Quantitative Ecotoxicology

Quantitative Ecotoxicology, Second Edition explores models and methods of quantitative ecotoxicology at progressively higher biological scales using worked examples and common software packages. It complements the author’s previous books, Fundamentals of Ecotoxicology, Third Edition and Ecotoxicology: A Comprehensive Treatment. Encouraging a more rigorous inferential approach to research, the book examines the quantitative features of the science of ecotoxicology. The first chapters lay the foundation by introducing fundamental concepts and definitions. The author traces the historical perspective, rationale, and characteristics of scientific ecotoxicology as well as the general measurement process. He also considers methodologies for defining and controlling variance, which could otherwise exclude valid conclusions from ecotoxicological endeavors. The book then discusses ecotoxicological concepts at increasing levels of ecological organization and outlines quantitative methods used to measure toxicant accumulation and effects. Reflecting the importance of establishing type I and type II error rates, it highlights design issues, particularly sample size and power estimation. The final chapter summarizes the book with a brief discussion of ecotoxicology from a nonregulatory perspective. Extensively updated, this second edition has been expanded to include terrestrial as well as aquatic ecotoxicology. Requiring only a basic knowledge of statistics, this highly readable book is suitable for graduate students and researchers as well as practicing environmental scientists and engineers. It guides readers to better understand the fate and effects of toxicants in the biosphere—and helps them frame this understanding in quantitative terms. What’s New in This Edition More than 40 new figures and 20 new worked examples Updated measurement quality methods and software Expanded coverage of synecological models and methods More integration of Bayesian concepts Appendices for power analysis and basic matrix methods Additional mixture toxicity and up-and-down methods Greatly expanded discussion of significance testing Expanded discussion of metapopulations Matrix tools for population demography Light isotope-based models for trophic transfer of toxicants Inclusion of metacommunity and SHE analysis techniques R script examples by Eduard Szöcs (University Koblenz-Landau) available at http://edild.github.io/blog/categories/quantitative-ecotoxicology-with-r/

IntroductionEcotoxicology as a Scientific DisciplineToxicants and BiosphereToxicant Effects in EcosystemsToxicant Fate in EcosystemsOrganization of Knowledge Based on Explanatory PrinciplesBayesian InferenceToward Strongest Possible Inference and Clear Ecological RelevanceReferencesThe Measurement ProcessGeneralRegions of QuantitationBlank CorrectionAccuracy and PrecisionVariance StructureSample SizeOutliersSummaryReferencesBioaccumulationGeneral Modeling Bioaccumulation: General ApproachModeling Bioaccumulation: Alternative ApproachesIntrinsic Factors Affecting BioaccumulationSummaryReferencesLethal and Other Quantal Responses to StressGeneralDose-Response at a Set EndpointTime to DeathQuantifying the Effects of Extrinsic FactorsQuantifying Effects of Intrinsic FactorsToxicant MixturesSummaryReferencesStatistical Tests for Detection of Chronic Lethal and Sublethal StressGeneralMethod SelectionOne-Way Analysis of VarianceTest of Normality: Shapiro–Wilk’s TestTest for Homogeneity of Variances: Bartlett’s TestTreatment Means Compared to the Control MeanMonotonic Trend: Williams’s TestSteel’s Multiple Treatment-Control Rank Sum TestWilcoxon Rank Sum Test with Bonferroni’s AdjustmentA Second Look at Statistical TestingInferring Biological Significance from Statistical SignificanceSummaryReferencesPopulation and Metapopulation EffectsGeneralEpidemiologyPopulation SizeDemographySpatial Distribution of IndividualsPopulation GeneticsSummaryReferencesCommunity EffectsGeneralSimple Species InteractionsCommunity Structure and FunctionComposite IndicesMetacommunitiesTrophic ExchangeSummaryReferencesSummaryApplicationFacilitating Growth of the ScienceReferences Appendices Appendix 1: Factors for Estimating Standard Deviation and Control Limits for Range Appendix 2: One-Sample Tolerance Probability Comparisons between nm* and nm Appendix 3: Critical Values of T Used to Test for Single Outliers (One-Sided Test) Appendix 4: Critical Values for ? Used to Test for Multiple Outliers (a = 0.05) Appendix 5: Response Metameters for Proportion Affected Appendix 6: Maximum Likelihood Values for Dixon’s Up-and-Down Method Appendix 7: E Values Used to Estimate 95% Confidence Intervals for LT50 with the Litchfield Method Appendix 8: Coefficients (an–i+1) for Shapiro–Wilk’s Test for Normality Appendix 9: Percentage Points of Shapiro–Wilk’s W Test for Normality Appendix 10: Dunnett’s t for One-Sided Comparisons between p Treatment Means and a Control for a = 0.05 Appendix 11: Dunnett’s t for Two-Sided Comparisons between p Treatment Means and a Control for a = 0.05 Appendix 12: Bonferroni’s Adjusted t Values for One-Sided Test and a = 0.01 Appendix 13: Bonferroni’s Adjusted t Values for One-Sided Test and a = 0.05 Appendix 14: Bonferroni’s Adjusted t Values for Two-Sided Test and a = 0.01 Appendix 15: Bonferroni’s Adjusted t Values for Two-Sided Test and a = 0.05 Appendix 16: Dunn–Šidák’s t for Comparisons between p Treatment Means and a Control for a = 0.01, 0.05, 0.10, and 0.20 (One-Sided Test) Appendix 17: Dunn–Šidák’s t for Comparisons between p Treatment Means and a Control for a = 0.01, 0.05, 0.10, and 0.20 (Two-Sided Test) Appendix 18: Williams’s ti,a for w = 1 and Extrapolation ßt (Superscript) for a One-Sided Test and a = 0.01 Appendix 19: Williams’s ti,a for w = 1 and Extrapolation ßt (Superscript) for a One-Sided Test and a = 0.05 Appendix 20: Williams’s ti,a for w = 1 and Extrapolation ßt (Superscript) for a Two-Sided Test and a = 0.01 Appendix 21: Williams’s ti,a for w = 1 and Extrapolation ßt (Superscript) for a Two-Sided Test and a = 0.05 Appendix 22: Significant Values of Steel’s Rank Sums for a One-Sided Test with a = 0.05 or 0.01 Appendix 23: Significant Values of Steel’s Rank Sums for a Two-Sided Test with a = 0.05 or 0.01 Appendix 24: Wilcoxon (Mann–Whitney) Rank-Sum Test Critical Values with Bonferroni’s Adjustments: One-Sided Test and a = 0.05 Appendix 25: Wilcoxon Rank-Sum Test Critical Values with Bonferroni’s Adjustments: Two-Sided Test and a = 0.05 Appendix 26: SAS Code for Implementing the Jonckheere–Terpstra Test Appendix 27: Balancing a, ß, and Effect Size (ES) Appendix 28: Basic Matrix Methods Appendix 29: Values of ? Used for Maximum Likelihood Estimation of Mean and Standard Deviation of Truncated Data Index

Michael C. Newman is currently the A. Marshall Acuff Jr. Professor of Marine Science at the College of William and Mary’s Virginia Institute of Marine Science, where he also served as Dean of Graduate Studies for the School of Marine Sciences from 1999 to 2002. His research interests include quantitative ecotoxicology, environmental statistics, risk assessment, population effects of contaminants, metal chemistry and effects, and bioaccumulation and biomagnification modeling. In addition to more than 125 articles, he has authored five books and edited another six on these topics. The English edition and Mandarin and Turkish translations of Fundamentals of Ecotoxicology have been adopted widely as the textbook for introductory ecotoxicology courses.