Artificial propagation efforts on the upper Willamette River began early this century, when the state of Oregon began operating a hatchery on the McKenzie River in 1902 (Olsen et al. 1992). From 1909 to 1942 eggs were collected from spring-run adults returning to the Santiam and Middle Fork Willamette Rivers, incubated at the state's Bonneville Hatchery, and the resulting fry returned to the Willamette River Basin (Howell et al. 1985). Egg collections from the four primary state-run stations on the Willamette River Basin--North Santiam, South Santiam, McKenzie, and Middle Fork Willamette River stations--totalled 668 million eggs during the 1918-42 period (Craig and Townsend 1946). These eggs were largely the source for the 382 million fingerlings released into the basin during that interval. Although there were introductions of non-native fish into this ESU during the first half of this century, the vast majority of the eggs used originated from fish returning to the upper Willamette River (Howell et al. 1985, Olsen et al. 1992). Cramer et al. (1996) provided a detailed description of hatchery development in the Willamette River watershed.
Although not located within the boundaries of the Upper Willamette River ESU, the Clackamas River contains several artificial propagation facilities that have been strongly associated with the upper Willamette River. The U.S. Fish Commission began operating a hatchery on the Clackamas River in 1888 (USCFF 1893). Several million eggs were obtained annually until 1893, when dam construction limited spawner access to the hatchery collection facilities. Egg collecting substations on the upper Clackamas and Salmon Rivers (a tributary of the Sandy River) were constructed in 1894 and 1895, respectively, to provide eggs for the main Clackamas Hatchery (Ravenel 1899). Spawning times for fish arriving at these substations, July-September, were considerably earlier than those recorded at the Clackamas River Hatchery, September-October (Ravenel 1899). Additionally, egg transfers from the Baird NFH (Sacramento River) and the Little White Salmon Hatchery substation were also used to maintain production from the Clackamas River Hatchery. Dam construction and habitat degradation in the Clackamas River Basin nearly eliminated the spring run of chinook salmon. Restoration efforts for the Clackamas River chinook salmon utilized transfers of Mackenzie River spring-run chinook salmon and the construction of new artificial propagation facilities: the USFWS Eagle Creek NFH in 1957, and the ODFW Clackamas Hatchery in 1979 (Delarm and Smith 1990a,c). The original broodstocks for both hatcheries were developed from stocks originating above Willamette Falls (Delarm and Smith 1990c, Willis et al. 1995). Between 1975 and 1987, about 1.2 million spring-run chinook salmon were released from Eagle Creek NFH; none have been released since then. The Clackamas River Hatchery continues to produce between 0.5 and 1.2 million fish per year (NRC 1996) (Appendix D). Several broodstocks were originally developed from populations in the Clackamas, Santiam, McKenzie, and Middle Fork Willamette Rivers; inter-hatchery stock transfers have been frequent and the broodstocks have become essentially a single, homogenized breeding unit (Kostow 1995, Cramer et al. 1996). Therefore, spring-run chinook salmon currently inhabiting the Clackamas River are thought to most closely resemble hatchery populations throughout the Willamette River (Cramer et al. 1996).
Current hatchery programs in this ESU were initiated or expanded to mitigate the loss of natural spawning and rearing areas lost due to the construction of dams in the 1950s and 1960s (Cramer et al. 1996). Most of the historical geographic range of spring-run chinook salmon in the Willamette River Basin has received introductions of hatchery fish (Cramer et al. 1996, NRC 1996). Due to the large and continuous nature of artificial propagation programs in the Willamette River system, wild populations are thought to be small and "vastly dominated by hatchery fish" (Kostow 1995, p. 44). Hatchery fish have been observed spawning in the wild and appear to be successfully reproducing (Cramer et al. 1996).
Hatchery practices have reduced the early and late segments of the spawning cycle in this ESU. Historically, the several wild populations of spring-run chinook salmon in the Willamette River spawned sometime between mid-July and late October. However, current Willamette River populations, both wild and hatchery, all spawn at the same time, during September. Therefore, the majority of natural spawners are now thought to be of recent hatchery origin (Cramer et al. 1996). In addition, hatchery strays are thought to have a significant impact on population dynamics in this ESU. It has been estimated that the straying rate of adults returning from releases of trucked juveniles can be as high as 75% (Cramer et al. 1996). These strays are thought to contribute to the naturally spawning population (Kostow 1995).
Although fall-run chinook salmon are not indigenous to the Willamette River Basin (Howell et al. 1985), large numbers have been introduced there. Since the 1950s, about 200 million fall-run chinook salmon have been introduced into this ESU, primarily from lower Columbia River stocks (e.g., the ODFW Bonneville Hatchery), in addition to a large number of fish from the Trask River (Appendix D). Fall-run chinook salmon have been distributed into nearly all watersheds formerly and currently occupied by spring-run chinook salmon (Appendix D). Currently, the only facility releasing Bonneville Hatchery fall-run chinook salmon stock into the Willamette River above the falls is the Stayton Pond, a satellite of the South Santiam Hatchery, which produces about 5 million fall-run chinook salmon each year for release into various Willamette River tributaries (Delarm and Smith 1990c, NRC 1996). Little is known about the impact of introduced fall-run chinook salmon, as no observations of upper Willamette River fall-run chinook salmon were included in a recent review of wild chinook salmon stocks in Oregon (Kostow 1995). However, a previous review reported that between 16% and 46% of the adult fall-run chinook salmon in the upper Willamette River were of natural origin, suggesting at least a moderate amount of successful reproduction by straying hatchery fall-run chinook salmon (Howell et al. 1985). Spawning of fall-run chinook salmon in the upper Willamette River has been observed to occur primarily during September (Howell et al. 1985), closely overlapping the spawning period of Willamette River spring-run chinook salmon. We found no studies that evaluated genetic or ecological interactions between fall- and spring-run chinook salmon in the upper Willamette River.
The artificial propagation of spring-run chinook salmon is a relatively new management strategy in this ESU. A hatchery program was initiated on the Klickitat River in 1899, but the facility was poorly sited and abandoned shortly thereafter (Mayhall 1925). It was not until 1950 that a hatchery was reestablished on the Klickitat River (Moore et al. 1960). This hatchery was the first Washington hatchery built under the Lower Columbia River Development Plan (Moore et al. 1960). Hatchery operations in the Deschutes River Basin began in 1947 with the construction of a hatchery and weir near Spring Creek on the Metolius River, a tributary to the Deschutes River (Nehlsen 1995). During the next 12 years, the Metolius Hatchery released an average of 125,000 spring-run chinook salmon juveniles annually (Nehlsen 1995). Additional spring-run chinook salmon hatcheries on the Deschutes River were built, in part, to mitigate for natural production lost as a result of the construction of Pelton and Round Butte Dams. The Round Butte Hatchery (1972), and Pelton Ladder (1974), a Round Butte satellite facility, are operated by ODFW (Delarm and Smith 1990c). The Warm Springs NFH (1977) is operated by the USFWS (Delarm and Smith 1990a). Additionally, the Deschutes River has received over 20 million fish since the late 1940s. The majority of these were derived from native Deschutes River spring-run chinook salmon (Howell et al. 1985), although a relatively limited number of fish from the Carson NFH and Willamette River hatcheries were released prior to 1969 (Olsen et al. 1992, Kostow 1995, NRC 1996).
Yakima River chinook salmon populations were not directly influenced by the artificial propagation efforts associated with the Grand Coulee Fish Maintenance Project during the 1940s. Despite irrigation diversion screening and improvements in fish ladders on the Yakima River from 1936 to 1941, massive water withdrawals for irrigation were the primary cause for the continuous decline in spring-run chinook salmon populations during most of this century (Davidson 1953), and eventually necessitated the use of artificial propagation to maintain fish numbers. Native Yakima River spring-run chinook salmon populations do not appear to have been significantly affected by hatchery supplementation or straying (Marshall et al. 1995), even though the number of hatchery smolts released into the Yakima River during the 1980s may have exceeded the number of naturally produced smolts migrating downstream (Fast et al. 1991, NRC 1996). While hatchery smolts were sometimes more numerous than wild smolts, they had only about 1/80th of the smolt-to-adult survival rate of naturally produced spring-run chinook salmon (Fast et al. 1991). The most commonly released stock in the Yakima River has been from the Leavenworth NFH (Appendix D), but these fish were apparently ill-adapted to the Yakima River (based on their extremely poor survival). In 1976, about 20,000 Klickitat Hatchery spring-run chinook salmon were introduced in Marion Drain, a tributary of the lower Yakima River (Appendix D). In general, spring-run chinook salmon populations in the Yakima River have been almost exclusively maintained by natural production (WDF et al. 1993). All transfers of spring-run chinook salmon into the Yakima ceased in 1988 (Appendix D).
The John Day River has been stocked with just a few fish, mostly from local stock, and has not been stocked at all since 1982 (Appendix D). Few hatchery strays from other river systems have been found there.
Native spring-run chinook salmon are thought to be extinct in the Hood, Umatilla, and Walla Walla Rivers on the Oregon side of this ESU (Kostow 1995). Reintroduction programs are currently underway in the Hood and Umatilla Rivers, with the Carson NFH (Wind River) and Lookingglass Hatchery (Grande Ronde River) being the predominant sources for spring-run chinook salmon used in these programs (Appendix D). The Umatilla River has received over 5 million Carson and Lookingglass Hatchery fish since 1986 (NRC 1996).
Large numbers of spring-run chinook salmon (approximately 11.8 million) have been released directly into the mainstem Columbia River since the 1970s, principally from WDFW Ringold Hatchery in the Hanford Reach, although smaller releases have occurred in the vicinity of Priest Rapids Dam (Appendix D). The stocks most commonly used in the Hanford Reach releases have been from the Carson NFH, and the WDFW Cowlitz and Klickitat River Hatcheries (Appendix D). There is no documented observation of spawning by spring-run chinook salmon in the Hanford Reach nor any other mainstem locations in the Columbia River (Fish and Hanavan 1948, Fulton 1968, WDF et al. 1993, Chapman et al. 1995). It is probable that many of the adults produced from these mainstem releases sought out tributary spawning areas. Stuehrenberg et al. (1995) observed adult hatchery spring-run chinook salmon from the Ringold Hatchery releases passing over Priest Rapids Dam. Spawned-out carcasses from Ringold Hatchery releases have been recovered in the Wenatchee River Basin (Peven 1994).
Artificial propagation in this ESU began in 1899, when hatcheries were constructed on the Methow and Wenatchee rivers (Mullan 1987). The Tumwater Hatchery on the Wenatchee River apparently released only 600,000 chinook salmon fry in 1903, while a hatchery on the Methow River produced primarily coho salmon, but a few chinook salmon were released as well before it was closed in 1913 (Craig and Suomela 1941, Nelson and Bodle 1990). The Leavenworth State Hatchery operated in the Wenatchee River Basin between 1913 and 1931. Eggs were procured from the Willamette River (spring-run chinook salmon), and from the Chinook Hatchery on the lower Columbia River (probably "tule" fall-run chinook salmon), apparently due to difficulties associated with collecting native stocks. In 1915, a hatchery at Pateros in the Methow River Basin released chinook salmon of lower river origin, but Craig and Suomela (1941) concluded that these fish probably were not able to successfully return to the Methow River . Between 1931 and 1939, no chinook salmon hatcheries were in operation above Rock Island Dam. Chinook salmon were released from the county trout hatchery at Kittitas, Washington from about 1923 to 1931. There is no record of any eggs being collected at this site, but approximately 6,500,000 chinook salmon fry (most likely fall-run chinook salmon from the Kalama River Hatchery) were released into the Yakima River Basin (WDF 1934).
The construction of Grand Coulee Dam (1941, RKm 959) prevented thousands of adult spring-run chinook salmon from reaching their natal streams. In an effort to mitigate the loss of spawning habitat above the dam, the Grand Coulee Fish Maintenance Project (GCFMP) was authorized by the federal government. The GCFMP sought to relocate all chinook salmon migrating past Rock Island Dam (RKm 730) into three of the remaining accessible tributaries to the Columbia River: the Wenatchee, Entiat, and Methow Rivers. As a part of this relocation, efforts were made to improve salmonid habitat (primarily through the screening of irrigation systems) and to increase run sizes through artificial propagation (Fish and Hanavan 1948). Several hatchery sites were designated as part of the GCFMP; the primary site on Icicle Creek, a tributary to the Wenatchee River, would later become the Leavenworth NFH (1940). Secondary substations were to be located on the Entiat (Entiat NFH, 1941), Methow (Winthrop NFH, 1941), and Okanogan Rivers. The hatchery on the Okanogan River was never developed due to the lack of a suitable site and wartime building restrictions (Fish and Hanavan 1948).
In 1938, the last salmon was allowed to pass upstream through the uncompleted Grand Coulee Dam. The trapping of adult salmon at Rock Island Dam began in May 1939 and continued until the autumn of 1943. Spring- and summer/fall-run fish were differentiated according to the time of their arrival at Rock Island Dam. A separation date of 9 July was established, based on weekly counts observed during 1933-38 (Fish and Hanavan 1948). However, Mullan (1987) estimated that 23 June was a more accurate discriminator between the two run times. It is likely that some summer-run fish were misidentified as belonging to the spring run. The GCFMP combined all late-run fish passing Rock Island Dam, including those destined for now-inaccessible spawning areas in Washington and British Columbia (Fish and Hanavan 1948). Offspring of these adults were reared at the newly constructed Leavenworth, Entiat, and Winthrop NFHs, and transplanted into the Wenatchee, Methow, and Entiat Rivers (Fish and Hanavan 1948). Furthermore, a number of late-run adults were transported to Nason Creek, a tributary to the Wenatchee River, and the Entiat River and allowed to spawn naturally.
The only tributary above Rock Island Dam that did not receive spawning adults or mixed-stock hatchery juveniles during the 5-year GCFMP was the Okanogan River (Fish and Hanavan 1948, Mullan et al. 1992). Chinook salmon adults destined for the Okanogan River from 1939 to 1943 were intercepted and included in the GCFMP mitigation efforts. With the exception of possibly a very small number of 6-year-old chinook salmon, native Okanogan River fish were eliminated or absorbed into other populations. The ocean-type chinook salmon now observed in the Okanogan River are likely strays originating from other tributaries or from the mainstem Columbia River (Mullan 1987).
Spawning channels were constructed near Wells, Rocky Reach, and Priest Rapids Dams in the mid-1960s and continued operations for several years, but were eventually abandoned due to high pre-spawning mortality and overall poor production of returning adults; these facilities were converted to conventional hatcheries and are currently in operation near these sites (Nelson and Bodel 1990). In addition, several acclimation ponds are now being used as a part of recent management changes to develop local stocks for Columbia River tributaries above Priest Rapids Dam (Chapman et al 1994).
Ocean-type chinook salmon in this ESU have been mixed considerably over the past five decades, not only among stocks, but among putative "runs" as well. This mixing was due to the variety of methods employed to collect broodstock at dams, hatcheries, or other areas and as a result of juvenile introductions into various areas (reviewed in Chapman et al. 1994). Recoveries of coded-wire-tagged adults derived from juvenile releases in the late 1970s and 1980s have indicated that wild and hatchery summer-run fish originating from above Rock Island Dam have spawned extensively with fall-run fish originating from the Hanford Reach and Priest Rapids Hatchery (Chapman et al. 1994). Similarly, a recent study of radio-tagged chinook salmon found that 10% of summer-run fish were distributed in the mainstem upper Columbia River (typically considered fall-run spawning habitat), while about 25% of fall-run chinook salmon (released from below the Priest Rapids Dam) were recovered as summer-run fish at Wells Hatchery and in the Okanogan River (Stuehrenberg et al. 1995). The possibility that substantial genetic exchange has taken place between chinook salmon populations above and below Rock Island Dam was hypothesized nearly 50 years ago (Fish and Hanavan 1948). Marshall et al. (1995) and Waknitz et al. (1995) reported that, partly as a result of hatchery practices, the genetic difference between summer- and fall-run chinook salmon in this ESU was "relatively small" and "essentially zero," respectively. Modifications in hatchery protocols and facilities in order to maintain discrete hatchery stocks have only recently been initiated (Utter et al. 1995).
There are currently no hatchery facilities on the Yakima River for ocean-type chinook salmon; however, the Yakima River has been heavily stocked with "upriver bright" ocean-type chinook salmon since 1980 (Appendix D). These transplanted stocks are reported to stray at substantial rates (Busack 1990, Hymer et al. 1992b, WDF et al. 1993). Similarities in the genetic composition among Yakima River, Hanford Reach, and Priest Rapids Hatchery ocean-type chinook salmon (Marshall et al. 1995, Waknitz et al. 1995) are thought to reflect the impact of hatchery releases of Hanford Reach/Priest Rapids fish on Yakima River chinook salmon (Busack et al. 1991). An average of 1 million "upriver bright" chinook salmon (none of which were derived from Yakima River returning adults) were released annually into the Yakima River Basin between 1980 and 1994 (Appendix D). In addition, strays from other programs, primarily the Umatilla River restoration effort, have been observed in the Yakima River (WDF et al. 1993). State and tribal management agencies have designated the Yakima River fall-run chinook salmon stock as of "unknown origin" and composite (mixed hatchery-derived and natural) production (WDF et al. 1993). There have been a limited number of unsuccessful summer-run chinook salmon introductions into the Yakima River as part of an effort to restore the early part of the ocean-type chinook salmon run (Appendix D).
Hatchery efforts with ocean-type chinook salmon in this ESU have been continuous and intensive since the implementation of the GCFMP, with numerous hatcheries constructed beginning in 1941 (Waknitz et al. 1995). From 1941 to the present, over 200 million ocean-type chinook salmon have been released into ESU 12 as either 0-age or yearling fish (Table 6). The percentage of non-indigenous stocks incorporated into this ESU has been low (about 3%), and does not appear to have had a significant impact on the integrity of this ESU (Chapman et al. 1995, Waknitz et al. 1995). However, the scale of hatchery chinook salmon elsewhere in the Columbia River Basin may pose risks for populations within this ESU. For example, as a result of large releases of ocean-type chinook salmon in the mainstem Columbia River and in the Yakima River in recent years, a substantial portion (approximately 50%) of the adults returning to ESU 12 appear to be of hatchery origin (Miller et al. 1990).
Early attempts to establish hatcheries on the Columbia River above the confluence of the Yakima River were generally unsuccessful. Beginning in 1899 with the construction of a fish hatchery on the Wenatchee River by the Washington Department of Fisheries and Game, hatcheries were constructed and subsequently abandoned on the Colville, Little Spokane, and Methow Rivers. Hatchery records indicate that relatively few chinook salmon were spawned (Craig and Suomela 1941). Attempts to improve the spring chinook salmon run with imported eggs (most notably from the upper Willamette River) were also apparently unsuccessful (Craig and Suomela 1941). By the 1930s, hatchery propagation of spring-run fish on the upper Columbia River had been terminated (WDF 1934).
The objectives and jurisdiction of the GCFMP are described in the previous ESU section. Adults collected for the GCFMP at Rock Island Dam were either transported to Nason Creek on the Wenatchee River to spawn naturally (1939-43), or to Leavenworth NFH for holding and subsequent spawning (1940-43). Over the course of 4 years, Nason Creek received 10,578 adult fish, of which an estimated 63.6% spawned successfully (Fish and Hanavan 1948). Beginning in 1940, some of the spring-run chinook salmon trapped at Rock Island Dam were spawned at the Leavenworth NFH. Eggs were incubated on site or transferred to the Entiat and Winthrop NFH. Almost 4 million fry and fingerlings were produced from adults collected at Rock Island Dam and subsequently released into the Wenatchee, Entiat, and Methow Rivers between 1940 and 1944 (Mullan 1987). In 1944, salmon were allowed to freely pass Rock Island Dam. In 1944 and 1945, a small number of spring-run adults returned to the Leavenworth and Winthrop NFHs; however, counts of fish migrating past Rock Island Dam indicated that a substantial number of fish probably spawned in the upriver tributaries (Fish and Hanavan 1948).
Artificial propagation efforts at Leavenworth NFH and Entiat NFH focused on the production of summer-run chinook salmon and other salmonids after 1943. In contrast, the culture of spring-run chinook salmon using local stocks continued at the Winthrop NFH through 1961. In the mid-1970s, there was a renewed effort to emphasize the production of spring-run chinook salmon at the three NFHs. In addition to the use of local stocks, there were large transfers of spring-run stocks from non-local sources: Carson NFH (Carson NFH stock), Little White Salmon NFH (Carson NFH stock), Klickitat WDFW hatchery (Klickitat River stock), and Cowlitz WDFW hatchery (Cowlitz River stock). In the early 1980s, imports of non-native eggs were reduced significantly, and thereafter the Leavenworth, Entiat, and Winthrop NFHs have relied on adults returning to their facilities for their egg needs (Chapman et al. 1995). Despite the current use of "local" fish in these hatcheries, a considerable amount of genetic introgression has probably occurred. Leavenworth, Entiat, and Winthrop NFH stocks are considered non-native (WDF et al. 1993), primarily derived from Carson NFH stocks (Hymer et al 1992b, Marshall et al. 1995). The current impact of hatchery fish on naturally spawning populations, especially those upriver from hatchery locations, appears to be slight, based on CWT recoveries from carcasses on the spawning grounds (Chapman et al. 1995).
Hatchery operations at the three NFHs in this ESU have been hampered by disease outbreaks, primarily BKD (Howell et al. 1985, Mullan et al. 1992, Hymer et al. 1992b, Chapman et al. 1995), which has been suggested as one of the causes of the generally low return rates observed for releases from these hatcheries (Mullan 1987, Chapman et al. 1995).
There are currently two hatcheries in this ESU operated by WDFW. The Methow Fish Hatchery Complex (MFHC, 1992) and Rock Island Fish Hatchery Complex (RIFHC, 1989) were both designed to implement supplementation programs for naturally-spawning populations on the Methow and Wenatchee Rivers, respectively (Chapman et al. 1995). The RIFHC uses broodstock collected at a weir on the Chiwawa River. Bugert (1998) discusses some of the difficulties these programs have experienced. Similarly, the MFHC uses returning adults collected at weirs on the Methow River and its tributaries, the Twisp and Chewuch Rivers (Chapman et al. 1995, Bugert 1998). Progeny produced from these programs are reared at and released from satellite sites on the tributaries where the adults were collected. Numerous other facilities have reared spring-run chinook salmon but on an intermittent basis.
In contrast to the lower and upper Columbia River, there was little effort directed toward the propagation of Snake River anadromous salmonids from the turn of the century through the 1960s, although a facility in the Grande Ronde River released an unknown number of fall-run chinook salmon between 1903 and 1907 (Howell et al. 1985). Early artificial propagation programs for fall-run chinook salmon in the Snake River were of limited scale and had little effect prior to 1976 (Howell et al. 1985, Waples et al. 1991b). Releases of marked fall-run chinook salmon (acquired from the Little White Salmon NFH) into the Salmon River in the 1920s did not result in any observed return of adults (Rich and Holmes 1928). In the early 1960s, eyed eggs from Snake River stocks were released above and below dams in the upper Snake River, but these efforts were apparently unsuccessful (Waples et al. 1991b).
In 1964, the Idaho Power Company was required to construct the Oxbow Hatchery below Oxbow Dam to mitigate the effects of the dam on fish returning to that section of the Snake River (Wahle and Smith 1979). Several million juveniles were released in the upper Snake River and in reservoirs above Oxbow Dam, but few returns were observed and the program was abandoned shortly thereafter. From 1955 to the present, fall-run chinook salmon juveniles have been released in reservoirs, apparently to provide sport fishing opportunities (Appendix D).
In 1960 and 1970, eyed eggs and juveniles, respectively, from the Spring Creek NFH were introduced into the Clearwater River Basin, but these efforts produced limited numbers of returning adults (Howell et al. 1985, Waples et al. 1991b). From 1960 to 1967, between 0.4 and 1.6 million eggs were collected annually at Oxbow Dam and transferred to the Clearwater River, but probably did not contribute many returning adults to the system (Waples et al. 1991b). Egg transfers to the Clearwater River were terminated in 1968.
Hatchery efforts to mitigate the effects of dam construction on fall-run chinook salmon populations in the Snake River Basin increased after the initiation of the Lower Snake River Compensation Plan (LSRCP) in 1976 (Mathews and Waples 1991). This program included the development of an egg bank program to ensure the genetic integrity of Snake River fall-run chinook salmon prior to the construction of propagation facilities dedicated to the compensation plan (Bugert and Hopley 1989, Nelson and Bodle 1990). This program involved, in part, the release of Snake River fall-run chinook salmon from the Kalama Falls Hatchery (WDFW) on the Kalama River, with additional egg incubation and early rearing being undertaken at the Hagerman NFH in Idaho (Waples et al. 1991b). As many as 1,500 adult Snake River fall-run chinook salmon returned annually to the Kalama Falls Hatchery or Ice Harbor Dam from 1981 to 1986 (Howell et al. 1985, Waples et al. 1991b).
Broodstock operations were transferred to the WDFW Lyons Ferry Hatchery when it began operations in 1984 (Delarm and Smith 1990d, Waples et al. 1991b). The Lyons Ferry Hatchery broodstock was derived from the Kalama Falls egg bank program and fish collected at Ice Harbor and Lower Granite Dams (Chapman et al. 1991). As a result of low numbers of naturally produced fall-run chinook salmon and an increasing number of hatchery-produced fish, the Snake River fall chinook salmon run was thought to be a composite of hatchery- and naturally produced fish by the mid-1980s (Howell et al. 1985). There are concerns that hatchery fish may now comprise a disproportionate number of naturally spawning fish throughout the Snake River Basin (ODFW 1991). Tagged fish from the Lyons Ferry Hatchery have been recovered from the mainstem Snake River and the Tucannon River (Nelson and Bodle 1990, Marshall et al. 1995). Between 7% and 67% (mean 38%) of fall-run chinook salmon passing over Lower Granite Dam have been first-generation hatchery fish (ODFW 1991). In addition, strays from the upper Columbia River Basin have recently been observed in substantial numbers (4% to 39%) at Lyons Ferry Hatchery, Lower Granite Dam, and on the spawning grounds (Waples et al. 1991b, Garcia et al. 1996, Mendel et al. 1996). There have not been any hatchery programs for fall-run chinook salmon on the Oregon side of the lower Snake River, although strays of mixed ancestry from the reintroduction program on the Umatilla River (Columbia River tributary) have been observed in the Snake River since the late 1980s (Chapman et al. 1991, Mendel et al. 1996). All Umatilla River hatchery fall-run chinook salmon are now being marked so they can be intercepted at the Snake River dams (Kostow 1995). Overall, with a few minor exceptions, native stocks have been used in Snake River fall-run chinook salmon hatchery programs (Table 6).
ODFW has also never had a fall-run chinook salmon hatchery on the Deschutes River (Kostow 1995). Small numbers of locally-derived and non-native fall-run chinook salmon were released into the Deschutes River up to the late 1970s; however, the success of these introductions is believed to have been very low (Howell et al. 1985). A limited number of strays from hatcheries on other rivers have been observed on the Deschutes River spawning grounds (Kostow 1995).
Artificial propagation efforts did not occur in ESU 15 as early as in other regions, nor in the same magnitude. From 1921 to 1934, the U.S. Fish and Fisheries Commission operated a hatchery at Salmon, Idaho. Eggs were collected from spring- and summer-run chinook salmon adults returning to the Lemhi and Pahsimeroi Rivers and the Yankee Fork of the Salmon River (Bowles and Leitzinger 1991). In all, 26,483,000 eggs were collected from local sources, incubated, and the progeny released into local waters. An additional 9,720,000 eggs were transferred to the Salmon River Hatchery (Idaho) substation from outside sources (7,720,000 from the McKenzie River and 2,000,000 eggs from the Little White Salmon NFH). The majority of juvenile fish were released as fingerlings. Following the 1934 broodyear, the Salmon hatchery was primarily devoted to trout production (Wahle and Smith 1979). Overall, stock transfers into the Snake River Basin were minimal prior to the mid-1900s (Matthews and Waples 1991).
Currently, the major spring- and summer-run chinook salmon propagation facilities (satellite facilities or adult collection weirs in parentheses) operating in the Snake River Basin area are: WDFW's Tucannon and Lyons Ferry Hatcheries; ODFW's Lookingglass and Wallowa (Big Canyon) Hatcheries; IDFG's Sawtooth (East Fork Salmon River), McCall, and Clearwater (Powell, Red River) Hatcheries; IPC's Rapid River and Pahsimeroi Hatcheries; and USFWS's Dworshak and Kooskia Hatcheries (Delarm and Smith 1990b). Stocks used in most ESU 15 hatcheries were derived from mixtures of non-indigenous stocks, or from a mix of non-indigenous and native stocks. Among the fish released into various Snake River Basins, there have been introductions from the Carson, Little White Salmon and Leavenworth NFHs, various Willamette River hatcheries, and the Cowlitz and Klickitat state hatcheries (Matthews and Waples 1991). The Tucannon River spring-run chinook salmon stock used at the Lyons Ferry Hatchery, the Imnaha River spring-run chinook salmon stock (reared at the Lookingglass Creek Hatchery, but released into the Imnaha River), and the Upper Salmon River Sawtooth Hatchery spring-run stock appear to have had minimal influence from out-of-basin stocks (Matthews and Waples 1991, Keifer et al. 1992). Additionally, the South Fork Salmon River summer-run chinook salmon stock reared at the McCall Hatchery has probably had minimal influence from outside sources (Matthews and Waples 1991, Keifer et al. 1992).
Spring- and summer-run stocks currently in the Clearwater River Basin are not part of this ESU, but artificial propagation activities for the basin are covered here because of their potential impact on the ESU. Native runs of spring- and summer-run chinook salmon on the Clearwater River were probably eliminated following the construction of the Lewiston Dam (1927) on the lower Clearwater River (Keifer et al. 1992). Modifications in the fish migration facilities at the dam were made in 1940, and from 1947 to 1953 approximately 100,000 spring-run chinook salmon eggs from the Middle Fork Salmon River were introduced annually into the Little North Fork of the Clearwater River (Fulton 1968, Keifer et al. 1992). Spawning channels on the Selway River were used in restoration efforts in the Clearwater River Basin. From 1961 to 1985 nearly 50 million eggs from the Rapid River Hatchery, Carson NFH, Spring Creek NFH, and the Salmon River were placed into various rearing/spawning channels (Keifer et al. 1992). The success of these transfers is unknown. In an effort to mitigate the effects of the construction of the Dworshak Dam, the Kooskia and Dworshak NFHs were constructed in 1967 and 1969, respectively (Keifer et al. 1992). Broodstock for these hatcheries came primarily from the Rapid River Hatchery, with significant contributions from Carson-stock hatcheries (Leavenworth, Little White Salmon, and Carson NFHs) and Willamette River hatcheries. Millions of fish have been released from the Dworshak and Kooskia Hatcheries, primarily as yearling smolts. More recently, these facilities have utilized adults returning to the hatcheries or satellite collection sites to supply gametes for their programs (Keifer et al. 1992).
Prior to 1985, the Tucannon River spring-run chinook salmon population was maintained entirely by natural production (Howell et al. 1985). A limited number of non-native fish were introduced in the Tucannon River--16,000 Klickitat River and 10,500 Willamette River spring-run chinook salmon in 1962 and 1964, respectively. Native broodstock were used to establish the Tucannon Hatchery spring-run chinook salmon population, although the number of fish available was limited (the total adult run size was approximately 200 fish during the early 1980s) (Howell et al. 1985). The absence of other spring-run chinook salmon propagation facilities nearby has probably limited introgression by non-native stocks, although a limited number of CWT-tagged hatchery-derived fish from the Umatilla River and Grande Ronde River (Rapid River stock) have been recovered (Marshall et al. 1995).
Spring-run chinook salmon hatchery programs were established in Oregon in the early 1980s as part of the LSRCP (ODFW 1991). The founding stocks used were transferred from the Carson NFH, and from the IDFG Rapid River Hatchery, which was founded from a mixture of Snake River populations (Howell et al. 1985, ODFW 1991). The Lookingglass Creek Hatchery initially utilized stock from the Carson NFH in 1982; however, adult returns were so poor and straying rates so high that the use of Carson stock was discontinued (Chapman et al. 1991, Kostow 1995). Carson NFH juveniles were also released into several non-hatchery streams and the returning adults may have interbred with native fish (ODFW 1991). Several years ago it was suggested that the hatchery programs "may be impeding the recovery of the wild populations in streams where hatchery facilities are located or where hatchery fish have been outplanted" (ODFW 1991, p. 14). Rapid River stock was subsequently imported during the late 1980s (Olsen et al. 1992). Beginning in 1989, returning adults (originating primarily from the Rapid River introductions) to Lookingglass Hatchery have provided gametes to produce subsequent releases (Olsen et al. 1992, Kostow 1995). Native stream-type chinook salmon populations in Lookingglass Creek are now thought to be extinct, and the location of current releases of the Lookingglass Hatchery stock has been restricted to prevent further introgression (Kostow 1995, Currens et al. 1996). For the past several years, stray hatchery fish of Rapid River stock origin have, on average, represented about half of all natural spawners throughout the Grande Ronde Basin (Crateau 1997). By contrast, the Imnaha River Acclimation Pond facility (1982) has collected gametes only from adults returning to the river, although the eggs have been incubated and juveniles reared at the Lookingglass Hatchery before being returned to the Imnaha site (Chapman et al. 1991, Olsen et al. 1992).
Several facilities for the propagation of spring- and summer-run chinook salmon exist in the Salmon River Basin. The Rapid River facility (1964) was constructed to mitigate the loss of spring-run chinook salmon spawning habitats resulting from the construction of the Hells Canyon Dam complex (Howell et al. 1985). Broodstock were collected from a trap at the Hells Canyon Dam on the Snake River from 1964 to 1969, and thereafter from broodstock returning to the hatchery weir on the Rapid River (Keifer et al. 1992). Fish from the Rapid River Hatchery and satellite facilities have been released in considerable numbers in the Rapid, Salmon, Snake, Clearwater, and Grande Ronde Rivers (Howell et al. 1985, Keifer et al. 1992). The Sawtooth Hatchery and satellite facilities (1985) on the Upper Salmon River have collected native returning spring chinook salmon for broodstock purposes (Howell et al. 1985, Delarm and Smith 1990b, Keifer et al. 1992). Rapid River fish were introduced into nearby watersheds through the 1980s (Keifer et al. 1992) and were used initially at the Sawtooth Hatchery.
Summer-run chinook salmon are propagated at McCall Hatchery (1980) and Pahsimeroi Hatchery (1969) (Delarm and Smith 1990b). The McCall Hatchery broodstock was initially collected at Little Goose and Lower Granite Dams and contained a mixture of Snake River summer-run stocks, with a lesser contribution by Snake River spring-run stocks (Chapman et al. 1991). Since 1981, a satellite facility on the South Fork Salmon River has collected adults (which consisted of returning McCall Hatchery releases and summer-run fish native to the South Fork Salmon River) to be used as broodstock for the McCall Hatchery (Keifer et al. 1992). The McCall Hatchery has been responsible for the majority of the 11 million juvenile summer chinook salmon released into the South Fork Salmon River (Appendix D). The Pahsimeroi Hatchery broodstock was founded with native summer-run fish returning to the Pahsimeroi River (Keifer et al. 1992). However, summer-run chinook salmon from the South Fork Salmon River (McCall Hatchery) were introduced into the Pahsimeroi River during 1985-90, and may have been integrated into the Pahsimeroi Hatchery broodstock (Keifer et al. 1992). Spring-run chinook salmon (Rapid River Hatchery stock) were also reared and released at the Pahsimeroi Hatchery for a limited time during the 1980s.
The Carson NFH stock has had a poor history in the Snake River Basin, not only for stock restoration, but also when used as a hatchery stock to increase harvest opportunities. Abundance in streams receiving Carson NFH fish is less than or no different than unenhanced streams (Chapman et al. 1991).