FISH ECOLOGY

Coho Salmon

Coho salmon are found along the Pacific Coast from Monterey Bay in central California to Point Hope, Alaska (Wydoski and Whitney 1979). In Washington, coho salmon spawn in streams in the Columbia River Basin, Puget Sound, and coastal drainages (Wydoski and Whitney 1979). In the Lake Washington system, coho salmon stocks have been divided into the Lake Washington/Sammamish Tributary stock and the Cedar River stock (Washington Department of Fisheries [WDF] et al. 1993). Adult coho salmon migrate through Lake Washington and Lake Sammamish to reach spawning grounds in the Cedar and Sammamish River systems, and in small tributaries to the lakes. Adults begin migrating into fresh water in August, and spawn from late October through December in most systems, and through mid-March in the Cedar River (WDF et al. 1993). Coho salmon eggs hatch after 45 to 60 days, depending on water temperature (Wydoski and Whitney 1979).

Life History Strategies

Coho juveniles typically rear in fresh water for one year, but may spend two years in fresh water or migrate to sea as fry during their first spring (Groot and Margolis 1991; Hartman et al. 1982). While in fresh water, juveniles utilize all accessible reaches of their natal stream systems for rearing, including lakes, seasonally wetted areas, off-channel ponds, sloughs, swamps, and their tributaries (Pollard et al. 1997; Bryant et al. 1996; Swales et al. 1988; Hartman and Brown 1987; Cederholm and Scarlett 1981; Skeesick 1970). Some physical characteristics of habitat typically selected by coho fry and parr include depths greater than 8 centimeters, low current velocity, and availability of cover (Fransen et al. 1993; Fausch 1993; Shirvell 1990; Bugert et al. 1991). Coho juveniles overwintering in two small Vancouver Island lakes were closely associated with the shoreline, with the highest CPUEs from traps set as close as possible to the shore (Swales et al. 1988).

Migration

Juvenile coho generally begin migrating to sea as smolts during their second spring, although Irvine and Ward (1989) found that 10 percent of the smolts migrating from the Keogh River watershed were age-2+ fish. In general, peak outmigration is in May and average smolt size is 90 to 115 mm (Weitkamp et al. 1995).

Studies at the mouth of the Cedar River have shown that most coho enter Lake Washington in May and June, and are 100 mm or greater (Fresh, pers. comm., 2 August 1999). Beyond the contribution of natural spawners, WDFW's Issaquah Creek hatchery has an annual production goal of releasing 1 million age-0+ coho and 450,000 yearlings into Issaquah Creek each spring, many of which enter Lake Washington via the Sammamish River. Additionally, 90,000 coho smolts are also released each May from the University of Washington hatchery (Tetrick, pers. comm., 10 July 2000).Emigrating coho smolts in a Lewis River reservoir traveled in schools of a few to several hundred fish, and generally remained in the upper 3 meters (Hamilton et al. 1970). Allen (1968) tested whether coho smolts migrating 56 km to saltwater from the Issaquah Creek hatchery through a lake system populated with piscivorous fish would experience higher mortality than coho from the same cohort released from the University of Washington hatchery (8 km from saltwater). Allen (1968) found that the fish migrating from the Issaquah Creek hatchery had an approximately 22 percent higher mortality rate than those from the University of Washington hatchery.

McMahon and Holtby (1992) found that coho smolts in a river and estuary aggregated and sought cover near large woody debris, overhanging banks, and riparian vegetation. Moser et al. (1991) reported that progress of the smolt migration in the Chehalis River was saltatory, with periods of movement with the current interspersed with periods of holding in low-velocity areas that also provided cover.

The distribution of juvenile coho salmon in Lakes Washington and Sammamish is poorly understood. There is evidence that juvenile coho are migrating and feeding along the Lake Washington shoreline (Fresh, pers. comm., 2 August 1999). Gill net sampling in all zones of Lake Washington by Bartoo (1972) indicated that coho juveniles were present during May, June, and July. Beauchamp (Univ. of Washington, unpubl. data) captured juvenile coho in all sampled littoral areas except during July and August. Beak Consultants Incorporated (1998) reported that peak smolt migration from the Sammamish River was April through mid-May, although their sampling ended in mid-June. Tabor and Chan (1996) found coho smolts in south Lake Washington from April to early June, with peak abundance in early May. Coho juveniles are less commonly encountered in sampling efforts than chinook and sockeye. In addition, some juveniles have been found to migrate out to sea as zero-age fish while some yearling-plus age coho may be residuals in the lake. (Warner, pers. com. 27 June 2000).

Temperature

Water temperature affects the distribution of coho salmon in lakes and reservoirs. Bjornn and Reiser (1991) reported the preferred temperature for coho as 12 to 14° C, and that temperatures from 23 to 25° C could be lethal and were actively avoided by most salmonids. In a stocked Wisconsin lake and in a Washington reservoir, coho inhabited nearshore areas in the spring and fall, and moved below the thermocline into the metalimnion in the summer where temperatures remained below 17° C (Engel and Magnuson 1976; Hamilton et al. 1970). However, coho fry remained in the littoral zone throughout the summer in Margaret Lake in southeast Alaska, especially near the cooler tributary mouths (Bryant et al. 1996). Summer afternoon surface temperatures occasionally exceeded 18° C in Margaret Lake, but most coho were caught while beach seining at dawn and dusk when temperatures were lower, and at the mouths of cool tributaries (Wright, pers. comm., 13 August 1999). This preference for lower water temperatures in the littoral zone during the summer, segregating themselves from shore-based sampling efforts. The late summer distribution of coho would also be restricted to the narrow metalimnion in Lake Sammamish, in a similar manner as chinook (see above). Like chinook, the result of this restricted distribution would be increased stress on holding adult fish, reduced juvenile foraging opportunities, and potentially higher predation mortality.

Diet

Diet may also be a factor in the distribution of juvenile coho in lakes. Hamilton et al. (1970) found that large coho juveniles were limnetic, with zooplankton as their primary prey. In Chignik Lake, Alaska, yearling coho fed heavily on newly emerged sockeye salmon fry around shoreline spawning and incubation areas (Ruggerone and Rogers 1992). Coho in the littoral zone of Margaret Lake, Alaska, fed entirely on insects (Cartwright and Beauchamp 1995). Aquatic insects comprised 75 percent of the diet of coho smolts in south Lake Washington during spring (February-June) 1995, and juvenile fish another 15 percent (Tabor and Chan 1996). Smolt-sized coho in a Wisconsin lake fed at the surface in the littoral zone during April, and ate aquatic and terrestrial insects (Engel and Magnuson 1971). In late spring and early summer, coho occupied both the littoral and epilimnetic zones, and continued to feed at the surface on terrestrial insects and some aquatic insects (Engel and Magnuson 1971). In late summer (mid-July to October), when water temperature in the epilimnion exceeded 16° C, coho avoided the littoral and epilimnetic zones, and fed on aquatic insects and zooplankton in the metalimnion (Engel and Magnuson 1971). Coho appear to adapt to existence in either the littoral or limnetic zones, exploiting available prey items. Because of this adaptability, temperature may be the most important determinant of coho distribution in lakes.