R: Fits a stock-recruitment curve for spawning stock and recruitment vectors.

stock.recruit {FSA}

R Documentation

Fits a stock-recruitment curve for spawning stock and recruitment vectors.

Description

Evaluates astock-recruitment curve for spawning stock and recruitment vectors. Models considered are four parameterizations of the Beverton-Holt and two parameterizations of the Ricker stock-recruitment models.

Usage

stock.recruit(R,S,type=c("BevertonHolt","Ricker"),param=1,correct.Ricker=TRUE,
              log.errors=FALSE,start=NULL)

## S3 method for class 'SR':
plot(x,xlab="Stock Size",ylab="Recruits",pch=19,col.pt="black",col.mdl="red",
     lwd=2,lty=1,replace.line=FALSE,col.replace="green",lwd.replace=2,
     lty.replace=2,xlim=c(0,max.S),ylim=c(0,max.R),...)

## S3 method for class 'SR':
summary(object,what=c("Estimates","Model"),...)

## S3 method for class 'SR':
coef(object,...)

## S3 method for class 'SR':
anova(object,...)

Arguments

`R`	A numerical vector containing the number of recruits.
`S`	A numerical vector containing the number of spawners (stock).
`type`	A string indicating which stock-recruitment family of models to use. See details.
`param`	A numeric indicating which parameterization of the model in `type` to use. See details.
`correct.Ricker`	A logical indicating whether a bias correction factor should be used on the results of the second parameterization of the Ricker model (this argument is ignored or other models and parameterizations).
`log.errors`	A logical indicating that lognormal errors should be used when fitting the Beverton-Holt family of models. See details.
`start`	A list containing starting values for the non-linear model fitting routine if lognormal errors and a Beverton-Holt model are used. See details.
`object`	An object saved from the `stock.recruit` call (i.e., of class `SR`).
`x`	An object saved from the `stock.recruit` call (i.e., of class `SR`).
`xlab`	A string containing a label for the x-axis.
`ylab`	A string containing a label for the y-axis.
`pch`	A numeric used to indicate the type of plotting character.
`col.pt`	a string used to indicate the color of the plotted points.
`col.mdl`	a string used to indicate the color of the fitted model.
`lwd`	a numeric used to indicate the line width of the fitted model.
`lty`	a numeric used to indicate the type of line used for the fitted model.
`replace.line`	A logical indicating that a replacement line should be drawn on the plot (default is FALSE).
`col.replace`	a string used to indicate the color of the replacement line.
`lwd.replace`	a numeric used to indicate the line width of the replacement line.
`lty.replace`	a numeric used to indicate the type of line used for the replacement line.
`xlim`	A numeric vector containing the minimum and maximum values over which to plot the x-axis values.
`ylim`	A numeric vector containing the minimum and maximum values over which to plot the y-axis values.
`what`	a indicating the type of summary to compute. If `what="Model"` then the typical summary for a `lm` or `nls` object is printed.
`...`	Additional arguments for methods.

Details

The type argument is used to choose either the "BevertonHolt" or "Ricker" models. Different parameterizations of these two models are chosen with the param= argument. There are four paramaterizations of the Beverton-HOld model (where S=stock and R=recruits):

`param="1"`	R = frac{S}{a+bS}
`param="2"`	R = frac{aS}{b+S}
`param="3"`	R = frac{aS}{1+bS}
`param="4"`	R = frac{aS}{1+frac{a}{b}S}

and two parameterizations of the Ricker model

`param="1"`	R = aSe^{-bS}
`param="2"`	R = Se^{a(1-frac{S}{b})}

Value

A list with the following items

`S`	The numerical vector containing the number of recruits that was provided.
`R`	The numerical vector containing the number of spawners that was provided.
`type`	The family of models used.
`param`	The parameterization of the model used.
`meth`	A label for the type of method used.
`log.errors`	The provided logical of whether lognormal errors were used.
`correct.Ricker`	The provided logical of whether the bias correction factor was used.
`lm`	The fitted `lm` object. Set to `NULL` of not fit.
`nl`	The fitted `nls` object. Set to `NULL` of not fit.
`est`	A 2x1 numeric vector containing the parameter estimates.

Author(s)

Derek H. Ogle, dogle@northland.edu

References

Ricker, W.E. 1975. Computation and interpretation of biological statistics of fish populations. Technical Report Bulletin 191, Bulletin of the Fisheries Research Board of Canada.

Examples

data(CodNorwegian)
attach(CodNorwegian)

# default -- first Beverton-Holt parameterization
sr.bh1 <- stock.recruit(recruits,stock)
summary(sr.bh1)
summary(sr.bh1,what="Model")
coef(sr.bh1)
plot(sr.bh1)

# second Beverton-Holt parameterization
sr.bh2 <- stock.recruit(recruits,stock,param=2)
summary(sr.bh2)
plot(sr.bh2)

# second Beverton-Holt parameterization with lognormal errors
sr.bh3 <- stock.recruit(recruits,stock,param=2,log.errors=TRUE)
summary(sr.bh3)
plot(sr.bh3)

# first Ricker parameterization
sr.r1 <- stock.recruit(recruits,stock,type="Ricker")
summary(sr.r1)
plot(sr.r1)

# second Ricker parameterization with bias correction
sr.r2 <- stock.recruit(recruits,stock,type="Ricker")
summary(sr.r2)
plot(sr.r2)

# second Ricker parameterization without bias correction
sr.r3 <- stock.recruit(recruits,stock,type="Ricker",correct.Ricker=FALSE)
summary(sr.r3)
plot(sr.r3)

detach(CodNorwegian)

[Package FSA version 0.0-13 Index]