Community organization of larval chironomid (Diptera) assemblages and environmental factors in a large river: prediction and validation of their interrelationships after three years
Fesl, Christian; Humpesch, Uwe H.
Assemblages of larval chironomids in a free-flowing section of the River Danube, east of Vienna, were examined over three years. Relationships between community attributes and environmental factors in the first year were used to predict assemblage characteristics for the subsequent two years. The accuracy of these predictions was then tested against the actual data obtained for the second and third years. Habitat stability and water velocity were the strongest predictors for total abundances (52 % of the variance in observed values explained by the predicted values), followed by community persistence (30 %, predicted from water temperature) and species aggregation (25 %, habitat stability and heterogeneity). Mean spatial resource width, evenness and ß-diversity accounted for 14 % predictability each, with again habitat stability as the dominant explanatory variable. In contrast, predictability was much higher among community attributes which accounted for up to 80 %. Since the slopes of the relationships were not significantly different between the years, deviations from the predictions were mainly due to differences in absolute values of community attributes. Comparisons of multivariate patterns in these relationships, obtained from Redundancy analyses by means of Procrustes analyses, showed no significant differences between the years. However, univariate and multivariate tests suggested a marked annual change in community composition, mainly with respect to patterns in dominance structure due to the replacement of the most abundant species. On the basis of discriminant functions obtained from environmental factors in the first year, 90 % of all samples from the two subsequent years were correctly classified to the four sampling sites, and correct classification on the basis of community attributes accounted for 52 % where evenness provided the highest discriminatory significance. It is concluded that most relationships, especially those among community attributes, were consistent for the three years. Because shifts in abundant species were reflected neither by total abundances nor by species richness, estimation of dominance structure has to be an essential tool for management purposes. Long-term investigations like this provide valuable information on the base-line variation of natural communities in terms of a data-basis for testable ecological theories and management strategies.