Science Service System

Summary of Proposal GEO2516

TitleRecognizing re-nascent volcanic activity along the American Cordillera using 6-beam Wide ScanSAR TerraSAR-X interferometry
Investigator Feigl, Kurt - University of Wisconsin-Madison, Geoscience
Team Members
Professor Feigl, Kurt - University of Wisconsin-Madison, Department of Geoscience
Professor Sandwell, David - Scripps Institution of Oceanography, Geophysics
SummaryScanSAR-to-ScanSAR interferometry is advantageous for monitoring volcanic hazard. In this mode, the swath of width 195 to 266 km allows more efficient coverage than does the standard strip-mode with a swath width of ~100 km. The drawback of the ScanSAR approach is that the data are acquired in bursts of several hundred echoes in each of six sub-swaths. Complete wide swath coverage is not possible because the time between echoes must be greater than the time it takes the radar pulse to traverse the swath. Incredibly, a SAR image for each sub-swath can be constructed from one sixth of the full along-track coverage. However, interferometric phase coherence can only be achieved if the bursts on the repeat pass are well aligned with the bursts of the reference pass. This requires the timing of the repeated radar bursts to be accurate to about 3 milliseconds based on GPS position and time information. We understand that the TerraSAR-X mission will be so synchronized when operated in the new 6-beam Wide ScanSAR mode. Theoretically, this improvement will increase the number of possible interferograms and lower the repeat interval by a factor of 5.

We have funding from the U.S. National Science Foundation (NSF) to develop an open-source software package called GMTSAR to develop tools to completely automate ScanSAR-to-ScanSAR interferometry for data from the ALOS-2 and Sentinel-1 SAR missions when they are launched in 2014 by JAXA and ESA, respectively. The key to automating ScanSAR-to-ScanSAR interferometry is to have robust software, combined with better than 50 cm accurate satellite orbits so that the subswaths can be interfered independently and combined seamlessly into a single large interferogram. This requires phase precision of better than 1/8 the radar wavelength (4 mm at X-band). Automation of the image co-registration requires that the geometric accuracy be better than about 10 m. In order to extend this capability to the TerraSAR-X mission, we request access to data that is acquired in 6-beam Wide ScanSAR mode. Adapting the GMTSAR to handle such data should be possible, as long as the image is well focused, burst-aligned and accompanied by accurate orbital information.

The InSAR results from TERRASAR-X will be validated with GPS and leveling data acquired on the ground, where available, as well as InSAR results from other missions.

Back to list of proposals

DLR 2004-2016