Field test of a bubble curtain deterrent system for common carp

Reducing the movement of invasive fish such as the common carp, Cyprinus carpio L., silver carp, Hypophthalmicthys molitrix (Val.), and bighead carp, Hypophthalmicthys nobilis (Richardson), is one of the greatest challenges fisheries managers are presently facing (Taylor et al. 2003, 2005; Ruebush et al. 2012). Because physical barriers (dams, screens) and electrical barriers are non-selective, can be prohibitively expensive and impractical to maintain, there has been an increasing interest in barriers that use behavioural deterrents (i.e. sound, light and bubbles) as they have the potential to be safer, inexpensive and species specific (Popper & Carlson 1998; Taylor et al. 2003, 2005; Noatch & Suski 2012; Ruebush et al. 2012). As behavioural deterrents do not provide a physical obstacle to movement, they are <100% effective. Accordingly, whether these deterrents can be useful depends on their relative efficacy vs cost relative to management goals (i.e. controlling an established population vs blocking a new invasion). This study focused on bubble curtains, a deterrent technology that uses a dense plume of (noisy) bubbles to repel fish with acute hearing (Zielinski et al. 2014). These systems are inexpensive, but their performance has not been well documented in the field for invasive carps (but see Ruebush et al. 2012). In a previous laboratory study, an optimised bubble curtain deterrent system was found to reduce the movement of small common carp [total length: 259 29 mm (mean SD)] by 75–80% (Zielinski et al. 2014). The objective of this study was to test the efficacy of this bubble curtain system in the field. Cyprinus carpio is a highly invasive fish from Eurasia (Balon 1995), which is already well established and implicated in degrading millions of acres of shallow lake and wetland ecosystems across the globe (Weber & Brown 2009). In Midwestern North America, C. carpio abundance appears to be attributable to the tendency of adult fish to leave lakes and use wetlands for spawning, which often lack native predators of juvenile carp (Bajer & Sorensen 2010). A key to the long-term control of this species could lie in blocking a large proportion of downstream-moving juveniles from leaving these wetland systems and/or stopping adults from entering them to spawn. Extant barrier technologies are not suited to this task because screening small enough to stop juvenile fish clogs easily (Bainbridge 1964) and electrical barriers are expensive and cannot stop downstream-moving fish which drift through them. This study tested the ability of a previously developed bubble curtain system (Zielinski et al. 2014) in Kohlman Creek, Maplewood, Minnesota, USA (45°01036′′ N 93°02048′′ W). This stream links a known C. carpio nursery and a chain of lakes. At the time of this study, the headwaters of Kohlman Creek supported nearly 30 000 young-of-year C. carpio (Osborne 2012), while the downstream lakes contained about 6000 adult C. carpio (Bajer et al. 2011). Kohlman Creek is about 6 m wide and 0.75 m deep. This study location allowed the testing of both upand downstream movement, but in other ways was less than ideal because it was narrowly confined and subject to heavy flooding at times. A bubble curtain was installed that had six perforated PVC pipes mounted on a 4-m-long steel frame, which was attached to docks on both sides of the stream (Fig. 1). Netting was placed around the docks along with V-traps to prevent fish from bypassing the bubble curtain. The