Statistics for Environmental Science and Management.

Statistics for Environmental Science and Management, by Bryan F. J. MANLY, Boca Raton, FL: Chapman and Hall/CRC, 2001, ISBN 1-58488-029-5, x + 326 pp., $49.95.

As manager for statistics technology support across a giant corporation, I have a number of distinct clienteles. One is our

cadre of environmental scientists, who work both in centralized groups and embedded within businesses. These people frequently have statistics applications, usually have a some basic statistics class in college, and generally are totally unprepared for the unique aspect of the statistical data in environmental science. Manly wrote this book for these people.

A former faculty member who has written books for environmental science most recently Manly (1997), reported in Ziegel (1998), and now a statistician for an environmental consulting company, the author knows the needs of environmental scientists. He also knows how to effectively present methodology to his audience. Presuming some previous basic statistics course, the author jumps right into sampling after the usual introductory chapter for motivation by examples. Focused more on traditional sampling than the excellent book by Thompson (1992), the chapter does not include sampling for spatial data. Next is a chapter on various types of statistical models for data: discrete distributions, continuous distributions, linear regression, analysis of variance (ANOVA), and generalized linear models.

The subsequent four chapters approach statistical methodology from the perspective of the environmental scientist. The first of these, "Drawing Conclusions from Data," discusses types of studies, types of inference, significance tests, confidence intervals, randomization tests, bootstrapping, multiple testing, meta-analysis, and Bayesian inference, all without much illustration. Next comes "Environmental Monitoring," which begins with some examples of spatial designs. Methods presented for detecting change include ANOVA. Shewhart control charts, and cumulative sum (CUSUM) charts. This chapter is mostly illustrations. The next chapter, "Impact Assessment," focuses on before-after, impact-control, and impact-gradient designs. The last chapter in this group, "Assessing Site Reclamation," has bioequivalence as its primary topic. Here bioequivalence refers to biological equivalence after a site is impacted.

The book's last four chapters present special statistical tools for environmental data that lie outside the scope of an introductory statistics course. First is "Time Series Analysis," which focuses primarily on serial correlation versus randomness and the detection of change points and trends. Additional topics include autoregressive integrated moving average models, frequency domain analysis, and forecasting. More material is needed on seasonality. Next comes "Spatial Data Analysis," which focuses mostly on quadrat counts and point patterns but concludes with several sections on geostatistics. "Censored Data" covers the important issue of nondetection in chemicals measurement. Last is a chapter on "Monte Carlo Risk Assessment."

Overall, this book has a huge variety of topics and numerous examples in most of the chapters. There are excellent references throughout. There is some discussion and even illustration of statistical computing. Hopefully, the second edition will incorporate more computing, but readers can use the book by Millard (2001), reviewed by Lumley (2001), as a guide. This book is a great and inexpensive library addition for statisticians and environmental scientists who analyze environmental data.

REFERENCES

Lumley, T. (2001), Review of Environmental Statistics With S-PLUS by S. Millard and N. Neerchal, Technometrics, 43, 495.

Manly, B. (1997), Randomization, Bootstrap and Monte Carlo Methods in Biology (2nd ed.), Boca Raton, FL: Chapman and Hall.

Millard, S., and Neerchal, N. (2001), Environmental Statistics With S-PLUS, Boca Raton, FL: CRC.

Thompson, S. (1992), Sampling, New York: Wiley.

Ziegel, E. (1998), Editor's Report for Randomization, Bootstrap and Monte Carlo Methods in Biology (2nd ed.) by B. Manly, Technometrics, 40, 84.