historical decline and current status of coho salmon in california

N 0 R T H A M E R I C A N

JOURNALFFISHERIESWAGEMENT

Volume 14 May 1994 Number 2

North Americann Journal of Fisheries Management 14237-26 1, 19940 Copyright by the American Fisheries Society 1994

Historical Decline and Current Status ofCoho Salmon in California

LARRY R. BROWN,' PETER B. MOYLE,2 AND RONALD M. YOSHIYAMA

Department of Wildlife and Fisheries BiologyUniversity of California, Davis, California 95616, USA

Abstract. -The southernmost populations of coho salmon Oncorhynchus kisutch occur in Cal-ifornia where native coho stocks have declined or disappeared from all streams in which they werehistorically recorded. Coho salmon previously occurred in as many as 582 streams, from the SmithRiver near the Oregon border to the San Lorenzo River on the central coast. Information on therecent presence or absence of coho salmon was available for only 248 (43%) of those streams. Ofthese 248 streams, 54% still contained coho salmon and 46% did not. The farther south a streamis located, the more likely it is to have lost its coho salmon population. We estimate that the totalnumber of adult coho salmon entering California streams in 1987-l991 averaged around 31,000fish per year, with hatchery populations making up 57% of this total. Thus, about 13,000 non-hatchery coho salmon have been spawning in California streams each year since 1987, an estimatethat includes naturalized stocks containing about 9,000 fish of recent hatchery ancestry. There arenow probably less than 5,000 native coho salmon (with no known hatchery ancestry) spawning inCalifornia each year, many of them in populations of less than 100 individuals. Coho populationstoday are probably less than 6% of what they were in the 1940s and there has been at least a 70%decline since the 1960s. There is every reason to believe that California coho populations, includinghatchery stocks, will continue to decline. The reasons for the decline of coho salmon in Californiainclude: stream alterations brought about by poor land-use practices (especially those related tologging and urbanization) and by the effects of periodic floods and drought, the breakdown ofgenetic integrity of native stocks, introduced diseases, overharvest, and climatic change. We believethat coho salmon in California qualify for listing as a threatened species under state law, andcertain populations may qualify for listing as threatened or endangered under federal law.

Populations of anadromous salmonids in thePacific Northwest, including California, have gen-erally declined in recent years, as indicated by de-creased commercial and sport harvests (Lufkin1991; Nehlsen et al. 1991). Coho salmon Onco-

1 Present address: U.S. Geological Survey, 2800 Cot-tage Way, Sacramento, California 95825, USA.

2 Author to whom correspondence should be ad-dressed.

rhynchus kisutch in particular have shown sub-stantial coastwide decreases (Konkel and Mc-Intyre 1987). In the U.S. Pacific Northwest, cohosalmon are now extinct in the eastern half of theirformer range, and badly depleted over most of thewestern half (Frissell 1993).

Although less abundant than chinook salmon0. tshawytscha in California, coho salmon havebeen an important component in both commercialand sport fisheries. In the 198Os, California’s com-bined commercial and sport catch averaged 8 3,000

237

MC Kier

scanned for KRIS

238 BROWN ET AL.

TABLE 1 . -Commercial landings of chinook and cohosalmon in California. Data are from the National MarineFisheries Service (NMFS 1977-l993).

Chinook salmon

TotalLandings ex-vessel

Year (pounds) value

1992 1 . 6 m i l l i o n $ 4 . 4 m i l l i o n1991 3 . 2 m i l l i o n $ 8 . 3 m i l l i o n1990 4.1 million $11.4 million1989 5.4 million $12.9 million1988 14.4 million $41.2 million1987 9.0 million $25.1million1986 7.2 million $21 .O million1985 4 . 5 m i l l i o n $11.6 million1984 2 . 6 m i l l i o n $ 7 . 3 m i l l i o n1983 2 . 1 m i l l i o n $ 4 . 4 m i l l i o n1982 7.4 million $19.0 million1981 5 . 5 m i l l i o n1980 5 . 6 m i l l i o n1979 7 . 6 m i l l i o n1978 5 . 3 m i l l i o n1977 4 . 3 m i l l i o n1976 4 . 2 m i l l i o n

Coho salmon

Tota lLandings ex-vessel(pounds) value

11,000 $18,000460,000 $701,000311,000 $617,000229,000 $320,000320,000 $707,000246,000 $263,000195,000 $101,000

8 1,000 $ 128,000346,000 $370,000266,000 $327,000545,000 $790,000482,0003 0 0 , 0 0 0

1,200,0001,300,0002 3 5 , 0 0 0

3,600,OOO

coho salmon annually, of which 30,200 were inthe sport fishery (Sheehan 1991). Commerciallandings of coho salmon in California averaged301,000 lb annually during the 1980s and 1.6 mil-lion pounds annually over the period 1976-l979(Table 1); the harvest in 1992 (11,000 lb) wassubstantially lower than in previous years. Asidefrom their economic importance, California cohosalmon stocks are of ecological and evolutionarysignificance because they are the southernmostpopulations of the species. These stocks undoubt-edly have at least some genetic distinctiveness andlocal adaptation to southern environmental con-ditions. Hence, they may be important reposito-ries of genetic variation that could promote thespecies’ expansion if large-scale climatic warmingof the Pacific Northwest region occurs. The po-tential economic significance of these southernstocks to future fisheries in more northern areascould be substantial.

There is general agreement among fisheries bi-ologists familiar with coho salmon in Californiathat native coho salmon stocks have declined sig-nificantly throughout the state in recent years(Moyle et al. 1989). The exact extent of that declineis unknown, in part because the species is dividedinto many small populations, few of which aremonitored closely. Nehlsen et al. (1991) listed 214naturally spawning native stocks of anadromoussalmonids that are declining. In that listing, Cal-ifornia coho salmon populations south of San

Francisco Bay were considered to be at high riskof extinction, and populations north of San Fran-cisco Bay were at moderate risk of extinction, ex-cept for populations in the Klamath River, whichwere classified as “of special concern” (decliningbut in no immediate danger).

Historical estimates of statewide coho salmonabundance are essentially guesses made by fish-eries managers, based on limited catch statistics,hatchery records, and personal observations. Inthe 1940s there were apparently between 200,000and 500,000 coho salmon spawning in the state(E. Gerstung, A~P.).~ The number decreased toabout 100,000 fish in the 1960s (California Ad-visory Committee on Salmon and Steelhead Trout1988) with 40,000 in the Eel River alone (U.S.Heritage Conservation and Recreation Service1980, App.). The statewide total of natural spawn-ing coho salmon during 1984-l985 was estimatedat 30,480 fish, compared with a total of about3,545,OOO natural spawners for the North Amer-ican Pacific Coast (Wahle and Pearson 1987). Un-fortunately, there is no way to test the reliabilityof the earlier estimates, and they are best regardedas accurate only within an order of magnitude.

This paper summarizes the published and un-published information on the distribution andabundance of coho salmon in California. Thissummary is the first attempt to assess the currentstatewide population status of coho salmon bymeans of a systematic evaluation of available in-formation from a variety of sources. Our intent isto establish a benchmark population estimate thatcan be compared with future population assess-ments and to emphasize that much better infor-mation is needed to guide management throughthe next century. We also discuss the probablecauses of the decline of coho salmon populationsin California.

Background Information and Methods

Life HistoryThe life history of the coho salmon in California

has been well documented by Shapavalov and Taft(1954), Hassler (1987), and Moyle et al. (1989). Acomprehensive account of coho salmon biologythroughout the species’ range is given by Sander-cock (199 1), and ocean-related aspects are coveredby Pearcy (1992). Coho salmon return to theirparent streams to spawn after spending 18 months

3 Citations followed by “App.” are given in the Ap-pendix.

,CALIFORNIA COHO SALMON STATUS 239

or more in the ocean (up to 3 years in Alaska).Jack males may, however, return after one growingseason in the ocean (at age 2). In general, the timeof freshwater spawning migration occurs progres-sively later in the year for more southern cohosalmon populations, although there are many ex-ceptions to this pattern (Sandercock 1991). In Cal-ifornia, the migrations normally occur from Oc-tober to March, peaking in November-January.Many of the small coastal streams in Californiaare blocked by sand bars at their mouths, and thecoho salmon cannot ascend until the sand barriersare breached by high streamflows that follow heavyautumn rains. Coho salmon use a variety ofspawning sites, but they characteristically entersmall coastal creeks or tributary headwaters oflarger rivers.

Optimal habitat for juveniles seems to be deeppools (2 1 m) containing logs, rootwads, or boul-ders in heavily shaded sections of stream. Thesehabitat characteristics are typical of streams in old-growth forests, and for that reason, the decline ofcoho salmon stocks in California can be tied tothe widespread elimination of old-growth foreston the California north coast. Juvenile coho salm-on in California typically live in the streams forabout 1 year. The juveniles begin to migrate down-stream to the ocean during late March and earlyApril, and outmigration usually peaks in mid-May,if conditions are favorable. The immature salmoninitially remain in inshore ocean waters close tothe parent stream. They gradually move north-ward, staying over the continental shelf. Cohosalmon range widely in the North Pacific, but themovements of California fish are poorly known.

DistributionCoho salmon spawning populations are distrib-

uted on both the Asian and North American coastsof the North Pacific, ranging southward to the Seaof Japan in Asia and into California in NorthAmerica (Sandercock 1991). Within California,coho salmon spawn in streams from the Oregonborder southward to Monterey Bay. The south-ernmost recorded spawning stream is the San Lor-enzo River, Santa Cruz County (Snyder 1908) butcoho salmon probably occurred in smaller streamsflowing into Monterey Bay and perhaps as far southas the Big Sur River. Presently, the most southernnaturally spawning populations are in Scott andWaddell Creeks, several kilometers north of theSan Lorenzo River.

Most of the coho salmon caught in the oceanfisheries of California originate in Oregon. Colum-

bia River fish apparently constitute the largestcomponent of the California ocean catch, andnorthern California coho salmon contribute onlyabout 10% (Baker and Reynolds 1986, App.). Con-versely, few fish from California are harvested outof state. Tagging studies indicated that only 6-7%of California native coho salmon stocks and 20%of nonnative stocks (imported from Oregon andWashington and released at California hatcheries)were caught in Oregon and Washington (Jensen1971). On the Oregon coast, 75% of the coho salm-on caught during 1977 had been released fromhatcheries (Scamecchia and Wagner 1980). In Cal-ifornia, the percentage of fish caught that wereproduced in hatcheries may be even higher, giventhe present low productivity of natural popula-tions.

MethodsOur approach in this survey of California coho

salmon stocks was simply to obtain as much in-formation as possible through searches of the pub-lished literature, file reports of fisheries agencies,and from mail or telephone interviews with per-sons involved in coho salmon research and man-agement in the state. Because of the extreme pau-city of published data on native coho salmonpopulations, we relied heavily on unpublished re-ports and personal communications; these sources(identified by “App.” in their citations) are listedafter the references in the Appendix. We soughtdata on the historical distribution of coho salmonpopulations, as well as more recent information,in order to evaluate the number of those popu-lations that still exist. Recent data, if they existed,were for 1987 or later for almost all streams.Streams for which there were several years of re-cent data were classified as having coho salmoneven if coho salmon were absent in some years.For many of the streams for which current dataon coho salmon were lacking, we asked surveyrespondents to give their opinions as to whetherstreams with which they were most familiar stillcontained coho salmon populations as of 1990-1991. The data used in this paper are compiled inBrown and Moyle (1991 a, App.).

Because hatchery-raised coho salmon constitutea significant portion of the population in somestreams, we classify coho salmon populations asthree stock types: (1) native stocks that have fewor no hatchery-raised fish in their ancestry; (2)naturalized stocks that included a large proportionof hatchery fish at one time, but are the progenyof naturally spawning fish; and (3) hatchery stocks

BROWN ET AL.

180

160

FIGURE 1 .-Weight of coho salmon caught in the California commercial troll fishery (open squares; data fromCalifornia Department of Fish and Game, unpublished records) and pounds of juvenile coho salmon planted eachyear (solid triangles; data from published hatchery reports).

that include large numbers of hatchery fish everyyear, usually of nonnative origin, and show littleevidence of successful natural reproduction. Many,if not most, of the streams containing stocks thatwe classify as naturalized may also contain sub-stantial numbers of hatchery strays, as indicatedby the greater abundance of coho salmon in streamsnear hatcheries. For example, the production ofcoho salmon in Mendocino County centers aroundthe Noyo River, which is stocked with hatchery-raised fish, and the number of coho salmon spawn-ing in county streams declines both to the northand south of the Noyo River (W. Jones, App.).Thus, runs in these streams may be less self-sup-porting than the limited data indicate.

In attempting to estimate numbers of cohosalmon for individual streams, we used proceduresthat most likely overestimate abundance to avoidexaggerating the extent of population depletion.For our enumeration, we assumed that each streamthat historically contained coho salmon or forwhich there were no data had a basal populationof 20 spawners (our “20-fish rule”). For each streamwhere an estimate of adult populations was avail-able, we used either the estimate itself or 20 fish,whichever was larger. For hatchery populations,we assumed the average population size, based onavailable data starting in 198l - 1982. For streamswhere hatcheries were located, we included both

the average hatchery population and the estimatednative or naturalized population.

Status of Coho Salmon Populations

Most of the coho salmon produced in Californiawaters are harvested there; however, the Califor-nia commercial catch includes fish produced instreams and hatcheries in both California and Or-egon (Hassler 1987). Increases in hatchery pro-duction are believed to be the major factor re-sponsible for the increased catches of the 1960sand 1970s; however, the commercial troll catchof coho salmon declined drastically in the late197Os, despite continued heavy plantings of hatch-ery fish (Figure 1; Brown and Moyle 1991 a, App.).Because counts of adult returns to hatcheries eitherincreased or fluctuated nondirectionally at thattime, it is probable that the decline in catch re-flected the general decline of California wild pop-ulations, as well as decreased production by Or-egon stocks. For example, the count of native cohosalmon at Benbow Dam on the South Fork EelRiver showed a gradual but steady drop from the1940s until the mid-1970s, when no fish werecounted (Brown and Moyle 1991a, App.). In con-trast, the hatchery-supported population in theMad River fluctuated at a low level through theearly 1960s, with no evident decline (Brown andMoyle 199la, App.). A general decline of native

CALIFORNIA COHO SALMON STATUS 241

OREGON

0 100 km

F I G U R E 2. -Map of northern Cal ifornia showing major spawning streams of cohothat have produced coho salmon in large numbers are indicated by solid circles.

coho salmon populations has occurred within theentire Oregon Production Area (Washington tonorthern California) since the mid-1960s (Scar-necchia and Wagner 1980; Nickelson 1986). Fur-thermore, total coho salmon adult production (in-cluding hatchery stocks) in this region has declinedabruptly since 1970, in spite of increasingly highernumbers of released hatchery smolts (Nickelson1986).

Hatchery Populations

Except in the Eel River, coho salmon stocks inthe larger rivers of California (Figure 2) are nowdominated by hatchery production, and even sev-eral small coastal streams receive regular plants ofhatchery fish. These hatchery stocks are of diverseorigin, but all have included fish derived fromoutside the river system receiving the plantingsand often from outside California. Hatchery stockshave also been used to reestablish extirpated pop-

salmon. Locations of hatcheries

ulations or to supplement depleted runs, whichmay partly explain the overall lack of genetic dif-ferentiation among coho salmon from differentCalifornia streams (Bartley et al. 1992). In the fol-lowing account, we review the records of the majorgovernment hatcheries. In addition, several pri-vate hatchery projects have attempted to rebuildlocal remnant stocks (Miller et al. 1990). The twolargest private producers of coho salmon are theHumboldt Fish Action Council and the MontereyBay Salmon and Steelhead Project, which annuallyrelease about 25,000 and 23,000 yearlings, re-spectively (Miller et al. 1990). In 1989, privateprojects, together with county and local programs,released more than 266,000 coho salmon yearlingsinto California waters.

Klamath River. -During 1963-l 968, adult re-turns to the Iron Gate Hatchery in the headwatersof the Klamath River never exceeded 500 fish (datain Brown and Moyle 199 la, App.). Following anintensive stocking program begun in 1966 (and

242 BROWN ET AL.

continued in 1967 and 1969) with Cascade River(Oregon) fish, adult returns to the hatchery wereover 1,000 fish in seven spawning seasons andexceeded 2,200 fish twice, most recently in 1987;numbers typically have ranged between 400 and1,500 (Hiser 1991). This hatchery run, therefore,is composed basically of an imported stock.

Trinity River. -The Trinity River Hatchery suc-cessfully established a run of coho salmon thatcontinued to increase in size until recently (Bedell1991). Prior to 197 1, adult returns rarely exceeded1,000 fish, but have consistently done so sincethen. Numbers of returning fish, including jacks(males returning after 4-6 months at sea), exceed-ed 5,000 in 1973 and in 1984-1988. Returns wereover 20,000 fish in 1987 and over 10,000 in 1988,but only about 5,000 in 1989 (Bedell 1991) andless than 3,000 in 1990 and 1991 (Ramsden 1993).Like the Iron Gate stock, the Trinity River stockis primarily of nonnative origin. The first signifi-cant planting was of Eel River stock in 1964, fol-lowed by Cascade River (Oregon) stocks in 1966,1967, and 1970. Noyo River (California) stock wasplanted along with Cascade River fish in 1970,and Alsea River (Oregon) stock was planted in1970. Significant numbers of fish (about 40% ofadult escapement) apparently spawned naturallyin the Trinity River or in tributaries above theNorth Fork confluence during 1969 and 1970,mainly in the area between Lewiston Dam andDouglas City (Rogers 1973). Downstream-mi-grant coho salmon that were not of hatchery originwere captured during 1968 in the Trinity River(Healey 1973), another indication that naturalspawning was occurring. The relative contribu-tions of naturalized and hatchery stocks to currentproduction in the Trinity River are unknown.

Mad River. -The Mad River Hatchery has beenless successful than the Klamath and Trinity fa-cilities in establishing a run of coho salmon (Brownand Moyle 199 la, App.). Adult returns have fluc-tuated, but have never exceeded 2,000 fish andseldom (2 out of 18 years) exceeded 1,000 fish.The Mad River Hatchery stock has the most di-verse heritage of any in California. Planting beganin 1970 with Noyo River (California) fish, andthere were additional plantings in seven subse-quent years. Klamath River fish (derived from Or-egon stocks) were planted in 1981, 1982, 1986,and 1987. Trinity River fish (derived from non-native stocks) were planted in 197 1. AdditionalOregon stocks were imported from the Trask Riv-er (1972), Klaskanine River (1973), Soos River(1978), and Sandy River (1979). Fish from Prairie

Creek (Redwood Creek drainage, California) wereplanted in 1987 and 1989.

Russian River. -Warm Springs Hatchery on theRussian River has not established a persistent runof coho salmon since it began planting fish in 1980(Brown and Moyle 1991a, App.). Adult returnshave varied from zero to just under 1,000 fish; themost recent returns have been 170 (1989) 277(1990) and 162 (1991) (B. Cox, App.). The WarmSprings Hatchery stock is derived from the IronGate Hatchery stock (in turn derived from Cas-cade River, Oregon, stock) and from the NoyoRiver, Hollow Tree Creek, and Prairie Creek (allCalifornia) stocks.

Noyo River. -The Noyo River egg-taking sta-tion on the South Fork Noyo River began oper-ation in 1962 for the purpose of establishing asupply of California coho salmon eggs to enhanceboth depleted naturally spawning stocks andhatchery production. The number of coho salmontrapped at the Noyo station varied between 1,500and 3,000 during 1964-1976 (Brown and Moyle1991 a, App.). The numbers declined during 1977-1986, exceeding 1,500 fish only in 1981. In 1987,about 2,650 adults were trapped (Grass 1990b).However, numbers trapped in subsequent yearshave been lower: 708 (1988), 1,011 (1989) 145(1990) 509 (199 l), 164 (1992) (S. Poe, App.). De-pending on the size of the run, a number of fishare allowed to pass over the dam to spawn natu-rally-either 10 pairs or 25% of each sex, which-ever is fewer, during each week of the spawningseason (CDFG 1987, App.). Significant naturalspawning also takes place downriver of the stationin the South Fork Noyo River and in the tributaryKaas Creek (Nielsen et al. 1991, App.), but thegenetic heritage of these spawners is unknown.Since 1964, the river has been routinely plantedwith yearling fish hatched from Noyo River eggsand raised at various hatcheries. Presently, 150,000eggs are reserved annually for maintaining the NoyoRiver run; coho salmon hatched and reared at theMad River Hatchery are released as yearlings inthe Noyo River. The management goal is to main-tain a minimum run of 1,500 adults (CDFG 1987,App.). Noyo River stock has also been planted atseveral hatcheries and in several coastal streams.

Prairie Creek (tributary of Redwood Creek,Humboldt County). -Most of the following infor-mation was obtained through personal commu-nications with S. Sanders, former manager of thePrairie Creek Hatchery. The Prairie Creek Hatch-ery did not have facilities to capture returning adultfish until 1972. The run to the hatchery generally


exceeded 100 fish and increased to 1,799 fish by1988; subsequent counts were 682 fish in 1989and 186 in 1990 (Brown and Moyle 199 la, App.).However, counts for 1988 and later are not strictlycomparable with earlier counts because the use ofa physical barrier or weir to trap adults was dis-continued. Only 154 adults were collected in 1991because of permit restrictions, but there reportedlywere many more fish in the system. Most adultstrapped in the hatchery had been previously re-leased as juveniles. In the early 1970s stray cohosalmon from the Columbia River were commonlycaptured but they are now rare in Prairie Creek.Since 1983, only Prairie Creek stock has beenplanted, but some nonnative stocks were intro-duced earlier; those stocks originated from the SoosRiver (1978) and Sandy River (1979) in Oregon,and the Noyo (1982) and Klamath (1981) Riversin California (the latter stock imported from theCascade River, Oregon). Prairie Creek coho salm-on have recently tended to return in January-Feb-ruary, which is later than in earlier years (1970s)when imported stocks were heavily used. ThePrairie Creek Hatchery was closed in 1992.

Native Populations

Little data exist on the status of native cohosalmon populations in California, but the avail-able information strongly suggests that nativestocks are at very low levels. We present infor-mation on adult coho salmon abundances for somemajor drainages (Table 2) to illustrate the natureof the available data and to convey some impres-sion of historical abundances. Brown and Moyle(199la, App.) give a more complete listing ofstreams, including data on juvenile abundances.A brief commentary on coho salmon populationsin the various river systems follows.

Smith River. -The Smith River drainage (DelNorte County) does not support a large run of cohosalmon (Waldvogel 1988, App.). Recent annualcounts of adults in a 2.7-km stretch of a smalltributary creek have yielded only low numbers(Table 2), but Hallock et al. (1952) seined 60,602juveniles from the creek in 1951, indicating thatit once supported a substantial coho salmon pop-ulation.

Klamath-Trinity River. --In the Klamath River(Del Norte County), coho salmon historically wereabundant, but the commercial fishery favored theeven more abundant chinook salmon (Snyder193 1); nonetheless, some early harvests of cohosalmon were substantial (Table 2). Historical an-nual spawning escapements for the Klamath River

system have been estimated at 15,400-20,000 fish,with 8,000 coming from the Trinity River (USPWS1979, App.). Presently, Iron Gate and TrinityHatcheries are considered to be the sources of mostKlamath River coho salmon and natural produc-tion is minor (Klamath River Fishery Manage-ment Council 1991, App.).

Coho salmon have been reported from 113 trib-utary streams in the combined Klamath-TrinityRiver system (Brown and Moyle 199la, App.).Many of the lower tributaries in the Klamathdrainage have been degraded by logging and road-building, and their coho salmon runs diminished.For example, surveys in 1989 failed to find cohosalmon in Tully and Pine Creeks, and no outmi-grants were found in Pecwan Creek, although ju-veniles werefound there in previous years (T. Kis-anuki, App.). Some tributary streams in the middleand upper Klamath River still support coho salm-on, and these populations may be native becauseavailable records do not indicate stocking. Of thelarger tributary systems, the Scott River probablyholds the largest number of native fish. The Salm-on River probably has few, if any, coho salmon(J. West, App.).

In the Trinity River drainage, coho salmon havebeen reported spawning in the main stem, the SouthFork Trinity River, and in the tributaries (Hassleret al. 1991, App.). An estimated 2,098 fish spawnedin the main stem below Trinity Hatchery during1970, but all or most of them probably were hatch-ery returns (Rogers 1973). Healy (1973) capturednaturally spawned downstream-migrant yearlingsin the Trinity River, but no juvenile coho salmonwere taken in the South Fork Trinity River, whichindicates that the native stock there is greatly di-minished or gone.

Redwood Creek. -Coho salmon were first re-ported in Redwood Creek (Humboldt County) bySnyder (1908). Juvenile and adults have been seenin the main stem of Redwood Creek, in PrairieCreek (its major tributary), and in several PrairieCreek tributaries. A 1973 survey by the U.S. Bu-reau of Reclamation estimated 2,000 spawners butreported extensive habitat damage above Red-wood National Park, which was attributed to log-ging (USBLM 1973, App.). The total populationin the Redwood Creek system may still numbermore than 2,000 fish in some years, but most occurin the Prairie Creek drainage and probably origi-nate from the Prairie Creek Hatchery (D. Ander-son, App.; S. Sanders, App.).

Mad River. -On the Mad River (HumboldtCounty), numbers of coho salmon passing over

244 BROWN ET AL.

TABLE 2.-Examples of historical and recent abundance data for adult coho salmon in some California rivers.

Drainage basinand stream Year(s) Coho salmon abundance Source

Smith RiverWest Branch Mill Creek

Klamath RiverLower River

Shasta River counting racks=

Klamathon racksb

Hoopa Valley Reservation

1919 11,162 in gill-net fishery1951 1,187 in sport catch1954 4,000 in sport catch1955 1,145 in sport catchAug-Nov 1955 None countedOct 1957 3 10 counted1923 Abundant1925 2 9 5 counted1956 None counted1988 588 in gill-net fishery1989 525 in gill-net fishery

Trinity RiverMain stem, Lewiston 1958-1963

Main stem, belowTrinity Hatchery

Redwood Creek

1970

1973 Estimated run of 2,000 U S B L M ( 1 9 7 3 , A p p . )1 9 9 0 s Perhaps 2,000+ fish in some years S. Sanders and D. Anderson

Mad River 1938-1943

Eel River, South Fork(Benbow Dam)

Noyo River

CoastalTen Mile River

Big R ive rGualala RiverWaddell Creek

Scott Creek

1980-1989 Annual mean = 11 fish (range 2-28) Waldvogel(l988, App.)

Snyder (193 1)Gibbs and Kimsey (1955)McCormick (1958 , A p p . )Gibbs and Kimsey (1955)coots (1957)Coots (1958a)Bryant (1923)Snyder (1931)Coots (1958b)Tuss et al. (1989, App.)Kisanuki et al. (1991, App.)

Average annual wild escapement =228 fish (range = 7-583 fish)

2,098 estimated natural spawners

Smith and SharpC

Rogers (1973, App.)

(App . )

19461949

Annual counts on fishway: 498; 725;73; 308; 378; 259

Annual counts on fishway: 4 15; 5 10;515; 512

1938-1949 Annual counts on fishway: 7,370;8,629; 11,073; 13,694; 15,037;13,030; 18,309; 16,731; 14,109;25,289; 12,872; 7,495

ThousandsEstimated 6,000 spawnersEstimated 901-950 spawners

Murphy and Shapovalov (195 1)



195819731989-1990

spawning sea-s o n

Holman and Evans (1964)U S B L M ( 1 9 7 3 , A p p . )Nielsen et al. (1991, App.)

19731989-1990

spawning sea-s o n

Estimated 6,000 spawnersEstimated 3 l-55 coho, based on car-

cass count; estimated 80-92 com-bined coho salmon/chinook s a l m -on/steelhead, based on live fishcounts

U S B L M ( 1 9 7 3 , A p p . )Nielsen et al. (199 1, App.)

1973 Estimated 6,000 spawners1973 Estimated 4,000 spawners1930-1940 120-633 spawners annually1990 50+ spawners198Os-1990s 30-40 spawners

U S B L M ( 1 9 7 3 , A p p . )U S B L M ( 1 9 7 3 , A p p . )Shapovaiov and Taft (1954)J . S m i t h ( A p p . )J . Smi th (App . ) ; D . S t re ig

(APP. )

a 282 km upstream from mouth.b 301 km upstream from mouth.c Personal communication, cited in Fredericksen et al. (1980, App.).

Sweasey Dam fluctuated between 0 and 1,000 fishduring 1938-1961; in 1962, 3,500 fishpassedoverthe dam, followed by 1,500 in 1963, and less than500 fish in 1964 (Murphy and Shapovalov 1951;Fredericksen et al. 1980, App.). Counts at the MadRiver Hatchery fluctuated in the same range (500-1,000 fish) during 1971-1988. Thus, it appearsthat overall numbers have remained fairly steady,

although the relative contributions of hatchery andnative fish to the population are not known.

Humboldt Bay streams. -Streams tributary toHumboldt Bay (Humboldt County) historicallyhave been important to the local sport fishery, butestimates of coho salmon abundance are few (Hullet al. 1989, App.). Hallock et al. (1952) seined8,642 juveniles from Freshwater Creek, 17,671


from Elk River, and 14,243 from Jacoby Creek,which indicates there were substantial populationsin those streams. Freshwater Creek has been thefocus of population and habitat recovery effortsby the Humboldt Fish Action Council, which be-gan rearing coho salmon and chinook salmon forlocal population enhancement in the early 1970s(Hull et al. 1989, App.). Total escapement in theFreshwater Creek drainage was estimated at 454adults in 1986-1987 and 834 in 1987-1988. Theentire 1986- 1987 run apparently was native fish,but 68% of the 1987-l 988 run consisted of hatch-ery fish (Hull et al. 1989, App.). Initial enhance-ment efforts used nonnative stocks from the Alsea,Trask, and Sandy Rivers of Oregon; the Skagit,Soos, and Minter Rivers of Washington; and theTrinity, Noyo, and Klamath Rivers of California.We estimate that the Elk River supports an annualrun of around 400 native coho salmon, based onrecent sporadic surveys of both adults and juve-niles (CDFG 1990-1993, App.)

Eel River. -The Eel River (Humboldt County)probably supports the largest remaining nativepopulations in California. One estimate places theEel River run at 40,000 annually (U.S. HeritageConservation and Recreation Services 1980, App.);however, this figure exceeds a more recent state-wide estimate of 33,500 spawners (Sheehan 1991).Presently, coho salmon are known to spawn main-ly in the tributaries of the South Fork Eel River.In the main-stem Eel River, coho salmon areknown to have spawned in several small tributar-ies of Outlet Creek as recently as the 1988-1989season (G. Flosi, App.; W. Jones, App.). Nielsenet al. (199 1, App.) conducted surveys on 69 kmof Outlet Creek and on 12 of its tributaries duringthe 1989-l990 season but were unable to find anycoho salmon. All but four tributaries had sup-ported coho salmon in the past (Brown and Moyle1991 a, App.). The lower main-stem Eel River ap-parently is not used as spawning or rearing habitatto any significant degree (Murphy and Dewitt1951, App.). In the Van Duzen River, juvenileshave been captured recently in small numbers fromthe main river and two small tributaries, Grizzlyand Cummings Creeks (Brown and Moyle 1991 b,App.).

Tributaries of the South Fork Eel River havebeen recently surveyed by Nielsen et al. (1991,App.). In the 1989-1990 spawning season, lessthan 300 adult coho salmon were counted in thesystem, which probably supports, at best, 1,320spawners (see next section). Very few juvenileshave been present in areas of the South Fork Eel

River drainage where there is adequate habitat tosupport large numbers. Early reports documentsamples of thousands of juveniles from somestreams-4,844 in the Bull Creek system in 1939(Shapovalov 1940) and 3,000 in 195 1 (Hallock etal. 1952); 3,475 in Ten Mile Creek in 1951 and4,369 in 1952 (Kimsey 1952, 1953); and 1,250 inDean Creek’in 1939 (Shapovalov 1940).

Coho salmon were formerly more widespreadin the Eel River drainage. California Departmentof Fish and Game (CDFG) files contain reports ofcoho salmon in Indian Creek (main-stem tributaryabove Outlet Creek) and several tributaries toTomki Creek. During the 1946-1947 season, 47coho salmon passed through the Van Arsdale fishfacility on the upper main-stem Eel, 315 km fromthe sea, but they have not been recorded theresince (Grass 1990a). The Tomki Creek drainagehas been intensively studied since 1986, but nocoho salmon have been captured or observed(Steiner Environmental Consultants 1990, App.).There are no recent records of coho salmon intributaries to the North Fork Eel and Middle ForkEel Rivers (W. Jones, App.; L. Brown, personalobservation), although they were formerly presentthere.

Mattole River.-The coho salmon run in theMattole River (Humboldt County) is much re-duced from historic levels-to less than 800 fishannually. Community-based restoration effortshave been underway for several years, but thereis a “good” run in only one out of three years (G.Petersen, App.). Plantings of hatchery fish havenot noticeably increased spawner returns, but theprogram has led to the establishment of popula-tions in tributary streams (Miller et al. 1990). Cohosalmon still migrate far up the Mattole River intothe South Fork Bear River (Preston 1988-1989,App.), but exact numbers are unknown.

Mendocino County streams. -In MendocinoCounty, juvenile coho salmon were found in 40of 146 streams recently surveyed (W. Jones, App.).At 71 stations, mean juvenile coho salmon densitywas 0.4 1 fish/m2 (range 0.0l - l .61 fish/m2). Bakerand Reynolds ( 1986, App.) reported coho salmonin only 21 of 70 major streams surveyed in Men-docino County. Thus, coho salmon appear to beabsent or very rare in many of the streams theyoccupied historically. For example, they have notbeen recently observed in Usal Creek, which wasonce used as a source of juvenile coho salmon toplant elsewhere. Numbers of juveniles removedfrom Usal Creek were 3,963 (1940), 60, 510 (1944),61,133 (1945), 11,455 (1951), and 13,864 (1952)

246 BROWN ET AL.

(Shapovalov 1940, 1945, 1949; Kimsey 1952,1953). Most recently, Nielsen et al. (1991, App.)found coho salmon populations to be low in all of82 streams (571 stream kilometers) surveyed inMendocino County. Only the Noyo River, whichis routinely planted with large numbers of fry andsmolts, had a population of more than 500 fish.Based on the number of coho salmon carcassesper stream-mile surveyed, the index of abundancefor recent spawning seasons for the Noyo Riverhas been 1.69 (1989-l990), 0.92 (1990-l991) and1.58 (1991-1992) (J. Nielsen, App.)

Sonoma County streams. - Coho salmon arepresent in Salmon Creek, Russian River, GualalaRiver, and their tributaries. Present numbers inSalmon Creek and the Gualala River are evidentlysmall (B. Cox, App.), although the 1973 spawningpopulation for the Gualala was estimated at 4,000fish (USBLM 1973, App.). Coho salmon have beenreported from the Russian River and 27 tributarystreams, but most of the streams no longer havepopulations (Brown and Moyle 1991 a, App.).

Marin County streams. -Several coastal streamsin Marin County have small coho salmon runs (B.Cox, App.), but there are insufficient historical datato determine trends. Olema Creek and its tribu-taries are believed to support a run of about 200native coho salmon, and Redwood Creek has arun of about 75 or more fish (B. Cox, App.). Thepresent population in Lagunitas Creek has beensignificantly reduced from historical levels (Smith1986, App.), despite efforts to preserve and en-hance the run (Brown and Moyle 1991 a, App.). Inthe early 1900s special trains brought anglers fromthe San Francisco Bay area to Lagunitas Creek tofish for coho salmon and steelhead (Smith 1986,App.), and the creek produced a state size recordfor coho salmon in 1959 (Giddings 1959). Thepopulation decline was associated with the con-struction of Kent and Nicasio Reservoirs, whichblocked access to the upper reaches of this system.Annual 1 -or 2-d counts of coho salmon in portionsof the Lagunitas-San Geronimo Creek system since1984-1985 have varied, but indicate that thespawning run is generally less than 100 fish (Kelley1991, App.), despite the stocking of coho salmonjuveniles (Noyo River stock) in 1985 (20,040 fish),1987 (3,888 fish) and 1988 (5,000 fish) (B. Cox,App.). Stream surveys indicate that about 400-500 adults were present in the Lagunitas Creeksystem (including Olema Creek) during the 199 l-1992 spawning season -the highest number seenin the last 7-8 years (B. Cox, App.). Approximately94-l16 coho salmon adults and 4l-53 redds were

observed during an extensive survey of Lagunitasand San Geronimo Creeks in January 1992 (W.Lifton, App.). An enhancement hatchery that useslocal spawners has recently operated on San Ge-ronimo Creek (L. Cronin, App.).

San Francisco Bay tributaries. -Within SanFrancisco Bay, coho salmon runs have been ex-tirpated, or nearly so. Skinner (1962) indicatedthat prior to human disturbance, most streamswith suitable habitat had coho salmon. Spawningmigrations were noted in Walnut Creek from the1950s to the mid- 1960s (Leidy 1984). Coho salm-on also have been recorded from Corte Madera(San Anselmo) Creek (Fry 1936; Hallock and Fry1967) and Mill Valley Creek (Hallock and Fry1967), and juvenile coho salmon were captured inboth streams during Leidy’s (1984) survey of SanFrancisco Bay streams. There have been no cohosalmon observed in Corte Madera Creek in thelast 7-8 years (B. Cox, App.). The threats to anyremaining fish are the degradation of habitat andwater quality caused by continued urban devel-opment.

Sacramento River drainage. - Coho salmon oc-curring in the Sacramento River during recent de-cades have been regarded as strays (Hallock andFry 1967; Fry 1973). Hallock and Fry (1967) re-ported that during 1949- 19 5 6 only two coho salm-on were seen in the Sacramento River, at the Cole-man National Fish Hatchery, and an additionalfish was reported there before 1949 (J. Pelnar, App.,cited in Hallock and Fry 1967). Older records sug-gest that coho salmon may once have been moreabundant in the Sacramento River. Jordan andJouy (1881) list a museum specimen from the Sac-ramento River, and Jordan and Gilbert (1881) de-scribe coho salmon as occurring from the Sacra-mento River northward. D. S. Jordan also reporteda fall run of coho salmon in the Sacramento River(U.S. Commission of Fish and Fisheries 1892).Eigenmann (1890) listed coho salmon as one ofthe four salmon species occurring in the Sacra-mento River and stated that runs occurred in thatriver in the summer and fall. He did not term cohosalmon rare as he did pink and chum salmon,which might indicate a higher level of abundanceof coho salmon at that time. The lack of moredefinitive information about the abundance of cohosalmon may be due to the difficulty of identifi-cation mentioned by Eigenmann (1890), as wellas a general lack of interest in coho salmon com-pared with the much more abundant chinooksalmon. For example, Snyder (1931) stated thatcoho salmon occurred in large numbers in the

Klamon coyear :

Inblish1956,CreeklingsFry 1Riveiscatteconcethe fithey 1Fry 1POPUbers (tified1978er RjAPP.

SO

tion (of Sadue tthe clby agAPP.ducehas cand tThe (are iLorePOPCrally1976latiothrowplanCreeturn!haveefforMOYtrapiH o whabiif thPOPI

TlCreeof C(ican


Klamath River, although up to that time statisticson coho salmon had been recorded only for theyear 1919.

In 1956-l958 the CDFG attempted to reesta-blish coho salmon in the Sacramento River. In1956, 43,025 yearlings were released into MillCreek (Tehama County), followed by 53,505 year-lings in 1957, and 48,000 in 1958 (Hallock andFry 1967; Fry 1973). These fish were of LewisRiver (Washington) stock. The returning adultsscattered throughout the drainage, with the largestconcentrations occurring at Battle Creek, wherethe fish had been raised, and Mill Creek, wherethey were planted. The fish spawned (Hallock andFry 1967) but failed to establish a self-sustainingpopulation in Mill Creek. Since then, small num-bers of coho salmon have been consistently iden-tified at Nimbus Hatchery (Jochimsen 1971 to1978c), and they also have been taken in the Feath-er River (Schlichting 1974; Painter et al. 1977,App.).

Southern populations. -Most natural produc-tion of coho salmon in the smaller streams southof San Francisco Bay appears to have been lostdue to the 1976-l 977 drought, which exacerbatedthe cumulative effects of stream alterations causedby agriculture, logging, and urbanization (D. Streig,App.). Apparently stray fish occasionally repro-duce in these small coastal streams (e.g., spawninghas occurred in San Gregorio Creek: Coots 1973),and they possibly could establish new populations.The only sustained coho salmon runs in this regionare in Waddell and Scott Creeks, and in the SanLorenzo River, which has a hatchery-maintainedpopulation. The San Lorenzo River lost its natu-rally spawning coho salmon population during1976-1977. However, much or all of that popu-lation was the result of stocking from the 1950sthrough the mid-1970s (J. Smith, App.). Smoltsplanted from Noyo River, Prairie Creek, and ScottCreek stocks have reestablished coho salmon re-turns to the system, and fish returning to the riverhave been trapped and artificially spawned in aneffort to establish a resident stock (Brown andMoyle 199la, App.). The number of adult fishtrapped from the river peaked in 1989 at 183 fish.However, this stream has undergone extensivehabitat loss and degradation, and it is not knownif there is adequate habitat for a self-sustainingpopulation to be established.

The coho salmon runs in Waddell and ScottCreeks (Santa Cruz County) are the most southernof coho salmon populations on the North Amer-ican Pacific Coast- the closest population is 159

km north (in Redwood Creek, Marin County).Waddell Creek, the site of Shapovalov and Taft’s(1954) classic study of coho salmon and steelheadlife history, presently maintains a much reducednatural run, although there is still extensive cohosalmon rearing habitat present (J. Smith, App.).This stream was heavily planted with juvenile cohosalmon in 1929, 1930, and 1933. In the early 1970sthere were additional introductions of fish fromWashington State (Taylor 1991, App.). Also, anumber of imported stocks have been introducedby private aquaculturists in recent years, but re-cords of egg sources were not kept (D. Streig, App.:cited in Bartley et al. 1992). Between 1930 and1940, the population varied between 120 and 633spawners (Shapovalov and Taft 1954). The pres-ent run is about 50 fish in better years, but muchless in poor years (J. Smith, App.); “good” runsoccur every third year, most recently in 1989- l990 .Lack of early rainfall in the 1990-l 991 spawningseason prevented adults from migrating upstreamat the usual time, and reproduction was probablyvery low (J. Smith, App.). There were approxi-mately 65 adults estimated for the 199 l-l 992 runin Waddell Creek, but at least three probable cohosalmon redds were destroyed by scouring follow-ing a postspawning storm (Smith 1992a, 1992b,App.). Scott Creek also maintains a natural cohosalmon run, which averages 30-40 fish per year,although it probably contains stray fish from near-by Waddell Creek. Scott Creek and its tributary,Big Creek, have been the focus of intensive re-habilitation efforts and may provide the best hab-itat for coho salmon south of San Francisco Bay(D. Streig and J. Smith, App.). At present there isan enhancement hatchery on Big Creek that useslocal spawners (Smith 1992a, App.); however, thenumber of adult spawners returning to the ScottCreek drainage appears to be extremely low (Mar-ston 1992, App.). The somewhat late run andspawning times (December-February; J. Smith,App.) of coho salmon in Waddell and Scott Creeksindicate adaptations to local conditions that havepersisted despite past interbreeding with importedstocks.

Extent of the Decline

Presence or Absence of Coho SalmonStream Populations

We classified all streams previously known tohave contained coho salmon either as (1) “pres-ent,” where coho salmon are present, whether nat-ural-spawning or hatchery fish, (2) “absent,” where

248 BROWN ET AL.

TABLE 3. -Counts for current presence of coho salmon in California streams known to have historically containedcoho populations. Results are listed by county; county classifications are based on the location of the mouth of thestream system. Streams wher e coho salmon are present some years but not in others are classified as havin g coho.Streams receiving hatchery plants are not counted as having coho salmon unless adult returns were documented.Corresponding percentages for the categories are given in parentheses.

Location SystemNumber of

streams Coho present Coho absent No data

Del Norte County Coastal 9 l ( l l%) 8 (89%)Smith River 4 1 2 (5%) 39 (95%)Klamath River 1 1 3 21(18%) 20 (18%) 72 (64%)

Humboldt County Coastal 3 4 7 (21%) 27 (79%)Redwood Creek 1 4 3 (21%) 11 (79%)Mad River 2 3 2 (9%) 21(91%)Eel River 1 2 4 34 (27%) 22 (18%) 68 (55%)Mattole River 3 8 3 (8%) 35 (92%)

Mendocino County Coastal 4 4 13 (30%) 22 (50%) 9 (20%)Ten Mile River 1 1 7 (64%) 3 (27%) 1(9%)Noyo River 1 3 11 (84%) 1(8%) 1(8%)Big River 1 6 11 (69%) 2 (13%) 3 (18%)Navarro River 1 9 4 (21%) 4 (21%) 11 (58%)

Sonoma County Coastal 1 0 1 (10%) 1 (10%) 8 (80%)Gualala 1 1 1(9%) 1(9%) 9 (82%)Russian River 3 2 2 (6%) 22 (69%) 8 (25%)

Marin County Coastal 1 0 7 (70%) 3 (30%)

San Francisco Baya Coastal 7 7 (100%)

South of San Francisco B a y Coastal 1 3 5 (38%) 8 (62%)

Total 5 8 2 135 (23%) 113(19%) 334 (58%)

a Includes Sacramento River and other tributaries to the Bay.

coho salmon are known to be very rare or extir-pated, or (3) “undetermined,” where sufficient in-formation is lacking to evaluate their coho salmonstocks. The time frame for which we consider cohosalmon to be present in a stream is from the early1980s up to 1991. For example, if coho salmonwere reported to occur in a stream during thatperiod or if our sources were of the opinion thatthe fish probably still existed there, we consideredthat stream to have coho salmon present. We notethat this time period includes years prior to therecent 6-year drought (1987-l 992); thus, streamsthat had lost their coho salmon runs by the late1980s did so for reasons other than, or in additionto, natural drought-related causes. The results ofour enumeration are summarized by county inTable 3. County classifications of streams werebased on the location of the mouth of the systemrather than by individual stream locations.

There are 5 82 California streams known to havesupported coho salmon at some time (a completelisting is given by Brown and Moyle 1991 a, App.).That number represents a minimum, because somestreams possibly had not been surveyed during theyears when they contained coho salmon and werenot included in the total. Furthermore, the number

of streams with coho salmon undoubtedly fluc-tuated as climatic conditions changed. Bearingthese points in mind, our assessment is as follows.We lack data on the recent status of coho salmonin 57% of the 582 identified coho salmon streams.Of the 248 streams for which we have some ideaof recent coho salmon presence or absence, 54%had coho salmon and 46% did not. The amountof data varied between counties; the percentage ofstreams that could not be classified was greatestin the northern part of the state and lowest in thesouth. This pattern may be partly owing to thegreater number of streams in the north, which arealso generally less accessible than those in the south.Differences in management emphasis among fish-eries managers also may be a contributing factor.The greatest concern for coho salmon appears tooccur in Mendocino County and southward, prob-ably because the smaller river systems in the southhistorically supported a higher proportion of cohosalmon to chinook salmon than did the largernorthern systems. Also, in the far south (SonomaCounty southward), there is concern for preservinghabitat for all species of anadromous fishes in theface of rapid urbanization.

In Del Norte County, on the Oregon border,


73% of the streams lacked information about theircoho salmon status. In the Klamath drainage, onlyabout half of the streams for which there was in-formation still had coho salmon; most of thestreams without coho salmon were tributaries tothe South Fork Trinity River and Salmon River.In all, 45% of the streams in Del Norte Countyfor which there were records no longer containcoho salmon.

In Humboldt County, 70% of the streams wereclassified as undetermined; if the Eel River drain-age is omitted, the percentage increases to 86%.Overall, 31% of Humboldt County streams forwhich there are recent records are without cohosalmon, all in the Eel River system. Of 103 formercoho salmon streams in Mendocino County, 24%were undetermined, 45% contained coho salmon,and 31% lacked coho salmon. Thus, 41% of clas-sified streams in Mendocino County evidently havelost their coho salmon stocks. Sonoma Countycontains 53 streams that historically containedcoho salmon. Of those, 47% could not be classi-fied; of the remaining 28 streams, 4 streams (14%)contained coho salmon and 24 (86%) did not. FromMarin County southward, only 30 streams his-torically contained coho salmon, not includingSacramento River tributaries. There were no datafor 3 streams, 15 have lost their populations, and12 streams still are used by coho salmon, at leastoccasionally. Most of the streams having cohosalmon in the region south of San Francisco Bayare very small, support few salmon, and are sup-plemented by hatchery stocks.

Estimates of Abundance

Little quantitative data exist on which to baseestimates of coho salmon abundance in California,and we therefore assumed that many streams forwhich data were lacking still contained coho salm-on (our 20-fish rule). In most cases when previousestimates of adult abundances were available forindividual streams, they were no greater than ourassumed basal estimate of 20 fish per stream.Numbers of coho salmon passing over SweaseyDam on the Mad River varied between 0 and1,000 fish during 1938-l961; our estimated num-ber of coho salmon in the system, discounting fishreturning to the hatchery, was 460 fish, which fitswell within the historic range of values (Table 4).We estimated the population in the Outlet Creekdrainage (tributary to the Eel River) to be 240 fish,although in 1989-1990 there was no evidence ofspawning there. In the South Fork Eel River drain-age, we estimated the Hollow Tree Creek popu-

lation to be 180, assuming 20 fish per tributary.That number is comparable to the 162 fish count-ed at the Hollow Tree Creek egg-taking station in1989-l990 and exceeds the counts for several oth-er years. Also in the South Fork Eel River, our20-fish rule predicted 140 coho salmon in the TenMile Creek drainage when none were seen in the1989-l990 season. Nielsen et al. (1991, App.) es-timated fewer than 100 spawners of all speciescombined in the coastal Ten Mile River system,for which our estimate was 160 fish. We similarlyoverestimated population sizes for the Big, Little,Garcia and Gualala Rivers. Cases in which the 20-fish-per-stream rule underestimated the popula-tion occurred mainly where there was ongoinghatchery supplementation, as in the Noyo River,tributaries around Humboldt Bay, Scott Creek,and the San Lorenzo River.

In most cases, therefore, our estimates for nativeand naturalized fish seem to be biased upwards.The degree of overestimation may be substantialbecause we assumed that all streams for whichthere were insufficient data still contained cohosalmon. Despite the probable overestimation ofnative and naturalized coho salmon populationsizes, those populations overall are at low levels.Furthermore, the recent (1986-l992) drought hasreduced some small populations to near extirpa-tion (e.g., in the Eel River system and some Klam-ath tributaries).

Our total population estimate for coho salmonin California is about 31,000 fish (Table 4). How-ever, hatchery populations contribute over halfthat number (57%) and natural spawners numberonly 13,240. Probably the largest concentration ofcoho salmon with little hatchery influence pres-ently occurs in the South Fork Eel River system,which we estimate to have 1,320 fish (using our20-fish rule for streams that lacked data). This isprobably a substantial overestimate, given the ab-sence of fish from many of the tributaries duringthe 1989-1990 survey by Nielsen et al. (1991,App.). The actual number probably is lower than1,320 fish but greater than 300 (roughly the num-ber counted by Nielsen et al. 199 1, App.). Simi-larly, our total estimate of 13,240 native and nat-uralized coho salmon statewide could easily beoverstated by 50% or more. This estimate is fur-ther reduced if naturalized (hatchery-influenced)stocks are discounted because of doubts over theirlong-term sustainability. Thus, native coho salm-on stocks in California probably have averagedless than 5,000 fish in recent years (Table 4). Manypopulations in the smaller drainages number less

250 BROWN ET AL.

TABLE 4.-Estimates of coho salmon abundance in California streams that historically contained coho salmon.Streams that are known or believed to currently support coho salmon, and streams for which we found no data onpresence or absence of coho salmon were all assumed to support 20 spawners, unless the available data indicateda larger population. Numbers for hatchery populations are the average population from the 198 l-l 982 spawningseason until the latest season (up to 1990-l991) for which data were available. For streams where hatcheries arelocated, both hatchery and natural-spawning fish are included. An asterisk (*) indicates a high probability that thenatural production is by native fish rather than naturalized fish with hatchery ancestry. An S in the hatchery columnindicates streams where it was difficult to classify fish as natural spawning or hatchery produced. Supplementationoccurs in these streams, but in the Noyo River most of the production is probably natural, and in Scott Creek(south of San Francisco Bay) only returning naturalized fish are spawned.

Numbers of fish

Location SystemNumber of Native and

streams naturalized Hatchery Total

Del Norte County

Humboldt County

Mendocino County

Coastal 9Smith River 4 1KIamath River 9 3

00

16,265aCoastal 3 4Redwood Creek 1 4Mad River 2 3Eel River 1 0 2Mattole River 3 8

180*820*

1,860680*2 8 0460

2,040*760*

05 2 53 6 6

00

Coastal 2 2 470 0Ten Mile River 8 160* 0Noyo River 1 2 3,740 SBig River 1 4 280 0Navarro River 1 5 3 0 0 0

Sonoma County

Marin CountySan Francisco Bayb

South of San Francisco Bay

TotalPercentage of total

Subtotal for native fishSubtotal for non-hatchery streamsd

CoastalGualala RiverRussian River

CoastalCoastal

Coastal

91 01 0

1 0

5

469

1 8 0 02 0 0 02 5 5 3 3 2

4 3 5 0

0 0

1 4 0 S

1 8 08 2 0

18,1256 8 08 0 58 2 6

2,0407 6 04701 6 0

3,7402803 0 01 8 02 0 05 8 7

4 3 50

1 4 0

1 3,240c4 3 %4,640

10,385

17,4885 7 %

30,728

a Number includes fish from Iron Gate Hatchery and Trinity Hatchery. Also included are hatchery fish spawning below TrinityHatchery, based on the assumption that only 60% of returning hatchery fish actually enter the hatchery, with the remainderspawning outside (Rogers 1973).

b Includes Sacramento River and other tributaries to the Bay.c Includes 4,640 fish of wild ancestry (35% of nonhatchery fish).d Excludes Prairie Creek and the Klamath, Mad, and Russian Rivers, but not the Noyo River, where fish were difficult to classify.

than 100 fish-probably lower than the populationsize necessary to preserve genetic integrity of thestock and to ensure its survival against randomenvironmental disasters. The abundance of nat-urally spawning coho salmon, especially nativestocks, clearly is at a low level, and the trendsindicate coho salmon numbers are continuing todecline statewide. Coho salmon in California (in-cluding hatchery stocks) presently are less than 6%of their abundance during the 1940s and probablyhave declined at least 70% in numbers since the1960s.

Naturally spawning coho salmon stocks on theOregon coast also appear to be numerically de-

pressed. For 1990, the standard index of cohosalmon abundance used by the Oregon Depart-ment of Fish and Wildlife (ODFW) was the thirdlowest value recorded in the 41-year observationperiod (Cooney and Jacobs 1992). The estimatedspawning escapement for standard sampling streamsegments in 1990 averaged 10 fish/km (the lowestsince 1983), compared with the l0-year averagesince 1981 of 19 fish/km (Cooney and Jacobs 1992).The total stock size of naturally spawning cohosalmon in Oregon coastal river and lake basinswas estimated at 103,964 fish in 1990, which was52% of the 200,000-fish spawning escapement goalset by the Pacific Fishery Management Council

(1tiaASC

B

dcC0inA

alinit,PIinalal

L

aan-lW

alcisathaIatthdigiit:Oi

PII!W

a1

In

it

:inPIinOi

bcVt


(PFMC) (Cooney and Jacobs 1992), and the es-timated escapement for 1991 was 109,000 fish,compared with the PFMC goal of 161,000 (L. Osis,App.). However, even these escapement values mayseriously overstate the true spawning numbers.Based on a new estimation method (stratified ran-dom sampling) employed by ODFW (Jacobs andCooney 199 l), the number of spawners for theOregon coast was estimated to be closer to 20,000in 1990 and 33,000 in 1991 (Engelmeyer 1992,App.).

Causes of the Decline

The causes of coho salmon decline in Californiaare multiple and interacting but can be dividedinto four broad categories: (1) loss of stream hab-itat, (2) interactions with hatchery fish, which canproduce a loss of genetic integrity, and increasesin competition and disease, (3) overexploitation,and (4) climatic factors, such as oceanic conditionsand precipitation.

Loss of Stream HabitatLoss of stream habitat is widely acknowledged

as the single biggest cause of declines of anadro-mous salmonids in general in the Pacific North-west and of coho salmon in particular (Nehlson etal. 1991; Reeves and Sedell 1992; Wilderness So-ciety 1993). In California, some of the loss of cohosalmon habitat has been the result of large damsthat eliminate access to upstream spawning areasand reduce flows in downstream areas (e.g., Klam-ath, Mad, and Russian Rivers). However, most ofthe habitat loss has been the result of watersheddisturbances associated with urbanization, log-ging, agriculture, mining, and other human activ-ities. Such activities are associated with severe lossof the habitat complexity that is characteristic ofproductive coho s almon streams (Sandercock199 l), especially reduction in the amount of largewoody debris and the sedimentation of spawningand rearing areas.

The loss of coho salmon habitat has been cu-mulative over at least the past 50 years. By 1956,it was estimated that over 1,600 km of streams inCalifornia had been lost as important fish habitat(Fisk et al. 1966). Damage was particularly severein coastal streams affected by logging. For exam-ple, 84 of 167 km of potential coho salmon habitatin the Garcia River (Mendocino County) and 110of 135 km of habitat in Redwood Creek (Hum-boldt County) were reported as moderately to se-verely damaged by ongoing logging practices (Fisk

et al. 1966). Habitat damage was attributed to ero-sion and land slippage that had been exacerbatedby the construction of logging roads and skid trailson steep slopes and by the removal of vegetativeground cover. Similarly, a 1973 survey of TenMile, Noyo, Big, and Gualala Rivers revealed thatpotential Coho salmon habitat in all those rivershad been negatively affected by logging, roadbuilding, grazing, or urbanization (USBLM 1973,App.). A review of the effects of logging on salmo-nids in California streams indicated that loggingseverely reduced the number of coho salmon smoltsemigrating out of a watershed (Bums 1972). Thesmolts also emigrated at much smaller sizes thanhad been noted for unlogged streams, presumablydue to stress associated with physical changes inthe habitat (Graves and Bums 1970).

Interactions with Hatchery Fish

Loss of genetic integrity. -The amount of in-terbreeding of nonnative hatchery strains and ofwild native strains of coho salmon is poorly un-derstood (Steward and Bjorrn 1990; Hindar et al.1991). In California, determination of the amountof interbreeding is particularly difficult becausepopulations of native coho salmon within the statedo not appear to be strongly differentiated genet-ically (Bartley et al. 1992). This pattern is consis-tent with other regional studies of coho salmon(Hjort and Schreck 1982; Wehrhahn and Powell1987). However, there is obviously genetic differ-entiation between coho salmon stocks from dif-ferent geographical regions, as indicated by proteinvariation (Utter et al. 1970; Hjort and Schreck1982; Wehrhahn and Powell 1987; Milner, inpress), life history and morphology (Hjort andSchreck 1982; Taylor and McPhail 1985a), differ-ential juvenile survival and growth in seawater(Murray et al. 1993), heritable differences in swim -ming performance (Taylor and McPhail 1985b),and relative survival in the wild (McIntyre 1984).Genetic differences between coho salmon stocksfrom different streams also have been observedfor agonistic behavior (Rosenau and McPhail1987) susceptibility to a myxosporean disease(Hemmingsen et al. 1986) transferrin genotype(Utter et al. 1970), and resistance to bacterial kid-ney disease (Winter et al. 1980). Bartley et al. (1992)observed several differences between Californiacoho salmon and more northern populations inthe frequency and occurrence of alleles at severalloci, and Milner (in press) was able to differentiateKlamath River coho salmon from coho salmon ofmore northern regions based on allozyme data. It

252 BROWN ET AL.

is likely that the full extent of genetic differentia-tion among coho salmon populations has yet tobe fully determined, given the considerable adap-tive variation in traits observed between localstocks of other salmon species (Ricker 1972; Tay-lor 199 1; Clarke et al. 1992).

Nevertheless, genetic studies on coho salmonthroughout their range indicate that overall levelsof variability are low compared with other Pacificsalmon and anadromous trout (Utter et al. 1980;Wehrhahn and Powell 1987; Bartley et al. 1992;Milner, in press). Although we cannot be certainof the reasons for this low variability, reduced ge-netic variation is a well-recognized consequenceof small population sizes (Meffe 1986; Stewardand Bjornn 1990). It is clear that coho salmonpopulations in some geographical areas experi-enced extreme depletion in the past (Johnson etal. 199 l), and many are currently in populationbottlenecks, particularly in California.

Bartley et al. (1992) used protein electrophoresisto study the genetic structure of 27 populations ofcoho salmon from throughout California. Allo-zyme variation occurred at 23 of 45 loci (5 1 % ) ,but much of the variation was due to rare alleles(frequency <5%) present in only a few samples.Of 39 variant alleles, 27 (69%) occurred at threeor fewer locations, and the distribution of thesealleles did not follow any particular geographicpattern. The average level of gene flow betweenCalifornia populations (Nm ) was 1.3 individualsper population per generation; this level is highfrom an evolutionary perspective but low in termsof the actual number of individuals being ex-changed between populations (Bartley et al. 1992).Thus, gene flow among California coho salmonpopulations is lower than that reported for localpopulations of British Columbia coho salmon (Nm= 5.8 + 1.2; Wehrhahn and Powell 1987), butcomparable to that for chinook salmon (Nm =1.16) and coastal rainbow trout Oncorhynchusmykiss (Nm 1 1.7) in California (Berg and Gall1988; Bartley and Gall 1990).

Bartley et al. (1992) noted that transplants ofdifferent stocks within California may have ob-scured whatever genetic differentiation formerlyexisted among them and that, in addition, manystreams have been planted with coho salmon fromOregon or Washington stocks. Possible effects ofsuch interbreeding include disruption of locallyadapted gene complexes, swamping and homog-enization of native gene pools, and transmittal ofnonadaptive traits from hatchery stocks to wild(native) stocks. Indirect genetic changes in wild

fish are also possible, resulting from the alteredenvironment or from processes, such as geneticdrift and inbreeding, that accompany reducedpopulation sizes caused by the presence of hatch-ery fish (Steward and Bjornn 1990; Krueger andMay 199 1; Waples 199 1). What is definitely known,however, is that the long history of stocking im-ported coho salmon strains in the lower ColumbiaRiver system has essentially resulted in the geneticextinction of native coho salmon runs in that sys-tem (Johnson et al. 1991).

The reasons for such extinctions are fairly wellunderstood (Fleming and Gross 1989; Skaala etal. 1990; Steward and Bjornn 1990; Waples 1990a,1990b, 1991; Waples and Teel 1990). Large dif-ferences in genetic structure of native and hatcherystocks can potentially lead to lower survival ofsubsequent hybrid generations compared with purewild fish (Steward and Bjornn 1990; Hindar et al.199 1). Even the simple homogenization of localstocks by an influx of genes from hatchery fishmay be detrimental in the long run. Such homog-enization might reduce between-population ge-netic diversity and lead to decreased overall pro-ductivity and increased vulnerability to localextirpation by unpredictable environmentalchanges (Waples 199 1). Obversely, supplementa-tion with hatchery fish may have positive effectson small wild stocks that already have lost, or willsoon lose, much of their genetic variability (Stew-ard and Bjornn 1990). For example, Bartley et al.(1992) noted that, of the two southernmost runsof coho salmon, Waddell Creek fish had the high-est level of heterozygosity (0.05, based on 10 in-dividuals) for any California coho salmon popu-lation, presumably as the result of interbreedingwith imported stocks. In contrast, nearby ScottCreek, which has not been planted with importedstocks, had the lowest heterozygosity (0.0). In thelong run, the Waddell Creek population shouldhave a greater probability of survival because theincreased heterozygosity should improve its abil-ity to adapt to changing environmental conditions.

Despite these potentially significant genetic con-sequences of stock supplementation, Steward andBjornn (1990) concluded that there was as yet noconclusive evidence that hatchery stocks causedactual genetic harm (or benefit) to native popu-lations of Pacific salmonids. However, Skaala etal. (1990) and Hindar et al. (199 1) cited exampleswhere there has been introgression of hatchery ge-notypes into native salmonid stocks, in spite oflower survival or reproductive success of hatcheryfish (see also Johnson et al. 199 l), and in some

insnifiet irephaythemaant

pr(o nthttoih aSig

b uferofre!

ticr oerpro fa’a lsejumSk

Wernc01tk

f-ih,t el edefSi

irt i

7

t lrlC

t1Itaa


instances hatchery (or farmed) fish composed sig-nificant fractions of in-river populations (Hindaret al. 199 1). The likelihood of detrimental geneticrepercussions on native stocks is great enough tohave led to explicit recommendations regardingthe use of hatchery supplementation in salmonidmanagement in the Pacific Northwest (Stewardand Bjornn 1990; Hindar et al. 199 1; Waples 199 1).

An additional, and largely unexpected, geneticproblem has been the hybridization of coho salm-on with chinook salmon in the Klamath River asthe result of hatchery practices and the crowdingtogether of the two species in limited spawninghabitat below a dam (Bartley et al. 1990). Thesignificance of such hybridization is not yet known,but Hedrick et al. (1987) speculate that gene trans-fer between the species may increase susceptibilityof coho salmon to diseases to which they are nowresistant.

Competition with hatchery stocks. -Introduc-tion of hatchery-raised fish into the natural envi-ronment can result in competition between hatch-ery and native fish. For example, release of hatcherypresmolts in Oregon streams reduced the densityof native juvenile coho salmon by 40-50%, witha subsequent reduction in adult returns (Miller etal. 1990). Similarly, Nickelson et al. (1986) ob-served reductions in density of native coho salmonjuveniles in streams previously stocked with pres-molts, and returning spawners in subsequent yearsshowed a shift toward an earlier spawning time,which is indicative of hatchery fish. There are sev-eral possible mechanisms leading to the observednet losses of native juveniles. Juvenile coho salm-on are territorial (Shapovalov and Taft 1954) andthe larger hatchery fish can displace smaller nativefish (Nickelson et al. 1986). Fish with territorieshave an energetic advantage over those lacking aterritory (Puckett and Dill 1985). Hungry fish areless responsive to predators, so mortality at highdensities would be higher (Dill and Fraser 1984),especially for displaced native fish. At high den-sities, growth of coho salmon is depressed throughintraspecific competition for resources, and mor-tality is increased (Fraser 1969). Shapovalov andTaft (1954) noted an inverse correlation betweenthe number of downstream migrants and adultreturns, implying that in years when intraspecificcompetition is low, downstream migrants are bet-ter able to survive ocean life. Also, Steward andBjomn (1990) cited studies showing that survivalto adulthood of hatchery-produced coho salmonand other salmon is positively related to their sizeat release, which further suggests that competition

(when it affects juvenile size) can affect productionof adults.

Competition among adults for spawning sitescan occur. When native stocks are small andhatchery supplementation occurs, hatchery fishmay outnumber native fish and monopolize theavailable spawning habitat. The negative effects ofsuch competition can be magnified when naturallyspawning hatchery stocks have lower spawningsuccess than native fish (Fleming and Gross 1992).Hatchery stocks also may produce fewer smoltsand returning adults (Steward and Bjomn 1990;Hindar et al. 199 1).

Disease. -The transmission of diseases fromhatchery to native stocks of coho salmon, or evenbetween salmon species (Hedrick et al. 1987) ispotentially a serious problem for wild coho salm-on. However, the effects of introduced diseases onnative stocks are not known. Documented casesof disease introductions in North America fromstocked or escaped salmonids are uncommon(Krueger and May 199 1). Steward and Bjomn(1990) likewise could find little evidence for theimportance of transmission of disease from hatch-ery to native stocks, primarily because little workhas been done, but they concluded that the fullimpact of disease on supplemented stocks is prob-ably underestimated. The protozoan Ceratomyxashasta occurs in the lower Columbia River, whereit has caused large losses in hatcheries, but notelsewhere on the Oregon coast (Johnson et al.199 1); the lower resistance of some coastal stocksto this parasite has been experimentally shown tobe heritable (Hemmingsen et al. 1986). Viral hem-orrhagic septicemia (VHS) has been reported fromhatchery coho salmon in Washington, and thereare a number of other virulent diseases that affectsalmonids (e.g., bacterial kidney disease, infec-tious hematopoietic necrosis, herpes virus infec-tions, infectious pancreatic necrosis) with the po-tential for transmission between hatchery andnative stocks (H&stein and Lindstad 199 1).

Overexploitation

Overfishing is often mentioned as a major factorcontributing to the decline of coho salmon, but itseffects are poorly known because catches of wildand hatchery fish are rarely separated (Steward andBjomn 1990). Hatchery production can indirectlycause severe depletion of native stocks by en-couraging intensified harvest or natural predation(Hindar et al. 199 1). Continued harvest of de-pleted wild stocks may prevent recovery and re-duce genetic variability or increase hybridization

254 BROWN ET AL.

rates with hatchery fish. Coho salmon in Californiaare particularly vulnerable to overfishing becausemost females have a 3-year life span (althoughsmall numbers of 4-and 5-year-olds have been ob-served in Mendocino County spawning streams;J. Nielsen, App.). California coho salmon popu-lations, therefore, lack the resilience to withstandexcessive harvest compared with other salmonidspecies (e.g., chum salmon Oncorhynchus keta orchinook salmon) or even some coho salmon stocksin more northern areas (Sandercock 199l), whichhave significant numbers of spawners returningover a wider range of ages. The southernmost cohosalmon populations in Waddell and Scott Creekseach have a 3-year cycle (J. Smith, App.); a strongrun occurs only every third year, and decimationof the spawning stock by overharvest during a sin-gle year could extirpate either or both of thosepopulations.

Climatic factors

Oceanic conditions. -The decline of coho salm-on in California has probably been exacerbated bynatural climatic events. A general warming trendin the northeast Pacific during 1976-1983 coin-cides with an abrupt drop in production of cohosalmon adults within the Oregon Production Area(Washington-northern California coast) and alsowith elevated sea-surface temperatures and re-duced biological productivity in the CaliforniaCurrent (Nickelson 1986; Lawson 1993). In ad-dition, El Niiio events have made oceanic con-ditions less favorable for coho salmon survival.The 1982-1983 El Nifio was the largest warmingevent on the North American Pacific Coast in thiscentury, and it had major impacts on primary andsecondary productivity and on coho salmon abun-dance off Oregon and Washington (Pearcy 1992).Coho salmon growth and survival were both af-fected to an unprecedented degree; perhaps 58%of the predicted number of coho salmon adults offOregon died during their last year in the ocean(Pearcy 1992). Significant statistical relationshipsof coho salmon abundance and ocean survival ratewith sea-surface temperature and strength of coastalupwelling have been identified for the Oregon Pro-duction Area (Scamecchia 198 1; Nickelson 1986;Pearcy 1992). Similarly, there is a positive cor-relation between upwelling and the survival andearly growth of coho salmon off northwest Van-couver Island, British Columbia (Holtby et al.1990). It has been suggested that upwelling affectscoho salmon survival by influencing coastal pro-ductivity and by transporting coho salmon smolts

offshore to areas where there are lower levels ofpredation (Nickelson 1986). Upwelling was weakoff the Oregon coast during the period 19 76 - 19 8 1,and the winds that drive upwelling off northernCalifornia were reduced between 1975 and 1988(Pearcy 1992). The time of transition between win-ter downwelling and spring upwelling conditionsalso may affect smolt survival, and on the Oregoncoast this transition time occurred earlier duringthe years of high coho salmon survival before 1976than in the low-survival years that followed (Pear-cy 1992). It is likely that unfavorable ocean con-ditions may persist for some time (Ware andThompson 199 l), which increases the importanceof improving coho salmon survival rates in fresh-water (Lawson 1993).

Precipitation. -The droughts of 1976-l 977 and1986-l 992 have clearly made conditions worse inmany streams, some of which completely driedup. On the other hand, heavy precipitation pro-duces stream scouring that can destroy both reddsand stream habitat for coho salmon. The effectsof the record 1964 floods on north coast streamscan still be seen in the streambeds and in the re-duced amount of high-quality habitat (W. Jones,App.), because natural recovery has been occur-ring very slowly in some drainages (Haskins 1982;Lisle 1982). Coho salmon in California undoubt-edly experienced catastrophic natural events in thedistant past-perhaps worse than those we reporthere- but they were not simultaneously confront-ed with widespread human-related degradation oftheir spawning streams (e.g., water diversions andincreased erosion).

Management

Most of the problems facing California cohosalmon populations have been well recognized formany years. One reason that little has been doneto remedy the situation seems to be that histori-cally coho salmon have been of secondary impor-tance to California fisheries. They presently rankbelow chinook salmon in the commercial fisheryand below both chinook salmon and steelhead inthe sport fishery. Coho salmon also are a verydiffuse resource, using a wide range of streamsalong the coast. As a result, management effortshave focused on chinook salmon and steelhead,apparently with the assumption that coho salmonwould be aided incidentally, Their wide distri-bution over many small coastal streams and theirrelatively short life span (3 years) make Californiacoho salmon more vulnerable to severe declinesduring drought years and, therefore, difficult to

manage iis that jupool habtection 0grazing, :water terimentatineeded 1changesmany yeexpensi

In 19the SalFisheriethat diiand Gamous sof thehas mafor retpopulaThe erand it-rwill ccerativcwhereare ap1996

(1) alridtc

(2) aSf

(3) t114

(4)

(5)

(6)

7

rnccoim


manage in a consistent manner. Another problemis that juvenile coho salmon require deep, cold,pool habitats for high survival. Inadequate pro-tection of watersheds from the effects of logging,grazing, and urbanization has resulted in increasedwater temperatures, the loss of pools through sed-imentation, and the removal of large woody debrisneeded for cover and pool formation. Once thesechanges have occurred, natural recovery can takemany years, and stream rehabilitation projects areexpensive and time consuming.

In 1988 the California State Legislature passedthe Salmon, Steelhead Trout and AnadromousFisheries Program Act (originally Senate Bill 226 1)that directed the California Department of Fishand Game to double the populations of anadro-mous species, including coho salmon, by the endof the century. The lack of accurate census datahas made it difficult to set specific numerical goalsfor recovery, but efforts are under way to increasepopulations as much as possible (T. Curtis, App.).The emphasis of these efforts is on the restorationand improvement of habitat. Hatchery productionwill continue at current levels, and private coop-erative fish-rearing projects will be encouragedwhere short-term, localized enhancement effortsare appropriate. Specific goals for the years 1992-1996 include (T. Curtis, App.):

(1) an inventory of streams within the historicrange of coho salmon to determine the presentdistribution and abundance of the species andto assess the condition of the habitat;

(2) a set of priorities for the improvement of cohosalmon streams on the basis of their potentialfor improvement;

(3) the identification of streams with the highestpotential for restoration and enhancement bythe Department of Fish and Game and ofstreams suitable for restoration and enhance-ment by private organizations;

(4) a set of priorities for restocking streams af-fected by droughts to speed recovery of thepopulation;

(5) the funding and completion of habitat resto-ration projects;

(6) the restocking of coho salmon streams ac-cording to priorities and in keeping with thedepartment’s genetic stock management pol-icy.

We would add to this outline a strong recom-mendation that hatchery production be carried outconservatively in order to minimize the negativeimpacts on native stocks (Steward and Bjomn

1990) and that efforts be made to increase the useof native coho salmon strains in hatcheries. Wewould also recommend a ban on all fisheries forcoho salmon until it can be demonstrated thatsignificant recovery is occurring. Furthermore, theaddition of a monitoring component seems nec-essary, for without a baseline it will be difficult todetermine the success or failure of enhancementand restoration efforts. A monitoring programshould include: (1) annual population surveys ofboth juvenile and spawner abundance in selectedstreams throughout the range of coho salmon, in-cluding streams both with and without ongoingmanagement efforts; (2) quick assessments of allhistoric coho salmon streams in the state at leastonce every 5 years to rate stream conditions andto determine if juveniles are present; (3) furthergenetic analysis of all stocks; (4) the marking ofall hatchery fish; and (5) evaluation of wild andhatchery populations for the presence of debili-tating diseases.

There should be greater coordination betweenthe different agencies and organizations concernedwith coho salmon, both within California and else-where, in order to prevent duplication of effortand failure. We urge that restoration goals be fo-cused on native coho salmon, and that hatcherystocks not be counted toward whatever numericalgoals are set.

Coho salmon statewide qualify for listing as athreatened species under state law, and certainpopulations may qualify under federal law forthreatened or endangered status. The southern-most coho salmon populations in Waddell andScott Creeks warrant special attention and qualifyfor federal listing as endangered. These popula-tions are extremely small and could be extirpatedeven by random demographic processes before afull evaluation of their biological distinctivenesscan be made. The severity of the coho salmondecline in California, as well as in Oregon andWashington, has led to the filing (October 1993)of a petition to the National Marine Fisheries Ser-vice to list the coho salmon as a threatened speciesunder the federal Endangered Species Act.

Conclusion

Despite the paucity of detailed quantitative dataon coho salmon stocks in California, it is clearfrom the information available that coho salmonpopulations statewide have undergone a dramaticdecline from historic levels. As a specific local ex-ample, the South Fork Eel River historically sup-ported a run of 5,000 - 25,000 coho salmon (Mur-

256 BROWN ET AL.

phy 1952) but now supports, at best, a run of about1,300. Our abundance estimates are optimistic be-cause we assumed coho salmon still occur instreams for which there are no current data; it islikely, therefore, that we have underestimated themagnitude of the decline.

The decline of coho salmon in California hasresulted from the combination of environmentaldegradation, climatic and ocean factors, and lackof comprehensive management focused on thisspecies. The absence of solid, long-term data onthe status of coho salmon populations is a reflec-tion of this lack of focused management. Of par-ticular concern is the fact that the decline has con-tinued despite substantial hatchery programs toraise coho salmon at five facilities in Californiaand despite several private supplementation pro-jects. Coho salmon, with their l-year freshwaterresidency as juveniles, are highly suited for hatch-ery augmentation of natural populations. The fail-ure of these hatchery programs to halt the declineis strong testimony that more effective manage-ment and conservation efforts must be pursued.Indeed, hatchery practices probably have contrib-uted to the decline of native stocks. The situationdemands immediate action, using the protectionavailable under state and federal endangered spe-cies laws, if necessary. Whether or not coho salm-on are formally listed under these laws, every na-tive coho salmon population in California shouldbe treated as a threatened species as part of a state-wide restoration effort.

Acknowledgments

We thank the many interested parties who con-tributed ideas, observations, manuscripts, and un-published data. Funding in support of this workwas provided by the National Marine FisheriesService; the Hewlett Foundation, through the Pub-lic Service Research and Dissemination Programat the University of California, Davis; and the

Mead Foundation.

References

Bartley, D. M., B. Bentley, and P. G. Olin. 1992. Pop-ulation genetic structure of coho salmon, Onco-rhynchus kisutch, in California. California Fish andGame 78:88-104.

Bartley, D. M., and G. A. E. Gall. 1990. Genetic struc-ture and gene flow in chinook salmon populationsof California. Transactions of the American Fish-eries Society 119:55-71.

Bartley, D. M., G. A. E. Gall, and B. Bentley. 1990.Biochemical genetic detection of natural and arti-ficial hybridization of chinook and coho salmon in

northern California. Transactions of the AmericanFisheries Society 119:43 l-437.

Bedell, G. W. 1991. Annual report. Trinity River salm-on and steelhead hatchery, 1989-1990. CaliforniaDepartment of Fish and Game, Inland Fisheries,Administrative Report 9l-8, Sacramento.

Berg, W. J., and G. A. E. Gall. 1988. Gene flow andgenetic differentiation among California coastalrainbow trout populations. Canadian Journal ofFisheries and Aquatic Sciences 45: 122-l31.

Bryant, H. C. 1923. Salmon fish cultural operations onthe Klamath River. California Fish and Game 9: 19-23.

FIe

FIe

Fr:

Bums, J. W. 1972. Some effects of logging and asso-ciated road construction on northern Californiastreams. Transactions of the American Fisheries So-ciety l0l:l-17.

Ft-i

California Advisory Committee on Salmon and Steel-head Trout. 1988. Restoring the balance, 1988annual report. Annual report to California Depart-ment of Fish and Game and the California Legis-lature, Sacramento.

Fr:

Clarke, W. C., R. E. Withler, and J. E. Shelboum.Fr

1992.Genetic control of juvenile life history pattern inchinook salmon (Oncorhynchus tshawytscha). Ca-nadian Journal of Fisheries and Aquatic Sciences49:2300-2306.

Gi

Cooney, C. X., and S. E. Jacobs. 1992. Oregon coastalsalmon spawning surveys, 1990. Oregon Depart-ment of Fish and Wildlife, Information Report 92-2, Portland.

Coots, M. 1957. Shasta River, Siskiyou County, 1955king salmon count and some notes on the 1956 run.California Department of Fish and Game, InlandFisheries, Administrative Report 57-5, Sacramento.

Coots, M. 1958a. Shasta River salmon count, 1957.California Department of Fish and Game, InlandFisheries, Administrative Report 58-4, Sacramento.

Coots, M. 1958b. Klamath River 1956 king salmoncount, Klamathon Racks, Siskiyou County. Cali-fornia Department of Fish and Game, Inland Fish-eries, Administrative Report 58-9, Sacramento.

Coots, M. 1973. A study ofjuvenile steelhead, Salmogairdnerii gairdnerii Richardson, in San GregorioCreek and lagoon, San Mateo County, March throughAugust, 197 1. California Department of Fish andGame, Anadromous Fisheries Branch, Administra-tive Report 73-4. Sacramento.

Dill, L. M., and A. H. G. Fraser. 1984. Risk of pre-dation and the feeding behavior of juvenile cohosalmon (Oncorhynchus kisutch). Behavioral Ecologyand Sociobiology 16:65- 71.

Gi

Gl

Gl

G

H

H

Eigenmann, C. H. 1890. The food fishes of the Cali-fornia fresh waters. Pages 53-65 in J. Routier andJ. D. Harvey, editors. Biennial report of the StateBoard of Fish Commissioners of the State of Cali-fomia for the years 1888-l890. State Board of FishCommissioners, Sacramento, California.

Fisk, L., E. Gerstung, and J. Thomas. 1966. Streamdamage surveys - 1966. California Department ofFish and Game, Inland Fisheries, Administrative

H

b

Report 66- 10, Sacramento.


Fleming, I. A., and M. R. Gross. 1989. Evolution ofadult female life history and morphology in a Pacificsalmon (coho: Oncorhynchus kisutch). Evolution 43:141-157.

Fleming, I. A., and M. R. Gross. 1992. Reproductivebehavior of hatchery and wild coho salmon (On-corhynchus kisutch): does it differ? Aquaculture 103:101-121.

Fraser, F. J. 1969. Population density effects on sur-vival and growth of juvenile coho salmon and steel-head trout in experimental stream channels. Pages253-268 in T. G. Northcote, editor. Symposium onsalmon and trout in streams. H. R. MacMillan Lec-tures in Fisheries, University of British Columbia,Vancouver.

Frissell, C. A. 1993. Topology of extinction and en-dangerment of native fishes in the Pacific Northwestand California (U.S.A.). Conservation Biology 7:342-354.

Fry, D. H., Jr. 1936. Life history of Hesperoleucusvenustus Snyder. California Fish and Game 22:65-98.

Fry, D. H., Jr. 1973. Anadromous fishes of California.California Department of Fish and Game, Sacra-mento.

Gibbs, E. D., and J. B. Kimsey. 1955. The 1951 creelcensus on the boat fishery of the Klamath Riverestuary, Del Norte County. California Departmentof Fish and Game, Inland Fisheries, AdministrativeReport 55-16, Sacramento.

Giddings, A. E. 1959. Record silver salmon taken inPapermill Creek, Marin County. California Fish andGame 45:353.

Grass, A. 1990a. Annual report, Van Arsdale fisheriesstation, 1987-88. California Department of Fish andGame, Inland Fisheries, Administrative Report 90-5, Sacramento.

Grass, A. 1990b. Annual report, Noyo River egg col-lecting station, 1987-1988. California Departmentof Fish and Game, Inland Fisheries, AdministrativeReport 90-3, Sacramento.

Graves, D. S., and J. W. Bums. 1970. Comparison ofyields of downstream migrant salmonids before andafter logging road construction on the South ForkCaspar Creek, Mendocino County. California De-partment of Fish and Game, Inland Fisheries, Ad-ministrat ive Report 70-3, Sacramento.

Hallock, R. J., and D. H. Fry, Jr. 1967. Five speciesof salmon, Oncorhynchus, in the Sacramento River,California. California Fish and Game 53:5-22.

Hallock, R. J., G. H. Warner, and D. H. Fry, Jr. 1952.California’s part in a three-state salmon fingerlingmarking program. California Fish and Game 38:301-332.

Haskins, D. M. 1982. Effects of valley inner gorge masswasting through time in the South Fork Trinity Riv-er, California. Pages 19-26 in K. A. Hashagen, ed-itor. Habitat disturbance and recovery. Proceedingsof a symposium. California Trout, San Francisco.

Hassler, T. J. 1987. Species profiles: life histories andenvironmental requirements of coastal fishes andinvertebrates (Pacific Northwest)-coho salmon.

U.S. Fish and Wildlife Service Biological Report82(11.70). (Also: U.S. Army Corps of Engineers, TREL-82-4, Washington, D.C.)

H&stein, T., and T. Linstad. 1991. Diseases in wild andcultured salmon: possible interaction. Aquaculture98:277-288.

Healey, T. P., Jr. 1973. Studies of steelhead and salmonemigration in the Trinity River. California Depart-ment of Fish and Game, Anadromous FisheriesBranch, Adminis t ra t ive Report 73- 1, Sacramento.

Hedrick, R. P. , S. E. LaPatra, J. L. Fryer, T. McDowell,and W. H. Wingfield. 1987. Susceptibility of coho(Oncorhynchus kisutch) and chinook (0. tshawyts-cha) salmon hybrids to experimental infection withinfectious hematopoietic necrosis virus (IHNV).Bulletin of European Association of Fish Patholo-gists 7:97-100.

Hemmingsen, A. R., R. A. Holt, R. D. Ewing, and J. D.McIntyre. 1986. Susceptibility of progeny fromcrosses among three stocks of coho salmon to in-fection by Ceratomyxa shasta. Transactions of theAmerican Fisheries Society 115:492-495.

Hindar, K., N. Ryman, and F. Utter. 1991. Geneticeffects of cultured fish on natural fish populations.Canadian Journal of Fisheries and Aquatic Sciences48:945-957.

Hiser, C. A. 1991. Annual report, Iron Gate salmonand steelhead hatchery, 1989-l990. California De-partment of Fish and Game, Anadromous FisheriesBranch, Administrative Report 9l-7, Sacramento.

Hjort, R. C., and C. B. Schreck. 1982. Phenotypic dif-ferences among stocks of hatchery and wild cohosalmon, Oncorhynchus kisutch, in Oregon, Wash-ington, and California. U.S. National Marine Fish-eries Service Fishery Bulletin 80: 105-119.

Holtby, L. B., B. C. Andersen, and R. K. Kadowaki.1990. Importance of smolt size and early oceangrowth to interannual variability in marine survivalof coho salmon (Oncorhynchus kisutch). CanadianJournal of Fisheries and Aquatic Sciences 47:2 18l-2194.

Holman, G., and W. A. Evans. 1964. Stream clearanceproject-completion report, Noyo River, Mendo-cino County. California Department of Fish andGame, Inland Fisheries, Administrative Report 64-10, Sacramento.

Jacobs, S. E., and C. X. Cooney. 1991. Improvementof methods used to estimate the spawning escape-ment of Oregon natural coho salmon. Fish ResearchProject, Annual Progress Report (October 199O-Oc-tober 1991) to Oregon Department of Fish andWildlife, Portland.

Jensen, P. T. 197 1. Fishery contribution and escape-ment of hatchery-produced silver salmon in Cali-fornia. California Department of Fish and Game,Anadromous Fisheries Branch, Administrative Re-port 7 1 - 10, Sacramento.

Jochimsen, W. H. 197 1. Annual report, Nimbus salm-on and steelhead hatchery (15th year), 1969-1970fiscal year. California Department of Fish and Game,Anadromous Fisheries Branch, Administrative Re-port 7 l-6, Sacramento.

r 258 BROWN ET AL.

Jochimsen, W. H. 1973a. Annual report, Nimbussalmon and steelhead hatchery, 1970-1971 fiscalyear. California Department of Fish and Game,Anadromous Fisheries Branch, Administrative Re-port 73-6, Sacramento.

Jochimsen, W. H. 1973b. Annual report, Nimbussalmon and steelhead hatchery, 1971-1972 fiscalyear. California Department of Fish and Game,Anadromous Fisheries Branch, Administrative Re-port 73-8, Sacramento.

Jochimsen, W. H. 1974. Annual report , Nimbus salm-on and steelhead hatchery, 1972-1973 fiscal year.California Department of Fish and Game, Anad-romous Fisheries Branch, Administrative Report 74-8, Sacramento.

Jochimsen, W. H. 1976. Annual report , Nimbus salm-on and steelhead hatchery, 1973-1974 fiscal year.California Department of Fish and Game, Anad-romous Fisheries Branch, Administrative Report 76-4, Sacramento.

Jochimsen, W. H. 1978a. Annual report, Nimbussalmon and steelhead hatchery, 1974-1975 fiscalyear. California Department of Fish and Game,Anadromous Fisheries Branch, Administrative Re-port 78-9, Sacramento.

Jochimsen, W. H. 1978b. Annual report, Nimbussalmon and steelhead hatchery, 1975-1976 fiscalyear. California Department of Fish and Game,Anadromous Fisheries Branch, Administrative Re-port 78-10, Sacramento.

Jochimsen, W. H. 1978c. Annual report, Nimbussalmon and steelhead hatchery, 1976-1977 fiscalyear. California Department of Fish and Game,Anadromous Fisheries Branch, Administrative Re-port 78-11, Sacramento.

Johnson, O. W., T. A. Flagg, D. J. Maynard, G. B. Mil-ner, and F. W. Waknitz. 199 1. Status review forlower Columbia River coho salmon. NOAA (Na-tional Oceanic and Atmospheric Administration)Technical Memorandum NMFS (National MarineFisheries Service) F/NWC-202, Seattle.

Jordan, D. S., and P. L. Jouy. 188 1. Checklist of du-plicates of fishes from the Pacific Coast of NorthAmerica, distributed by the Smithsonian Institutionin behalf of the United States National Museum,1881. Proceedings of the United States NationalM u s e u m 1881:1-18.

Jordan, D. S., and C. H. Gilbert. 1881. Notes on thefishes of the Pacific Coast of the United States. Pro-ceedings of the United States National Museum1881:29-70.

Kimsey, J. B. 1952. Fish rescue and stream improve-ment work in the north coast area in 1951. Cali-fornia Department of Fish and Game, Bureau ofFish Conservat ion, Adminis t ra t ive Report , Sacra-mento.

Kimsey, J. B. 1953. Fish rescue and stream improve-ment work in the north coast area in 1952. Cali-fornia Department of Fish and Game, Inland Fish-eries , Adminis trat ive Report 53-9, Sacramento,

Konkel, G. W., and J. D. McIntyre. 1987. Trends inspawning populations of Pacific anadromous salmo-

nids. U.S. Fish and Wildlife Service, Fish and Wild-life Technical Report 9.

Krueger, C. C., and B. May. 199 1. Ecological and ge-netic effects of salmonid introductions in NorthAmerica. Canadian Journal of Fisheries and Aquat-ic Sciences 48 (Supplement 1):66-77.

Lawson, P. W. 1993. Cycles in ocean productivity,trends in habitat quality and the restoration of salm-on runs in Oregon. Fisheries (Bethesda) 18(8):6-10.

Leidy, R. A. 1984. Distribution and ecology of streamfishes in the San Francisco Bay drainage. Hilgardia52: 1-175.

Lisle, T. E. 1982. The recovery of stream channels innorth coastal California from recent large floods.Pages 3 l-4 1 in K. A. Hashagan, editor. Habitat dis-turbance and recovery. Proceedings of a sympo-sium. California Trout, San Francisco.

Lufkin, A., editor. 1991. California’s salmon and steel-head. The struggle to restore an imperiled resource.University of California Press, Berkeley.

MacIntyre, J. D. 1984. Differentiation of anadromoussalmonid stocks. Pages 9-15 in J. M. Walton andD. B. Houston, editors. Proceedings of the Olympicwild fish conference. Peninsula College, FisheriesTechnologies Program, Port Angeles, Washington.

McCormick, R. B. 1958. Observations of the sportfishery for salmon in tidewater of the Klamath Riv-er, 1954. California Department of Fish and Game,Inland Fisheries, Administrative Report 58-25, Sac-ramento.

Meffe, G. K. 1986. Conservation genetics and the man-agement of endangered fishes. Fisheries (Bethesda)11(1):14-23.

Miller, W. H., T. C. Coley, H. L. Burge, and T. T. Kis-anuki. 1990. Analysis of salmon and steelheadsupplementation: emphasis on unpublished reportsand present programs. Bonneville Power Admin-istration, Project 88- 100, Portland, Oregon.

Milner, G. B. In press. Isozyme variation of coho salmon(Oncorhynchus kisutch) and its potential to estimatestock compositions of mixed-stock fisheries. In Pro-ceedings of a workshop on coho salmon. U.S. Na-tional Marine Fisheries Service, Northwest Fish-eries Science Center, Seattle.

Moyle, P. B., J. E. Williams, and E. D. Wikramanayake.1989. Fish species of special concern of California.Report to the California Department of Fish andGame, Sacramento.

Murphy, G. I. 1952. An analysis of silver salmon countsat Benbow Dam, South Fork Eel River, California.California Fish and Game 38:105-l12.

Murphy, G. I . , and L. Shapovalov. 1951. Prel iminaryanalysis of northern California salmon and steel-head runs. California Fish and Game 37:497-507.

Murray, C. B., T. D. Beacham, and L. W. Bamer. 1993.Growth and survival of newly emerged and juvenilecoho salmon (Oncorhynchus kisutch) reared at dif-ferent salinities. Canadian Journal of Zoology 71:1230-1237.

NMFS (National Marine Fisheries Service). 1977-l993.Current fishery statistics. Fisheries of the UnitedStates, 1975... .1992. U.S. Department of Com-

N

N


merce, National Oceanic and Atmospheric Admin-istration, U.S. Department of Commerce, Wash-ington, D.C.

Nehlsen, W., J. E. Williams, and J. A. Lichatowich. 1991.Pacific salmon at the crossroads: stocks at risk fromCalifornia, Oregon, Idaho, and Washington. Fish-eries (Bethesda) 16(2):4-21.

Nickelson, T. E. 1986. Influences of upwelling, oceantemperature, and smolt abundance on marine sur-vival of coho salmon (Oncorhynchus kisutch) in theOregon Production Area. Canadian Journal of Fish-eries and Aquatic Sciences 43:527-535.

Nickelson, T. E., M. F. Solazzi, and S. L. Johnson. 1986.Use of hatchery coho salmon (Oncorhynchus kis-utch) presmolts to rebuild wild populations in Or-egon coastal streams. Canadian Journal of Fisheriesand Aquatic Sciences 43:2443-2449.

Pearcy, W. G. 1992. Ocean ecology of North Pacificsalmonids. University of Washington Press, Seattle.

Puckett, K. J., and L. M. Dill. 1985. The energetics offeeding territoriality in juvenile coho salmon (On-corhynchus kisutch). Behaviour 92:97-l11.

Ramsden, G. 1993. Annual report, Trinity River salm-on and steelhead hatchery, 1991-1992. CaliforniaDepartment of Fish and Game, Inland Fisheries,Administrative Report 93-3. Sacramento.

Reeves, G. H., and J. R., and Sedell. 1992. An eco-system approach to the conservation and manage-ment of freshwater habitat for anadromous salmo-nids in the Pacific Northwest. Transactions of theNorth American Wildlife and Natural ResourcesConference 57:408-4 15.

Ricker, W. E. 1972. Hereditary and environmental fac-tors affecting certain salmonid populations. Pages19-160 in R. C. Simon and P. A. Larkin, editors.The stock concept in Pacific salmon. H. R. Mac-Millan Lectures in Fisheries, University of BritishColumbia, Vancouver.

Rogers, D. W. 1973. King salmon (Oncorhynchustshawytscha) and silver salmon (Oncorhynchus kis-utch) spawning escapement and spawning habitat inthe upper Trinity River, 1970. California Depart-ment of Fish and Game, Anadromous FisheriesBranch, Administrative Report 73-10, Sacramento.

Rosenau, M. L., and J. D. McPhail. 1987. Inheriteddifference in agonistic behavior between two pop-ulations of coho salmon. Transactions of the Amer-ican Fisheries Society 116:646-654.

Sandercock, R. K. 1991. Life history of coho salmon(Oncorhynchus kisutch). Pages 395-445 in C. Grootand L. Margolis, editors. Pacific salmon life histo-ries. University of British Columbia Press, Vancou-ver.

Scamecchia, D. L. 198 1. Effects of streamflow and up-welling on yield of wild coho salmon (Oncorhynchuskisutch) in Oregon. Canadian Journal of Fisheriesand Aquatic Sciences 38:47 l-475.

Scamecchia, D. L., and H. H. Wagner. 1980. Contri-bution of wild and hatchery reared coho salmon,Oncorhynchus kisutch, to the Oregon ocean sportfishery. U.S. National Marine Fisheries ServiceFishery Bulletin 77:6 17-624.

Schlichting, D. 1974. Annual report, Feather Riversalmon and steelhead hatchery, 1970-1971. Cali-fornia Department of Fish and Game, AnadromousFisheries Branch, Administrative Report 74-11,Sacramento.

Shapovalov, L. 1940. Fish rescue work in the NorthCoast district in 1939. California Department ofFish and Game, Bureau of Fish Conservation Ad-ministrative Report 40-3, Sacramento.

Shapovalov, L. 1945. Fish rescue and stream improve-ment work in the North Coast area in 1944. Cali-fornia Department of Fish and Game, Bureau ofFish Conservation Administrative Report 45-7,Sacramento.

Shapovalov, L. 1949. Fish rescue and stream improve-ment work in the North Coast area in 1945. Cali-fornia Department of Fish and Game, Bureau ofFish Conservation Administrative Report 49-26,Sacramento.

Shapovalov, L., and A. C. Taft. 1954. The life historiesof the steelhead rainbow trout (Salmo gairdnerigairdneri) and silver salmon (Oncorhynchus kisutch)with special reference to Waddell Creek, California,and recommendations regarding their management.California Department of Fish and Game, Fish Bul-letin 98.

Sheehan, J. 1991. Snags and twists abound in managingsport fish. Outdoor California 52 (May-June): 15-25.

Skaala, 0., G. Dahle, K. E. Jorstad, and G. Nzvdal.1990. Interactions between natural and farmed fishpopulations: information from genetic markers.Journal of Fish Biology 36:449-460.

Skinner, J. E. 1962. An historical review of the fishand wildlife resources of the San Francisco Bay area.California Department of Fish and Game, WaterProjects Branch Report 1, Sacramento.

Snyder, J. 0. 1908. The fishes of the coastal streamsof Oregon and northern California. U.S. Bureau ofFisheries Bulletin 27: 153-189.

Snyder, J. 0. 1931. Salmon of the Klamath River,California. California Division of Fish and Game,Fish Bulletin 34.

Steward, C. R., and T. C. Bjomn. 1990. Supplemen-tation of salmon and steelhead stocks with hatcheryfish: a synthesis of published literature. BonnevillePower Administration, Technical Report 90- 1,Portland, Oregon.

Taylor, E. B. 1991. A review of local adaptation inSalmonidae, with particular reference to Pacific andAtlantic salmon. Aquaculture 98: 185-207.

Taylor, E. B., and J. D. McPhail. 1985a. Variation inbody morphology among British Columbia popu-lations of coho salmon, Oncorhynchus kisutch. Ca-nadian Journal of Fisheries and Aquatic Sciences42:2020-2028.

Taylor, E. B., and J. D. McPhail. 1985b. Variation inburst and prolonged swimming performance amongBritish Columbia populations of coho salmon, On-corhynchus kisutch. Canadian Journal of Fisheriesand Aquatic Sciences 42:2029-2033.

U.S. Commission of Fish and Fisheries. 1892. Report

2 6 0 BROWN ET AL.

of the Commissioner for 1888, Part XVI. U.S. Gov-ernment Printing Office, Washington, D.C.

Utter, F., W. E. Ames, and H. 0. Hodgins. 1970.Transferrin polymorphism in coho salmon (Onco-rhynchus kisutch). Journal of the Fisheries ResearchBoard of Canada 27:237l-2373.

Utter, F., D. Campton, S. Grant, G. Milner, J. Seeb, andL. Wishard. 1980. Population structures of indig-enous salmonid species of the Pacific Northwest.Pages 285-304 in W. J. McNeil and D. C. Him-sworth, editors. Salmonid ecosystems of the NorthPacific. Oregon State University Press, Corvallis.

Wahle, R. J., and R. E. Pearson. 1987. A listing ofPacific coast spawning streams and hatcheries pro-ducing chinook and coho salmon. NOAA (NationalOceanic and Atmospheric Administration) Tech-nical Memorandum NMFS (National Marine Fish-eries Service) F/NWC- 122, Seattle.

Waples, R. S. 1990a. Conservation genetics of Pacificsalmon. II. Effective population size and the rate ofloss of genetic variability. Journal of Heredity 81:267 -276 .

Waples, R. S. 1990b. Conservation genetics of Pacificsalmon. III. Estimating effective population size.Journal of Heredity 8 1:277-289.

Waples, R. S. 1991. Definition of “species” under the

endangered species act: application to Pacific salm-on. NOAA (National Oceanic and Atmospheric Ad-ministration) Technical Memorandum NMFS (Na-tional Marine Fisheries Service) F/NWC- 194,Seattle.

Waples, R. S., and D. J. Teel. 1990. Conservation ge-netics of Pacific salmon. I. Temporal changes inallele frequency. Conservation Biology 4: 144-156.

Ware, D.M., and R. E. Thomson. 1991. Link betweenlong-term variability in upwelling and fish produc-tion in the northeast Pacific Ocean. Canadian Jour-nal of Fisheries and Aquatic Sciences 48:2296-2306.

Wehrhahn, C. F., and R. Powell. 1987. Electrophoreticvariation, regional differences and gene flow in thecoho salmon (Oncorhynchus kisutch) of southernBritish Columbia. Canadian Journal of Fisheries andAquatic Sciences 44:822-83 1.

Wilderness Society. 1993. The living landscape, vol-ume 2. Pacific salmon and federal lands. The Wil-derness Society, Washington, D.C.

Winter, G. W., C. B. Schreck, and J. D. McIntyre. 1980.Resistance of different stocks and transferrin ge-notypes of coho salmon, Oncorhynchus kisutch, andsteelhead trout, Salmo gairdneri, to bacterial kidneydisease and vibriosis. U.S. National Marine Fish-eries Service Fishery Bulletin 77:795-802.

Mur

N i e

Pail

Pre

Sm

Sm

Appendix: Unpublished Documents and Personal Communications

The l i s t s that follow contain the sources of unpublished reports and documents citedin the text and the affiliations and addresses of persons who provided unpublishedinformation and personal communications.

Unpublished Reports and Documents

Baker, P., and F. Reynolds. 1986. Life history, habitatrequirements, and status of coho salmon in Cali-fornia. Report to the California Fish and GameCommission, Sacramento. 37 pages.

Brown, L. R., and P. B. Moyle. 199 la. Status of cohosalmon in California. Report to the National MarineFisheries Service, Terminal Island, California. 114pages.

Brown, L. R., and P. B. Moyle. 1991b. Eel River sur-vey: final report. Contract Report FG-7054 to theCalifornia Department of Fish and Game, Sacra-mento. 200 pages.

CDFG (California Department of Fish and Game). 19 8 7.Noyo River fisheries station coho salmon manage-ment plan. Yountville, California. 2 pages.

CDFG (California Department of Fish and Game).1990-1993. Unpublished data on file. CDFG, Eu-reka.

Engelmeyer, P. 1992. Oregon coastal natural coho-distinct populations at risk. Report to Tenmile CreekAssociation, Yachats, Oregon. 5 pages.

Fredericksen, Kamine and Associates. 1980. ProposedTrinity River basin fish and wildlife managementprogram, appendix C, final, volume 1. Fredericksen,Kamine and Associates, Sacramento, California.

Hassler, T. J., C. M. Sullivan, and G. R. Stem. 199 1.Distribution of coho salmon in California. AnnualReport to California Department of Fish and Game,Contract FG7292. Arcata, California. 24 pages.

Kelley, D. W. 199 1. September 24, 199 1, memoran-dum to B. Cox. California Department of Fish andGame, Sacramento.

Hull, D., J. Ellinwood, C. Toole, and T. Flannigan. 1989.Humboldt Fish Action Council, final report projectnumber C- 1538, October 1986 through August 1988.Report to Harvey Reading, Cal ifornia Departmentof Fish and Game, Sacramento. 27 pages.

Kisanuki, T., J. Polos, and D. Wills. 1991. Annualreport, Klamath River fisheries assessment program1989. U.S. Fish and Wildlife Service, Report AFFI-FRO-9 1- 14, Arcata, California.

Klamath River Fishery Management Council. 1991.The Klamath Fishery Management Council s trate-gic plan for management of harvest of anadromousfish populations of the Klamath River basin. Publicreview draft prepared by the Klamath River FisheryManagement Council, Yreka, California. 45 pages.

Marston, D. 1992. June-July 1992 stream survey re-port of lower Scott Creek, Santa Cruz County. Cal-ifornia Department of Fish and Game, Sacramento.53 pages.

Sll-

Ste

Ta

Tt

UI

U


Murphy, G. I., and J. W. De Witt Jr. 1951. Notes onthe fishes and fishery of the lower Eel River, Hum-boldt County, California. Report to Bureau of FishConservation, California Division of Fish and Game,Sacramento.

Nielsen, J. L., M. Maas, and G. Balding. 1991. Anad-romous salmonid resources of Mendocino coastaland inland rivers, 1989-l990. An evaluation of re-habilitation efforts based on carcass recovery andspawning activity. Draft Report to California De-partment of Fish and Game, Inland Habitat Sur-veys, Standard Agreement FG9364, Sacramento. 98pages.

Painter, R. E., L. H. Wixom, and S. N. Taylor. 1977.An evaluation of fish populations and fisheries inthe post-Oroville project Feather River. Report byCalifornia Department of Fish and Game to theCalifornia Department of Water Resources, Sacra-mento.

Preston, L. 1988-1989. Unpublished data on file. Cal-ifornia Department of Fish and Game, Eureka.

Smith, G. E. 1986. Instream flow requirements, anad-romous salmonids spawning and rearing, LagunitasCreek, Marin County. California Department of Fishand Game, Environmental Services Branch, StreamEvaluation Report 86-2, Sacramento.

Smith, J. J. 1992a. Distribution and abundance of ju-venile coho salmon and steelhead in Waddell, Scottand Gazos Creeks in 1992. San Jose State Univer-sity, Unpublished report. San Jose, California. 9pages.

Smith, J. J. 1992b. Summary of trapping results onWaddell Creek for 199l - l992. Unpubl. report. SanJose State University, San Jose, California. 11 pages.

Steiner Environmental Consultants. 1990. Potter Val-ley project monitoring program (FERC No. 77, Ar-ticle 39): effects of operations on upper Eel Riveranadromous salmonids, 1988-l989. Progress re-port. Steiner Environmental Consulting, Potter Val-ley, California. 300 pages.

Taylor, T. 1991. November 15, 1991, memorandumto B. Getty. California Department of Parks andRecreation, Sacramento.

Tuss, C. J. Larson, T. Kisanuki, J. Polos, and J. Craig.1989. Annual report, Klamath River fisheries as-sessment program. U.S. Fish and Wildlife Service,Report AFF/FAO-89-13, Arcata, California.

USBLM (U.S. Bureau of Reclamation). 1973. Fisheryimprovement study, Eureka Division north coastproject, California. U.S. Department of the Interior,Bureau of Reclamation, Mid-Pacific Region, Sac-ramento, California. 44 pages + appendix.

USFWS (U.S. Fish and Wildlife Service). 1979. Klam-ath River fisheries investigations: progress, prob-lems and prospects. USFWS, Annual Report, Ar-cata, California. 49 pages.

U.S. Heritage Conservation and Recreation Service.1980. Final environmental impact statement. Pro-posed designation of five California rivers in thenational wild and scenic rivers system, volume 1.U.S. Department of the Interior, U.S. Heritage Con-servation and Recreation Service, Washington, D.C.322 pages.

Waldvogel, J. 1988. Fall chinook salmon spawningescapement estimate for a tributary of the SmithRiver, California, 2nd interim report (1980-l 987).University of California, Sea Grant Extension Pro-gram 88-5, Crescent City, California. 21 pages.

Personal Communications

Anderson, David G. Redwood National Park, Arcata,California.

Cox, Bill. California Department of Fish and Game,Sebastopol, California.

Cronin, Leo. Trout Unlimited, Corte Madera Lagun-itas Creek Restoration, Fairfax, California.

Curtis, Tim. California Department of Fish and Game,Sacramento, California.

Flosi, Gary. California Department Fish and Game,Weott, California.

Gerstung, Eric. California Department of Fish andGame, Sacramento, California.

Jones , Weldon. California Department of Fish andGame, Ukiah, California.

Kelley, Don W. D. W. Kelley and Associates, Post Of-fice Box 634, Newcastle, California.

Kisanuki, Tom. U.S. Fish and Wildlife Service, Arcata,California.

Lifton, Wayne. Entrix, Inc., 590 Ygnacio Valley Road,Walnut Creek, California.

Nielsen, Jennifer. Hopkins Marine Laboratory, Stan-ford University, Department of Biological Sciences,Pacific Grove, California.

Osis, Laimons. Oregon Department of Fish and Wild-life, South Beach, Oregon.

Petersen, Gary. The Mattole Restoration Council, Pe-trolia, California.

Poe, Sidney. California Department of Fish and Game,Yountville, California.

Preston, Larry. California Department of Fish andGame, Eureka, California.

Sanders, Steven. Prairie Creek Fish Hatchery, Orick,California.

Smith, Jerry J. Department of Biology, California StateUniversity, San Jose, California.

Streig, Dave. Monterey Bay Salmon and Trout Project,324 Swanton Road, Davenport, California.

Taylor, Thomas. California Department of Parks andRecreation, Sacramento, California.

West, John R. U.S. Forest Service, Yreka, California.