U.S. Dept Commerce/NOAA/NMFS/NWFSC/Publications
Tech Memo-22: Status Review for Mid-Columbia River Summer Chinook Salmon

Discussion and Conclusions

A key question in the biological evaluation of MCR late-run chinook salmon is whether the summer- and fall-run fish in the MCR are reproductively isolated. Whether they are considered separately or jointly is pivotal in determing the ESU and its numeric stability. We have evaluated a variety of biological characters and have summarized them in Table 7.

Reproductive Isolation
The petitioners argue that MCR summer chinook salmon are reproductively isolated with respect to three other groups of chinook salmon within the Columbia River Basin: 1) Snake River summer chinook salmon, 2) mid-Columbia River spring chinook salmon, and 3) mid-Columbia River fall chinook salmon.

Genetic Characters

Genetic data based on protein electrophoresis are consistent with the existing view of the genetic affinities in MCR summer chinook salmon. This group of fish is clearly reproductively isolated from Snake River summer chinook salmon (previously determined by NMFS to be part of a different ESU), and from the spring chinook salmon in the MCR as well. Although there is a lack of genetic information specifically for fall-run fish from the rivers identified in the petition (Methow, Okanogan, and Wenatchee), it has been known for some time that there is a close genetic similarity between mid-Columbia River summer chinook salmon and fall chinook salmon from the Hanford Reach area.

Table 7. Summary of biological characteristics of stream- and ocean-type chinook salmon in the mid-Columbia region

Stream type Ocean-type
Characteristic Spring-run Summer-run Fall-run
Adult run timing
at Bonneville Dam
Mar.-May June-July Aug.-Oct.
General spawning
upper tributaries mid tributaries
lower tributaries
and mainstem
lower tributaries
and mainstem
Spawning time
August to
late September
to mid-November
mid-October to
early December
Seaward juvenile
yearling subyearling subyearling
Ocean distribution
extended coastal coastal
a b b
Meristic similarity*
a b b
Genetic similarity*
a b b

*Run-types with the same letter are not significantly different. Morphometric and meristic data from Schreck et al. (1986); genetic information from Schreck (1986), Utter et al. (1989), Matthews and Waples (1991), and Marshall (1993, 1994a and b).

New data from WDF (Marshall 1993, 1994a and b) included Hanford Reach natural population and Priest Rapids Hatchery. Allele frequencies in the Priest Rapids sample did not differ significantly from those in summer-run adults that returned to Wells Hatchery, the Wells Dam trap, or the Similkameen River (a tributary of the Okanogan River).

Phenotypic Characters

Data from one multivariate study of meristic and morphometric characters of chinook salmon in the Columbia River Basin indicated that summer and fall chinook salmon from the MCR are similar to one another but different from other groups of chinook salmon in the MCR and the Snake River.

Life History Characters

Among MCR and Snake River chinook salmon populations, major juvenile and adult life history characters match the patterns of genetic variation. In both rivers, spring chinook salmon are stream-types and fall chinook salmon are ocean-types. In the Snake River, summer chinook salmon are stream-types like spring chinook salmon, while in the MCR, they are ocean-types like fall chinook salmon. Genetically, Snake River summer chinook salmon are closely related to spring chinook salmon in both rivers while MCR summer chinook salmon are closely related to fall chinook salmon in both rivers.

Spawn and Run Timing

Although groups of fish returning over dams in the MCR were once thought to spawn in different habitats at different times, this notion has not been supported by the data collected for this Biological Status Review. Both types of fish were found to spawn in tributaries as well as mainstem areas, with substantial overlap in the spawn timing and duration.

Evolutionary Significance

The petitioners argue that summer-run chinook salmon in the Methow and Wenatchee Rivers are the last remnants of the large population of summer-run fish that once spawned in areas above Grand Coulee Dam. The summer run of fish in the Columbia River was renowned for its large size (hence the term "June hogs") and abundance. Additionally, the petitioners argue that the Okanogan River, which did not receive transplants from the GCFMP, contains the "only documented native stock of summer chinook in the mid-Columbia basin" (NEDC et al. 1993. p. 6).

Phenotypic and Life History Traits

Given their extensive upriver migration and presumably distinctive habitat characteristics, it seems likely that chinook salmon spawning in the now inaccessible Lake Windermere region of the upper Columbia River historically comprised an ESU. However, for several reasons, it is unlikely that any appreciable remnants of this gene pool remain in the petitioned streams.

First, reports of spawn timing by early settlers suggested stream-type, not ocean-type fish. Comparison with upriver populations from the Fraser River, which has headwaters near those of the upper Columbia River, also suggests that the upper Columbia River chinook salmon may have been stream-type fish. Thus, it is uncertain what relationship the ocean-type fish currently residing in the mid-Columbia River have to the original upriver populations.

Second, it is likely that "June hogs" taken in the lower Columbia River were comprised of large fish from many different populations. There are conflicting accounts from the 1800s regarding the size of chinook salmon in the upper Columbia River. For example, historical accounts from the early 1800s suggest that chinook salmon taken by aboriginals at Kettle Falls in the upper Columbia River averaged only 7 kg in weight (Mullan et al. 1992b), whereas Bryant and Parkhurst (1950) reported that early settlers witnessed fish averaging 18 kg. In any case, by the time the GCFMP was initiated, wild adult ocean-type fish taken at Rock Island Dam averaged only about 8 kg (Fulton and Pearson 1981). Thus, even if a run of large fish did occur historically in the upper Columbia River, it no longer existed by 1939. "June hogs" from all populations may have been largely eliminated by the turn of the century as a result of heavy fishing pressure.

Third, adults collected for the GCFMP were a mixture of summer- and fall-run fish from all areas above Rock Island Dam. It seems likely, therefore, that genetic admixture, in addition to translocation, has occurred in existing summer-run chinook salmon from the mid-Columbia River. Nevertheless, abundance of chinook salmon increased at least temporarily in rivers that received transplants from the GCFMP, so presumably some genes from upriver stocks were incorporated into mid-Columbia River populations. Whether any genetic traits that distinctly characterized upper Columbia River chinook salmon still exist in the MCR is not known with certainty, but we found no empirical evidence to support this hypothesis.

Finally, although the Okanogan River is the only major stream in the petitioned area that did not receive transplants from the GCFMP, it is unlikely that summer-run fish in this river represent an essentially pure native stock. All late-run (summer and fall) chinook salmon adults reaching Rock Island Dam were taken for the GCFMP for a period of 5 years. According to age data for late-run chinook salmon from the mid-Columbia River, less than 1% of returning adults are older than 5 years. Therefore, the current population in the Okanogan River, which is upstream from Rock Island Dam, must be derived largely, if not entirely, from recolonization with manipulated populations of the GCFMP.

Genetic Data

Several published genetic studies and new data from WDF presented here have shown a strong genetic similarity between summer-run fish from the Wenatchee and Okanogan Rivers and Wells Hatchery; there is no indication that the Okanogan stock is representative of a remnant upstream chinook salmon stock.

Conclusion: Species Determination

Genetic and life-history information, as well as data from various tagging experiments, all fail to demonstrate reproductive isolation between summer- and fall-run chinook salmon in the mid- Columbia River (Table 7). In addition, coded wire tag data suggest that the two forms have a similar ocean distribution. Therefore, we concluded that all late-run, ocean-type chinook salmon from the mid-Columbia River are part of the same ESU as defined by the ESA.

We evaluated the relationship of this ESU to three other groups of Columbia River Basin chinook salmon and found substantial genetic and life-history differences: 1) stream- type, spring chinook salmon from the mid-Columbia River were clearly part of a separate ESU; 2) Snake River summer chinook salmon were much more closely related to Snake River spring chinook salmon than to summer chinook salmon from the mid- Columbia River; and 3) the closest relatives of mid-Columbia River late-run chinook salmon are other groups of Columbia River fall chinook salmon.

Two major types of fall chinook salmon were found in the basin: "tules" and "upriver brights." Substantial genetic differences have been demonstrated between these two types of chinook salmon in the Columbia River. The "upriver bright" group includes late-run chinook salmon from the mid-Columbia River as well as Snake River fall chinook salmon.

Previously, ecological, genetic, and ocean distribution data were used to demonstrate that Snake River fall chinook salmon are distinct from mid-Columbia River fall chinook salmon and represent a separate ESU. Therefore, we concluded that late-run (summer and fall), ocean-type chinook salmon from the mid- Columbia River represent an ESU separate from all other chinook salmon in the Columbia River Basin. Some of the distinctive habitat features of this ESU are discussed above.

Threshold Determination

Dam and redd count information indicated that although depression in some individual runs of late-run chinook salmon in the mid-Columbia River is cause for concern, the mid-Columbia River late-run, ocean-type chinook salmon ESU as a whole is relatively healthy, with little risk of extinction in the foreseeable future. Even if we considered populations from only those rivers identified in the petition (Okanogan including Similkameen, Methow, and Wenatchee), we would find little risk of extinction for those populations.

While redd counts in two of these rivers (Okanogan and Methow) have exhibited substantial declines since the late 1960s, they have been relatively stable since 1980, and counts in both the Wenatchee and Similkameen Rivers have exhibited long-term (1956-93) increasing trends. Based on Rock Island Dam adult counts, this group of populations is certainly more abundant than it was in the 1930s and 1940s.


While we do not believe that this ESU is at significant risk of extinction or endangerment, the low numbers of ocean-type chinook salmon above Rocky Reach Dam (including the Methow and Okanogan Rivers), despite the virtual elimination of inriver harvest on these stocks, are of concern. Declines since the mid- 1970s in this region may simply be a return of the populations to normal carrying capacity following substantial hatchery supplementation, or may indicate local problems with habitat, dam passage, or hatchery practices. Special management consideration of these particular stocks may be warranted.

Some aspects of artificial propagation also pose risks for populations within the ESU. For example, there have been large releases of fall chinook salmon in the mainstem Columbia River and in the Yakima River in recent years. We do not believe that the potential genetic and ecological consequences of these releases have been adequately addressed.

Table of Contents