The Dependence of Magnitude Uncertainty on Station Coverage

A useful event characterization parameter for relatively large (e.g., mb > 4) teleseismic events is mb-Ms, provided the events are not too deep. To use mb-Ms as a parameter with which to screen events of natural seismicity, at a given confidence level, the uncertainty (standard deviation) in mb-M...

Full description

Saved in:
Bibliographic Details
Main Authors: Bottone, Steven, Fisk, Mark D, Gray, Henry L, McCartor, Gary D
Format: Report
Language:English
Subjects:
CARTESIAN COORDINATES
MULTIVARIATE ANALYSIS
PE62301E
SEISMIC ARRAYS
SEISMIC DATA
SEISMIC SIGNATURES
SEISMOGRAPHS
SEISMOLOGICAL STATIONS
Seismology
SEISMOMETERS
STATISTICAL SAMPLES
UNCERTAINTY
WUPLNM95GMAA
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A useful event characterization parameter for relatively large (e.g., mb > 4) teleseismic events is mb-Ms, provided the events are not too deep. To use mb-Ms as a parameter with which to screen events of natural seismicity, at a given confidence level, the uncertainty (standard deviation) in mb-Ms must be quantified. This report presents a generalization to the current method of screening events using mb-Ms by including correlation between station observations due to station coverage. After a description of the current method used to estimate the uncertainty in mb-Ms, an analysis is presented, using all events currently available in the Reviewed Event Bulletin (REB) for which Ms was measured at six or more stations, to determine the dependence on station coverage. Generally, Ms measurements at stations which are located along the same direction from the event are positively correlated, while measurements at stations which are located at near right angles from the event are negatively correlated. Using the data to estimate the correlation between stations as a function of angle relative to the event location, a multivariate statistical model for the uncertainty in mb-Ms is given and applied to each of the 400 relevant events in the REB. An interpretation of the results and recommendations for treatment of station coverage is provided.