Search

Coregonus populations across their Holarctic range have often undergone unexplained collapses. Here, we document causes of collapse for two of the largest Coregonus populations in the world, cisco (C. artedi) in Green Bay, Lake Michigan, and Saginaw Bay, Lake Huron, during the 1950s. We used simulation models, rank‐order correlations, and historical literature to evaluate effects of overfishing, interactions with introduced rainbow smelt (Osmerus mordax), and overwinter bottom hypoxia on both cisco populations. Fully selected exploitation rates for cisco were alarmingly high (0.82–0.84), correlations and anecdotal evidence suggested negative interactions with rainbow smelt, and extensive areas of hypoxia coincident with cisco declines suggested incubation of cisco embryos was impaired regionally. Our findings suggested that negative effects of rainbow smelt were mainly responsible for the cisco population collapse in Green Bay and that overfishing and hypoxia had suppressed the Saginaw Bay cisco population before rainbow smelt caused its final collapse. [ABSTRACT FROM AUTHOR]

Historically, Cisco Coregonus artedi and deepwater ciscoes Coregonus spp. were the most abundant and ecologically important fish species in the Laurentian Great Lakes, but anthropogenic influences caused nearly all populations to collapse by the 1970s. Fishery managers have begun exploring the feasibility of restoring populations throughout the basin, but questions regarding hatchery propagation and stocking remain. We used historical and contemporary stock-recruit parameters previously estimated for Ciscoes in Wisconsin waters of Lake Superior, with estimates of age-1 Cisco rearing habitat (broadly defined as total hectares ≤ 80 m depth) and natural mortality, to estimate how many fry (5.5 mo posthatch), fall fingerling (7.5 mo posthatch), and age-1 (at least 12 mo posthatch) hatchery-reared Ciscoes are needed for stocking in the Great Lakes to mimic recruitment rates in Lake Superior, a lake that has undergone some recovery. Estimated stocking densities suggested that basin-wide stocking would require at least 6.41 × 108 fry, 4.69 × 108 fall fingerlings, or 3.43 × 108 age-1 fish for a simultaneous restoration effort targeting historically important Cisco spawning and rearing areas in Lakes Huron, Michigan, Erie, Ontario, and Saint Clair. Numbers required for basin-wide stocking were considerably greater than current or planned coregonine production capacity, thus simultaneous stocking in the Great Lakes is likely not feasible. Provided current habitat conditions do not preclude Cisco restoration, managers could maximize the effectiveness of available production capacity by concentrating stocking efforts in historically important spawning and rearing areas, similar to the current stocking effort in Saginaw Bay, Lake Huron. Other historically important Cisco spawning and rearing areas within each lake (listed in no particular order) include 1) Thunder Bay in Lake Huron, 2) Green Bay in Lake Michigan, 3) the islands near Sandusky, Ohio, in western Lake Erie, and 4) the area near Hamilton, Ontario, and Bay of Quinte in Lake Ontario. Our study focused entirely on Ciscoes but may provide a framework for describing future stocking needs for deepwater ciscoes. [ABSTRACT FROM AUTHOR]