Abstract

The Mw 7.6 Chi-Chi earthquake struck central western Taiwan in 1999. The rupture was complex with several dislocations along the 100-km long Chelungpu thrust fault. Revisiting this earthquake is a challenge, as the precision and coverage of the data sets available are quite poor. Furthermore, the topographic and vegetation coverage complexity of the area prevents coherent radar images. In this paper, radar and optical images, and terrestrial geodetic measurements, were utilised to study the fault. The Particle Swarm Optimization and Okada Inversion (PSOKINV) geodetic inversion package was used with the generalized Akaike’s Bayesian Information Criterion (gABIC) to precisely determine the slip distribution and relative weighting of datasets. Differences in results using the data sets jointly or separately (e.g., under-estimation due to InSAR, inconsistencies in SPOT offsets, smoother slip distribution with gABIC weighting) are observable. Most of the energy was released in the northern part of the fault, where the strike veers toward the east, and mainly at depths less than 4 km. The PSOKINV-gABIC approach is viable for the study of complicated cases such as the Chi-Chi earthquake and can significantly benefit the weight determination and physical realism of the fault geometry.

Highlights

  • Taiwan is the emerged part of an orogen formed by the collision between the Eurasian and Philippine Sea plates [1] entering in convergence at a rate of 82 mm/year with an orientation of 310◦

  • The main steps of the procedure are an ortho-rectification of the primary image with Tandem-X DEM due to precise selection of ground control points (GCPs), ortho-rectification of the secondary image using GPS data values in the far field to correct for long wavelength bias

  • A nonlinear inversion is performed to determine the fault parameters minimizing the square misfit under the hypothesis of a uniform slip on a rectangular fault using the multi-peak particle swarm optimization (M-PSO) [36], a non linear algorithm approach based on the elastic Okada dislocation models [26]

Read more

Summary

Introduction

Taiwan is the emerged part of an orogen formed by the collision between the Eurasian and Philippine Sea plates [1] entering in convergence at a rate of 82 mm/year with an orientation of 310◦. Two subductions are consuming the island—at the south, the Eurasian plate goes beneath the Philippine sea plate in the east direction at the Manila trench, and the Philippine sea plate subducts towards the north near the Ryukyu trench It is a high-risk seismic area (Figure 1) with the major and most destructive inland earthquake of the 20th century being the Chi-Chi event in central western Taiwan, occurring on 21st September 1999 with a magnitude of Mw 7.6 and at a depth of 33 km (according to United States Geological Survey (USGS)). Reference [10] used SPOT (Satellite Pour l’Observation de la Terre) images and interpreted the coseismic slip as shallow and detected more displacement on the hanging-wall than the footwall They estimated that the deeper part of the fault was not activated during the event and the western front of the area should produce a Mw 7 to 7.5 event every 150–250 years. The method presented in this paper shows the impact of the different data sets on the slip distribution as well as the necessity to properly determine the relative weights

Geodetic Data Sets and Processing
GPS Measurements
Levelling Data
InSAR: ERS-2 Interferometric Processing Chain
28 October 1999 21 January 1999
Sub-Pixel Correlation of SPOT Images
29 January 1999 23 November 1999
Earthquake Modelling
Modelling Strategy
Benefit of Multiple Data Sets
Seismic Hazard
Conclusions

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.