Home
This Title All WIREs
WIREs RSS Feed
How to cite this WIREs title:
WIREs Comp Stat

Seismic event identification

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract Earthquakes and explosions generate seismic waveforms that have different characteristics. However, the challenge of confidently differentiating between these two signatures is complex, and requires the integration of physical and statistical techniques. This article reviews the methods for constructing discrimination features from diverse physical observations. These discrimination features are appropriate for many statistical classification frameworks. Under the null hypothesis an event is an explosion, we discuss strategies for constructing P‐values which can be interpreted as standardized discrimination features. We develop standardized discriminants for both teleseismic (simple propagation path in the mantle) and regional (complicated propagation path in the crust) events, following the trend toward characterizing increasingly smaller single‐point explosions. Copyright © 2010 John Wiley & Sons, Inc. This article is categorized under: Applications of Computational Statistics > Computational Physics and Computational Geophysics

Teleseismic event and station locations.

[ Normal View | Magnified View ]

Infrasound discriminant P‐values for the conceptual null hypothesis H0 : Single‐point fully contained explosions. Explosions are shown in red, and shallow earthquakes are yellow.

[ Normal View | Magnified View ]

P‐values for the regional phase amplitude hypothesis H0 : µ − µ ≥ 0.43. Single‐point fully contained explosions are shown in red, and shallow earthquakes are yellow.

[ Normal View | Magnified View ]

Bivariate plot of complexity and depth P‐values. The complexity P‐values (typicality indices) are for a multivariate test of the hypothesis H0 : Station complexity discriminants are from the single‐point fully contained explosion population. Explosions are shown in red, shallow earthquakes are yellow, and deep earthquakes are green. Arrays used in the analysis are EKA, GBA, WRA and YKA listed in Table 1.

[ Normal View | Magnified View ]

P‐values for the first motion hypothesis H0 : The source mechanism is a single‐point fully contained explosion. The abscissa is the network (average) event b. Single‐point fully contained explosions are shown in red, shallow earthquakes are yellow, deep earthquakes are green, and mining explosions are gray.

[ Normal View | Magnified View ]

Polarity of first motion discriminant P‐value as a function of the number of stations (N = n) that have positive first motion for M = {6, 7}.

[ Normal View | Magnified View ]

P‐value calculation for the order statistic formulation of the pP discriminant. The joint probability model for stepout and number of observed pP is integrated over the dark gray region and subtracted from one, giving the P‐value.

[ Normal View | Magnified View ]

P‐values for the order statistics and regression formulation of the pP discriminant are top and bottom respectively. The abscissa is the average epicentral distance (degrees) between event and seismic stations. Shallow earthquakes are yellow, and deep earthquakes are green.

[ Normal View | Magnified View ]

Order statistic pP discriminant P‐value as a function of R = r stepout in seconds for n = 2 observed pP.

[ Normal View | Magnified View ]

IASPEI travel time curves for pPP are shown in raw form (left) and in the log domain (right) for a series of different event depths. Superimposed on the curves are data for 10‐ and 50‐km‐deep events.

[ Normal View | Magnified View ]

The direct P phase and the surface reflected pP phase are shown as ray paths.

[ Normal View | Magnified View ]

Bivariate plot of mb versus MS (mb/Ms) and depth P‐values. Single‐point fully contained explosions are shown in red, shallow earthquakes are yellow, and deep earthquakes are green.

[ Normal View | Magnified View ]

P‐values for H0 : µ − µ ≥ 1.35. The abscissa is the average epicentral distance (degrees) between event and seismic stations. Single‐point fully contained explosions are shown in red, shallow earthquakes are yellow, and deep earthquakes are green.

[ Normal View | Magnified View ]

P‐values for the hypothesis H0: event depth ≤ 50 km. The abscissa is the established seismic catalog event depth. Single‐point fully contained explosions are shown in red, shallow earthquakes are yellow, deep earthquakes are green, and mining explosions are gray.

[ Normal View | Magnified View ]

P, S, and Rayleigh (R) phases for a seismic waveform, both as raw data (top) and filtered (bottom) at 25 s. Information about the event, and about MS measurement can be found in Ref 15.

[ Normal View | Magnified View ]

Regional event and station locations.

[ Normal View | Magnified View ]

Browse by Topic

Applications of Computational Statistics > Computational Physics and Computational Geophysics

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts