Anderson, Joseph H. 2006. Colonization of newly accessible habitat by coho salmon (Oncorhynchus kisutch). 2006.
Read the full thesis here (870 KB) 
Post-glacial colonization by Pacific salmon (Oncorhynchus spp.) in their native range and rapid dispersal by some introduced populations reveal that salmon are remarkably capable of exploiting unoccupied habitat. However, the initial stages of colonization events are seldom documented. Landsburg Dam on the Cedar River, Washington had excluded salmon from over 27 km of high quality habitat for over a century but was modified to allow passage in 2003. In Chapter I, adult coho salmon (O. kisutch) were sampled as they entered the newly accessible habitat and a subset of fish received radio transmitters to assess three main objectives. The first objective was to document the ecological attributes of the initial colonists (chiefly abundance, timing, body size and hatchery or wild origin) and analyze the proximate environmental stimuli for upriver ladder movement. Annual counts of coho salmon circumventing the dam steadily increased over the period of study (N = 47 in 2003, 99 in 2004 and 170 in 2005) and ladder movement was positively related to discharge in two of three years. The second objective was to evaluate reach scale patterns in spawning site selection and a hypothesis that adult coho would spawn primarily in tributaries of the Cedar River, in particular Rock Creek. In two seasons of telemetry data, over 40% of radio tagged salmon entered Rock Creek at least briefly but all identified spawning sites were in the mainstem Cedar River, contrary to the prediction. Our third objective was to quantify exploratory movements of salmon within a habitat to which they were naïve. Movement patterns differed markedly between the two sexes; males extensively within the new habitat (average minimum estimate = 34.8 km) whereas females moved much less (average minimum estimate = 5.8 km). Roughly half of tagged males (49%) and fewer
females (17%) eventually moved downstream below the dam after spending some time upriver. Regression analysis revealed that male movements within the new habitat were related to breeding opportunities, as males moved more when females were scarce. Total individual movements of tagged salmon were not detectably related to body size, timing, or Cedar River discharge, although tributary entrances tended to occur on days of high discharge. These results suggest that breeding opportunities dominated exploratory movements of salmon colonists within the new habitat. The number of colonists and widespread movements used by many salmon suggest that salmon are capable of finding new habitat if it is made available to them, are promising for future efforts to restore access to lost salmon habitat.
Most research on salmon population expansion has focused on straying and colonization by adults but dispersal by juveniles of stream-rearing species such as coho salmon (Oncorhynchus kisutch) can be an important part of the overall colonization process. The major objectives of Chapter II were to describe the spatial distribution and growth patterns of the first two generations of juvenile coho produced in the newly accessible habitat. Snorkel surveys revealed that reach scale densities in the main Cedar River largely matched the spatial distribution of spawning the previous fall, as densities were generally highest in the lower reaches. However, sequential snorkel surveys indicated that juvenile coho immigrated into Rock Creek, a tributary of the Cedar River where few, if any, adults spawned. Competition was unlikely to be the only driver of movement as Rock Creek immigrants faced higher densities of resident trout, and end of summer coho densities were higher in the creek than the mainstem, suggesting a role for habitat preference in dispersal. Juvenile coho salmon were also captured and measured from mainstem and creek habitats to assess growth. There was a significant difference in juvenile coho size between year classes but not between mainstem and tributary habitats. Within the Cedar River (both seasons) and Rock Creek (pattern observed one season; no pattern one season), larger fish tended to occur farther upriver. An interannual comparison and the longitudinal pattern of size were consistent with density dependent growth at a broad scale, but the most direct test of density dependence yielded no relationship between the number of coho in mainstem river aggregations and average body size within the aggregation. Dispersal away from nest sites into areas of lower
conspecific density, both within relatively “open” mainstem habitats and into new areas such as Rock Creek, may have reduced the effects of conspecific density on growth, and played a role in overall expansion of the species into new habitat.