Extending interferometric synthetic aperture radar measurements from one to two dimensions.

摘要

Interferometric synthetic aperture radar (InSAR), a very effective technique for measuring crustal deformation, provides measurements in only one dimension, along the radar line of sight. Imaging radar measurements from satellite-based systems are sensitive to both vertical and across-track displacements, but insensitive to along-track displacement. Multiple observations can resolve the first two components, but the along-track component remains elusive. The best existing method to obtain the along-track displacement involves pixel-level azimuth cross-correlation. The measurements are quite coarse (typically 15 cm precision), and they require large computation times. In contrast, across-track and vertical InSAR measurements can reach centimeter-level precision and are readily derived.; We present a new method to extract along-track displacements from InSAR data. The new method, multiple aperture InSAR (MAI), is based on split-beam processing of InSAR data to create forward- and backward-looking interferograms. The phase difference between the two modified interferograms provides the along-track displacement component. Thus, from each conventional InSAR pair we extract two components of the displacement vector: one along the line of sight, the other in the along-track direction. Multiple MAI observations, either at two look angles or from the ascending and descending radar passes, then yield the three-dimensional displacement field.; We analyze precision of our method by comparing our solution to GPS and offset-derived along-track displacements from interferograms of the M7.1 1999, Hector Mine earthquake. The RMS error between GPS displacements and our results ranges from 5 to 8.8cm. Our method is consistent with along-track displacements derived by pixel-offsets, themselves limited to 12-15cm precision. The theoretical MAI precision depends on SNR and coherence. For SNR=100 the expected precision is 3, 11cm for coherence of 0.8, 0.4, respectively.; Finally, we evaluate how the new measurements improve the determination of the earthquake coseismic slip distribution by comparison of models derived from multiple data types. We find that MAI data help constrain the southern portion of the lip distribution, by adding information where GPS data are sparse and the deformation is below the azimuth pixel-offsets detection threshold.