Many studies have demonstrated the potential of satellite synthetic-aperture-radar (SAR) interferometry to map surface displacements. In the context of earthquakes, it is particularly interesting that reference data typically exist that were acquired before the event. In addition, data can be obtained reliably within a relatively short time after the earthquake because of the all-weather capability of radar systems.
To map co-seismic displacement during the 12 January 2010 Haiti earthquake,1 we used data collected by the Phased Array-type L-band SAR (PALSAR) instrument onboard the Advanced Land Observing Satellite (ALOS), which is operated by the Japan Aerospace Exploration Agency (JAXA). We preferred L-band (central frequency: 1270MHz) compared to C- (5.3GHz: ERS-2, ENVISAT, Radarsat-1, or Radarsat-2) or X-band (8–12GHz) data because of both the lower decorrelation of L-band data over vegetated areas and its reduced sensitivity to displacement. By mid-February 2010, suited measurement pairs containing at least one scene taken before and one after the earthquake became available from one descending and several ascending orbit tracks.
For each suited pair, we processed a differential interferogram using the Shuttle Radar Topographic Mission digital elevation model as altitude reference in the simulation of the topographic phase component. Figures 1 and 2 show the wrapped interferometric phase for two selected pairs. A color cycle corresponds to one phase cycle (2π).
Figure 1. Haiti earthquake's co-seismic Phased Array-type L-band Synthetic Aperture Radar (PALSAR) differential interferogram 20090309–20100125 (path 447D). One color cycle corresponds to 11.8cm line-of-sight (LOS) displacement. (© JAXA and the Japanese Ministry of Economy, Trade, and Industry, METI.)
As Figure 1
, but for interferogram 20080209–20100214 (path 137). (© JAXA, METI.)
Except for signal noise, this differential interferometric phase relates mainly to ground-surface displacement in the line-of-sight direction of the SAR and path-delay effects in the atmosphere. The phase to line-of-sight displacement sensitivity is a phase cycle per half a wavelength. (The PALSAR wavelength is 23cm.) The dominant phase signal in the two interferograms (fringe patterns below the image centers) clearly relate to co-seismic displacement. The much smaller phase signals observed in other areas of the interferograms may relate to atmospheric path-delay effects caused by heterogeneity in atmospheric water vapor concentrations.
For a quantitative interpretation, we unwrapped the differential interferometric phases starting from a given spatial reference point. We estimated line-of-sight displacement components for several pairs acquired in ascending and descending orbits. After geocoding this information to a common map geometry, we combined the two components estimated from the descending orbit pair 20090309–20100125 and the ascending orbit pair 20080209–20100214 to calculate the vertical and east–west displacement components (see Figures 3 and 4, respectively).
Figure 3. Vertical component of the Haiti earthquake's co-seismic displacement, derived from PALSAR differential interferograms 20090309–20100125 (path 447D) and 20080209–20100214 (path 137). Positive values correspond to uplift. (© JAXA, METI.)
Figure 4. East–west component of the Haiti earthquake's co-seismic displacement, derived from PALSAR differential interferograms 20090309–20100125 (path 447D) and 20080209–20100214 (path 137). Positive values correspond to a displacement to the west. (© JAXA, METI.)
The maximum vertical displacement observed is an uplift of approximately 60cm (see the red area in Figure 3). Given that we applied some spatial filtering and that unwrapping in the areas of the highest spatial gradients was critical, this may slightly underestimate the true displacement.
Our results demonstrate the potential of ALOS PALSAR differential interferometry to map large co-seismic displacements. Using measurement pairs acquired in ascending and descending paths, we resolved the vertical and east–west components of the displacement. We have provided our results to geoscientist colleagues for further study of the earthquake mechanism.
We acknowledge PALSAR Announcement of Opportunity project 094 and European Space Agency Category 1 project C1F.6549. This work was supported by the European Commission's Seventh Framework Programme Global Monitoring for Environment and Security (GMES) project ‘Services and Applications for Emergency Response’ (SAFER).
Gamma Remote Sensing AG
Urs Wegmüller is chief executive officer of Gamma Remote Sensing AG, a company involved in developing innovative regular and interferometric SAR applications, related services, and consulting. The company also provides licenses for GAMMA software and develops microwave instruments.