Relating Fish Assemblage Variability To Environmental Gradients In Small And Mid-Sized Streams Of The Elbe Basin (Czech Republic)


Martin Humpl


Institute for Environmental Protection, Faculty of Science, Charles University in Prague, Benátská 2, CZ-128 01 Praha 2, Czech Republic



Abstract: Fish assemblages and environmental variables were studied in 567 sites in small and mid-sized streams of the Elbe basin. A presence-absence data of 28 fish species were analyzed to characterize ichthyocenoses in diverse environments. Several multivariate techniques were used: 1. Detrended correspondence analysis (DCA) for examination of species occurrence changes in the longitudinal profiles in relation to “fish zones”; 2. Partial Canonical correspondence analyses (CCA), and Cluster analysis for the assessment of the fish assemblages’ regional and temporal variability; 3. Direct gradient CCA for investigation of the influence of environmental variables on fish assemblages, and partial CCA’s for interpretation of the most important fish species-enviroment relations.

 The biggest proportion of variability in species composition (13%) was explained by the regional variability (CCA). Major environmental factors determining species distributions (the number of ponds in the catchment, effect of higher order stream, the distance from the source, stream slope, elevation, and stream order) were used for characterization of the environmental influence. All these six variables together explained another 9% of the total variability. The strongest gradients of change in fish assemblages revealed to inversely related distance from the source and stream slope. The second gradient was based on the number of ponds and/or on the effect of higher order stream. The patterns of fish assemblage composition and the associated fish species-environment relations provided a framework for developing an ecologically meaningful classification of small and mid-sized stream sites in the Elbe basin.


Key words:  fish assemblages variability; environmental factors; small and mid-sized streams; detrended correspondence analysis; canonical correspondence analysis


Full text (PDF)