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

NOAA-NMFS-NWFSC TM-33: Sockeye Salmon Status Review (cont)
Back to Previous Section

Discussion and Conclusions for ESU Determinations

Based on genetic, life-history, and ecological evidence presented above, the BRT identified six ESUs for sockeye salmon in the state of Washington; fish from one of these ESUs spawns in British Columbia, rears in a lake that straddles the U.S./Canadian border, and migrates to and from the sea through Washington. In addition, sockeye salmon from Big Bear Creek in the Lake Washington/Lake Sammamish Basin of Washington were provisionally identified as an ESU. There was insufficient evidence to determine the ESU status for two additional groupings of sockeye salmon, one in Washington and the other in Oregon. In the following discussion, we describe these ESUs and outline the issues that were important to the BRT in making each ESU determination.

Status of Transplanted Populations

Available artificial propagation data and transplantation records provide evidence that within the Lake Washington Basin, the current Cedar River and Issaquah Creek sockeye salmon are introduced populations, originating from transplants of Baker Lake stock that had been perpetuated at the Birdsview Hatchery. Allozyme data based on 29 allozyme loci indicate a close genetic relationship between Cedar River, Lake Washington beach-spawning, and Baker River sockeye salmon. Similar artificial propagation data and transplantation records provide evidence that spawning aggregations of sockeye salmon that are annually seen in Icicle Creek and the Methow and Entiat Rivers in the Columbia River Basin are the result of transplants that occurred during the Grand Coulee Fish Maintenance Project.

Waples (1991a, p. 18-19), in the NMFS "Definition of Species" paper, suggested that

In general, populations resulting from the introduction of fish into a local area not occupied by the biological species... are probably not ESU's because they do not contribute to maintaining diversity of the species in its native habitat.
However, Waples (1991a, p. 19) went on to say that
Some introduced populations should not be excluded from ESA consideration, and these include populations occupying habitat that is ecologically similar and geographically proximate to the source population, and those that represent the only remaining component of the native gene pool.

These ecological and geographic criteria do not appear to apply in the present case. Although introduced sockeye salmon populations in the Lake Washington/Lake Sammamish system and native Baker River sockeye salmon are both geographically within Puget Sound, the natural ecosystems are dissimilar. Baker Lake is a cold-water, glacially influenced, low productivity mountain lake system, and the Lake Washington/Lake Sammamish drainage is a relatively warm-water, high productivity coastal lake system.

The BRT concluded that historical records, stocking history, and genetic data indicate that sockeye salmon that spawn in the Cedar River, Issaquah Creek, on lakeshore beaches in Lake Washington (in the Lake Washington drainage), and in Icicle Creek and the Methow and Entiat Rivers (in the Columbia River Basin) originated from transplants from outside the basins. As these populations are considered non-native and not part of any ESU, they are not considered to be an ESA issue.

Status of Nonanadromous Oncorhynchus nerka

Within the range of west coast sockeye salmon, "kokanee-sized" O. nerka occur sympatrically with several sockeye salmon populations in their respective spawning and/or juvenile rearing environments. In the Okanogan River, Lake Wenatchee, Sammamish River tributaries, Cedar River, Lake Pleasant, and Ozette Lake populations, sockeye salmon and kokanee-sized O. nerka are often observed together on the spawning grounds at the same time and in the same place. In addition, kokanee-sized O. nerka are often observed at the time of sockeye salmon spawning in the channel which drains the upper spawning beaches at Baker Lake, but these fish cannot gain access to these artificial sockeye salmon spawning beaches. Kokanee-sized fish have not been reported on the sockeye salmon spawning grounds of Quinault Lake to any appreciable degree. On the other hand, native kokanee that have relatively easy access to the ocean spawn in the absence of anadromous sockeye salmon in several tributary streams of Ozette Lake and in Issaquah Creek, a tributary of Lake Sammamish. Juvenile kokanee from these two populations rear sympatrically with sockeye salmon in Ozette Lake and Lake Sammamish, respectively.

Several native and numerous introduced populations of kokanee exist within the geographic range of west coast sockeye salmon. Several of these native kokanee populations may be genetically distinct and reproductively isolated from one another and from other O. nerka populations. The BRT acknowledged that it has long been known that kokanee can produce anadromous fish; however, the number of outmigrants that successfully return as adults is typically quite low. In three instances where populations of native kokanee occur in coastal lakes in the Pacific Northwest and where access from the ocean is relatively easy (Ozette Lake and Lake Sammamish in Washington, and Cowichan Lake on Vancouver Island), the sockeye salmon morphology is absent on the kokanee spawning grounds. If kokanee in these populations were producing anadromous outmigrants in any appreciable numbers that were surviving to adulthood, the sockeye salmon morphology should be visible on the kokanee spawning grounds.

Occasionally, a proportion of the juveniles in an anadromous sockeye salmon population will remain in the rearing-lake environment throughout life and will be observed on the spawning grounds together with their anadromous siblings. For the purposes of this review, we have defined these fish as "resident sockeye salmon," to indicate that they are the progeny of anadromous sockeye salmon parents, yet spend their adult life in freshwater and are observed together with their anadromous siblings on the spawning grounds.

Foote et al. (1989), Wood and Foote (1990), Foote et al. (1994), Taylor and Foote (1991), Taylor et al. (1996), Wood and Foote (1996), and Winans et al. (1996) provide evidence that sympatric populations of sockeye salmon and kokanee can be both genetically distinct and reproductively isolated. In the following compilation of sockeye salmon ESUs, the status of kokanee-sized O. nerka that spawn together with sockeye salmon will be addressed on a case-by-case basis for each ESU.

In considering the ESU status of resident forms of O. nerka, a key issue is evaluating the strength and duration of reproductive isolation between resident and anadromous forms. Many kokanee populations appear to have been strongly isolated from sympatric sockeye salmon populations for long periods of time. Since the two forms experience very different selective regimes over their life cycle, reproductive isolation provides an opportunity for adaptive divergence in sympatry. Kokanee populations that fall into this category will generally not be considered part of sockeye salmon ESUs. On the other hand, resident fish appear to be much more closely integrated into some sockeye salmon populations. For example, in some situations anadromous fish may give rise to progeny that mature in freshwater (as is the case with residual sockeye salmon), and some resident fish may have anadromous offspring. In these cases, where there is presumably some regular or at least episodic genetic exchange between resident and anadromous forms, they should be considered part of the same ESU.

Sockeye Salmon ESUs

1) Okanogan River

This ESU consists of sockeye salmon that return to Lake Osoyoos through the Okanogan River via the Columbia River and spawn primarily in the Canadian section of the Okanagan River above Lake Osoyoos. Genetic, environmental, and life history information were the primary factors in distinguishing this ESU. Factors important to the BRT in identifying this ESU were these: 1) the very different environmental and habitat conditions encountered by sockeye salmon in Lakes Osoyoos and Wenatchee during juvenile rearing; 2) the near absence of 3-year-old sockeye salmon returns to Lake Wenatchee coupled with the tendency for a large percentage of 3-year-olds to return to the Okanogan population; 3) the apparent 1 month separation in juvenile outmigration-timing between Okanogan and Wenatchee-origin fish; and 4) the adaptation of Okanogan River sockeye salmon to much higher temperatures during adult migration in the Okanogan River. Protein electrophoretic data (with the exception of Utter's (1995) preliminary report) also indicate that this population is genetically distinct from other sockeye salmon currently in the Columbia River drainage. Utter's (1995) data, which show sockeye salmon collected in 1994 (at Wells Dam on the Columbia River and presumably bound for the Okanogan River) and Wenatchee sockeye salmon (collected in multiple years) to be genetically indistinguishable, is at odds with all other genetic studies that have shown high levels of genetic differentiation between Okanogan River and Lake Wenatchee sockeye salmon (Winans et al. 1996, Wood et al. 1996, Thorgaard et al. 1995)

The overall effect of the Grand Coulee Fish Maintenance Project (GCFMP) on the current composition of sockeye salmon in this ESU is difficult to determine. A majority of the sockeye salmon returning to the mid- to upper Columbia River prior to the operation of Grand Coulee Dam most likely spawned in the Arrow Lakes region of British Columbia. The redistribution and long-term propagation of mixed Arrow Lakes, Okanogan, and Wenatchee stocks of adult sockeye salmon originally captured at Rock Island Dam, as well as introductions of Quinault Lake sockeye salmon stocks, may have altered the genetic make-up of indigenous sockeye salmon in the Okanogan River, particularly considering the relatively low estimated returns of native sockeye salmon immediately prior to the beginning of the GCFMP. However, electrophoretic analysis of current Okanogan River sockeye salmon reveals little affinity with either Lake Wenatchee or Quinault Lake sockeye salmon or with kokanee currently residing in Lower Arrow Lake (see statement above concerning Utter (1995)).

Kokanee are reported to occur in Lake Osoyoos, and one known plant of 195,000 kokanee from an unknown source occurred in this lake in the years 1919-1920. Kokanee-sized fish (L. LaVoy52) or residuals (Chapman et al. 1995) with a reportedly olive drab or "typically dark" coloration, respectively, have been observed spawning with sockeye salmon in the Okanogan River. Genetic samples of kokanee-sized fish from Lake Osoyoos have not been obtained. However, kokanee from Okanagan Lake, above Vaseux Dam and Vaseux Lake on the Okanagan River, are genetically quite distinct from Okanogan River sockeye salmon (Wood et al. 1994, Thorgaard et al. 1995, Utter 1995, Robison 1995, Winans et al. 1996). Robison (1995) suggested that Okanagan Lake kokanee may have been transplanted into this system, as they appear genetically similar to Shuswap Lake kokanee.

The BRT concluded that if "kokanee-sized" O. nerka observed spawning with sockeye salmon on the Okanogan River are identified as resident sockeye salmon they are to be considered part of this sockeye salmon ESU. Based on the large genetic distance between Okanagan Lake kokanee and Okanogan River sockeye salmon, the BRT decided that Okanagan Lake kokanee are not part of the Okanogan sockeye salmon ESU. The BRT concluded that spawning aggregations of sockeye salmon that are occasionally observed downstream from Lake Osoyoos and below Enloe Dam on the Similkameen River are most likely wanderers from the Okanogan River population and are therefore to be considered part of this ESU.

2) Lake Wenatchee

This ESU consists of sockeye salmon that return to Lake Wenatchee through the Wenatchee River via the Columbia River and spawn primarily in tributaries above Lake Wenatchee (the White River, Napeequa River, and Little Wenatchee River). Genetic, environmental, and life history information were the primary factors in distinguishing this ESU. Allozyme data indicate that, of the populations examined, the Lake Wenatchee sockeye salmon population is genetically very distinctive (but see discussion above concerning Utter (1995)). Several ecological and biological factors were important in distinguishing the Okanogan River and Lake Wenatchee sockeye salmon ESUs. These include: 1) the very different environmental conditions encountered by sockeye salmon in Lakes Wenatchee and Osoyoos, 2) the near absence of 3-year-old sockeye salmon returns to Lake Wenatchee coupled with the tendency for a large percentage of 3-year-olds to return to the Okanogan population, and 3) the apparent 1 month separation in juvenile outmigration-timing between Okanogan- and Wenatchee-origin fish.

The overall effect of the GCFMP on the current make-up of sockeye salmon in this ESU is difficult to determine. The redistribution and long-term propagation of mixed Arrow Lakes, Okanogan, and Wenatchee stocks of sockeye salmon originally captured at Rock Island Dam, as well as introductions of Quinault Lake sockeye salmon stocks, may have altered the genetic makeup of indigenous sockeye salmon in the Lake Wenatchee system, particularly considering the low estimated returns of native sockeye salmon to Lake Wenatchee immediately prior to the beginning of the GCFMP. It is possible that a significant portion of the current gene pool of Lake Wenatchee sockeye salmon is derived from introduced Arrow Lakes sockeye salmon. However, electrophoretic analysis of current Lake Wenatchee sockeye salmon reveals little affinity with either Okanogan River (but see discussion of Utter (1995) in Okanogan River ESU section) or Quinault Lake sockeye salmon or with kokanee from Lower Arrow Lake.

Spawning aggregations of sockeye salmon that appear in the Entiat and Methow Rivers and in Icicle Creek (a tributary of the Wenatchee River) were presumed by the BRT to be nonnative and the result of transplants carried on during the GCFMP. Both the Methow and Entiat Rivers had no history of sockeye salmon prior to stocking (WDF et al. 1938, Mullan 1986). Leavenworth National Fish Hatchery is located on Icicle Creek, and between 1942 and 1969 over 1.5 million sockeye salmon juveniles (of mixed Columbia, Entiat, Methow Rivers heritage) were liberated from this facility into Icicle Creek (Mullan 1986, Chapman et al. 1995).

Kokanee-sized fish with a reportedly olive-drab coloration have been observed spawning with sockeye salmon in the White, Napeequa, and Little Wenatchee Rivers (L. LaVoy 53). Over 23 million Lake Whatcom kokanee were released in Lake Wenatchee between 1934 and 1983; however, the current genetic make-up of the Lake Wenatchee sockeye salmon population reveals little or no affinity with Lake Whatcom kokanee. Genetic samples of kokanee-sized fish from Lake Wenatchee have not been obtained.

The BRT concluded that if "kokanee-sized" O. nerka observed spawning with sockeye salmon on the White and Little Wenatchee Rivers are identified as resident sockeye salmon they are to be considered part of the Lake Wenatchee sockeye salmon ESU.


3) Quinault Lake

This ESU consists of sockeye salmon that return to Quinault Lake and spawn in the mainstem of the upper Quinault River, in tributaries of the upper Quinault River, and in a few small tributaries of Quinault Lake itself. The BRT felt that Quinault Lake sockeye salmon deserved separate ESU status based on its unique life history characteristics and degree of genetic differentiation from other sockeye salmon populations.

Key factors in identifying this ESU were: 1) the distinctive early river-entry timing, 2) the protracted adult run-timing, 3) the long 3- to 10-month lake-residence period prior to spawning, 4) the unusually long spawn timing, and 5) the genetic differences from other coastal Washington sockeye salmon. In addition, the relative absence of red skin pigmentation and the presence of an olive-green spawning coloration by the majority of the Quinault stock appear to be unique among major sockeye salmon stocks in Washington (Storm et al. 1990, D. Boyer, Jr.54), although at least two sockeye salmon stocks in British Columbia appear more green than red at spawning (C. C. Wood55). The rather large genetic distance between U.S. and Vancouver Island sockeye salmon, together with the apparently unique life-history characters of Quinault Lake sockeye salmon (very early, yet protracted run-timing, and lengthy lake-residency as adults), persuaded the BRT to exclude Vancouver Island stocks from this ESU.

Kokanee-sized O. nerka have not been identified within the Quinault River Basin; however, stocking history reveals over 300,000 kokanee transplanted into Quinault Lake between 1917 and 1922 from an unknown source and 260,000 kokanee eggs transferred from Lake Whatcom to the "Quinault, Washington Station" in 1925.

4) Ozette Lake

This ESU consists of sockeye salmon that return to Ozette Lake through the Ozette River and currently spawn primarily in lakeshore upwelling areas in Ozette Lake (particularly at Allen's Bay and Olsen's Beach). Minor spawning may occur below Ozette Lake in the Ozette River or in Coal Creek, tributary of the Ozette River. Sockeye salmon do not presently spawn in tributary streams to Ozette Lake, although they may have spawned there historically. Genetic, environmental, and life history information were the primary factors in distinguishing this ESU. The BRT felt that Ozette Lake sockeye salmon were a separate ESU based on the degree of genetic differentiation from other sockeye salmon populations and on life history characteristics.

Ozette Lake sockeye salmon are genetically distinct from all other sockeye salmon stocks in the Northwest. Sockeye salmon stocks from west coast Vancouver Island were excluded from this ESU, in part because of the large genetic distance between Vancouver Island and Ozette Lake sockeye salmon. On the other hand, Ozette Lake kokanee proved to be the most genetically distinct O. nerka stock examined in the contiguous United States (based on 29 allozyme loci). However, Ozette Lake kokanee were closely allied to several sockeye salmon stocks on Vancouver Island (based on analysis of a nine allozyme loci data set).

Kokanee are very numerous in Ozette Lake and spawn in inlet tributaries, whereas sockeye salmon spawn on lakeshore upwelling beaches. Sockeye salmon have not been observed on the inlet spawning grounds of kokanee in Ozette Lake, although there are no physical barriers to prevent their entry into these tributaries. On the other hand, kokanee-sized O. nerka are observed together with sockeye salmon on the sockeye salmon spawning beaches at Allen's Bay and Olsen's Beach. One recorded plant of over 100,000 kokanee from an unknown source stock occurred in 1940, and anecdotal reports of another kokanee plant in 1958 were found.

Based on the very large genetic distance between Ozette Lake kokanee that spawn in tributaries and Ozette Lake sockeye salmon that spawn on shoreline beaches, the BRT excluded Ozette Lake kokanee from this sockeye salmon ESU. In addition, the BRT concluded that if "kokanee-sized" O. nerka observed spawning with sockeye salmon on sockeye salmon spawning beaches in Ozette Lake are identified as resident sockeye salmon, then they are to be considered as part of the Ozette Lake sockeye salmon ESU.

5) Baker River

This ESU consists of sockeye salmon that return to the barrier dam and fish trap on the lower Baker River after migrating through the Skagit River. Adults are trucked to one of three artificial spawning beaches above either one or two dams on the Baker River, and are held in these enclosures until spawning.

The BRT felt that Baker River sockeye salmon are a separate ESU based on genetic, life history, and environmental characters. Baker River sockeye salmon are genetically distinct from sockeye salmon populations that spawn in the lower Fraser River (allozyme data based on 9 loci) and are genetically distinct from all other native populations of Washington sockeye salmon (allozyme data based on 29 loci). Prior to inundation behind Upper Baker Dam, Baker Lake was a typical cold, oligotrophic, well-oxygenated, glacially turbid sockeye salmon nursery lake, in contrast to other sockeye salmon systems under review, with the exception of Lake Wenatchee.

The Birdsview Hatchery population on Grandy Creek in the Skagit River Basin was established from Baker Lake sockeye salmon together with a probable mixture of Quinault Lake stock and an unknown Fraser River stock. This stock was the ultimate source for the apparently successful transplants of sockeye salmon to the Lake Washington/Lake Sammamish system in the mid-1930s to early 1940s (Royal and Seymour 1940, Kolb 1971).

Numerous reports indicate that residual or resident sockeye salmon began appearing in Baker Lake and Lake Shannon Reservoir following the installation of Lower Baker Dam in 1925 (Ward 1929, 1930, 1932; Ricker 1940; Kemmerich 1945). A spring-time recreational kokanee fishery exists in Baker Lake, although substantial aggregations of spawning kokanee have yet to be identified. We found no historical records of kokanee stocking in Baker Lake. However, approximately 40-100 kokanee-sized O. nerka spawn each year in the outlet channel that drains the two upper sockeye salmon spawning beaches at Baker Lake.

6) Lake Pleasant

A majority of the BRT concluded that Lake Pleasant sockeye salmon constituted a separate ESU, while a minority thought that insufficient information existed to make a decision as to its ESU status. Allozyme data for Lake Pleasant sockeye salmon indicate genetic distinctiveness from other sockeye salmon populations. Sockeye salmon in this population enter the Quillayute River from May through September and hold in the Sol Duc River before entering Lake Pleasant, usually in early November, when sufficient water depth is available in Lake Creek. Spawning occurs on beaches from late November to early January. Kemmerich (1945) indicated that native sockeye salmon occurred in Lake Pleasant prior to 1932 and that they were of an "individual size comparable with the size of the fish of the Lake Quinault and Columbia River populations"; however, sockeye salmon currently in Lake Pleasant are said to be small and no bigger than about 2 to 3 pounds (0.9 to 1.4 kg) (J. Haymes56). Adult male and female Lake Pleasant sockeye salmon have an average fork length of 460 mm or less for all ages combined, which is the smallest body size of any anadromous O. nerka population in the Pacific Northwest. In some years, a majority of Lake Pleasant sockeye salmon spend 2 years in freshwater prior to migrating to sea (Appendix Table C-1).

Over 0.5 million sockeye salmon fry from Baker Lake and the Birdsview Hatchery in the Skagit River Basin were released in Lake Pleasant in the 1930s; however, electrophoretic analysis of current Lake Pleasant sockeye salmon reveals little genetic affinity with Baker Lake sockeye salmon. It is assumed that poisoning of Lake Pleasant during "lake rehabilitation" activities in the 1950s and 1960s (see previous information section) may have impacted one or two broodyears of sockeye salmon in Lake Pleasant. Sockeye salmon escapement to Lake Pleasant was between about 760 and 1,500 fish in the early 1960s; indicating that "lake rehabilitation" failed to eliminate sockeye salmon from this system. Although kokanee-sized O. nerka spawn together with sockeye salmon on the beaches in Lake Pleasant, only anecdotal reference to kokanee being stocked in Lake Pleasant during the 1930s were found.

The BRT concluded that if "kokanee-sized" O. nerka observed spawning with sockeye salmon on sockeye salmon spawning beaches in Lake Pleasant are identified as resident sockeye salmon, then they are to be considered as part of the Lake Pleasant sockeye salmon ESU.

Sockeye Salmon of Provisional ESU Status

Big Bear Creek

The BRT was divided on the ESU status of sockeye salmon that currently spawn in Big Bear Creek and its two tributaries, Cottage Lake and Evans Creeks. Members did agree that the available evidence does not clearly resolve this issue. In spite of various uncertainties, about half of the BRT felt that the current sockeye salmon population in Big Bear and Cottage Lake Creeks is a separate ESU that represents either an indigenous Lake Washington/Lake Sammamish sockeye salmon population or a native kokanee population that has naturally re-established anadromy. About half the members felt that the available information was insufficient to determine the ESU status of sockeye salmon in Big Bear Creek. This issue is particularly difficult due to the equivocal nature of historical accounts concerning the presence and distribution of sockeye salmon within the Lake Washington/Lake Sammamish Basin.

Genetically, Big Bear and Cottage Lake Creek sockeye salmon are quite distinct from other stocks of sockeye salmon in the Lake Washington/Lake Sammamish Basin; they are genetically more similar to Okanogan River sockeye salmon than they are to any other sockeye salmon population examined. It was acknowledged that the genetic distinctiveness of the current Big Bear Creek/Cottage Lake Creek sockeye salmon as revealed through analysis of allozyme data could have resulted from genetic change (founder effect and/or subsequent genetic drift) following the recorded transplant of Baker Lake sockeye salmon in 1937 and observation of 2 adults in October 1940, or it could be indicative of a native population of O. nerka indigenous to the Lake Washington/Lake Sammamish Basin.

A native kokanee population once spawned in Big Bear Creek and its tributaries, although it is uncertain whether a remnant of this native stock still exists in this drainage. Big Bear Creek was once the largest producer of kokanee for artificial propagation in Washington, although relatively few kokanee currently spawn there. Currently a small number of kokanee-sized O. nerka spawn in Big Bear Creek together with sockeye salmon. The spawn timing of kokanee in Big Bear Creek is currently much later than the only remaining recognized native kokanee stock in the Lake Washington Basin (early entry Issaquah Creek kokanee). There were over 35 million Lake Whatcom kokanee fry released in Big Bear Creek between 1917 and 1969, and what effect this stocking program had on the native kokanee is open to speculation. In addition, potential genetic interactions of these introduced kokanee with sockeye salmon are unknown.

Sockeye Salmon of Uncertain ESU Status

Riverine-spawning sockeye salmon

Spawning ground survey data of the Washington Department of Fish and Wildlife and numerous anecdotal references dating back to the turn of the century indicate that riverine-spawning aggregations of sockeye salmon exist in certain rivers within Washington that lack lake-rearing habitat. Consistent riverine- spawning aggregations of sockeye salmon have been documented over a period of decades in the North and South Fork Nooksack, Skagit, Sauk, North Fork Stillaguamish, Samish (D. Hendrick57), and Green Rivers. Riverine-spawning sockeye salmon have also been reported in the Nisqually, Skokomish, Dungeness, Calawah, Hoh, Queets, and Clearwater Rivers, and are occasionally seen in small numbers in a number of other rivers and streams in Washington.

Protein electrophoretic data for riverine spawners from the Nooksack, upper Skagit, and Sauk Rivers indicate that these aggregations are genetically similar to one another and genetically distinct from other sockeye salmon in Washington. Genetic data of equal resolution (29 allozyme loci) for comparison with river/sea-type and lake-type sockeye salmon populations in the lower Fraser River are not available.

The BRT considered these hypotheses that might explain river-spawning aggregations of sockeye salmon in Washington: 1) they represent multiple U.S. populations, 2) they represent one U.S. population, 3) they represent strays from U.S. lake-type sockeye salmon, 4) they represent strays from British Columbia lake-type sockeye salmon, and 5) they represent strays from river-type populations in British Columbia. Genetic data for river spawning sockeye salmon in the Nooksack, Skagit, and Sauk Rivers do not support hypothesis 3. The disjunct timing and geographic distance between individual aggregations of riverine-spawning sockeye salmon suggest that more than one process may be responsible for the occurrence of these aggregations.

The small size of the spawning aggregations of sockeye salmon periodically reported in rivers without lake-rearing habitat in Washington raises the question of historic population size and persistence of Pacific salmon over evolutionarily significant time scales. Because many populations of Pacific salmon show large temporal fluctuations in abundance, Waples (1991a, p. 19) argued in the NMFS "Definition of Species" paper that

there must be some size below which a spawning population is unlikely to persist in isolation for a long period of time. The fact that small spawning aggregations are regularly observed may reflect a dynamic process of extinction, straying, and recolonization. Such small populations are unlikely to be ESU's, although a collection of them might be.

However, Waples (1991a, p. 19) went on to say that

In making this evaluation, the possibility should be considered that small populations observed at present are still in existence precisely because they evolved mechanisms for persisting at low abundance.

The BRT acknowledged the evolutionary importance of existing river/sea-type sockeye salmon in British Columbia and Alaska but felt that the evidence was insufficient to determine whether sockeye salmon seen in rivers without lake-rearing habitat in Washington were distinct populations. The ESU status of riverine-spawning sockeye salmon in Washington remains an open question.

Deschutes River, Oregon

The BRT concluded that sockeye salmon that historically migrated up the Deschutes River via the Columbia River to spawn in Suttle Lake were a separate ESU, but it is uncertain whether remnants of this ESU exist. Fish passage into and out of Suttle Lake was blocked sometime around 1930. Currently, sockeye salmon adults that are consistently seen each year in the Deschutes River below the regulatory dam downstream from Pelton Dam may be derived from 1) a self-sustaining population of sockeye salmon that spawn below Pelton Dam on the Deschutes River, 2) strays from elsewhere in the Columbia River, or 3) outmigration of smolts from populations of "kokanee-sized" O. nerka that exist above the Pelton/Round Butte Dam complex. Two kokanee populations are present above the dams: one population resides in Suttle Lake and spawns in the lake inlet stream (Link Creek), and a second population resides in Lake Billy Chinook, behind Round Butte Dam, and spawns in the upper Metolius River. Both kokanee populations have a distinctive blue-black body coloration that distinguishes them from hatchery kokanee that are released in Lake Simtustus and other Deschutes River Basin hatcheries.

Allozyme data for Deschutes River sockeye salmon do not exist; however, mtDNA data (Brannon 1996), suggest the possibility that Lake Billy Chinook kokanee and Deschutes River sockeye salmon are related. Protein electrophoretic data indicate that kokanee in Suttle Lake and in Lake Billy Chinook cluster together genetically (unpublished data, NMFS, Northwest Fisheries Science Center, 2725 Montlake Blvd East, Seattle, WA 98112). Over 1.2 million sockeye salmon were planted in the Metolius River and its tributaries before 1962, and a significant portion of the adult sockeye salmon returns recorded at the Pelton Dam fish trap, starting in 1956, may have been descended from these plantings.

The majority of the BRT concluded that a remnant component of this historical population cannot be identified with any certainty. A minority of the BRT felt that the extensive transplant history of non-native sockeye salmon into this basin explains the continued occurrence of anadromous O. nerka in the Deschutes River Basin and, as the descendants of transplants, these sockeye salmon are not an ESA issue. It should be noted at this point that sockeye salmon continue to return to the base of reservoir dams on the Middle and South Santiam Rivers in Oregon, long after sockeye salmon fry were released into these reservoirs. Sockeye salmon that return to the Santiam River are the putative progeny of residualized sockeye salmon. The majority of the BRT agreed that the possibility exists that recent sockeye salmon in the Deschutes River may result from some remnant outmigrants of residualized sockeye salmon or kokanee. The ESU status of sockeye salmon returning to the Deschutes River remains an open question.



On to Next Section

Table of Contents