Morphological and Ecological Differences Between Shallow- and Deep-water Lake Trout in Lake Mistassini, Quebec

Lake trout (Salvelinus namaycush) in Lake Mistassini, Quebec, were investigated to determine whether they resembled the lean and siscowet morphotypes of Lake Superior and Great Slave Lake. Lake trout caught in deep water were predicted to resemble the siscowet morphotype and to be better adapted for vertical migration (i.e., low percent buoyancy) than those caught in shallow water. The research objectives were to 1) identify groups based on shape, and 2) determine whether shape was associated with other morphological traits (fin length, buoyancy, color), ecology (habitat depth, diet), and life history (size at adulthood). Eighty-five lake trout were collected from three depth zones. At least two phenotypes exist in Lake Mistassini. A shallow-water form (< 50-m depth), identified by its streamlined shape, was dark in color and high in percent buoyancy. A deep-water form (> 50-m depth), identified by a deep anterior-body profile, was light in color and lower in percent buoyancy than the shallow-water form. Absolute buoyancies were relatively high in both forms; therefore, the deep-water form did not appear well-adapted for vertical migration. Opossum shrimp (Mysis relicta) were more frequent and abundant in stomachs of deep-bodied trout. All deep-bodied trout (minimum 32-cm SL) had reached adulthood, whereas immature streamlined individuals were as long as 49 cm in SL. The deep-bodied form resembled humper lake trout, a lesser-known third morphotype from Lake Superior. A humper-like morphotype in Lake Mistassini, and the apparent absence of a siscowet-like morphotype, challenges the previously-held hypothesis that humpers resulted from an introgression of leans and siscowets. INDEX WORDS: Adaptation, buoyancy regulation, humper, lean, phenotypic diversity, siscowet. *Corresponding author. E-mail: [email protected] Address correspondence to Great Lakes Fishery Commission, 2100 Commonwealth Blvd., Suite 100, Ann Arbor, MI 48105-1563 INTRODUCTION An abundance of available niches is hypothesized to have facilitated morphological divergence in fish species colonizing recently deglaciated lakes (Skulason and Smith 1995, Smith and Skulason 1996). In at least eight fish families, resource polymorphism is attributed to divergent selection in benthic and pelagic habitats (Robinson and Wilson 1994, Smith and Skulason 1996). A less-recognized ecological variable, habitat depth, is also associated with the radiation of freshwater fishes (Eshenroder et al. 1999, Turgeon et al. 1999, Power et al. 2005). For example, lake trout (Salvelinus namaycush) phenotypic diversity in the Laurentian Great Lakes once included shallowand deep-water forms in Lakes Superior, Michigan, Huron, and Erie (Brown et al. 1981, Goodier 1981, Krueger and Ihssen 1995). However, this diversity was lost from Lakes Michigan, Huron, and Erie during the past century. Some of the original phenotypic diversity remains in Lake Superior (maximum depth ~400 m) in forms known as lean, humper, and siscowet lake trout (Khan and Qadri 1970, Lawrie and Rahrer 1973, Moore and Bronte 2001). The origin and maintenance of lake trout phenotypic diversity has been difficult to investigate due to the loss of diversity from most of the Laurentian Great Lakes and the highly disturbed nature of remaining communities. Therefore, recent lake trout Phenotypic Diversity of Lake Trout in Lake Mistassini 157 studies have been conducted on other large, deep North American lakes. Descriptions of lake trout communities in Great Bear (maximum depth ~450 m) and Great Slave (maximum depth ~600 m) lakes have provided a broader perspective on the association of lake trout phenotypes with ecological variables such as habitat depth. In Great Slave Lake, for example, a siscowet-like trout with high-fat-content tissue recently was caught in deep-water habitats more than 50 m in depth (Zimmerman et al. 2006). In Great Bear Lake, deep-water forms have not been identified; however, shallow-water trout are exceptionally diverse in their morphology (Blackie et al. 2003, Alfonso 2004). Further, inter-lake comparisons of large, deep lakes would provide a better understanding of lake trout phenotypic diversity and its association with deep-water habitat. The siscowet phenotype from Lake Superior has high-lipid tissue (Eschmeyer and Phillips 1965), which may be an adaptation for buoyancy regulation in deep-water habitats (Thurston 1962, Henderson and Anderson 2002). The compressed swim bladder of a vertically migrating fish, when the fish has descended, is ineffective at regulating buoyancy (Alexander 1972, 1993). High-lipid content lowers the density of body tissue in water and, consequently, minimizes the energetic costs of swimming with a compressed swim bladder (Henderson and Anderson 2002). Diel vertical migrations have been observed in siscowet predators (Hrabik et al. 2006, Jensen et al. 2006) as well as in deep-water lacustrine prey such as bloater Coregonus hoyi (Eshenroder et al. 1998, TeWinkel and Fleischer 1999) and opossum shrimp Mysis relicta (Beeton 1960, Bowers 1988). The ubiquity of a siscowet-like phenotype in large, deep North American lakes would suggest that this form is best adapted to deep-water living and to foraging on vertically migrating prey such as deep-water coregonines. A second deep-water form, the humper trout, is far less abundant than siscowets in Lake Superior (Peck 1975). Humpers have been considered to be a “form” (Eschmeyer and Phillips 1965) or a “morphological variant” (Lawrie and Rahrer 1973, Burnham-Curtis and Bronte 1996) of the lean trout and have been hypothesized to have resulted from the introgression of lean and siscowet trout during a period of fluctuating water levels (Burnham-Curtis 1993, Burnham-Curtis and Smith 1994). If humperlike trout occur as distinct populations outside Lake Superior in the absence of the siscowet form, the existing hypothesis on their origin would need to be reconsidered. The purpose of this study was to quantitatively describe the phenotypic diversity of lake trout in Lake Mistassini, a large, deep lake (maximum depth > 150 m) in central Quebec. The abundance of deep-water habitat in this lake was proportionally comparable to other lakes where lake trout phenotypic diversity was known; however, this lake was smaller in surface area (2,150 km2) than Lake Superior (82,100 km2) or Great Slave Lake (28,568 km2) (Beeton 1984). Two hypotheses were tested in this study. First, lake trout occupying deep-water habitat are predicted to have lower percent buoyancy, and thus be better adapted for vertical migration, than those occupying shallow-water habitat. Second, deep-water forms, if they exist, will likely resemble the shape and buoyancy of siscowet trout from Lake Superior. The research objectives were to 1) identify lake trout groups based on shape, and 2) determine whether lake trout shape was associated with other morphological traits (fin length, buoyancy, color), ecology (habitat depth, diet), and life history (size of adulthood). MATERIAL AND METHODS