Lardeau-Lower Duncan River Juvenile Rainbow Trout Abundance and Stock-Recruitment Analysis 2016

The suggested citation for this analytic report is:

Thorley, J.L. and Hogan, P.M. (2016) Lardeau-Lower Duncan River Juvenile Rainbow Trout Abundance and Stock-Recruitment 2016. A Poisson Consulting Analysis Report. URL: https://www.poissonconsulting.ca/f/6848898.

Background

Juvenile Age-1 Gerrard Rainbow Trout from Kootenay Lake rear in the Lardeau and Lower Duncan rivers. From 2006 to 2014 and in 2016 spring snorkel surveys were done to estimate the abundance of Age-1 fish. From 2006 and 2010 the surveys were conducted at fixed index site. From 2010 fish observations were mapped to sites on the river by georeferencing all counts which allowed the swimmers to cover more of the river.

The primary aims of the current analyses are to:

  1. Estimate the observer efficiency when conducting spring snorkel surveys for Age-1 fish.
  2. Estimate the spring abundance of Age-1 fish by year.
  3. Estimate the stock-recruitment relationship between the number of spawners in a given year and the number of Age-1 recruits the following spring.

Methods

Data Preparation

The snorkel count data was provided by Redfish Consulting Ltd. The Ministry of Forests, Lands and Natural Resource Operations (MFLNRO) provided the AUC-based spawner escapement estimates for Gerrard.

Length-Bias Correction

Biases in length estimation were estimated manually; for each observer for each unique day of observations, a fork-length cut-off between Age-1 and Age-2+ fish was estimated, which was then used to classify each fish by lifestage.

Statistical Analysis

Hierarchical Bayesian models were fitted to the data using R version 3.2.4 (R Core Team 2015) and JAGS 4.2.0 (Plummer 2015) which interfaced with each other via jaggernaut 2.3.2 (Thorley 2013). For additional information on hierarchical Bayesian modelling in the BUGS language, of which JAGS uses a dialect, the reader is referred to (Kery and Schaub 2011, 41–44).

Unless specified, the models assumed vague (low information) prior distributions (Kery and Schaub 2011, 36). The posterior distributions were estimated from a minimum of 1,000 Markov Chain Monte Carlo (MCMC) samples thinned from the second halves of three chains (Kery and Schaub 2011, 38–40). Model convergence was confirmed by ensuring that Rhat (Kery and Schaub 2011, 40) was less than 1.1 for each of the parameters in the model (Kery and Schaub 2011, 61).

The posterior distributions of the fixed (Kery and Schaub 2011, 75) parameters are summarised in terms of a point estimate (mean), lower and upper 95% credible limits (2.5th and 97.5th percentiles), the standard deviation (SD), percent relative error (half the 95% credible interval as a percent of the point estimate) and significance (Kery and Schaub 2011, 37, 42).

Variable selection was achieved by dropping insignificant (Kery and Schaub 2011, 37, 42) fixed (Kery and Schaub 2011, 77–82) variables and uninformative random variables. A fixed variable was considered to be insignificant if its significance was \(\geq\) 0.05 while a random variable was considered to be uninformative if its percent relative error was \(\geq\) 80%. The Deviance Information Criterion (DIC) was not used because it is of questionable validity when applied to hierarchical models (Kery and Schaub 2011, 469).

The results are displayed graphically by plotting the modelled relationships between particular variables and the response with 95% credible intervals (CRIs) with the remaining variables held constant. In general, continuous and discrete fixed variables are held constant at their mean and first level values respectively while random variables are held constant at their typical values (expected values of the underlying hyperdistributions) (Kery and Schaub 2011, 77–82). Where informative the influence of particular variables is expressed in terms of the effect size (i.e., percent change in the response variable) with 95% CRIs (Bradford, Korman, and Higgins 2005).

Observer Efficiency

The observer efficiency for Age-1 fish was estimated using a mark-resight binomial model (Kery and Schaub 2011, 134–36, 384–88).

Key assumptions of the observer efficiency model include:

  • The observer probability varies with study design (Index and GPS).
  • There is no tag loss.
  • There is no emigration of marked fish.
  • The number of marked fish that are resighted is described by a binomial distribution.

Abundance

The abundance was estimated from the length bias-corrected observer count data using an overdispersed Poisson model (Kery and Schaub 2011, 55–56). The annual abundance estimates represent the total number of fish in the study area.

Key assumptions of the abundance model include:

  • The lineal fish density varies with year, useable width and river kilometer as a polynomial, and randomly with site.
  • The observer efficiency at marked sites was as estimated by the observer efficiency model (which varied by study design).
  • The observer efficiency also varies with visit type (standard count versus presence of marked fish) within study design, and randomly with swimmer.
  • The expected count at a site is the expected lineal density multiplied by the site length, the observer efficiency and the proportion of the site surveyed.
  • The residual variation in the actual count, which is gamma-Poisson distributed, varies with the annual lineal fish density.

Stock-Recruitment

The relationship between the number of spawners in a given year (\(S\)) and the number of Age-1 recruits the following spring (\(R\)) was estimated using a Bayesian Beverton-Holt stock-recruitment model (Walters and Martell 2004):

\[ R = \frac{a \cdot S}{1 + b \cdot S} \quad,\]

where \(a\) is the maximum reproductive performance per spawner, and \(b\) determines the population size scaling.

Key assumptions of the stock-recruitment model include:

  • The prior probability \(a\) is normally distributed with a mean of 500 and a SD of 250; this mean is based on an average of 8,000 eggs per female spawner, a 50:50 sex ratio, 50% egg survival, 50% post-emergence fall survival and 50% overwintering survival.
  • The residual variation in the number of recruits is log-normally distributed.

We may determine the maximum recruit population \(K\) that the environment can sustain indefinitely, the carrying capacity, by the relation:

\[ K = \frac{a}{b} \quad.\]

Model Code

The JAGS model code, which uses a series of naming conventions, is presented below.

Capture Efficiency

Variable/Parameter Description
bEfficiency logit(eEfficiency) intercept
bEfficiencyStudyDesign[ii] Effect of iith study design on logit(eEfficiency)
eEfficiency[ii] Expected capture efficiency on iith visit
Marked[ii] Number of marked fish prior to iith visit
Resighted[ii] Number of marked fish resighted on iith visit
StudyDesign[ii] Study design of iith visit
Capture Efficiency - Model1
model{

  bEfficiency ~ dnorm(0, 5^-2)

  bEfficiencyStudyDesign[1] <- 0
  for (ii in 2:nStudyDesign) {
    bEfficiencyStudyDesign[ii] ~ dnorm(0, 2^-2)
  }

  for(ii in 1:length(Marked)){
    logit(eEfficiency[ii]) <- bEfficiency
                            + bEfficiencyStudyDesign[StudyDesign[ii]]

    Resighted[ii] ~ dbin(eEfficiency[ii], Marked[ii])
  }
}

Abundance

Variable/Parameter Description
bDensityMarking Effect of Marking on log(eDensity)
bDensityRkmX Polynomial coefficients of effect of river kilometer on log(eDensity)
bDensitySite[ii] Effect of iith site on log(eDensity)
bDensityWidth Effect of site width on log(eDensity)
bDensityYear[yr] Estimate of log(eDensity) for yrth year
bEfficiencySwimmer[ii] Effect of iith swimmer on logit(eEfficiency)
bEfficiencyVisitStudy[ii, jj] Effect of iith visit type within jjth study design on logit(eEfficiency)
bSDispersion0 Estimate of log(eSDispersion)
bSDispersion1 Effect of bDensityYear on log(eSDispersion)
eAbundance[ii] Expected abundance of fish at site of iith visit
eCount[ii] Expected total number of fish at site of iith visit
eDensity[ii] Expected lineal density of fish at site of iith visit
eDispersion[ii] Expected overdispersion of Count[ii]
eEfficiency[ii] Expected observer efficiency on iith visit
eSDispersion[ii] Expected SD of overdispersion of Count[ii]
logit(bEfficiencyStudy[ii]) Effect of iith study design on logit(eEfficiency)
Marking Whether a site has been chosen as a marking site under the different study designs
Rkm[ii] River kilometer of iith visit
sDensitySite SD of effect of site on log(eDensity)
sEfficiencySwimmer SD of effect of swimmer on logit(eEfficiency)
Site[ii] Site of iith visit
SiteLength[ii] Length of site of iith visit
StudyDesign[ii] Study design of iith visit
SurveyProportion[ii] Proportion of site surveyed on iith visit
Swimmer[ii] Swimmer of iith visit
VisitType[ii] Visit type of iith visit
Width[ii] Site width of iith visit
Year[ii] Year of iith visit
Abundance - Model1
model{
    for(yr in 1:nYear){
    bDensityYear[yr] ~ dnorm(0, 5^-2)
  }

    bDensityWidth ~ dnorm(0, 2^-2)

    sDensitySite ~ dunif(0, 5)
  for(st in 1:nSite){
    bDensitySite[st] ~ dnorm(-sDensitySite^2 / 2, sDensitySite^-2)
  }

  bDensityMarking[1] <- 0
  for(mk in 2:nMarking) {
    bDensityMarking[mk] ~ dnorm(0, 5^-2)
  }

    sEfficiencySwimmer ~ dunif(0, 5)
  for(sw in 1:nSwimmer){
    bEfficiencySwimmer[sw] ~ dnorm(0, sEfficiencySwimmer^-2)
  }

    bDensityRkm1 ~ dnorm(0, 2^-2)
  bDensityRkm2 ~ dnorm(0, 2^-2)
  bDensityRkm3 ~ dnorm(0, 2^-2)
  bDensityRkm4 ~ dnorm(0, 2^-2)

    for(sd in 1:nStudyDesign){
    bEfficiencyStudy[sd] ~ dnorm(Efficiency[sd], Efficiency.sd[sd]^-2) T(Efficiency.lower[sd], Efficiency.upper[sd])
  }

    for(sd in 1:nStudyDesign){
    bEfficiencyVisitStudy[1, sd] <- 0
    for(vt in 2:nVisitType){
      bEfficiencyVisitStudy[vt, sd] ~ dnorm(0, 2^-2)
    }
  }

    bSDispersion0 ~ dnorm(0, 2^-2)
  bSDispersion1 ~ dnorm(0, 2^-2)

    for(ii in 1:length(Count)){
    log(eDensity[ii]) <- bDensityYear[Year[ii]]
                       + bDensityWidth * log(Width[ii])
                       + bDensitySite[Site[ii]]
                       + bDensityRkm1 * Rkm[ii]
                       + bDensityRkm2 * Rkm[ii]^2
                       + bDensityRkm3 * Rkm[ii]^3
                       + bDensityRkm4 * Rkm[ii]^4
                       + bDensityMarking[Marking[ii]]

    eAbundance[ii] <- eDensity[ii] * SiteLength[ii]

    logit(eEfficiency[ii]) <- logit(bEfficiencyStudy[StudyDesign[ii]])
                            + bEfficiencyVisitStudy[VisitType[ii], StudyDesign[ii]]
                            + bEfficiencySwimmer[Swimmer[ii]]

    eCount[ii] <- eAbundance[ii] * eEfficiency[ii] * SurveyProportion[ii]

    log(eSDispersion[ii]) <- bSDispersion0 + bSDispersion1 * bDensityYear[Year[ii]]

    eDispersion[ii] ~ dgamma(1/eSDispersion[ii]^2, 1/eSDispersion[ii]^2)

    Count[ii] ~ dpois(eCount[ii] * eDispersion[ii])
  }
}

Stock-Recruitment

Variable/Parameter Description
a Maximum reproductive performance per spawner
b Population size scaling parameter
eRecruits[i] Expected number of recruits in ith spawn year
Recruits[i] Number of recruits in ith spawn year
Spawners[i] Number of spawners in ith spawn year
sRecruits Standard deviation of residual variation in log(eRecruits)
Stock-Recruitment - Model1
model {

  a ~ dnorm(8000 * 0.5^4, 250^-2) T(0, )
  b ~ dunif(0, 0.1)

  sRecruits ~ dunif(0, 5)

  for(i in 1:length(Spawners)){
    eRecruits[i] <- a * Spawners[i] / (1 + Spawners[i] * b)
    Recruits[i] ~ dlnorm(log(eRecruits[i]), sRecruits^-2)
  }
}

Results

Stock-Recruitment

The maximum reproductive performance per spawner was estimated to be 603 (219 - 1050 95% CRI).

The environmental recruit carrying capacity was estimated to be 107,000 (50,500 - 211,200 95% CRI).

Model Parameters

The posterior distributions for the fixed (Kery and Schaub 2011 p. 75) parameters in each model are summarised below.

Capture Efficiency - Fry

Parameter Estimate Lower Upper SD Error Significance
bEfficiency 0.2055 -0.1089 0.5121 0.1565 150 0.1807
bEfficiencyStudyDesign[2] -0.9755 -1.3770 -0.5923 0.2051 40 0.0010
Convergence Iterations
1 5000

Abundance - Fry

Parameter Estimate Lower Upper SD Error Significance
bDensityMarking[2] 0.39970 0.0605 0.71770 0.16650 82 0.0232
bDensityMarking[3] 0.28220 0.0804 0.49100 0.10640 73 0.0097
bDensityRkm1 -0.47620 -0.6462 -0.31540 0.08600 35 0.0010
bDensityRkm2 0.28750 0.0482 0.50670 0.11830 80 0.0155
bDensityRkm3 0.05130 -0.0258 0.13300 0.04090 160 0.2068
bDensityRkm4 -0.19750 -0.2750 -0.11640 0.04110 40 0.0010
bDensityWidth 0.19940 0.0923 0.29220 0.05150 50 0.0010
bDensityYear[1] -0.20800 -0.8540 0.52800 0.34700 330 0.5392
bDensityYear[10] -2.78120 -3.1332 -2.44960 0.18230 12 0.0010
bDensityYear[2] -1.09200 -1.7830 -0.36700 0.36100 65 0.0020
bDensityYear[3] -0.77700 -1.4610 -0.07000 0.34800 90 0.0310
bDensityYear[4] -1.05900 -1.7700 -0.34100 0.36200 68 0.0020
bDensityYear[5] -1.02900 -1.7700 -0.23700 0.39200 75 0.0097
bDensityYear[6] -1.00100 -1.3650 -0.62590 0.18860 37 0.0010
bDensityYear[7] -1.03520 -1.3712 -0.69060 0.17940 33 0.0010
bDensityYear[8] -1.37640 -1.6902 -1.05010 0.16990 23 0.0010
bDensityYear[9] -1.83290 -2.1605 -1.48350 0.17540 18 0.0010
bEfficiencyStudy[1] 0.54680 0.4866 0.61120 0.03260 11 0.0010
bEfficiencyStudy[2] 0.31698 0.2688 0.36487 0.02528 15 0.0010
bEfficiencyVisitStudy[2,1] -1.61300 -2.4710 -0.78400 0.40800 52 0.0010
bEfficiencyVisitStudy[2,2] -0.33710 -0.6132 -0.02040 0.15150 88 0.0387
bSDispersion0 0.03960 -0.1712 0.29640 0.11250 590 0.7208
bSDispersion1 0.16260 0.0140 0.34770 0.08210 100 0.0329
sDensitySite 0.59180 0.4957 0.68120 0.04600 16 0.0010
sEfficiencySwimmer 0.44080 0.2471 0.76430 0.13670 59 0.0010
Convergence Iterations
1.05 20000

Stock-Recruitment

Parameter Estimate Lower Upper SD Error Significance
a 603.30000 218.90000 1049.60000 213.70000 69 0.001
b 0.00663 0.00143 0.01555 0.00382 110 0.001
sRecruits 0.79030 0.47830 1.39130 0.22950 58 0.001
Convergence Iterations
1 5e+05

Figures

Capture Efficiency - Fry

figures/efficiency/Fry/efficiency-studydesign.png
Figure 1. Predicted capture efficiency for Age-1 Rainbow Trout by study design (with 95% CRIs).

Abundance - Fry

figures/abundance/Fry/density-marking.png
Figure 2. Predicted lineal density of Age-1 Rainbow Trout by site marking type (with 95% CRIs).
figures/abundance/Fry/efficiency-type.png
Figure 3. Predicted observer efficiency for Age-1 Rainbow Trout by visit type and study design (with 95% CRIs).
figures/abundance/Fry/density-width.png
Figure 4. Predicted lineal density of Age-1 Rainbow Trout in 2014 by useable width (with 95% CRIs).
figures/abundance/Fry/density-site.png
Figure 5. Predicted lineal density of Age-1 Rainbow Trout by river kilometre (with 95% CRIs).
figures/abundance/Fry/density-year.png
Figure 6. Predicted lineal density of Age-1 Rainbow Trout by year (with 95% CRIs).
figures/abundance/Fry/abundance-year.png
Figure 7. Predicted abundance of Age-1 Rainbow Trout by year (with 95% CRIs).
figures/abundance/Fry/abundance-river-year.png
Figure 8. Predicted abundance of Age-1 Rainbow Trout by river and year (with 95% CRIs).

Stock-Recruitment

figures/sr/stock-recruitment.png
Figure 9. Predicted stock-recruitment relationship between AUC spawners and Age-1 recruits (with 95% CRIs).
figures/sr/recruits-per-spawner.png
Figure 10. Predicted reproductive performance per AUC spawner by AUC spawners (with 95% CRIs).
figures/sr/percent-carry-capacity.png
Figure 11. Predicted percent of Age-1 recruits carry capacity by AUC spawners (with 95% CRIs).

Acknowledgements

The organisations and individuals whose contributions have made this analysis report possible include:

References

Bradford, Michael J, Josh Korman, and Paul S Higgins. 2005. “Using Confidence Intervals to Estimate the Response of Salmon Populations (Oncorhynchus Spp.) to Experimental Habitat Alterations.” Canadian Journal of Fisheries and Aquatic Sciences 62 (12): 2716–26. https://doi.org/10.1139/f05-179.

Kery, Marc, and Michael Schaub. 2011. Bayesian Population Analysis Using WinBUGS : A Hierarchical Perspective. Boston: Academic Press. http://www.vogelwarte.ch/bpa.html.

Plummer, Martyn. 2015. “JAGS Version 4.0.1 User Manual.” http://sourceforge.net/projects/mcmc-jags/files/Manuals/4.x/.

R Core Team. 2015. “R: A Language and Environment for Statistical Computing.” Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org/.

Thorley, J. L. 2013. “Jaggernaut: An R Package to Facilitate Bayesian Analyses Using JAGS (Just Another Gibbs Sampler).” Nelson, B.C.: Poisson Consulting Ltd. https://github.com/poissonconsulting/jaggernaut.

Walters, Carl J., and Steven J. D. Martell. 2004. Fisheries Ecology and Management. Princeton, N.J: Princeton University Press.