Bycatch, bait, anglers, and roads: quantifying vector activity and propagule introduction risk across lake ecosystems

Abstract

Long implicated in the invasion process, live-bait anglers are highly mobile species vectors with frequent overland transport of fishes. To test hypotheses about the role of anglers in propagule transport, we developed a social-ecological model quantifying the opportunity for species transport beyond the invaded range resulting from bycatch during commercial bait operations, incidental transport, and release to lake ecosystems by anglers. We combined a gravity model with a stochastic, agent-based simulation, representing a 1-yr iteration of live-bait angling and the dynamics of propagule transport at fine spatiotemporal scales (i.e., probability of introducing n propagules per lake per year). A baseline scenario involving round goby (Neogobius melanostomus) indicated that most angling trips were benign; irrespective of lake visitation, anglers failed to purchase and transport propagules (benign trips, median probability P = 0.99912). However, given the large number of probability trials (4.2 million live-bait angling events per year), even the rarest sequence of events (uptake, movement, and deposition of propagules) is anticipated to occur. Risky trips (modal P = 0.00088 trips per year; approximately 1 in 1136) were sufficient to introduce a substantial number of propagules (modal values, Poisson model = 3715 propagules among 1288 lakes per year; zero-inflated negative binomial model = 6722 propagules among 1292 lakes per year). Two patterns of lake-specific introduction risk emerged. Large lakes supporting substantial angling activity experienced propagule pressure likely to surpass demographic barriers to establishment (top 2.5% of lakes with modal outcomes of five to 76 propagules per year; 303 high-risk lakes with three or more propagules, per year). Small or remote lakes were less likely to receive propagules; however, most risk distributions were leptokurtic with a long right tail, indicating the rare occurrence of high propagule loads to most waterbodies. Infestation simulations indicated that the number of high-risk waterbodies could be as great as 1318 (zero-inflated negative binomial), whereas a 90% reduction in bycatch from baseline would reduce the modal number of high risk lakes to zero. Results indicate that the combination of invasive bycatch and live-bait anglers warrants management concern as a species vector, but that risk is confined to a subset of individuals and recipient sites that may be effectively managed with targeted strategies.

Range expansion and molecular confirmation of the Asian fish tapeworm in the lower Great Lakes and St. Lawrence River with notes on infections in baitfish

“…The discovery of S. acheilognathi in emerald shiner purchased from bait retailers within the Great Lakes basin confirms that the cestode has reached sufficient prevalence in the wild to be harvested and transported throughout the commercial supply chain. Although a low prevalence was documented in bait retail tanks in this study (4 and 20%), previous models of the baitfish pathway in Ontario suggest that even low prevalence of non-target fishes in retail tanks can lead to a substantial number of introduction events to inland lakes each year (Drake and Mandrak, 2014a). Ontario supports a large volume of emerald shiner harvest (approximately 47 million emerald shiner were commercially harvested in 2005; OMNR and BAO, 2006) with fish captured primarily from nearshore areas and tributaries from the lower Great Lakes (Erie, Huron, Ontario) and Lake Simcoe (Drake and Mandrak, 2014b).…”

“…The volume of commercial harvest from infected waterbodies, combined with the apparent preference for emerald shiner as host, indicates that future commercial catches of infected emerald shiner from the lower Great Lakes may occur. While most angler movement of baitfish from infected retail localities would likely be to the Great Lakes proper, a subset of inland lakes also are likely to be visited, representing logical early detection surveillance targets (e.g., Lake Simcoe, Belwood Lake, and other high-ranking lakes in Drake and Mandrak, 2014a, Appendix H, Table H1). Surveillance of emerald shiner, bluntnose minnow, and other minnow species for S. acheilognathi at point of harvest or within retail facilities may be warranted, especially given the recent discovery of several additional infected bait species in Lake Huron (mimic shiner, Notropis volucellus, sand shiner, Notropis stramineus, spottail shiner, Notropis hudsonius, plus emerald shiner; Muzzall et al, 2016).…”

“…2, Table 1), using the same beach seine deployed by hand or partially from a boat. Additionally, eight other species of fish were collected from various localities in the St. Lawrence River in June 2014 (Table 2), eight species from inland lakes and rivers in Ontario in July 2014 in areas of low and high predicted baitfish use (Drake and Mandrak, 2014a) (Table 3), and five species from localities in the lower Great Lakes in August through October 2014 (Table 2). Lastly, formalin-fixed museum specimens of emerald shiner and bluntnose minnow collected from the Great Lakes were provided by the Royal Ontario Museum.…”

The invasive bighead goby Ponticola kessleri displays large‐scale genetic similarities and small‐scale genetic differentiation in relation to shipping patterns

“…Bait‐bucket transfers have been shown to be relevant in Northern America (Drake et al . ). However, fishermen in Switzerland were not yet routinely catching invasive gobies when sample collections started.…”

Coalescent models characterize sources and demographic history of recent round goby colonization of Great Lakes and inland waters

“…Collections represent populations from the Great Lakes (e.g., Little Traverse Bay [LTB], Saginaw Bay [SAB], LKE, Lake STC) and populations from three inland river systems (Flint River, Au Sable River, and Cheboygan River systems; Table , Figure ). These three river basins were chosen because they vary in levels of recreational use, distance from populated areas, and number or type of human‐constructed barriers that impede natural upstream movement, all of which have been identified as important factors associated with the secondary spread of aquatic invasive species (Drake & Mandrak, ; Johnson, Olden, & Vander Zanden, ). All tissue samples were preserved in 95% ethanol in the field and returned to the lab for DNA extraction.…”