Abstract

This study explores the environmental factors underlying variation in abundance of common and rare freshwater taxa. Hybrid multidimensional scaling is used to model variation in distribution and abundance of freshwater microinvertebrate taxa over 17 sample sites in the upper catchment of the LaTrobe River, Victoria, Australia. Initial analysis of 40 common taxa revealed high correlations of the ordination space with physico-chemical variables related to temperature, stream order, particle size and water chemistry. Analysis of all 269 taxa, or alternatively of the 229 rarer taxa alone, resulted in ordination spaces that showed high correlations for additional physico-chemical variables, particularly relating to water chemistry. Monte Carlo significance tests supported this finding in demonstrating that the analysis of all taxa produced a greater number of significant correlations between the ordination space and physico-chemical variables. The additional important environmental correlates revealed by the analysis of the rare taxa suggested that there might be differences in the set of environmental variables that are related to patterns of distribution and abundance of rare versus common taxa. A Monte Carlo test was carried out to test the null hypothesis that the failure to recover some environmental correlates in the analysis of common taxa simply resulted from the small (40) number of taxa involved. Results of the test generally showed that rareness versus commonness could not be implicated in the greater recovery of these water chemistry variables in the analysis of the rare taxa. The recovery of additional environmental correlates with the inclusion of rare taxa has implications for conservation studies at the community level. Ordination can be used for survey extension where complete information on distribution and abundance of taxa is unavailable. The ability of ordination methods to summarise distribution and abundance of rare taxa, and incorporate their additional information on environmental variation, suggests that representativeness of the ordination space is a useful criterion for reserve selection.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call