Reintroduction of beaver (Castor spp) may facilitate rehabilitation of freshwater habitats providing a cost‐effective sustainable means of improving ecological conditions. Despite extensive research, debate and consultation, a general consensus on the impact of beaver on fishes has proven elusive because of variability in biological response. This paper provides a systematic review of the impacts of beaver dams on fishes and fish habitat based on a meta‐analysis of the literature and expert opinion. Research is regionally biased to North America (88%). The most frequently cited benefits of beaver dams were increased habitat heterogeneity, rearing and overwintering habitat and flow refuge, and invertebrate production. Impeded fish movement because of dams, siltation of spawning habitat and low oxygen levels in ponds were the most often cited negative impacts. Benefits (184) were cited more frequently than costs (119). Impacts were spatially and temporally variable and differed with species. The majority of 49 North American and European experts considered beaver to have an overall positive impact on fish populations, through their influence on abundance and productivity. Perceived negative effects related to the movement of aquatic organisms in tributary streams, including upstream and downstream migrating salmonids, and the availability of suitable spawning habitat.

Nature-based solutions are widely advocated for freshwater ecosystem conservation and restoration. As increasing amounts of river restoration are undertaken, the need to understand the ecological response to different measures and where measures are best applied becomes more pressing. It is essential that appraisal methods follow a sound scientific approach. Here, experienced restoration appraisal experts review current best practice and academic knowledge to make recommendations and provide guidance that will enable practitioners to gather and analyse meaningful data, using scientific rigor to appraise restoration success. What should be monitored depends on the river type and the type and scale of intervention. By understanding how habitats are likely to change we can anticipate what species, life stages, and communities are likely to be affected. Monitoring should therefore be integrated and include both environmental/habitat and biota assessments. A robust scientific approach to monitoring and appraisal is resource intensive. We recommend that appraisal efforts be directed to where they will provide the greatest evidence, including ‘flagship’ restoration schemes for detailed long-term monitoring. Such an approach will provide the evidence needed to understand which restoration measures work where and ensure that they can be applied with confidence elsewhere.

Based on a paired analysis, we describe a method for evaluating the potential of rivers with different physical characteristics to provide ecosystem services. Scores based on an extensive scientific literature review and expert opinion were applied to four sets of rivers in Scotland, with each pair comprising one river with a statutory nature conservation designation and one where such designations were largely absent. Data on physical habitat features and land cover were extracted manually from Google Earth™, based upon a previously published method expanded here to take account of cultural ecosystem services. Twenty physical habitat features and land‐cover types and 13 ecosystem services (four provisioning, three regulating, and six cultural) were used in the analysis. Notable developments on the earlier approach included the full integration of cultural ecosystem services alongside provisioning, regulating, and supporting services; introduction of confidence levels to river feature–ecosystem service linkages; and incorporation of valley floor surface area into one of the two scoring systems. Ecosystem scores for 500 m reaches along each river from source to mouth were calculated using Microsoft Excel, with results showing high reach‐to‐reach variability within individual rivers and significant differences between paired rivers. The four rivers with statutory nature conservation designations provided a greater range and typically higher ecosystem service scores than those with little or no designation, a result that has significant implications for river conservation and for framing catchment‐level conservation policy.

1. Freshwater pearl mussels (Margaritifera margaritifera) in the River Dee (north-east Scotland) are in decline, and now occur at low densities with poor recruitment levels. The population's deterioration has been attributed to reduced water quality, degraded river-bed habitat and pearl fishing but the specific importance of each of these impacts has not previously been investigated.2. This study focused on the impacts of artificial structures on the River Dee's freshwater pearl mussel population. Natural hydromorphological features were also included in the analysis. Relationships between the spatial distribution of the freshwater pearl mussel population and 31 explanatory variables were investigated using logistic regression. Three different model selection strategies were compared and contrasted.3. An absence of freshwater pearl mussels was found to be significantly associated with the following factors: (1) presence of bank protection; (2) proximity of the nearest upstream bridge; (3) presence of a wastewater treatment works; (4) number of upstream tributaries; (5) proximity of the nearest upstream tributary; and (6) presence of sediment deposition.4. The results strongly suggest that artificial structures on the main stem of the River Dee have contributed to the decline of its freshwater pearl mussel population. Hypotheses are proposed for the causal relationships involved. Requirements for further exploratory analyses, model testing, and experimentation are discussed.5. This work opens a new line of investigation into the causes of the decline of freshwater pearl mussel populations, providing formal support at a catchment scale for an impact previously identified only on a site-by-site basis.

No abstract