|
|
|
Final
results - Barcelona test site |
 |
|
 |
| |
In order to validate the ERS –
ASAR phase continuity in the Barcelona area the
subsidence trend between both sensors during common
periods with and without mixing the ERS and ASAR
data in the processing was compared.
|
|
| |
The main idea was to compare the movements estimated
by the two sensors, and especially the absolute
displacement along the satellites line of sight.
As can be seen in Figure 6 one may notice that
most of the metropolitan area of Barcelona is
not subjected to subsidence. On the other hand,
there is a major subsidence area in the
South part of the city, in the delta
of the Llobregat River, which covers about 50
Km2. From the geological viewpoint this is a very
young alluvial terrain (less then 2000 years),
which undergoes compaction phenomena.
Figure 6: Geocoded mean subsidence
velocity
from April 1995 to December 2000.
................................
The ASAR results of the Barcelona
test side have been obtained by the processing
of 5 Envisat ASAR descending images, which cover
a period of about one year, from April 2003 to
March 2004. The mean subsidence velocity fields
estimated by using the ERS series and the short
ASAR series are shown in Figure 7.
One may observe that the two maps show a very
similar subsidence pattern. The results are also
quite similar in terms of estimated mean velocity
of the subsidence, with maximum values of about
1 cm/yr in the delta of the Llobregat river.
Figure 7: Geocoded mean subsidence
velocity fields estimated with the ERS1 and 2
series (left), and geocoded mean subsidence velocity
estimated from a short series of five ASAR images.
The observed period for the ERS dataset spans
from April 1995 to December 2000, while it covers
about one year for the ASAR data, from April 2003
to March 2004.
................................
The basic idea for the validation
was to compare the high resolution maps of subsidence
rates detected by the SPN software using the dataset
ERS alone, and those derived by using the ASAR
dataset alone. In Figure 8 a zoomed area is illustrated.
It can be noted that the detected rates are in
very good agreement. It can be also seen that
there are few false alarms in the ASAR dataset
(only 5 images).
These good results are probably due to the type
of reflection that is taking place over
the top of the considered industrial buildings:
their uniform and metallic structures keep the
phase continuity between repeat passes. This makes
this area an interesting site where the phase
continuity between the two sensors can be validated.
Figure 8: Comparison of the ERS
detected subsidence vs. the ASAR results on the
delta of the Llobregat river. The subsidence velocity
values are expressed in cm/yr. The dark dots correspond
to subsidence velocity values between –1.3
and –1.1 cm/yr, which are not included in
the color bar.
................................
Other two interesting areas were firefly
analysed:
• the subsidence
of the Bon Pastor area.
• the subsidence
of the main dike of the Port of Barcelona:
the Trencaones dike.
The Bon Pastor area is placed at the north of
the city, close to the Besós River. Thanks
to the precise geocoding of the points over
1:5000 topographic maps of the city at
very high resolution (0.8 meters/pixel), it was
easy to identify in situ the risky buildings.
Some big cracks in the external walls were observed
on buildings affected by high subsidence rates
(close to 1 cm/year). Some pictures provided by
IG show these cracks of few cm in some cases.
Figure 9: Zoom of the Bon Pastor
subsidence rate area over the geocoded radar mean
amplitude (25 m/pixel) and geocoded final good
points over a topographic map of the city 1/5000
(0.8 m/pixel) in order to perfectly identify the
risky structures in situ. Results in cm/years.
................................
In figure 10 are illustrated
the results of the in situ validation. The pictures
taken from the point B show a very big crack in
the middle of the external wall along the street.
Looking at the topographic map of the figure
9 it can be noted that the red points
are placed exactly over the crack. In some parts
of the crack there are few centimetres of separation
, which is correspond exactly with measured rates
of about 0.8 to 1.0 cm/years.
Figure 10: Cracks at the external
walls perfectly visible over the risky buildings
detected by the high subsidence rate
................................
The Port of Barcelona represents one of
the most important infrastructures of Catalonia.
The Port is partially located on the delta of
the river Llobregat. Some of its infrastructures
are known to be subjected to subsidence phenomena.
Some parts of the Port are difficult to analyse
by DInSAR because either they are continuously
changing (e.g. the deposits of the containers)
or their particular geometry. The validation has
been focused on a particular structure of the
Port, where geodetic measures taken by the Topographic
Service of the Port are available. This structure
is the so-called Trencaones, the main dike of
the Port which protects all the other infrastructures.
The location of the Trencaones dike is indicated
in Figure 11.
Figure 11: Location of the Trencaones
dike
over the geocoded mean amplitude radar.
................................
On the Trencaones dike four points have been measured
using the ERS and ENVISAT SAR time series.
Their location is indicated by a yellow dot in
Figure 12.
One may notice that these points are quite close
to the last of the reference points, which are
indicated by green crosses and blue dots in the
same image.
Figure 12:
Location of the reference points
(blue and green)
and of the SPN measured points
(yellow)
................................
The subsidence values estimated by the
four SPN points have been compared with the value
given by the closest reference points.
In Figure 13 the subsidence profiles of the four
points are plot together with the reference measure.
Two of these points are characterized by a very
high coherence, one may notice that the SPN estimates
and the reference value are in very good agreement.
Figure 13: Comparison between
the ERS and ASAR
subsidence profiles and the reference measurement.
................................
PRECISE GEOCODING
The geocoding procedure was carried out using
the DEM error obtained by means of the SPN software
without any prior information. In order to achieve
an accurate geocoding, the final point position
was corrected by taking into account the detected
vertical height of every point in the radar geometry.
Thus, the obtained final results was superimposed
on a topographic map of the city at scale 1:5000.
The final position of the points was analysed
using aerial ortophotos (at 0.5m/pixel)
in order to check if the final point
reflection is placed at the correct place. Therefore,
the validation of the precise point geocoding
will assess implicitly the DEM error detected
by the SPN, i.e.: the difference between the DEM
used to substract the interferometric topography
and the actual height where the reflection of
the radar signal took place.
In figure 14 is showed a zoom
in the Bon Pastor area with a geocoded SAR points
superposed to a 1:5000 map, and an aerial orthoimage
of the same area. The focused area is corresponded
with a little bridge used for people to cross
the Besos River. The bridge has no more than 3
meters of width. It can be seen how the radar
points are perfectly aligned over the bridge.
Looking at Figure 15 it can stated that the point
geocoding errors on ground are approximately of
1.6 m, less then one third of the pixel footprint.
Figure 14: Geocoded topographic
map with the radar points at 0.8m/pixel compared
with an aerial ortophoto at 0.5m/pixel of the
same area.
Figure 15: Geocoded topographic
map with the radar points at 0.8m/pixel (upper
left) compared with an aerial ortofoto at 0.5m/pixel
(upper right) of the same area. The lower image
is a zoom of the upper left previous image.
................................
A study of the sensibility of the SPN measures
has been conducted on the Palau Sant Jordi,
analysing the deformation signal related to the
thermal dilation. The study was focused on the
top of that building because it presents a lot
of metallic parts as shown in Figure 16.
A point profiles is showed in figure 17
jointly with the scaled version of the measured
temperature. It shows a similar trend in terms
of thermal dilation as a function of the temperature.
The correlation coefficient is higher than 0.6.
Figure16: Picture of the Palau
Sant Jordi (left)
and geocoded image (right) with the points analysed.
Figure 17: Palau Sant Jordi,
point 3. Temporal evolution of the point deformation
and profiles of the mean and maximum temperatures
of the days of acquisition of the ERS SAR images.
The temperature values have been multiplied by
1/100. One may notice the seasonal behaviour of
the deformations, probably due to thermal dilation.
The correlation coefficient between temperatures
and deformation equals 0.62.
|
|
|
| |
|
|
|
