U.S. Dept Commerce/NOAA/NMFS/NWFSC/Publications

NOAA-NWFSC Tech Memo-27: Status Review of West Coast Steelhead

Steelhead Life History and Ecology

Oncorhynchus mykiss exhibit perhaps the most complex suite of life history traits of any species of Pacific salmonid. They can be anadromous or freshwater resident (and under some circumstances, apparently yield offspring of the opposite form). Resident forms are usually called rainbow, or redband, trout. Those that are anadromous can spend up to 7 years in fresh water prior to smoltification, and then spend up to 3 years in salt water prior to first spawning. The half-pounder life history type in southern Oregon and northern California spends only 2 to 4 months in salt water after smoltification, then returns to fresh water and outmigrates to sea again the following spring without spawning. Another life history variation is the ability of this species to spawn more than once (iteroparity), whereas all other species of Oncorhynchus, except O. clarki, spawn once and then die (semelparity).

Migration and Spawn Timing

The most widespread run type of steelhead is the winter (ocean-maturing) steelhead. Winter steelhead occur in essentially all coastal rivers of Washington, Oregon, and California, south to Malibu Creek. Summer (stream-maturing) steelhead, including spring and fall steelhead in southern Oregon and northern California, are less common; for example, on the Oregon coast only the Rogue, Umpqua, and Siletz Rivers have natural populations of summer steelhead. Inland steelhead of the Columbia River Basin, however, are essentially all stream-maturing steelhead; as discussed earlier, these inland steelhead are referred to in terms of A-run and B-run.

Available information for natural populations of steelhead (Table 3) reveals considerable overlap in migration and spawn timing between populations of the same run type. Moreover, there is a high degree of overlap in spawn timing between populations regardless of run type. California steelhead generally spawn earlier than those in areas to the north; both summer and winter steelhead in California generally begin spawning in December, whereas most populations in Washington begin spawning in February or March. Relatively little information on spawn timing is available for Oregon and Idaho steelhead populations. Among inland steelhead, Columbia River populations from tributaries upstream of the Yakima River spawn later than most downstream populations.

Ageing

Steelhead exhibit great variation in smolt age and ocean age both within and between populations, but there are some trends.

Smolt age--Smolt age discussed here is based on scale and otolith data from adult steelhead. The emphasis on adult steelhead is based on the assumption that fish surviving to spawning age are expressing the successful and adaptive life history strategy for steelhead in a given geographical location. Steelhead from British Columbia and Alaska most frequently smolt after 3 years in fresh water (Withler 1966, Narver 1969, Sanders 1985). In most other populations for which there are data, the modal smolt age is 2 years (Table 4). Hatchery conditions usually allow steelhead to smolt in 1 year; this difference is often used by biologists to distinguish hatchery and wild steelhead. There appears to be an increase in the frequency of naturally produced 1-year-old smolts in the southern portion of the steelhead range (Table 4). Withler (1966) suggested that there may be a latitudinal cline in steelhead smolt age; however, Titus et al. (in press) found no statistical evidence for a latitudinal cline in steelhead smolt age from California to British Columbia.

Ocean age--North American steelhead most commonly spend 2 years (2-ocean) in the ocean before entering fresh water to spawn (Table 5). Populations in Oregon and California have higher frequencies of age-1-ocean steelhead than populations to the north, but age-2-ocean steelhead generally remains dominant. Withler (1966) and Titus et al. (in press) found that ocean age at spawning (and mean adult length) increased with increasing latitude.

Total age--For most steelhead populations, total age at maturity can be estimated by adding the smolt age and saltwater age. However, summer steelhead (especially in the Columbia River Basin) enter fresh water up to a year prior to spawning, and that year is generally not accounted for in the saltwater age designation; for example, a 2-ocean steelhead from the Yakima River may actually have 3 years between smolting and spawning. Table 6 shows the most common life history patterns expressed by North American steelhead from several river basins. Most steelhead in Alaska and British Columbia are 3/2 (smolt age/ocean age) and have a total age of 5 years at first spawning. For coastal steelhead in Washington, Oregon, and northern California, the modal total age at maturity is 4 years (2/2). Central and southern California steelhead appear to spend less time in the ocean, and they are dominated by 3-year-old (2/1) spawners. Complete life history data for southern California steelhead are lacking; however, it appears that it is common for these fish to smolt in 1 year (CDFG 1995). If they only have one ocean year, as neighboring populations to the north do, then adults may be spawning as 2-year-olds (1/1) in this region.


Table 6. Most common life history patterns reported for selected steelhead populations; frequency of occurrence in sample is shown in parentheses. Format used is freshwater age/ocean age at first spawning migration. Populations are generally arranged from north to south.
Life history (frequency) Sample
Population Run typea Primary Secondary size Reference
Alaska
Karluk RiverS 3/2 (0.42) 2/2 (0.36) 62 Sanders 1985
Anchor RiverS 3/2 (0.61) 3/1 (0.23) 80 Sanders 1985
Copper RiverS 3/2 (0.73) 3/1 (0.10) 30 Sanders 1985
Situk RiverS/O 3/2 (0.43) 3/3 (0.32) 211 Sanders 1985
Sitkoh CreekO 3/2 (0.38) 3/3 (0.27) 497 Sanders 1985
Karta RiverO 3/2 (0.46) 3/3 (0.20) 542 Sanders 1985
British Columbia (mainland)
Babine RiverS 3/2 (0.62) 3/3 (0.17) 100 Narver 1969
Cheakamus RiverO 3/2 (0.34) 2/3 (0.25) 64 Withler 1966
Capilano RiverO 3/2 (0.40) 2/2 (0.26) 70 Withler 1966
Capilano RiverS 3/2 (0.49) 3/3 (0.31) 86 Withler 1966
Seymour RiverO 3/2 (0.38) 3/3 (0.22) 58 Withler 1966
Seymour RiverS 3/2 (0.48) 2/3 (0.24) 25 Withler 1966
British Columbia (Fraser River Basin)
Coquitlam RiverO 3/2 (0.49) 2/2 (0.23) 146 Withler 1966
Alouette RiverO 2/2 (0.32) 2/3 (0.32) 131 Withler 1966
Chilliwack RiverO 2/2 (0.31) 2/3 (0.31) 770 Maher and Larkin 1955
Chehalis RiverO 3/3 (0.34) 3/2 (0.33) 111 Withler 1966
Coquihalla RiverO 3/2 (0.49) 3/3 (0.18) 39 Withler 1966
Coquihalla RiverS 3/2 (0.63) 2/2 (0.15) 150 Withler 1966

Life history (frequency) Sample
Population Run typea Primary Secondary size Reference

British Columbia (Vancouver Island)
Keogh RiverO? 3/2 (0.40) 3/3 (0.19) 1391 Ward and Slaney 1988
Nanaimo River? 2/1 (0.41) 3/1 (0.26) 228 Narver and Withler 1974
Nahmint RiverS 3/2 (0.71) 2/2 (0.19) 58 Narver 1974
Washington
Skagit RiverO 2/2 (0.48) 2/3 (0.33) n/ab WDFW 1994b
Deer CreekS 2/1 (0.95) 3/1 (0.05) n/a WDFW 1994b
Snohomish RiverO 2/2 (0.47) 2/3 (0.36) n/a WDFW 1994b
Green RiverO 2/2 (0.52) 2/3 (0.17) 100 Larson and Ward 1954
Puyallup RiverO 2/2 (0.61) 2/3 (0.28) n/a WDFW 1994b
Nisqually RiverO 2/2 (0.51) 2/3 (0.28) n/a WDFW 1994b
Hoh RiverO 2/2 (0.74) 2/3 (0.14) n/a WDFW 1994b
Quillayute RiverO 2/2 (0.46) 2/3 (0.40) n/a WDFW 1994b
Chehalis RiverO 2/2 (0.66) 2/3 (0.15) 100 Larson and Ward 1954
Columbia River Basin
Toutle RiverO 2/2 (0.73) 2/3 (0.11) 37 Howell et al. 1985
Cowlitz RiverO 2/2 (0.55) 2/3 (0.34) 56 Howell et al. 1985
Kalama RiverO 2/2 (0.65) 2/3 (0.18) 1363 Howell et al. 1985
Kalama RiverS 2/2 (0.67) 2/1 (0.17) 909 Howell et al. 1985
Willamette RiverO 2/2 (0.92) 3/2 (0.08) 141 Howell et al. 1985
Washougal RiverS 2/2 (0.71) 2/1 & 2/3 (0.14) 7 Howell et al. 1985
Wind RiverS 2/2 (0.58) 2/3 (0.26) 19 Howell et al. 1985
Klickitat RiverS 2/2 (0.75) 2/1 (0.14) 148 Howell et al. 1985
Deschutes RiverS 2/1 (0.35) 1/2 (0.22) 100 Howell et al. 1985
Yakima RiverS 2/1 (0.47) 2/1 (0.42) 64 BPA 1992
Wenatchee RiverS 2/1 (0.65) 3/1 & 3/2 (0.12) 17 Howell et al. 1985
Entiat RiverS 2/1 (0.88) 2/2 (0.12) 8 Howell et al. 1985
above Wells DamS 2/2 (0.41) 3/2 (0.24) 349 Mullan et al. 1992

Life history (frequency) Sample
Population Run typea Primary Secondary size Reference

Snake River Basin
Clearwater RiverS 2/1 (0.34) 2/2 (0.25) 510 Whitt 1954
S.F. Salmon RiverS 3/3 (0.49) 2/3 (0.31) 65 BPA 1992
Lemhi RiverS 2/2 (0.86) 2/1 (0.09) 353 BPA 1992
Oregon
Nehalem RiverO 2/2 (0.73) 2/3 (0.08) 310 Weber and Knispel 1977
Alsea RiverO 2/2 (0.52) 2/3 (0.22) 978 Chapman 1958
Siuslaw RiverO 2/2 (0.67) 2/3 (0.16) 125 Lindsay et al. 1991
Rogue RivercO 2/2 (0.60) 3/2 (0.17) 547 ODFW 1990
California
Klamath RiverS 2/1 (0.52) 1/1 (0.19) 391 Kesner and Barnhart 1972
Mad RiverO 2/2 (0.69) 2/1 (0.26) 35 Forsgren 1979
Jacoby CreekO 2/2 (0.50) 2/1 (0.26) 109 Harper 1980
Van Duzen RiverS 1/2 (0.62) 1/3 (0.29) 58 Puckett 1975
M.F. Eel RiverS 2/1 (0.45) 2/2 (0.33) 82 Puckett 1975
Sacramento RiverO? 2/1 (0.36) 2/2 (0.31) 83 Hallock 1989
Waddell CreekO 2/1 (0.39) 2/2 (0.30) 3,888 Shapovalov and Taft 1954

aO = Ocean maturing; S = Stream maturing.
bSample size not indicated in reference.
cAdults with half-pounder life history (spawning migrants, ODFW 1990) are included as age-2-ocean; these comprise 26% of the sample size, but had no effect on the ranking of most common life history patterns.

Determining total age at maturity for inland steelhead of the Columbia River Basin is complicated by variations in reporting methods. Generally, these fish spend a year in fresh water prior to spawning and this is not included in the age designation. Therefore, by adding 1 year after freshwater entry (indicated here as +1), most Columbia River inland steelhead are 4 years old at maturity (2/1+1). An exception is the Klickitat River; if these steelhead also spend a year in fresh water before spawning, they are dominated by 5-year-old spawners (2/2+1). Most of the available age data for Snake River steelhead are based on length frequency; smolt age is often assumed or not reported. The data that are available from scales show a high degree of variability in age structure, from 4-year-old spawners (2/1+1) in the Clearwater River (Whitt 1954) to 7 year-old spawners (3/3+1) in the South Fork Salmon River (BPA 1992).

Repeat Spawning

As noted above, most species of Oncorhynchus die after spawning, whereas O. mykiss may spawn more than once. The frequency of multiple spawnings is variable both within and among populations (Table 7). For North American steelhead populations north of Oregon, repeat spawning is relatively uncommon, and more than two spawning migrations is rare. In Oregon and California, the frequency of two spawning migrations is higher, but more than two spawning migrations is still unusual. The largest number of spawning migrations for which we found data was five, from the Siuslaw River, Oregon (Bali 1959). Iteroparous steelhead are predominately female.


Table 7. Repeat spawning frequency for selected steelhead populations. Data were collected from scale samples. Numbers indicate the proportion of steelhead collected in each study during a given spawning migration; for example,89% of the steelhead collected by Chapman (1958) in the Alsea River were on their first spawning migration. Populations are generally arranged from north to south.
Spawning migration Sample
Population Run typea 1 2 3 4 5 size Reference
British Columbia (mainland)
Babine RiverS 0.97 0.03 -- -- -- 121 Narver 1969
Cheakamus RiverO 0.69 0.26 0.05 -- -- 64 Withler 1966
Capilano RiverS 0.94 0.06 -- -- -- 99 Withler 1966
Seymour RiverO 0.95 0.05 -- -- -- 41 Withler 1966
Seymour RiverS 0.96 0.04 -- -- -- 45 Withler 1966
British Columbia (Fraser River Basin)
Coquitlam RiverO 0.95 0.03 0.02 -- -- 148 Withler 1966
Coquihalla RiverO 0.94 0.03 0.03 -- -- 31 Withler 1966
Coquihalla RiverS 0.94 0.06 <0.01 -- -- 158 Withler 1966
Washington
Skagit RiverO 0.92 0.07 0.01 -- -- n/ab WDFW 1994b
Snohomish RiverO 0.92 0.06 0.01 -- -- n/a WDFW 1994b
Green RiverO 0.93 0.07 <0.01 -- -- n/a WDFW 1994b
Puyallup RiverO 0.89 0.10 <0.01 -- -- n/a WDFW 1994b
Nisqually RiverO 0.93 0.06 0.01 -- -- n/a WDFW 1994b
Quillayute RiverO 0.91 0.07 0.01 -- -- n/a WDFW 1994b
Columbia River Basin
Cowlitz RiverO 0.96 0.04 -- -- -- 56 Howell et al. 1985
Toutle RiverO 0.89 0.05 0.05 -- -- 37 Howell et al. 1985
Kalama RiverO 0.93 0.06 <0.01 <0.01 -- 1,363 Howell et al. 1985
Kalama RiverS 0.94 0.06 <0.01 -- -- 909 Howell et al. 1985
Klickitat RiverS 0.97 0.02 0.01 -- -- 148 Howell et al. 1985
Oregon
Alsea RiverO 0.89 0.09 0.02 -- -- 1,223 Chapman 1958
Siuslaw RiverO 0.86 0.11 0.02 -- 0.01 125 Lindsay et al. 1991
Rogue RiverS 0.79 0.17 0.04 -- -- 4,058 ODFW 1994d
California
Mad RiverO 0.77 0.17 0.06 -- -- 35 Forsgren 1979
Jacoby CreekO 0.83 0.17 -- -- -- 109 Harper 1980
Sacramento RiverO 0.83 0.14 0.02 0.01 -- n/a Hallock 1989
Waddell CreekO 0.83 0.15 0.02 <0.01 -- 3,888 Shapovalov and Taft 1954

aO = Ocean maturing; S = Stream maturing.
bSample size not indicated in reference.


Resident Fish

Although we have defined steelhead as anadromous O. mykiss, there are areas where the separation between rainbow or redband trout and steelhead is obscured. In areas where anthropogenic barriers have isolated populations of O. mykiss, these landlocked populations could conceivably residualize (footnote 5) and, therefore, continue to exist in the nonanadromous form. Similarly, the mouths of some rivers in Oregon and California close seasonally, forming lagoons (during droughts, these rivers may remain closed for extended periods of time--even years). Again, landlocked O. mykiss in these systems could residualize. In some inland populations, growth rate can cause O. mykiss to residualize (Mullan et al. 1992); this apparently involves both fish that grow too quickly and those that grow too slowly.


Table of Contents