Event Identification - Applications to Area of Interest Events
Abstract
An integrated event identification system has been developed to operate on an 'Area of Interest' data base. This system includes an automated measurement and analysis section using the IBM 360/44 computer and an interactive editing and identification section using the PDP-15/50 computer. Amplitude measurements are made on both short-period and long-period data and serve as the basis for computing most of the discriminants. Other discriminants include mean instantaneous frequency and mean instantaneous phase standard deviation, and complexities. Ringdal's maximum likelihood technique is used to generate 'unbiased' event measurements from the individual station measurements obtained for each event. At present, a set of 35 events has been processed with this system. A multivariate analysis technique was used to classify the events either as earthquakes or as particular types of anomalous events. The results indicate that the system makes it possible to achieve good separation between anomalous events and a selected set of normal shallow-depth earthquakes.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 13, 1978
- Accession Number
- ADA072956
Entities
People
- A. C. Strauss
- A. G. R. Bell
- D. L. Dietz
- J. S. Shaub
- R. L. Sax
Organizations
- Texas Instruments