Assessing the efficiency of connectivity measures with regard to the EU-Water Framework Directive in a Danube-tributary system

The effect of restoring connectivity for fish by the construction of 11 fish ladders in the Pielach and Melk rivers, both tributaries to the Danube in Austria, was monitored using electric fishing and fish traps between 1999 and 2004. In order to assess the efficiency of connectivity rehabilitation measures pre- and post-project data combining electric fishing and trap catch data were analyzed by means of three fish-based assessment methods: a MUlti-Level concept for a Fish-based, river-type-specific Assessment of ecological integrity (MULFA), the Fish Index Austria (FIA) and the European Fish Index (EFI). The effect of adding qualitative trap catch data to electric fishing data on metrics and indices was also tested and the magnitude of the effect was related to the distance of the sites from the river mouth. The results clearly demonstrated the significant contribution of connectivity rehabilitation measures to the ecological integrity of rivers like the River Pielach where morphological conditions are good, whereas remaining channelization still limits the success of connectivity measures in the River Melk. Trap catch data were found to represent an essential source of additional information to assess the efficiency of connectivity measures shortly after their implementation. The negative correlations of the magnitude of the effect of different indices and metrics with the distance of assessment sites from river mouths obviously underline the importance of the river Danube as a source for the re-colonization process. While the indices tested were found to have limited ability to reflect short-term response of fish assemblages to continuum rehabilitation, guild metrics were able to detect improvements of the ecological status shortly after the implementation of connectivity measures. Six metrics showed significant differences between pre and post-project data reflecting the expected increase of the ecological integrity: (1) Fish Region Index (FRI; FIA, MULFA), (2) number of subdominant species and (3) number of flow-guilds (FIA), (4) number of type specific species (MULFA), (5) number of benthic species and (6) number of potamodromous species (EFI); the FRI differences were only significant when trap catch data were added. The EFI indicated a decline of ecological integrity through increases in the density of omnivorous species and the relative number of tolerant species as well as a decrease in the relative number of intolerant species. Significantly de