A comparative analysis of restoration measures and their effects on hydromorphology and benthic invertebrates in 26 central and southern European rivers

P>1. Hydromorphological river restoration usually leads to habitat diversification, but the effects on benthic invertebrates, which are frequently used to assess river ecological status, are minor. We compared the effects of river restoration on morphology and benthic invertebrates by investigating 26 pairs of non-restored and restored sections of rivers in Austria, Czech Republic, Germany, Italy and the Netherlands. 2. Sites were grouped according to (1) region: central Europe vs. southern Europe; (2) river type: mountain vs. lowland rivers; (3) restoration approach: active vs. passive restoration and (4) a combination of these parameters. All sites were sampled according to the same field protocol comprising hydromorphological surveys of river and floodplain mesohabitats, microhabitats at the river bottom and habitat-specific sampling of benthic invertebrates. Restoration effects were compared using Shannon-Wiener Indices (SWIs) of mesohabitats, microhabitats and invertebrate communities. Differences in metric values between non-restored and restored sites were compared for 16 metrics that evaluated hydromorphology and the benthic invertebrate community. 3. Mean SWIs differed for both mesohabitats (1 center dot 1 non-restored, 1 center dot 7 restored) and microhabitats (1 center dot 0 non-restored, 1 center dot 3 restored), while SWIs for invertebrate communities were not significantly different (2 center dot 4 non-restored, 2 center dot 3 restored). Meso- and microhabitat metrics in the restored sections were usually higher compared with the non-restored sections, but the effects on invertebrate metrics were negligible. 4. Measures in southern Europe and mountainous regions yielded larger differences between non-restored and restored sections of rivers. Differences in the meso- and microhabitat metrics were largest for actively restored sections of central European mountain rivers and rivers from southern Europe, followed by passively restored mountain rivers in central Europe. The smallest differences were observed for lowland sites. There was no significant restoration effect on invertebrate metrics in any categories. 5. Synthesis and applications. Restoration measures addressing relatively short river sections (several hundred metres) are successful in terms of improving habitat diversity of the river and its floodplain. Active restoration measures are suitable if short-term changes in hydromorphology are desired. To realize changes in benthic invertebrate community composition, habitat restoration within a small stretch is generally not sufficient. We conclude that restoring habitat on a larger scale, using more comprehensive measures and tackling catchment-wide problems (e.g. water quality, source populations) are required for a recovery of the invertebrate community.