Could artificial structures such as fish passes facilitate the establishment and spread of the ‘killer shrimp’ Dikerogammarus villosus (Crustacea: Amphipoda) in river systems?

Linkages between habitat structure, invasibility and direct species interactions were examined in a Central European river system, the Schipbeek in the Netherlands, invaded by the ‘killer shrimp’ amphipod Dikerogammarus villosus . The invader's arrival in other fresh waters is linked to declines in macroinvertebrate biodiversity, with previous studies indicating that the invader's impacts may be mediated by habitat heterogeneity and substrate type. Using an extensive 50 km field survey, potential habitat factors influencing distribution were identified. Of 12 sites surveyed, two contained fish ‘passes’ (instream sections creating faster flows via concrete structures covered in cemented rocks). It was only in these sites that D. villosus was found and then only in the fish pass areas, as opposed to the surrounding river, which, like the remaining 10 sites of the survey, was characterized by sluggish flows and a muddy/silty substrate. Gammarus roeselii was the numerically dominant amphipod in these latter areas, with no D. villosus being detected. A repeated survey more than 2 years later, revealed the same distribution pattern. Within both fish pass areas, D. villosus relative abundance was highest in faster flowing, less macrophyte dense patches, with the opposite true for G. roeselii . D. villosus numerical abundance decreased with increasing macrophyte cover in both passes. A transplant experiment indicated high survivorship of D. villosus outside fish pass areas and showed that slow‐flowing watercourses with mud/silt substrates, previously deemed ‘unsuitable’, are not effective barriers to D. villosus spreading to more ‘suitable’, faster, rocky substrate watercourses. We recommend that when the location and cost‐benefit analysis of instream rehabilitation structures, such as fish passes, is being considered, attention must be paid to the risk of hastening the spread of this invader. We also recommend that when monitoring programmes are designed for D. villosus in newly invaded regions, instream artificial structures should be included as part of the survey. Copyright © 2013 John Wiley & Sons, Ltd.