Detection of Crustal Deformations by SAR Interferometry in Combination with Dense GPS Array

Abstract

We aim to improve the accuracy of space-borne synthetic aperture radar (SAR) interferometry making advantages of Japanese nationwide GPS array. In order to detect crustal deformations in high accuracy without delay, the Geographical Survey Institute is constructing dense GPS array that consists of 900 permanent GPS stations all over the country. The average interval of the stations is about 25km. Though this is one of the densest GPS arrays in the world, the surface change observed by GPS is the data “at a point”. So it can miss detecting the surface changes associated with geological activities at the most important spot, such as deformations near an epicenter. To supplement this disadvantage, we employ SAR interferometry (InSAR). Though the InSAR measurements are less accurate than the GPS measurements, it provides better spatial resolution of about 100m, measuring dislocations at millions of points over areas that can not be covered by GPS. Since approximately nine (9) GPS stations are included in a SAR image of about 76km by 76km square obtained by JERS-1, we can use them to constrain the amount of surface change from InSAR result. We analyzed by this technique a fault model of an earthquake (M=6.3) that occurred in Kagoshima-ken, Japan on March 26, 1997. Though the maximum amount of detected deformations is as small as 5 cm, we have obtained a unique solution by using no other data than GPS and InSAR.