**ABSTRACT NOT FOR CITATION WITHOUT AUTHOR PERMISSION. The title, authors, and abstract for this completion report are provided below.  For a copy of the full completion report, please contact the author via e-mail at michaelhansen@usgs.gov. Questions? Contact the GLFC via email at frp@glfc.org or via telephone at 734-662-3209.**



 Melissa J. Johnson1, Andrea L. Akins2, Michael J. Hansen3, and Michael J. Seider4



1Wisconsin Department of Natural Resources, Wild Rose State Fish Hatchery, N5871 State Rd 22, Wild Rose, WI 54984

2University of Wisconsin-Stevens Point, College of Natural Resources, 800 Reserve St., Stevens Point, WI 54481

3 United States Geological Survey, Great Lakes Science Center, Hammond Bay Biological Station, 11188 Ray Rd., Millersburg, MI 49759

4 United States Fish and Wildlife Service, Ashland Fish and Wildlife Conservation Office, 2800 Lake Shore Drive East, Ashland, WI 54806



August 2014





Lake Superior supported one of the largest and most diverse lake trout fisheries in the Laurentian Great Lakes until stocks collapsed because of excessive fishery exploitation and sea lamprey predation, after which stocking, sea lamprey control, and fishery regulations were used to restore stocks.  In the Apostle Islands region of Lake Superior, a refuge was established around Gull Island Shoal in 1976 to enable recovery of lake trout stock that spawned on this historically important spawning shoal in the Apostle Islands region.  In the early 2000s, evidence suggested that stocks in the Apostle Islands Region were rehabilitated and near carrying capacity.  Our objectives were: (1) to determine if lake trout abundance, growth, maturity, and mortality differed inside and outside the refuge, before and after the refuge was created; and (2) to determine if future sustainability of lake trout stocks will depend on the Gull Island Shoal refuge.  To address our first objective, we compared lake trout abundance, growth, age structure, and mortality from survey data collected inside and outside the refuge, before and after the refuge was established.  Wild adult lake trout abundance increased significantly whereas stocked adult lake trout abundance decreased significantly after the refuge was created.  Wild adult lake trout were significantly more abundant inside than outside the refuge, and stocked adult lake trout were less abundant inside than outside the refuge.  Wild juvenile lake trout were significantly more abundant inside than outside the refuge, whereas stocked juvenile lake trout abundance did not differ significantly inside and outside the refuge.  Lake trout outside the refuge grew from a theoretical age at zero length of −1.781 years at an instantaneous rate of 0.098/year to an asymptotic length of 35.8 inches, whereas lake trout inside the refuge grew from a younger theoretical age at length (−0.915 years) at a faster instantaneous rate (0.144/year) to a shorter asymptotic length (33.4 inches).  Lake trout inside the refuge were younger than outside refuge.  Lake trout sampled inside the refuge suffered significantly lower mortality than lake trout sampled outside refuges.  Therefore, in response to our first objective, we conclude that the refuge enhanced recovery of the wild lake trout population in the Apostle Islands region.  To address our second objective, we constructed an age-structured simulation model to assess the effect of excluding and including the refuge, as a harvest management tool, on sustainability.  The model was used to estimate median abundance, probability of collapse, and time to extinction, over a range of commercial and recreational fishing mortality rates.  Natural mortality was modeled as a fixed-base rate and a randomly varying sea lamprey mortality rate.  Recruit abundance was modeled as a density-dependent function of adult lake trout abundance, with random process error and parameter uncertainty.  Movement rates from refuge to non-refuge waters were varied to assess effectiveness of the refuge on lake trout sustainability.  In general, median abundance of lake trout collapsed at lower fishing mortality rates with the refuge than without the refuge.  When movement was low, too many fish stayed in the refuge, and conversely, when movement was high, too many fish became vulnerable to fishing, so the refuge was ineffective at both low and high rates of movement.  The refuge prevented extinction at all levels of fishing mortality, but when the refuge was removed, extinction occurred at all combinations of commercial and recreational fishing mortality.  Therefore, in response to our second objective, we conclude that the lake trout population in Eastern Wisconsin waters of Lake Superior would be sustained by the refuge under impossibly high fishing mortality rates, whereas removal of the refuge would likely increase risk of collapse at present fishing mortality rates.  Overall, we conclude that retaining the refuge is the safest action to ensure long-term sustainability of the lake trout population in eastern Wisconsin waters of Lake Superior.