- © 2006 by the Seismological Society of America
Probabilistic seismic hazard analysis (PSHA) is the methodology used to estimate how often some chosen level of earthquake ground motion can be reached or exceeded at one or more places of interest. It synthesizes knowledge about where earthquakes occur, how often they occur, and how big they can be; knowledge of the ground motion that is excited by earthquakes and how it changes with earthquake magnitude, distance, and (occasionally) azimuth; and the uncertainties that attend both the models of seismogenesis and the models of ground-motion excitation/propagation (epistemic uncertainty), as well as the intrinsic variability in the way earthquakes actually occur and in the way that ground motion actually happens (aleatory uncertainty). Risk analysis involves the calculation of loss, generally reckoned through damage and loss functions in terms of dollars of replacement costs and/or loss that will be equaled or exceeded per year, given the hazard analysis.
Novices to hazard and risk analysis should note that seismic hazard—or any other natural hazard—cannot be reduced; natural hazard analysis is simply a quantitative assessment of what nature serves up in the way of earthquakes, volcanoes, hurricanes, tornadoes, wildfires, droughts, and the like. Risk, however, can be reduced, most notably through land-use planning and building codes, but increasingly through rational self-interest enlightened by modern hazard and risk analysis. Even so, total risk always seems to be on the rise, due to an ever-growing population and the ever-growing built environment needed to support it.
Robin McGuire is a top practitioner in these fields, and he also enjoys excellent research credentials in the topics covered by Seismic Hazard and Risk Analysis. This book discusses in a comprehensive and quantitative way all of the essentials for calculating seismic hazard and risk, and it presents this material in a readable and engaging manner. More importantly, McGuire integrates seismic hazard and risk into a single, medium-length monograph (178 pages of well-illustrated text), a first in this business. Seismic Hazard and Risk Analysis also has been written for both earthquake scientists and earthquake engineers, and few books have been successful in reaching both audiences, at least at this level of analysis. I believe this one will—or at least should—be of value to both groups, the many problems of inconsistent notation used by earthquake scientists and earthquake engineers notwithstanding. (Why do seismologists use base-10 and engineers use base-e in describing the same exponential Gutenberg-Richter frequency-of-occurrence relationships?) With the appropriate supplementary materials from its references, this book would make an excellent graduate-level course in seismic hazard and risk analysis, in either earth sciences or engineering. Judging by the misconceptions, misinformation, and just plain nonsense about PSHA found in publications as diverse as EOS (American Geophysical Union), Civil Engineering, Engineering Geology, and this one, I will hazard a guess, so to speak, that such courses are in desperately short supply.
There are always caveats for books of this scope and dimension, and so to appear even-handed, I will mention three. McGuire tells us very early on p. 3: “This monograph is intended as a first step in connecting seismic hazard and seismic risk. Without this connection, seismic hazard studies would be merely an interesting avocation.” Readers of this book, especially those from earth-sciences backgrounds, are advised not to bite on this hook. The PSHA methodology has great value in synthesizing and analyzing all manner of earth-sciences data, models, and uncertainties in probabilistic portrayals, a value independent of any risk analysis that might follow the hazard analysis.
Structural engineers presumably will be disappointed that there is so little structural engineering in this book. Between earthquake ground motion and the resulting damage and loss functions lies the important area of structural response and the demand/capacity issues that determine the damage and loss. This topic, however, is probably worthy of a monograph in its own right.
I wish McGuire had spent more time explaining and discussing epistemic uncertainty and the importance of viable alternative models, concepts apparently impenetrable to most earth scientists and the source of much of the confusion mentioned three paragraphs ago. Viable alternative models simply means the set of all models that “fit” some data set within its aleatory variability (for example, ground-motion amplitudes as a function of distance and magnitude), without predicting exactly the same things, like the median values of the ground motion as a function of distance and magnitude. It is just an earth-sciences truism to say that more than one model, and often many different models, can fit the same set of data—or at least are not inconsistent with data that, in the earth sciences, are almost always inadequate, incomplete, and inconclusive. But these models are not the same, and the differences between and among them are a measure of epistemic uncertainty, our ignorance of how the earth really works—or at least that piece of the earth supposedly illuminated by the models. “Viable alternative models” have their intellectual origins in the philosophy of “multiple working hypotheses” first espoused by G. K. Gilbert and T. C. Chamberlin more than a century ago, the idea being that all hypotheses are to be considered until the observations themselves rule out one or more of them. And 120 years later, we still don't get it.
But none of these digressions materially detracts from what is a very good book on matters that are taking on increasing importance not only in the earthquake sciences and earthquake engineering, but also in a wide range of natural hazards and risks of interest to many different fields in the earth and engineering sciences. For those who are or would like to be professionals in the earthquake business and need to know more about seismic hazard and risk, I recommend this book.
Earthquake Engineering Research Institute, Oakland, 2004. Hardcover: ISBN 0-943198-01-1, $45