1 Purpose
The primary purpose of space borne SAR interferometry is to measure small surface displacements as indicators for internal magmatic activity. Furthermore, surface changes occurring during an eruption can be imaged and mapped. The advantage of the method is that no ground instruments are required and several volcanoes can be imaged quasi simultaneously. The disadvantage of current systems is that the observation sampling interval is restricted to the satellite repeat cycles, e.g. 11 days for TerraSAR-X.
Space borne Synthetic Aperture Radar (SAR) is an active coherent microwave imaging system with side-looking geometry. Because microwaves are not significantly affected by clouds and volcanic ash plumes, SAR satellites provide a robust method to monitor volcanoes with a spatial resolution of 1 to 20 meters. The SAR interferometry (InSAR) method uses carrier phase differences between images acquired in the orbital repeat period (11-45 days). It’s the phase difference (interferogram) contains the information of topography, ground movements, atmospheric propagation delays and noise [1]. The deformation signal can be partially isolated from the interferogram by subtracting the topographic phase component using a digital elevation model (DEM). Such a differential interferogram is shown in Fig 1.
- Fig 1. Acquisition footprint (green rectangle) of TerraSAR-X high resolution (TSX-HS) spotlight data from Stromboli volcano (Italy) with descending orbit and 43° incidence angle.
- Fig 2. Differential interferogram from selected geometry with 11 days temporal baseline from 2008-01-28 to 2008-02-08, critical baseline 35.8 m.
1.1 Theoretical background
SAR interferometry uses the differential measurement between two SAR acquisitions from slightly different geometry. Its phase observations (interferogram) contain (1) the information of topography, ground movements, atmospheric signal and noises [1]. If we neglect the atmospheric disturbance, the deformation signal could be derived from differential phase using additional height information e.g. from shuttle radar mission DEM (SRTM).
φdefo =φ - (φtopo + φatmo+ φorbit + φnoise ) (1)
φdefo : possible displacement during the observation period; φtopo: the topographic phase; φatmo: phase delay difference due to ionospheric and atmospheric propagation conditions. Atmospheric effect has a strong influence on interferograms and must be compensated to obtain reliable deformation measurements; φorbit: flat earth contribution; φnoise: e.g. thermal noise.
An example of differential interferogram is selected from Stromboli volcano, see Fig 2. The interferogram is generated with TerraSAR-X high resolution spotlight image (TSX-HS) with descending orbit and 43° incidence angle, see Fig 1.
Unfortunately, DEMs of volcanoes are often of insufficient quality and volcanoes close to coastal areas show very strong atmospheric water vapor delay variations. For example, in the case of the Stromboli volcano it is hard to separate the phase components because of the inaccurate reference DEM and atmospheric delay [2]. In order to detect the deformation signal, a more sophisticated methods based on time series analysis is developed. It uses just the long-time coherent scatter from a stack of image and models the deformation and atmospheric signal with its phase observations, so called persistent scatterer interferometry (PSI) [3][4]. Using 16 TSX-HS data from the same geometry (Fig 1) the topographic update for the SRTM and deformation velocity map are generated with PS processing based on Kampes, 2004 [3].
- Fig 3. topographic update based on the SRTM DEM with colorbar from -50 to 50 meters.
- Fig 4. deformation velocity map with linear deformation model from colorbar from -30 to 30 mm/year.
1.2 Literature
[1] R. Bamler, and P. Hartl, Synthetic aperture radar interferometry, Inverse Problems, 14, pp1-R54, 1998
[2] R.F. Hanssen, Radar Interferometry – Data Interpretation and Error Analysis, Kluwer Academic Publishers, 2001
[3] B.M. Kampes, Radar Interferometry – Persistent Sactterer Technique, Springer, 2006
[4] A. Ferretti, C. Prati, and F. Rocca, Nonlinear subsidence rate estimation using permanent scatterers in differential SAR interferometry, IEEE Transcations on Geoscience and Remote Sensing 38(5), 2202-2212, 2000