Identifying Historical Populations of Steelhead Within the Puget Sound Distinct Population Segment

The Puget Sound Steelhead Technical Recovery Team (PSS TRT) convened in March 2008 to review information relevant to the identification of historical demographically independent populations (DIPs) of steelhead (Oncorhynchus mykiss) in the Puget Sound steelhead distinct population segment (DPS). The PSS TRT identified three major population groups (MPGs) containing a total of 32 steelhead DIPs in Puget Sound.

Steelhead in the Puget Sound DPS exhibit two distinct life history strategies: summer-run and winter-run migrations. Winter-run steelhead, also known as ocean-maturing steelhead, return to freshwater during the winter and early spring months and spawn relatively soon after entering freshwater. Alternatively, summer-run (stream-maturing) steelhead return to freshwater during late spring and early summer in a relatively immature state and hold there until spawning in the following winter/spring. Generally, but not necessarily, summer-run steelhead return-timing is coordinated with river flow patterns that allow access past barriers to headwater spawning areas. Presently and historically, winter-run steelhead numerically represent the predominant life history type in Puget Sound.

Steelhead exhibit considerable diversity in age at smoltification, age at return or maturation, and spawning timing and repeat spawning (iteroparity). Overall, there were few clear trends in these life history traits across the Puget Sound DPS. Steelhead in lowland, rain-dominated streams tended to spawn earlier than fish in upland or headwater, snowfall-dominated streams. Information on life history characteristics is limited for all but a few DIPs and completely absent for others, especially for summer-run populations. Additionally, there is little information available on ocean migratory patterns outside of Puget Sound and, until recently, steelhead tagging studies have not been undertaken to any great degree.

The PSS TRT reviewed available information on Puget Sound steelhead, which included life history and genetic data. This information was not universally available for all populations and, in many cases, ecological information was used to estimate life history characteristics. In the absence of historical demographic information (e.g., abundance, spatial structure), the TRT also used basin characteristics to estimate the potential historical size and level of interaction between prospective populations. The TRT initially utilized an expert panel system to develop criteria for establishing DIP criteria, but ultimately incorporated these criteria into a decision support system to identify DIPs. DIPs were in turn organized in MPGs. These larger scale units delineate DPS-wide spatial structure. The TRT identified MPGs based on the geographic and ecological characteristics of the DPS and the genetic clustering of existing steelhead populations in Puget Sound.

As a preliminary filter for putative DIPs, the TRT only considered basins with intrinsic productivity (based on stream area) equal to or greater than that estimated for Snow Creek, an apparently self-sustaining, small, wild population located on the northeastern corner of the Olympic Peninsula. The decision support system relied on basin intrinsic potential, basin elevation, snow cover, distances between potential DIPs, genetic differences between potential DIPs, life history differences between potential DIPs, and the presence of temporal migrational barriers between potential DIPs. The decision support system, or gatekeeper model, required that the TRT estimate for each factor a threshold value that indicated populations were demographically independent with a very high certainty. One of the benefits of this system was that missing information did not bias the outcome.

The boundaries for historical DIPs were in part established using information related to two isolating mechanisms: homing fidelity and migration timing. Homing fidelity was examined to estimate the extent of adult exchange among putative spawning populations. Analysis of the terminal recoveries of adult marked hatchery fish indicates that less than 10% of the recoveries occur more than 50 km from the mouth of their natal stream (stream of release). Within a basin, temporal differences in return migration and spawn timing provided mechanisms for establishing demographically and reproductively isolated populations. Adult run and spawn timing are often coordinated with stream hydrology and temperature, which in turn are strongly affected by basin elevation. Major run-timing (e.g., summer and winter) differences were used as one criterion for distinguishing DIPs in the gatekeeper decision support system, especially where temporal barriers provided a reproductive barrier between presumptive DIPs.

In the Puget Sound DPS, three MPGs were identified: Northern Cascades, Central and South Puget Sound, and Hood Canal and Strait of Juan de Fuca. Within the Northern Cascades MPG, 16 DIPs (8 winter run, 3 summer/winter run, 5 summer run) were identified as historically present. In the Central and South Puget Sound MPG, 8 winter-run DIPs were historically present. There was some discussion regarding the presence of an additional historical summer-run DIP in the Green River or, alternatively, that the Green River winter-run DIP should be designated as a mixed summer/winter-run DIP, although the information available was not considered compelling. Additionally, while there are no known native-origin summer-run steelhead currently in the Green River (i.e., the summer-run steelhead currently released into and naturally spawning in the Green River originated from the Skamania Hatchery in the Columbia River basin), it is possible that resident O. mykiss above Howard Hansen Dam may contain the genetic legacy of a summer run. The Hood Canal and Strait of Juan de Fuca MPG historically contained 8 DIPs (1 summer/winter run and 7 winter run, with 2 of these winter runs possibly historically including summer-run components).

Where steelhead population information was available, especially genetic information, it was possible to identify steelhead DIPs with a relatively high degree of certainty. In other cases, ecological information provided a reasonable proxy for population data. The TRT strongly recommends further life history and genetics sampling and evaluation, especially in those areas currently less well studied. For some populations, basic abundance data are still lacking and need to be collected. It is likely that, in the process of collecting additional information on these populations, some revision in the DPS population structure will be necessary and should be undertaken.