Abstract
Macrofaunal and environmental data were collected from 97 sampling sites in ditches. In total 28 oligochaete taxa were found, 22 of which were determined to species level and further processed. About 33 environmental variables were measured at each site. The resulting multivariate data set is described by three techniques, clustering (FLEXCLUS), ordination (CANOCO) and log linear regression (exponential and Gaussian). In this way a detailed description of environmental preferences of the dominant species collected is given and also the possible species combinations which can be found under certain environmental circumstances. Some advantages and disadvantages of the multivariate analysis techniques are reported. Detrended canonical correspondence analysis is a promising technique, especially in combination with clustering and regression analysis.
Original language | English |
---|---|
Pages (from-to) | 283-292 |
Number of pages | 10 |
Journal | Hydrobiologia |
Volume | 155 |
Issue number | 1 |
DOIs | |
Publication status | Published - 1 Dec 1987 |
Keywords
- clustering
- ditches
- multivariate analysis
- Oligochaetes
- ordination
- regression