**ABSTRACT NOT FOR CITATION WITHOUT AUTHOR PERMISSION.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 Jim.Reist@dfo-mpo.gc.ca. Questions? Contact the GLFC via email at frp@glfc.org or via telephone at 734-662-3209.**

Patterns and processes of cisco differentiation in relation to food web structuring using ecological tracers

Reist, J. D.¹, A. M. Muir², P. Vecsei³, T. C. Pratt⁴, M. Power⁵, M. T. Arts⁶, and N. E. Mandrak⁷

¹Fisheries and Oceans Canada, 501 University Crescent, Winnipeg, Manitoba

R3T 2N6

²Great Lakes Fishery Commission, 2100 Commonwealth Blvd., Suite 100

Ann Arbor, Michigan 48105

³Golder Associates, 9 – 4905 – 48 Street Yellowknife, Northwest Territories,

X1A 3S3

⁴Fisheries and Oceans Canada, 1219 Queen St. E., Sault Ste. Marie, Ontario

P6A 2E5

⁵University of Waterloo, 200 University Avenue West, Ontario N2L 3G1

⁶Ryerson University, 350 Victoria St., Toronto, Ontario M5B 2K3

⁷Department of Fisheries and Oceans Canada, 867 Lakeshore Road,

Burlington, Ontario L7R 4A6

February 2016

ABSTRACT:

The coregonine ciscoes of North America show extensive morphological and life history diversity with multiple phenotypes occurring across their range. By the mid-1900s, much of that diversity had been lost from the Laurentian Great Lakes and they are currently under consideration for restoration. We implemented a comparative approach to resolve links between cisco morphs and their niches within Great Slave Lake and Lake Superior to assess whether the ecological opportunity to promote and sustain cisco diversity currently exists within the Great Lakes. Great Slave Lake contained a suite of cisco morphs comparable to those described for the Laurentian Great Lakes, with two notable exceptions. We provided the first record of Coregonus sardinella (a western assemblage cisco) in Great Slave Lake as well as the first description of an adfluvial morph that differed morphologically from its lacustrine conspecifics. The Great Slave morphs differed in body shape, linear measures and counts, δ13C and δ15N, and growth. Our results suggest that ontogenetic niche shifts occurred within the lacustrine C. artedi group. Small (≤ 299 mm), low-gillraker (≤ 44) and large (≥ 300mm), high-gillraker (≥ 45) C. artedi had the same geometric body shape and were phenotypically similar based on 23 size-corrected linear measures. Differences in gillraker number and phenotype with age and size were explained by shifts in trophic resource and habitat use. When comparing the Great Slave and Lake Superior cisco assemblages, species or morph discrimination was surprisingly robust within lakes, with 86% of the fish correctly assigned and only 1%

(23 of 1011 fish) assigned to the incorrect lake. Between lakes, species or ecomorphs for each cisco group had distinct signatures. Our results are consistent with the hypothesis that the observed patterns of diversity in North American ciscoes were caused by parallel sympatric radiation within each lake after initial colonization, although our results do not preclude that ecological forces shaped morphology after an initial allopatric expansion. We found that fatty acids biomarkers may provide evidence of vertical spatial segregation based on depth of feeding of fishes. Relatively homogeneous quality and availability of prey suggest that spatial segregation may be a key axis along which niches are partitioned among sympatric morphotypes/species within Great Slave Lake and Lake Superior. We found little evidence that trophic resources differ between the two lakes, suggesting that perhaps invasive species have each competitively excluded the most similar native fishes and assumed their role in niche space.