Search

1. Species assemblages of diatoms, rotifers, chydorids, planktonic crustaceans and chironomids were studied in 235 alpine lakes in the Alps, Pyrenees, Tatras (Western Carpathians), Retezat (Southern Carpathians) and Rila Mountains (Balkans). 2. For all taxonomic groups we found a hierarchical structure in the community assemblage using distinct scales of lake clustering (number of k-means groups) based on species composition similarity (Hellinger distance). We determined the optimal partition in assemblage types (i.e. number of lake clusters) for each taxonomic group by maximising the sum of the taxon indicative value (IndVal) and performed discriminant analyses, using environmental variables not conditioned by geographical patterns. Relevant environmental variables differed among and within taxonomic groups. Therefore the assemblages respond to a complex environmental mosaic, with the exception of diatom assemblages, which followed an acid–base gradient. 3. The significant environmental variables could be grouped into four general factors: lake size, tropho-dynamic status, acid–base balance and ice-cover duration (i.e., altitudinal gradient). Lake size was significant for the highest number of assemblage types; however, the most significant factor differed among taxonomic groups: acid–base balance for diatoms, lake size for rotifers, ice-cover duration for chydorids and planktonic crustaceans and tropho-dynamic status for chironomids. No single environmental typology accounted for the assemblage structure of all taxonomic groups. 4. However, defining ecological thresholds as values within environmental gradients at which the rate of change in assemblages is accelerated relative to points distant from that threshold, we were able to find specific threshold values for each of the four main general environmental factors identified, which were relevant across several taxonomic groups: 3 ha for lake area; 0.6 mg L−1 for dissolved organic carbon; 190 days for ice-cover duration and 200 μeq L−1 for acid neutralising capacity. Above and below these values ecosystem organisation change substantially. They have direct applications in establishing lake typologies for environmental quality and biodiversity conservation programmes, and in improving predictions about global change impacts. [ABSTRACT FROM AUTHOR]