An Investigation into the Migratory Behavior, Habitat Use and Genetic Composition of Fluvial and Resident Bull Trout (Salvelinus confluentus) in the Yakima River Basin

Uncertainty regarding the distribution, migratory patterns, and habitat preferences of some populations of adult and sub-adult bull trout (Salvelinus confluentus) in the Yakima Basin constrains effective management for this species. Adfluvial populations have been studied extensively, but information is lacking on fluvial and resident forms of bull trout. This radio telemetry study was initiated in 2003 to increase our knowledge of Yakima Basin fluvial and resident bull trout populations.

Fish movement patterns were intensively monitored from September 2003 through December 2006. A total of 96 fluvial bull trout were captured. Seventy-one of these were radio-tagged and tissue samples taken for DNA analysis. Sixty-two bull trout were radio-tagged in the Naches drainage and seven in the smaller, adjacent Ahtanum Creek drainage. Only two fish were captured and tagged from the upper Yakima drainage. These latter two were chance encounters, as few bull trout are found in the upper Yakima River mainstem or tributaries outside of the Bureau of Reclamation storage reservoirs. Of the 62 bull trout tagged in the Naches drainage, 29 were captured from various holding pools in the mainstem Naches River, 16 from the Tieton River below Rimrock Reservoir (Tieton Dam), seven from Rattlesnake Creek, three from the Bumping River, two from the American River and five from Union Creek (American River tributary). In the Ahtanum Creek drainage, four were from the N. Fork and three from the Middle Fork Ahtanum Creek.

Surgical implantation training was conducted on 32 hatchery rainbow trout before tagging bull trout. Rainbow trout were held as a �"control group” for nine months at the WDFW Naches Hatchery. Although no mortalities occurred with the control group, we observed some shed tags, unabsorbed sutures, suture injury and keloidal extrusion of tags. With tagged bull trout, we found evidence of only six shed tags that we could directly attribute to our surgeries. However, we did find additional tags during the project, some of which appeared to be associated with natural (i.e., spawning) and/or predator induced mortalities that occurred months or years after initial tagging.

We used information derived from DNA tissue analysis to confirm population/stream assignments for radio-tagged fish. In both the Ahtanum and the Naches drainages, bull trout stayed true to their natal spawning areas. Site fidelity was extraordinarily high, near 100% on radio-tracked fish. Genetic analysis revealed that bull trout in the Tieton River stilling pool below the dam at Rimrock Reservoir were from various populations. Some were from adfluvial populations above the dam that were entrained out of the reservoir. Since there were no fish passage facilities, they could not move back upstream to natal spawning grounds. Other bull trout in the dam stilling pool were from various Naches Basin fluvial populations, likely moving into the pool to feed on the abundant forage (entrained kokanee salmon).

Adult bull trout in the Naches River spent lengthy periods of time in their over-wintering habitat, generally moving very little after settling in for the winter (November thru March). The prime locations consisted of a series of large pools in the mid to upper Naches River from the Wapatox Diversion Dam (RM 17.1) upstream to Pool #6 (R.M. 40.5). Fluvial bull trout from several tributary populations over-wintered in these pools, which included adults from Rattlesnake Creek, American River/Union Creek and Crow Creek. Pool selection seemed to be driven by velocity and surface opacity, especially where other forms of cover were absent, but where prey was readily available. Velocity preferences in these holding areas consistently ranged from 1.4 â€" 1.9 feet per second (fps). Adult bull trout were found in these same pools or moving slowly upstream between them throughout the spring (April, May). By early to mid-summer (June, July) the fish were moving into their natal tributary streams, where they continued to move slowly upstream to the spawning grounds. Conversely, resident bull trout found in the Ahtanum Creek drainage stayed year-round in their respective natal streams. Spawning occurs in the month of September in both drainages.

Although adult bull trout often migrated into their natal tributary streams well before the actual spawning period, the time spent on the actual spawning grounds was relatively short, usually only 2-3 weeks, compared to the overwintering (5-6 months) and foraging/migration (5 months) periods. Habitat preference was also different. Spawning habitat consisted of gravel, small cobble and sand substrate found in smaller, higher gradient stream reaches. Also, spawning areas typically contained more woody debris. Spawning bull trout often preferred areas where cover was abundant and vegetation draped over the water or lay completely across the channel. Although redds were often in open exposed areas of the stream; adults were usually close to some form of concealment cover. Access to woody debris (log jams), large boulders, undercut banks, shoreline vegetative cover, deep pools, as well as substrate composition, water flow and temperature all played a crucial role in the selection of spawning areas. Water depth on spawning grounds was much shallower than the water depths associated with foraging, migration and over-wintering (FMO) pool-glide habitat. Bull trout were often observed in shallow water riffles from just a few inches deep to pool tail outs that were two feet deep.