26
Geotechnical News • June 2012
GEOTECHNICAL INSTRUMENTATION NEWS
Remote monitoring of deformation using
Satellite SAR Interferometry
Francesca Bozzano and Alfredo Rocca
Principle of operation
Satellite SAR (Synthetic Aperture
Radar) Interferometry (SInSAR) is a
technique able to produce displace-
ment maps of the ground surface both
night and day and in the presence of
clouds by using microwave signals.
Taking advantage of the orbit of the
satellite, the SAR sensor mounted on
it can capture an image of an area,
when it passes over it. The phase
value, contained in every pixel of the
image, is correlated to the sensor-tar-
get distance. Thus, given two or more
images acquired at different times,
information about the displacement
occurred in a pixel in the time interval
between the acquisitions, is achieved
by computing the corresponding phase
difference.
Main fields of application Classical
Differential Interferometry (DInSAR)
approach (using only pairs of SAR
images) has been already used suc-
cessfully in the past, in particular to
investigate regional displacements
phenomena (e.g. earthquakes). Today,
Advanced DInSAR (A-DInSAR)
techniques, for instance Persistent
Scatterers Interferometry (PSI) and
Small Baseline Subset (SBAS), which
make use of multitemporal SAR data
and displacement models, are most
common approaches.
Main fields of application are related
to monitoring of buildings, structures
and land affected by landslides, sub-
sidence and any other process which
leads to a displacement of the ground
surface, as long as not too fast.
Accuracy and pixel resolution
SInSAR spatial resolution depends on
sensor characteristics. For most com-
mon monitoring uses, pixel size spans
from 25 m (e.g. ERS1/2 and Envisat
satellites) to 1 m (COSMO Sky-Med,
TerraSAR X, Radarsat satellites).
DInSAR accuracy is in the order
of centimetres, while A-DInSAR
methods are able to achieve accuracy
of few millimetres, from 1 to 5, for a
single displacement value, depending
on the used techniques. The accuracy
of trend displacement average velocity
for the whole analysed period, is from
0.1 to 1 mm/yr.
Main advantages
Main advantage of SInSAR is the
possibility to obtain measurements
of displacements occurred in the
past starting from 1992 (ERS1). This
great result can be achieved using
archived data acquired by the Space
Agencies during past decades. In this
case a frequency acquisition with a
maximum of generally one image per
month has to be considered. Further-
more, SInSAR monitoring can be
continued in the future, if a new data
capture campaign is planned. In this
case, thanks to shorter satellite revisit
time, more images will be available
for shorter time.
Other advantages are: SInSAR data
cover wide areas (a single frame has
tens of km on each side); Modern
A-DInSAR methods allows displace-
ment information spatially widespread
over the area of interest; There is no
need to install anything on the area
under study (although some corner
reflectors can be useful sometimes).
Main limitations
Main technical limitations are caused
by the geometrical configuration, thus
the sensor can observe movements
only along the Line Of Sight (LOS).
As consequences, image distortions
caused by steep topography and dif-
ficulty to observe displacements along
N-S direction have to be considered.
Moreover, the so-called “phase
ambiguity” effect (i.e. the inability to
recognize too fast displacements) as a
function of the signal wavelength and
the satellite revisiting time is a typical
SInSAR limitation.
Other limitations in terms of feasibil-
ity are: the difficulty to investigate
vegetated areas and the cost of SAR
data in particular for new acquisitions
by new sensors.
Future challenges
Data cost reduction would be desir-
able in order to allow A-DInSAR to
be used more frequently as a tool for
monitoring. Another interesting chal-
lenge for the future is the development
of models of displacement better able
to detect non-linear trends.
Commercial sources in North
America
In the authors knowledge, the follow-
ing companies provide this service in
North America: TRE Canada (www.
treuropa.com) and Altamira (www.
altamira-information.com). Further
companies such as FUGRO (www.
fugro-npa.com) and Egeos (www.
eurimage.com) can be found in
Europe.
Francesca Bozzano, Alfredo Rocca
Earth Science Department
“Sapienza” Università di Roma, p.le
Aldo Moro, 5, Rome, Italy.
E:
