In andrew-edwards/sizeSpectra: Fitting Size Spectra to Ecological Data Using Maximum Likelihood

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.width = 5.7,
  fig.height = 7
)

Recommended plotting approach from the MLEbins method

This vignette has some of the calculations used to explain the recommended plotting approach, reproduces Figures S.4 and Figure 7 and presents Figures S.5 to S.34 as a movie.

library(sizeSpectra)
# library(tibble)  # Else prints all of a tibble

Schematic diagram for two species

Picking two example species (coloured in red in Figure 6). Codes are 127205 (Moustache Sculpin) and 154675 (Snakeblenny).

sculpin = 127205
snakeblenny = 154675
specForFig = c(snakeblenny, sculpin)

dataTwoSpec = dplyr::filter(dataBin, SpecCode %in% specForFig)

Combine the data across all years for illustrative purposes (to reduce the number of empty bins in the illustration):

dataComb = dplyr::summarise(dplyr::group_by(dataTwoSpec,
                                            SpecCode,
                                            wmin),
                            wmax = unique(wmax),
                            number = round(sum(Number), dig=4))

dataComb = dplyr::mutate(dataComb, j = dplyr::min_rank(wmin))   # works since
# they're grouped

# dataComb[nrow(dataComb), ]$j <- 25                       # manually, because there's a gap
# TODO THIS NEEDS UNDOING
nrow(dataComb)

dataComb = dplyr::mutate(dataComb, wmid = (wmin + wmax)/2)  # midpoint for plotting

Now create Figure S.4, with semi-automated text for the caption:

target = 7                 # target bin (for species 2)

target.wmin = dplyr::filter(dataComb,
                            SpecCode == specForFig[2],
                            j == target)$wmin
target.wmax = dplyr::filter(dataComb,
                            SpecCode == specForFig[2],
                            j == target)$wmax
# Smallest and then largest possible inclusions of being > target bin:
dataComb = dplyr::mutate(dataComb,
                         wmin.gt.tmax = (wmin >= target.wmax))
dataComb = dplyr::mutate(dataComb,
                         wmax.gt.tmin = (wmax > target.wmin))

xLim = c(15, 43)           # range to show, must be integers
dataComb = dplyr::filter(dataComb,
                         wmin > xLim[1])  # only show complete bins

numSpec = length(specForFig)
yLimMax = 1
yVals = c(0.3, 0.7)        # y values for horizontal mass bins

colSpec = c("red", "pink")
thick = 30                 # y thickness of bins
cex.sub = 0.7              # font size for subscripts

```r
#postscript("binCountRangeSchematic.eps",
#           height = 4, width = 6.5,
#           horizontal=FALSE,  paper="special")
par(mgp=c(2.0, 0.5, 0))    # puts axes labels closer I think
par(lend="butt")           # To have butted line caps, need for thick lines.
par(mar = c(4, 2.5, 3, 2)) # outer margins

plot(0, 0, xlab="Body mass, g", ylab="", xlim=xLim,
     ylim=c(0, yLimMax), yaxt="n", type="n", xaxs="i") # set up axes.

axis(1, at = xLim[1]:xLim[2], labels = rep("", xLim[2]-xLim[1]+1), tck=-0.01)

mtext("Species Code", side=2, line=1.5, cex.lab=1)
axis(2, at = yVals, labels = c(1,2), las = 2,
        tck = -0.005, cex.axis=0.8)

abline(v = target.wmin, col = "grey")
abline(v = target.wmax, col = "grey")

for(ii in 1:length(specForFig))   # loop over species, plot bins for each
   {
    yVal = yVals[ii]         # where to have horizontal bars
    bins.wmin = dplyr::filter(dataComb, SpecCode == specForFig[ii])$wmin  # for this sp
    bins.wmax = dplyr::filter(dataComb, SpecCode == specForFig[ii])$wmax
    bins.wmid = dplyr::filter(dataComb, SpecCode == specForFig[ii])$wmid
    # bins for which wmin >= target.wmax
    bins.wmin.gt.tmax = dplyr::filter(dataComb, SpecCode == specForFig[ii])$wmin.gt.tmax
    bins.wmin.gt.tmax.yn = ifelse(bins.wmin.gt.tmax,"Y", "N")
    # bins for which wmax > target.wmin
    bins.wmax.gt.tmin = dplyr::filter(dataComb, SpecCode == specForFig[ii])$wmax.gt.tmin
    bins.wmax.gt.tmin.yn = ifelse(bins.wmax.gt.tmin,"Y", "N")

    ## TODO these are all commented out to get vignette to build on install
    ##  properly, yet it still builds locally fine (and used to)
    ## segments(y0 = yVal,
    ##          x0 = bins.wmin,
    ##          x1 = bins.wmax,
    ##          col=colSpec, lwd=thick)    # recycles col
    # Bin break labels (cannot do , with vector for labels, it seems):

    ## for(iiii in 1:length(bins.wmin))
    ##   {
    ##     text(x = bins.wmin[iiii],
    ##          y = yVal,
    ##          labels = "XXX",
    ##          # TODO commenting out this as it stopped working when building with
    ##          #  R version 4.3.0 (something between 4.2.1 and 4.3.0). Though
    ##          #  build_vignettes() works locally.
    ##          # bquote(paste(w[.(ii)*","*.(dplyr::filter(dataComb, SpecCode == specForFig[ii])$j[iiii])])),
    ##          pos = 1,
    ##          offset = 0.03*thick,
    ##          cex = cex.sub)
    ##   }

##     # Do final wmax manually:
##     text(x = bins.wmax[length(bins.wmax)],
##          y = yVal,
##          labels = bquote(paste(w[.(ii)*","*
##                      .(max(dplyr::filter(dataComb, SpecCode == specForFig[ii])$j)+1)])),
##          pos = 1,
##          offset = 0.03*thick,
##          cex = cex.sub)
##     # Put counts for each bin
##     text(x = bins.wmid,
##          y = yVal,
##          labels = f(dplyr::filter(dataComb, SpecCode == specForFig[ii])$number, ii+1),
##                   # ii+1 happens to give 3 d.p.s for species 2 and 2 for species 1,
##                   #  as needed on figure
##          pos = NULL,
##          offset = 0.03*thick,
##          cex = cex.sub*1.2)
##     # Whether to include in counts
##     eps = 0.23                  # Offset from wmid for Y's and N's
##     text(x = bins.wmid - eps,
##          y = yVal,
##          labels = bins.wmin.gt.tmax.yn,
##          pos = 3,
##          offset = 0.03*thick,
##          cex = cex.sub*1.2)
##     text(x = bins.wmid + eps,
##          y = yVal,
##          labels = bins.wmax.gt.tmin.yn,
##          pos = 3,
##          offset = 0.03*thick,
##          cex = cex.sub*1.2)
}

## # Need w_1,24 manually, assumes this is the only one with j=23
## text(x = dplyr::filter(dataComb, j==23)$wmax,
##      y = yVals[1],
##      labels = bquote(paste(w[.(1)*","*24])),
##      pos = 1,
##      offset = 0.03*thick,
##      cex = cex.sub)

## # Calculations for the caption:
lowCount = sum(dataComb$wmin.gt.tmax * dataComb$number)
highCount = sum(dataComb$wmax.gt.tmin * dataComb$number)

Figure S.4 caption: Schematic diagram to explain how we calculate the range of counts of individuals that are larger than those in a given bin. Red and pink body-mass bins are those for Snakeblenny and Moustache Sculpin (labelled species 1 and 2, respectively) from Figure 5 in the paper. Bin breaks are denoted by $w_{s,j}$ and the number inside each bin is the number observed per hour of trawling. For illustration the data are combined across all years and only bins with minima $>15$ g are shown. The target bin has $s=2$ and $j=7$ and therefore has bin breaks $w_{2,7}$ and $w_{2,8}$ and is indicated by the vertical grey lines. The first letter in each pair (NN, NY or YY) denotes whether or not each bin is included in the low count $E_{2,7,1}$, i.e. its lower bound is $\geq$ the upper bound of the target bin. Similarly, the second letter denotes whether or not each bin is included in the high count $E_{2,7,2}$, i.e. its upper bound is $>$ the lower bound of the target bin. Summing the respective counts as per (S.30) and (S.31) gives $E_{2,7,1} =$ r f(lowCount,2) and $E_{2,7,2} =$ r f(highCount,2).

Results plotted for each year

These calculations are to get the required input for the recommended plot (see ?sizeSpectra::ISD_bin_plot for the structure). This could maybe be functionalised like the plotting function ISD_bin_plot().

dataRecommend.isd = dplyr::select(dataBin,
                                  Year,
                                  wmin,
                                  wmax,
                                  Number)


data.year.list = list()                # to save results for each year

fullYears = sort(unique(dataBin$Year))
for(i in 1:length(fullYears))
  {
    data.year = dplyr::filter(dataRecommend.isd,
                              Year == fullYears[i])

    data.year = dplyr::arrange(data.year,
                               desc(wmin))
    sumNumber = sum(data.year$Number)

    # data.year = dplyr::mutate(data.year,
    #                          cumSum = cumsum(Number))
    # This is wrong when we have two species with the same
    #  length-weight coefficients in the same year, so need countGTEwmin below
    # Have to do not with dplyr:
    wmin.vec = data.year$wmin
    wmax.vec = data.year$wmax
    num.vec  = data.year$Number

    countGTEwmin = rep(NA, length(num.vec)) # to do a manual count
    lowCount = countGTEwmin
    highCount = countGTEwmin

    for(iii in 1:length(countGTEwmin))
    {
      countGTEwmin[iii]    = sum( (wmin.vec >= wmin.vec[iii]) * num.vec)
      lowCount[iii]  = sum( (wmin.vec >= wmax.vec[iii]) * num.vec)
      highCount[iii] = sum( (wmax.vec >  wmin.vec[iii]) * num.vec)
    }
    data.year = cbind(data.year,
                      "countGTEwmin" = countGTEwmin,
                      "lowCount" = lowCount,
                      "highCount" = highCount)
    data.year = dplyr::tbl_df(data.year) # This is one of the desired input for
                                         #  the plotting function below

    data.year.list[[i]] = data.year
}

xlim.global = c(min(dataRecommend.isd$wmin),
                max(dataRecommend.isd$wmax))   # x-axis limits to be common for
                                               # all plots

Here is the code to give an animation for Figures 7 and S.5-S.34 for the IBTS data. For each year (given in top-right corner) the plot shows the individual size distribution and MLEbins fit (red solid curve) with 95\% confidence intervals (red dashed curves, may be hard to see). For each bin, the horizontal green line shows the range of body sizes, with value on the y-axis corresponding to the total number of individuals in bins whose minima are $\geq$ the bin's minimum. For each bin, the vertical span of the grey rectangle shows the possible range of the number of individuals with body mass $\geq$ the body mass of individuals in that bin (horizontal span is the same as for the green lines). The text in (a) gives the year, the MLE for the size-spectrum exponent $b$, and the sample size $n$.

Here is the code to build the movie, but it is commented out since it causes Travis to fail -- Travis is the continuous integration service that automatically checks the packages builds every time a change is committed to GitHub, giving the little green symbol passing icon on the main page. The trick to showing the movie is to run the code uncommented (you need to install the gifski package), right-click on the animation in the html viewer and save it. I've done that and so am leaving this code commented, and will just refer to the saved animation file.

# ```r
# fig.width is 0.67 * fig.height (which is 8)
#
# for(i in 1:length(fullYears))  # Not tested new options here (from _all vignette)
#    {
#    ISD_bin_plot(data.year = data.year.list[[i]],
#                 b.MLE = dplyr::filter(MLEbins.res, Year == fullYears[i])$b,
#                 b.confMin = dplyr::filter(MLEbins.res, Year ==
#                                                      fullYears[i])$confMin,
#                 b.confMax = dplyr::filter(MLEbins.res, Year ==
#                                                      fullYears[i])$confMax,
#                 year = fullYears[i],
#                 xlim = xlim.global,
#                 xmin = dplyr::filter(MLEbins.res, Year ==
#                                                      fullYears[i])$xmin,
#                 xmax = dplyr::filter(MLEbins.res, Year ==
#                                                   fullYears[i])$xmax,
#                 IBTS_MEPS_figs = TRUE
#                 )
#  }
# ```

The resulting animation is

.

To save each year as it's own figure do this (not it running here; you have to make the IBTS-ISD-figs directory first):

for(i in 1:length(fullYears))
  {
  postscript(paste0("../IBTS-ISD-figs/IBTS-ISD", fullYears[i], ".eps"),
           height = 8, width = 5.36,
           horizontal=FALSE, paper="special")

  ISD_bin_plot(data.year = data.year.list[[i]],
               b.MLE = dplyr::filter(MLEbins.res, Year == fullYears[i])$b,
               b.confMin = dplyr::filter(MLEbins.res, Year ==
                                                     fullYears[i])$confMin,
               b.confMax = dplyr::filter(MLEbins.res, Year ==
                                                     fullYears[i])$confMax,
               year = fullYears[i],
               xlim = xlim.global,
               xmin = dplyr::filter(MLEbins.res, Year ==
                                                    fullYears[i])$xmin,
               xmax = dplyr::filter(MLEbins.res, Year ==
                                                 fullYears[i])$xmax,
               IBTS_MEPS_figs = TRUE
               )
  dev.off()
}

Do this to include (in a LaTeX file) a loop that plots each figure one at a time, like in our Supporting Information.

% Include this in latex preamble:
\newcommand\isdfig[2]{    % filename is #1, text is #2
  \begin{figure}[tp]
  \begin{center}
  \epsfbox{IBTS-ISD-figs/IBTS-ISD#1.eps}
  \end{center}
  \vspace{-5mm}
  \caption{#2}
  \label{fig:ISD-#1}
  \end{figure}
  \clearpage
}

% Include this after the figures have already been created in an R loop
\isdfig{\Sexpr{fullYears[1]}}{Include caption as above for year \Sexpr{fullYears[i]}}

\newcounter{loop}

\newcommand{\loopMax}{\Sexpr{max(fullYears)+1}}
\forloop{loop}{\Sexpr{fullYears[2]}}{\value{loop} < \loopMax}
        {\isdfig{\arabic{loop}}{Individual size distribution and MLEbins fit
            with 95\% confidence intervals for IBTS data in \arabic{loop}.
            Details as in Figure~\ref{fig:ISD-1986}.
}}

andrew-edwards/sizeSpectra documentation built on June 28, 2023, 7:09 p.m.