CSRC InSAR Introduction

		CSRC \| SOPAC \| Index \| Mail \| Forums \| General \| Reports \| Contacts

CSRC InSAR Introduction

The following article is a term paper from a remote sensing class held in 1998 at UCSD.

Synthetic aperture radar interferometry for crustal deformation studies.
Karen Watson
Interferometric synthetic aperture radar (InSAR) has successfully been used for a range of crustal deformation studies. These applications include determination of co- and post-seismic slip (e.g. Massonnet et al, 1993 and Hernandez et al, 1997), volcanic deformation (e.g. Massonnet et al, 1995 and Lu et al, 1997) and deformation due to rifting (e.g. Vadon and Sigmundsson, 1997 and Sigmundsson et al 1997). Other applications of InSAR include studies of ice deformation and flow (e.g. Joughin et al, 1996 and Li et al, 1997). The key to InSAR is the extraction of the phase change between two images of the same region taken at times spanning the deformation period. This phase change is a measure of the line-of-sight deformation between the two images (plus orbital effects, topographic effects and atmospheric effects).

Synthetic aperture radar (SAR) is a side-looking radar instrument which emits microwave pulses and measures the amplitude and phase of signals reflected (returned) from the surface of the Earth. The system utilises the Doppler effect for target positioning. That is, it uses the effect whereby targets perpendicular to the satellite flight path at some instant in time have return signal frequencies equal to the Doppler centre frequency (fDc) and targets fore (aft) of perpendicular have greater (lesser) frequencies, respectively.

Figure 1 - Interferometric imaging geometry. (After Zebker et al, 1997, p.7549)

For InSAR, the imaging positions of the satellite or satellites (A1 for the master or initial pass and A2 for the slave or secondary pass; Figure 1) are separated by a baseline B. The satellite-target range ρ of a target in the master image will change by some amount

for the slave; the induced phase change

(for two-way travel), assuming that the target backscattering phase for the two frames is essentially equal, will be

(1)

(Zebker et al, 1994). This phase change consists of four components; the topographic, orbital and atmospheric components must be removed (or at least minimised) before the component due to deformation can be determined.

There are two methods of subtracting the topographic signal; the first is to use a digital elevation model (DEM; e.g. Massonnet et al, 1993), while the other is to utilise a second interferometric pair (spanning a period of no deformation; e.g. Rosen et al, 1996) which therefore requires at least one extra image. The DEM must be coregistered with the radar images and so relies on an ability to discern features in the images. This method also relies on the availability of a suitably accurate DEM for the region.

The resultant interferogram, however, is less sensitive to topography z

(2)

(where

is the radar wavelength,

is the look angle from nadir and

is the baseline orientation angle) than to deformation

(3)

since, at least for satellite observations,

>> B and therefore

(4)

(Zebker et al, 1994). The sensitivity of an interferometric pair to topography is therefore measured by the ambiguity height ha which is the topographic height required to introduce one cycle (2

radians) of phase difference

(5)

(Zebker et al, 1997).

The orbital effects are minimised by setting the phase at the outer edges (where deformation is assumed to be negligible) to zero. This can, however, be an inaccurate assumption (Figure 2; note that the southern edge of the interferogram lies in a region with in excess of 100mm deformation (this figure also highlights the greater spatial resolution of InSAR techniques over ground-based techniques such as GPS)), with the result that a fraction of the deformation signal is removed along with the orbital effect. In the future, other data (e.g. continuous GPS data (Bock and Williams, 1997)) could be used to predetermine deformation at selected sites which may then be used to set non-zero deformation at the edges of the interferogram.

InSAR is an active microwave system (that is, it provides its own illumination source), and as such is capable of imaging the Earth under any weather conditions. These images can, however, contain spurious signals due to atmospheric perturbations. Using extra interferometric pairs the atmospheric effects may be identified; transient patterns which can be traced to a single image are probably due to the atmosphere (Massonnet and Feigl, 1995). If such a pair exists outside the period of deformation, the topography and atmospheric effects could possibly be subtracted together.

InSAR relies on correlation between the images making up the pair. That is, the imaged surface must not be overly deformed at the small wavelength scale. This is a disadvantage of InSAR - some regions which would benefit from remote sensing methods of deformation determination (e.g. hazardous volcanic regions) may be covered by materials which do not retain correlation on long enough time scales e.g. snow, ice, volcanic ash, or even vegetation. Seasonal observations might be possible, for example during the summer when the snow/ice cover has melted (e.g. Lu et al, 1997).

The use of InSAR alleviates the need for monumentation required by conventional surveying techniques (and thus is not susceptible to errors at the local scale such as random walk of the monument or instrumental positioning errors). As a result there is no requirement for extensive reconnaissance surveys to determine site suitability (factors such as access, stability of ground surface, available power supply etc.).

Figure 2 - Comparison of interferogram, GPS-derived displacements and predictions of a dislocation model for the 1992 Landers earthquake. (After Bock and Williams, 1997, p.293)

In conclusion, InSAR is capable of providing a spatially dense series of measurements of line-of-sight crustal deformation. The various advantages and disadvantages of the system are summarised below (Table 1).

Advantages	Disadvantages
Greater spatial measurement density*	Lower accuracy*
Does not require ground control	Lower temporal measurement density*
All-weather system	Decorrelation of images
-	Orbital, atmospheric and topographic errors
-	Measures line-of-sight displacement*

Table 1 - Advantages and disadvantages of InSAR. (*Compared to e.g. continuous
GPS measurements. See Bock and Williams, 1997)
References
Bock, Y. and S. Williams (1997). Integrated Satellite Interferometry in Southern California. Eos, Trans., AGU. 78(29): 293,299-300.
Hernandez, B., F. Cotton, M. Campillo and D. Massonnet (1997). A comparison between short term (co-seismic) and long term (one year) slip for the Landers earthquake: Measurements from strong motion and SAR interferometry. Geoph. Res. Lett. 24(13): 1579-1582.
Joughin, I., R. Kwok and M. Fahnestock (1996). Estimation of ice-sheet motion using satellite radar interferometry: method and error analysis with application to Humboldt Glacier, Greenland. J. Glaciol. 42(142): 564-575.
Li, S., C. Benson, L. Shapiro and K. Dean (1997). Aufeis in the Ivishak river, Alaska, Mapped from Satellite Radar Interferometry. Remote Sens. Environ. 60: 131-139.
Lu, Z., R. Fatland, M. Wyss, S. Li, J. Eichelberger, K. Dean and J. Freymueller (1997). Deformation of New Trident volcano measured by ERS-1 SAR interferometry, Katmai National Park, Alaska. Geoph. Res. Lett. 24(6): 695-698.
Massonnet, D. and K. L. Feigl (1995). Discrimination of geophysical phenomena in satellite radar interferograms. Geoph. Res. Lett. 22(12): 1537-1540.
Massonnet, D., M. Rossi, C. Carmona, F. Adragna, G. Peltzer, K. Feigl and T. Rabaute (1993). The displacement field of the Landers earthquake mapped by radar interferometry. Nature. 364: 138-142.
Massonnet, D., P. Briole and A. Arnaud (1995). Deflation of Mount Etna monitored by spaceborne radar interferometry. Nature. 375: 567-570.
Rosen, P. A., S. Hensley, H. A. Zebker, F. H. Webb and E. J. Fielding (1996). Surface deformation and coherence measurements of Kilauea Volcano, Hawaii, from SIR-C radar interferometry. J. Geoph. Res. 101(E10): 23,109-23,125.
Sigmundsson, F., H. Vadon and D. Massonnet (1997). Readjustment of the Krafla spreading segment to crustal rifting measured by Satellite Radar Interferometry. Geoph. Res. Lett. 24(15): 1843-1846.
Vadon, H. and F. Sigmundsson (1997). Crustal Deformation from 1992 to 1995 at the Mid-Atlantic Ridge, Southwest Iceland, Mapped by Satellite Radar Interferometry. Science. 275: 193-197.
Zebker, H. A. , P. A. Rosen, R. M. Goldstein, A. Gabriel and C. L. Werner (1994). On the derivation of coseismic displacement fields using differential radar interferometry: The Landers earthquake. J. Geoph. Res. 99(B10): 19,617-19,634.
Zebker, H. A., P. A. Rosen and S. Hensley (1997). Atmospheric effects in interferometric synthetic aperture radar surface deformation and topographic maps. J. Geoph. Res. 102(B4): 7547-7563.