Most of the ESUs described below include multiple spawning populations of chinook salmon, and most also extend over a considerable geographic area (Figs. 23 and 24). This result is consistent with NMFS' species definition paper, which states that, in general, "ESUs should correspond to more comprehensive units unless there is clear evidence that evolutionarily important differences exist between smaller population segments" (Waples 1991b, p. 20). However, considerable diversity in genetic or life-history traits or habitat features exists within most ESUs, and maintaining this diversity is critical to their overall health. The descriptions below briefly summarize some of the notable types of diversity within each ESU, and this diversity is considered in the next section in evaluating risk to the ESUs as a whole.
This run was determined to be a distinct population segment by NMFS in 1987, prior to development of the NMFS species policy. The BRT concluded that this run meets the criteria to be considered an ESU. It includes chinook salmon entering the Sacramento River from November to June and spawning from late-April to mid-August, with a peak from May to June. No other chinook salmon populations have a similar life-history pattern. In general, winter-run chinook salmon exhibit an ocean-type life-history strategy, with smolts emigrating to the ocean after five to nine months of freshwater residence (Johnson et al. 1992b) and remaining near the coasts of California and Oregon. Winter-run chinook salmon also mature at a relatively young age (2-3 years old). DNA analysis indicates substantial genetic differences between winter-run and other chinook salmon in the Sacramento River.
Historically, winter-run populations existed in the Upper Sacramento, Pit, McCloud, and Calaveras Rivers. The spawning habitat for these stocks was primarily located in the Sierra Nevada Ecoregion (Omernik 1987). Construction of dams on these rivers in the 1940s led to the extirpation of populations in the San Joaquin River Basin and displaced the Sacramento River population to areas below Shasta Dam.
Extant populations in this ESU spawn in the Sacramento River and its tributaries. Historically, spring-run chinook salmon were the dominant run in the Sacramento and San Joaquin River Basins (Clark 1929), but native populations in the San Joaquin River have apparently all been extirpated (Campbell and Moyle 1990). This ESU includes chinook salmon entering the Sacramento River from March to July and spawning from late August through early October, with a peak in September. Spring-run fish in the Sacramento River exhibit an ocean-type life history, emigrating as fry, subyearlings, and yearlings. Coded-wire-tag (CWT) recoveries are primarily from ocean fisheries off the California and Oregon coast. There were minimal differences in the ocean distribution of fall- and spring-run fish from the Feather River Hatchery (as determined by CWT analysis); however, due to hybridization in the hatchery between these two runs, this similarity in ocean migration may not be representative of wild runs. The BRT noted substantial ecological differences in the historical spawning habitat for spring-run vs. fall- and late-fall-run fish. The spring chinook salmon run timing was suited to gaining access to the upper reaches of river systems (up to 1,500 m elevation) prior to the onset of prohibitively high water temperatures and low flows that inhibit access to these areas during the fall. Differences in adult size, fecundity, and smolt size are also observed between spring- and fall-run chinook salmon in the Sacramento River.
No allozyme data are available for naturally spawning Sacramento River spring-run chinook salmon. A sample from Feather River Hatchery spring-run fish, which may have undergone substantial hybridization with fall chinook salmon, shows modest (but statistically significant) differences from fall-run hatchery populations. DNA data show moderate genetic differences between the spring and fall/late-fall runs in the Sacramento River; however, these data are difficult to interpret in the context of this broad status review because comparable data are not available for other geographic regions.
There were lengthy discussions by the BRT concerning the disposition of spring runs in the Sacramento River, and a number of different scenarios were considered. The majority of the BRT felt that the spring-run chinook salmon in the Sacramento River represented a separate ESU. A minority felt that the spring-run fish are part of a larger ESU that also includes the fall and late-fall runs. Based largely on environmental factors, the BRT also considered the possibility that spring-run fish from the San Joaquin River were historically part of a separate ESU, but little life-history and genetic information was available to evaluate this hypothesis. The BRT felt that it was important to develop additional genetic information to elucidate the status of the remnant spring-run populations in Butte, Deer, and Mill Creeks and their relationship to spring-run fish from the mainstem Sacramento and Feather Rivers.
This ESU includes fall and late-fall chinook salmon spawning in the Sacramento and San Joaquin Rivers and their tributaries. These populations enter the Sacramento and San Joaquin Rivers from July through April and spawn from October through February. Both runs are ocean-type chinook salmon, emigrating predominantly as fry and subyearlings and remaining off the California coast during their ocean migration. All chinook salmon in the Sacramento/San Joaquin Basin are genetically and physically distinguishable from coastal forms (Clark 1929, Snyder 1931). Ecologically, the Central Valley also differs in many important ways from coastal areas.
There were a number of life-history differences noted between Sacramento and San Joaquin River Basin fall-run populations. In general, San Joaquin River populations tend to mature at an earlier age and spawn later in the year than Sacramento River populations. These differences could have been phenotypic responses to the generally warmer temperature and lower flow conditions found in the San Joaquin River Basin relative to the Sacramento River Basin. There was no apparent difference in the distribution of marine CWT recoveries from Sacramento and San Joaquin River hatchery populations, nor were there genetic differences between Sacramento and San Joaquin River fall-run populations (based on DNA and allozyme analysis) of a similar magnitude to that used in distinguishing other ESUs. This apparent lack of distinguishing life-history and genetic characteristics may be due, in part, to large-scale transfers of Sacramento River fall-run chinook salmon into the San Joaquin River Basin. There was some concern expressed by the BRT that the information available may not be representative of fish historically occupying the San Joaquin River Basin.
A majority of the BRT felt that fall and late-fall chinook salmon in the Sacramento River represented a single ESU. Contrasting minority viewpoints were that: 1) Spring-run fish are part of the same ESU that includes the fall and late-fall runs; 2) fall and late-fall runs constituted separate ESUs; and 3) fall-run fish in the San Joaquin River Basin constituted their own ESU.
All coastal spring and fall chinook salmon spawning from Cape Blanco (south of the Elk River) to the southern extent of the current range comprise this ESU. The Cape Blanco region is a major biogeographic boundary for numerous species. The Southern Oregon and California Coastal ESU extends to the southern limit of the Coastal Range Ecoregion. Populations from the Central Valley and Klamath River Basin upstream from the Trinity River confluence are in separate ESUs. Chinook salmon in this ESU exhibit an ocean-type life-history; ocean distribution (based on marine CWT recoveries) is predominantly off the California and Oregon coasts. Life-history information on smaller populations, especially in the southern portion of the ESU, is extremely limited. Additionally, there was anecdotal or incomplete information on the existence of several spring-run populations, including the Chetco, Winchuck, Smith, Mad, and Eel Rivers. Allozyme data indicate that this ESU is genetically distinguishable from the Oregon Coast, Upper Klamath and Trinity River, and Central Valley ESUs.
Ecologically, the majority of the river systems in this ESU are relatively small and heavily influenced by a maritime climate. Low summer flows and high temperatures in many rivers result in seasonal, physical, and thermal barrier bars that block movement by anadromous fish. The Rogue River is the largest river basin in this ESU and extends inland, into the Sierra Nevada and Cascades Ecoregions.
A minority of the BRT felt that coastal chinook salmon from south of the Klamath River should be considered a separate ESU. Allozyme data, which show some level of genetic divergence between coastal chinook salmon populations north and south of the Klamath River, support this argument, as do the establishment of ESU boundaries for steelhead south of the Klamath River and for coho salmon south of Punta Gorda. A nearly total lack of biological information for chinook salmon south of the Eel River makes this issue difficult to resolve.
The BRT also considered arguments for the creation of separate fall- and spring-run ESUs in this and other coastal regions, but the consensus of the BRT was that this was not warranted.
Included in this ESU are all Klamath River Basin populations from the Trinity River and the Klamath River upstream from the confluence of the Trinity River. These populations include both spring- and fall-run fish that enter the Upper Klamath River Basin from March through July and July through October and spawn from late August through September and September through early January, respectively. Body morphology (vertebral counts, lateral-line scale counts, and fin-ray counts) and reproductive traits (egg size and number) for populations from the Upper Klamath River differ from those of populations in the Sacramento River Basin. Genetic analysis indicated that populations from the Upper Klamath River Basin form a unique group that is quite distinctive compared to neighboring ESUs. The Upper Klamath River crosses the Coastal Range, Sierra Nevada, and Eastern Cascades Ecoregions, although dams prevent access to the upper river headwaters of the Klamath River in the Eastern Cascades Ecoregion.
Within the Upper Klamath River Basin, there are statistically significant, but fairly modest, genetic differences between the fall and spring runs. The majority of spring- and fall-run fish emigrate to the marine environment primarily as subyearlings, but have a significant proportion of yearling smolts. Recoveries of CWTs indicate that both runs have a coastal distribution off the California and Oregon coasts. There was no apparent difference in the marine distribution of CWT recoveries from fall-run (Iron Gate and Trinity River Hatcheries) and spring-run populations (Trinity River Hatchery). The BRT discussed at some length the proposition that spring- and fall-run populations should be in separate ESUs based on differences in run timing and habitat utilization and reproductive isolation. The majority of the BRT concluded that both run types should be considered part of the same ESU; a minority felt that separation into two ESUs was warranted; and some BRT members were undecided on this issue. The BRT was concerned that the only estimate of the genetic relationship between spring and fall runs in this ESU is from a comparison of hatchery stocks that may have undergone some introgression during hatchery spawning operations. The BRT acknowledged that the ESU determination should be revisited if substantial new information from natural spring-run populations becomes available.
This ESU contains coastal populations of spring- and fall-run chinook salmon from the Elk River north to the mouth of the Columbia River. These populations exhibit an ocean-type life history and mature at ages 3, 4, and 5. In contrast to the more southerly ocean distribution pattern shown by populations from the lower Columbia River and farther south, CWT recoveries from populations within this ESU are predominantly from British Columbia and Alaska coastal fisheries. There is a strong genetic separation between Oregon Coast ESU populations and neighboring ESU populations. This ESU falls within the Coastal Ecoregion and is characterized by a strong maritime influence, with moderate temperatures and high precipitation levels.
A minority of the BRT felt that, because of similarities in life-history traits and environmental features, populations from the Oregon and Washington coasts were part of a single ESU. A separate minority felt that, based primarily on genetic information, the Oregon Coast ESU should be divided into two units, with populations north of the Umpqua River being in separate ESUs.
Coastal populations spawning north of the Columbia River and west of the Elwha River are included in this ESU. These populations can be distinguished from those in Puget Sound by their older age at maturity and more northerly ocean distribution. Allozyme data also indicates geographical differences between populations from this area and those in Puget Sound, the Columbia River, and the Oregon coast ESUs. Populations within this ESU are ocean-type chinook salmon and generally mature at ages 3, 4, and 5. Ocean distribution for these fish is more northerly than that for the Puget Sound and Lower Columbia River ESUs. The boundaries of this ESU lie within the Coastal Ecoregion, which is strongly influenced by the marine environment: high precipitation, moderate temperatures, and easy migration access. As noted above, a minority of the BRT felt that this ESU should be combined with chinook salmon from the Oregon coast.
This ESU encompasses all runs of chinook salmon in the Puget Sound region from the North Fork Nooksack River to the Elwha River on the Olympic Peninsula. Chinook salmon in this area all exhibit an ocean-type life history. Although some spring-run chinook salmon populations in the Puget Sound ESU have a high proportion of yearling smolt emigrants, the proportion varies substantially from year to year and appears to be environmentally mediated rather than genetically determined. Puget Sound stocks all tend to mature at ages 3 and 4 and exhibit similar, coastally-oriented, ocean migration patterns. There are substantial ocean distribution differences between Puget Sound and Washington coast stocks, with CWTs from Washington Coast fish being recovered in much larger proportions from Alaskan waters. The marine distribution of Elwha River chinook salmon most closely resembled other Puget Sound stocks, rather than Washington coast stocks. The BRT concluded that, on the basis of substantial genetic separation, the Puget Sound ESU does not include Canadian populations of chinook salmon. Allozyme analysis of North Fork and South Fork Nooksack River spring-run chinook salmon identified them as outliers, but most closely allied with other Puget Sound samples. DNA analysis identified a number of markers that appear to be restricted to either the Puget Sound or Washington coastal stocks. Some allozyme markers suggested an affinity of the Elwha River population with the Washington coastal stocks, while others suggested an affinity with Puget Sound stocks.
The boundaries of the Puget Sound ESU correspond generally with the boundaries of the Puget Lowland Ecoregion. Despite being in the rainshadow of the Olympic Mountains, the river systems in this area maintain high flow rates due to the melting snowpack in the surrounding mountains. Temperatures tend to be moderated by the marine environment. The Elwha River, which is in the Coastal Ecoregion, is the only system in this ESU which lies outside the Puget Sound Ecoregion. Furthermore, the boundary between the Washington Coast and Puget Sound ESUs (which includes the Elwha River in the Puget Sound ESU) corresponds with ESU boundaries for steelhead and coho salmon. In life history and genetic attributes, the Elwha River chinook salmon appear to be transitional between populations from Puget Sound and the Washington Coast ESU.
A majority of the BRT considered that Elwha River chinook salmon were part of the Puget Sound ESU. A minority of the BRT felt that the Elwha River chinook salmon belonged in the Washington Coast ESU, and a further minority was undecided.
This ESU includes all native populations from the mouth of the Columbia River to the crest of the Cascade Range, excluding populations above Willamette Falls. Celilo Falls, which corresponds to the edge of the drier Columbia Basin Ecosystem and historically may have presented a migrational barrier to chinook salmon at certain times of the year, is the eastern boundary for this ESU. Not included in this ESU are "stream-type" spring-run chinook salmon found in the Klickitat River (which are considered part of the Mid-Columbia River Spring-Run ESU) or the introduced Carson spring-chinook salmon strain. "Tule" fall chinook salmon in the Wind and Little White Salmon Rivers are included in this ESU, but not introduced "upriver bright" fall-chinook salmon populations in the Wind, White Salmon, and Klickitat Rivers. Available information suggests that spring-run chinook salmon presently in the Clackamas and Sandy Rivers are predominantly the result of introductions from the Willamette River ESU and are thus probably not representative of spring-run chinook salmon historically found in these two rivers.
In addition to the geographic features mentioned above, genetic and life-history data were important factors in defining this ESU. Populations in this ESU are considered ocean type. Some spring-run populations have a large proportion of yearling migrants, but this trend may be biased by yearling hatchery releases. Subyearling migrants were found to contribute to the escapement. CWT recoveries for Lower Columbia River ESU populations indicate a northerly migration route, but with little contribution to the Alaskan fishery. Populations in this ESU also tend to mature at ages 3 and 4, somewhat younger than populations from the coastal, upriver, and Willamette ESUs. Ecologically, the Lower Columbia River ESU crosses several ecoregions: Coastal, Willamette Valley, Cascades and East Cascades.
This ESU includes native spring-run populations above Willamette Falls. Fall chinook salmon above the Willamette Falls were introduced and are not considered part of this ESU. Populations in this ESU have an unusual life history that shares features of both the stream and ocean types. Scale analysis of returning fish indicate a predominantly yearling smolt life-history and maturity at 4 years of age, but these data are primarily from hatchery fish and may not accurately reflect patterns for the natural fish. Young-of-year smolts have been found to contribute to the returning 3-year-old year class. The ocean distribution is consistent with an ocean-type life history, and CWT recoveries occur in considerable numbers in the Alaskan and British Columbian coastal fisheries. Intrabasin transfers have contributed to the homogenization of Willamette River spring-run chinook salmon stocks; however, Willamette River spring-run chinook salmon remain one of the most genetically distinctive groups of chinook salmon in the Columbia River Basin.
The geography and ecology of the Willamette Valley is considerably different from surrounding areas (see discussion of the Willamette Valley Ecoregion). Historically, the Willamette Falls offered a narrow temporal window for upriver migration, which may have promoted isolation from other Columbia River stocks.
Included in this ESU are stream-type chinook salmon spawning in the Klickitat, Deschutes, John Day, and Yakima Rivers. Historically, spring-run populations from the Hood, Walla Walla, and Umatilla Rivers may have also belonged in this ESU, but these populations are now considered extinct. Chinook salmon from this ESU emigrate to the ocean as yearlings and apparently migrate far off-shore, as they do not appear in appreciable numbers in any ocean fisheries. The majority of adults spawn as 4-year-olds, with the exception of fish returning to the upper tributaries of the Yakima River, which return predominantly at age 5. Populations in this ESU are genetically distinguishable from other stream-type chinook salmon in the Columbia and Snake Rivers. Streams in this region drain desert areas east of the Cascades (Columbia Basin Ecoregion) and are ecologically differentiated from the colder, less productive, glacial streams of the upper Columbia River Spring-Run ESU and from the generally higher elevation streams of the Snake River.
There were two different minority BRT opinions regarding fish from this area. Some BRT members felt that all stream-type chinook salmon populations from the Columbia River Basin (or all populations outside the Snake River) are part of a single ESU. A separate minority felt that the Yakima River populations should be considered a separate ESU from spring-run populations downstream from the Snake River.
(Formerly known as the Mid-Columbia River Summer/Fall Chinook salmon ESU.)
Waknitz et al. (1995) and NMFS (1994a) identified an ESU that included all ocean-type chinook salmon spawning in areas between McNary Dam and Chief Joseph Dam. The BRT for the current status review concluded that the boundaries of this ESU do not extend downstream from the Snake River. In particular, the BRT concluded that Deschutes River fall chinook salmon are not part of this ESU. The ESU status of the Marion Drain population from the Yakima River is still unresolved. The BRT also identified the importance of obtaining more definitive genetic and life-history information for naturally spawning fall chinook salmon elsewhere in the Yakima River drainage.
Fish from this ESU primarily emigrate to the ocean as subyearlings but mature at an older age than ocean-type chinook salmon in the Lower Columbia and Snake Rivers. Furthermore, a greater proportion of CWT recoveries for this ESU occur in the Alaskan coastal fishery than is the case for Snake River fish. The status review for Snake River fall chinook salmon (Waples et al. 1991b, NMFS 1992) also identified genetic and environmental differences between the Columbia and Snake Rivers. Substantial life-history and genetic differences distinguish fish in this ESU from stream-type spring-run chinook salmon from the mid- and upper-Columbia Rivers.
This ESU falls within part of the Columbia Basin Ecoregion. The area is generally dry and relies on Cascade Range snowmelt for peak spring flows. Historically, this ESU may have extended farther upstream; spawning habitat was compressed down-river following construction of Grand Coulee Dam.
This ESU includes stream-type chinook salmon spawning above Rock Island Dam--that is, those in the Wenatchee, Entiat, and Methow Rivers. All chinook salmon in the Okanogan River are apparently ocean-type and are considered part of the Upper Columbia River Summer- and Fall-Run ESU. These upper Columbia River populations exhibit classical stream-type life-history strategies: yearling smolt emigration with only rare CWT recoveries in coastal fisheries. These populations are genetically and ecologically well separated from the summer- and fall-run populations that exist in the lower parts of many of the same river systems. Morphological differences and meristic traits also distinguish stream and ocean types in the Columbia and Snake River Basins (Schreck et al. 1986).
Rivers in this ESU drain the east slopes of the Cascade Range and are fed primarily by snowmelt. The waters tend to be cooler and less turbid than the Snake and Yakima Rivers to the south. Although these fish appear to be closely related genetically to stream-type chinook salmon in the Snake River, the BRT recognized substantial ecological differences between the Snake and Columbia Rivers, particularly in the upper tributaries favored by stream-type chinook salmon. Allozyme data demonstrate even larger differences between spring-run chinook salmon populations from the mid- and upper Columbia River.
Artificial propagation programs have had a considerable influence on this ESU. During the Grand Coulee Fish-Maintenance Project (GCFMP 1939-43), all spring-run chinook salmon reaching Rock Island Dam, including those destined for areas above Grand Coulee Dam, were collected, and they or their progeny were dispersed into streams in this ESU (Fish and Hanavan 1948). Some ocean-type fish were undoubtedly also incorporated into this program. Spring-run escapements to the Wenatchee, Entiat, and Methow Rivers were severely depressed prior to the GCFMP but increased considerably in subsequent years, suggesting that the effects of the program may have been substantial. Subsequently, widespread transplants of Carson stock spring-run chinook salmon (derived from a mixture of Columbia River and Snake River stream-type chinook salmon) have also contributed to erosion of the genetic integrity of this ESU. Nevertheless, the majority of the BRT felt that, in spite of considerable homogenization, this ESU still represents an important genetic resource, in part because it presumably contains the last remnants of the gene pools for populations from the headwaters of the Columbia River. A minority of the BRT felt that chinook salmon in this area should be considered part of a larger ESU that includes other Columbia River (and perhaps Snake River) populations of stream-type chinook salmon.
This ESU, which includes ocean-type fish, was identified in an earlier status review (Waples et al. 1991b, NMFS 1992) based on genetic, life history, and ecological differences between Columbia and Snake River populations. In that status review and in a later review of mid-Columbia River summer-run chinook salmon (Waknitz et al. 1995), the ESU status of populations from Marion Drain and the Deschutes River was not resolved, so these issues were considered in the current review. Both populations show a greater genetic affinity to Snake River fall chinook salmon than to other ocean-type Columbia River populations.
As the origin of both of these populations is uncertain, the BRT considered several possible alternative hypotheses. The Marion Drain is an irrigation channel dug early in this century that is used to return irrigation water to the Yakima River. Perhaps because of the relative inhospitability of the mainstem Yakima River, the channel appears to be favored by spawning chinook salmon and other species. Obviously, the current population is not native to this artificial channel, but it may represent a native population that at one time inhabited the mainstem Yakima River or other nearby areas. Under this scenario, the fish in Marion Drain might better reflect the historical Yakima River fall chinook salmon than do fish currently spawning in the mainstem, which is heavily stocked with fish from the Priest Rapids/Bonneville Hatchery upriver "bright" stock. The genetic affinity between the Marion Drain and Snake River fish thus might reflect a historical link between areas that share some ecological similarities (e.g., relatively high summer water temperatures). Alternatively, the current population might have colonized Marion Drain from the Snake River more recently, perhaps as Snake River fish were displaced from their historic spawning areas by the series of impassable dams in Hells Canyon or by flooding of habitat by the four dams on the lower Snake River. Finally, the current Marion Drain population might be the result of stock transfers during the past several decades. Several possible scenarios involving stock transfers have been hypothesized, but the BRT found no direct evidence to substantiate them. In either of these latter two scenarios, the Marion Drain fish would be considered an introduced population and therefore not an ESA issue, except perhaps as a reserve source of genetic material for the listed Snake River population.
After considerable discussion, the majority of the BRT concluded that chinook salmon spawning in the Marion Drain could not with any certainty be assigned to any historic or current ESU.
The Deschutes River historically supported a population of fall chinook salmon, as evidenced by counts of fish at Sherars Falls in the 1940s. Genetic and life-history data for the current population indicate a closer affinity to fall chinook salmon in the Snake River than to those in the Columbia River. Similarities were observed in the distribution of CWT ocean recoveries for Snake River and Deschutes River fall-run chinook salmon; however, information on Deschutes River fish was based on a limited number of releases over a relatively short time frame. One hypothesis is that these similarities reflect a historic relationship between populations in the Deschutes and Snake Rivers. Another hypothesis is that displacement of Snake River fish by construction of John Day Dam and/or the lower Snake River dams led to colonization of the Deschutes River by Snake River fish and interbreeding with, or replacement of, the native fish. There was a considerable increase in the run-size of fall chinook salmon in the Deschutes River following the construction of John Day Dam, although it has been suggested that these fish may have been local mainstem spawners whose spawning areas were inundated (Nehlsen 1995). Coded-wire-tag data indicate that straying by non-native chinook salmon into the Deschutes River is very low and does not appear to be disproportionately influenced by Snake River fall-run chinook salmon (Hymer et al. 1992b).
After considerable discussion, a plurality of the BRT concluded that the Deschutes River population should be considered part of the Snake River Fall-Run ESU. Separate minorities favored two other scenarios: 1) The Deschutes River population is part of a separate ESU that historically also included ocean-type fish in the Umatilla, John Day, and Walla Walla Rivers. Populations in the later three rivers are considered to be extinct (Kostow 1995). 2) All ocean-type chinook salmon upstream of the historical site of Celilo Falls (approximately the location of the Dalles Dam) belonged to one ESU. A further minority was undecided on the ESU status of these populations. All of the BRT members were concerned about the lack of definitive information for the Deschutes River population(s).
This ESU, which includes populations of spring- and summer-run chinook salmon from the Snake River Basin (excluding the Clearwater River), was identified in a previous status review (Waples 1991, NMFS 1992). These populations show modest genetic differences, but substantial ecological differences, in comparison with Columbia River stream-type populations. Populations from this ESU emigrate to the ocean as yearlings, mature at ages 4 and 5, and are rarely taken in ocean fisheries. The majority of the spawning habitat occurs in the Northern Rockies and Blue Mountains ecoregions. A minority of the BRT felt this ESU should be combined with stream-type spring-run chinook salmon from the Columbia River.
Marshall et al. (1995) identified Major Ancestral Lineages (MALs) and Genetic Diversity Units (GDUs=subsets of MALs) for chinook salmon in Washington State. This effort, which seeks to identify the existing amount and patterns of genetic diversity within the state, supports the goals of the Wild Salmonid Policy under development by state and tribal fishery managers and is intended to facilitate its implementation. The terminology (GDUs and MALs) differs somewhat from that of previous documents prepared by WDW and WDFW (Leider et al. 1995). According to Busack and Marshall (1995), GDU designations were based on a combination of genetic, life history/ecological, and physiographic/ecoregion data.
ODFW has designated Gene Conservation Groups (GCGs) for salmonid and non-salmonid fishes (Kostow 1995). These designations are part of the implementation of the Oregon Wild Fish Management Policy and Wild Fish Gene Resource Conservation Policy. The definition of the GCG is roughly equivalent to WDFW's GDU and considers similar criteria: genetic, meristic, geographic, and life-history differences. In addition, ODFW has presented NMFS with specific recommendations for ESU boundaries (ODFW 1995).
Comparison of proposed ESUs with state conservation management groups is complicated in some cases by the restricted scope of the state evaluations. For example, ESUs can extend across state (or even international) borders, but Washington and Oregon generally only considered populations within their respective state boundaries. Nevertheless, comparison of proposed ESUs for chinook salmon with Washington's GDUs and MALs supports the prediction by Marshall et al. (1995) that individual ESUs would often include multiple GDUs but would be unlikely to include multiple MALs. The Puget Sound ESU and Washington Coast ESU generally correspond to the WDFW Puget Sound Chinook salmon MAL and Coastal and Strait of Juan de Fuca Chinook salmon MAL, with the exception of the Elwha and Dungeness River populations, which WDFW placed in the Coastal and Strait of Juan de Fuca MAL (Table 4).
The boundaries for ESUs on the Oregon coast correspond with one of the scenarios recommended by ODFW. The Oregon Coast ESU includes five GCGs from the Elk River to the Nehalem River and Elk Creek. The Oregon portion of the Southern Oregon and California Coastal ESU is composed of a single GCG (Table 5).
The Lower Columbia River ESU incorporates several GCGs and generally agrees with the ODFW recommendation for an ESU. The Willamette River ESU also corresponds to an ESU suggested by ODFW; however, whereas ODFW considers spring-run chinook salmon in the Clackamas and Sandy Rivers to be part of this ESU, the BRT considered these to be introduced populations.
The Mid-Columbia Spring-Run ESU contains portions of the Upper Columbia and Snake Spring Chinook Salmon MAL and Upper Columbia Summer and Fall, Snake Fall, and Mid & Lower Columbia MAL. The Klickitat River was determined by WDFW to belong to a separate Lower and Mid-Columbia MAL relative to the other rivers in this ESU, in contrast to ODFW's recommendation to group the Klickitat, Deschutes, and John Day Rivers into one ESU. ODFW grouped the Deschutes River and John Day River spring-run chinook salmon into the Mid-Columbia Spring GCG, which historically would have also included the now extinct Hood, Umatilla, and Walla Walla River spring chinook salmon runs. It is not clear whether ODFW considered the Yakima River in their evaluations. The Upper Columbia Spring-Run ESU corresponds with the Upper Columbia Spring Genetic Diversity Unit (GDU), which is a subunit of the larger Upper Columbia and Snake Spring Chinook salmon MAL designated by WDFW.
The Upper Columbia Summer- and Fall-Run ESU boundaries incorporate two GDUs designated by WDFW within the Upper Columbia Summer and Fall, Snake Fall, and Mid & Lower Columbia MAL. The WDFW GDUs include introduced "upriver bright" fall chinook salmon in the Klickitat, White Salmon, and Wind Rivers that were not considered by the BRT.
The Snake River Fall-Run ESU is geographically a component of the Mid-Columbia and Snake Fall Chinook salmon GDU designated by WDFW. This GDU includes upriver "brights" from the Hanford Reach, lower Yakima River, and Marion Drain, in addition to the Snake River fall-run chinook salmon. ODFW has designated separate GCGs for Deschutes and Snake River fall chinook salmon, and recommend that the Deschutes River fall chinook salmon constitutes its own ESU.
The Snake River Spring- and Summer-Run ESU includes the WDFW Snake River Spring GDU, ODFW Snake Spring/Summer GCG, and other populations in Idaho.
The historic distribution and life history of chinook salmon most closely resembles those of coho salmon and steelhead. Ocean-type chinook salmon prefer to spawn in mainstem rivers and larger tributaries with relatively low gradients and generally have a shorter freshwater residence time than do coho salmon and steelhead in the same geographic area. In comparing
coastal ESU boundaries, because of their preference for smaller systems to spawn in and extended freshwater rearing period, steelhead and coho salmon probably exhibit a finer scale of ecological adaptation than do ocean-type chinook salmon. Conversely, in inland regions stream-type chinook salmon and steelhead express similar life-history strategies and there is a greater similarity in ESU boundaries. Differences in ESU boundaries among these species may also be related to artificial propagation practices and anthropogenic changes in habitat quality or access.
The boundaries for the Central Valley Fall-Run ESU correspond to those for the Central Valley Steelhead ESU. Chinook and coho salmon (Weitkamp et al. 1995) and steelhead (Busby et al. 1996) ESU designations for coastal California and southern Oregon are quite different, except that all three share a common boundary at Cape Blanco, on the Oregon Coast (Fig. 25). Cape Blanco is a recognized biogeographical transition zone for aquatic organisms. In the steelhead and coho salmon ESU determinations, the Klamath River Basin was incorporated with coastal systems, whereas it is proposed as a separate ESU for chinook salmon. In other coastal areas the Oregon Coast and Puget Sound ESUs were generally the same for all three species.
The ESU boundaries for the chinook salmon Washington Coast ESU encompasses the steelhead Olympic Peninsula ESU and a portion of the Southwest Washington ESU, as well as the coho salmon Olympic Peninsula and Southwest Washington Coast ESUs.
The Lower Columbia River ESU incorporates portions of ESUs designated for coho salmon and steelhead, but most notably shares similar geographic boundaries at the Willamette Falls, the Oregon Coast, and the Cascade Crest. The Willamette River, above Willamette Falls, forms a geographically defined area that contains separate chinook salmon and steelhead ESUs.
Beyond the Cascade Crest, native coho salmon populations have been extirpated. The three stream-type chinook salmon ESUs east of the Cascades correspond almost exactly with those for steelhead (Fig. 26). The ESUs for ocean-type chinook salmon east of the Cascades have no analogue in steelhead ESU designations.