|
<< Click to Display Table of Contents >> Interferometric Stacking Module - Overview |
![]() ![]()
|
A Note on the Interferometric Stacking module
This module is intended for the generation of Displacement Maps and Digital Elevation Models (DEMs) from multitemporal interferometric SAR data series.
Two different processing approaches are foreseen:
1.PS (Persistent Scatterers) - It is intended for the analysis of point targets. The resulting product is relevant to the measurements of linear displacements and the derivation of precise heights of local scatterers, which are typically characterized by high coherence. The number of input images is crucial for the pixel coherence estimate, which determines the identification of suitable PSs. The use of an insufficient number of acquisitions will produce a high coherence estimate throughout the entire scene, which would result in a PS number overestimation. The application of this technique can be considered reliable when 20 or more acquisitions, which are characterized by a regular temporal separation, are available. This approach should exclusively be used in urban areas, or in general, where scatterers remain stable in radiometric and interferometric phase terms. Depending upon the scatterer stability (time coherence), the displacement measurement accuracy can reach the precision of millimeters, while the maximum detectable displacement velocity depends on both the minimum time distance between consecutive acquisitions and the SAR wavelength. Finally concerning the height estimates, this technique provides a better accuracy than the SBAS approach; the difference between the two methodologies is particularly evident in layover areas (e.g. skyscrapers in urban zones) where the SBAS, due to both the interferogram filtering and the phase unwrapping processes (none of the two is performed in the PS processing), tends to smooth the elevations.
1.1 E-PS (Enhanced-PS) - The E-PS technique identifies Distributed Scatterers and Permanent Scatterers pixels allowing to enrich the InSAR stacking analysis with a wider and denser pixel coverage. The application of an adaptive filter allows to preserve the level of detail that distinguish the PS technique.
2.SBAS (Small Baseline Subset) - It is intended for the analysis of distributed targets. The resulting products resemble those coming from a conventional DInSAR processing; the key difference is that SBAS enables the analysis of large time-series, while the classical DInSAR is limited to the 2-, 3- and 4-acquisitions (refer also to the Dual Pair Differential Interferometry). With respect to the PS, the SBAS technique is less sensitive to the number of acquisitions; this is because the SBAS exploits the spatially distributed coherence, instead of estimating the coherence exclusively on local scatterers (PS characteristic). It remains anyhow that, also with the SBAS technique, the availability of more acquisitions allow to achieve a better product quality; in this case the improvement is mostly related to the better estimate (and removal) of the atmospheric phase component. Concerning the displacement assessment, while the PS is limited to linear models, the SBAS can cope with linear, quadratic and cubic models (i.e. when the displacement velocity and/or acceleration change over the time). Moreover the SBAS technique can also be exploited only for the terrain elevation estimate and, in such case, the “no model” option is adopted. In terms of maximum measurable displacement, there are not relationships with the temporal distance between consecutive acquisitions, whilst there are limitations with respect to the displacement spatial variability; this is due to the phase unwrapping intrinsic constraints. We can finally state that, in several cases, the SBAS approach is more robust than the PS, as the former takes advantage of the higher redundancy (i.e. number of connections of each acquisition), which eventually allows to generate many more interferograms.
2.1 E-SBAS (Enhanced-SBAS) - The E-SBAS technique identifies Permanent Scatterers and Distribuited Scatterers pixels allowing to enrich the InSAR stacking analysis with a wider and denser pixel coverage. The E-SBAS processing allows to retrive the Non-linear motion on both PS and DS pixel time series.
The following basic requirements have to be fulfilled in the input data series:
-All data must be acquired by the same sensor.
-All data must be acquired with the same viewing geometry.
-In case of multi-polarization acquisitions, the same polarization must be selected for all data. It must be noted that it is possible to enter, in the same input temporal series, both single polarization and multi-polarization acquisitions; for instance, in case of ALOS PALSAR data, we can make a series using the HH channels of Fine Beam Single (FBS) and the HH channels of Fine Beam Dual (FBD). When the SBAS processing is performed, we suggest to select an FBD acquisition as Super Reference file.
Note that:
–SAR data must be imported (see Basic module).
–In case of SAR RAW products, the data must be imported and focussed (refer to Focusing module).
–Default setting for selected parameters can be specified in the Preferences panel.
–Data geocoded to GEO-GLOBAL cartographic reference system can be automatically displayed into the Google Earth environment by double clicking on the output .kml file.
–Co-ordinate decimal values must be entered using the dot (e.g. 29.30) and not the comma (e.g. 29,30) character.
Support material
Please consult the tutorial material dedicated to the Interferometric Stacking module, available on the sarmap website, for more details on the usage and applications of all the stacking tecniques included in this module.
References
P. Pasquali, P. Riccardi, A. Cantone, M. Defilippi, F. Ogushi & S. Gagliano, “Quantitative comparison of methods and sensors for monitoring land subsidence phenomena based on satellite SAR interferometric stacking”, Proceedings of GRSG Annual General Meeting 201, Including the 2011 Oil and Gas industry workshop, 7th –9th December 2011, ESA ESRIN, Frascati, Italy