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

Statistical methods in seismology

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract On July 16, 1945 the first nuclear weapon test (code named Trinity) began an evolution in warfare that led to the cold war where mutually assured destruction (MAD) prevented nuclear conflict between nations. The generation that fought the cold war understood the destructive power of a nuclear weapon—many had observed first hand the ruins of Nagasaki and Hiroshima. The modern nuclear weapon has the potential to kill over 1,000,000 people in seconds if detonated in a large metropolitan city. In contrast to the man‐made threat of nuclear weapons, there are an estimated 230,000 people dead or presumed so as a result of the 2004 Indian Ocean tsunami. Seismology is the core science in monitoring for nuclear weapon tests worldwide—an essential function in global efforts to eliminate nuclear weapons. Equally important, seismology provides the theory and methods to monitor and warn for natural threats such as the Indian Ocean tsunami. Today, many scientists continue research and development efforts to more effectively monitor natural seismic activity and eliminate nuclear weapons from the globe. This article reviews an important aspect of that research—mathematical statistics contributions to seismic monitoring with emphasis on underground nuclear weapon test monitoring. Copyright © 2010 John Wiley & Sons, Inc. This article is categorized under: Applications of Computational Statistics > Computational Physics and Computational Geophysics

Typical seismic waveform with the onset of P and S phases identified. Many phases can appear in a waveform, each generated by reflection and/or refraction. Teleseismic P waves are red and teleseismic S waves are blue. (Figures courtesy of Professor Edward Garnero, Arizona State University).

[ Normal View | Magnified View ]

Scatter plots relating log base 10 yield to event body wave magnitude (here denoted mb) and regional mb(Lg) magnitudes with fitted regression lines. Yields are assumed to be exact.

[ Normal View | Magnified View ]

Intuition of likelihood‐based seismic identification. Population membership defines four possible identification statements: Explosion, Earthquake, Indeterminate, and Unidentified.

[ Normal View | Magnified View ]

Probability density function for observed mb at a station (Eq. (13)). The parameter values for this example are (background), σ = 0.25 (signal variability), µ = 1 (average seismicity), and δ = 0.5 (detection threshold).

[ Normal View | Magnified View ]

Example of detection threshold contours. Contours are constructed with a fictitious network of stations providing simulated calibrations of detection threshold models.

[ Normal View | Magnified View ]

Empirical detection threshold estimation. SLR coupled with a leverage analysis of the data determine the regression line. Ordinate units are natural logarithms. For this example, the estimated detection threshold is mb = 5.3.

[ 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