Long Range Plan for the Klamath River Basin Conservation Area Fishery Restoration Program
Part 1 of

CHAPTER 4: FISH POPULATION PROTECTION

CHAPTER 4: FISH POPULATION PROTECTION

* How do we identify the distinct salmonid stocks of the Klamath Basin and how do we protect their remaining genetic diversity?

* There is a need for the Klamath Fisheries Management Council to help protect the locally adapted stocks needed for population rebuilding while still providing for fisheries.

* Enforcement in the Klamath Basin is a huge problem: more wardens are needed, as well as stricter enforcement of possession limits.

* Is there a significant impact by high seas drift-netting on Klamath River salmon and steelhead?

* What is the impact of predators on salmon and steelhead populations?

* Should native stocks of steelhead be protected by catch-and-release regulations?

* We should judge the success of the Restoration Program on increases in native fish, not hatchery stocks.

INTRODUCTION

This section deals with the identification of anadromous fish stocks and trends in their run strength. Discussions concerning the protection of various stock groups from overharvesting, predation, and threats related to habitat destruction are also included.

It may seem that the matter of depletion is overstressed in this report, since its progress has been evident for years. A condition of increasing depletion was not sufficiently evident on the Klamath, however to be convincing to those most interested. In fact, opinions to the contrary were commonly held, some asserting that the runs were gradually building up. There is very little exact information concerning....the Klamath River previous to 1912.

J.O. Snyder

Thus wrote Dr. J.O. Snyder for the California Division of Fish and Game in 1931 about trends in run strength on the Klamath in the 1920's. The comments have a striking similarity to those of biologists around 1980. One need only substitute "1978" for 1912. The lack of exact information still holds true today for many of the river's fish stocks. Snyder was concerned that two- and three-year-old chinook salmon were dominating the ocean and river catch and that six-year-old fish had disappeared from the runs. It was not the first downturn in the river's fish populations (Hume in Snyder 1931).

Before Europeans settled in the Klamath Basin, the Yurok, Hoopa, and Karuk Indians had been sustained by the river's fishes for thousands of years. Weirs were constructed annually at various sites in the Hoopa Valley, at Red Cap Creek, and the largest at Cappell Creek below Weitchpec. Conservation of salmon populations was insured by use of harvest methods governed in accordance with a complex set of social and religious customs (Kroeber 1974). The behavior may have evolved from past experiences with food shortages after periods of overharvest (McEvoy 1986).

Mining was the first major impact of European culture on the Klamath watershed. The first wave of degradation changed the balance of the river's chinook stocks from predominantly spring chinook to fall chinook runs (Hume in Snyder 1931). The primary cause of the decline may have been the heavy sediment loads unleashed by hydraulic mining which filled the deep pools required by spring chinook for holding during summer (see effects of mining in Chapter 2). Sediment problems from mining were probably exacerbated by a large flood in 1861. Miners may have been heavily reliant on salmon as a food source. Snyder (1931) claimed that "large numbers of salmon were speared or otherwise captured as they neared their spawning beds, and if credence be given to the reports of old miners, there then appeared to be the first and perhaps major cause of early depletion." A splash dam was constructed across the Klamath at Klamathon in 1889 which blocked spring chinook passage into the upper Klamath basin until it was washed out by a flood in 1902 (Fortune et al. 1966). By 1892 spring chinook were thought to be almost extinct (Hume in Snyder 1931).

It is unlikely that the Indian harvest contributed substantially to the early decline of the spring run of chinook salmon. Spring chinook were not a high priority for subsistence harvest by Indians because the fish's high body fat made it unsuitable for drying and smoking. Because the river was often swollen and surging in the spring due to snow melt, spring chinook may have been difficult to harvest even with gill nets. The Yurok began to fish commercially at the mouth of the Klamath in 1876. Only Indians were allowed to fish and the first pack for the new canneries in the lower river was in 1881 (McEvoy 1986).

Gold mining in the Klamath Basin dwindled at the turn of the century due to decreased profits. As habitat began to recover, the fall chinook in the river started to rebound. The runs rebuilt to a peak in abundance around 1912, as indicated by the cannery pack (Snyder 1931). The Yurok began to modernize and increase their fishing efforts about 1915 and continued to do so until 1928 (McEvoy 1986).

Commercial gill net harvest in the Sacramento River was greatly reduced in the 1880's as a result of pressure from sport fishermen (McEvoy 1986). With the resurgence of salmon populations in both the Sacramento and the Klamath Rivers, the ocean troll fishery grew. Trolling efforts were fairly primitive, at first involving sailboats in the Monterey and San Francisco Bay areas. By 1915 boats with motors were in use, and both catch and effort were rising sharply (McEvoy 1986). Snyder and Schofield (1924) tagged salmon from the Klamath and noted that they were being caught as far south as Monterey. The combined efficiency of the new troll fishery, which by 1920 covered the entire coast, and the modern gill net fishery proved too much for the salmon. Snyder's observations were correct. Klamath stocks reached an extreme low in the early 1930's. The canneries on the river were ordered closed in 1933, and commercial fishing in the river was outlawed (Moffett and Smith 1950).

After Snyder's work, little information about Klamath River run sizes was collected. The California salmon troll fishery had declining catches through the 1930's reaching a record low in 1938 (McEvoy 1986). After World War II, the ocean salmon fishery rebounded strongly. Runs in the Klamath during the postwar period probably reflected this general trend. In 1955, alone, the sport catch on the river was estimated to be 95,000 chinook and 100,000 steelhead (Coots 1967).

Timber harvest activities were greatly increased after World War II. Disturbances associated with logging and the 1955 flood caused substantial damage to salmon and steelhead habitat. The flood and the poor ocean conditions (El Nino) in 1956-57 resulted in a downturn in salmon spawning escapement. The 1964 flood was a catastrophic event which caused major habitat losses throughout the Klamath River Basin. Entire watersheds turned into debris flows in some areas of the basin (MacCleery 1979). From 1964 to 1984, the river's anadromous fish declined further. The habitat loss above Trinity and Iron Gate dams, the reduced flows in the Trinity, lingering effects from the 1964 flood, further habitat degradation, continued fishing pressure, and natural cycles like El Nino and the 1976-77 drought drove the river's stocks to new lows.

From 1985 to 1988, salmon runs in the Klamath and Trinity Rivers rebounded, with particularly large returns to the Trinity River and Iron Gate hatcheries. Evidence suggests that many of the native stock groups of salmon, steelhead, and other anadromous fishes of the basin may not have experienced increases similar to the hatchery stocks of chinook and coho salmon. As in Snyder's day, opinions vary as to whether stocks in the river are building up or in further decline.

STOCK IDENTIFICATION

Ricker (1972) defined a stock as "the fish spawning in a particular lake or stream (or portion thereof) at a particular season, which ... to a substantial degree do not interbreed with any group spawning in a different place or in the same place at a different time." Through evolutionary time stocks adapt through natural selection to home streams and the wider environment experienced throughout their life history (Helle 1981). While some information has been gathered on chinook salmon (Oncorhynchus tshawytscha) and steelhead (Oncorhynchus mykiss) stocks since the Klamath River Basin Fisheries Resources Plan (CH2M Hill 1985), no attempt has been made to better understand the Basin's coho salmon (Oncorhynchus kisutch), coastal cutthroat trout (Oncorhynchus clarkii), green sturgeon (Acipenser medirostris), American shad (Alosa sapidissima), eulachon, or candlefish, (Thaleichthys pacificus), or Pacific lamprey (Lampetra tridentata) population groups.

Stock identification can be determined by using genetic information analyzed by a laboratory technique known as "gel electrophoresis" (Ryman and Utter 1986). Genetic changes are representative of the length of time that populations have evolved separately. The time it takes for genetic change or mutation seems to be fairly fixed at about 10,000 years for each easily detectable change (Wilson and Sarich 1966). Where stocks have been separated for a short period on an evolutionary time scale, different behaviors and physiological features necessary for survival, the animal's "phenotype," may change faster than its genetic structure, or "genotype." Electrophoresis is, therefore, actually a more appropriate tool for defining regional stocks which have been isolated from one another for longer periods, as opposed to finer stock-group distinction within basins (Eric Loudenslager personal communication).

No genetic basis for some traits, such as fall- and spring-run timing in chinook or steelhead stocks within the same basin, can be found (Riesenbechler and Phelps 1989). Varying physiological or behavioral characters may be better indicators of stocks within the Klamath Basin. Nicholas and Hankin (1988a) used the season-of-return to native stream, spawning date, age at maturity, ocean migration pattern, number and size of eggs, resistance to disease, and juvenile life history as characters with which to define stocks of chinook salmon of the Oregon coast.

Fishery managers tend to think of stocks in the broadest sense, such as "fall chinook" or "spring chinook." Using Ricker's definition, however, numerous stock groups tuned to the tributaries and geographic regions of the Klamath River seem to be present. The "stock concept" that recognizes that salmon and steelhead are divided into discrete subpopulations has wide acceptance in fisheries science (Berst and Simon 1981). Similar stock groups have been identified by Saunders (1981) for Atlantic salmon, including several stocks within one river system. Parkinson (1984) found distinct genetic strains of steelhead in all the British Columbia streams that he studied. His work suggests that the steelhead stocks that he studied had colonized a wide area as glaciers retreated in relatively recent geologic time. While the stocks he studied were very similar in overall genetic makeup, differences had evolved in local populations, even between adjacent streams.

Utter (1981) credited the evolution of genetic, morphological, and behavioral differences to reproductive isolation allowed by homing to natal streams. Recent work by Nicholas and Hankin (1988a) found distinct populations of chinook in every coastal drainage in Oregon, with some streams harboring several stocks.

Problems with the Current Concept of "Natural" Spawners

The current convention for both the Klamath River and Trinity River restoration programs is to call fish spawning outside the hatchery environment "natural" spawners. Tuss (USFWS 1988a) pointed out that surplus hatchery fish, straying up streams near the hatchery, or spawning in the main river below, contributed substantially to "natural" escapement. Recent investigations suggest that there can be substantial differences in growth and survival of offspring of native or locally adapted salmon and steelhead compared to those of hatchery fish spawned in the wild (Riesenbichler and McIntyre 1977, Altukhov and Salmenova 1986, Chilcote et al. 1986, Solazzi et al. 1983).

The use of the term "natural" to include both groups of fish obscures these differences and can mask whether the goal of preserving viable native populations is being met (USFWS 1988a). As an example, studies of chinook salmon spawning above Junction City in the Trinity River indicated that 60 percent were first-generation Trinity River Hatchery fish in 1987 (Stempel 1988). This high degree of straying would overwhelm any genetic difference between hatchery stocks and other salmon present, yet these fish make up the majority of "natural" chinook salmon spawning in the Trinity River in this area. McIntyre et al. (1988) used a more restrictive definition of "natural fish" as those "produced by natural spawning, but with at least one parent of hatchery origin."

Many areas in the Klamath River Basin still have discrete groups of salmon and steelhead that are not of hatchery origin. These stocks may have been returning to the Klamath Basin for millions of years. In Oregon's Natural Production and Wild Fish Management Rules (Chilcote 1990), wild fish are defined as "any naturally spawning fish belonging to indigenous populations." Indigenous fish were those descended from ancestral populations which had spawned in the same geographic area prior to 1800, which excludes fish populations established by man. The term "native" will be used here when referring to the self-replicating populations that return to various tributaries and at various times that do not coincide with the range or timing of hatchery stocks. If this use of native were adopted, "natural" spawners might be those fish with run timing and distribution similar to hatchery fish.

Various salmon and steelhead stocks from outside the Klamath Basin have been imported and planted in basin tributaries. Fish from the large hatcheries within the basin have also been transplanted widely. Stock transfers of salmon and steelhead, or straying, do not necessarily change the genetic structure of locally adapted populations, however. If the introduced fish do not have critically important survival adaptations to the local environment, none of their offspring will survive, thereby preventing "gene flow" from occurring (Riggs 1990). Further, a few strays per generation will not cause appreciable genetic change, although large numbers of strays can change a local population. Genetic purity of stocks may not ultimately be the issue, however. If stocks remain self-perpetuating in various streams of the basin, they are adapted to local stream conditions. The may prove to be essential building blocks for restoring runs either through artificial culture or for recolonization after habitat restoration.

The current fall chinook population in Bluff Creek was established from Iron Gate Hatchery fish. Similar populations have been established in all tributaries from pond rearing programs (see Chapter 5). Whether these transplanted fish will be self sustaining without continuing pond rearing programs is unknown.

Use Of Stock Groups For Recognition and Protection of Populations

McIntyre (1983) suggested the use of "management units" for salmon management that might represent from one to several stocks. He offered this option for stock groups in deference to the fact that management of all creeks on an individual basis, although ideal from a stock conservation and genetic preservation perspective, was not possible due to costs and logistics. Some fall chinook salmon stocks have been accepted de facto in management, such as those fish returning to the Shasta, Scott, and Salmon rivers and the South Fork of the Trinity. These populations have been monitored with weirs.

Detailed identification of stock diversity of anadromous fish in the Klamath Basin has yet to be attempted. What is offered below is a conservative approach using "stock groups" parallel to the concept of management units used by McIntyre (1983). These stock groups also meet Ricker's definition of run timing and destination and, where electrophoretic information and those characters used by Nicholas and Hankin (1988a) are available, they are used, as well. A complete listing is found in Table 4-1. The boundaries may seem arbitrary when one splits stock designation for fall chinook in small streams immediately upstream and downstream from Weitchpec, for example. If one considers geographic centers of these group boundaries, such as Blue Creek and Clear Creek, the differences can be more demonstrable. Snyder (1931) noted differences in run timing and body shape between these two stock groups calling the former "Blue Creekers" and the latter "hookbills."

The stock groups should be thought of as locally adapted subpopulations that may have evolved appropriate characteristics to survive in different regions of the Klamath Basin. Factors such as climate and geology vary widely over the basin, giving rise to varied fish habitat conditions. Adaptations to regional stream flows, water temperatures, stream gradients, as well as to the disease organisms present, may be captured in the genetic information that different runs possess. The stock groups proposed here cover wide areas. It is possible that considerable diversity, worthy of preserving, may be found on a smaller geographic scale between streams within these areas. A similar recognition of stocks is emerging from the Columbia River Basin Salmon and Steelhead Production Plan (Riggs 1990): "Because natural populations of salmon and steelhead have evolved somewhat independently in response to environmental conditions in different parts of a varied ecosystem like the Columbia River Basin, each population may represent an efficient production unit for its historic location and a potentially valuable resource for other similar locations." In implementing gene conservation for the Columbia Basin program, Riggs suggests that "stock assessment is fundamental to the process, but must not become an obstacle to the use of best available information" for planning and program implementation.

FALL CHINOOK

The evidence suggests that fall chinook stock groups in the Klamath River include those fish returning to: 1) Iron Gate Hatchery, 2) Bogus Creek, 3) the Shasta River, 4) the Scott River, 5) the Salmon River, in addition to the distinctly late runs to 6) the middle Klamath tributaries below Iron Gate Dam, and 7) the lower Klamath River tributaries below Weitchpec.

Electrophoretic Studies

Milner et al. (USFWS unpublished report), as by the National Council on Gene Resources (1982), found that genetic differences between Trinity and Klamath chinook were greater than the differences between four Sacramento River stocks tested. The differences between Klamath and Trinity River chinook reflect the fact that these populations have evolved separately for some time. The similarity of Sacramento tributary stocks may be the result of the continuing stock transfers between subbasins there. Recent electrophoretic analysis of ocean troll catches have defined differences between the Klamath stock complex of fall chinook, those of other California coastal systems, Central Valley stocks, and those of southern Oregon (Gall et al. 1989).

Chinook salmon (Oncorhynchus tshawytscha)

More detailed work was conducted within the basin by Gall et al. (1989) as background. Samples were taken from Camp Creek, Bogus Creek, Horse Linto Creek, the Iron Gate Hatchery, the South Fork of the Trinity River, the Trinity River Hatchery, and rearing ponds holding "late run" fall chinook from the lower Klamath. The purpose of the study was not to determine the genetic relationships of chinook stocks within the Klamath Basin, but rather to distinguish basin stocks from others in the mixed-stock ocean harvest. All samples, however, showed some genetic differences from one another. Those closest geographically showed the greatest genetic similarity, although the differences were not statistically significant (Devon Bartley personal communication).

Life History Studies

Sullivan (unpublished) collected scales from adult fall chinook salmon captured at weirs on the South Fork of the Trinity River, Salmon River, Scott River, Shasta River, and Bogus Creek. The patterns of the innermost areas of the scales were analyzed to determine the early life history of each fish. He found that three life histories exist for fall chinook:

1. Type I,in which outmigration begins immediately and juveniles entered the ocean in the spring.

2. Type II,which spends the spring and summer in the river or estuary and enters theocean in the fall.

3. Type III, which occurs only rarely, in which chinook juveniles spend an entire year in freshwater, entering the ocean as yearlings in spring.

Sullivan concluded that "major differences of relative frequencies of life history types were apparent between different tributaries studied." He found high frequencies of Type II life histories in the Scott and Salmon drainages. The South Fork of the Trinity and Shasta River fall chinook showed a higher incidence of Type I patterns. These differences may reflect a difference in genetic structure, but they may also be behavioral responses to environmental conditions. Do more Type I fish in the Shasta and South Fork Trinity simply reflect the fact that most juvenile chinook that remain in these streams fail to survive? Are Type II and Type III fish still present in these two stock groups and will they be reexpressed if habitat conditions improve? Life history patterns are used as partial criteria for stock group identification here, but further study is needed.

Most adults returning to spawn in upper Klamath tributaries and at Iron Gate Hatchery enter the river early (USFWS 1982, Hubbell et al. 1979). Migration peaks in the last week of August or toward the beginning of the first week in September. The time of entry into the Klamath for the various stock groups and the time of entry into their home streams follow characteristic patterns which may vary somewhat with river conditions. Rates of upstream migration may be effected by water temperatures, for instance, in the main stem of the Klamath. The following describe the fall chinook population groups.

Iron Gate

The hatchery stock may represent upper basin stocks that once returned to the Upper Klamath and its tributaries above Iron Gate Dam (Fortune et al. 1966). These fish arrive at the hatchery beginning in the third week in September, peak in abundance toward mid-October and have all arrived by the second week in November. Their average fecundity is about 3,100 eggs per female.

Bogus Creek

While straying has increased from Iron Gate Hatchery into Bogus Creek in recent years (Randy Baxter personal communication), Gall et al. (1989) still found genetic difference between Bogus stocks and those of the Iron Gate Hatchery. Mills et al. (unpublished) has found that the outmigration of juveniles begins in mid-February and continues through early June. Sullivan (unpublished) found that three-year-old Bogus Creek chinook returned to spawn at a smaller size than three-year-old Shasta, Scott, or Salmon River fish.

Shasta River

Department of Fish and Game operations at the Shasta Racks show that fall chinook enter the Shasta River from mid-September to mid-October. Snyder (1931) reported that spawning activity on the Shasta peaked in mid-October. CDFG reports from the operation of the racks suggest little straying from Iron Gate Hatchery, indicating a strong likelihood of the continuing genetic integrity of this stock group. Mills et al. (unpublished) found only early outmigration of juvenile chinooks, beginning in early January and complete by the end of April.

Scott River

Weir operation by CDFG (Hubbell, et al. 1985) on the Scott indicated a peak in spawning run near the end of October. Again, incidences of straying are low, indicating

little intermixing with Iron Gate Hatchery stocks. Sullivan (unpublished) found predominantly Type II life histories in the fall chinook returning to the Scott.

Salmon River

This major Klamath tributary has adult fall chinook returning as soon as early September. Large adults have also been seen spawning as late as January (J. West personal communication), which may represent a second fall chinook run in this system. Early life histories of Salmon River fall chinook were also predominantly Type II (Sullivan unpublished).

Middle Klamath Tributaries

Snyder (1931) described a late run of fall chinook for the area above the Trinity River's confluence with the Klamath, calling them "hookbills." He said that spawning took place between November and January. Leidy and Leidy (1984) also described a run of fall chinook in this region with this late timing. Current efforts by the Karuk Tribe to trap late fall chinook for breeding are directed at this stock group.

Lower Klamath Tributaries

Snyder (1931) noted that larger fish showed up at the mouth of the Klamath beginning in October and entered the lower river tributaries to spawn. Recent observations have noted spawning as late as January by this stock group (USFWS 1990c). The Indian fishermen called these fish "Blue Creekers." Snyder (1931) found them to be very similar to Smith River fish in body size, shape, and coloration. Gall et al. (1989) found these fish to be more similar genetically to Smith River or southern Oregon stocks than to other Klamath groups. USFWS (1990b) found that juvenile chinook outmigration extended from April at least through July (sampling ended in July) with peaks in mid-April and mid-June. Some yearling (Type III) chinook juveniles have been found in the lower Klamath tributaries (USFWS 1990a). Yurok Tribe enhancement projects are attempting to increase runs of these "Blue Creekers."

SPRING CHINOOK

The runs of spring salmon in the Klamath Basin were very important historically, outnumbering fall chinook stocks substantially (Hume in Snyder 1931). Snyder (1931) described a spring run that began in late March and continued through mid-June, followed by a summer run. Some spring chinook have returned as early as February, even in recent years (USFWS 1990d). Moffett and Smith (1950) described two distinct peaks at Lewiston, on the Trinity River, in spring chinook migrations prior to dam and hatchery construction. One run was most abundant in June, while the second peaked in August. Today's runs are supported in large part by the Trinity River Hatchery, which was founded on these ancestral stocks. These stocks return to the mouth of the Klamath River beginning in April and continue entering the river into July.

TABLE 4-1 -- Tentative stock groups of Anadromous Fishes on the Klamath River Basin.

FALL CHINOOK

Upper Klamath (Iron Gate Hatchery)

Bogus Creek

Shasta River

Scott River

Salmon River

Middle Klamath tributaries (from Weitchpec to Iron Gate Dam)*

Lower Klamath tributaries (below Weitchpec)*

Trinity River Hatchery/Upper Trinity (above Junction City)

South Fork Trinity

North Fork Trinity

Middle Trinity tributaries (from South Fork to Junction City)

Lower Trinity tributaries (South Fork to Weitchpec)

SPRING CHINOOK

Upper Trinity/Trinity River Hatchery

South Fork Trinity

North Fork Trinity

New River

Salmon River

Wooley Creek

Elk Creek

Clear Creek

Dillon Creek

COHO

Iron Gate Hatchery

Trinity River Hatchery

Lower Klamath tributaries

Scott River

Shasta River (?)

Salmon River

Middle Klamath tributaries

Lower Trinity tributaries (?)

SUMMER STEELHEAD

New River

South Fork Trinity River

North Fork Trinity River

Canyon Creek

Bluff Creek

Salmon River

Wooley Creek

Elk Creek

Dillon Creek

Red Cap Creek

Clear Creek

Indian Creek

FALL/WINTER STEELHEAD (from Leidy and Leidy 1984, in part)

Upper Klamath (Iron Gate Hatchery)

Upper Trinity (Trinity River Hatchery)

Shasta River

Scott River

Salmon River

Middle Klamath tributaries

Lower Klamath tributaries

Lower Trinity tributaries (Weitchpec to North Fork)

Upper Trinity tributaries (North Fork to Lewiston Dam)

New River

North Fork Trinity River

South Fork Trinity River

CUTTHROAT TROUT: Lower Klamath tributaries

GREEN STURGEON: Unknown

PACIFIC LAMPREY: Unknown

EULACHON: Unknown

AMERICAN SHAD: East Coast in origin

* The stock boundaries used here are the same as used to define the basin areas in this Plan except for the Lower and Middle Klamath tributary fall chinook stocks, due to the information from Snyder (1931) and Gall et al. (1989).

A few dozen spring chinook were still returning to the upper Klamath at the time that Iron Gate Hatchery was begun, 25 years ago (Curt Hiser personal communication). Fortune et al. (1966) described upper Klamath spring chinook stocks as having special abilities to migrate and home through Klamath Lake. From 1962 to 1968 the return of this distinct run of fish went unrecognized. Efforts to maintain these runs were begun in 1968, but were not successful and this stock group was lost (CH2M Hill 1985).

The Salmon River and its Wooley Creek tributary support what may be the last viable native spring chinook salmon population in the Klamath Basin. Streams that support summer steelhead in the mid-Klamath, such as Indian Creek, Elk Creek, and Clear Creek, have small, highly variable populations of spring chinook salmon.

Trinity River tributaries such as the North Fork, New River, the South Fork, and Canyon Creek also have runs of spring chinook. Canyon Creek is not included in the stock groups listed in Table 4-1 because it is suspected that its run is made up largely of hatchery strays. Salmon River stocks seem to enter this major tributary from mid-April to early June, but run timing may vary with river temperature and flows.

COHO SALMON

Snyder (1931) reported significant coho salmon runs, particularly in the lower Klamath Basin tributaries. He noted a migration of coho to the Klamathon Racks on the upper Klamath in the 1920's, although they were never used for broodstock at the Fall Creek Hatchery. More recently, Harry (1966) described coho salmon populations in the Shasta, Scott, and Salmon Rivers and some coho have been counted at weirs in these systems in recent years (CDFG unpublished). Coho once returned to the Stuart's Fork of the upper Trinity River (USFWS 1979) and native coho were trapped at a weir to establish a broodstock just prior to the completion of the Trinity River Hatchery (Bedell 1968). Moffett and Smith (1950) noted that coho spawned in smaller tributaries below the South Fork on the Trinity River.

Coho salmon (Oncorhynchus kisutch)

Hoopa Fisheries Department surveys (1988) note the incidence of adult and juvenile coho salmon in the Trinity River, but whether viable reproducing coho salmon populations still exist on the Reservation remains questionable. The question of whether native coho stocks remain in this area is somewhat clouded because of releases of Trinity River Hatchery coho in 1981-82 (Mike Orcutt personal communication). Native coho are still present in the Klamath tributaries below Weitchpec. Unpublished CDFG field reports note the presence of coho in Hunter Creek and Terwer Creek. Small numbers of coho juveniles are found in downstream migrant traps operated by USFWS (1990a) in creeks in this area. Native coho migration and spawning is later than hatchery populations, with adults captured in the lower river in November and December (R. Pierce personal communication). In some years coho at the trapping station in Camp Creek outnumber the returning chinook salmon (Leaf Hillman personal communication).

The hatchery runs of coho for both Iron Gate and Trinity River hatcheries were created from broodstock from the Cascade Hatchery in the Columbia River Basin. This stock returns to the lower river in September and October, with the peak generally occurring in the second week of October (Hubbell 1979). Coho yearlings from Iron Gate Hatchery were transplanted to Indian Creek, Beaver Creek, and Elk Creek between 1985 and 1989 and have resulted in at least some spawning activity in Indian Creek (Dennis Maria personal communication).

STEELHEAD

Although steelhead are very important to the economy of the Klamath Basin, little is known about their stock groups. Distinguishing between steelhead stock groups by time of return to the river becomes very problematic (Roelofs 1983). Everest (1973) found that steelhead entering in early fall spawned with earlier returning summer steelhead in the Rogue River. Similarly, fall fish may sometimes wait until after the rains to move into their tributaries and spawning could overlap with early winter steelhead. "Half pounders," small, sexually immature steelhead that have spent less than one year in the ocean, may be the

Steelhead (Oncorhynchus mykiss)

offspring of fall, summer or winter steelhead stocks (Everest 1973). Half pounders run only in the Klamath, Rogue, Eel, and Mad Rivers (Barnhardt 1986). Ceratomyxa shasta, a deadly protozoan fish disease is present in the upper Klamath. Buchanan (in press) has found native trout in the Klamath above Iron Gate Dam to be resistant to this disease. Steelhead in the Middle and Upper Klamath would also be exposed to high levels of C. shasta and have evolved a resistance.

The only attempts to identify stocks of steelhead in the Klamath Basin using electrophoresis were conducted on the South Fork of the Trinity River. The study compared South Fork stock groups with those of the upper mainstem of the Trinity and found significant difference between stocks in the two streams (Baker 1988). Lesser differences were noted among steelhead juveniles in South Fork tributaries, but Baker pointed out that the diversity might be indicative of important local adaptations to environmental conditions.

Both Iron Gate and Trinity River hatcheries release steelhead that return in fall and winter. Trinity River Hatchery steelhead broodstock included stock imported from the Eel River, three Oregon hatcheries, and Washington hatchery Skamania steelhead (CH2M Hill 1985). Iron Gate Hatchery steelhead stocks were founded from native fish but some steelhead eggs from Trinity River Hatchery and the Cowlitz River Hatchery in Washington were imported (CH2M Hill 1985). Recently, large numbers of Iron Gate Hatchery steelhead have been transferred to Trinity River Hatchery (Bedell 1984, 1985). Studies by Satterthwaite (1988) indicated that half-pounders from both hatcheries were present in the Rogue River. This indicates that a native component remains in both hatchery broodstocks as pure non-native stocks would probably not exhibit the half-pounder life history.