1998 Sunset Pond Survey: The Warmwater Fish Community in a Disturbed, Urban System and Salmonid Migration Route

Sunset Pond is a small, constructed body of water located in Whatcom County within the city limits of Bellingham, Washington. Sunset Pond, occurring on river km 4 of Squalicum Creek, and Bug Lake, occurring on river km 3.5, were carved directly into the watercourse during the construction of Interstate-5 in the 1960’s, and now function as part of a flood control system for the city of Bellingham. Sunset Pond is eleven m deep and five ha in surface area (Table 1). Bug Lake is six m deep and 2.5 ha, but much of the lake is only one to two m deep resulting in a lower volume than Sunset Pond. Both of these riverine ponds are productive, highly stratified throughout most of the year, and fished year round for introduced largemouth bass (Micropterus salmoides), yellow perch (Perca flavescens), bluegill (Lepomis macrochirus), brown bullhead (Ameiurus nebulosus), yellow bullhead (Ameiurus natalus), native steelhead trout (Oncorhynchus mykiss), and native cutthroat trout (Oncorhynchus clarki). These water bodies represent potential rearing habitat and refugia from summer low flows and winter high flows for juvenile salmonids (R.W. Beck and Associates 1994). However, illegally introduced species of the families Centrarchidae, Ictaluridae, and Percidae are well suited to conditions in these ponds, and may compete with and prey upon salmonids.

The Squalicum Creek system historically supported populations of anadromous fish and formed a migration route for chum salmon (Oncorhynchus keta), coho salmon (Oncorhynchus kisutch), steelhead trout, and sea-run cutthroat trout. Squalicum Creek is one of the largest independent drainages in Whatcom County. Originating in the hills west of the Nooksack Basin it flows westward, draining 6,750 hectares of early successional forested, agricultural, residential, and urban areas around and within the city of Bellingham into Bellingham Bay. The watershed is independent of the Cascade foothills and is not influenced by mountain snowmelt. A third order stream, Squalicum Creek and its adjoining tributaries consist of 84 km of stream habitat. However, culverts bar salmonid migration from as much as 12 km of stream reaches that include adequate salmonid spawning and rearing habitat in the upper watershed (WDFW Spawner Escapement Survey Report 1998, Downen 1999). Native lotic fish species include coho salmon, steelhead trout, cutthroat trout, three-spine stickleback (Gasterosteus aculeatus), prickly sculpin (Cottus asper), western brook lamprey (Lamptera richardsoni), and Pacific lamprey (Lampetra tridentata).

Squalicum Creek and Sunset Pond lie in a watershed dominated by human activity and associated disturbance regimes. Historically, Squalicum Creek flowed through lowland forests of western red cedar, Douglas fir, big leaf maple, alder, and willow (R.W. Beck and Associates 1994). Today, upper portions of the watershed, above river km 6 on Squalicum Creek, are agricultural, residential, or forested with second growth vegetation. These land uses impose disturbance regimes characterized by vegetation removal, elevated water temperatures, summer low flow, lowered dissolved oxygen, encroachment of invasive grasses, sedimentation, and streambed disturbance (Brown and Krygier 1970; Lenth 1995). Impervious surfaces have expanded steadily in the lower portion of the watershed below river km 6 of Squalicum Creek, that channel storm water runoff, tainted with fertilizers, trace heavy metals, hydrocarbons, and organic wastes (Herrera Environmental Consultants Inc. 1991). Land uses throughout the lower watershed are primarily urban, residential, and industrial, and these impose disturbance regimes characterized by heavy storm water runoff, extreme discharge peaks, sedimentation, scouring, and non-point source pollution (Casper 1994; R.W. Beck and Associates 1994).

The relevance of these conditions to salmonid restoration and persistence in lowland stream systems resulted in an examination of outmigration timing, species composition, environmental conditions, growth, mortality and life histories of salmonids inhabiting the watershed. Downen (1999) monitored the outmigration of salmonids with a complete smolt trap between 21 March and 21 June 1998. Although fish appeared in the trap in small numbers, beginning in March when the full-spanning smolt trap was installed, 83% of the run occurred between 25 May and 1 June during a single discharge event, coincident with a new moon (Figure 1). Positive correlations for all species were significant with respect to timing. The outmigration consisted of 7,168 coho salmon, 1,270 cutthroat trout, 180 steelhead trout, 6 fall-run chinook salmon, and 423 marked Skookum hatchery stock coho salmon with estimated instantaneous growth rates of 3.7, 3.8, 4.2, 11.1 and 4.9, respectively. In view of conformity with assumptions and using Peterson mark and recapture data collected from two release groups, exploitation and initial population estimates were calculated for the coho salmon parr population as 0.57 and 12,575, and 0.42 and 17,066, respectively. Estimated instantaneous mortality for marked coho salmon released in Sunset Pond, marked coho salmon released in Bug Lake, and system-reared coho salmon were 2.2, 3.5, and 2.3, respectively. Ninety-six percent of coho salmon, 41% of steelhead trout, and 14% of cutthroat trout migrated at age 1 (Table 2). The balance of these populations migrated at age 2 with the exception of 4% of cutthroat trout and 1% of steelhead trout migrating at age 3. All fall-run chinook were age-0 fingerlings. Seventy-nine adult cutthroat trout and four steelhead trout were also retained in the trap between 21 March and 21 June. Life history patterns for adult fish were characterized by short freshwater residence times and rapid growth prior to their first seaward migration.

Downen collected extensive data on salmonid populations and some limited data from introduced fish populations the previous spring. Rapid growth, high mortality, and decreased freshwater residence time for salmonids appeared correlated with environmental conditions in this disturbed system. The presence of some of the previously mentioned introduced species was confirmed, and downstream dispersal of age-1 largemouth bass feeding on coho salmon fry was observed. However, the relative abundance, growth, and condition of introduced species were unknown in Sunset Pond, their principal habitat in the Squalicum Creek system, Assessing the potential for ecological interactions between salmonids and introduced fish species required an assessment of introduced populations as well. To this end the Warmwater Enhancement Program of the Washington Department of Fish and Wildlife (WDFW) conducted a warmwater stock assessment of Sunset Pond during fall of 1998.