University of Western Ontario - Electronic Thesis and Dissertation Repository

Location of Thesis Examination

Room 4185 Support Services Building

Degree

Doctor of Philosophy

Program

Geophysics

Supervisor

Dr. Kristy Tiampo

Abstract

In the following work, I address the problem of coherence loss in standard Differential Interferometric SAR (DInSAR) processing, which can result in incomplete or poor quality deformation measurements in some areas. I incorporate polarimetric information with DInSAR in a technique called Polarimetric SAR Interferometry (PolInSAR) in order to acquire more accurate and detailed maps of surface deformation.

In Chapter 2, I present a standard DInSAR study of the Ahar double earthquakes (Mw=6.4 and 6.2) which occurred in northwest Iran, August 11, 2012. The DInSAR coseismic deformation map was affected by decorrelation noise. Despite this, I employed an advanced inversion technique, in combination with a Coulomb stress analysis, to find the geometry and the slip distribution on the ruptured fault plane. The analysis shows that the two earthquakes most likely occurred on a single fault, not on conjugate fault planes. This further implies that the minor strike-slip faults play more significant role in accommodating convergence stress accumulation in the northwest part of Iran.

Chapter 3 presents results from the application of PolInSAR coherence optimization on quad-pol RADARSAT-2 images. The optimized solution results in the identification of a larger number of reliable measurement points, which otherwise are not recognized by the standard DInSAR technique. I further assess the quality of the optimized interferometric phase, which demonstrates an increased phase quality with respect to those phases recovered by applying standard DInSAR alone.

Chapter 4 discusses results from the application of PolInSAR coherence optimization from different geometries to the study of creep on the Hayward fault and landslide motions near Berkeley, CA. The results show that the deformation rates resolved by PolInSAR are in agreement with those of standard DInSAR. I also infer that there is potential motion on a secondary fault, northeast and parallel to the Hayward fault, which may be creeping with a lower velocity.