Oncorhynchus tshawytscha 1...We divided each trap season into temporal 510 Walters et al. strata corresponding to parr and smolt movement tim-ing, and these strata were further subdivided

summer Chinook salmon Oncorhynchus tshawytschathat spawn in central Idaho, USA. We examine pat-terns in relevant population parameters (movement,growth, survival) to understand the processes thatproduce the overall pattern. Insight into the role ofdensity dependence can help guide and prioritisemanagement strategies for these threatened popula-tions (Einum et al. 2008).

Methods

Study populations

We examined juvenile production, survival, growthand movement in Chinook salmon populationsthroughout the anadromous portion of Idaho over17 years (brood years 1991–2007) (Fig. 1). Weselected populations where there was establishedinfrastructure for monitoring juvenile emigration fromspawning reaches paired with adult abundance esti-mates. There were 18 such places (e.g., Venditti et al.2010). We chose nine populations based on the quan-tity and quality of the data with the aim of maximis-ing spatial dispersion of study sites (Fig. 1). Thesepopulations spawn in stream sections that averaged34.7 km in length and ranged in length from 15.3 to56.5 km.The study populations are considerably below his-

torical capacity for adult spawners. It is estimatedthere were greater than 1.5 million adult spawners inlate 1800s, but this dropped to approximately100,000 in the 1950s and to less than 10,000 duringthe 1980s (Matthews & Waples 1991). All popula-tions are part of the Snake River spring/summer runChinook salmon Evolutionarily Significant Unit(ESU), which was listed as threatened under the U.S.Endangered Species Act in 1992. At the time of list-ing, the adult population was estimated to be 0.5% ofthe historical abundance (Matthews & Waples 1991).Note that this period of extreme low abundance coin-cides with the beginning of our study. In terms ofjuvenile capacity, yearly smolt production for theESU reaches an asymptote of approximately 1.6 mil-lion since 1990 (Copeland et al. 2009) comparedwith an average of 2.5 million during 1964–1969(Raymond 1979).Snake River spring/summer run Chinook salmon

are considered to have a stream-type life history(Good et al. 2005); that is, they have an extendedfreshwater rearing phase and enter the ocean as year-lings. In one of our study populations (PahsimeroiRiver), a sizeable fraction of the juveniles emigrate tothe ocean as subyearlings (Copeland & Venditti2009), but this group has very low adult return ratesand was excluded from this study. In general, juve-nile Chinook salmon emigrants display two distinct

migratory phenotypes: leaving the spawning groundsas subyearlings during June through November (parr)and emigrating one full year after emergence duringtheir second spring (age-1 smolts). Parr emigrantsspend the winter in main stem reaches and passLower Granite Dam (Fig. 1, the first of eight damsthat juveniles must pass to reach the Pacific Ocean)the following spring. Age-1 smolts leave the streamsbetween March and June and travel quickly to passLower Granite Dam that spring.

Data collection

Initial cohort abundance for each population in eachyear was indexed by multiple pass redd surveys. Reddsare nests constructed in the stream gravel by spawn-ing females and are a surrogate for the number ofeggs spawned. The streams were surveyed three tofive times annually between early August and earlyOctober each year. Surveys began at the respectivescrew trap and proceeded upstream to the upperextent of known spawning. Trained observers walkedthe bank, scanning the stream substrate using polar-ised sunglasses to identify redds. To avoid doublecounting, each redd was marked by flagging a nearbybush or tree. The redd count was the sum of the newredds seen during each survey.Estimates of juvenile production, survival, growth

and migration were all based on monitoring data. Inall populations, emigrating juveniles were collectedusing rotary screw traps. The fish caught at the trapswere counted and measured (fork length), and a sub-sample was PIT-tagged. Traps on the Lemhi River,Pahsimeroi River, upper Salmon River, Marsh Creek,South Fork Salmon, Red River and Crooked ForkCreek were operated by the Idaho Department of Fishand Game. The Secesh River trap was operated bythe Nez Perce Tribe Department of FisheriesResources, and the East Fork Salmon River trap wasoperated by the Shoshone–Bannock Tribes FisheriesDepartment.Total juvenile production is a combination of abun-

dance and survival estimates. More specifically, weestimated the total abundance of each life stage (i.e.,parr and age-1 smolt) at the rotary screw traps, esti-mated survival for each life stage to Lower GraniteDam using a subsample of PIT-tagged fish andapplied these survival rates to the screw trap abun-dance estimates.We calculated life-stage abundance estimates

within the brood year from rotary screw trap opera-tions with Bailey’s modification of the Lincoln–Pet-ersen estimator (Steinhorst et al. 2004). To detectchanges in trap efficiency, we conducted efficiencytrials using marked fish released upstream of thetraps. We divided each trap season into temporal

510

Walters et al.

strata corresponding to parr and smolt movement tim-ing, and these strata were further subdivided intoshorter substrata in response to changes in environ-mental conditions (e.g., flow and temperature). Tomaintain robustness for analysis, we set a lower limitof seven mark recaptures for any substrata (Steinhorstet al. 2004). If a substratum did not contain a suffi-cient number of recaptures, it was included with theprevious or subsequent strata depending on streamand trap conditions. Young-of-the-year Chinook sal-mon fry (identified by readily observable differencesin size, coloration and lateral body markings) werenot included in smolt estimates for the spring season,and precocial Chinook salmon (large juveniles thatfreely expressed milt when handled) were notincluded in parr abundance.Juvenile survival from emigration from the traps to

Lower Granite Dam (calculated separately for parrand age-1 smolts) was estimated from the detectionof PIT-tagged individuals in the lower Snake andColumbia rivers. Daily detection records wereobtained by querying the PTAGIS database (www.ptagis.org) for all observations of fish tagged at eachtrap by calendar year. We estimated survival toLower Granite Dam by emigrant type within eachcohort using a Cormack–Jolly–Seber model imple-mented by SURPH software (Lady et al. 2010).Model inputs were records of the PIT tags released ateach trap and their subsequent detection at down-stream sites. Model outputs were the probability ofbeing detected at Lower Granite Dam (based ondetections there and downstream) and the probabilityof survival to Lower Granite Dam. The number ofeach emigrant type surviving to Lower Granite Damwas computed by multiplying abundance estimates atthe trap by survival probability. Juvenile productionwas estimated as the total number of juveniles(includes both parr and age-1 smolt emigrants) thatpass Lower Granite Dam.We used length at time of PIT tagging as our

growth estimate. We queried PTAGIS for each screwtrap and used average length in October, Novemberand December as our index of growth for parr andaverage length in February, March and April as ourindex of growth for smolts (Copeland & Venditti2009). Obvious data entry mistakes (e.g., lengthsgreater than 250 mm) were deleted.For movement, we focused on emigration from the

spawning grounds as estimated by the date of capturein the screw trap. For parr, we looked at the start ofemigration or the day that 10% of the population hadpassed the trap, as we hypothesised that at high den-sities, parr would start emigrating earlier. In contrast,for smolts, we used the end of emigration or the daythat 90% of the population had passed the trap. Wehypothesised that the start of smolt emigration in the

spring would be controlled by environmental factors,but in high-density years, emigration would continuefor a longer period of time. We also examined theratio of juveniles (calculated at the screw trap) thatemigrate as parr relative to smolts. We analysed theratio of parr abundance to smolt abundance andhypothesised that in high-density years, a higher pro-portion would emigrate as parr.

Data analyses

To assess density dependence, we combined datafrom the nine populations collected over 17 yearsinto a Bayesian hierarchical analysis to evaluatewhether there is data support for shared stock–recruit relationships. We used redd counts as ourindex of initial densities (stock) because juvenile sal-mon surveys are not carried out early enough to givereliable estimates of initial densities. We focused onjuvenile production at Lower Granite Dam (recruits)instead of returning spawners as we were interestedin density dependence during the freshwater rearingstage.Hierarchical models have become widely used in

fisheries and ecology over the last decade (Myerset al. 2001; Bolker 2009). In hierarchical models,parameters for each grouping factor (individual, pop-ulation, etc.) are assumed to be distributed aroundglobal or shared parameters (Gelman & Hill 2007).The strength of hierarchical models or meta-analysesis that by combining data from multiple sources, theprecision of estimated global parameters can begreatly improved (Liermann & Hilborn 1997). It alsoprovides better estimates for populations with limitedor variable data, because it can draw on the data fromthe other populations.We considered four potential models: a linear

model forced through the origin, a linear model withestimated intercept, the Ricker model and the Bever-ton–Holt model (Table 1). For the linear models, wetreated the slope and intercept as hierarchicalparameters, and for the Ricker and Beverton–Holtmodels, the ‘a’ and ‘b’ parameters were treated as

Table 1. Global parameter estimates and deviance information criterion(DIC) scores for the data fit to the four competing models.

Model Global a Global b DIC

Linear (throughorigin)

ln(juv) ~ ln(a*redd) 134 327.2

Linear juv ~ a + b*redd 8.5 0.0040 324.4

Beverton Holt ln!juv"# ln a$redd1%b$redd

! "275.1 0.0107 285.9

Ricker ln(juv) ~ ln((a*redd)*exp( b*redd))

198.1 0.0029 293.7

511

Density dependence in salmon

hierarchical. We used deviance information criterion(DIC), a Bayesian model selection technique, to seewhich of the four models the data supported (Spiegel-halter et al. 2002). The posterior parameter estimatesfrom the best model provided a global model andestimates of parameters for each individual popula-tion. Analyses were run using JAGS (Just AnotherGibbs Sampler) in R utilising the runjags and R2 jagspackages. All analyses were run in R (version 2.13.1;R Development Core Team 2011).The population-level effect of density dependence

occurs via processes affecting individuals. Therefore,we also considered the effect of density on growth,survival and movement for the two emigrant life-his-tory strategies (parr and age-1 smolts). We plottedeach metric versus the redd count for the correspond-ing year and found the negative power function(y = ax b) provided a good fit to the data (Grant &Imre 2005; Vincenzi et al. 2010). We loge-trans-formed the data to linearise the relationship (ln(y) = ln(a) & b*ln(x)) and used a linear model to fitthe data using the same hierarchical modellingapproach for parameter estimation as describedabove. For the linear models, the slope (b) providesan indicator of the degree of density dependence. Anegative slope suggests density dependence, withsteeper slopes indicative of greater density depen-dence, a slope of zero suggests no density depen-dence, and a positive slope suggests inverse densitydependence. We considered the slope to be signifi-cantly different from zero if the 95% confidenceintervals for b did not overlap zero.

Results

There was a wide range in abundance of adults andjuveniles between both streams and years during thestudy period. Redd counts varied over two orders ofmagnitude, from 0 to 718. Redd counts were gener-ally highest in 2001 and 2003 and lowest in 1995and 1999. Juvenile emigration estimates ranged fromeight to 759,567 parr and from six to 9,055 smolts.Parr survival from trap to Lower Granite Dam wasbetween 0.04 and 0.58, and smolt survival wasbetween 0.09 and 1. Total juvenile production atLower Granite Dam ranged from 235 to 91,813 indi-viduals. The range in these data provides the basisfor credible estimates of the parameters of the stock–recruit functions. We also believe these data encom-pass the likely range of values that will occur in thefuture.There was strong support for density dependence

in juvenile production. The Beverton–Holt modelbest fit the data, and the Ricker model was the nextbest fit (Table 1). In the Beverton–Holt model, juve-nile production is a function of the initial cohort size

(redd count), intrinsic productivity (a) and the percapita strength of density-dependent mortality (b).The global model had parameter estimates for the Be-verton–Holt model of a = 275.1 and b = 0.0107(Table 1). In other words, on average 275 juvenilesper redd would survive to migrate past Lower GraniteDam in the absence of density dependence. Togethera and b parameters provide an estimate of averagecarrying capacity (a/b = 275.1/0.0107 = 25,710juveniles). The individual populations showed sub-stantial variation in both a (228–358) and b(0.0037–0.0409) parameters (Table 2). In terms ofjuvenile production, the Lemhi River, East Fork Sal-mon River and Crooked Fork Creek populationsshowed the strongest degree of density dependence(high b values). For those populations, the asymptoticproduction was lower due to density dependencecausing a greater curvature in the stock–recruit rela-tionship. In contrast, the upper Salmon River and theSecesh River populations had low b values and highasymptotic production and did not approach asymp-totic production as quickly (Fig. 2, Table 2).We also found evidence of density-dependent

effects on growth and survival, but not on migra-tion. The average length of parr and smolts for allpopulations showed a significant decrease withincreasing redd count suggesting strong density-dependent growth (Fig. 3, Table 3). Both parr andsmolt survival decreased with increasing redd count,but the slopes were steeper for parr survival (Fig.4, Table 3). The end of smolt migration occurredlater at higher redd counts, as hypothesised, for allpopulations, but was not significant for any, sug-gesting the timing of smolt migration is not highlydensity dependent (Fig. 5, Table 3). The beginningof parr migration occurred significantly earlier intwo populations suggesting the potential for den-sity-dependent movement in these populations, butthere were negligible shifts in timing for the

Table 2. Beverton Holt ln!juv"# ln a$redd1%b$redd

! "! "parameter estimates for

the global model and each population. The a parameter is indicative ofpopulation productivity; the b parameter, of density-dependent mortality;and a/b, of the carrying capacity. The parameter values for the globalmodel are in bold.

Population a b a/b

Global 275.1 0.0107 25710Lemhi 228.6 0.0332 6886Marsh 301.3 0.0096 31385Upper Salmon 340.1 0.0059 57644East Fork 272.5 0.0409 6663Pahsimeroi 249.0 0.0137 18175South Fork 268.5 0.0091 29505Secesh 357.6 0.0037 96649Red 270.7 0.0183 14792Crooked Fork 338.0 0.0335 10090

512

Walters et al.

density dependence (Lemhi River, East Fork, andCrooked Fork) also experience a higher degree ofanthropogenic disturbance; therefore, habitat restora-tion may be effective in some situations. However,many of the populations studied are in relativelyundisturbed areas suggesting that habitat quality isnot a universal limiting factor.The idea that there are insufficient resources due to

nutrient limitation also has proponents (Achord et al.2003). In Atlantic salmon Salmo salar, variation inprey biomass has been shown to be one of the mostimportant factors driving growth (Ward et al. 2009).In the Pacific Northwest, it is estimated that returningspawners supply only 6–7% of the historic load ofmarine-derived nutrients (Gresh et al. 2000) resultingin nutrient-limited streams (Sanderson et al. 2009). Inaddition, studies of nutrient additions in the form ofsalmon carcasses have often resulted in increased sal-mon growth (Wipfli et al. 2003; Rinella et al. 2009).Lowered resource availability could also be interact-ing with abiotic factors; Crozier et al. (2010) foundthat density-dependent growth was more prevalent athigher temperatures, which they attributed toincreased resource requirements at higher tempera-tures.The importance of spatial clustering, either due to

clustered redd distribution and limited movement ordue to predation risk, is less well explored. Two pop-ular management techniques, population supplemen-tation with hatchery fish and restoration of spawninghabitat, will be ineffective if juvenile clustering is notaddressed. Increased numbers of juveniles due tostocking can actually lead to increased severity ofdensity-dependent growth and survival if juveniles donot disperse (Walters & Juanes 1993; Buhle et al.2009). Renovation of spawning habitat is also inef-fective unless adults colonise restored areas, andthere is sufficient rearing habitat available.Conservation actions should focus on life-history

stages that are most susceptible to density depen-dence. Our analyses suggest that winter mortality isimportant for these populations. Use of winter refugiacan be density dependent (Armstrong & Griffiths2001); however, winter habitats have not been identi-fied for parr. If these are delineated, managers couldconsider how to increase refugia quantity. Giventhese areas are likely in larger main stem habitat, thisproblem needs careful and creative planning to makethe issue tractable. Bjornn (1971) found that fewerfish emigrated in the fall from experimental channelsprovided with rubble substrate rather than gravel,showing the importance of appropriate habitat to fishpreparing for winter. Another approach to reducewinter mortality is to increase juvenile quality goinginto that period. Several authors (Gresh et al. 2000;Wipfli et al. 2003; Kohler et al. 2012) have proposed

augmenting nutrient levels to increase growth. To ourknowledge, none have demonstrated a population-level benefit to date. Nutrient augmentation wouldlikely be most successful if complemented by anincrease in refugia in summer rearing reaches, whichmay allow juveniles to safely access resources cur-rently too risky to use (Walters & Juanes 1993).The scale of the actions necessary to address den-

sity dependence across the study area is likely to beconsiderable and varies between streams. While somestreams would clearly benefit from local habitat res-toration efforts, many of these populations inhabitminimally impacted areas in remote settings. If den-sity dependence is a natural feature of these popula-tions, then it will be hard to generate many moresmolts from the currently occupied areas. Indeed, Pet-rosky et al. (2001) found that the productivity rate(smolts per spawner) of the aggregated Snake Riverspring/summer Chinook salmon populations did notchange significantly between the 1962 and 1999brood years, indicating that the quality of currentlyoccupied habitats has not changed greatly in the lastfew decades. Hilborn (1985) posited that stressedpopulations lose the less productive subunits first;surviving subunits are more productive and respondquickly but might not fill former areas. Therefore,recovery to levels capable of sustainable harvest maydepend on an increase in smolt-to-adult survivalrates, so there are sufficient adults to recolonise non-core areas.In summary, effective conservation and manage-

ment of these populations will require a thoroughconsideration of density dependence. We foundstrong evidence for density-dependent growth andsurvival in multiple populations across a wide rangeof spawner and juvenile abundances and environmen-tal conditions. Density in spawning reaches affectsgrowth of all juveniles, which in turn affects survivalof parr emigrants downstream and overwinter sur-vival of smolt emigrants before they start their move-ments in spring. There are several reasons whydensity dependence could be occurring. Of these,habitat loss and degradation are being addressed,while further research is needed into the role ofresource availability, spatial clustering and life-his-tory trade-offs due to predation risk. Density depen-dence at the population level is common inanadromous salmonids with substantial freshwaterresidence time, even if the population has experi-enced serious declines, and must be considered indemographic analysis and management.

Acknowledgements

Bruce Barnett assisted with collection and collation of muchof the data. The Nez Perce Tribe and the Shoshone Bannock

516

Walters et al.

Tribes provided data for the Secesh River and East ForkSalmon River, respectively. Paul Bunn provided the mapfigure. Eric Ward wrote the code for the Bayesian hierarchicalmodelling. Michelle McClure generously gave guidance toAWW. The study benefitted from review by Charlie Petroskyand two anonymous reviewers. Funding for field work andsupport for DAV and TC were provided by the BonnevillePower Administration (projects 1989 098 00 and 1990 05500). AWW was supported by a postdoctoral fellowship fromthe National Research Council. Any use of trade, firm or product names is for descriptive purposes only and does not implyendorsement by the U.S. Government.

References

Achord, S., Levin, P.S. & Zabel, R.W. 2003. Density dependent mortality in Pacific salmon: the ghost of impacts past?Ecology Letters 6: 335 342.

Armstrong, J.D. & Griffiths, S.W. 2001. Density dependentrefuge use among over wintering wild Atlantic salmon juveniles. Journal of Fish Biology 58: 1524 1530.

Barnthouse, L.W., Boreman, J., Christensen, S.W., Goodyear,C.P., Van Winkle, W. & Vaughan, D.S. 1984. Populationbiology in the courtroom: the Hudson River controversy.BioScience 34: 14 19.

Beard, T.D. & Carline, R.F. 1991. Influence of spawning andother stream habitat features on spatial variability of wildbrown trout. Transactions of the American Fisheries Society120: 711 722.

Beckerman, A., Benton, T.G., Ranta, E., Kaitala, V. & Lundberg, P. 2002. Population dynamic consequences of delayedlife history effects. Trends in Ecology and Evolution 17:263 269.

Bentzen, P., Olsen, J.B., McLean, J.E., Seamons, T.R. &Quinn, T.P. 2001. Kinship analysis of Pacific salmon:insights into mating, homing, and timing of reproduction.Journal of Heredity 92: 127 136.

Berryman, A.A., Lima Arce, M. & Hawkins, B.A. 2002. Population regulation, emergent properties, and a requiem fordensity dependence. Oikos 99: 600 606.

Bjornn, T.C. 1971. Trout and salmon movements in two Idahostreams as related to temperature, food, stream flow, cover,and population density. Transactions of the American Fisheries Society 100: 423 438.

Bjornn, T.C. 1978. Survival, production, and yield of troutand Chinook salmon in the Lemhi River. Moscow, ID: University of Idaho, College of Forestry, Wildlife and RangeSciences Bulletin. 27

Bolker, B.M. 2009. Learning hierarchical models: advice forthe rest of us (forum response). Ecological Applications 19:588 592.

Bowersox, B.J., Banks, R. & Crawford, E. 2011. PotlatchRiver steelhead monitoring and evaluation, annual performance report, 2009. Idaho Department of Fish and GameReport 11 103.

Buhle, E.R., Holsman, K.K., Scheuerell, M.D. & Albaugh, A.2009. Using an unplanned experiment to evaluate the effectsof hatcheries and environmental variation on threatened populations of wild salmon. Biological Conservation 142:2449 2455.

Connor, W.P. & Tiffan, K.F. 2012. Evidence for parr growthas a factor affecting parr to smolt survival. Transactions ofthe American Fisheries Society 141: 1207 1218.

Copeland, T. & Venditti, D.A. 2009. Contribution of three lifehistory types to smolt production in a Chinook salmon(Oncorhynchus tshawytscha) population. Canadian Journalof Fisheries and Aquatic Sciences 66: 1658 1665.

Copeland, T., Johnson, J., Apperson, K., Flinders, J. & Hand,R. 2009. Idaho natural production monitoring and evaluation. Idaho Department of Fish and Game Report 09 06.

Crozier, L.G., Zabel, R.W., Hockersmith, E.E. & Achord, S.2010. Interacting effects of density and temperature on bodysize in multiple populations of Chinook salmon. Journal ofAnimal Ecology 79: 342 349.

Einum, S., Sundt Hansen, L. & Nislow, K.H. 2006. The partitioning of density dependent dispersal, growth and survivalthroughout ontogeny in a highly fecund organism. Oikos113: 489 496.

Einum, S., Nislow, K.H., Reynolds, J.D. & Sutherland, W.J.2008. Predicting population responses to restoration ofbreeding habitat in Atlantic salmon. Journal of AppliedEcology 45: 930 938.

Foldvik, A., Finstad, A.G. & Einum, S. 2010. Relating juvenilespatial distribution to breeding patterns in anadromous salmonid populations. Journal of Animal Ecology 79: 501 509.

Fowler, C.W. 1981. Density dependence as related to life history strategy. Ecology 62: 602 610.

Francis, R.I.C.C. 1997. Comment: how should fisheries scientists and managers react to uncertainty about stock recruitrelationships? Canadian Journal of Fisheries and AquaticSciences 54: 982 983.

Gelman, A. & Hill, J. 2007. Data analysis using regressionand multilevel/ hierarchical models. Cambridge: CambridgeUniversity Press.

Gilbert, D.J. 1997. Towards a new recruitment paradigm forfish stocks. Canadian Journal of Fisheries and Aquatic Sciences 54: 969 977.

Good, T.P., Waples, R.S. & Adams, P.B. 2005. Updated status of federally listed ESUs of West Coast salmon and steelhead. Seattle, WA: U.S. Department of Commerce, NOAATechnical Memorandum, NMFS NWFSC 66.

Grant, J.W.A. & Imre, I. 2005. Patterns of density dependentgrowth in juvenile stream dwelling salmonids. Journal ofFish Biology 67: 100 110.

Greene, C.M. & Beechie, T.J. 2004. Consequences of potentialdensity dependent mechanisms on recovery of ocean typeChinook salmon (Oncorhynchus tshawytscha). CanadianJournal of Fisheries and Aquatic Sciences 61: 590 602.

Gresh, T., Lichatowich, J. & Schoonmaker, P. 2000. An estimation of historic and current levels of salmon production inthe Northeast Pacific Ecosystem: evidence of a nutrient deficit in the freshwater systems of the Pacific Northwest. Fisheries 25: 15 21.

Hall, C.A.S. 1988. An assessment of several of the historicallymost influential theoretical models used in ecology and ofthe data provided for their support. Ecological Modelling43: 5 31.

Hamann, E.J. & Kennedy, B.P. 2012. Juvenile dispersalaffects straying behaviors of adults in a migratory population. Ecology 93: 733 740.

517

Density dependence in salmon

Hilborn, R. 1985. Apparent stock recruitment relationships inmixed stock fisheries. Canadian Journal of Fisheries andAquatic Sciences 42: 718 723.

Hilborn, R. 1997. Comment: recruitment paradigms for fishstocks. Canadian Journal of Fisheries and Aquatic Sciences54: 984 985.

Hilborn, R. & Walters, C.J. 1992. Quantitative fisheries stockassessment: choices, dynamics, and uncertainty. Boston,MA: Kluwer Academic Publishers.

Hixon, M.A., Pacala, S.W. & Sandin, S.A. 2002. Populationregulation: historical context and contemporary challengesof open vs. closed systems. Ecology 83: 1490 1508.

Hurst, T.P. 2007. Causes and consequences of winter mortality in fishes. Journal of Fish Biology 71: 315 345.

Iles, T.C. & Beverton, R.J.H. 2000. The concentration hypothesis: the statistical evidence. ICES Journal of Marine Science 57: 216 227.

Isaak, D.J. & Thurow, R.F. 2006. Network scale spatial andtemporal variation in Chinook salmon (Oncorhynchustshawytscha) redd distributions: patterns inferred from spatially continuous replicate surveys. Canadian Journal ofFisheries and Aquatic Sciences 63: 285 296.

Kohler, A.E., Pearsons, T.N., Zendt, J.S., Mesa, M.G., Johnson, C.L. & Connolly, P.J. 2012. Nutrient enrichment withsalmon carcass analogs in the Columbia River basin, USA:a stream food web analysis. Transactions of the AmericanFisheries Society 141: 802 824.

Krebs, C.J. 1995. Two paradigms of population regulation.Wildlife Research 22: 1 10.

Lady, J., Westhagen, P. & Skalski, J.R. 2010. SURPH (Survival Under Proportional Hazards). Version 3.1.1. [online].Available at: www.cbr.washington.edu/analysis/apps/surph/.Last accessed 1 October 2010.

Liermann, M. & Hilborn, R. 1997. Depensation in fish stocks:a hierarchic Bayesian meta analysis. Canadian Journal ofFisheries and Aquatic Sciences 54: 1976 1984.

Lob!on Cervi!a, J. 2007. Density dependent growth in streamliving brown trout Salmo trutta L. Functional Ecology 21:117 124.

Lob!on Cervi!a, J. 2009. Recruitment as a driver of productiondynamics in stream resident brown trout (Salmo trutta).Freshwater Biology 54: 1692 1704.

Lorenzen, K. 2005. Population dynamics and potential forfisheries stock enhancement: practical theory for assessmentand policy analysis. Philosophical Transactions of the RoyalSociety B 360: 171 189.

Matthews, G.M. & Waples, R.W. 1991. Status review forSnake River spring and summer Chinook salmon. Seattle,WA: US Department of Commerce, National Marine Fisheries Service.

Milner, N.J., Elliott, J.M., Armstrong, J.D., Gardiner, R., Welton, J.S. & Ladle, M. 2003. The natural control of salmon andtrout populations in streams. Fisheries Research 62: 111 125.

Myers, R.A. 1997. Comment and reanalysis: paradigms forrecruitment studies. Canadian Journal of Fisheries andAquatic Sciences 54: 978 981.

Myers, R.A. 2002. Recruitment: understanding density dependence in fish populations. In: Hart, P., Reynolds, J., eds.Handbook of fish biology and fisheries, Volume 1: FishBiology. Oxford: Blackwell Publishing, pp. 123 148.

Myers, R.A., MecKenzie, B.R., Bowen, K.G. & Barrowman,N.J. 2001. What is the carrying capacity for fish in theocean? A meta analysis of population dynamics of NorthAtlantic cod. Canadian Journal of Fisheries and Aquatic Sciences 58: 1464 1476.

Paulsen, C.M. & Fisher, T.R. 2001. Statistical relationshipbetween parr to smolt survival of Snake River spring summer Chinook salmon and indices of land use. Transactionsof the American Fisheries Society 130: 347 358.

Peery, C.A. & Bjornn, T.C. 2000. Dispersal of hatchery rearedChinook salmon parr following release into four Idahostreams. North American Journal of Fisheries Management20: 19 27.

Peterman, R.M. 1987. Review of the components of recruitment of Pacific salmon. American Fisheries Society Symposium 1: 417 429.

Petrosky, C.E., Schaller, H.A. & Budy, P. 2001. Productivityand survival rate trends in the freshwater rearing stage ofSnake River Chinook salmon (Oncorhynchus tshawytscha).Canadian Journal of Fisheries and Aquatic Sciences 58:1196 1207.

Quinn, T.J. & Deriso, R.B. 1999. Quantitative fish dynamics.New York: Oxford University Press.

R Development Core Team. 2011. R: a language and environment for statistical computing. Vienna: R Foundation forStatistical Computing. Available at: http://www.R project.org. Last accessed 1 October 2012.

Raymond, H.L. 1979. Effects of dams and impoundments onmigrations of juvenile Chinook salmon and steelhead fromthe Snake River, 1966 1975. Transactions of the AmericanFisheries Society 108: 505 529.

Richards, C. & Cernera, P.J. 1989. Dispersal and abundanceof hatchery reared and naturally spawned juvenile Chinooksalmon in an Idaho stream. North American Journal of Fisheries Management 9: 345 351.

Rinella, D.J., Wipfli, M.S., Stricker, C.A., Heintz, R.A. &Rinella, M.J. 2009. Pacific salmon (Oncorhynchus spp.)runs and consumer fitness: growth and energy storage instream dwelling salmonids increase with salmon spawnerdensity. Canadian Journal of Fisheries and Aquatic Sciences69: 73 84.

Rose, K.A., Cowan, J.H., Winemiller, K.O., Myers, R.A. &Hilborn, R. 2001. Compensatory density dependence in fishpopulations: importance, controversy, understanding andprognosis. Fish and Fisheries 2: 293 327.

Sanderson, B.L., Coe, H.J., Tran, C.D., Macneale, K.H., Harstad, D.L. & Godwin, A.B. 2009. Nutrient limitation ofperiphyton in Idaho streams: results from nutrient diffusingexperiments. Journal of the North American BenthologicalSociety 28: 832 845.

Sekulich, P.T. 1980. The carrying capacity of infertile foreststreams for rearing juvenile Chinook salmon. Doctoral dissertation Moscow, ID: University of Idaho.

Sibly, R.M., Barker, D., Denham, M.C., Hone, J. & Pagel, M.2005. On the regulation of populations of mammals, birds,fish, and insects. Science 309: 607 610.

Spiegelhalter, D.J., Best, N.G., Carlin, B.R. & Van Der Linde,A. 2002. Bayesian measures of model complexity and fit.Journal of the Royal Statistical Society Series B StatisticalMethodology 64: 583 616.

518

Walters et al.

Stamps, J.A. 2006. The silver spoon effect and habitat selection by natal dispersers. Ecology Letters 9: 1179 1185.

Steinhorst, K., Wu, Y., Dennis, B. & Kline, P. 2004. Confidence intervals for fish out migration estimates using stratified trap efficiency methods. Journal of Agricultural,Biological, and Environmental Statistics 9: 284 299.

Teichert, M.A.K., Kvingdal, E., Forseth, T., Ugedal, O. &Finstad, A.G. 2010. Effects of discharge and local densityon the growth of juvenile Atlantic salmon Salmo salar. Journal of Fish Biology 76: 1751 1769.

Thompson, W.L. & Lee, D.C. 2002. A two stage information theoretic approach to modeling landscape level attributes and maximum recruitment of Chinook salmon inthe Columbia River Basin. Natural Resource Modeling15: 227 257.

Turchin, P. 1999. Population regulation: a synthetic view.Oikos 84: 153 159.

Utz, R.M. & Hartman, K.J. 2009. Density dependent individual growth and size dynamics of central Appalachian brooktrout (Salvelinus fontinalis). Canadian Journal of Fisheriesand Aquatic Sciences 66: 1072 1080.

Venditti, D.A., Apperson, K.A., Kinzer, R., Flinders, J.,Teton, A., Bretz, C., Bowersox, B. & Barnett, B. 2010.Idaho supplementation studies, brood year 2007 cooperatorreport. Boise, ID: Idaho Department of Fish and GameReport. 10 14 Boise, Idaho.

Vincenzi, S.A., Crivelli, A.J., Jesenek, D. & De Leo, G.A.2008. The role of density dependent individual growth inthe persistence of freshwater salmonid populations. Oecologia 156: 523 534.

Vincenzi, S., Crivelli, A.J., Jesenek, D. & De Leo, G.A. 2010.Detection of density dependent growth at two spatial scalesin marble trout (Salmo marmoratus) populations. Ecology ofFreshwater Fish 19: 338 347.

Vincenzi, S., Satterthwaite, W.H. & Mangel, M. 2012.Spatial and temporal scale of density dependent bodygrowth and its implications for recruitment, populationdynamics, and management of stream dwelling salmonidpopulations. Reviews in Fish Biology and Fisheries 22:813 825.

Vollestad, L.A. & Olsen, E.M. 2008. Non additive effects ofdensity dependent and density independent factors on browntrout vital rates. Oikos 117: 1752 1760.

Walters, C.J. & Juanes, F. 1993. Recruitment limitation as aconsequence of natural selection for use of restricted feedinghabitats and predation risk taking be juvenile fishes. Canadian Journal of Fisheries and Aquatic Sciences 50:2058 2070.

Ward, D.M., Nislow, K.H. & Folt, C.L. 2009. Increased population density and suppressed prey biomass: relative impactson juvenile Atlantic salmon growth. Transactions of theAmerican Fisheries Society 138: 135 143.

Werner, E.E. & Anholt, B.R. 1993. Ecological consequencesof the trade off between growth and mortality rates mediatedby foraging activity. American Naturalist 142: 242 272.

Werner, E.E. & Gilliam, J.F. 1984. The ontogenetic niche andspecies interactions in size structured populations. AnnualReview of Ecology and Systematics 15: 393 425.

Wipfli, M.S., Hudson, J.P., Caouette, J.P. & Chaloner, D.T.2003. Marine subsidies in freshwater ecosystems: salmoncarcasses increase the growth rates of stream resident salmonids. Transactions of the American Fisheries Society 132:371 381.

Zabel, R.W. & Achord, S. 2004. Relating size of juveniles tosurvival within and among populations of Chinook salmon.Ecology 85: 795 806.

519

Density dependence in salmon