Numerical Ecology, 20
- P. Legendre, Département de Sciences Biologiques, Université de Montréal, H3C 3J7, Québec, Canada
- L. Legendre, Département de Biologie, Université Laval, Québec, G1K 7P4, Canada
The book describes and discusses the numerical methods which are successfully being used for analysing ecological data, using a clear and comprehensive approach. These methods are derived from the fields of mathematical physics, parametric and nonparametric statistics, information theory, numerical taxonomy, archaeology, psychometry, sociometry, econometry and others. Compared to the first edition of Numerical Ecology, this second edition includes three new chapters, dealing with the analysis of semiquantitative data, canonical analysis and spatial analysis. New sections have been added to almost all other chapters. There are sections listing available computer programs and packages at the end of several chapters. As in the previous English and French editions, there are numerous examples from the ecological literature, and the choice of methods is facilitated by several synoptic tables.View full description
For practicing ecologists, graduate students and professional researchers.
- Imprint: ELSEVIER
- ISBN: 978-0-444-89249-2
...The text is well written and very clear, it is so clear that the reader is seduced to read further; I found myself happily reading page after page. The authors are masters of their subject and as they delineate it, from chapter to chapter, they define it. At every corner of the text there is an 'ecological application' and this is useful in revealing the limits and advantages of a method.
D.H. Cushing, Journal of Plankton Research, Volume 21, Number 7
...The first English edition of Numerical Ecology appeared in 1983. It was already an important reference for multi-dimensional analysis in ecology, providing the most comprehensive explanations of matrix algebra, eigenanalysis, measures of association (similarity and dissimilarity), cluster analysis and ordination for ecology. The new edition is much more than this. The last 15 years have seen an explosion of new statistical non-parametric methods, particularly permutation methods, and computer programs that have made the analysis of 'misbehaving' data more possible than ever before.
This new edition of Numerical Ecology provides a very impressive overview of these complex new methods. It still provides an (up-dated) summary of the fundamentals of measures of association, clustering and ordination, but it also provides a plethora of new material, consistent with the recent explosion of new methods. It is comprehensive in the sheer quantity of the new methods it covers, such as cluster analysis with spatial contiguity constraints, multivariate Mantel tests for spatial autocorrelation, redundancy analysis and permutation tests for complex linear models. It is, without a doubt, the most extensive current review of the most up-to-date multivariate numerical techniques in experimental ecology.
Along with providing understandable introductions to and examples of these new methods, it is a great source to access the primary literature on these topics. This is especially important because many of the numerical methods useful to ecologists come from advances in other fields, such as agriculture, economics, psychology, archaeology, sociology, physics, geology, geography or from other branches of biology, such as genetics, taxonomy, morphometrics.... This book provides a superb line of access for those who approach complex statistics with a weary heart. Have you never understood an eigenvalue before? Look here! It also provides an exceptional up-to-date reference for the experts...This book offers ecologists the means to make the most of their numerical methods, particularly for descriptive multi-dimensional analysis. My prediction: it will take more than ten years for the rest of us to catch up to this book, to explore its information content and the many non-parametric approaches to unique situations. It is a 'must-have' reference for any researcher or graduate student studying multivariate ecological systems.M.J. Anderson, Journal of Experimental Marine Biology and Ecology, Volume 239
...Considerable appraisal has to be given on the extensiveness of the subjects treated, the clear presentations with i.e. helpful indications in the margin, clear figures and tables, all demonstrating the thoroughness with which the authors have prepared this second English edition. The fact that previous English and French editions, even if much less extensive than the second English edition, were published, shows that students and scientists around the world are interested in in-depth studies of numerical ecology.E.K. Duursma, Oceanologica Acta
...one of those few books that is worth every page it is written on. It is truly a monumental work, both in its size (853 pages) and thematic content. It is an ambitious and timely overview of the quantitative analysis of ecological data at a time when the increasing availability of computer programs makes choosing among the dizzying number of techniques a confusing task. Clearly, Pierre and Louis Legendre have produced an excellent book on numerical ecology.(...)The book is well produced, with few errors. It will be essential reading for ecologists, whether beginners or seasoned professionals, who conduct quantitative analyses of ecological and environmental data. This book can be used at several levels, from an introduction, to the in-depth teaching/reading of the topics covered.P. Bourgeron, Arctic, Antarctic, and Alpine Research, Vol. 32/2
...This volume made me wish that I taught a course for which it was the recommended textbook. It is a lucid guide to a particular area of quantitative biology. The book is devoted to numerical methods that can be used for the analysis of multidimensional data, particularly the sort of data collected in studies of assemblages of organisms. The outstanding feature of the book is the clarity with which these methods are described. As someone who has fought a losing battle with techniques such as the terrible twins Decorana and Twinspan, this book was a revelation. Anyone who is concerned with analysing fish assemblages and the relationships between assemblage composition and environmental characteristics should find this book valuable....It should sit alongside Underwood's Experiments in Ecology (CUP, 1997) on the desk of any ecologist who undertakes quantitative field studies.R.J. Wootton, Journal of Fish Biology
...The main message of this review is that this volume should not be absent from the bookshelf of any quantitatively minded community ecologist.J. Podani, Journal of Classification
Table of ContentsChapter headings and selected parts: Preface. Complex Ecological Data Sets. Numerical analysis of ecological data. Statistical testing by permutation. Ecological descriptors. Matrix Algebra: A Summary. The ecological data matrix. Vectors and scaling. Eigenvalues and eigenvectors. Dimensional Analysis in Ecology. Fundamental principles and the Pi theorem. Scale factors and models. Multidimensional Quantitative Data. Multidimensional variables and dispersion matrix. Multinormal distribution. Tests of normality and multinormality. Multidimensional Semiquantitative data. Nonparametric statistics. Quantitative, semiquantitative, and qualitative multivariates. Multidimensional Qualitative Data. Multiway contingency tables. Species diversity. Ecological Resemblance. The basis for clustering and ordination. Association coefficients. R mode: coefficients of dependence. Cluster Analysis. The basic model: single linkage clustering. Cophenetic matrix and ultrametric property. Hierarchical divisive clustering. Ordination in Reduced Space. Projecting data sets in a few dimensions. Principal component analysis (PCA). Nonmetric multidimensional scaling (MDS). Interpretation of Ecological Structures. Ecological structures. The mathematics of ecological interpretation. The 4th-corner problem. Canonical Analysis. Redundancy analysis (RDA). Canonical correspondence analysis (CCA). Canonical analysis of species data. Ecological Data Series. Characteristics of data series and research objectives. Trend extraction and numerical filters. Periodic variabilty: spectral analysis. Detection of discontinuities on multivariate series. Box-Jenkins models. Spatial Analysis. Unconstrained and constrained ordination maps. Causal modelling: partial canonical analysis. Causal modelling: partial Mantel analysis. Bibliography. Tables. Subject index.