Altogether 71 samples collected in 1968–2004 from the Rokytná and Loučka rivers were used to estimate proportional fish species richness from single-pass electro-fishing and probabilities of detection for individual fish species. Mean estimated species richness from single-pass sampling (p ́s1) was 95.4% and 96.2% of estimated total species richness for Rokytná and Loučka rivers, respectively. However, p ́s1 values for individual sites ranged from 71.4% to 100% of estimated total species richness. Additional species unique to the second pass were collected in 38% (Rokytná) and 22% (Loučka) of the samples. When assessing the fish assemblage structure by a multivariate technique (PCA), considerable inaccuracy between the single-pass and two-pass electro-fishing data was also observed. Decisions regarding standardized sampling effort and whether to conduct one or more electro-fishing passes must be based on study objectives. For more accurate evaluations of fish assemblage structure two-pass electro-fishing is recommended. Three-pass electro-fishing did not produce significant increase of species richness.
In 1975–2005, we utilized data on water quality (N-NH4+, O2, Cl-) and number of carp caught (CC) as environmental data, and sport fishery statistics as species data, fromthree fishing stretches of the Berounka River, Czech Republic. An indirect gradient analysis (DCA) for calculating the strongest gradient affecting the species data, and direct linear gradient analysis (RDA) were used for assessing the relationships among environmental and species data. All RDA models were significant and consecutively explained 45.5, 39.6, and 39.7% of the variability of species data. The strength of environmental variables was evaluated by partial RDA. In the first stretch below the City of Plzeň, with the poor water quality and low carp yields, all environmental variables were significant and the CC data alone explained 26.2% variability of species data. In the second stretch, with intermediate water quality and carps yields, only two environmental variables (O2 and Cl-) were significant and in the third stretch near the city of Prague, with the same water quality and exceptionally high yields, the environmental variables did not significantly influence the species data. These results indicate that sport fishery statistics can be used as at least one source of information for studies of fish assemblages in streams where other data is lacking.