaldsbygda are presented and com-
pared with measurements from in situ instruments
and Radarsat-2 InSAR. The data coverage is poor in
the upper graben area where we measure the largest
displacement with in situ instruments. The coverage
is very good in the central part of the landslide, while
it is more sporadic in the lower part due to dense
forest. The pattern of displacement is similar from
year to year but we see an important change in 2012,
detailed information of the displacement of the
graben area. The GB InSAR measurements are pro-
viding the only data we have on the very steep slope
south west of the graben (Fig 10).
USE OF GB INSAR FOR EARLY WARNING
The high temporal and spatial resolution of a GB
InSAR system, as well as it is not affected by weath-
er and poor visibility, suggest that it is very useful
to include in an early warning system. It has been
used for early warning purposes at Ruinon rockslide
(eg. C
rosta
et alii, 2012), and also at the Norwegian
rockslide Mannen. However, the strong atmospheric
disturbance of the signal recorded at Oaldsbygda,
and the need for assisted processing, makes it dif-
ficult to use it for early warning. Furthermore, most
of Åknes is almost entirely covered by snow in the
winter time, leading to poor conditions for radar
measurements for half of the year.
Instead the newly installed radar point at Fjel-
lvåken is planned to be used in critical conditions;
for example during acceleration phases of the rock-
background image
GB INSAR MEASUREMENT AT THE ÅKNES ROCKSLIDE, NORWAY
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
347
which suggest that the upper graben area is further
developing towards east. The measured displacement
compares very well with the measurements by the
total station when the GB InSAR measurements are
corrected for the discrepancy between LOS and the
displacement vector. The GB InSAR measurements
from Oaldsbygda give the most complete displace-
ment pattern, but due to strong atmospheric effects
REFERENCES
a
ntoneLLo
g., C
asagLi
n., F
arina
p., L
eva
d., n
iCo
g., s
ieber
a.J. & t
arChi
d. (2004) - Ground based SAR interferometry for
monitoring mass movements. Landslides, 1: 21-28.
b
LiKra
L. h., h
oLe
J. & a
nda
e. (2010) - Scenario og prognoser for fjellskred og flodbølger fra Åknes og Hegguraksla. Åknes
report 1.2010.
b
ozzano
F., C
ipriani
i., M
azzanti
p. & p
restininzi
a. (2011) - Displacement patterns of a landslide affected by human activities:
insights from ground-based InSAR monitoring. Nat. Hazards, 59: 1377-1396.
C
asagLi
n., C
atani
F., v
entisette
C.d. & L
uzi
g. (2010) - Advances in Geosciences InSAR processing for the recognition of
landslides. Landslides, 7: 291-301.
C
rosta
g.b., a
gLiardi
F., s
osio
r., r
ivoLta
C., L
eva
d. & d
ei
g
as
L. (2012) - Long term real-time monitoring of large alpine
rockslides by GB-InSAR: mechanisms, triggers, scenario assessment and EarlyWarning. Geophysical Research Abstracts,
14: EGU2012-11496.
F
arina
p., L
eoni
L., b
abboni
F., C
oppi
F., M
ayer
L. & r
iCCi
p. (2011) - IBIS-M, an Innovative Radar for Monitoring Slopes in
Open-Pit Mines. Proceedings, Slope Stability: International Symposium on Rock Slope Stability in Open Pit Mining and
Civil Engineering, Vancouver, Canada
F
erretti
a., M
onti
g
uarnieri
a., p
rati
C. & r
oCCa
F. (2007) - InSAR Principles: Guidelines for SAR Interferometry Processing
and Interpretation. ESA.
g
anerød
g.v., g
røneng
g., r
ønning
J.s., d
aLsegg
e., e
LvebaKK
, h., t
ønnesen
J.F., K
veLdsviK
v., e
iKen
t. & b
LiKra
L.h.
(2008) - Geological model of the Åknes rockslide, western Norway. Engineering Geology, 102: 1-18.
h
arbitz
C. & g
LiMsdaL
s. (2011) - Numerical simulations of tsunamis from potential and historical rock slides in Storfjorden:
hazard zoning and comparison with 3D laboratory experiments. Revised NGI report 20051018-00-1-R.
J
aboyedoFF
M., o
ppiKoFer
t., d
erron
M.h., b
LiKra
L.h. b
öhMe
M & s
aintot
A. (2011) - Complex landslide behavior and
structural control: a three-dimensional conceptual model of Åknes rockslide, Norway. Geological Society London. Special
Publications 2011, 351: 147-161.
K
ristensen
L. & b
LiKra
L.h. (2011) - Monitoring displacement on the Mannen rockslide in Western Norway. Proceedings of the
2
nd
World Landslide Forum - 3-7 October 2011, Rome.
K
ristensen
L. & r
ivoLta
C. (2011) - Instability mapping by Ground Based inSAR along a road, Oppdølsstranda, Norway. Poster.
Fringe 2011 Workshop, Frascati, Italy, 19-23 September.
K
ristensen
L. (2011) - Ground based radar measurements at Gamanjunni 3 and Oksfjellet, Troms. Åknes report 09.2011.
K
ristensen
L. (2012a) - Radarkartlegging av bevegelse på Jettan og Indre Nordnes, 2011. Åknes report 04.2012
K
ristensen
L. (2012b) - GB radar measurements at Joasete/Furekamben in Flåm, Sogn og Fjordane, 2012. Åknes report 06.2012.
K
ristensen
L. (2012c) - GB InSAR measurements at Storhaugen in Manndalen, Troms, 2012. Åknes report 11.2012.
L
ingua
a., p
iatti
d. & r
inaudo
F. (2008) - Remote monitoring of a landslide using an integration of GB InSAR and LIDAR
techniques. The International Archive of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XXXVll.
Part B1. Beijing 2001.
r
iedeL
b & W
aLther
a. (2008) - Monitoring, prediction, and early warning using ground-based radar interferometry. Adv.
Geosci., 14: 189-194.
s
ChuLz
W.h., C
oe
J.a., s
hurtLeFF
b.L. & p
anosKy
J. (2012) - Kinematics of the Slumgullion landslide revealed by ground-
based InSAR surveys. In: e
berhardt
et alii (
eds
.). Landslides and engineered slopes: protecting society through improved
along the fjords, it is difficult to use for early warning.
Measurements from Fjellvåken do not suffer the same
atmospheric effects. This site provides measurements
of the graben and upper flank area, which has the
largest deformations and probably the highest prob-
abilities for a catastrophic failure. This site will be
used for early warning, in connection with accelera-
tion of movements and after a major event.
background image
L. KRISTENSEN, C. RIVOLTA, J. DEHLS & L.H. BLIKRA
348
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
understanding. Taylor & Francis Group, London.
s
everin
J., e
berhardt
e., L
eoni
L. & F
ortin
s. (2011) - Use of ground-based synthetic aperture radar to investigate complex
3-D pit slope kinematics. Proceedings, Slope Stability 2011: International Symposium on Rock Slope Stability in Open Pit
Mining and Civil Engineering, Vancouver, Canada.
t
arChi
d., C
asagLi
n., L
eva
d., M
oretti
s. & s
ieber
a.J. (2003) - Monitoring landslide displacements by using ground-based
SAR interferometry: application to the Ruinon landslide in the Italian Alps. Journal Geophysical Research, 108: 2387-2401.
t
arChi
d., a
ntoneLLo
g., C
asagLi
n. F
arina
p., F
ortuny
-g
uasCh
J., g
uerri
L & L
eva
d. (2005) - On the use of roundbased
SAR interferometry for slope failure early warning: the Cortenova rock slide, Italy. In: s
assa
et alii (e
ds
.). Landslides, risk
analysis and sustainable disaster management. Springer, 337-342.
Statistics