Wild chinook salmon survive better than hatchery salmon in a period of poor production R. J. Beamish & R. M. Sweeting & C. M. Neville & K. L. Lange & T. D. Beacham & D. Preikshot Received: 21 August 2010 / Accepted: 8 March 2011 / Published online: 6 April 2011 # The Author(s) 2011. This article is published with open access at Springerlink.com Abstract The population dynamics of chinook salmon (Oncorhynchus tshawytscha) from the Cowichan River on Vancouver Island, British Columbia, Canada are used by the Pacific Salmon Commission as an index of the general state of chinook salmon coast wide. In recent years the production declined to very low levels despite the use of a hatchery that was intended to increase production by improving the number of smolts entering the ocean. In 2008, we carried out an extensive study of the early marine survival of the hatchery and wild juvenile chinook salmon. We found that both rearing types mostly remained within the Gulf Islands study area during the period when most of the marine mortality occurred for the hatchery fish. By mid September, approximately 1.3% of all hatchery fish survived, compared to 7.8%–31.5% for wild fish. This six to 24 times difference in survival could negate an estimated increased egg-to-smolt survival of about 13% that is theorized to result through the use of a hatchery. Estimates of the early marine survival are approximate, but sufficient to show a dramatic difference in the response of the two rearing types to the marine nursery area. If the declining trend in production continues for both rearing types, modifications to the hatchery program are needed to improve survival or an emphasis on improving the abundances of wild stocks is necessary, or both. The discovery that the juvenile Cowichan River chinook salmon remain within a relatively confined area of the Gulf Islands within the Strait of Georgia offers an excellent opportunity to research the mechanisms that cause the early marine mortalities and hopefully contrib- ute to a management that improves the production. Keywords Chinook salmon . Salmon hatchery . Early marine survival . Strait of Georgia Introduction The Strait of Georgia, located between Vancouver Island and the British Columbia mainland is a major nursery area for Pacific salmon (Oncorhynchus spp., Fig. 1). Historically, about 35% to 40% of the commercial and recreational catch of Pacific salmon in British Columbia reared as juveniles in the Strait of Georgia (Beamish and Neville 1999). The Strait of Georgia ecosystem is changing, as indicated by an increasing temperature of about 1°C in the past 50 years (Fig. 2) and by the changing composition of the major species of Pacific salmon. In recent years, the marine survival of coho (O. kisutch) and chinook (O. tshawytscha) salmon dropped to the lowest levels in recorded history (Beamish et al. 1995, 2008, 2010; Beamish and Neville 1999) while the abundances of pink (O. gorbuscha) and chum (O. keta) salmon are close to historic high levels Environ Biol Fish (2012) 94:135–148 DOI 10.1007/s10641-011-9783-5 R. J. Beamish (*) : R. M. Sweeting : C. M. Neville : K. L. Lange : T. D. Beacham : D. Preikshot Fisheries & Oceans Canada, Pacific Biological Station, 3190 Hammond Bay Road, Nanaimo, BC V9T 6N7, Canada e-mail: [email protected]
Transcript
Wild chinook salmon survive better than hatchery salmonin a period of poor production
R. J. Beamish & R. M. Sweeting & C. M. Neville &
K. L. Lange & T. D. Beacham & D. Preikshot
Received: 21 August 2010 /Accepted: 8 March 2011 /Published online: 6 April 2011# The Author(s) 2011. This article is published with open access at Springerlink.com
Abstract The population dynamics of chinook salmon(Oncorhynchus tshawytscha) from the Cowichan Riveron Vancouver Island, British Columbia, Canada areused by the Pacific Salmon Commission as an index ofthe general state of chinook salmon coast wide. In recentyears the production declined to very low levels despitethe use of a hatchery that was intended to increaseproduction by improving the number of smolts enteringthe ocean. In 2008, we carried out an extensive study ofthe early marine survival of the hatchery and wildjuvenile chinook salmon. We found that both rearingtypes mostly remained within the Gulf Islands studyarea during the period when most of the marinemortality occurred for the hatchery fish. By midSeptember, approximately 1.3% of all hatchery fishsurvived, compared to 7.8%–31.5% for wild fish. Thissix to 24 times difference in survival could negate anestimated increased egg-to-smolt survival of about 13%that is theorized to result through the use of a hatchery.Estimates of the early marine survival are approximate,but sufficient to show a dramatic difference in theresponse of the two rearing types to the marine nurseryarea. If the declining trend in production continues forboth rearing types, modifications to the hatcheryprogram are needed to improve survival or an emphasis
on improving the abundances of wild stocks is necessary,or both. The discovery that the juvenile Cowichan Riverchinook salmon remain within a relatively confined areaof the Gulf Islands within the Strait of Georgia offers anexcellent opportunity to research the mechanisms thatcause the early marine mortalities and hopefully contrib-ute to a management that improves the production.
The Strait of Georgia, located between VancouverIsland and the British Columbia mainland is a majornursery area for Pacific salmon (Oncorhynchus spp.,Fig. 1). Historically, about 35% to 40% of thecommercial and recreational catch of Pacific salmonin British Columbia reared as juveniles in the Strait ofGeorgia (Beamish and Neville 1999). The Strait ofGeorgia ecosystem is changing, as indicated by anincreasing temperature of about 1°C in the past50 years (Fig. 2) and by the changing compositionof the major species of Pacific salmon. In recentyears, the marine survival of coho (O. kisutch) andchinook (O. tshawytscha) salmon dropped to thelowest levels in recorded history (Beamish et al.1995, 2008, 2010; Beamish and Neville 1999) whilethe abundances of pink (O. gorbuscha) and chum(O. keta) salmon are close to historic high levels
Environ Biol Fish (2012) 94:135–148DOI 10.1007/s10641-011-9783-5
R. J. Beamish (*) : R. M. Sweeting : C. M. Neville :K. L. Lange : T. D. Beacham :D. PreikshotFisheries & Oceans Canada, Pacific Biological Station,3190 Hammond Bay Road,Nanaimo, BC V9T 6N7, Canadae-mail: [email protected]
(Beamish and Neville 1999; Beamish et al. 2007).Sockeye salmon (O. nerka) returns to the FraserRiver drainage declined since about 1994 and in2009 were the lowest ever recorded. These reasonsfor the recent poor returns are being investigated by aJudicial Inquiry (http://www.commissioncohen.ca/en/).
In the mid 1980s, Canada initiated a cooperativeprogram with the United States to reverse the declineof chinook salmon abundance (PSC 1987). All of thereasons for the decline were not understood, but itwas apparent that fishing had an important impact.The management of fishing required internationalcooperation with the United States and this wasaccomplished through the International Pacific SalmonCommission (PSC 1987). Chinook salmon from theCowichan River were selected as a population thatcould be used as an indicator of the health of chinooksalmon in general and the ability of the program toincrease the abundance of chinook salmon in particular.In 2000, the escapement target was recommended to be7400 (95% confidence intervals: 4185–18 915; Riddellet al. 2000). More recent escapement goals wereestimated to be 6500 and 6600 (Tompkins et al.2005; Parken et al. 2006). Escapements increased inthe early 1990s and generally exceeded the recom-mended target (Fig. 3). However, beginning in the late1990s, escapements declined to the very low abun-dance of 981 in 2008. The escapement estimates donot include chinook salmon that are removed fromthe river for brood stock for the hatchery. In 2008and 2009, the hatchery brood stock was 667 and612 fish, respectively. Part of the reason for the
recent decline in escapement is the declining marinesurvival (Fig. 4). This decline in survival is likelyrelated to ecosystem changes within the Strait ofGeorgia (Hinke et al. 2005; Beamish et al. 2008,2010; Beauchamp 2009).
In the late 1970s, the Department of Fisheries andOceans Canada established a program to use hatcheriesto increase the abundances of coho and chinook salmon(Fisheries and Environment Canada 1978). The basisfor the program was a belief that there was additionalcapacity within the Strait of Georgia to produce morePacific salmon and that hatcheries could add more fishto the ocean faster than could be added through amanaged natural production. A hatchery along theCowichan River started to produce chinook salmon in1979 using adults returning to the Cowichan River(Cross et al. 1991). The number of adults removedfrom the population ranged from 175 to 678 up to1990 (Fig. 5). Beginning in 1991, the number of adultsremoved from the spawning population to producejuveniles increased to an average of 1235 between1991 and 2008. In recent years, the number ofjuveniles released from the hatchery into the riveraveraged 1.9 million from 2000 to 2009 and rangedfrom 3.2 million to 0.5 million (Fig. 6). The hatcheryon the Cowichan River has not only been unable toincrease the abundance, it has also not been able tosustain the abundances that existed at the time theprogram started.
In this paper we report the results of an intensivestudy to compare the early marine survival ofhatchery and wild chinook salmon from the CowichanRiver in the Strait of Georgia. This report is restrictedto 2008 because all fish released from the hatchery inthis year could be readily identified as they received acoded wire tag (CWT) and had the adipose finclipped. We consider that all naturally spawning fishare wild. We use this terminology only for consistencyand do not consider that our use is the correct definitionof a wild Pacific salmon.
Methods
The trawl survey methodology and trawl net designare reported in Beamish et al. (2000) and Sweeting etal. (2003). The set locations within the Gulf Islands,Strait of Georgia and Juan de Fuca Strait are shown inFig. 7. Each set was approximately 30 min in length
Strait of Georgia
Juan de Fuca Strait
Gulf Islands U. S. A.
A
Fig. 1 Standard track lines (solid lines) for trawl surveys in theStrait of Georgia. Sets were evenly spaced along the track lines.Black box shows location of the Gulf Islands
and fished an average of about 4.3 km. Catch per uniteffort (CPUE) is the catch expanded to one hour.Most sets were at the surface, but sets were made withthe head rope at 14 and 29 m in the Gulf Islands anddeeper sets were included in the surveys in the Straitof Georgia. Abundance was determined using theprocedures in Beamish et al. (2000) and modified inBeamish et al. (2008). The procedures for the stratumvolumes are in Thomson and Foreman (1998). In theGulf Islands, we estimated a volume of 9.3 km3 forthe strata 0–14 m, 15–29 m and 30–44 m. The area isrelatively shallow compared of the open Strait ofGeorgia with about 35% deeper than 45 m. The totalvolume of water fished in each stratum was divided
Fig. 2 Average seasurface temperaturesfrom lighthouses in theStrait of Georgia from1960 to 2010 for a)winter (December toMarch) and b) Spring(April to June)
0
5
10
15
1975 1980 1985 1990 1995 2000 2005
Brood Year
Spa
wni
ng E
scap
emen
t (×
103)
Fig. 3 Spawning escapement of chinook salmon in theCowichan River, for brood years 1978–2008 or ocean entryyear 1979–2009 and return year 1982–2012. Data fromTompkins et al. (2005) and the Department of Fisheries andOceans (www.pac.dfo-mpo.gc.ca/gis-sig/maps-cartes-eng.htm)
into the total volume of water in the particular 15 mlayer to estimate the percentage of water fished. Anassumption in the calculation was that the catchabilityof fish by the net was 1.0 (Beamish et al. 2000). If thecatchability was less than 1.0, actual abundancewould be underestimated. Catches of juvenile chinooksalmon in the standard trawl surveys in the Strait ofGeorgia (Beamish et al. 2008, 2010) were used toshow that juvenile chinook salmon originating fromthe Cowichan River were rarely found in the Strait ofGeorgia in July and September.
A purse seine survey was conducted within theGulf Islands area from June 20–27, 2008. The 38 ft(11.7 m) vessel fished a 420 ft (129.2 m) purse seinethat was 60 ft (18.5 m) deep and had ¼ inch (6 mm)mesh in the bunt. Approximately 10 sets were
completed each day, with sets spaced throughout theGulf Islands, including Cowichan Bay (Fig. 7b, d).All fish were identified and up to 50 juvenile chinooksalmon per set were measured for fork length andchecked for the presence of a CWT. Otolith and scalesamples were taken from up to 10 fish per set. Asample of the operculum was taken for DNA analysisand preserved in 95% ethanol.
A beach seine survey was carried out from April8 to June 6 in the Cowichan River estuary (CowichanBay). Approximately seven sets were made each day,two times a week. Two teams fished concurrently onthe north and south side of Cowichan Bay. Themarking of all hatchery fish with a clipped adiposefine facilitated the recognition of hatchery and wildjuvenile chinook salmon. All species of Pacificsalmon were examined, but only the catches andlengths of chinook salmon are reported here.
Fish were sampled for DNA using pieces of theoperculum preserved in 95% ethanol or dried materialfrom around the otolith. Up to 50 fish were sampledfrom each set and the resulting samples were analyzedfor stock composition using the procedures describedin Beacham et al. (2006). Briefly summarized, 12microsatellites (Ots 102 not included) were analyzedfor all individuals in the samples. A baseline of 280populations ranging from the Alsek River in northernBritish Columbia to the Sacramento River in Californiawas used as the basis for estimating stock compositionof mixed-stock samples with cBayes (Neaves et al.2005). In the analysis, ten 20 000-iteration MonteCarlo Markov chains of estimated stock compositionswere produced, with initial starting values for each
0
1
2
3
1975 1980 1985 1990 1995 2000 2005 2010
Release year
Num
ber
of s
mol
ts (
×10
6 )
Fig. 6 Numbers of smolts released by Cowichan RiverHatchery, by release year 1980–2009. Data are from theDepartment of Fisheries and Oceans (www.pac.dfo-mpo.gc.ca/gis-sig/maps-cartes-eng.htm). Note that there were no data for2005.
0
2
4
6
1975 1980 1985 1990 1995 2000 2005 2010
Brood Year
Sur
viva
l (%
)
Fig. 4 Marine survival of Cowichan River chinook salmon, bybrood year 1985–2002, or ocean entry year 1986–2003 andreturn year 1989–2006, using coded wire tag (CWT) informa-tion from hatchery fish, reported in Tompkins et al. (2005).Note that there were no data for 1986 and 1987
0
1
2
1975 1980 1985 1990 1995 2000 2005 2010
Spawning year
Bro
od S
tock
(nu
mbe
r ×
103 )
Fig. 5 Number of chinook salmon in the Cowichan River usedas hatchery brood stock, 1981–2008. Data are from Tompkinset al. (2005) and the Department of Fisheries and Oceans(www.pac.dfo-mpo.gc.ca/gis-sig/maps-cartes-eng.htm)
chain set at 0.90 for a particular population that wasdifferent for each chain. Estimate stock compositionswere considered to have converged when the shrink
factor was <1.2 for the 10 chains (Pella and Masuda2001). The last 1000 iterations from each of the 10chains were then combined and for each fish the
Fig. 7 Survey and set locations for (a) Strait of Georgia (b) Gulf Islands (c) Juan de Fuca Strait and (d) Cowichan Bay. Black dotsindicate the locations of sets
Environ Biol Fish (2012) 94:135–148 139
probability of originating from each population in thebaseline was determined. These individual probabilitieswere summed over all fish in the sample anddivided by the number of fish sampled to providethe point estimate of stock composition. Standarddeviations of estimated stock compositions weredetermined from the last 1000 iterations from eachof the 10 chains incorporated in the analysis. Theaccuracy of the stock compositions was examinedby comparing the determinations to known stockcompositions using CWTs. We used all samplesfrom the July and September trawl surveys in theStrait of Georgia from 2007, 2008 and 2009 for thecomparison. The population identification resultsfrom DNA analysis from 2007 to 2009 werecompared to known identifications from CWTs for215 chinook salmon. Additionally, we evaluatedaccuracy of the estimated stock composition for asample of 133 coded-wire-tagged Cowichan Riverjuveniles captured and sampled during 2007–2009.
Strait of Georgia sea surface temperature (SST, °C)and salinity (SSS, ppm) were obtained from the OceanScience webpage maintained at the Institute of OceanSciences in Sidney, British Columbia (http://www.pac.dfo-mpo.gc.ca.science/oceans/data-donnees/lighthouses-phares.index-eng.html). Data are collected daily atlighthouse stations throughout the Strait of Georgiaand monthly averages are posted on the webpage.To obtain monthly values for the entire Strait ofGeorgia, we averaged the data from 1960 to presentfrom the following stations: Cape Mudge (1960–1985), Chrome Island, Departure Bay, Entrance Island,Sisters Island, and from West Vancouver (1980–1985).Cape Mudge and West Vancouver stations were closedin 1985.
The Cowichan River flows from Cowichan Lakeabout 50 km from Cowichan Bay in the Strait ofGeorgia (Fig. 7b). A fishway at a partial obstructionallows the passage of adult chinook salmon intoCowichan Lake, although most spawning occurs inthe river. Fall spawning chinook salmon return to theriver from mid August through to October, and insome years until November. Coho and chum salmonalso spawn in the river. The hatchery is about 2 kmfrom the estuary, however, the juvenile chinooksalmon were transported to a release location about25–30 km from the estuary. A small number wereheld in seapens and released into the estuary in lateMay and early July.
Results
DNA analysis
The population identification results from DNAanalysis from 2007 to 2009 were compared to knownidentifications from CWTs for 215 chinook salmon.The results from both the DNA analysis and the CWTidentification were grouped into the following eightcommon stock areas: Upper Fraser River, Lower FraserRivere, South Thompson River, North ThompsonRiver, East Coast Vancouver Island, West Strait ofGeorgia, Puget Sound, and Columbia River. Of the215 fish identified using DNA analysis, 204 (95%)came from the same stock area as the CWT. Sevenof the 11 fish that had disagreement between theCWT identification and the DNA stock allocationwere fish that were either identified from CWTs asfish from East Coast Vancouver Island but allocatedto Puget Sound from DNA analysis, or fish thatwere identified from CWTs as Puget Sound fish butallocated to East Coast Vancouver Island fromDNA analysis.
The stock composition of a sample of 133-CWTCowichan River individuals estimated with a 280-population baseline (Cowichan River was one popu-lation in the baseline) was estimated at 98% CowichanRiver origin. On an individual level, all but one of the133 individuals examined were assigned to CowichanRiver origin, with individual probability levels rangingfrom 0.71 to 1.00. We concluded that accurate identi-fication of the Cowichan River component of the catchof juvenile chinook salmon was achieved.
Juvenile Pacific salmon surveys
The purse seine survey in the Gulf Islands in late June2008 captured 186 juvenile chinook salmon through-out the Gulf Islands. The largest catches of 61chinook salmon occurred within Cowichan Bay.DNA analysis indicated that all except one of the 61fish were from the Cowichan River and CWTsindicated that 25 or 41% were from the hatchery.The sample of 115 chinook salmon from all otherareas consisted of 75.7% Cowichan River chinooksalmon of which 30.5% were from the hatchery(Table 1). A preliminary trawl survey on June 24–26caught 47 chinook salmon in 24 sets throughout theGulf Islands (Table 1). Only 12 fish were from the
Cowichan River, and 75.0% (9 of 12) of these werefrom the hatchery. The mid-July 2008 trawl surveywas about three weeks later than the June surveys andhad the largest catches of juvenile chinook salmon ofall Gulf Island surveys in 2008 (Table 1). Chinooksalmon from the Cowichan River comprised 66.2% ofthe catch, or 452 fish. We recovered 40 juveniles witha CWT from the Cowichan hatchery which indicatesthat the percentage of Cowichan hatchery fish in Julywas 8.8%. Catches declined in the mid September andearly October trawl surveys, but the percentage ofhatchery fish from the Cowichan River remainedabout the same at 8.3% and 11.8%, in September andOctober, respectively (Table 1). In the mid-July trawlsurvey, most juvenile chinook salmon were capturedin the top 30 m (Table 2). In September, catches in thedepth stratum from 30 m to 44 m (head rope depth of30 m) increased slightly (Table 2). However, by earlyOctober, juvenile chinook salmon were more evenlydistributed within the top 44 m (Table 2).
The beach seine study from April 8 to June 6captured 579 juvenile chinook salmon over 11 daysof sampling. Wild chinook salmon were captured ineach day while hatchery chinook salmon werecaptured for several days immediately after theirrelease from the hatchery (Table 3). Over the2 months of beach seining, hatchery fish represented45% of the catch of all juvenile chinook salmon.Hatchery fish were consistently larger than the wildfish (Table 4). In mid July, the average lengths of
Tab
le1
Catch
ofchinoo
ksalm
onof
Cow
ichanRiver
origin
intheGulfIsland
s,Straitof
Georgia
andJuan
deFucaStraitsurveysin
2008
Survey
Num
ber
ofsets
Catch
Num
ber
forDNA
analysis
Num
berof
Cow
ichanchinook
salm
onin
DNA
analysis
Num
berof
CWTs
from
Cow
ichan
River
hatchery
incatch
Percentageof
DNA
sample
from
Cow
ichan
River
Num
berof
Cow
ichan
River
chinook
salm
onin
catch,
usingDNA
%
Percentageof
Cow
ichan
River
chinooksalm
onfrom
hatchery
using
CWTsin
catch
Percentageof
wild
Cow
ichan
River
chinook
salm
on
GulfIslands
Purse
seineJune
20–27
69186
115
8743
75.7%
141
30.5%
69.5%
TrawlJune
24–26
2447
4411
925.0%
1275.0%
25.0%
TrawlJuly
16–17
18683
207
137
4066.2%
452
8.8%
91.2%
TrawlSeptember10
–12
20422
302
698
22.8%
968.3%
91.7%
TrawlOctober
3–5
21170
9720
420.0%
3411.8%
88.2%
Straitof
Georgia
TrawlJune
27–Jul6
901770
461
03
0%0
0%0%
TrawlSeptember13
–24
801825
261
40
1.5%
270%
0%
Juan
deFucaStrait
TrawlJuly
109
70
00
0%0
0%0%
TrawlSeptember28,30
18180
210
00%
00%
0%
Table 2 Catches, catch per unit effort (CPUE) for each depthstratum for the trawl survey, 2008
Date Depth Numberof sets
Catch CPUE
June 24–26 0–14 m 15 42 5.6
15–29 m 6 5 1.5
30–44 m 3 0 0
July 16–17 0–14 m 14 547 78.1
15–29 m 3 130 91.8
30–44 m 1 6 12.0
September 10–12 0–14 m 10 300 59.6
15–29 m 7 96 27.4
30–44 m 3 26 17.3
October 3–5 0–14 m 9 62 13.9
15–29 m 9 73 16.5
30–44 m 3 35 24.7
Environ Biol Fish (2012) 94:135–148 141
hatchery and wild fish had increased by 58 mm and19 mm, respectively. Hatchery fish continued to belarger than wild fish in all the samples, although thecatch of hatchery fish from the Cowichan hatcherywas small (Table 1).
The preliminary trawl survey captured very fewchinook salmon, indicating that most juveniles werestill in the nearshore areas. Therefore, we combinedthe two late June surveys to get an estimate of thehatchery and wild percentages in the catch. Theestimated percentage of hatchery fish, weighted forthe differences in the catches, was 36.1%. Theweighted estimate of 36.1% hatchery fish was usedto estimate the number of wild smolts that wereproduced in the Cowichan River in 2007–2008. Thehatchery released 460 000 chinook salmon smolts in2008. If these fish represented 36.1% of all Cowichan
River chinook salmon, there would be 814 200 wildchinook salmon smolts. Another estimate of wildsmolt production was made using the reportedescapement of 1860 adults, a sex ratio of 50% malesand females, an average fecundity of 3700 eggs andan egg-to-smolt survival of 6% (Bradford 1995,Tompkins et al. 2005). This calculation produced anestimate of 206 500 wild smolts. A third estimate ofwild smolt abundance was made using the hatcherypercentage of 45% in the beach seine study. Ifhatchery and wild chinook salmon smolts had asimilar mortality during the beach seine study, therewould have been 562 200 wild smolts produced.
The abundance estimates (Table 5) can be used toestimate the early marine survival of the hatchery andwild fish. Estimates of wild juvenile chinook salmonsurvival are a range produced using the threeestimates of wild smolt production. The early marinesurvival of hatchery fish from ocean entry to mid July,mid September and early October were 6.9%, 1.3%and 0.8%, respectively (Table 5). Wild fish had a28.2%–158.5%, 7.8%–31.5% and 3.6%–14.3% survivalfrom ocean entry to mid July, mid September andearly October, respectively.
The extensive surveys in the Strait of Georgia inJuly (Fig. 7) captured larger numbers of juvenilechinook salmon than in the Gulf Islands (Table 1).Samples used for the DNA analysis were distributedthroughout the Strait of Georgia (Fig. 8). None of the461 fish analyzed for DNA came from the CowichanRiver, however, there were three fish with a CWTfrom the Cowichan hatchery that were captured in theStrait of Georgia near a pass leading to the GulfIslands area. Similar large catches of juvenile chinooksalmon were obtained in the September trawl survey.Four (1.7%) wild chinook salmon from the CowichanRiver were captured (Table 1). Neither of the twosurveys in Juan de Fuca Strait in July or Septembercaptured any wild or hatchery chinook salmon from
Table 3 Catches of hatchery and wild juvenile chinook salmonfrom the Cowichan River in the beach seine study in 2008
Date Numberof sets
Number ofhatchery fish
Percentage ofhatchery fish
Number ofwild fish
April 8 7 0 0% 1
April 18 4 0 0% 5
May 6 10 9 90% 1
May 8 9 10 77% 3
May 13 6 2 67% 1
May 15 7 0 0% 2
May 20 8 0 0% 3
May 22 8 5 71% 2
May 27 7 8 67% 4
May 29 8 0 0% 30
June 6 11 225 46% 268
Total 259 45% 320
April 25, hatchery released 204 000 smolts
May 22, hatchery released 230 400 smolts
June 2, hatchery released 25 300 smolts
Date Hatchery Wild
Method Length (mm) ± SD Number Length (mm) ± SD Number
May 22–29 Beach seine 96±5.3 13 67±10.2 36
July 10–12 Trawl 148±21.8 14 94±18.5 94
September 10–12 Trawl 227±13.2 7 162±27.4 63
October 3–4 Trawl 282±35.8 3 205±17.7 17
Table 4 Mean length(± SD) of hatchery andwild juvenile chinooksalmon from the CowichanRiver in the Gulf Islandsarea of the Strait of Georgiain 2008
142 Environ Biol Fish (2012) 94:135–148
the Cowichan River, as indicated from the CWTs andfrom a sample of DNA (Table 1).
Hatchery fish were released with different codesfor the CWTs. The tagged fish were recovered in thevarious surveys in the Gulf Islands from June toNovember 2008. Fewer fish were recovered from theApril 25 release than the May 29 release in the river(Fig. 9). The rate of recapture varied among tag codeswith the most consistent rate occurring for the May 29release. No recaptures were made for the May 29release from the sea pen. Recapture rates for the July2 release from a sea pen ranged from the highest tolowest (Fig. 9). There were 204 000 fish released intothe river in April 2008 and 18 CWTs were recoveredin all surveys. There were 205 000 fish released intothe river in May 2008 and 68 CWTs were recovered.The rate of recapture of CWTs from the May 29release was 3.8 times larger than from the April 25release.
Spring (April to June) sea surface temperatures in theStrait of Georgia demonstrated a clear warming trend
since 1960 of approximately 0.28°C / decade (Fig. 2a).Winter (December to March) sea surface temperaturesshowed a similar albeit slightly lower trend of about0.16°C / decade (Fig. 2b). Summer (June to August)sea surface temperatures showed an even steeperincrease in warming, averaging 0.39°C / decade since1960, with average summer SSTs now exceeding 17°C.Within the long-term trend, there were periods ofcooling such as in the 1960s and early 1970s orfrom 2005 to 2008, but the general trend has beenan increase of about 1°C over the past 50 years.
Discussion
The DNA analysis and all surveys indicated that mostjuvenile hatchery and wild chinook salmon from theCowichan River remained within the Gulf Islandsuntil at least the end of the survey in September. Weare confident that our identifications of hatchery fishwere correct because all hatchery fish were tagged
Table 5 Abundance estimates from the trawl survey, and the number and percentage of hatchery and wild chinook salmon
Date Abundance±2 SD Number ofCowichan Riverchinook salmon
Fig. 8 The samplesanalyzed for DNA in theStrait of Georgia in the 2008June 27 to July 6 trawlsurvey showing thatsamples were distributedthroughout the Strait ofGeorgia
Environ Biol Fish (2012) 94:135–148 143
with a CWT. We are also confident that the DNAanalysis was reliable as the DNA results were similarto 95%–98% of the CWTs from a number of differentstocks, including a sample of 133 CWT fish from theCowichan River. During this period, the percentage ofhatchery fish that survived declined to an estimated1.3% in mid September and 0.8% in early October.There were three Cowichan hatchery fish captured inthe standard survey in July in the Strait of Georgianear the Gulf Islands. This indicates that someCowichan hatchery fish were leaving the survey areain July. However, catches of juvenile hatchery andwild chinook salmon from the Cowichan River wererare outside of the Gulf Islands area as indicated bythe DNA analysis. Even if the abundances outside ofthe Gulf Islands were equal to the abundances in oursurveys within the Gulf Islands, the estimated marinesurvival of the hatchery fish through to the fall wouldonly be about two percent. The recovery of CWTs inthe surveys within the Gulf Islands showed that thehighest mortalities were for the early releases from thehatchery. It is most likely that these mortalitiesoccurred soon after ocean entry and within the areaof the Gulf Islands close to the Cowichan estuary. Thevery poor early marine survival indicates that most ofthe marine mortality that affects the brood yearstrength of the Cowichan River hatchery chinooksalmon occurred within the Gulf Islands and withinabout 5 months of ocean residence.
Not all fish released from the hatchery into theriver will survive to enter the ocean, thus theestimated marine mortality includes some freshwatermortality. Perhaps, the most important error in theestimate of early marine survival of hatchery fish isour estimate of abundance. The confidence limits arelarge and we use a catchability of 1.0 which assumesthat all fish in front of the net opening are caught. Atrue catchability is probably lower, which wouldincrease the abundance and the marine survival.However, despite the imprecision of some estimates,it was clear that very few hatchery fish survivedthrough to early October. This interpretation issupported by an acoustic tagging study in the GulfIslands in mid July 2008 (Neville et al. 2010). In theacoustic tagging study 70 juvenile chinook salmonwere tagged and approximately 63 were hatchery andwild chinook salmon from the Cowichan River. Onlyone fish of the 70 (from the Big Qualicum hatchery)was detected leaving the Strait of Georgia and thiswas through Juan de Fuca Strait. The battery life ofthe acoustic tag was approximately 4 months whichwould last until at least mid November. The conclu-sion of the acoustic tagging study was none of theCowichan River fish were detected after they weretagged and they probably died within the Strait ofGeorgia.
Estimates of the early marine mortality of the wildchinook salmon juveniles were more problematic. We
0
10,000
20,000
30,000
40,000
April 25River
May 29Seapen
May 29River
July 2Seapen
Num
ber
Rel
ease
d
0.000%
0.010%
0.020%
0.030%
0.040%
0.050%
Per
cent
Rec
over
ed
(2)
(4)
(3)
(6)
(3)
(6)
(17)(16)
(11)
(12)
(6)
(5)
(2)
Fig. 9 The number of fish released (bars) and the percentage of CWTs recovered (black line) for each group of fish produced by theCowichan River Hatchery in 2008. The number of CWTs recovered is shown in brackets. Each group of fish (bars) had a unique tag code
144 Environ Biol Fish (2012) 94:135–148
had three estimates of wild smolt abundance thatranged from 814 200 to 206 500. The estimate of 814200 was a simple calculation that assumed thathatchery and wild fish entered the ocean at the sametime and had the same mortality up to the time thatthe hatchery percentages were measured in late June.Our estimate of hatchery percentage from the beachseine survey was 45%, indicating that there wereslightly more wild smolts in the ocean than hatcherysmolts. However, there were two release periods ofhatchery fish in the river on April 25 and May 29 andtwo releases from a net pen in the ocean on May 29and July 2. The higher mortalities associated with theearlier releases, as identified by the reduced recapturerate of CWTs, would indicate that it is unlikely thathatchery and wild juveniles had an equal mortalityprior to the June surveys. Thus, the estimate of 728600 wild juveniles would be too large. Also,considering that the escapement-fecundity estimate isabout 3 ½ times smaller, it is most likely that thelarger estimate of wild smolt production is high. Thebeach seine based estimate of 562 200 may be a moreaccurate estimate of wild smolt production, althoughit may also be too large. An important observation,however, is that all smaller estimates of wild smoltproduction increase the survival estimates of wildchinook salmon which increases the difference in theearly marine survival between wild and hatcherychinook salmon. If we use the larger estimate of wildjuvenile chinook salmon abundance, the survival ofwild Cowichan chinook salmon was estimated to be3.6% to early October or four times larger than thehatchery survival. If the lower estimate of wild smoltabundance is used, the early marine wild smoltsurvival is 18 times larger than the hatchery survivalthrough to early October. The intermediate estimate ofwild smolt abundance indicates that wild smolt earlymarine survival is about 6.5 times larger than forhatchery fish. The trawl study in early Octoberresulted in a total catch that was about 50% smallerthan observed 3 weeks earlier. The fish were alsodeeper in the water column, indicating a change inbehaviour had occurred. There were no comprehen-sive surveys after this date, so it was not possible toidentify movements out of the Gulf Islands area. It isprobable that the reduced catch and the changes in thepercentage of wild chinook salmon were influencedby movements out of the study area in the GulfIslands. Thus, the estimates of juvenile chinook
salmon survival in early October may be low becausefish were starting to migrate out of the Gulf Islands. Ifwe compare the survival estimates of hatchery andwild chinook salmon using the September survey, thewild survival is between six, nine and 24 times largerthan the hatchery survival, depending on the numberof wild smolts. The September samples may provide abetter comparison of the early marine survival of thetwo rearing types, with wild juvenile chinook salmonsurviving between six and 24 times better thanhatchery-released juveniles.
There were no Cowichan CWTs recovered in theSeptember survey in the Strait of Georgia. Therewere, however, four wild Cowichan juvenile chinooksalmon identified in the DNA sample. This indicatesthat there was some movement out of the GulfIslands, but the catches were rare. Any abundanceoutside of the Gulf Islands would increase theestimate of survival which would increase thedifference in survival between hatchery and wild fish.
The errors associated with the estimates of earlymarine survival of wild chinook salmon are similar tothe errors associated with the estimates for hatcheryfish. Additionally, there was a substantial error in theestimate of the numbers of wild smolts entering theocean. The estimate of a hatchery percentage of 45%in the beach seine survey may be closer to the truepercentage. The true early marine survival of juvenilewild chinook salmon from the Cowichan River in theGulf is not clear; however, it is clear that the estimateis substantially larger than for hatchery fish. Thedifferences in early marine survival between hatcheryand wild fish should be a clear indicator that the wildfish are better adapted to survive in the ecosystemwithin the Gulf Islands.
A number of studies (Bilton et al. 1982; Bilton1984; Ward and Slaney 1988; Martin and Wertheimer1989; Ward et al. 1989; Henderson and Cass 1991;Beckman et al. 1998; Friedland et al. 2009) haveshown that larger juvenile Pacific salmon survivebetter in the early marine environment than smallerindividuals. Throughout our study, hatchery fish wereconsistently larger than the better surviving wildchinook salmon. The large mortality of hatchery fishthat occurred from ocean entry until mid Septemberwould result from some kind of selection from theagent or agents causing the mortality. If the agent wasa predator, there was some quality that made thehatchery fish more accessible. It would seem reasonable
Environ Biol Fish (2012) 94:135–148 145
that future hatchery related research should focus onidentifying the attributes that result in hatchery fishhaving the very large early marine mortalities and beingmore susceptible to the sources of mortalities than thewild fish.
Hatcheries around the Strait of Georgia achievean average egg-to-smolt production of 70–80%(MacKinlay et al. 2010) compared to about 6% forwild chinook salmon (Bradford 1995). This isapproximately 12–13 times better than observed forwild chinook salmon. However, if the hatchery fishhave between a six and 24 times greater early marinemortality as found using the September survey data,there may be little value in removing the wild fishfrom the naturally spawning population. This con-clusion is not a criticism of hatcheries; rather it ismeant to identify the need to be more experimental.Experimentation is not limited to size and time ofrelease studies, as it is necessary to understand why wildfish can survive better in the early days in the ocean. Theimportance of the first few months in determining therecruitment of Pacific and Atlantic salmon is wellrecognized (Parker 1962; Ricker 1976; Pearcy 1992;Hansen and Quinn 1998; Friedland et al. 2009). It isalso known that hatchery-reared salmon do not surviveas well as their wild counterparts (Cross et al. 1991;Jonsson et al. 1991). These differences in survival mayindicate that there were ecological differences betweenthe hatchery and wild fish, as observed in other studies(Beauchamp 2009; Buhle et al. 2009).
We propose that the declining production ofCowichan chinook salmon stocks and possibly all of thedeclining chinook salmon stocks is a consequence of achanging environment in the early marine period, asothers have reported (Coronado and Hilborn 1998).Temperature is correlated with Pacific salmon survival(Mueter et al. 2002; Hinke et al. 2005; Beauchamp2009) and the temperature in the Strait of Georgia hasbeen increasing. The mechanisms linking temperatureto the decreasing marine survivals remain to beidentified, but the existing temperatures could beconsidered stressful for chinook salmon (Hinke et al.2005; Beauchamp 2009). The declines in the marinesurvival of chinook salmon are similar to the declinesobserved for coho salmon in the Strait of Georgiaindicating that large-scale ecosystem changes probablyare occurring possibly as a consequence of theincreasing trend in temperature (Beamish et al. 2008,2010). Thus, we propose that the general warming of
the surface waters is an indicator that juvenilePacific salmon will continue to be under increasingstress as the temperatures approach critical values(Beauchamp 2009). It appears reasonable to assumethat the trend in warming in the next 50 years willbe similar or greater that the past 50 years. Thiswould mean that the environment within the GulfIslands will become even more stressful for thejuvenile chinook salmon. In the 1970s it wasbelieved that there was capacity within the oceanto produce more salmon if more juveniles wereadded. Most scientists today believe that the challengein the future is to manage Pacific salmon so they canadapt to a more stressful ocean environment. Thus, thefuture objectives of hatchery programs may be toproduce fish that are able to survive in changing oceanenvironments.
The recent studies of Volk et al. (2010) showedthat chinook salmon from the Salmon River had aseries of life history types that were characterized by adiversity of estuarine entry times, sizes and periods ofnearshore residency. It is possible that this variation inthe use of the early marine environment provides theresiliency needed to adapt to a more variableenvironment as the ocean warms. It is known thatchinook salmon have a wide plasticity in smoltingbehaviour, with sizes ranging from 1 g to 30 g andages from 30 days to 14 months post emergence(Healey 1991; Beckman et al. 2003). Thus, it ispossible to consider that the population structure infresh water is an evolution of adaptation to con-ditions in the ocean in the immediate area of theocean adjacent to the river. The concern is that as theStrait of Georgia continues to warm, it may be theevolved resiliency of the wild fish that are best ableto adapt to the variability associated with thechanging nearshore environment.
Our message is not to encourage the shutting downof hatcheries, but to encourage everyone to recognizethe complexities of managing populations of chinooksalmon and all populations of Pacific salmon in achanging environment. Continuing to do what we aredoing and hoping that the next year will be bettermakes little sense. We think that the Cowichan Riverchinook population provides a perfect opportunity toidentify exactly what has caused the declining trendand exactly why wild chinook salmon survive betterthan hatchery chinook salmon. This information willmake better use of hatcheries and show British
146 Environ Biol Fish (2012) 94:135–148
Columbians how their impacts on climate are affectingboth wild and hatchery Pacific salmon.
Acknowledgments We thank the Pacific Salmon Commissionfor funding this study and Paul Rickard and Wilf Luedke for theiradvice and support. Kim Jonsen conducted the DNA analysis andJohn Candy managed the data analysis for the chinook salmonsamples at the Micro Genetics Laboratory at the Pacific BiologicalStation. Lana Fitzpatrick assisted with field work, figures andmanuscript preparation.
Open Access This article is distributed under the terms of theCreative Commons Attribution Noncommercial License whichpermits any noncommercial use, distribution, and reproductionin any medium, provided the original author(s) and source arecredited.
References
Beacham TD, Candy JR, Jonsen KL, Supernault J, Wetklo M,Deng L, Miller KM, Withler RE (2006) Estimation ofstock composition and individual identification of Chinooksalmon across the Pacific Rim basis using microsatellitevariation. Trans Am Fish Soc 135:861–888
Beamish RJ, Neville C-EM (1999) Large-scale climate-relatedchanges in the carrying capacity of salmon in the Straitof Georgia and northern North Pacific ecosystems. In:Sherman K, Tang Q (eds) Large marine ecosystems of thePacific Rim. Blackwell Science, Malden, pp 27–41
Beamish RJ, Riddell BE, Neville C-EM, Thomson BL, ZhangZY (1995) Declines in chinook salmon catches in theStrait of Georgia in relation to shifts in the marineenvironment. Fish Oceanogr 4:243–256
Beamish RJ, McCaughran D, King JR, Sweeting RM,McFarlane GA (2000) Estimating the abundance ofjuvenile coho salmon in the Strait of Georgia by meansof surface trawls. N Am J Fish Manage 20:369–375
Beamish RJ, Sweeting RM, Neville CM, Lange K (2007)Ocean changes in the Strait of Georgia indicate a need tolink hatchery programs, fishing strategies and early marinestudies of ocean carrying capacity into an ecosystemapproach to manage coho salmon. Extended abstractNPAFC Tech Rep 7:49–51
Beamish RJ, Sweeting RM, Lange KL, Neville CM (2008)Changing trends in the population ecology of hatchery andwild coho salmon in the Strait of Georgia. Trans Am FishSoc 137:503–520
Beamish RJ, Sweeting RM, Lange KL, Noakes DJ, PreikshotD, Neville CM (2010) Early marine survival of cohosalmon in the Strait of Georgia declines to very low levels.Mar Coast Fish 2:424–439
Beauchamp DA (2009) Bioenergetic ontogeny: Linking climateand mass-specific feeding to life-cycle growth andsurvival of salmon. Am Fish Soc Symp 70:1–19
Beckman BR, Larsen DA, Lee-Pawlik B, Dickhoff WW(1998) Relation of fish size and growth rate tomigration of spring chinook salmon smolts. N Am JFish Manage 18:537–546
Beckman BR, Larsen DA, Dickhoff WW (2003) Life historyplasticity in chinook salmon: relation of size and growthrate to autumnal smolting. Aquaculture 222:149–165
Bilton HT (1984) Returns of chinook salmon in relation tojuvenile size at release. Can Tech Rep Fish Aquat Sci1245
Bilton HT, Alderdice DF, Schnute JT (1982) Influence of timeand size at release of juvenile coho salmon (Oncorhynchuskisutch) on returns at maturity. Can J Fish Aquat Sci39:426–447
Bradford MJ (1995) Comparative review of Pacific salmonsurvival rates. Can J Fish Aquat Sci 52:1327–1338
Buhle ER, Holsman KK, Scheuerell MD, Albaugh A (2009)Using an unplanned experiment to evaluate the effects ofhatcheries and environmental variation on threatenedpopulations of wild salmon. Biol Conserv 142:2449–2455
Coronado C, Hilborn R (1998) Spatial and temporal factorsaffecting survival in coho and fall chinook salmon in thePacific Northwest. Bull Mar Sci 62:409–425
Cross CL, Lapi L, Perry EA (1991) Production of Chinook andcoho salmon from British Columbia hatcheries, 1971through 1989. Can Tech Rep Fish Aquat Sci 1816:48p
Fisheries and Environment Canada (1978) The salmonidenhancement program. A public discussion paper. Infor-mation Branch. Fisheries and Marine Science Service,Vancouver
Friedland KD, Moore D, Hogan F (2009) Rtrospective growthanalysis of Atlantic salmon (Salmo salar) from theMiramichi River, Canada. Can J Fish Aquat Sci 66:1294–1308
Hansen LP, Quinn TP (1998) The marine phase of theAtlantic salmon (Salmo salar) life cycle, with comparisonsto Pacific salmon. Can J Fish Aquat Sci 55(Supp. 1):104–118
Healey MC (1991) Life history of chinook salmon. In: Groot C,Margolis L (eds) Pacific salmon life histories. Universityof British Columbia Press, Vancouver, pp 311–393
Henderson MA, Cass AJ (1991) Effect of smolt size on smolt-to-adult survival for Chilko Lake sockeye salmn (Oncorhynchusnerka). Can J Fish Aquat Sci 48:988–994
Hinke JR, Watters GM, Boehlert GW, Zedonis P (2005) Oceanhabitat use in autumn by chinook salmon in coastal watersof Oregon and California. Mar Ecol Prog Ser 285:181–192
Jonsson B, Jonsson N, Hansen LP (1991) Difference in lifehistory and migratory behaviour between wild andhatchery reared Atlantic salmon in nature. Aquaculture98:69–78
MacKinlay D, Lehman S, Bateman J, Cook R (2010) PacificSalmon Hatcheries in British Columbia. Report to theSalmon Enhancement and Habitat Advisory Board.Available at www.sehab.org/pdf/hatcheries
Martin RM, Wertheimer A (1989) Adult production of chinooksalmon reared at different densities and released as twosmolt sizes. Prog Fish Culturist 51:194–200
Mueter FJ, Ware DM, Peterman RM (2002) Spatial correlationpatterns in coastal environmental variables and survivalrates of Pacific salmon in the Northeast Pacific Ocean.Fish Oceanogr 11:205–218
Neaves PI, Wallace CG, Candy JR, Beacham TD (2005)CBayes: Computer program for mixed stock analysis ofallelic data. Version v4.02. Free program distributed by the
Environ Biol Fish (2012) 94:135–148 147
authors over the internet from http://www.pac.dfo-mpo.gc.ca/sci/mgl/Cbayes_e.htm
Neville CM, Beamish RJ, Chittenden CM (2010) The use ofacoustic tags to monitor the movement and survival ofjuvenile chinook salmon in the Strait of Georgia. N PacAnadromous Fish Comm Document 1286:19
Parken CK, McNicol RE, Irvine JR (2006) Habitat-basedmethods to estimate escapement goals for data limitedChinook salmon stocks in British Columbia. CanadianScience Advisory Secretariat Research Document 2006/083, p. 74
Parker RR (1962) Estimations of ocean mortality rates forPacific salmon (Oncorhynchus). J Fish Res Board Can19:561–589
Pearcy WG (1992) Ocean Ecology of North Pacific Salmonids.University of Washington Press, Washington Sea Grant, 179
Pella J, Masuda M (2001) Bayesian method for analysis of stockmixtures from genetic characters. Fish Bull 99:151–167
Ricker WE (1976) Review of the rate of growth and mortalityof Pacific salmon in salt water, and noncatch mortalitycaused by fishing. J Fish Res Board Can 33:1483–1524
Riddell BE, Nagtegaal DA, Chen D (2000) A biologically-based escapement goal for Cowichan River Fall Chinook
salmon (Oncorhynchus tshawyscha). Pac Sci Advice RevCommittee Working Pap S00–17:30p
Sweeting RM, Beamish RJ, Noakes DJ, Neville CM (2003)Replacement of wild coho salmon by hatchery-reared cohosalmon in the Strait of Georgia over the past three decades.N Am J Fish Manage 23:492–502
Thomson RE, Foreman MGG (1998) Basin areas and volumesfor coastal southwest British Columbia and northwestWashington. Can Tech Rep Hydrol Ocean Sci 196:17
Tompkins A, Riddell B, Nagtegaal DA (2005) A biologically-based escapement goal for Cowichan River Fall chinooksalmon (Oncorhynchus tshawtyscha). Canadian ScienceAdvisory Secretariat Research Document 2005/095, p. 42.
Volk EC, Bottom DL, Jones KK, Simenstad CA (2010)Reconstructing juvenile Chinook salmon life history inthe Salmon River estuary, Oregon, using otolith micro-chemistry and microstructure. Trans Am Fish Soc139:535–549
Ward BR, Slaney PA (1988) Life history and smolt-to-adultsurvival of Keogh River steelhead (Salmo gairdneri) andthe relationship to smolt size. Can J Fish Aquat Sci45:1110–1122
Ward BR, Slaney PA, Facchin AR, Land RW (1989) Size-basedsurvival in steelhead trout: back-calculated lengths fromadults’ scales compared to migrated smolts at the KeoghRiver, British Columbia. Can J Fish Aquat Sci 46:1953–1858
