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

NOAA-NWFSC Tech Memo-24: Status Review of Coho Salmon
Discussion and Conclusions on ESU Determinations

In this section, we summarize evidence developed in the status review that is relevant to the two criteria that must be met for a population(s) to be considered an ESU and, hence, a species under the ESA: reproductive isolation and contribution to ecological/genetic diversity.

Reproductive Isolation

Coho salmon are generally believed to have strong tendencies to home to their natal stream (Donaldson and Allen 1958, Quinn and Tolson 1986, Sandercock 1991, Labelle 1992). Fish that do stray are most commonly found in streams near their natal stream (Shapovalov and Taft 1954, Jacobs 1988b, Labelle 1992). Because a low level of natural straying is expected to occur, some exchange of fish between adjacent rivers (within-ESU exchanges) probably occurs. The ESUs defined below are relatively large and identify regions among which we believe gene flow to be limited.

Genetic data (from studies of protein electrophoresis and DNA) provide further evidence for the reproductive isolation criterion. Genetic information presented in this status review indicate that there are several locations where genetic discontinuity/transition occurs. These locations are approximately: Punta Gorda, California; Cape Blanco, Oregon; the north Oregon coast/Columbia River; the area between the Chehalis and Queets Rivers; and areas between Puget Sound/Strait of Georgia and the upper Fraser River and Alaska. These genetic discontinuities or transition zones indicate a restriction of gene flow across these areas, suggesting a reasonable degree of reproductive isolation from each other.

There is also evidence for genetic heterogeneity within many of the areas defined by these boundary locations, such as the greater Olympic Peninsula/Puget Sound/Strait of Georgia area, and Oregon coast north of Cape Blanco. This heterogeneity suggests fairly high reproductive isolation of individual populations or groups of populations. In the case of the greater Olympic Peninsula/Puget Sound/Strait of Georgia area, Olympic Peninsula fish are both geographically and genetically distinct, indicating they form a major subgroup within the larger unit. In contrast, Oregon coast fish lack clear geographic patterns to the genetic structuring that would allow us to identify major subgroups within this area.

Ecological/Genetic Diversity

The physical environments to which west coast coho salmon are adapted, and the life history traits and genetic characteristics exhibited by these fish indicate a substantial degree of ecological and genetic diversity. Physical environments in the six ESUs summarized below are unique within the range of the species, and are expected to exert distinctive selection pressures. These environments range from the relatively dry climate in central California with strong and consistent upwelling offshore, to the extremely wet Olympic Peninsula with its snow-fed rivers.

Between these extremes, the environments in the other ESUs present their own particular challenges: Oregon coastal rivers receive considerable rain but little snowmelt and flow into an unpredictable ocean environment; fish inhabiting Columbia River tributaries must navigate one of the largest rivers along the West Coast of North America; and coho salmon in the Puget Sound/Strait of Georgia inhabit glacially-fed rivers which flow into a productive and stable marine environment. Ocean migration patterns, as inferred from marine CWT recovery patterns, are also distinctive in each of the six ESUs. The distinctiveness of these patterns indicates the unique adaptations made by the six groups of fish to their environments.


Based on information discussed above and summarized below, the BRT identified six ESUs for west coast coho salmon populations (Fig. 34). The proposed ESUs are briefly described and characterized below.

1) Central California coast--The geographic boundaries of this ESU extend from Punta Gorda in northern California south to the San Lorenzo River in central California, and include tributaries to San Francisco Bay, excluding the Sacramento-San Joaquin River system. These boundaries encompass coho salmon populations from the present southern extreme range of the species. This area is characterized by very erosive soils, and redwood forest is the dominant vegetation for these coastal drainages. Precipitation is much lower and less prolonged here than in areas to the north, and elevated stream temperatures (>20° C) are common in the summer. Freshwater fishes in this area are derived from the Sacramento River fauna. Coastal upwelling in this region is strong and consistent, resulting in a relatively productive nearshore marine environment.

Both run and spawn timing of coho salmon in this region are very late (both peaking in January), with little time spent in freshwater between river entry and spawning. This compressed adult freshwater residency appears to coincide with the single, brief peak of river flow characteristic of this area. Coho salmon released from the Warm Springs Hatchery on the Russian River have a much more southern distribution than fish released north of the ESU. Whether this pattern reflects a unique migration pattern for the ESU as a whole or just the southerly location of the hatchery is not known. However, genetic data indicate that most samples from this region differ substantially from coho salmon north of Punta Gorda.

The northern boundary of this ESU, Punta Gorda, was selected primarily because of the clear shift in terrestrial and marine environments that occurs in the vicinity of Punta Gorda and Cape Mendocino. The freshwater environment of the Mattole River, which enters the Pacific Ocean between these two points, is more similar to rivers north of Cape Mendocino, and coho salmon populations from the Mattole River are likewise more similar to populations farther north. However, there is scant information on coho salmon from the numerous small basins between Punta Gorda and the Ten Mile River, some 90 km to the south, which might indicate their greater similarity to populations to the north or south.

Available information indicates that the San Lorenzo River currently is the southernmost population of coho salmon, and this is the southern geographic boundary for the proposed ESU. However, it should be recognized that any coho salmon found spawning south of the San Lorenzo River that have not resulted from stock transfers should also be considered part of this ESU.

2) Southern Oregon/northern California coasts--This ESU includes coho salmon from Cape Blanco in southern Oregon to Punta Gorda in northern California. Geologically, this region includes the Klamath Mountains Geologic Province, which has soils that are not as erosive as those of the Franciscan Formation to the south. Dominant vegetation along the coast is redwood forest, while some interior basins are much drier than surrounding areas and are characterized by many endemic plant species. Elevated stream temperatures are a factor in some areas, but not to the extent that they are in areas south of Punta Gorda.

Rivers in this ESU are relatively long compared to those to the south. With the exception of major river basins such as the Rogue and Klamath, most streams in this region have short duration of peak flows and relatively low flows given both peak flow levels and basin sizes, compared to rivers farther north. Freshwater fishes include elements of the Sacramento River fauna as well as from the Klamath-Rogue ichthyofaunal region. Strong and consistent coastal upwelling begins around Cape Blanco and continues south into central California, resulting in a relatively productive nearshore marine environment. In contrast to coho salmon from north of Cape Blanco, which are most frequently captured off the Oregon coast, coho salmon from this region are captured primarily in California waters.

Genetic data indicate that most samples from this region differ substantially from coho salmon from south of Punta Gorda. In general, populations from southern Oregon also differ from coastal Oregon populations north of Cape Blanco. However, some samples from the Rogue River show an unexplained genetic affinity to samples from outside the region, including some from the Columbia River. In addition, a sample from the Elk River (just south of Cape Blanco) clustered with samples from the Umpqua River.

The southern boundary of this ESU is farther south than the boundary designated for the Klamath Mountains Province steelhead ESU, which includes the Klamath River but not drainages to the south (Busby et al. 1994). Both the steelhead and coho salmon ESUs share the northern boundary of Cape Blanco. Although the Klamath River (inclusive) serves as the southern boundary for the Klamath Mountains Geological Province and for freshwater fish faunas, major changes in ocean currents and environmental characteristics, as well as the southern limit of the steelhead half-pounder life history strategy, occur at Cape Mendocino/Punta Gorda.

Consequently, the southern limit of the steelhead ESU was based primarily on strong genetic discontinuity between Klamath River steelhead and steelhead populations to the south (Busby et al. 1994). In contrast, Punta Gorda serves as the southern boundary of the southern Oregon/northern California coho salmon ESU because of the strong environmental transition at Punta Gorda, and because genetic data indicate Punta Gorda, rather than the Klamath River, as an approximate transition area for coho salmon.

3) Oregon coast--This ESU covers much of the Oregon coast, from Cape Blanco to the mouth of the Columbia River, an area with considerable physical diversity ranging from extensive sand dunes to rocky outcrops. With the exception of the Umpqua River, which extends through the Coast Range to drain the Cascade Mountains, rivers in this ESU have their headwaters in the Coast Range. These rivers have a single peak of flow in December or January and relatively low flow in late summer. Upwelling north of Cape Blanco is much less consistent and weaker than in areas south of Cape Blanco. Sitka spruce is the dominant coastal vegetation and extends to Alaska. Precipitation in coastal Oregon is higher than in southern Oregon/northern California but lower than on the Olympic Peninsula. Oregon coast coho salmon are caught primarily in Oregon marine waters and have a slightly earlier adult run timing than populations farther south.

Genetic data indicate that Oregon coast coho salmon north of Cape Blanco form a discrete group, although there is evidence of differentiation within this area. However, because there is no clear geographic pattern to the differentiation, the area is considered to be a single ESU with relatively high heterogeneity.

4) Lower Columbia River/southwest Washington coast--The BRT concluded that historically, there was probably an ESU that included coho salmon from all tributaries of the Columbia River below the Klickitat River on the Washington side and below the Deschutes River on the Oregon side (including the Willamette River as far as the Willamette Falls), as well as coastal drainages in southwest Washington between the Columbia River and Point Grenville (between the Copalis and Quinault Rivers). The Columbia River estuary, Willapa Bay, and Grays Harbor all have extensive intertidal mud and sandflats and similar estuarine fish faunas, and they differ substantially from estuaries to the north and south. This similarity results from the shared geology of the area and the transportation of Columbia River sediments northward along the Washington coast.

Rivers draining into the Columbia River have their headwaters in increasingly drier areas moving from west to east. Columbia River tributaries that drain the Cascade Mountains have proportionally higher flows in late summer and early fall than rivers on the Oregon coast. Coded- wire-tag (CWT) data indicate a distinctive oceanic distribution pattern for Columbia River coho salmon, with a higher percentage of recoveries from Washington compared to recovery patterns for Oregon coastal stocks and a much lower percentage of recoveries from British Columbia compared to recovery patterns for Washington coastal populations, including populations from the southwest Washington coast. Genetic data indicate that Columbia River coho salmon are distinct from coastal Oregon populations but similar to populations from several coastal streams in southwest Washington.

The question of where the southwest Washington coast (areas draining Willapa Bay and Grays Harbor) should be placed with respect to coho salmon ESUs prompted considerable debate within the BRT: Should it be part of the lower Columbia River ESU, the Olympic Peninsula ESU, or its own ESU? The southwest Washington coast has many traits in common with the lower Columbia River: the hydrology, topography, and climate of river basins in the two areas are similar; Grays Harbor, Willapa Bay and the Columbia River estuary are physically and biologically similar; and coho salmon from southwest Washington are genetically most similar to Columbia River fish. However, the southwest Washington coast also shares traits with the Olympic Peninsula: tributaries draining the north side of the Chehalis River Basin share the same hydrology, topography, and climate as Olympic Peninsula rivers, and marine CWT recovery patterns of coho salmon released from southwest Washington coast hatcheries are more similar to those released from Olympic Peninsula facilities than to releases from the Columbia River.

The southwest Washington coast also has distinctive features--it forms the transition zone between the moderately wet Oregon coast and the extremely wet Olympic Peninsula, and although its rivers share many characteristics with lower Columbia River tributaries, they drain directly into the Pacific Ocean. Why the CWT recovery patterns of coho salmon from southwest Washington do not follow the genetic patterns remains to be determined. It was concluded that because of the similarity of southwest Washington coast and the lower Columbia River, and the genetic similarity of coho salmon from the two areas, southwest Washington coast should be part of the lower Columbia River ESU.

Once it was decided that coho salmon from the southwest Washington coast and the lower Columbia River form a single ESU, the location of its border with the Olympic Peninsula had to be identified. This also prompted debate within the BRT because of the broad transition zone between southwest Washington and the Olympic Peninsula. In particular, tributaries draining the northern part of the Chehalis River Basin are typical of Olympic Peninsula basins with respect to hydrology, topography, and climate, while in most other respects the Chehalis River Basin is physically and biologically similar to other southwest Washington coast basins. In addition, river basins between the Chehalis and Quinault Rivers (Humptulips, Copalis, and Moclips Rivers) drain low-elevation coastal areas and have flow characteristics typical of rivers farther south. Although some Chehalis River tributaries share traits with Olympic Peninsula rivers, BRT members ultimately decided that the region between Point Grenville and Grays Harbor was most similar to southwest Washington, so the northern boundary of the lower Columbia River/southwest Washington coast ESU was placed at Point Grenville.

In the status review for lower Columbia River coho salmon (excluding the Clackamas River), NMFS concluded that, historically, at least one ESU of coho salmon probably occurred in the lower Columbia River Basin, but that the agency was unable to identify any remaining natural populations that warranted protection under the ESA (Johnson et al. 1991, NMFS 1991a). This status review has not uncovered substantial new information on coho salmon populations considered by that earlier status review. However, the BRT considered further information regarding coho salmon in the Clackamas River and along the southwest Washington coast. Evidence of extensive hatchery production and outplanting and high harvest rates led the BRT to conclude that, similar to the lower Columbia River, we cannot at this time identify any remaining natural populations of coho salmon along the Washington coast south of Point Grenville that warrant protection under the ESA.

Evidence regarding the history of hatchery influence on late-run Clackamas River coho salmon is not as clear, and the BRT could not reach a definite conclusion as to whether these fish represent the historical lower Columbia River/southwest Washington ESU.

5) Olympic Peninsula--The geographic boundaries of this ESU are entirely within Washington, including coastal drainages from Point Grenville to Salt Creek. This region is characterized by high levels of precipitation and streams with cold water, high average flows, and a relatively long duration of peak flows, including a second peak later in the year resulting from snow melt. In contrast to inland areas, such as Puget Sound, where western hemlock is the dominant forest cover, vegetation in this region is dominated by the Sitka spruce. Coho salmon from the Olympic Peninsula ESU have a more northerly ocean distribution than populations from the Columbia River or coastal regions in Oregon, and they are more commonly captured in Canadian waters than are coho salmon from the Puget Sound region. Genetic data show that coho salmon from this region are distinct from populations to the south and somewhat differentiated from populations in the Puget Sound area.

Like the southern boundary of this ESU discussed earlier, the eastern boundary of the Olympic Peninsula ESU also overlays an extended transition zone between the extremely wet Olympic Peninsula and the much drier Puget Sound/Strait of Georgia. The transition point between the wet Olympic Peninsula and the rainshadow farther east is thought to occur east of the Elwha River. However, the Elwha River is physically more similar to the Dungeness River than it is to basins farther west; the Elwha and Dungeness Rivers are both relatively long and begin in alpine areas of the Olympic Mountains, while rivers west of the Elwha River are much shorter, draining the low ridge that separates the Soleduck River from the Strait of Juan de Fuca.

Coded-wire-tag recovery patterns of coho salmon released from the Elwha River are also typical of those from stocks to the east. However, Dungeness River coho salmon have been extensively planted in the Elwha River, and the effect of these plants on Elwha River CWT recovery patterns is unknown. Although the climate of the Elwha River Basin may be considered more similar to that of the Olympic Peninsula than to areas farther east, we felt the Elwha River's physical similarity to the Dungeness River and the similarity of CWT recovery patterns from the two basins provided sufficient evidence to keep the two rivers in the same ESU. Therefore, the boundary between the Olympic Peninsula and Puget Sound/Strait of Georgia ESUs was placed at Salt Creek, the basin immediately west of the Elwha River.

The west coast of Vancouver Island in British Columbia shares many of the physical and environmental features of the Olympic Peninsula ESU. However, we have little biological information for coho salmon from this area. The Strait of Juan de Fuca is potentially a strong isolating mechanism, and although comparable data are not available for coho salmon, genetic data for chinook salmon show that populations from the west coast of Vancouver Island differ genetically from those on the northern Washington coast. Therefore, until more complete information becomes available, we concluded that the geographic boundaries of this ESU do not extend across the Strait of Juan de Fuca.

6) Puget Sound/Strait of Georgia--This ESU includes coho salmon from drainages of Puget Sound and Hood Canal, the eastern Olympic Peninsula (east of Salt Creek), and the Strait of Georgia from the eastern side of Vancouver Island (north to and including Campbell River) and the British Columbia mainland (north to and including Powell River), excluding the upper Fraser River above Hope. This region is drier than the rainforest area of the western Olympic Peninsula and the west side of Vancouver Island and is dominated by western hemlock forests. Streams are similar to those of the Olympic Peninsula, being characterized by cold water, high average flows, a relatively long duration of peak flows, and a second snow-melt peak, although flow levels per basin area are much lower than in the Olympic Peninsula. Genetic and CWT data both show substantial differences between coho salmon from this region and those from the Columbia River and more southern coasts, and more modest differences between coho salmon from this region and populations from the Olympic Peninsula. Coho salmon samples from Puget Sound and the Strait of Georgia form a coherent genetic cluster. The few samples we have examined from Alaska and the upper Fraser River showed substantial genetic differences from all Washington, Oregon, and California populations.

Drainages entering the Strait of Georgia from both sides share many of the physical and environmental features that characterize the Puget Sound area. From the Queen Charlotte Strait north, the prevalence of coho salmon smolting at age 2 (rather than age 1) begins to increase. At about this point, the British Columbia mainland assumes more of the physical and environmental characteristics of the outer coast of Vancouver Island. However, genetic and life history data for populations between the Strait of Georgia and Queen Charlotte Strait are insufficient to indicate relationships between coho salmon in this area and those to the north and south. Therefore, we concluded that until further information is available, the geographic boundaries of this ESU extend into Canada to include drainages from both sides of the Strait of Georgia as far as the north end of the strait.

Additional Comments

Historically, coho salmon have been reported to occur in U.S. waters outside of the geographic areas covered by proposed ESUs, and a brief discussion of this topic is necessary.

It is generally believed that at least some coho salmon populations may have existed in the Sacramento River Basin prior to 1880 (Brown et al. 1994, Bryant 1994). After that time, placer mining, dams, diversions, and other perturbations caused extreme habitat degradation throughout the basin, and any coho salmon there became extinct (Brown and Moyle 1991). In recent decades, attempts have been made to reintroduce coho salmon to the basin, but these attempts have not been successful. Intermittent reports of small numbers of coho salmon in the Sacramento River are generally attributed to strays or to remnants of these stocking programs. We found no evidence that coho salmon eligible for ESA consideration presently occur in the Sacramento River.

Although several tributaries in the upper Columbia River Basin, including the Snake River, once supported coho salmon runs, we are not aware of any native coho salmon production in the upper basin at the present time. Consequently, although the petitioners included Idaho coho salmon in the petition, there are no coho salmon in Idaho which would qualify for listing under ESA. Columbia River stock summary reports (Columbia River Coordinated Information System (CIS) 1992) identify no coho salmon of native origin in this region except in the Hood and Deschutes Rivers in Oregon. According to Nehlsen et al. (1991), all coho salmon above Bonneville Dam are extinct except those spawning in the Hood River. Both the Hood and Deschutes Rivers have had extensive outplanting of hatchery coho salmon, and no recent natural production estimates are available.

Artificial Propagation

The effects of artificial propagation can be relevant to ESA listing determinations, as discussed in the Introduction. In this section, we present information on the magnitude and patterns of artificial propagation of west coast coho salmon. The importance of this information to risk assessments for each ESU will be discussed in the subsequent section.

Patterns of Artificial Propagation and Stock Transfers

Artificial propagation of coho salmon and stock transfers, both between production facilities and out-of-basin, off-station plants, have been and continue to be common within the petitioned range. The nature and magnitude of these transfers varies by ESUs, as discussed below. However, the true impact of these transfers and plants remain unclear for two reasons.

First, because there have been transfers and planting of fish which were unrecorded, the best compilation of hatchery data will remain incomplete. Second, there has generally been little evaluation of the success of hatchery plants and stocks, especially prior to the widespread application of CWTs. Accordingly, although there are fairly good records of which fish were released, there are almost no comprehensive reports of which fish came back. In addition, the pedigrees of many stocks, or individual year classes of stocks, change over time as additional stocks are deliberately or unintentionally added to them, further complicating the compilation of hatchery records. Despite these complications and limitations, however, comparing the relative magnitude, pattern, and frequency of stock transfers and plants remains the best approximation of the potential for hatchery impacts to west coast coho salmon ESUs.

The following compilations of coho salmon hatchery production information are not complete, but are sufficiently thorough to indicate trends in the magnitude, frequency, and types of stock transfers between ESUs. Table 5 presents the average annual release of coho salmon from selected production facilities in 1987-91, summarized by ESU. Figures 35-37 display coho salmon stocks used by selected hatcheries or production facilities within the geographic range of the six ESUs, while Figures 38-40 present out-of-basin stocks planted in selected river basins in ESUs in Oregon and Washington.

The actual data for out-of-basin plants in Oregon and Washington are presented in Appendix Tables E-1, E-2. Out-of-basin plants for California were not compiled because most coho salmon released from California Department of Fish and Game (CDFG) hatcheries occur within basin, and planting records for private production facilities are incomplete. Figure 41 indicates stocks used at selected saltwater-release facilities in California, Oregon and Washington.

Table 5. Average annual releases of coho salmon juveniles (fry and smolts) by evolutionarily significant unit (ESU) from selected production facilities in California, Oregon, Washington, and British Columbia, during release years 1987-91. Releases from facilities for which only one year of data were available were not used. Releases from earlier time periods were used when information from 1987-91 was not available. Production facilities followed by an asterisk (*) are no longer in operation.

5-Year Average

Central California coast ESU
Monterey Bay Salmon and Trout25,764 Streig 1993
Silver King* 95,074a Streig 1991
Noyo Eggtake Station 107,918 Poe 1988; Grass 1990-92b
Warm Springs 123,157 Gunter 1988b, 1990a-91; Cartwright 1992
        Total 351,913
Southern Oregon/northern California coasts ESU
Cochran Ponds (Humboldt Fish Action Council) 35,931b Hull et al. 1989
Mad River 372,863 Barngrover 1988, 1990a-91, Gallagher 1992
Prairie Cr.* 89,009c NRC 1995
Trinity 496,813 Bedell 1990a-91b
Iron Gate 147,272 Hiser 1990-92
Cole Rivers 271,492 R. Beamesderfer 1994 App.
        Total 1,413,380
Oregon coast ESU
Butte Fallsd 379,353 R. Beamesderfer 1994 App.
Bandon 396,521 R. Beamesderfer 1994 App.
Oregon Aqua Foods* and Anadromous, Inc.*, Coos Bay 2,173,625e Wagoner et al. 1990
Rock Cr. (Umpqua) 328,573 R. Beamesderfer 1994 App.
Fall Cr. 1,360,284 R. Beamesderfer 1994 App.
Oregon Aqua Foods*, Yaquina Bay4,840,000f Borgerson et al. 1991
Salmon River 1,305,576 R. Beamesderfer 1994 App.
Cedar Cr. (Nestucca) 42,680 R. Beamesderfer 1994 App.
Trask 1,087,587 R. Beamesderfer 1994 App.
Nehalem 786,164 R. Beamesderfer 1994 App.
        Total 12,700,363
Lower Columbia River/southwest Washington coast ESU
Klaskanine 1,146,128 R. Beamesderfer 1994 App.
Clatsop Economic Development Committee 1,097,210g S. Allen 1994 App.
Big Creek 675,593 R. Beamesderfer 1994 App.
Eagle Cr. NFH 1,859,477 S. Allen 1994 App.
Sandy 1,049,055 R. Beamesderfer 1994 App.
Bonneville 2,481,746 R. Beamesderfer 1994 App.
Cascade 1,417,881 R. Beamesderfer 1994 App.
Oxbow/Wahkeena Pond 1,184,569 R. Beamesderfer 1994 App.
Klickitat 1,532,498g S. Allen 1994 App.
Willard 3,269,220g S. Allen 1994 App.
Washougal 3,885,612g S. Allen 1994 App.
Speelyai 1,356,904g S. Allen 1994 App.
Lewis 6,180,000g S. Allen 1994 App.
Kalama Falls 990,000g S. Allen 1994 App.
Lower Kalama 831,605g S. Allen 1994 App.
Toutle 478,090g S. Allen 1994 App.
Cowlitz 7,956,089g S. Allen 1994 App.
Elokomin 2,013,032gS. Allen 1994 App.
Grays River 744,655g S. Allen 1994 App.
Sea Resources 125,500g S. Allen 1994 App.
Naselle 3,028,580 NRC 1995
Nemah 1,148,940 NRC 1995
Willapa 1,140,865 NRC 1995
Willapa Bay Gillnetters 529,167h NRC 1995
Westport Pens 154,615i NRC 1995
Aberdeen Net Pens 103,574j WDFW 1994b
Ocean Shores Pens* 100,000k WDFW 1994b
Pacific Trollers (Grays Harbor)454,931 NRC 1995
Simpson 4,411,488 NRC 1995
Humptulips 3,667,491 NRC 1995
        Total 55,214,270

5-Year Average


Olympic Peninsula ESU
Quinault NFH 1,167,595 NRC 1995
Quinault Lake Tribal Pens 1,207,296 NRC 1995
Soleduck 1,714,747 NRC 1995
Makah NFH 684,405 NRC 1995
        Total 4,774,043
Puget Sound/Strait of Georgia ESU
Lower Elwha 743,395 NRC 1995
Dungeness 704,751 NRC 1995
Port Gamble Pens 373,567 WDFW 1994b
Quilcene Bay Pens 200,482 WDFW 1994b
Quilcene NFH 1,155,107 NRC 1995
Hood Canal 131,400 NRC 1995
George Adams 932,160 NRC 1995
Squaxin Coop 1,049,758 NRC 1995
S. Sound Pens 1,402,783 NRC 1995
Garrison Springs 883,986 NRC 1995
Fox Island Pens 270,789 WDFW 1994b
Minter Cr. 3,723,359 NRC 1995
Kalama Cr. 823,793 NRC 1995
Puyallup 2,083,898 NRC 1995
Green River 1,316,681 NRC 1995
Crisp Cr. (Green R. satellite)1,957,936 NRC 1995
Elliott Bay Pens 137,312 WDFW 1994b
Suquamish Pens (Agate Pass) 412,826 WDFW 1994b
Grovers Cr. 190,914 NRC 1995
Issaquah 2,205,195 NRC 1995
Univ. Washington 49,434 NRC 1995
Skykomish 1,270,302 NRC 1995
Tulalip 820,224 NRC 1995
S. Whidbey Salmon 105,833 NRC 1995
Baker 110,670 C. Feldman 1994 App.
Skagit 308,622 NRC 1995
Swinomish Channel 980,055 NRC 1995
Lummi Sea Ponds 801,277i NRC 1995
Skookum Cr. 1,628,901 NRC 1995
Nooksack 2,497,740 NRC 1995
British Columbia insidel production facilities 13,918,384e Cross et al. 1991
        Total 43,191,534
Total for all areas 117,645,503

aAverage from 4 years of data: 1984-88.
bAverage from 2 years of data: 1987-88.
cAverage from 4 years of data: 1987-88, 1990-91.
dButte Falls Hatchery, although located within the southern Oregon/northern California coast ESU, is considered to be part of the Oregon coast ESU because it rears Umpqua River coho salmon and releases them into the Umpqua River.
eAverage from 5 years of data: 1985-89.
fAverage from 5 years of data: 1986-90.
gSmolt releases only.
hAverage from 3 years of data: 1987, 1990-91.
iAverage from 4 years of data: 1988-91.
jAverage from 3 years of data: 1989-91.
kAverage from 2 years of data: 1990-91.
l"Inside" production area includes the following: Johnstone Strait, Georgia Strait--Vancouver Island, Strait of Georgia--mainland, and Lower Fraser.

Unless specified, these tables and figures were compiled from the following sources: Murray 1959, 1961-62b, 1964-68; U.S. Dep. Interior 1959; Wallis 1961a-c, 1963a-64b, 1966; Riley 1967a-68, 1970; Bedell 1970a-b, 1972a-b, 1974-76, 1978-80, 1982-84, 1987a-b, 1990a- 91b; Marshall 1970; Arnold 1972a-b, 1974-75, 1977a-c; Will 1973a-c, 1975-76, 1978a-79; Hiser 1979, 1982a-83, 1985a-b, 1987a-b, 1990-93; Snyder and Sanders 1979; Willis 1979; Ducey 1980, 1982a-83; Sanders 1980, 1982a-83b; Estey 1982a-84, 1986; Grass 1982, 1990-92c; ODFW 1982, 1993a; Barngrover 1983, 1986-88, 1990a-92; Houston 1983; Poe 1984, 1988; USFWS 1985; Kenworthy 1986a-b; Milligan 1986-87; Gunter 1988a-b, 1990a-91; Zajac 1988-92; Cross et al. 1991; Cartwright 1992; Fuss et al. 1993; Ramsden 1993;

S. Allen 1994 App.; R. Beamesderfer 1994 App.; R. Berry 1994 App.; Bryant 1994; D. Dailey 1994 App.; H. Fuss 1994 App.; J. Harp 1994 App.; WDFW 1994c; and Natural Resources Consultants (NRC) 1995.

Central California coast ESU--Compared to areas farther north, the few hatcheries located in the central California ESU are relatively small, with approximately 350,000 coho salmon released annually between 1987 and 1991. The largest production facilities each release about 100,000 coho salmon each year (Table 5). There has been considerable movement of coho salmon between hatcheries or egg-taking stations in central and northern California, with the fish eventually outplanted in either area. These transfers primarily involved California hatchery stocks and may have also included Oregon and Washington stocks that were not identified as such.

Southern Oregon/northern California coasts ESU--Hatchery production of coho salmon in the southern Oregon/northern California coasts ESU is greater than in the central California coast ESU, but considerably less than in more northerly ESUs. Large hatcheries within this ESU (e.g., Mad, Trinity) have released 400,000-600,000 coho salmon annually in recent years, with total annual production at approximately 1.4 million coho salmon between 1987 and 1991 (). Aside from considerable movement of coho salmon between hatcheries or egg-taking stations in central and northern California, northern California hatcheries have also received fairly large transplants of coho salmon from hatcheries in areas outside the ESU, including the Oregon coast, lower Columbia River/southwest Washington coast, and Puget Sound/Strait of Georgia (Fig. 35). In contrast, Cole Rivers Hatchery (Rogue River), appears to have relied almost exclusively on native stocks (Fig.35). The frequency and magnitude of out-of- basin plants in the southern Oregon/northern California coasts ESU appears to be relatively low (Fig. 38), although records are incomplete.

Oregon coast ESU--Each large public hatchery along the Oregon coast (Trask, Salmon, Fall Creek) has released just over 1 million coho salmon annually between 1987 and 1991. In recent years, private production facilities released between 2.2 and 4.8 million coho salmon annually (Wagoner et al. 1990, Borgerson et al. 1991), for total annual production in the Oregon coast ESU of about 12.7 million coho salmon (Table 5). Most transfers of coho salmon between public hatcheries used stocks from within the area, most commonly from Tenmile Lake and Trask (Fig. 35). Some transfers into these hatcheries have occurred from the lower Columbia River/southwest Washington coast, but these were relatively infrequent and minor. In contrast, private Oregon coast hatcheries began coho salmon production using Puget Sound stocks, which were later mixed with Oregon coastal stocks (Fig. 41) (Wagoner et al. 1990, Borgerson et al. 1991). These private facilities are presently not in operation. Most outplants of coho salmon into Oregon coastal rivers have used Oregon coastal stocks, with outplants of stocks from outside the Oregon coast being relatively small and infrequent (Fig. 38, Appendix Tables E-1, E-2). Plants of sexually mature adults were common within this ESU in the late 1960s and early 1970s, and they used stocks from within and outside of the Oregon coast (Appendix Table E-2). Recipient basins of stock transfers along the Oregon coast, either to hatcheries or off-station plants, are generally those closest to the source stock.

Lower Columbia River/southwest Washington coast ESU--Hatchery production of coho salmon in the lower Columbia River/southwest Washington coast ESU far exceeds that of any other area with respect to the number of hatcheries and quantities of fish produced; total annual production was just over 55 million fish between 1987 and 1991 (Table 5). Many hatcheries within this ESU released 1-3 million smolts annually, with the two largest hatcheries, Cowlitz and Lewis, releasing an average of 6-7 million smolts annually (Table 5). Coho salmon production from Washington-side Columbia River hatcheries (29.4 million smolts per year) provides about 53% of the total annual production, with the remainder split between Oregon-side Columbia River (10.9 million smolts) and southwest Washington coast (14.7 million fish) facilities.

Extensive stock transfers have occurred within the lower Columbia River/southwest Washington coast ESU. Most transfers of coho salmon have used stocks from within the ESU, although transfers from outside the ESU have also occurred, including those from the Oregon coast, Olympic Peninsula, and Puget Sound/Strait of Georgia ESUs (Fig. 36). Outplanting records show a similar pattern to transfers between hatcheries, with extensive use of within-ESU stocks, in addition to less frequent use of stocks from the same three ESUs (Fig. 39). Most movement of coho salmon, either as hatchery transfers or off-station releases, has occurred within each of the three areas of this ESU (Oregon-side Columbia River, Washington-side Columbia River, and southwest Washington coast), with little movement of fish among the three areas. The Clackamas River has also been extensively outplanted with early-running Columbia River stocks and was outplanted with coho salmon from the Oregon coast in 1967 (Cramer and Cramer 1994).

Olympic Peninsula ESU--Coho salmon production facilities in the Olympic Peninsula ESU each produce 1-2 million coho salmon annually, with a total annual release of 4.8 million coho salmon between 1987 and 1991 (Table 5). Natural production in the ESU is relatively high, due in large part to nearly pristine habitat within the Olympic National Park. Hatcheries within this ESU have relied on native stocks, with contributions of stocks from the lower Columbia River/southwest Washington and Puget Sound/Strait of Georgia (Fig. 37). Olympic Peninsula drainages are primarily outplanted with Olympic Peninsula stocks; however, some outplants of fish from adjacent ESUs have also occurred (Fig. 40).

Puget Sound/Strait of Georgia ESU--Hatchery production in the Puget Sound/Strait of Georgia ESU is extensive, with 43 million coho salmon released annually in 1987- 91 (Table 5). Many of the larger Washington hatcheries (Minter Creek, Puyallup, Crisp Creek, Issaquah, and Nooksack) each recently released 2-4 million coho salmon annually, while the larger net-pen facilities (South Sound Net Pens and Squaxin Coop) released a total of over 2 million coho salmon annually (Table 5). Net-pen operations, unlike hatcheries, generally do not attempt to attract or recover returning adults. Most coho salmon released from Puget Sound net- pen facilities are of local origin and produced at local hatcheries, with Skykomish and Skagit River stocks most commonly used. About one-third of the artificial production in this ESU occurs in British Columbia. In addition to extensive hatchery production in the ESU, there is also considerable natural production.

Stock transfers and outplants within the Washington portion of the Puget Sound/Strait of Georgia ESU have been extensive. Most stocks involved have been derived from within the geographic boundaries of the ESU, but stocks from the lower Columbia River/southwest Washington and Olympic Peninsula have also been used (Figs. 37, 40-41). The stocks which are most commonly transferred between Washington hatcheries in the area include Skagit, Skykomish, Green River, and Dungeness, while the most commonly planted out-of-basin stocks include Samish, Skagit, Skykomish, Green River, Minter Creek, Puyallup, and George Adams. In British Columbia, stock transfers have also occurred (Aro 1979), although many hatcheries there have relied primarily on native stocks. In some areas of British Columbia, virtually all production releases are marked (Hilborn and Winton 1993), and there is an emphasis on using wild broodstock (Miller 1990).

Run Timing

A common phenomenon in hatchery populations is advancement and compression of run timing, and these changes can affect future generations of naturally spawning fish. Fry of early-spawning adults generally hatch earlier, grow faster, and can displace fry of later-spawning natural fish. Conversely, early spawning coho salmon redds are more prone to destruction from early fall floods. Consequently, early-spawning individuals may be unable to establish permanent, self-sustaining populations but may nevertheless adversely affect existing natural populations (Nickelson et al. 1986). A recent study found that over a period of 13 years, spawn timing of coho salmon at five Washington hatcheries decreased from 10 to 3 weeks, causing the period of return to hatcheries to decrease by half (Fig. 42) (Flagg et al. 1995).

Juvenile Outplants

Another common hatchery practice with coho salmon is release of excess hatchery juveniles into natural habitat as fry or parr (Flagg et al. 1995). Outplanting large numbers of large hatchery juveniles into streams already occupied by naturally produced juveniles may place the resident fish at a competitive disadvantage and force them into marginal habitats with low survival potential (Chapman 1962, Solazzi et al. 1990). The overstocked hatchery fish may also have low survival potential under these conditions and fail to return as adults in adequate numbers. This practice may cause streams planted with presmolts to remain below juvenile carrying capacity (Nickelson et al. 1986). Stocking of over 10 million presmolts annually into small streams of Columbia River tributaries has occurred (ODFW 1991, WDF 1991), far exceeding expected carrying capacity.

Table of Contents