Global consequences of dam-induced river fragmentation on diadromous migrants: a systematic review and meta-analysis.

Abstract

The global proliferation of dams has altered flow and sediment regimes in rivers, presenting a major threat to freshwater biodiversity. Diadromous species, such as fishes, decapod crustaceans and gastropods, are particularly susceptible to fragmentation because dams obstruct their breeding migrations between coastal waters and rivers. Although dams have contributed to significant declines in abundance of some commercially important diadromous fishes (salmonids and anguillids) and Macrobrachium shrimps, understanding of the impacts of fragmentation on the majority of diadromous animals is limited. Moreover, the number of species known to have diadromous life cycles has risen substantially during the last four decades, from ~250 to more than 800. This synthesis aims to consolidate the global impacts of fragmentation on diadromous animals and highlight potential knowledge gaps. We identified 338 publications documenting the impacts of dams on diadromous fishes and decapods, but this was reduced to 65 publications after application of our strict selection criteria. Specifically, we only included studies that compared unfragmented (e.g. undammed) or restored (e.g. dams with fish passes) with fragmented (e.g. site above dams) rivers. To assess statistical significance, the results of studies that were replicated sufficiently to enable calculation of standardised effect sizes were also subject to meta-analysis focusing on three topics: impacts of dam-induced fragmentation; efficacy of fish passes; and the mitigative potential of dam removal. Study outcomes were evaluated from five key variables: abundance; species richness; assemblage composition; population genetic diversity; and population genetic structure. We found that fragmentation led to net negative effects across all key variables for diadromous fishes. Fishes with limited jumping or climbing ability and obligate diadromous migrants that cannot persist as landlocked populations were more threatened by fragmentation. However, fishes that were capable climbers or jumpers and facultatively diadromous were nonetheless susceptible to impacts, particularly in their abundance and gene flow between fragmented populations. Installation of fish passes did not lead to positive outcomes, whereas dam removal was effective in restoring connectivity for fishes, suggesting that it is a more effective, albeit potentially contentious, approach (e.g. the dam may serve an important societal need), for restoring habitat connectivity. A smaller number of publications investigated diadromous decapods (seven versus 61 on fishes), and our synthesis of their findings suggests that decapods were vulnerable to habitat alteration by dams, but were less sensitive to their barrier effects because they were better climbers than fishes. Gastropods were the least studied diadromous taxon, and none met our criteria for systematic review or meta-analysis. The imbalance in information about diadromous taxa was compounded by a scarcity of studies from the tropics, particularly in South America, Africa, South Asia, Southeast Asia, and East Asia. These regions support diverse aquatic assemblages so the impacts of dams may be underestimated given existing knowledge gaps. The conservation of diadromous migrants would be best served by avoiding the construction of dams while improving mitigation strategies, such as fish passage design, to limit the most damaging effects of river fragmentation.