Theory and application of Synthetic Aperture Radar

A lecture on theory and application of Synthetic Aperture Radar given at Banda Informasi Geospasial, Bogor, Indonesia.

The resolution of figures is reduced for reduced file size. If you want an original file, please send an email to yaoki_at_eri.u-tokyo.ac.jp.

Transcript

1. Theory and applications of Synthetic Aperture Radar Yosuke Aoki Earthquake Research Institute, The University of Tokyo Email: yaoki@eri.u-tokyo.ac.jp 29 November 2018 Banda Informasi Geospasial Bogor, Indonesia
2. Image of an earthquake Nature Cover page of the 8 July 1993 issue (Vol. 364, No. 6433) Massonnet et al. (Nature, 1993) ü Coseismic deformation of the 1992 Landers (California, USA; Mw=7.3) earthquake measured by Synthetic Aperture Radar (SAR). ü Amazing spatial resolution (~3-5 meters) ü No need for a ground-based instruments ü Available day and night. All weather. Compare with optical measurements.
3. The 1995 Kobe earthquake (Mw=6.9) the world. In Japan, for example, the deformation due to the 1995 Kobe earthquake was measured by InSAR (Ozawa et al., 1997; Fig. 2). In spite of this worldwide trend, however, InSAR studies did Figure 2: An interferogram associated with the 1995 Kobe earthquake. (a) Aftershock distribution due the Kobe earthquake. Triangle denotes the epicenter of the mainshock. (b) Observed coseismic deformation. (c) Calculated deformation from a fault model. Taken from Ozawa et al., (1997). not become very popular among Japanese scientists mainly because of the lack of a satellite with a long wavelength (L-band) which is required to measure deformation in vegetated regions such as the Japanese islands. JERS-1, a L-band satellite, ended its operation in 1998. Ozawa et al. (GRL, 1997) ü A L-band SAR satellite JERS-1 was available between 1992 and 1998. ü No L-band SAR satellite available between 1998 and 2006, stagnating research in Japan. ü What is L-band? Why L-band satellites are so important in Japan (and Indonesia)?
4. Frequencies of electromagnetic waves Wavelength Frequency Microwave G 1200–1500 mm 0.2–0.25 GHz P 600–1200 mm 0.25–0.5 GHz L 200–600 mm 0.5–1.5 GHz S 75–150 mm 2–4 GHz C 38–75 mm 4–8 GHz X 25–38 mm 8–12 GHz Ku 17–25 mm 12–18 GHz K 12–17 mm 18–26 GHz Ka 7.5–12 mm 26–40 GHz V 4.0–7.5 mm 40–75 GHz W 2.7–4.0 mm 75–111 GHz Infrared Far-infrared 15–1000 µm 0.3–20 THz Thermal infrared 2.5–15 µm 20–120 THz Near-infrared 0.7–2.5 µm 120–400 THz Light 380–750 nm 400–790 THz Ultraviolet 10–380 nm 790–30,000 THz ü Electromagnetic waves of longer wavelength are better at transmitting vegetation. Big advantage for vegetated areas such as Indonesia and Japan! ü L-band satellites are better than C- or X- band satellites in vegetated areas.
5. Previous and current SAR satellites Satellite (sensor) name Band Agency Period SeaSAT L US Department of Defense 1978 ERS-1 C European Space Agency 1992–1996 JERS-1 L Japan Aerospace Exploration Agency 1992–1998 RADARSAT-1 C Canadian Space Agency 1995–2013 ERS-2 C European Space Agency 1996–2011 Space shuttle (SRTM) C, X National Aeronautics and Space Administration 2000 ENVISAT (ASAR) C European Space Agency 2002–2011 ALOS (PALSAR) L JAXA 2006–2011 RADARSAT-2 C Canadian Space Agency 2007–present TerraSAR-X X German Space Agency (DLR) 2007–present COSMO-SkyMed X Italian Space Agency (ASI) 2007–present ALOS-2 L Japan Aerospace Exploration Agency (JAXA) 2014–present SENTINEL-1 C European Space Agency 2014–present Table 2: SAR satellites launched so far. ü ALOS-4 (L-band) is to be launched in 2019. ü NISAR (L-band) is to be launched in 2020 (http://nisar.jpl.nasa.gov).
6. ALOS (2006-2011) An interferogram due to the 2007 Chuetsu-oki earthquake (Aoki et al., 2008). Blue dots denote of aftershocks. (a) Observed changes of line-of sight shown by an arrow. Note that vertical i) east-west displacements cannot be separated. Gray depicts incoherent areas mainly due all. (b) Calculated displacement field by a fault model. (c) Residual between observed and displacements. 3 The derived relaxation times do not vary much between western and eastern points; they are mostly between 1.5 and 2.5 years (Fig. 4). Note that a relaxation time is estimated only when an F-test (e.g., Menke, 2012, pp. 111–112) shows that the time series better than a linear fit with a relaxation time is not obtained from noisy or -7˚35 -7˚30 (a) 427-7030 a b c d LOS -7˚35 -7˚30 (b) 430-7020 a b c d LOS 112˚40 112˚45 -7˚35 -7˚30 (c) 91-3770 a b c d LOS 112˚40 -7˚35 -7˚30 (d) 92-3770 a b c d LOS 112˚45 5 km -100 -50 0 50 100 150 200 LOS displacement (mm) -7˚30 (e) 89-3780 2007 Chuetsu-oki earthquake (Mw=6.8) Aoki et al. (EPS, 2008) 2011 Tohoku-oki earthquake (Mw=9.0) Feng & Jónsson (GRL, 2012) LUSI Aoki & Sidiq (JVGR, 2014)
7. ALOS-2 (2014-) 2018 Palu earthquake (Mw=7.5) 2015 Wolf volcano (Galápagos) eruption Xu, Jónsson, Ruch & Aoki (GRL, 2016)
8. Palu earthquake with ALOS-2
9. SAR amplitude image ü Biwa Lake, Japan ü SAR image is complex with phase and amplitude. ü Larger amplitude is represented by white. ü Higher amplitude in the cities. ü Lower amplitude on the lake.
10. How SAR works h’s surface. By convention, the direction of flight of the satellite is called ”azimuth” and the dir is the emission of microwaves is called ”range” (Fig. 8). A SAR sensor emits pulses of microw ü Radar = Radio detection and ranging ü The satellite trasmits electromagnetic wave obliquely to the ground and observes reflected waves. ü The transmission needs to be oblique to distinguish different points by different line-of- sight distance. ü Flat surface does not generate much reflected waves.
11. Shadow and layover age (called ”foreshortening” or ”layover”) (Fig. 9). By the way, microw matic view of layover and shadow resulting from an oblique incidence o pography. ü Layover: Different points cannot be separated because they are at the same distance from the satellite. ü Shadow: Rugged topography does not allow the electromagnetic wave to reach.
12. Resolution in azimuth direction ü The resolution in azimuth direction is a function of antenna size. ü Moving antenna enhance the resolution as if the target were viewed by a big antenna. h’s surface. By convention, the direction of flight of the satellite is called ”azimuth” and the dir is the emission of microwaves is called ”range” (Fig. 8). A SAR sensor emits pulses of microw
13. SAR interferometry and Young’s interference experiment schematic view of Young’s interference experiment. (b) An extension of (a) to three ellite orbits in two different timings are denoted by black lines dots. the point P in Fig. 13a, redoubling the strength or destroying each other depending r = r1 r2 ' xB R x = R B Optical path difference Width of fringes which is inversely propotional To the separation of two satellites (baseline) and proportional to the wavelength of the electromagnetic wave.
14. Effect of orbital separation schematic view of Young’s interference experiment. (b) An extension of (a) to three ellite orbits in two different timings are denoted by black lines dots. the point P in Fig. 13a, redoubling the strength or destroying each other depending ath difference. This results in fringes on the wall. When R x ± d/2, The optical path r1 − r2 (Fig. 13a) is given by Critical baseline Satellites with longer wavelength has larger critical baseline. L-band satellites require less strict orbital controls than C- band and X-band satellites. x = R B < r Range resolution Bc = R r
15. Effect of topography Figure 15: Geometry of SAR interferometry. φ = 4π λ (R2 − R1 ) = 4πR1 λ 1 − 2B cos β R1 − B2 R2 1 1/2 − 1 4πB λ cos β = 4πB λ sin (θ − α) where λ is the wavelength of microwave and R1 and R2 represent satellite t1 and t2 , respectively. An approximation from equation (??) to (??) is due Phase difference
16. Effect of topography Height difference over which the phase difference is one cycle Sensitivity to topography is higher with longer wavelength. where λ is the wavelength of microwave and R1 and R2 represent sate t1 and t2 , respectively. An approximation from equation (??) to (??) is β = π/2 − (θ − α). Equation (25) and h = H − R1 cos θ gives the height phase shift is one cycle or 2π as ha = 2π · ∂h ∂φ = 2π · ∂h ∂θ ∂θ ∂φ = λR1 sin θ 2B sin β = λR1 sin θ 2B⊥ B⊥ = B sin β Smaller ha , more sensitive to topography. Equation (26) also indicates sensitive to topography than a L-band SAR because of the shorter wavele
17. Effect of topography Topography of Etna volcano Massonet & Feigl (Rev. Geophys., 1998) Longer baseline is better to measure topography in higher sensitivity, but the baseline should not exceed the critical baseline.
18. Effect of Digital Elevation Model (DEM) The 2007 Chuetsu-oki earthquake Furuya, Takada, and Aoki (2010) Interferogram with higher-resolution DEM gives more detailed deformation field. (top) GSI 50 m (bottom) ASTER 15 m erferograms with different DEMs. The upper panel shows the one with GSI DEM with
19. Correcting interferograms Figure 14: Flow of an InSAR processing. Taken from Pritchard (2006). (a)(b) Two SLC images to gen- erate an interferogram. (c) An interferogram without corrections. (d) An interferogram after removing an orbital effect. (e) The final interferogram after removing topographical effects as well. (f) Coherence Interferogram = Orbit separation + Topography + Deformation Pritchard (Phys. Today, 2006)
20. Phase unwrapping LOS = 4⇡ (2N⇡ + ) phase change integer ambiguity What we get by InSAR is the phase fraction (wrapped phase). We need to estimate integer ambiguity to extract the real line-or-sight changes (unwrapped phase).
21. Limitation of InSAR: Line of sight (LOS) acements themselves like Figs. 4 and 5 may be more natural but it tends to lose small-scale deformation. nSAR s ons measure all three component displacements, InSAR measures a single com- ch pixel. Sensitivity of LOS to the three component displacement depends on taken from the ascending or descending orbit. The satellite flies from roughly ascending orbit and the other way around in the descending orbit (Fig. 18) . The ht-looking SAR satellite such as ALOS and many others is roughly from west to ng orbit and from east to west from the descending orbit. When the orbit is offset a degree of φ (Fig. 18), the LOS change is written by ∆Ra = ue cos φ sin θ + un sin φ sin θ − uv cos θ (28) ∆Rd = −ue cos φ sin θ + un sin φ sin θ − uv cos θ (29) ending orbits of the satellit. LOS is for a right-looking SAR satellit ü InSAR measures the line-of- sight component of the displacement, not the 3D displacement as in GNSS. ü Insensitive to north-south displacements. ü More sensitive to vertical displacements, but impossible to separate vertical and horizontal displacements. LOS change from ascending and descending orbits
22. Limitation of InSAR: Swath width Now we move onto how to generate a SLC image. SAR satellites take a orbit 600–800 km above th’s surface. By convention, the direction of flight of the satellite is called ”azimuth” and the direc- is the emission of microwaves is called ”range” (Fig. 8). A SAR sensor emits pulses of microwave ure 8: (a) Geometry of the observation by a SAR satellite. A target is seen only when it is within w of the satellite. (b) Center of microwave emission, denoted by a thick line is not perpendicular to direction of flight but offset by an angle of θs . length of 0.02–0.04 milliseconds every 0.5–1 milliseconds, or a frequency of 1000–2000 Hz, to the ge direction with an incident angle of 20–50 degrees from vertical. It travels to the azimuth direction Swath width Figure 19: An interferogram associated with the 2008 Wenchan earthquake obtained by SAR images from the ALOS satellite. One cycle of fringes corresponds to LOS changes of about 118 millimeters. Taken from Hao et al. (2009). 2008 Wenchuan (Mw=8.0) earthquake Hao et al. (GRL, 2009) ü The swath width is 50-70 km for stripmap mode. It takes some time to observe the whole deformation field if the deformation is extended in east-west direction. ü ALOS-2 has ScanSAR mode with a width of 350 km.
23. Limitation of InSAR: Decorrelation 141˚50' 142˚00' 142˚10' 142˚20' 42˚30' 42˚40' 42˚50' 43˚00' 141˚50' 142˚00' 142˚10' 142˚20' 42˚30' 42˚40' 42˚50' 43˚00' LOS 122−850 HH 20180825−20180908 141˚50' 142˚00' 142˚10' 142˚20' 42˚30' 42˚40' 42˚50' 43˚00' 10 km 0.00 0.05 0.10 Coherence 2018 Hokkado Eastern Iburi earthquake (Mw=6.5) ü Change in surface feature caused by landslide, surface faulting, volcanic ash, etc, decrease the coherence to degrade the observation. ü Temporal decorrelation is severe in vegetated regions such as in Indonesia and Japan. Images with temporal separation of 1 year can be incoherent in Indonesia.
24. Limitations of InSAR: Atmospheric disturbance ü Electromagnetic waves refract in the presence of water vapor, making the apparent line-of-sight distance longer. ü Interferograms contain long-wavelength patterns even if no real deformation is present. ü Precise correction of the atmospheric effect requires a precise knowledge of the spatial variations of water vapor. ü Removal of altitude-correlated signals does a reasonably good job. Lohman & Simons (Geochem. Geophys. Geosyst., 2005)
25. Limitations of InSAR: Ionospheric disturbance Massonnet & Feigl (GRL, 1995) experiment with SIR-C did not detect ionospheric effects (Rosen et al., 1998). Th of the lower altitude of the SIR-C satellite (∼220 km) than other satellites so that t underestimated. Interferograms associated with the 1992 Landers earthquake (F patterns due to the ionospheric effects (Fig. 22a, b, c). Acquisition time with resp ü Electromagnetic waves propagate slower than speed of light in the presence of ionized atmosphere, making apparent line-of-sight distance longer. ü The ionospheric effect is frequency-dependent, so multiple frequencies of electromagnetic waves allow us to extract the ionospheric effect. ü GNSS has two frequencies, but current SAR satellites use only single frequency. ü L-band waves are more susceptible to the ionospheric effect than C- and X-band waves.
26. Other limitations of InSAR ü Oblique incidence of electromagnetic waves makes layover and shadow in areas of steep topography. ü Temporal resolution is limited by the recurrence time of the satellite. ALOS-2: >14 days ALOS: 46 days Sentinel-1: 6 days (originally 12 days) Figure 8: (a) Geometry of the observation by a SAR satellite. A target is seen only when it is within view of the satellite. (b) Center of microwave emission, denoted by a thick line is not perpendicular to the direction of flight but offset by an angle of θs . of a length of 0.02–0.04 milliseconds every 0.5–1 milliseconds, or a frequency of 1000–2000 Hz, to the range direction with an incident angle of 20–50 degrees from vertical. It travels to the azimuth direction by emitting microwaves and receiving back-scattered waves (Fig. 8). Note that microwaves emitting to a flat surface such as sea surface or an runway reflect forward much more than backward, resulting in a very weak signal received by the SAR sensor. In addition, an oblique incidence of microwaves yields invisible targets (called ”shadow”) or different targets with the same range distance, resulting in a skewed image (called ”foreshortening” or ”layover”) (Fig. 9). By the way, microwaves must be
27. Earthquake deformation Figure 24: Deformation field due to a M 5 earthquake in Zagros Mountains, Iran, derived b Zagros Mountains, Iran Lohman & Simons (Geochem. Geophys. Geosyst., 2005) ü InSAR is capable of detecting displacements by earthquakes down to M5.0- 5.5. ü InSAR is capable of precisely locating M5.0-5.5 earthquakes whose location error is up to a few tens of kilometers with teleseismic waveforms only.
28. Postseismic deformation Gourmelen & Amelung (Science, 2005) ü Three possible mechanisms of postseismic deformation - Deep afterslip (a few years) - Viscoelastic relaxation (long) - Poroelastic (< a few months) ü Gourmelen & Amelung (2005) detected postseismic deformation of earthquakes that occurred a few tens of years before in Nevada, USA. The observed deformation is due to viscoelastic relaxation of the lower crust and upper mantle.
29. Figure 28: Postseimic deformation soon after earthquakes in South Iceland on 17 and 21 June, 2000. (a) Deformation between 19 June and 24 July, 2000. Only the area closed to the epicenter of the 17 June earthquake is shown. (b) Calculated deformation due to poroelastic relaxation. (c) Calculatd deformation due to afterslip. (d) Calculated deformation due to viscoelastic relaxation. (e) Comparison between observed and modeled deformation. (f) Deformation between 24 June and 29 July. Areas of epicenters of both 17 June and 21 June earthquakes are shown. (g) Comparison between observed and modeled deformation. After J´ onsson et al. (2003). Poroelastic deformation Jónsson et al. (Nature, 2003) ü InSAR is capable of discriminating postseismic deformation of different origin. ü Jónsson et al. (2003) succeeded in constraining the mechanism of early postseismic deformation of two M6.5 earthquakes in South Iceland Seismic Zone.
30. Interseismic deformation Fialko (Nature, 2006) ü Stacking large numbers of interferograms allows us to infer interseismic displacements as low as ~3-5 mm/yr by reducing noise. ü Fialko (2006) succeeded in separating contribution from San Jucinto and San Andreas faults in Southern California. ü Applicable to the Sumatra fault and major faults elsewhere? Spatially variable fault slip rates on an active fault is an important information for hazard assessment!
31. Volcano deformation Amelung et al. (Nature, 2000) ü Volcanoes are inaccessible in some areas because of steep topography and danger from volcanic activity. ü InSAR is capable of measuring deformation of where ground- based measurements are not available. With this background, SAR observations give us valuable informat tific interests as the study of Amelung et al. (2000) was on the cover o Figure 30: Cover of Nature on 26 October
32. Heuristic volcano deformation study Pritchard & Simons (Nature, 2002) ü Pritchard & Simons (2002) found that some volcanoes in South America which were recognized inactive are actually accumulating magma. ü A discovery only done by remote sensing technique! ü Applicalble to Indonesian volcanoes? We did it! Figure 32: An example of deformation in volcanoes which have been considered dormant.
33. Application to Indoneisan volcanoes Chaussard, Amelung & Aoki (JGR Solid Earth, 2013) Many volcanoes have sub- optimal ground-based network, so a systematic monitoring by remote sensing techniques is required!
34. Post-eruptive thermal contraction of Usu volcano, Japan ü The 2000 vent: LOS velocity of about 38 mm/yr in the ALOS-1 period (2006-2011). Negligible deformation in the ALOS2 period (2014-2017). ü The 1977 vent: Maximum LOS velocities of about 66, 45 and 43 mm/yr in the JERS (1992-1998), ALOS-1 and ALOS-2 periods. ü The 1943 vent: Steady deformation with a maximum LOS velocity of about 20 mm/yr in 1992-2017. Ascending Descending 2000 1977 1943 Wang & Aoki (JGR Solid Earth, in revision)
35. Temporal evolution of vertical displacements: The 1943 vent
36. Temporal evolution of vertical displacements: The 1977 vent
37. Temporal evolution of vertical displacements: The 2000 vent
38. Optimum deformation parameters ü Shallow sources (<400 m bsl) are responsible. ü Thermal diffusivity much larger than that derived in lab experiments (0.1–1×10-5 m2/s) . Longitude (deg) Latitude (deg) Depth ( m bsl) Thermal diffusivity (×10-5 m2/s) Volume (×106 m3) 2000 vent 140.8034 42.5541 213±19 8.21±1.01 6.67±0.21 140.8118 42.5563 100±13 8.06±1.20 2.05±0.13 1977 vent 140.8353 42.5416 369±29 10.05±1.09 132.18±8.01 1943 vent 140.8662 42.5426 92±12 1.65±0.22 49.51±3.12
39. Time-varying volcano deformation: LUSI n times do not vary much between western ey are mostly between 1.5 and 2.5 years ation time is estimated only when an F-test (e.g., Menke, 2012, pp. 111–112) shows that an exponential curve fits the time series better than a linear fit with a confidence of 95%. Thus a relaxation time is not obtained from noisy or short time series. (Fig. 4). -7˚35 -7˚30 (a) 427-7030 a b c d LOS -7˚35 -7˚30 (b) 430-7020 a b c d LOS 112˚40 112˚45 -7˚35 -7˚30 (c) 91-3770 a b c d LOS 112˚40 -7˚35 -7˚30 (d) 92-3770 a b c d LOS 112˚45 5 km -100 -50 0 50 100 150 200 LOS displacement (mm) (e) 89-3780 2006 2008 2010 −300 −200 −100 0 100 200 300 427−7030 τ=2.52±0.34 yr 430−7020 89−3780 91−3770 τ=2.47±0.66 yr 92−3770 τ=2.09±0.03 yr (a) −7.5111°S 112.6583° E LOS change (mm) Year 2006 2008 2010 427−7030 τ=2.22±0.07 yr 430−7020 89−3780 91−3770 τ=1.35± 0.03 yr 92−3770 τ=1.83±0.03 yr (b) −7.5125°S 112.6778° E Year 2006 2008 2010 −300 −200 −100 0 100 200 300 427−7030 τ=2.93±0.47 yr 430−7020 89−3780 91−3770 92−3770 τ=1.63±0.02 yr (c) −7.5167°S 112.7000° E Year 2006 2008 2010 −300 −200 −100 0 100 200 300 427−7030 τ=2.21±0.24 yr 430−7020 89−3780 91−3770 τ=1.84±0.30 yr 92−3770 τ=1.64±0.15 yr (d) −7.5417°S 112.7000° E Year Fig. 4. Temporal evolution of LOS changes for four representative points whose location is shown in Fig. 3. Solid curves denote the calculated LOS changes by exponential curve fit the observed LOS changes more than two years after the onset of the eruption (t N 2008.408). They are shown only when an F-test (Menke, 2012, p. 111–112) gives a bett exponential fit than a straight-line regression by a confidence level of 95%. Y. Aoki, T.P. Sidiq / Journal of Volcanology and Geothermal Research 278–279 (2014) 96–102 Aoki & Sidiq (JVGR, 2014)
40. Land subsidence to 2015 with the deformation value reach up to 10 cm on points A and 6 cm on point D. r, in the northern part (Kenjeran sub-district) subsidence reach up to 2.2 cm/ year and the part is more lower reach up to 0.6 cm/ year. Generally the coherence value of ALOS-2 images than ALOS but the temporal resolution is opposite for the area in Indonesia. Figure 6. Plot of Specified Point in Surabaya lusions rk has presented an analysis of the ground subsidence phenomena in Surabaya City. The d multi-temporal InSAR technique is applied to this site using 8 ALOS-2 PALSAR-2 images from 2014 to 2017. We identified a few locations undergoing subsidence at rates up to 2,2 Figure 7. Identification of subsided areas in Jakarta based on DInSAR technique. (a) Interferogram from 2007-2011 based on ALOS-PALSAR data, (b) subsided area in Pluit region from 2007-2011 or 1472 days based on ALOS-PALSAR data, (c) subsided area in Pluit area from 2014 -2016 or 658 days based on ALOS-2 data, image not rectified. 4. Conclusions This research shows the ability of SAR data to identify ground deformation in Jakarta area by analysing the amplitude and phase components. Sentinel-1A data and S1TBX software are useful to obtain the land surface changes based on amplitude analysis. This research also found that atmospheric phase affects much to C-band SAR data as already identify by previous studies [11] and c b a Surabaya, ~30 mm/yr of subsidence Aditiya, Takeuchi & Aoki (2017) North Jakarta, Up to 260 mm/yr (!) of subsidence Agustan et al. (2016) “If we look at our models, by 2050 about 95% of North Jakarta will be submerged.” – Heri Andreas (ITB), BBC Indonesian, Aug. 2018
41. Damage detection = Cpre Cco = Cpre Cco Cpre + Cco mum case the temporal baseline between each acquisition is only one repeat so Section 4). (or more) SAR images, fulfilling the requirements mentioned above, are co-registered master image and resampled to its reference grid (Figure 4). Additionally, a common sures that only the overlapping parts of the spectrums are used. Thereby, the spatial fect (see Section 2.1.) is reduced [74]. In the next step, interferograms between he two slave images are generated: One pre-disaster InSAR pair (t1 and t2 ) and one R pair (t2 and t3 ). Then, for both InSAR pairs the coherence is computed according to e Section 2.1). Moreover, as described by Equation (9), also two SAR intensity computed using again the co-registered pre- (t1 and t2 ) and co-disaster (t2 and t3 ) airs [8]. The damage caused by the natural disaster is then assessed by detecting the he corresponding image pairs (see Section 2.3 for more details). pre co Two possible metrics for damage detection as a function of coherence C e.g., Arciniegas et al. (IEEE TGRS, 2007) Hoffmann et al. (Int. J. Remote Sens., 2007) Here we only look at co- disaster coherence. Pre-disaster coherence should also be used to for damage detection to evaluate temporal decorrelation. Source of decorrelation ü Surface rupture ü Too much deformation ü Too much vegetation ü Landslide, ashfall ü Snow
42. The 2016 Kumamoto earthquake Earthquakes of MJMA >5.5 Mo d hr min lon lat depth MJMA 04 14 21 26 130.8087 32.7417 11.39 6.5 04 14 22 07 130.8495 32.7755 8.26 5.8 04 15 00 03 130.7777 32.7007 6.71 6.4 04 16 01 25 130.7630 32.7545 12.45 7.3 04 16 01 45 130.8990 32.8632 10.55 5.9 04 16 03 03 131.0868 32.9638 6.89 5.9 04 16 03 55 131.1910 33.0265 10.89 5.8 04 18 20 41 131.1998 33.0020 8.64 5.8 04 19 17 52 130.6353 32.5352 9.96 5.5
43. SAR Interferometry (InSAR) and pixel offset (or offset tracking) InSAR Pixel offset Measurement Phase Amplitude Error 20-30 mm ~300 mm Large deformation Incoherent Capable
44. Decorrelation and landslides Landslide sites are identified from optical images by Geospatial Information Authority. Source of decorrelation Surface rupture, Too much deformation Too much vegetation Landslide, ashfall, Snow
45. Surface fractures Identified from an optical image by Geospatial Information Authority
46. New (unconventional) technique: Amplitude changes Tobita et al. (EPS, 2006) ü Tobita et al. (2006) compared amplitude of SAR images before and after the 2004 Sumatra- Andaman earthquake to detect uplifted and subsided regions on the coast. ü Subsidence decreases amplitude of the SAR image on the coast. ü Applicable to e.g., identifying tsunami- inundated areas.
47. New (unconventional) technique: pixel offset InSAR Pixel offset Measurement Phase Amplitude Error 20-30 mm ~300 mm Large deformation Incoherent Capable The 2016 Kumamoto earthquake
48. Pixel offset analysis: Kumamoto earthquake ü Pixel offset analysis can give two-component displacements.
49. The Palu earthquake
50. What to do: The Palu earthquake ü Numerical rupture simulation to be consistent with SAR measurements. The Palu earthquake may have been a super- shear earthquake. ü Damage detection from the coherence of interferograms. Establishing a system to systematically process the incoming data may be beneficial. ü Postseismic deformation? ü Combining SAR measurements with GNSS.
51. What to do: Volcano deformation ü Monitoring all Indonesian (and beyond?) volcanoes systematically with SAR images by processing incoming images. ü Detecting ashfall, pyroclastic flow, and lava flow from InSAR coherence. Establishing a system to systematically process the incoming data may be beneficial. ü Combine InSAR with GNSS measurements (if any) to correct InSAR measurements.
52. Summary ü SAR can measure Earth’s surface day and night, regardless of the weather. ü SAR can measure surface displacements associated with various phenomena with a high spatial resolution, but with some limitations. ü SAR can be a powerful tool in detecting damages caused by landslide, flooding, tsunami, …. Terima Kasih!