Lecture 7: InSAR
-
Upload
truonghanh -
Category
Documents
-
view
246 -
download
2
Transcript of Lecture 7: InSAR
![Page 1: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/1.jpg)
Lecture 7: InSAR
GEOS 655Tectonic GeodesyJeff Freymueller
Thanks to Ramón Hanssen, Mark Simons, Dörte Mann, and Howard Zebker for slides and images
![Page 2: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/2.jpg)
The Picture That Started the Excitement
Landers earthquake, first shown by Massennet et al. (1993)
![Page 3: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/3.jpg)
Early ground based radar
RAdio Detection and Ranging
![Page 4: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/4.jpg)
Radio waves, active sensor
Wavelength range is cm-m
24 cm 3 cm 5 cm
![Page 5: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/5.jpg)
Penetration (weather independent)
C b
and
Radar waves penetrate the atmosphere and clouds
![Page 6: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/6.jpg)
InSAR Platforms
Both from: JPL From: H. Zebker
Satellites: Repeat pass Fly over once, repeat days-years later * Measures deformation and topography
Space shuttle: Shuttle Radar Topography Mission (SRTM)
Aircraft: Shown here: AIRSAR Measures topography, ocean currents
![Page 7: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/7.jpg)
ERS-1, ERS-2, Envisat: C Band Dt > 35 days Right looking
JERS-1 (and ALOS): L Band Dt > 45 days Right looking
Radarsat 1 (and 2): C Band DT > 23 days Right looking
Early SAR/InSAR Missions
![Page 8: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/8.jpg)
SAR Satellite Timeline
www.unavco.org
![Page 9: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/9.jpg)
• Landsat optical• Shuttle L-band radar• What do we see?
Sahara desert
Radar penetrates material with a low dielectric constant (dep. on wavelength)
Here about 3 m.
Sahara, NW Sudan (SIR-A)
Physics
![Page 10: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/10.jpg)
Physics - Scattering
• Scattering is dominated by wavelength-scale structures
• Wavelength shorter: image brighter• Specular and Bragg scattering• Speckle
![Page 11: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/11.jpg)
Backscatter
![Page 12: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/12.jpg)
SPECULAR
Radar signal return depends on:
• Slope
• Roughness
• Dielectric constant
Scattering
Smooth Rough
![Page 13: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/13.jpg)
![Page 14: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/14.jpg)
Physics – scattering phase
Imag
inar
y
Real
Radar pixel phase is superposition of near-random scattering elements:
Unpredictable!
![Page 15: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/15.jpg)
Scattering MechanismsSynthetic Aperture Radar – Systems and Signal Processing
![Page 16: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/16.jpg)
Rule of Thumb in SAR images
Synthetic Aperture Radar – Systems and Signal Processing
• Backscattering Coefficient
• Smooth – Black • Rough surface – white
• Calm water surface – black • Water in windy day – white
• Hills and other large-scale surface variations tend to appear bright on one side and dim on the other.
• Human-made objects - bright spots (corner reflector) • Strong corner reflector- Bright spotty cross (strong sidelobes)
![Page 17: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/17.jpg)
Geometry
Terminology – Foreshortening, layover, shadow– Why side-looking?– Incidence angle, – Coordinates range, azimuth
![Page 18: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/18.jpg)
Geometry
Ground range
![Page 19: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/19.jpg)
Geometry
JERS-1 data (M.Shimada)
![Page 20: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/20.jpg)
Real Aperture vs. Synthetic Aperture
Synthetic Aperture Radar – Systems and Signal Processing
• Real Aperture : resolution ~ Rλ/L
• Synthetic Aperture: resolution ~ L/2
Irrespective of R Smaller, better?! - Carl Wiley (1951)
![Page 21: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/21.jpg)
Resolution I: RAR• Real Aperture Radar
• Resolution dependent on antenna dimension/pulse length
• Beam width (half power width) is ratio of wavelength over antenna size:
![Page 22: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/22.jpg)
Antenna dimensions
1.3 m
10 m
Calculate Ground Resolution
C-band λ= ~0.05 m
D=10 m antenna
Beam angle = 5.10-2/10 = 5.10-3 rad
R=850 km
times 8.5.105 = 42.102[m] = 4.2 km
Dλθ =
![Page 23: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/23.jpg)
![Page 24: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/24.jpg)
antenna
pulse
![Page 25: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/25.jpg)
antenna
pulse length
swath
ground range
![Page 26: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/26.jpg)
Improvement in Resolution �(Crimea, Ukraine)
Real Aperture Radar
5x14 km pixels
Massonnet and Feigl, 1998
![Page 27: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/27.jpg)
Improvement of along-track resolution
Synthetic antenna
Physical antenna
Resolution cell
![Page 28: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/28.jpg)
Improvement in Resolution �(Crimea, Ukraine)
Real Aperture Radar Synthetic Aperture Radar
5x14 km pixels 4x20 m pixels
Massonnet and Feigl, 1998
![Page 29: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/29.jpg)
Radar Interferometry
![Page 30: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/30.jpg)
![Page 31: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/31.jpg)
![Page 32: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/32.jpg)
![Page 33: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/33.jpg)
Example in 2D: interferogram
![Page 34: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/34.jpg)
SAR Interferometry (InSAR)
Amplitude 1
Phase 1 Phase 2
Amplitude 2 =
Interferometric Phase
Average Amplitude
![Page 35: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/35.jpg)
Basics: Interferogram Formation Basics: Interferogram Formation I
![Page 36: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/36.jpg)
Effect of SAR Frequency
L band: longer λ, fewer fringes, better coherence
![Page 37: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/37.jpg)
Why is Coherence Better at L-band?• Coherence is a measure of how well the phase of adjacent pixels agree
with each other– Random phase = low or zero coherence
• C-band radars usually have poor coherence in vegetated areas.– Leaves on trees tend to have size ~ wavelength
• Therefore are effective scatterers• Leaves move (wind, growth) and change (fall off)
• L-band scattering is dominated by features 20-25 cm in scale– More likely to be ground features or unchanging features of vegetation
(large branches)• Some other notes on coherence
– Plowed fields have bad coherence– Permanent structures (buildings, roads) usually have good coherence.– Surprisingly, certain kinds of swamps seem to have good coherence with returns
that depend on the water level.
![Page 38: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/38.jpg)
RangeExpressed as phase (radians)
![Page 39: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/39.jpg)
Range
Expressed as integer cycles + fractional phase
![Page 40: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/40.jpg)
Phase-range relationship
€
2R1λ
= −ϕ12π
![Page 41: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/41.jpg)
Phase-height relationship(Far-field approximation)
Ellipsoid
Topographic phase is (inversely) scaled by the perpendicular baseline!
![Page 42: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/42.jpg)
Height ambiguity
Height difference related to 1 phase cycle:
0
100
200
300
400
500
20 100
180
260
340
420
500
580
660
740
820
900
Perpendicular baseline [m]
Hei
ght a
mbi
guity
[m]
![Page 43: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/43.jpg)
![Page 44: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/44.jpg)
Baseline dependency, height ambiguity
Bperp 173 m, Bt= 1day Bperp 531 m, Bt= 1 day
H2pi=45m H2pi=16m
![Page 45: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/45.jpg)
Phase-deformation relationship
Subsidence ∆z
1 cycle LOS deformation is equal to half the physical wavelength
![Page 46: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/46.jpg)
Interferometric phase - deformation
Relative uplift of the center (9 cm)
Distance satellite-ground has become shorter towards center
Interferometric phase decreases towards center
![Page 47: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/47.jpg)
Topography and deformation
Ellipsoid
Sensitivity to deformation 1000x higher than for topography
![Page 48: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/48.jpg)
To satellite (LOS vector)
E
N
U
U Us
One interferogram (igram) only provides one component (LOS) of the displacement field - pay attention to LOS vector.
![Page 49: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/49.jpg)
Atmospheric disturbance• Spatially varying
disturbance signal• Can be ~5 cm over 20
km• Spatially correlated but
temporally uncorrelated (Δt>1 day)
• Introduces covariances in stochastic model
![Page 50: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/50.jpg)
Example Interferometric
Radar Meteorology
![Page 51: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/51.jpg)
Geometric decorrelation• Baselines vary• Relative scattering
mechanisms change
• Images become uncomparable
• Function of baseline, Doppler centroid, and terrain slope
Note the trade-off between height sensitivity (large baseline) and noise reduction (small baseline)!
![Page 52: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/52.jpg)
Temporal decorrelation
Temporal baseline
1 day 1 year 2 years 3 years 6 years
Perpendicular baseline (m)
29 112 93 185 166
![Page 53: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/53.jpg)
Envisat interferograms (single master)
![Page 54: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/54.jpg)
Volcanoes
![Page 55: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/55.jpg)
Mapping Out Large Swaths
• From Pritchard and Simons (2002)
• Found inflation at 4 volcanic centers in Andes
• None were really expected
![Page 56: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/56.jpg)
Volcanic source models
Point inflation source (Mogi, 1958)
Pressure and volume change
d
r
![Page 57: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/57.jpg)
Anchorage
Okmok Volcano, Alaska-Aleutian arc
Okmok
![Page 58: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/58.jpg)
Deformation interferograms1993 - 1995
1995 - 1997 1997 - 1998
1992 - 1993
![Page 59: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/59.jpg)
Source locations and depths
Inflation Deflation
2 km
![Page 60: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/60.jpg)
Time dependence of deformation
![Page 61: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/61.jpg)
• Eruption volume in 1997: 70 x 106 m3 • Average magma accumulation rate during inflation periods: 3-6 x 106 m3/yr Estimated recurrence time : 15 years • Eruptive history:
Observed magma accumulation rate is typical for the long term, suggesting continuous supply from a deeper source
Magma Volume and Eruption Frequency
1938 1946 1958 1997 1983-88
Mann et al. (2002)
![Page 62: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/62.jpg)
1992 Mw 7.3 Landers 1999 Mw 7.1 Hector Mine
![Page 63: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/63.jpg)
Descending orbit Ascending orbit
1999 Mw 7.1 Hector Mine EQ
![Page 64: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/64.jpg)
pixel tracking
1999, Mw 7.1 Hector Mine, CA Earthquake
Tracking features in imagery: How much did a boulder/cactus move?
Ø Find shift (offset) that maximizes cross-correlation of small ensembles of pixels in two (before/after) images
Ø With radar data, along track component of offsets is perpendicular to LOS phase
Ø Does not need to be phase unwrapped
Ø Sensitivity much less than InSAR
Ø Can use optical data too (satellite and airphotos)
Ø Other applications (glaciers, …)
![Page 65: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/65.jpg)
Ø Significant vertical fault slip
Ø Fault slip concentrated at shallow depth (7 to 10 km)
Ø Use as input into post-seismic models Inferred subsurface coseismic fault slip
1999 Mw 7.1 Hector Mine EQ
3D displacement field
![Page 66: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/66.jpg)
Flow of Rutford Ice Stream
D. Goldstein, JPL
6 days of displacement, each fringe ~28 mm LOS
![Page 67: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/67.jpg)
Patagonia Ice Velocities �from Shuttle Imaging Radar (SIR-C)
Isacks et al. (1997)http://www.geo.cornell.edu/geology/SIRC_Pat/patagonia.html
![Page 68: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/68.jpg)
InSAR Glacier Velocity
![Page 69: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/69.jpg)
DInSAR �Land Subsidence
![Page 70: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/70.jpg)
PS principle• Pixels with strong and consistent reflections
in time.• Multi-pass InSAR – time series necessary.• Estimate atmospheric signal:
– Spatially, not temporally correlated.– Independent of baseline. (topography is)
![Page 71: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/71.jpg)
Persistent Scatterers processing
Preprocessing: Selection test areas
Selection potential PS points
![Page 72: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/72.jpg)
Persistent Scatterers processing chain
• Persistent Scatterer Candidates selection, based on amplitude dispersion (Ferretti et al., 2001)
• Construction network by Delaunay triangulation
2 km
10 km
• Integer LSQ estimation of ambiguities and parameters • Testing of residuals • Spatial unwrapping (path integration) • Separation of atmosphere and residual deformation by filtering and Kriging • Removal Atmospheric Phase Screen • Selection of Persistent Scatterers
![Page 73: Lecture 7: InSAR](https://reader033.fdocuments.net/reader033/viewer/2022051104/58a2d3181a28abf8698b6760/html5/thumbnails/73.jpg)
PS Result for L.A. (U. Milano)