Ordination procedure

The term ordinations encompasses many mathematical practices that aim to graphically illustrate certain data in a coordinate system .

These are z. B. used in ecology or vegetation science to compare different vegetation images and to illustrate their differences along one or more gradients . To arrange the data, an n-dimensional space is initially assumed (the number of dimensions corresponds in the example to the number of plant species found). A so-called dimension reduction is carried out using mathematical processes, so that the position of the data (vegetation images) can be represented in a two- or three-dimensional coordinate system. The axes of the coordinate system should represent gradients (or summarize several gradients; here e.g. water and nutrient content of the soil) that explain as much of the variance as possible . The environmental variables can be analyzed directly or deduced indirectly from the occurrence of the plant species.

Examples of ordination procedures are:

• CA - Correspondence Analysis , Correspondence Analysis
• CCA - Canonical Correspondence Analysis , Canonical Correspondence Analysis
• DCA - Detrended Correspondence Analysis
• PCA - Principal Component Analysis , Principal Component Analysis
• NMDS - Nonmetric multidimensional scaling , non-metric multidimensional scaling

swell

• Leyer, I. & K. Wesche (2007) Multivariate Statistics in Ecology: An Introduction. - Berlin (Springer)
• Palmer, M. Ordination Methods for Ecologists .