Abstract.In this study, we develop an procedure to predict displacements causing by earthquakes at strong motion recorder sites without the use of experts' opinion. The algorithm combines methods from signal process and edge detection. We adopt methods to filter noise and detect the jumps in the strong motion data, then determine the baseline and integrate the data to find the displacement.

A set of strong motion records of earthquakes that occurred in Taiwan is used to test the algorithm. The results will be compared to the near by GPS records and prior study [#!ywu!#]. Because of the loss of experts' opinions, the result may seem to have larger errors. However, the purpose of this algorithm is to create a general process that can efficiently predict any displacement regardless the location of the strong motion. Therefore, we will tolerate a larger margin of error when comparing the computing results and the GPS records.

Received: August 17, 2007

AMS Subject Classification: 86A17, 86A15, 94A08, 41A10

Key Words and Phrases: edge detection, signal processing, surface deformation, semiology

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2008
Volume: 42
Issue: 3