plot_data creates plot(s) of the biotracer data and saves the plot(s)
to file(s) in the working directory. All 3 required data files must have been
loaded by load_mix_data, load_source_data,
and load_discr_data. Behavior depends on the number of tracers:
1 tracer: calls plot_data_one_iso to create a 1-D plot.
2 tracers: calls plot_data_two_iso to create a biplot.
>2 tracers: calls plot_data_two_iso in a loop to create
biplots for each pairwise combination of biotracers.
plot_data( filename, plot_save_pdf, plot_save_png, mix, source, discr, return_obj = FALSE )
| filename | name of the plot file(s) to save (e.g. "isospace_plot") |
|---|---|
| plot_save_pdf | T/F, save the plot(s) as a pdf? |
| plot_save_png | T/F, save the plot(s) as a png? |
| mix | output from |
| source | output from |
| discr | output from |
| return_obj | T/F, whether or not to return ggplot object for further modification, defaults to F |
An important detail is that plot_data_two_iso and plot_data_one_iso
plot the raw mix data and add the TDF to the source data, since this is
the polygon that the mixing model uses to determine proportions. The plotted
source means are:
$$\mu_source + \mu_discr$$
The source error bars are +/- 1 standard deviation, calculated as a
combination of source and TDF variances:
$$\sqrt{\sigma^2_source + \sigma^2_discr}$$
plot_data looks for 'C', 'N', 'S', and 'O' in the biotracer column
headers and assumes they are stable isotopes, labeling the axes with, e.g.,
expression(paste(delta^13, "C (u2030)",sep="")).