In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Voi. XXXVIII, Part 7B 
156 
Figure 2. Example of the sampling capability of TerraSAR-X StripMap imagery. A photo of the Agbar tower, an emblematic 144 m 
height skyscraper of Barcelona, is shown in the left image. The mean SAR amplitude image over the same tower is shown in the right 
image. In this case, the mean amplitude image was obtained using 20 co-registered images. One sample every 0.75 m along the 
vertical direction can be obtained thanks to the resolution in range of these images (0.9094 m). 
3. SPATIAL SAMPLING 
As mentioned above, the major advantage of the X-band 
imagery captured by the sensors onboard the TerraSAR-X and 
COSMO-SkyMed satellites is its high spatial resolution (see 
Adam et al., 2008). Figure 1 shows an illustrative example of a 
skyscraper: a very dense sampling of vertical facades is 
provided by TerraSAR-X. However, it is worth emphasising 
that the actual PSI spatial sampling capability, and hence the 
PSI deformation measurement capability, is usually much lower 
than the resolution of the original SAR imagery. In fact, PSI 
deformation measurements can only be achieved for the points 
where PSI phases maintain good quality over time. An example 
of PSI sampling density is shown in Figure 3. It is evident that 
many of the original SAR pixels cannot be exploited to estimate 
deformation in this case (e.g. see the central part of the image 
where no measurements were obtained). However, a very dense 
set of deformation measurement points can be obtained where 
the PSI phase maintains good quality. 
4. PSI GEOCODING 
An interesting aspect of the analysis of the TerraSAR-X data 
described in this work is the accurate PSI geocoding capability, 
which is a fundamental step to correctly interpret and exploit 
the PSI results. A key step to achieve this geocoding is the 
estimation, for each PSI measured point, of the so-called 
residual topographic error (see Crosetto et al., 2010), which is 
one of the main PSI products. 
An example of PSI geocoding is shown in Figure 4, which 
shows the residual topographic error estimated over the Camp 
Nou stadium. Note that the optical image used as background 
(to show the geocoding quality) shows the typical geometric 
distortions of the standard ortho-images (i.e. it is not a so-called 
true ortho-image). Another interesting example of PSI 
geocoding is shown in Figure 5. 
5. DEFORMATION ANALYSIS 
The estimation of the deformation of the area at hand and 
during the observed period is the main outcome of the PSI 
analysis. It usually yields two main products: (i) the average 
displacement rates over the observed period, and (ii) the time 
series of the deformation. In this paper we only consider the 
first product. An example of average displacement rates is 
shown in Figure 3. The most important characteristic of this 
example is the high density of measurements achieved through 
PSI analysis. This example is complemented by another 
deformation measurement result, shown in Figure 6, not 
obtained by means of a PSI analysis. With a simple standard 
interferometric analysis based on two images, interesting 
deformation phenomena may be found. This result, achieved 
with just two images, is important, especially considering the 
high commercial price of the TerraSAR-X images. 
Even though the results shown in this paper are rather 
preliminary, they are very promising. Additional work is needed 
to make a comprehensive assessment of PSI TerraSAR-X: 
defining the key achievements in terms of new applications, 
their advantages and limitations, their monitoring performances 
with respect to costs, etc.