《Data Analysis in Vegetation Ecology》 by Otto Wildi



Contents
1 Introduction
2 Patterns in Vegetation Ecology
  2.1 Pattern recognition
  2.2 Interpretation of patterns
  2.3 Sampling for pattern recognition
  2.3.1 Getting a sample
  2.3.2 Organizing the data
3 Transformation
  3.1 Data types
  3.2 Scalar transformation and the species enigma
  3.3 Vector transformation
  3.4 Example: Transformation of plant cover data
4 Multivariate Comparison
  4.1 Resemblance in multivariate space
  4.2 Geometric approach
  4.3 Contingency testing
  4.4 Product moments
  4.5 The resemblance matrix
  4.6 Assessing the quality of classifications
5 Ordination
  5.1 Why ordination?
  5.2 Principal component analysis (PCA)
  5.3 Principal coordinates analysis (PCOA)
  5.4 Correspondence analysis (CA)
  5.5 The horseshoe or arch effect
  5.5.1 Origin and remedies
  5.5.2 Comparing DCA, FSPA and NMDS
  5.6 Ranking by orthogonal components
  5.6.1 Method
  5.6.2 A numerical example
  5.6.3 A sampling design based on RANK (example)
6 Classification
  6.1 Group structures
  6.2 Linkage clustering
  6.3 Minimum-variance clustering
  6.4 Average-linkage clustering: UPGMA, WPGMA, UPGMC and WPGMC
  6.5 Forming groups
  6.6 Structured synoptic tables
  6.6.1 The aim of ordering tables
  6.6.2 Steps involved
  6.6.3 Example: Ordering Ellenberg’s data
7 Joining Ecological Patterns
  7.1 Pattern and ecological response
  7.2 Analysis of variance
  7.2.1 Variance testing
  7.2.2 Variance ranking
  7.2.3 How to weight cover abundance (example)
  7.3 Correlating resemblance matrices
  7.3.1 The Mantel test
  7.3.2 Correlograms: Moran’s I
  7.3.3 Spatial dependence: Schlaenggli data revisited
  7.4 Contingency tables
  7.5 Constrained ordination
8 Static Explanatory Modelling
  8.1 Predictive or explanatory?
  8.2 The Bayes probability model
  8.2.1 The discrete model
  8.2.2 The continuous model
  8.3 Predicting wetland vegetation (example)
9 Assessing Vegetation Change in Time
  9.1 Coping with time
  9.2 Rate of change and trend
  9.3 Markov models
  9.4 Space-for-time substitution
  9.4.1 Principle and method
  9.4.2 The Swiss National Park succession (example)
  9.5 Dynamics in pollen diagrams (example)
10 Dynamic Modelling
  10.1 Simulating time processes
  10.2 Including space processes
  10.3 Processes in the Swiss National Park (SNP)
  10.3.1 The temporal model
  10.3.2 The spatial model
  10.3.3 Simulation results
11 Large Data Sets: Wetland Patterns
  11.1 Large data sets differ
  11.2 Phytosociology revisited
  11.3 Suppressing outliers
  11.4 Replacing species with new attributes
  11.5 Large synoptic tables?
12 Swiss Forests: A Case Study
  12.1 Aim of the study
  12.2 Structure of the data set
  12.3 Methods
  12.4 Selected questions
  12.4.1 Is the similarity pattern discrete or continuous?
  12.4.2 Is there a scale effect from plot size?
  12.4.3 Does the vegetation pattern reflect the environmental conditions?
  12.4.4 Is tree species distribution man-made?
  12.4.5 Is the tree species pattern expected to change?
  12.5 Conclusions

Appendix A On Using Software
  A.1 Spreadsheets
  A.2 Databases
  A.3 Software for multivariate analysis
Appendix B Data Sets Used
References
Index