Statistics for Earth and Environmental Scientists

The goal of this book is to present skills and knowledge of important statistical concepts through data analytic tools and then apply them to real world problems in the environmental sciences. Since there is no single approach that works best in all research circumstances, the authors introduce models using the frequentist approach, but also discuss Bayesian, nonparametric, and computer intensive methods. The book begins with an introduction to types of data, evaluation of data, modeling and estimation, random variation, sampling all of which are explored in the context of case studies, which use real data from earth science applications. Subsequent chapters focus on general principles of modeling (namely formulation, estimation, and evaluation), interval estimation and hypothesis testing of means; regression; methods for analyzing time series data; spatial statistics; multivariate analysis; discrete distributions; and experimental design. Most statistical models are introduced by concept and application, given as equations, and then accompanied by heuristic justification rather than a formal proof. In addition, graphs throughout the book illustrate concepts, identify unusual observations, and assist in model evaluation. Each chapter concludes with exercises that are designed to elicit understanding of the processes of data analysis, model building, and statistical inference. The authors encourage readers to collect their own data and incorporate them into these chapter problems, facilitating an intricate and hands-on understanding of the presented concepts. Solutions to most problems are available on the book's related Web site, along with additional data sets and R-code for most graphs and analyses.

John H. Schuenemeyer, PhD, is President of Southwest Statistical Consulting, LLC and Professor Emeritus of Statistics, Geography, and Geology at the University of Delaware. A Fellow of the American Statistical Association, Dr. Schuenemeyer has more than thirty years of academic and consulting experience and was the recipient of the 2004 John Cedric Griffiths Teaching Award, awarded by the International Association for Mathematical Geosciences. Lawrence J. Drew, PhD, is Research Scientist at the U.S. Geological Survey. Dr. Drew has published more than 200 scientific papers on the role of quantitative methods in petroleum and mineral resource assessment, and he is currently is working on an analysis of environmental data. Dr. Drew is the winner of the 2005 Krumbein Medal, awarded by the International Association for Mathematical Geosciences.