Package 'FSA'

Description

Adds zeros for catches of species that were not captured in a sampling event but were captured in at least one other sampling event (i.e., adds zeros to the data.frame for capture events where a species was not observed).

Usage

addZeroCatch(df, eventvar, specvar, zerovar, na.rm = TRUE)

Arguments

Details

The data.frame in df must contain a column that identifies a unique capture event (given in eventvar), a column with the name for the species captured (given in specvar), and a column that contains the number of that species captured (potentially given to zerovar; see details). All sampling event and species combinations where catch information does not exist is identified and a new data.frame that contains a zero for the catch for all of these combinations is created. This new data.frame is appended to the original data.frame to construct a data.frame that contains complete catch information – i.e., including zeros for species in events where that species was not captured.

The data.frame may contain other information related to the catch, such as number of recaptured fish, number of fish released, etc. These additional variables can be included in zerovar so that zeros will be added to these variables as well (e.g., if the catch of the species is zero, then the number of recaptures must also be zero). All variables not given in eventvar, specvar, or zerovar will be assumed to be related to eventvar and specvar (e.g., date, gear type, and habitat) and, thus, will be repeated with these variables.

In situations where no fish were captured in some events, the df may contain rows that have a value for eventvar but not for specvar. These rows are important because zeros need to be added for each observed species for these events. However, in these situations, a <NA> species will appear in the resulting data.frame. It is unlikely that these “missing” species are needed so they will be removed if na.rm=TRUE (DEFAULT) is used.

One should test the results of this function by creating a frequency table of the eventvar or specvar. In either case, the table should contain the same value in each cell of the table. See the examples.

Value

A data.frame with the same structure as df but with rows of zero observation data appended.

IFAR Chapter

2-Basic Data Manipulations

Note

An error will be returned if either specvar or eventvar are factors with any NA levels. This usually arises if the data.frame was subsetted/filtered prior to using addZeroCatch. See droplevels for descriptions of how to drop unused levels.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

See Also

complete in tidyr package.

Examples

## Example Data #1 (some nets missing some fish, ancillary net data)
df1 <- data.frame(net=c(1,1,1,2,2,3),
                  eff=c(1,1,1,1,1,1),
                  species=c("BKT","LKT","RBT","BKT","LKT","RBT"),
                  catch=c(3,4,5,5,4,3))
df1
# not all 1s
xtabs(~net+species,data=df1)

df1mod1 <- addZeroCatch(df1,"net","species",zerovar="catch")
df1mod1
# check, should all be 3
xtabs(~net,data=df1mod1)
# check, should all be 1
xtabs(~net+species,data=df1mod1)
# correct mean/sd of catches
Summarize(catch~species,data=df1mod1)
# incorrect mean/sd of catches (no zeros)
Summarize(catch~species,data=df1)

# Same as example 1 but with no ancillary data specific to the net number
df2 <- df1[,-2]
df2
df1mod2 <- addZeroCatch(df2,"net","species",zerovar="catch")
df1mod2
# check, should all be 1
xtabs(~net+species,data=df1mod2)

## Example Data #3 (All nets have same species ... no zeros needed)
df3 <- data.frame(net=c(1,1,1,2,2,2,3,3,3),
                  eff=c(1,1,1,1,1,1,1,1,1),
                  species=c("BKT","LKT","RBT","BKT","LKT",
                            "RBT","BKT","LKT","RBT"),
                  catch=c(3,4,5,5,4,3,3,2,7))
df3
# should all be 1 for this example
xtabs(~net+species,data=df3)

# should receive a warning and table should still all be 1
df3mod1 <- addZeroCatch(df3,"net","species",zerovar="catch")
xtabs(~net+species,data=df3mod1)

## Example Data #4 (another variable that needs zeros)
df4 <- df1
df4$recaps <- c(0,0,0,1,2,1)
df4
# not all 1s
xtabs(~net+species,data=df4)

df4mod1 <- addZeroCatch(df4,"net","species",zerovar=c("catch","recaps"))
# note zeros in both variables
df4mod1
# check, should all be 1
xtabs(~net+species,data=df4mod1)
# observe difference from next
Summarize(catch~species,data=df4)
Summarize(catch~species,data=df4mod1)
# observe difference from next
Summarize(recaps~species,data=df4)
Summarize(recaps~species,data=df4mod1)

## Example Data #5 (two "specvar"s)
df5 <- df1
df5$sex <- c("m","m","f","m","f","f")
df5
# not all 1s
xtabs(~sex+species+net,data=df5)

df5mod1 <- addZeroCatch(df5,"net",c("species","sex"),zerovar="catch")
df5mod1
# all 1s
xtabs(~sex+species+net,data=df5mod1)
str(df5mod1) 

## Example Data #6 (three "specvar"s)
df6 <- df5
df6$size <- c("lrg","lrg","lrg","sm","lrg","sm")
df6

df6mod1 <- addZeroCatch(df6,"net",c("species","sex","size"),zerovar="catch")
df6mod1

Description

Constructs age-agreement tables, statistical tests to detect differences, and plots to visualize potential differences in paired age estimates. Ages may be from, for example, two readers of the same structure, one reader at two times, two structures (e.g., scales, spines, otoliths), or one structure and known ages.

Usage

ageBias(
  formula,
  data,
  ref.lab = tmp$Enames,
  nref.lab = tmp$Rname,
  method = stats::p.adjust.methods,
  sig.level = 0.05,
  min.n.CI = 3
)

## S3 method for class 'ageBias'
summary(
  object,
  what = c("table", "symmetry", "Bowker", "EvansHoenig", "McNemar", "bias", "diff.bias",
    "n"),
  flip.table = FALSE,
  zero.print = "-",
  digits = 3,
  cont.corr = c("none", "Yates", "Edwards"),
  ...
)

## S3 method for class 'ageBias'
plot(
  x,
  xvals = c("reference", "mean"),
  xlab = ifelse(xvals == "reference", x$ref.lab, "Mean Age"),
  ylab = paste(x$nref.lab, "-", x$ref.lab),
  xlim = NULL,
  ylim = NULL,
  yaxt = graphics::par("yaxt"),
  xaxt = graphics::par("xaxt"),
  col.agree = "gray60",
  lwd.agree = lwd,
  lty.agree = 2,
  lwd = 1,
  sfrac = 0,
  show.pts = NULL,
  pch.pts = 20,
  cex.pts = ifelse(xHist | yHist, 1.5, 1),
  col.pts = "black",
  transparency = 1/10,
  show.CI = FALSE,
  col.CI = "black",
  col.CIsig = "red",
  lwd.CI = lwd,
  sfrac.CI = sfrac,
  show.range = NULL,
  col.range = ifelse(show.CI, "gray70", "black"),
  lwd.range = lwd,
  sfrac.range = sfrac,
  pch.mean = 19,
  pch.mean.sig = ifelse(show.CI | show.range, 21, 19),
  cex.mean = lwd,
  yHist = TRUE,
  xHist = NULL,
  hist.panel.size = 1/7,
  col.hist = "gray90",
  allowAdd = FALSE,
  ...
)

Arguments

Details

Generally, one of the two age estimates will be identified as the “reference” set. In some cases this may be the true ages, the ages from the more experienced reader, the ages from the first reading, or the ages from the structure generally thought to provide the most accurate results. In other cases, such as when comparing two novice readers, the choice may be arbitrary. The reference ages will form the columns of the age-agreement table and will be the “constant” age used in the t-tests and age bias plots (i.e., the x-axis). See further details below.

The age-agreement table is constructed with what="table" in summary. The agreement table can be “flipped” (i.e., the rows in descending rather than ascending order) with flip.table=TRUE. By default, the tables are shown with zeros replaced by dashes. This behavior can be changed with zero.print.

Three statistical tests of symmetry for the age-agreement table can be computed with what= in summary. The “unpooled” or Bowker test as described in Hoenig et al. (1995) is constructed with what="Bowker", the “semi-pooled” or Evans-Hoenig test as described in Evans and Hoenig (1998) is constructed with what="EvansHoenig", and the “pooled” or McNemar test as described in Evans and Hoenig (1998) is constructed with what="McNemar". All three tests are computed with what="symmetry".

The age bias plot introduced by Campana et al. (1995) can be constructed with plotAB. Muir et al. (2008) modified the original age bias plot by plotting the difference between the two ages on the y-axis (still against a reference age on the x-axis). McBride (2015) introduced the Bland-Altman plot for comparing fish ages where the difference in ages is plotted on the y-axis versus the mean of the ages on the x-axis. Modifications of these plots are created with plot (rather than plotAB) using xvals= to control what is plotted on the x-axis (i.e., xvals="reference" uses the reference ages, whereas xvals="mean" uses the mean of the two ages for the x-axis). In both plots, a horizontal agreement line at a difference of zero is shown for comparative purposes. In addition, a histogram of the differences is shown in the right margin (can be excluded with yHist=FALSE.) A histogram of the reference ages is shown by default in the top margin for the modified age bias plot, but not for the modified Bland-Altman-like plot (the presence of this histogram is controlled with xHist=).

By default, the modified age bias plot shows the mean and range for the nonreference ages at each of the reference ages. Means shown with an open dot are mean age differences that are significantly different from zero. The ranges of differences in ages at each reference age can be removed with show.range=FALSE. A confidence interval for difference in ages can be added with show.CI=FALSE. Confidence intervals are only shown if the sample size is greater than the value in min.n.CI= (also from the original call to ageBias). Confidence intervals plotted in red with an open dot (by default; these can be changed with col.CIsig and pch.mean.sig, respectively) do not contain the reference age (see discussion of t-tests below). Individual points are plotted if show.pts=TRUE, where there color is controlled by col.pts= and transparency=. See examples for use of these arguments.

The main (i.e., not the marginal histograms) can be "added to" after the plot is constructed if allowAdd=TRUE is used. For example, the Bland-Altman-like plot can be augmented with a horizontal line at the mean difference in ages, a line for the regression between the differences and the means, or a generalized additive model that describes the relationship between the differences and the means. See the examples for use of allowAdd=TRUE. It is recommended to save the plotting parameters (e.g., op <- par(no.readonly=TRUE)) before using plot with allowAdd=TRUE so that the original parameters can be reset (e.g., par(op)); see the examples.

Individual t-tests to determine if the mean age of the nonreference set at a particular age of the reference set is equal to the reference age (e.g., is the mean age of the nonreference set at age-3 of the reference set statistically different from 3?) are shown with what="bias" in summary. The results provide a column that indicates whether the difference is significant or not as determined by adjusted p-values from the t-tests and using the significance level provided in sig.level (defaults to 0.05). Similar results for the difference in ages (e.g., is the mean reference age minus nonreference age at a nonreference age of 3 different from 0?) are constructed with what="diff.bias" in summary.

The sample size present in the age-agreement table is found with what="n".

Value

ageBias returns a list with the following items:

• data A data.frame with the original paired age estimates and the difference between those estimates.

• agree The age-agreement table.

• bias A data.frame that contains the bias statistics.

• bias.diff A data.frame that contains the bias statistics for the differences.

• ref.lab A string that contains an optional label for the age estimates in the columns (reference) of the age-agreement table.

• nref.lab A string that contains an optional label for the age estimates in the rows (non-reference) of the age-agreement table.

summary returns the result if what= contains one item, otherwise it returns nothing. Nothing is returned by plot or plotAB, but see details for a description of the plot that is produced.

Testing

Tested all symmetry test results against results in Evans and Hoenig (2008), the McNemar and Evans-Hoenig results against results from compare2 in fishmethods, and all results using the AlewifeLH data set from FSAdata against results from the online resource at http://www.nefsc.noaa.gov/fbp/age-prec/.

IFAR Chapter

4-Age Comparisons. Note that plot has changed since IFAR was published. Some of the original functionality is in plotAB.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Campana, S.E., M.C. Annand, and J.I. McMillan. 1995. Graphical and statistical methods for determining the consistency of age determinations. Transactions of the American Fisheries Society 124:131-138. [Was (is?) available from http://www.bio.gc.ca/otoliths/documents/Campana%20et%20al%201995%20TAFS.pdf.]

Evans, G.T. and J.M. Hoenig. 1998. Testing and viewing symmetry in contingency tables, with application to readers of fish ages. Biometrics 54:620-629.

Hoenig, J.M., M.J. Morgan, and C.A. Brown. 1995. Analysing differences between two age determination methods by tests of symmetry. Canadian Journal of Fisheries and Aquatic Sciences 52:364-368.

McBride, R.S. 2015. Diagnosis of paired age agreement: A simulation approach of accuracy and precision effects. ICES Journal of Marine Science 72:2149-2167.

Muir, A.M., M.P. Ebener, J.X. He, and J.E. Johnson. 2008. A comparison of the scale and otolith methods of age estimation for lake whitefish in Lake Huron. North American Journal of Fisheries Management 28:625-635. [Was (is?) available from http://www.tandfonline.com/doi/abs/10.1577/M06-160.1]

See Also

See agePrecision for measures of precision between pairs of age estimates. See compare2 in fishmethods for similar functionality. See plotAB for a more traditional age-bias plot.

Examples

ab1 <- ageBias(scaleC~otolithC,data=WhitefishLC,
               ref.lab="Otolith Age",nref.lab="Scale Age")
summary(ab1)
summary(ab1,what="symmetry")
summary(ab1,what="Bowker")
summary(ab1,what="EvansHoenig")
summary(ab1,what="McNemar")
summary(ab1,what="McNemar",cont.corr="Yates")
summary(ab1,what="bias")
summary(ab1,what="diff.bias")
summary(ab1,what="n")
summary(ab1,what=c("n","symmetry","table"))
# flip table (easy to compare to age bias plot) and show zeroes (not dashes)
summary(ab1,what="table",flip.table=TRUE,zero.print="0")


#############################################################
## Differences Plot (inspired by Muir et al. (2008))
# Default (ranges, open circles for sig diffs, marginal hists)
plot(ab1)
# Show CIs for means (with and without ranges)
plot(ab1,show.CI=TRUE)
plot(ab1,show.CI=TRUE,show.range=FALSE)
# show points (with and without CIs)
plot(ab1,show.CI=TRUE,show.range=FALSE,show.pts=TRUE)
plot(ab1,show.range=FALSE,show.pts=TRUE)
# Use same symbols for all means (with ranges)
plot(ab1,pch.mean.sig=19)
# Use same symbols/colors for all means/CIs (without ranges)
plot(ab1,show.range=FALSE,show.CI=TRUE,pch.mean.sig=19,col.CIsig="black")
# Remove histograms
plot(ab1,xHist=FALSE)
plot(ab1,yHist=FALSE)
plot(ab1,xHist=FALSE,yHist=FALSE)
## Suppress confidence intervals for n < a certain value
##   must set this in the original ageBias() call
ab2 <- ageBias(scaleC~otolithC,data=WhitefishLC,min.n.CI=8,
               ref.lab="Otolith Age",nref.lab="Scale Age")
plot(ab2,show.CI=TRUE,show.range=FALSE)

 
#############################################################
## Differences Plot ( inspired by Bland-Altman plots in McBride (2015))
plot(ab1,xvals="mean")
## Modify axis limits
plot(ab1,xvals="mean",xlim=c(1,17))
## Add and remove histograms
plot(ab1,xvals="mean",xHist=TRUE)
plot(ab1,xvals="mean",xHist=TRUE,yHist=FALSE)
plot(ab1,xvals="mean",yHist=FALSE)

#############################################################
## Adding post hoc analyses to the main plot
# get original graphing parameters to be reset at the end
op <- par(no.readonly=TRUE)

# get raw data
tmp <- ab1$d
head(tmp)

# Add mean difference (w/ approx. 95% CI)
bias <- mean(tmp$diff)+c(-1.96,0,1.96)*se(tmp$diff)
plot(ab1,xvals="mean",xlim=c(1,17),allowAdd=TRUE)
abline(h=bias,lty=2,col="red")
par(op)

# Same as above, but without marginal histogram, and with
#   95% agreement lines as well (1.96SDs)
#   (this is nearly a replicate of a Bland-Altman plot)
bias <- mean(tmp$diff)+c(-1.96,0,1.96)*se(tmp$diff)
agline <- mean(tmp$diff)+c(-1.96,1.96)*sd(tmp$diff)
plot(ab1,xvals="mean",yHist=FALSE,allowAdd=TRUE)
abline(h=bias,lty=2,col="red")
abline(h=agline,lty=3,lwd=2,col="blue")
par(op)

# Add linear regression line of differences on means (w/ approx. 95% CI)
lm1 <- lm(diff~mean,data=tmp)
xval <- seq(0,19,0.1)
pred1 <- predict(lm1,data.frame(mean=xval),interval="confidence")
bias1 <- data.frame(xval,pred1)
head(bias1)
plot(ab1,xvals="mean",xlim=c(1,17),allowAdd=TRUE)
lines(lwr~xval,data=bias1,lty=2,col="red")
lines(upr~xval,data=bias1,lty=2,col="red")
lines(fit~xval,data=bias1,lty=2,col="red")
par(op)

# Add loess of differences on means (w/ approx. 95% CI as a polygon)
lo2 <- loess(diff~mean,data=tmp)
xval <- seq(min(tmp$mean),max(tmp$mean),0.1)
pred2 <- predict(lo2,data.frame(mean=xval),se=TRUE)
bias2 <- data.frame(xval,pred2)
bias2$lwr <- bias2$fit-1.96*bias2$se.fit
bias2$upr <- bias2$fit+1.96*bias2$se.fit
head(bias2)
plot(ab1,xvals="mean",xlim=c(1,17),allowAdd=TRUE)
with(bias2,polygon(c(xval,rev(xval)),c(lwr,rev(upr)),
                   col=col2rgbt("red",1/10),border=NA))
lines(fit~xval,data=bias2,lty=2,col="red")
par(op)
                  
# Same as above, but polygon and line behind the points
plot(ab1,xvals="mean",xlim=c(1,17),col.pts="white",allowAdd=TRUE)
with(bias2,polygon(c(xval,rev(xval)),c(lwr,rev(upr)),
                   col=col2rgbt("red",1/10),border=NA))
lines(fit~xval,data=bias2,lty=2,col="red")
points(diff~mean,data=tmp,pch=19,col=col2rgbt("black",1/8))
par(op)

# Can also be made with the reference ages on the x-axis
lo3 <- loess(diff~otolithC,data=tmp)
xval <- seq(min(tmp$otolithC),max(tmp$otolithC),0.1)
pred3 <- predict(lo3,data.frame(otolithC=xval),se=TRUE)
bias3 <- data.frame(xval,pred3)
bias3$lwr <- bias3$fit-1.96*bias3$se.fit
bias3$upr <- bias3$fit+1.96*bias3$se.fit
plot(ab1,show.range=FALSE,show.pts=TRUE,col.pts="white",allowAdd=TRUE)
with(bias3,polygon(c(xval,rev(xval)),c(lwr,rev(upr)),
                   col=col2rgbt("red",1/10),border=NA))
lines(fit~xval,data=bias3,lty=2,col="red")
points(diff~otolithC,data=tmp,pch=19,col=col2rgbt("black",1/8))
par(op)

Description

Computes overall measures of precision for multiple age estimates made on the same individuals. Ages may be from two or more readers of the same structure, one reader at two or more times, or two or more structures (e.g., scales, spines, otoliths). Measures of precision include ACV (Average Coefficient of Variation), APE (Average Percent Error), AAD (Average Absolute Deviation), and ASD (Average Standard Deviation), and various percentage difference values.

Usage

agePrecision(formula, data)

## S3 method for class 'agePrec'
summary(
  object,
  what = c("precision", "difference", "absolute difference", "details"),
  percent = TRUE,
  trunc.diff = NULL,
  digits = 4,
  show.prec2 = FALSE,
  ...
)

Arguments

Details

If what="precision" in summary then a summary table that contains the following items will be printed:

• n Number of fish in data.

• validn Number of fish in data that have non-NA data for all R age estimates.

• R Number of age estimates given in formula.

• PercAgree The percentage of fish for which all age estimates perfectly agree.

• ASD The average (across all fish) standard deviation of ages within a fish.

• ACV The average (across all fish) coefficient of variation of ages within a fish using the mean as the divisor. See the IFAR chapter for calculation details.

• ACV2 The average (across all fish) coefficient of variation of ages within a fish using the median as the divisor. This will only be shown if R>2 or show.prec2=TRUE.

• AAD The average (across all fish) absolute deviation of ages within a fish.

• APE The average (across all fish) percent error of ages within a fish using the mean as the divisor. See the IFAR chapter for calculation details.

• APE2 The average (across all fish) percent error of ages within a fish using the median as the divisor. This will only be shown if R>2 or show.prec2=TRUE.

• AD The average (across all fish) index of precision (D).

Note that ACV2 and APE2 will not be printed when what="precision" if only two sets of ages are given (because mean=median such that ACV=ACV2 and APE=APE2). If what="difference" is used in summary, then a table that describes either the percentage (if percent=TRUE, DEFAULT) or frequency of fish by the difference in paired age estimates. This table has one row for each possible pair of age estimates.

If what="absolute difference" is used in summary, then a table that describes either the percentage (if percent=TRUE, DEFAULT) or frequency of fish by the absolute value of the difference in paired age estimates. This table has one row for each possible pair of age estimates. The “1” column, for example, represents age estimates that disagree by one year (in either direction).

If what="detail" is used in summary, then a data.frame of the original data along with the intermediate calculations of the mean age, median age, modal age (will be NA if a mode does not exist (e.g., all different ages) or multiple modes exist), standard deviation of age (SD), coefficient of variation using the mean as the divisor (CV), coefficient of variation using the median as the divisor (CV2), absolute deviation using the mean as the divisor (AD), absolute deviation using the median as the divisor (AD2), average percent error (PE), and index of precision (D) for each individual will be printed.

All percentage calculations above use the validn value in the denominator.

Value

The main function returns a list with the following items:

• detail A data.frame with all data given in data and intermediate calculations for each fish. See details

• rawdiff A frequency table of fish by differences for each pair of ages.

• absdiff A frequency table of fish by absolute differences for each pair of ages.

• AAD The average absolute deviation.

• APE The average percent error (using the mean age as the divisor).

• APE2 The average percent error (using the median age as the divisor).

• ASD The average standard deviation.

• ACV The average coefficient of variation (using the mean age as the divisor).

• ACV2 The average coefficient of variation (using the median age as the divisor).

• AD The average index of precision.

• R The number of readings for each individual fish.

• n Number of fish in data.

• validn Number of fish in data that have non-NA data for all R age estimates.

The summary returns the result if what= contains only one item, otherwise it returns nothing. See details for what is printed.

Testing

Tested all precision results against published results in Herbst and Marsden (2011) for the WhitefishLC data and the results for the AlewifeLH data set from FSAdata against results from the online resource at http://www.nefsc.noaa.gov/fbp/age-prec/.

IFAR Chapter

4-Age Comparisons.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Beamish, R.J. and D.A. Fournier. 1981. A method for comparing the precision of a set of age determinations. Canadian Journal of Fisheries and Aquatic Sciences 38:982-983. [Was (is?) available from http://www.pac.dfo-mpo.gc.ca/science/people-gens/beamish/PDF_files/compareagecjfas1981.pdf.]

Campana, S.E. 1982. Accuracy, precision and quality control in age determination, including a review of the use and abuse of age validation methods. Journal of Fish Biology 59:197-242. [Was (is?) available from http://www.denix.osd.mil/nr/crid/Coral_Reef_Iniative_Database/References_for_Reef_Assessment_files/Campana,%202001.pdf.]

Campana, S.E., M.C. Annand, and J.I. McMillan. 1995. Graphical and statistical methods for determining the consistency of age determinations. Transactions of the American Fisheries Society 124:131-138. [Was (is?) available from http://www.bio.gc.ca/otoliths/documents/Campana%20et%20al%201995%20TAFS.pdf.]

Chang, W.Y.B. 1982. A statistical method for evaluating the reproducibility of age determination. Canadian Journal of Fisheries and Aquatic Sciences 39:1208-1210. [Was (is?) available from http://www.nrcresearchpress.com/doi/abs/10.1139/f82-158.]

McBride, R.S. 2015. Diagnosis of paired age agreement: A simulation approach of accuracy and precision effects. ICES Journal of Marine Science, 72:2149-2167.

See Also

See ageBias for computation of the full age agreement table, along with tests and plots of age bias.

Examples

## Example with just two age estimates
ap1 <- agePrecision(~otolithC+scaleC,data=WhitefishLC)
summary(ap1)
summary(ap1,what="precision")
summary(ap1,what="difference")
summary(ap1,what="difference",percent=FALSE)
summary(ap1,what="absolute")
summary(ap1,what="absolute",percent=FALSE)
summary(ap1,what="absolute",trunc.diff=4)
summary(ap1,what=c("precision","difference"))
summary(ap1,what="detail")

barplot(ap1$rawdiff,ylab="Frequency",xlab="Otolith - Scale Age")
plot(AD~mean,data=ap1$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Mean Age",ylab="Absolute Deviation Age")
plot(SD~mean,data=ap1$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Mean Age",ylab="Standard deviation Age")
plot(SD~AD,data=ap1$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Absolute Deviation Age",ylab="Standard deviation Age")
plot(CV~PE,data=ap1$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Percent Error Age",ylab="Coefficient of Variation Age")

## Example with three age estimates
ap2 <- agePrecision(~otolithC+finrayC+scaleC,data=WhitefishLC)
summary(ap2,digits=3)
summary(ap2,what="precision")
summary(ap2,what="difference")
summary(ap2,what="absolute",percent=FALSE,trunc.diff=4)
summary(ap2,what="detail",digits=3)

plot(AD~mean,data=ap2$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Mean Age",ylab="Absolute Deviation Age")
plot(SD~mean,data=ap2$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Mean Age",ylab="Standard Deviation Age")
plot(SD~AD,data=ap2$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Absolute Deviation Age",ylab="Standard Deviation Age")
plot(CV~PE,data=ap2$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Percent Error Age",ylab="Coefficient of Variation Age")
plot(median~mean,data=ap2$detail,pch=19,col=col2rgbt("black",1/5),
     xlab="Mean Age",ylab="Median Age")

Description

Computes the proportions-at-age (with standard errors) in a larger sample based on an age-length-key created from a subsample of ages through a two-stage random sampling design. Follows the methods in Quinn and Deriso (1999).

Usage

alkAgeDist(key, lenA.n, len.n)

Arguments

Details

The age-length key in key must have length intervals as rows and ages as columns. The row names of key (i.e., rownames(key)) must contain the minimum values of each length interval (e.g., if an interval is 100-109 then the corresponding row name must be 100). The column names of key (i.e., colnames(key)) must contain the age values (e.g., the columns can NOT be named with “age.1”, for example).

The length intervals in the rows of key must contain all of the length intervals present in the larger sample. Thus, the length of len.n must, at least, equal the number of rows in key. If this constraint is not met, then the function will stop with an error message.

The values in lenA.n are equal to what the row sums of key would have been before key was converted to a row proportions table. Thus, the length of lenA.n must also be equal to the number of rows in key. If this constraint is not met, then the function will stop with an error message.

Value

A data.frame with as many rows as ages (columns) present in key and the following three variables:

• age The ages.

• prop The proportion of fish at each age.

• se The SE for the proportion of fish at each age.

Testing

The results from this function perfectly match the results in Table 8.4 (left) of Quinn and Deriso (1999) using SnapperHG2 from FSAdata. The results also perfectly match the results from using alkprop in fishmethods.

IFAR Chapter

5-Age-Length Key.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Lai, H.-L. 1987. Optimum allocation for estimating age composition using age-length key. Fishery Bulletin, 85:179-185.

Lai, H.-L. 1993. Optimum sampling design for using the age-length key to estimate age composition of a fish population. Fishery Bulletin, 92:382-388.

Quinn, T. J. and R. B. Deriso. 1999. Quantitative Fish Dynamics. Oxford University Press, New York, New York. 542 pages.

See Also

See alkIndivAge and related functions for a completely different methodology. See alkprop from fishmethods for the exact same methodology but with a different format for the inputs.

Examples

## Example -- Even breaks for length categories
WR1 <- WR79
# add length intervals (width=5)
WR1$LCat <- lencat(WR1$len,w=5)
# get number of fish in each length interval in the entire sample
len.n <- xtabs(~LCat,data=WR1)
# isolate aged sample and get number in each length interval
WR1.age <- subset(WR1, !is.na(age))
lenA.n <- xtabs(~LCat,data=WR1.age)
# create age-length key
raw <- xtabs(~LCat+age,data=WR1.age)
( WR1.key <- prop.table(raw, margin=1) )

# use age-length key to estimate age distribution of all fish
alkAgeDist(WR1.key,lenA.n,len.n)

Description

Use either the semi- or completely-random methods from Isermann and Knight (2005) to assign ages to individual fish in the unaged sample according to the information in an age-length key supplied by the user.

Usage

alkIndivAge(
  key,
  formula,
  data,
  type = c("SR", "CR"),
  breaks = NULL,
  seed = NULL
)

Arguments

Details

The age-length key in key must have length intervals as rows and ages as columns. The row names of key (i.e., rownames(key)) must contain the minimum values of each length interval (e.g., if an interval is 100-109, then the corresponding row name must be 100). The column names of key (i.e., colnames(key)) must contain the age values (e.g., the columns can NOT be named with “age.1”, for example).

The length intervals in the rows of key must contain all of the length intervals present in the unaged sample to which the age-length key is to be applied (i.e., sent in the length portion of the formula). If this constraint is not met, then the function will stop with an error message.

If breaks=NULL, then the length intervals for the unaged sample will be determined with a starting interval at the minimum value of the row names in key and a width of the length intervals as determined by the minimum difference in adjacent row names of key. If length intervals of differing widths were used when constructing key, then those breaks should be supplied to breaks=. Use of breaks= may be useful when “uneven” length interval widths were used because the lengths in the unaged sample are not fully represented in the aged sample. See the examples.

Assigned ages will be stored in the column identified on the left-hand-side of formula (if the formula has both a left- and right-hand-side). If this variable is missing in formula, then the new column will be labeled with age.

Value

The original data.frame in data with assigned ages added to the column supplied in formula or in an additional column labeled as age. See details.

Testing

The type="SR" method worked perfectly on a small example. The type="SR" method provides results that reasonably approximate the results from alkAgeDist and alkMeanVar, which suggests that the age assessments are reasonable.

IFAR Chapter

5-Age-Length Key.

Author(s)

Derek H. Ogle, [email protected]. This is largely an R version of the SAS code provided by Isermann and Knight (2005).

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Isermann, D.A. and C.T. Knight. 2005. A computer program for age-length keys incorporating age assignment to individual fish. North American Journal of Fisheries Management, 25:1153-1160. [Was (is?) from http://www.tandfonline.com/doi/abs/10.1577/M04-130.1.]

See Also

See alkAgeDist and alkMeanVar for alternative methods to derived age distributions and mean (and SD) values for each age. See alkPlot for methods to visualize age-length keys.

Examples

## First Example -- Even breaks for length categories
WR1 <- WR79
# add length categories (width=5)
WR1$LCat <- lencat(WR1$len,w=5)
# isolate aged and unaged samples
WR1.age <- subset(WR1, !is.na(age))
WR1.len <- subset(WR1, is.na(age))
# note no ages in unaged sample
head(WR1.len)
# create age-length key
raw <- xtabs(~LCat+age,data=WR1.age)
( WR1.key <- prop.table(raw, margin=1) )
# apply the age-length key
WR1.len <- alkIndivAge(WR1.key,age~len,data=WR1.len)
# now there are ages
head(WR1.len)
# combine orig age & new ages
WR1.comb <- rbind(WR1.age, WR1.len)
# mean length-at-age
Summarize(len~age,data=WR1.comb,digits=2)
# age frequency distribution
( af <- xtabs(~age,data=WR1.comb) )
# proportional age distribution
( ap <- prop.table(af) )

## Second Example -- length sample does not have an age variable
WR2 <- WR79
# isolate age and unaged samples
WR2.age <- subset(WR2, !is.na(age))
WR2.len <- subset(WR2, is.na(age))
# remove age variable (for demo only)
WR2.len <- WR2.len[,-3]
# add length categories to aged sample
WR2.age$LCat <- lencat(WR2.age$len,w=5)
# create age-length key
raw <- xtabs(~LCat+age,data=WR2.age)
( WR2.key <- prop.table(raw, margin=1) )
# apply the age-length key
WR2.len <- alkIndivAge(WR2.key,~len,data=WR2.len)
# add length cat to length sample
WR2.len$LCat <- lencat(WR2.len$len,w=5)
head(WR2.len)
# combine orig age & new ages
WR2.comb <- rbind(WR2.age, WR2.len)
Summarize(len~age,data=WR2.comb,digits=2)

## Third Example -- Uneven breaks for length categories
WR3 <- WR79
# set up uneven breaks
brks <- c(seq(35,100,5),110,130)
WR3$LCat <- lencat(WR3$len,breaks=brks)
WR3.age <- subset(WR3, !is.na(age))
WR3.len <- subset(WR3, is.na(age))
head(WR3.len)
raw <- xtabs(~LCat+age,data=WR3.age)
( WR3.key <- prop.table(raw, margin=1) )
WR3.len <- alkIndivAge(WR3.key,age~len,data=WR3.len,breaks=brks)
head(WR3.len)
WR3.comb <- rbind(WR3.age, WR3.len)
Summarize(len~age,data=WR3.comb,digits=2)

Description

Computes the mean value-at-age in a larger sample based on an age-length-key created from a subsample of ages through a two-stage random sampling design. The mean values could be mean length-, weight-, or fecundity-at-age, for example. The methods of Bettoli and Miranda (2001) or Quinn and Deriso (1999) are used. A standard deviation is computed for the Bettoli and Miranda (2001) method and standard error for the Quinn and Deriso (1999) method. See the testing section notes.

Usage

alkMeanVar(
  key,
  formula,
  data,
  len.n,
  method = c("BettoliMiranda", "QuinnDeriso")
)

Arguments

Details

The age-length key key must have length intervals as rows and ages as columns. The row names of key (i.e., rownames(key)) must contain the minimum values of each length interval (e.g., if an interval is 100-109, then the corresponding row name must be 100). The column names of key (i.e., colnames(key)) must contain the age values (e.g., the columns can NOT be named with “age.1”, for example).

The length intervals in the rows of key must contain all of the length intervals present in the larger sample. Thus, the length of len.n must, at least, equal the number of rows in key. If this constraint is not met, then the function will stop with an error message.

Note that the function will stop with an error if the formula in formula does not meet the specific criteria outlined in the parameter list above.

Value

A data.frame with as many rows as ages (columns) present in key and the following three variables:

• age The ages.

• mean The mean value at each age.

• sd,se The SD if method="BettoliMiranda" or SE of the mean if method="QuinnDeriso" for the value at each age.

Testing

The results of these functions have not yet been rigorously tested. The Bettoli and Miranda (2001) results appear, at least, approximately correct when compared to the results from alkIndivAge. The Quinn and Deriso (1999) results appear at least approximately correct for the mean values, but do not appear to be correct for the SE values. Thus, a note is returned with the Quinn and Deriso (1999) results that the SE should not be trusted.

IFAR Chapter

5-Age-Length Key.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Bettoli, P. W. and Miranda, L. E. 2001. A cautionary note about estimating mean length at age with subsampled data. North American Journal of Fisheries Management, 21:425-428.

Quinn, T. J. and R. B. Deriso. 1999. Quantitative Fish Dynamics. Oxford University Press, New York, New York. 542 pages

See Also

See alkIndivAge and related functions for a completely different methodology. See alkAgeDist for a related method of determining the proportion of fish at each age. See the ALKr package.

Examples

## Example -- Even breaks for length categories
WR1 <- WR79
# add length intervals (width=5)
WR1$LCat <- lencat(WR1$len,w=5)
# get number of fish in each length interval in the entire sample
len.n <- xtabs(~LCat,data=WR1)
# isolate aged sample
WR1.age <- subset(WR1, !is.na(age))
# create age-length key
raw <- xtabs(~LCat+age,data=WR1.age)
( WR1.key <- prop.table(raw, margin=1) )

## use age-length key to estimate mean length-at-age of all fish
# Bettoli-Miranda method
alkMeanVar(WR1.key,len~LCat+age,WR1.age,len.n)

# Quinn-Deriso method
alkMeanVar(WR1.key,len~LCat+age,WR1.age,len.n,method="QuinnDeriso")

Description

Various plots to visualize the proportion of fish of certain ages within length intervals in an age-length key.

Usage

alkPlot(
  key,
  type = c("barplot", "area", "lines", "splines", "bubble"),
  xlab = "Length",
  ylab = ifelse(type != "bubble", "Proportion", "Age"),
  xlim = NULL,
  ylim = NULL,
  showLegend = FALSE,
  lbl.cex = 1.25,
  leg.cex = 1,
  lwd = 2,
  span = 0.25,
  grid = TRUE,
  col = NULL,
  buf = 0.45,
  add = FALSE,
  ...
)

Arguments

Details

A variety of plots can be used to visualize the proportion of fish of certain ages within length intervals of an age-length key. The types of plots are described below and illustrated in the examples.

• A “stacked” bar chart where vertical bars over length intervals sum to 1 but are segmented by the proportion of each age in that length interval is constructed with type="barplot". The ages will be labeled in the bar segments unless showLegend=TRUE is used.

• A “stacked” area chart similar to the bar chart described above is constructed with type="area".

• A plot with (differently colored) lines that connect the proportions of ages within each length interval is constructed with type="lines".

• A plot with (differently colored) lines, as estimated by loess splines, that connect the proportions of ages within each length interval is constructed with type="splines".

• A “bubble” plot where circles whose size is proportional to the proportion of fish of each age in each length interval is constructed with type="bubble". The color of the bubbles can be controlled with col= and an underlying grid for ease of seeing the age and length interval for each bubble can be controlled with grid=. Bubbles from a second age-length key can be overlaid on an already constructed bubble plot by using add=TRUE in a second call to alkPlot.

Note that all plots are “vertically conditional” – i.e., each represents the proportional ages WITHIN each length interval.

Value

None, but a plot is constructed.

IFAR Chapter

5-Age-Length Key.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

See Also

See alkIndivAge for using an age-length key to assign ages to individual fish. See hcl.colors for a simple way to choose other colors.

Examples

## Make an example age-length key
WR.age <- droplevels(subset(WR79, !is.na(age)))
WR.age$LCat <- lencat(WR.age$len,w=5)
raw <- xtabs(~LCat+age,data=WR.age)
WR.key <- prop.table(raw, margin=1)
round(WR.key,3)

## Various visualizations of the age-length key
alkPlot(WR.key,"barplot")
alkPlot(WR.key,"barplot",col="Cork")
alkPlot(WR.key,"barplot",col=heat.colors(8))
alkPlot(WR.key,"barplot",showLegend=TRUE)
alkPlot(WR.key,"area")
alkPlot(WR.key,"lines")
alkPlot(WR.key,"splines")
alkPlot(WR.key,"splines",span=0.2)
alkPlot(WR.key,"bubble")
alkPlot(WR.key,"bubble",col=col2rgbt("black",0.5))

Description

Uses one of three methods to compute a confidence interval for the probability of success (p) in a binomial distribution.

Usage

binCI(
  x,
  n,
  conf.level = 0.95,
  type = c("wilson", "exact", "asymptotic"),
  verbose = FALSE
)

Arguments

Details

This function will compute confidence interval for three possible methods chosen with the type argument.

Note that Agresti and Coull (2000) suggest that the Wilson interval is the preferred method and is, thus, the default type.

Value

A #x2 matrix that contains the lower and upper confidence interval bounds as columns and, if verbose=TRUE x, n, and x/n .

Author(s)

Derek H. Ogle, [email protected], though this is largely based on binom.exact, binom.wilson, and binom.approx from the old epitools package.

References

Agresti, A. and B.A. Coull. 1998. Approximate is better than “exact” for interval estimation of binomial proportions. American Statistician, 52:119-126.

See Also

See binom.test; binconf in Hmisc; and functions in binom.

Examples

## All types at once
binCI(7,20)

## Individual types
binCI(7,20,type="wilson")
binCI(7,20,type="exact")
binCI(7,20,type="asymptotic")
binCI(7,20,type="asymptotic",verbose=TRUE)

## Multiple types
binCI(7,20,type=c("exact","asymptotic"))
binCI(7,20,type=c("exact","asymptotic"),verbose=TRUE)

## Use with multiple inputs
binCI(c(7,10),c(20,30),type="wilson")
binCI(c(7,10),c(20,30),type="wilson",verbose=TRUE)

Description

Each line consists of the capture history over two samples of Bluegill (Lepomis macrochirus) in Jewett Lake (MI). This file contains the capture histories for only Bluegill larger than 6-in.

Format

A data frame with 277 observations on the following 2 variables.

first

a numeric vector of indicator variables for the first sample (1=captured)

second

a numeric vector of indicator variables for the second sample (1=captured)

Topic(s)

• Population Size

• Abundance

• Mark-Recapture

• Capture-Recapture

• Petersen

• Capture History

Source

From example 8.1 in Schneider, J.C. 1998. Lake fish population estimates by mark-and-recapture methods. Chapter 8 in Schneider, J.C. (ed.) 2000. Manual of fisheries survey methods II: with periodic updates. Michigan Department of Natural Resources, Fisheries Special Report 25, Ann Arbor. [Was (is?) from http://www.michigandnr.com/publications/pdfs/IFR/manual/SMII%20Chapter08.pdf.] CSV file

See Also

Used in mrClosed examples.

Examples

str(BluegillJL)
head(BluegillJL)

Description

Catch-at-age in fyke nets from 1996-1998 for “Coaster” Brook Trout (Salvelinus fontinalis) in Tobin Harbor, Isle Royale, Lake Superior.

Format

A data frame with 7 observations on the following 2 variables.

age

A numeric vector of assigned ages

catch

A numeric vector of number of Brook Trout caught

Topic(s)

• Mortality

• Catch Curve

• Chapman-Robson

Source

Quinlan, H.R. 1999. Biological Characteristics of Coaster Brook Trout at Isle Royale National Park, Michigan, 1996-98. U.S. Fish and Wildlife Service Ashland Fishery Resources Office report. November 1999. CSV file

See Also

Used in catchCurve and chapmanRobson examples.

Examples

str(BrookTroutTH)
head(BrookTroutTH)
plot(log(catch)~age,data=BrookTroutTH)

Description

Capitalizes the first letter of first or all words in a string.

Usage

capFirst(x, which = c("all", "first"))

Arguments

Value

A single string with the first letter of the first or all words capitalized.

Author(s)

Derek H. Ogle, [email protected]

Examples

## Capitalize first letter of all words (the default)
capFirst("Derek Ogle")
capFirst("derek ogle")
capFirst("derek")

## Capitalize first letter of only the first words
capFirst("Derek Ogle",which="first")
capFirst("derek ogle",which="first")
capFirst("derek",which="first")
## apply to all elements in a vector
vec <- c("Derek Ogle","derek ogle","Derek ogle","derek Ogle","DEREK OGLE")
capFirst(vec)
capFirst(vec,which="first")

## check class types
class(vec)
vec1 <- capFirst(vec)
class(vec1)
fvec <- factor(vec)
fvec1 <- capFirst(fvec)
class(fvec1)

Description

Use to convert between simple versions of several capture history data.frame formats – “individual”, “frequency”, “event”, “MARK”, and “RMark”. The primary use is to convert to the “individual” format for use in capHistSum.

Usage

capHistConvert(
  df,
  cols2use = NULL,
  cols2ignore = NULL,
  in.type = c("frequency", "event", "individual", "MARK", "marked", "RMark"),
  out.type = c("individual", "event", "frequency", "MARK", "marked", "RMark"),
  id = NULL,
  event.ord = NULL,
  freq = NULL,
  var.lbls = NULL,
  var.lbls.pre = "event",
  include.id = ifelse(is.null(id), FALSE, TRUE)
)

Arguments

Details

capHistSum requires capture histories to be recorded in the “individual” format. In this format, the data frame contains (at least) as many columns as sample events and as many rows as individually tagged fish. Optionally, the data.frame may also contain a column with unique fish identifiers (e.g., tag numbers). Each cell in the capture history portion of the data.frame contains a ‘0’ if the fish of that row was NOT seen in the event of that column and a ‘1’ if the fish of that row WAS seen in the event of that column. For example, suppose that five fish were marked on four sampling events; fish ‘17’ was captured on the first two events; fish ‘18’ was captured on the first and third events; fish ‘19’ was captured on only the third event; fish ‘20’ was captured on only the fourth event; and fish ‘21’ was captured on the first and second events. The “individual” capture history date.frame for these data looks like:

The “frequency” format data.frame (this format is used in Rcapture) has unique capture histories in separate columns, as in the “individual” format, but also includes a column with the frequency of individuals that had the capture history of that row. It will not contain a fish identifier variable. The same data from above looks like:

The “event” format data.frame has a column with the unique fish identifier and a column with the event in which the fish of that row was observed. The same data from above looks like:

MARK (http://www.phidot.org/software/mark/index.html) is the “gold-standard” software for analyzing complex capture history information. In the “MARK” format the 0s and 1s of the capture histories are combined together as a string without any spaces. Thus, the “MARK” format has the capture history strings in one column with an additional column that contains the frequency of individuals that exhibited the capture history of that row. The final column ends with a semi-colon. The same data from above looks like:

The RMark and marked are packages used to replace some of the functionality of MARK or to interact with MARK. The “RMark” or “marked” format requires the capture histories as one string (must be a character string and called ‘ch’), as in the “MARK” format, but without the semicolon. The data.frame may be augmented with an identifier for individual fish OR with a frequency variable. If augmented with a unique fish identification variable then the same data from above looks like:

However, if augmented with a frequency variable then the same data from above looks like:

Each of the formats can be used to convert from (i.e., in in.type=) or to convert to (i.e., in out.type=) with the exception that only the individual fish identifier version can be converted to when out.type="RMark".

Value

A data frame of the proper type given in out.type is returned. See details.

Warning

capHistConvert may give unwanted results if the data are in.type="event" but there are unused levels for the variable, as would result if the data.frame had been subsetted on the event variable. The unwanted results can be corrected by using droplevels before capHistConvert. See the last example for an example.

IFAR Chapter

9-Abundance from Capture-Recapture Data.

Note

The formats as used here are simple in the sense that one is only allowed to have the individual fish identifier or the frequency variable in addition to the capture history information. More complex analyses may use a number of covariates. For these more complex analyses, one should work directly with the Rcapture, RMark, or marked packages.

This function also assumes that all unmarked captured fish are marked and returned to the population (i.e., no losses at the time of marking are allowed).

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

See Also

See capHistSum to summarize “individual” capture histories into a format usable in mrClosed and mrOpen. Also see Rcapture, RMark, or marked packages for handling more complex analyses.

Examples

## A small example of 'event' format
( ex1 <- data.frame(fish=c(17,18,21,17,21,18,19,20),yr=c(1987,1987,1987,1988,1988,1989,1989,1990)) )
# convert to 'individual' format
( ex1.E2I <- capHistConvert(ex1,id="fish",in.type="event") )
# convert to 'frequency' format
( ex1.E2F <- capHistConvert(ex1,id="fish",in.type="event",out.type="frequency") )
# convert to 'MARK' format
( ex1.E2M <- capHistConvert(ex1,id="fish",in.type="event",out.type="MARK") )
# convert to 'RMark' format
( ex1.E2R <- capHistConvert(ex1,id="fish",in.type="event",out.type="RMark") )

## convert converted 'individual' format ...
# to 'frequency' format (must ignore "id")
( ex1.I2F <- capHistConvert(ex1.E2I,id="fish",in.type="individual",out.type="frequency") )
# to 'MARK' format
( ex1.I2M <- capHistConvert(ex1.E2I,id="fish",in.type="individual",out.type="MARK") )
# to 'RMark' format
( ex1.I2R <- capHistConvert(ex1.E2I,id="fish",in.type="individual",out.type="RMark") )
# to 'event' format
( ex1.I2E <- capHistConvert(ex1.E2I,id="fish",in.type="individual",out.type="event") )

#' ## convert converted 'frequency' format ...
# to 'individual' format
( ex1.F2I <- capHistConvert(ex1.E2F,freq="freq",in.type="frequency") )
( ex1.F2Ia <- capHistConvert(ex1.E2F,freq="freq",in.type="frequency",include.id=TRUE) )
# to 'Mark' format
( ex1.F2M <- capHistConvert(ex1.E2F,freq="freq",in.type="frequency",
                            out.type="MARK") )
# to 'RMark' format
( ex1.F2R <- capHistConvert(ex1.E2F,freq="freq",in.type="frequency",
                            out.type="RMark") )
( ex1.F2Ra <- capHistConvert(ex1.E2F,freq="freq",in.type="frequency",
                             out.type="RMark",include.id=TRUE) )
# to 'event' format
( ex1.F2E <- capHistConvert(ex1.E2F,freq="freq",in.type="frequency",
                            out.type="event") )

## convert converted 'MARK' format ...
# to 'individual' format
( ex1.M2I <- capHistConvert(ex1.E2M,freq="freq",in.type="MARK") )
( ex1.M2Ia <- capHistConvert(ex1.E2M,freq="freq",in.type="MARK",include.id=TRUE) )
# to 'frequency' format
( ex1.M2F <- capHistConvert(ex1.E2M,freq="freq",in.type="MARK",out.type="frequency") )
# to 'RMark' format
( ex1.M2R <- capHistConvert(ex1.E2M,freq="freq",in.type="MARK",out.type="RMark") )
( ex1.M2Ra <- capHistConvert(ex1.E2M,freq="freq",in.type="MARK",out.type="RMark",include.id=TRUE) )
# to 'event' format
( ex1.M2E <- capHistConvert(ex1.E2M,freq="freq",in.type="MARK",out.type="event") )
 
## convert converted 'RMark' format ...
# to 'individual' format
( ex1.R2I <- capHistConvert(ex1.E2R,id="fish",in.type="RMark") )
# to 'frequency' format
( ex1.R2F <- capHistConvert(ex1.E2R,id="fish",in.type="RMark",out.type="frequency") )
# to 'MARK' format
( ex1.R2M <- capHistConvert(ex1.E2R,id="fish",in.type="RMark",out.type="MARK") )
# to 'event' format
( ex1.R2E <- capHistConvert(ex1.E2R,id="fish",in.type="RMark",out.type="event") )

## Remove semi-colon from MARK format to make a RMark 'frequency' format
ex1.E2R1 <- ex1.E2M
ex1.E2R1$freq <- as.numeric(sub(";","",ex1.E2R1$freq))
ex1.E2R1
# convert this to 'individual' format
( ex1.R2I1 <- capHistConvert(ex1.E2R1,freq="freq",in.type="RMark") )
( ex1.R2I1a <- capHistConvert(ex1.E2R1,freq="freq",in.type="RMark",include.id=TRUE) )
# convert this to 'frequency' format
( ex1.R2F1 <- capHistConvert(ex1.E2R1,freq="freq",in.type="RMark",out.type="frequency") )
# convert this to 'MARK' format
( ex1.R2M1 <- capHistConvert(ex1.E2R1,freq="freq",in.type="RMark",out.type="MARK") )
# convert this to 'event' format
( ex1.R2E1 <- capHistConvert(ex1.E2R1,freq="freq",in.type="RMark",out.type="event") )


########################################################################
## A small example using character ids
( ex2 <- data.frame(fish=c("id17","id18","id21","id17","id21","id18","id19","id20"),
                    yr=c(1987,1987,1987,1988,1988,1989,1989,1990)) )
# convert to 'individual' format
( ex2.E2I <- capHistConvert(ex2,id="fish",in.type="event") )
# convert to 'frequency' format
( ex2.E2F <- capHistConvert(ex2,id="fish",in.type="event",out.type="frequency") )
# convert to 'MARK' format
( ex2.E2M <- capHistConvert(ex2,id="fish",in.type="event",out.type="MARK") )
# convert to 'RMark' format
( ex2.E2R <- capHistConvert(ex2,id="fish",in.type="event",out.type="RMark") )

## convert converted 'individual' format ...
# to 'frequency' format
( ex2.I2F <- capHistConvert(ex2.E2I,id="fish",in.type="individual",out.type="frequency") )
# to 'MARK' format
( ex2.I2M <- capHistConvert(ex2.E2I,id="fish",in.type="individual",out.type="MARK") )
# to 'RMark' format
( ex2.I2R <- capHistConvert(ex2.E2I,id="fish",in.type="individual",out.type="RMark") )
# to 'event' format
( ex2.I2E <- capHistConvert(ex2.E2I,id="fish",in.type="individual",out.type="event") )

## demo use of var.lbls
( ex2.E2Ia <- capHistConvert(ex2,id="fish",in.type="event",var.lbls.pre="Sample") )
( ex2.E2Ib <- capHistConvert(ex2,id="fish",in.type="event",
              var.lbls=c("first","second","third","fourth")) )

## demo use of event.ord
( ex2.I2Ea <- capHistConvert(ex2.E2Ib,id="fish",in.type="individual",out.type="event") )
( ex2.E2Ibad <- capHistConvert(ex2.I2Ea,id="fish",in.type="event") )
( ex2.E2Igood <- capHistConvert(ex2.I2Ea,id="fish",in.type="event",
                 event.ord=c("first","second","third","fourth")) )

## ONLY RUN IN INTERACTIVE MODE
if (interactive()) {

########################################################################
## A larger example of 'frequency' format (data from Rcapture package)
data(bunting,package="Rcapture")
head(bunting)
# convert to 'individual' format
bun.F2I <- capHistConvert(bunting,in.type="frequency",freq="freq")
head(bun.F2I)
# convert to 'MARK' format
bun.F2M <- capHistConvert(bunting,id="id",in.type="frequency",freq="freq",out.type="MARK")
head(bun.F2M)
# convert converted 'individual' back to 'MARK' format
bun.I2M <- capHistConvert(bun.F2I,id="id",in.type="individual",out.type="MARK")
head(bun.I2M)
# convert converted 'individual' back to 'frequency' format
bun.I2F <- capHistConvert(bun.F2I,id="id",in.type="individual",
           out.type="frequency",var.lbls.pre="Sample")
head(bun.I2F)


########################################################################
## A larger example of 'marked' or 'RMark' format, but with a covariate
##   and when the covariate is removed there is no frequency or individual
##   fish identifier.
data(dipper,package="marked")
head(dipper)
# isolate males and females
dipperF <- subset(dipper,sex=="Female")
dipperM <- subset(dipper,sex=="Male")
# convert females to 'individual' format
dipF.R2I <- capHistConvert(dipperF,cols2ignore="sex",in.type="RMark")
head(dipF.R2I)
# convert males to 'individual' format
dipM.R2I <- capHistConvert(dipperM,cols2ignore="sex",in.type="RMark")
head(dipM.R2I)
# add sex variable to each data.frame and then combine
dipF.R2I$sex <- "Female"
dipM.R2I$sex <- "Male"
dip.R2I <- rbind(dipF.R2I,dipM.R2I)
head(dip.R2I)
tail(dip.R2I)

} # end interactive


## An example of problem with unused levels
## Create a set of test data with several groups
( df <- data.frame(fish=c("id17","id18","id21","id17","id21","id18","id19","id20","id17"),
                   group=c("B1","B1","B1","B2","B2","B3","B4","C1","C1")) )
#  Let's assume the user wants to subset the data from the "B" group
( df1 <- subset(df,group %in% c("B1","B2","B3","B4")) )
#  Looks like capHistConvert() is still using the unused factor
#  level from group C
capHistConvert(df1,id="fish",in.type="event")
# use droplevels() to remove the unused groups and no problem
df1 <- droplevels(df1)
capHistConvert(df1,id="fish",in.type="event")

Description

Use to summarize a capture history data file that is in the “individual” fish format (see capHistConvert for a discussion of data file format types). Summarized capture history results may be used in the Lincoln-Petersen, Schnabel, Schumacher-Eschmeyer, or Jolly-Seber methods for estimating population abundance (see mrClosed and mrOpen).

Usage

capHistSum(df, cols2use = NULL, cols2ignore = NULL)

is.CapHist(x)

## S3 method for class 'CapHist'
plot(x, what = c("u", "f"), pch = 19, cex.pch = 0.7, lwd = 1, ...)

Arguments

Details

This function requires the capture history data file to be in the “individual” fish format. See capHistConvert for a description of this (and other) formats and for methods to convert from other formats to the “individual” fish format. In addition, this function requires only the capture history portion of the data file. Thus, if df contains columns with non-capture history information (e.g., fish ID, length, location, etc.) then use cols2use= to identify which columns contain only the capture history information. Columns to use can be identified by listing the column numbers (e.g., columns 2 through 7 could be included with cols2use=2:7). In many instances it may be easier to identify columns to exclude which can be done by preceding the column number by a negative sign (e.g., columns 1 through 3 are excluded with cols2use=-(1:3)).

The object returned from this function can be used directly in mrClosed and mrOpen. See examples of this functionality on the help pages for those functions.

The plot function can be used to construct the two diagnostic plots described by Baillargeon and Rivest (2007). The what="f" plot will plot the log of the number of fish seen i times divided by choose(t,i) against i. The what="u" plot will plot the log of the number of fish seen for the first time on event i against i. Baillargeon and Rivest (2007) provide a table that can be used to diagnosed types of heterogeneities in capture probabilities from these plots.

Value

If the capture history data file represents only two samples, then a list with the following two components is returned.

• caphist A vector summarizing the frequency of fish with each capture history.

• sum A data.frame that contains the number of marked fish from the first sample (M), the number of captured fish in the second sample (n), and the number of recaptured (i.e. previously marked) fish in the second sample (m).

If the capture history data file represents more than two samples, then a list with the following five components is returned

• caphist A vector summarizing the frequency of fish with each capture history.

• sum A data frame that contains the the number of captured fish in the ith sample (n), the number of recaptured (i.e. previously marked) fish in the ith sample (m), the number of marked fish returned to the population following the ith sample (R; this will equal n as the function currently does not handle mortalities); the number of marked fish in the population prior to the ith sample (M); the number of fish first seen in the ith sample (u); the number of fish last seen in the ith sample (v); and the number of fish seen i times (f).

• methodB.top A matrix that contains the top of the Method B table used for the Jolly-Seber method (i.e., a contingency table of capture sample (columns) and last seen sample (rows)).

• methodB.bot A data.frame that contains the bottom of the Method B table used for the Jolly-Seber method (i.e., the number of marked fish in the sample (m), the number of unmarked fish in the sample (u), the total number of fish in the sample (n), and the number of marked fish returned to the population following the sample (R).

• m.array A matrix that contains the the so-called “m-array”. The first column contains the number of fish captured on the ith event. The columns labeled with “cX” prefix show the number of fish originally captured in the ith row that were captured in the Xth event. The last column shows the number of fish originally captured in the ith row that were never recaptured.

IFAR Chapter

9-Abundance from Capture-Recapture Data.

Note

This function assumes that all unmarked captured fish are marked and returned to the population (i.e., no losses at the time of marking are allowed).

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Baillargeon, S. and Rivest, L.-P. (2007). Rcapture: Loglinear models for capture-recapture in R. Journal of Statistical Software, 19(5):1-31.

See Also

See descriptive in Rcapture for m.array and some of the same values in sum. See capHistConvert for a descriptions of capture history data file formats and how to convert between them. See mrClosed and mrOpen for how to estimate abundance from the summarized capture history information.

Examples

# data.frame with IDs in the first column
head(PikeNYPartial1)

# Three ways to ignore first column of ID numbers
( ch1 <- capHistSum(PikeNYPartial1,cols2use=-1) )
( ch1 <- capHistSum(PikeNYPartial1,cols2ignore=1) )
( ch1 <- capHistSum(PikeNYPartial1,cols2ignore="id") )

# diagnostic plots
plot(ch1)
plot(ch1,what="f")
plot(ch1,what="u")

# An examle with only two sample events (for demonstration only)
( ch2 <- capHistSum(PikeNYPartial1,cols2use=-c(1,4:5)) )
( ch2 <- capHistSum(PikeNYPartial1,cols2use=2:3) )
( ch2 <- capHistSum(PikeNYPartial1,cols2ignore=c(1,4:5)) )

Description

Fits a linear model to the user-defined descending limb of a catch curve. Method functions extract estimates of the instantaneous (Z) and total annual (A) mortality rates with associated standard errors and confidence intervals. A plot method highlights the descending limb, shows the linear model on the descending limb, and, optionally, prints the estimated Z and A.

Usage

catchCurve(x, ...)

## Default S3 method:
catchCurve(
  x,
  catch,
  ages2use = age,
  weighted = FALSE,
  negWeightReplace = 0,
  ...
)

## S3 method for class 'formula'
catchCurve(
  x,
  data,
  ages2use = age,
  weighted = FALSE,
  negWeightReplace = 0,
  ...
)

## S3 method for class 'catchCurve'
summary(object, parm = c("both", "all", "Z", "A", "lm"), ...)

## S3 method for class 'catchCurve'
coef(object, parm = c("all", "both", "Z", "A", "lm"), ...)

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

## S3 method for class 'catchCurve'
confint(
  object,
  parm = c("all", "both", "Z", "A", "lm"),
  level = conf.level,
  conf.level = 0.95,
  ...
)

## S3 method for class 'catchCurve'
rSquared(object, digits = getOption("digits"), percent = FALSE, ...)

## S3 method for class 'catchCurve'
plot(
  x,
  pos.est = "topright",
  cex.est = 0.95,
  round.est = c(3, 1),
  ylab = "log(Catch)",
  xlab = "Age",
  ylim = NULL,
  col.pt = "gray30",
  col.mdl = "black",
  lwd = 2,
  lty = 1,
  ...
)

Arguments

Details

The default is to use all ages in the age vector. This is appropriate only when the age and catch vectors contain only the ages and catches on the descending limb of the catch curve. Use ages2use to isolate only the catch and ages on the descending limb.

If weighted=TRUE then a weighted regression is used where the weights are the log(number) at each age predicted from the unweighted regression of log(number) on age (as proposed by Maceina and Bettoli (1998)). If a negative weight is computed it will be changed to the value in negWeightReplace and a warning will be issued.

Value

A list that contains the following items:

• age The original vector of assigned ages.

• catch The original vector of observed catches or CPUEs.

• age.e A vector of assigned ages for which the catch curve was fit.

• log.catch.e A vector of log catches or CPUEs for which the catch curve was fit.

• W A vector of weights used in the catch curve fit. Will be NULL unless weighted=TRUE.

• lm An lm object from the fit to the ages and log catches or CPUEs on the descending limb (i.e., in age.e and log.catch.e).

Testing

Tested the results of catch curve, both unweighted and weighted, against the results in Miranda and Bettoli (2007). Results for Z and the SE of Z matched perfectly. Tested the unweighted results against the results from agesurv in fishmethods using the rockbass data.frame in fishmethods. Results for Z and the SE of Z matched perfectly.

IFAR Chapter

11-Mortality.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Maceina, M.J., and P.W. Bettoli. 1998. Variation in Largemouth Bass recruitment in four mainstream impoundments on the Tennessee River. North American Journal of Fisheries Management 18:998-1003.

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. [Was (is?) from http://www.dfo-mpo.gc.ca/Library/1485.pdf.]

See Also

See agesurv in fishmethods for similar functionality. See chapmanRobson and agesurvcl in fishmethods for alternative methods to estimate mortality rates. See metaM for empirical methods to estimate natural mortality.

Examples

plot(catch~age,data=BrookTroutTH,pch=19)

## demonstration of formula notation
cc1 <- catchCurve(catch~age,data=BrookTroutTH,ages2use=2:6)
summary(cc1)
cbind(Est=coef(cc1),confint(cc1))
rSquared(cc1)
plot(cc1)
summary(cc1,parm="Z")
cbind(Est=coef(cc1,parm="Z"),confint(cc1,parm="Z"))

## demonstration of excluding ages2use
cc2 <- catchCurve(catch~age,data=BrookTroutTH,ages2use=-c(0,1))
summary(cc2)
plot(cc2)

## demonstration of using weights
cc3 <- catchCurve(catch~age,data=BrookTroutTH,ages2use=2:6,weighted=TRUE)
summary(cc3)
plot(cc3)

## demonstration of returning the linear model results
summary(cc3,parm="lm")
cbind(Est=coef(cc3,parm="lm"),confint(cc3,parm="lm"))

## demonstration of ability to work with missing age classes
df <- data.frame(age=c(  2, 3, 4, 5, 7, 9,12),
                 ct= c(100,92,83,71,56,35, 1))
cc4 <- catchCurve(ct~age,data=df,ages2use=4:12)
summary(cc4)
plot(cc4)

## demonstration of ability to work with missing age classes
## evein if catches are recorded as NAs
df <- data.frame(age=c(  2, 3, 4, 5, 6, 7, 8, 9,10,11,12),
                 ct= c(100,92,83,71,NA,56,NA,35,NA,NA, 1))
cc5 <- catchCurve(ct~age,data=df,ages2use=4:12)
summary(cc5)
plot(cc5)

Description

Computes the Chapman-Robson estimates of annual survival rate (S) and instantaneous mortality rate (Z) from catch-at-age data on the descending limb of a catch-curve. Method functions extract estimates with associated standard errors and confidence intervals. A plot method highlights the descending-limb, shows the linear model on the descending limb, and, optionally, prints the estimated Z and A.

Usage

chapmanRobson(x, ...)

## Default S3 method:
chapmanRobson(
  x,
  catch,
  ages2use = age,
  zmethod = c("Smithetal", "Hoenigetal", "original"),
  ...
)

## S3 method for class 'formula'
chapmanRobson(
  x,
  data,
  ages2use = age,
  zmethod = c("Smithetal", "Hoenigetal", "original"),
  ...
)

## S3 method for class 'chapmanRobson'
summary(object, parm = c("all", "both", "Z", "S"), verbose = FALSE, ...)

## S3 method for class 'chapmanRobson'
coef(object, parm = c("all", "both", "Z", "S"), ...)

## S3 method for class 'chapmanRobson'
confint(
  object,
  parm = c("all", "both", "S", "Z"),
  level = conf.level,
  conf.level = 0.95,
  ...
)

## S3 method for class 'chapmanRobson'
plot(
  x,
  pos.est = "topright",
  cex.est = 0.95,
  round.est = c(3, 1),
  ylab = "Catch",
  xlab = "Age",
  ylim = NULL,
  col.pt = "gray30",
  axis.age = c("both", "age", "recoded age"),
  ...
)

Arguments

Details

The default is to use all ages in the age vector. This is only appropriate if the age and catch vectors contain only the ages and catches on the descending limb of the catch curve. Use ages2use to isolate only the catch and ages on the descending limb.

The Chapman-Robson method provides an estimate of the annual survival rate, with the annual mortality rate (A) determined by 1-S. The instantaneous mortality rate is often computed as -log(S). However, Hoenig et al. (1983) showed that this produced a biased (over)estimate of Z and provided a correction. The correction is applied by setting zmethod="Hoenigetal". Smith et al. (2012) showed that the Hoenig et al. method should be corrected for a variance inflation factor. This correction is applied by setting zmethod="Smithetal" (which is the default behavior). Choose zmethod="original" to use the original estimates for Z and it's SE as provided by Chapman and Robson.

Value

A list with the following items:

• age the original vector of assigned ages.

• catch the original vector of observed catches or CPUEs.

• age.e a vector of assigned ages used to estimate mortalities.

• catch.e a vector of catches or CPUEs used to estimate mortalities.

• age.r a vector of recoded ages used to estimate mortalities. See references.

• n a numeric holding the intermediate calculation of n. See references.

• T a numeric holding the intermediate calculation of T. See references.

• est A 2x2 matrix that contains the estimates and standard errors for S and Z.

Testing

Tested the results of chapmanRobson against the results in Miranda and Bettoli (2007). The point estimates of S matched perfectly but the SE of S did not because Miranda and Bettoli used a rounded estimate of S in the calculation of the SE of S but chapmanRobson does not.

Tested the results against the results from agesurv in fishmethods using the rockbass data.frame in fishmethods. Results for Z and the SE of Z matched perfectly for non-bias-corrected results. The estimate of Z, but not the SE of Z, matched for the bias-corrected (following Smith et al. (2012)) results. FSA uses equation 2 from Smith et al. (2012) whereas fishmethods appears to use equation 5 from the same source to estimate the SE of Z.

IFAR Chapter

11-Mortality.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

Chapman, D.G. and D.S. Robson. 1960. The analysis of a catch curve. Biometrics. 16:354-368.

Hoenig, J.M. and W.D. Lawing, and N.A. Hoenig. 1983. Using mean age, mean length and median length data to estimate the total mortality rate. International Council for the Exploration of the Sea, CM 1983/D:23, Copenhagen.

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. [Was (is?) from http://www.dfo-mpo.gc.ca/Library/1485.pdf.]

Robson, D.S. and D.G. Chapman. 1961. Catch curves and mortality rates. Transactions of the American Fisheries Society. 90:181-189.

Smith, M.W., A.Y. Then, C. Wor, G. Ralph, K.H. Pollock, and J.M. Hoenig. 2012. Recommendations for catch-curve analysis. North American Journal of Fisheries Management. 32:956-967.

See Also

See agesurv in fishmethods for similar functionality. See catchCurve and agesurvcl in fishmethods for alternative methods. See metaM for empirical methods to estimate natural mortality.

Examples

plot(catch~age,data=BrookTroutTH,pch=19)

## demonstration of formula notation
cr1 <- chapmanRobson(catch~age,data=BrookTroutTH,ages2use=2:6)
summary(cr1)
summary(cr1,verbose=TRUE)
cbind(Est=coef(cr1),confint(cr1))
plot(cr1)
plot(cr1,axis.age="age")
plot(cr1,axis.age="recoded age")
summary(cr1,parm="Z")
cbind(Est=coef(cr1,parm="Z"),confint(cr1,parm="Z"))

## demonstration of excluding ages2use
cr2 <- chapmanRobson(catch~age,data=BrookTroutTH,ages2use=-c(0,1))
summary(cr2)
plot(cr2)

## demonstration of ability to work with missing age classes
age <- c(  2, 3, 4, 5, 7, 9,12)
ct  <- c(100,92,83,71,56,35, 1)
cr3 <- chapmanRobson(age,ct,4:12)
summary(cr3)
plot(cr3)

Description

Lengths and weights for Chinook Salmon from three locations in Argentina.

Format

A data frame with 112 observations on the following 3 variables:

tl

Total length (cm)

w

Weight (kg)

loc

Capture location (Argentina, Petrohue, Puyehue)

Topic(s)

• Weight-Length

Source

From Figure 4 in Soto, D., I. Arismendi, C. Di Prinzio, and F. Jara. 2007. Establishment of Chinook Salmon (Oncorhynchus tshawytscha) in Pacific basins of southern South America and its potential ecosystem implications. Revista Chilena d Historia Natural, 80:81-98. [Was (is?) from http://www.scielo.cl/pdf/rchnat/v80n1/art07.pdf.] CSV file

See Also

Used in lwCompPreds examples.

Examples

str(ChinookArg)
head(ChinookArg)
op <- par(mfrow=c(2,2),pch=19,mar=c(3,3,0.5,0.5),mgp=c(1.9,0.5,0),tcl=-0.2)
plot(w~tl,data=ChinookArg,subset=loc=="Argentina")
plot(w~tl,data=ChinookArg,subset=loc=="Petrohue")
plot(w~tl,data=ChinookArg,subset=loc=="Puyehue")
par(op)

Description

Norwegian cod (Gadus morhua) stock and recruitment by year, 1937-1960.

Format

A data frame of 24 observations on the following 3 variables:

year

Year of data

recruits

Recruits – year-class strength index

stock

Spawning stock index

Topic(s)

• Stock-Recruit

• Recruitment

Source

From Garrod, D.J. 1967. Population dynamics of the Arcto-Norwegian Cod. Journal of the Fisheries Research Board of Canada, 24:145-190. CSV file

See Also

Used in srStarts, srFuns, and nlsTracePlot examples.

Examples

str(CodNorwegian)
head(CodNorwegian)
op <- par(mfrow=c(1,2),pch=19,mar=c(3,3,0.5,0.5),mgp=c(1.9,0.5,0),tcl=-0.2)
plot(recruits~year,data=CodNorwegian,type="l")
plot(recruits~stock,data=CodNorwegian)
par(op)

Description

Converts an R color to RGB (red/green/blue) including a transparency (alpha channel). Similar to col2rgb except that a transparency (alpha channel) can be included.

Usage

col2rgbt(col, transp = 1)

Arguments

Value

A vector of hexadecimal strings of the form "#rrggbbaa" as would be returned by rgb.

Author(s)

Derek H. Ogle, [email protected]

See Also

See col2rgb for similar functionality.

Examples

col2rgbt("black")
col2rgbt("black",1/4)
clrs <- c("black","blue","red","green")
col2rgbt(clrs)
col2rgbt(clrs,1/4)
trans <- (1:4)/5
col2rgbt(clrs,trans)

Description

S3 methods are provided to construct non-parametric bootstrap confidence intervals, predictions with non-parametric confidence intervals, hypothesis tests, and plots of the parameter estimates for objects returned from Boot from car.

Usage

## S3 method for class 'boot'
confint(
  object,
  parm = NULL,
  level = conf.level,
  conf.level = 0.95,
  type = c("bca", "norm", "basic", "perc"),
  plot = FALSE,
  err.col = "black",
  err.lwd = 2,
  rows = NULL,
  cols = NULL,
  ...
)

## S3 method for class 'boot'
htest(
  object,
  parm = NULL,
  bo = 0,
  alt = c("two.sided", "less", "greater"),
  plot = FALSE,
  ...
)

## S3 method for class 'boot'
predict(object, FUN, conf.level = 0.95, digits = NULL, ...)

## S3 method for class 'boot'
hist(
  x,
  same.ylim = TRUE,
  ymax = NULL,
  rows = round(sqrt(ncol(x$t))),
  cols = ceiling(sqrt(ncol(x$t))),
  ...
)

Arguments

Details

confint is largely a wrapper for Confint from car (see its manual page).

predict applies a user-supplied function to each row of object and then finds the median and the two quantiles that have the proportion (1-conf.level)/2 of the bootstrapped predictions below and above. The median is returned as the predicted value and the quantiles are returned as an approximate 100conf.level% confidence interval for that prediction. Values for the independent variable in FUN must be a named argument sent in the ... argument (see examples). Note that if other arguments are needed in FUN besides values for the independent variable, then these are included in the ... argument AFTER the values for the independent variable.

In htest the “direction” of the alternative hypothesis is identified by a string in the alt= argument. The strings may be "less" for a “less than” alternative, "greater" for a “greater than” alternative, or "two.sided" for a “not equals” alternative (the DEFAULT). In the one-tailed alternatives the p-value is the proportion of bootstrapped parameter estimates in object$coefboot that are extreme of the null hypothesized parameter value in bo. In the two-tailed alternative the p-value is twice the smallest of the proportion of bootstrapped parameter estimates above or below the null hypothesized parameter value in bo.

Value

If object is a matrix, then confint returns a matrix with as many rows as columns (i.e., parameter estimates) in object and two columns of the quantiles that correspond to the approximate confidence interval. If object is a vector, then confint returns a vector with the two quantiles that correspond to the approximate confidence interval.

htest returns a two-column matrix with the first column containing the hypothesized value sent to this function and the second column containing the corresponding p-value.

hist constructs histograms of the bootstrapped parameter estimates.

plot constructs scatterplots of all pairs of bootstrapped parameter estimates.

predict returns a matrix with one row and three columns, with the first column holding the predicted value (i.e., the median prediction) and the last two columns holding the approximate confidence interval.

Author(s)

Derek H. Ogle, [email protected]

References

S. Weisberg (2005). Applied Linear Regression, third edition. New York: Wiley, Chapters 4 and 11.

See Also

Boot in car.

Examples

fnx <- function(days,B1,B2,B3) {
  if (length(B1) > 1) {
    B2 <- B1[2]
    B3 <- B1[3]
    B1 <- B1[1]
  }
  B1/(1+exp(B2+B3*days))
}
nl1 <- nls(cells~fnx(days,B1,B2,B3),data=Ecoli,
           start=list(B1=6,B2=7.2,B3=-1.45))

if (require(car)) {
  nl1.bootc <- car::Boot(nl1,f=coef,R=99)  # R=99 too few to be useful
  confint(nl1.bootc,"B1")
  confint(nl1.bootc,c(2,3))
  confint(nl1.bootc,conf.level=0.90)
  confint(nl1.bootc,"B1",plot=TRUE)
  htest(nl1.bootc,1,bo=6,alt="less")
  htest(nl1.bootc,1,bo=6,alt="less",plot=TRUE)
  predict(nl1.bootc,fnx,days=1:3)
  predict(nl1.bootc,fnx,days=3)
  hist(nl1.bootc)
}

Description

Individual capture histories of Cutthroat Trout (Oncorhynchus clarki) in Auke Lake, Alaska, from samples taken in 1998-2006.

Format

A data frame with 1684 observations on the following 10 variables.

id

Unique identification numbers for each fish

y1998

Indicator variable for whether the fish was captured in 1998 (1=captured)

y1999

Indicator variable for whether the fish was captured in 1999 (1=captured)

y2000

Indicator variable for whether the fish was captured in 2000 (1=captured)

y2001

Indicator variable for whether the fish was captured in 2001 (1=captured)

y2002

Indicator variable for whether the fish was captured in 2002 (1=captured)

y2003

Indicator variable for whether the fish was captured in 2003 (1=captured)

y2004

Indicator variable for whether the fish was captured in 2004 (1=captured)

y2005

Indicator variable for whether the fish was captured in 2005 (1=captured)

y2006

Indicator variable for whether the fish was captured in 2006 (1=captured)

Topic(s)

• Population Size

• Abundance

• Mark-Recapture

• Capture-Recapture

• Jolly-Seber

• Capture History

Note

Entered into “RMark” format (see CutthroatALf in FSAdata) and then converted to individual format with capHistConvert

Source

From Appendix A.3 of Harding, R.D., C.L. Hoover, and R.P. Marshall. 2010. Abundance of Cutthroat Trout in Auke Lake, Southeast Alaska, in 2005 and 2006. Alaska Department of Fish and Game Fisheries Data Series No. 10-82. [Was (is?) from http://www.sf.adfg.state.ak.us/FedAidPDFs/FDS10-82.pdf.] CSV file

See Also

Used in mrOpen examples.

Examples

str(CutthroatAL)
head(CutthroatAL)

Description

Computes the Leslie or DeLury estimates of population size and catchability coefficient from paired catch and effort data. The Ricker modification may also be used.

Usage

depletion(
  catch,
  effort,
  method = c("Leslie", "DeLury", "Delury"),
  Ricker.mod = FALSE
)

## S3 method for class 'depletion'
summary(object, parm = c("all", "both", "No", "q", "lm"), verbose = FALSE, ...)

## S3 method for class 'depletion'
coef(object, parm = c("all", "both", "No", "q", "lm"), ...)

## S3 method for class 'depletion'
confint(
  object,
  parm = c("all", "both", "No", "q", "lm"),
  level = conf.level,
  conf.level = 0.95,
  ...
)

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

## S3 method for class 'depletion'
rSquared(object, digits = getOption("digits"), percent = FALSE, ...)

## S3 method for class 'depletion'
plot(
  x,
  xlab = NULL,
  ylab = NULL,
  pch = 19,
  col.pt = "black",
  col.mdl = "gray70",
  lwd = 1,
  lty = 1,
  pos.est = "topright",
  cex.est = 0.95,
  ...
)

Arguments

Details

For the Leslie method, a linear regression model of catch-per-unit-effort on cumulative catch prior to the sample is fit. The catchability coefficient (q) is estimated from the negative of the slope and the initial population size (No) is estimated by dividing the intercept by the catchability coefficient. If Ricker.mod=TRUE then the cumulative catch is modified to be the cumulative catch prior to the sample plus half of the catch of the current sample.

For the DeLury method, a linear regression model of log (catch-per-unit-effort) on cumulative effort is fit. The catchability coefficient (q) is estimated from the negative of the slope and the initial population size (No) is estimated by dividing the intercept as an exponent of e by the catchability coefficient. If Ricker.mod=TRUE then the cumulative effort is modified to be the cumulative effort prior to the sample plus half of the effort of the current sample.

Standard errors for the catchability and population size estimates are computed from formulas on page 298 (for Leslie) and 303 (for DeLury) from Seber (2002). Confidence intervals are computed using standard large-sample normal distribution theory with the regression error df.

Value

A list with the following items:

• method A string that indicates whether the "Leslie" or "DeLury" model was used.

• catch The original vector of catches.

• effort The original vector of efforts.

• cpe A computed vector of catch-per-unit-effort for each time.

• KorE A computed vector of cumulative catch (K; Leslie method) or effort (E; DeLury method).

• lm The lm object from the fit of CPE on K (Leslie method) or log(CPE) on E (DeLury method).

• est A 2x2 matrix that contains the estimates and standard errors for No and q.

testing

The Leslie method without the Ricker modification and the DeLury method with the Ricker modification matches the results from deplet in fishmethods for the darter (from fishmethods), LobsterPEI and BlueCrab from FSAdata, and SMBassLS for N0 to whole numbers, the SE for No to one decimal, q to seven decimals, and the SE of q to at least five decimals.

The Leslie method matches the results of Seber (2002) for N0, q, and the CI for Q but not the CI for N (which was so far off that it might be that Seber's result is incorrect) for the lobster data and the q and CI for q but the NO or its CI (likely due to lots of rounding in Seber 2002) for the Blue Crab data.

The Leslie and DeLury methods match the results of Ricker (1975) for No and Q but not for the CI of No (Ricker used a very different method to compute CIs).

IFAR Chapter

10-Abundance from Depletion Data.

Author(s)

Derek H. Ogle, [email protected]

References

Ogle, D.H. 2016. Introductory Fisheries Analyses with R. Chapman & Hall/CRC, Boca Raton, FL.

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. [Was (is?) from http://www.dfo-mpo.gc.ca/Library/1485.pdf.]

Seber, G.A.F. 2002. The Estimation of Animal Abundance. Edward Arnold, Second edition (reprinted).

See Also

See removal for related functionality and deplet in fishmethods for similar functionality.

Examples

## Leslie model examples
# no Ricker modification
l1 <- depletion(SMBassLS$catch,SMBassLS$effort,method="Leslie")
summary(l1)
summary(l1,verbose=TRUE)
summary(l1,parm="No")
rSquared(l1)
rSquared(l1,digits=1,percent=TRUE)
cbind(Est=coef(l1),confint(l1))
cbind(Est=coef(l1,parm="No"),confint(l1,parm="No"))
cbind(Est=coef(l1,parm="q"),confint(l1,parm="q"))
summary(l1,parm="lm")
plot(l1)

# with Ricker modification
l2 <- depletion(SMBassLS$catch,SMBassLS$effort,method="Leslie",Ricker.mod=TRUE)
summary(l2)
cbind(Est=coef(l2),confint(l1))
plot(l2)

## DeLury model examples
# no Ricker modification
d1 <- depletion(SMBassLS$catch,SMBassLS$effort,method="DeLury")
summary(d1)
summary(d1,parm="q")
summary(d1,verbose=TRUE)
rSquared(d1)
cbind(Est=coef(d1),confint(d1))
summary(d1,parm="lm")
plot(d1)

# with Ricker modification
d2 <- depletion(SMBassLS$catch,SMBassLS$effort,method="DeLury",Ricker.mod=TRUE)
summary(d2)
cbind(Est=coef(d2),confint(d2))
cbind(Est=coef(d2,parm="q"),confint(d2,parm="q"))
plot(d2)