README.md

In jstanley26/thermochronplotr: Make Common Plots in AHe Thermochronology

thermochronplotr

The goal of thermochronplotr is to make several commonly used plots for low temperature thermochronology, particularly apatite (U-Th)/He thermochronology, from data files. Ideally this makes plotting more streamlined, while giving the user the option to customize aesthetics using the wide range of tools in the ggplot2 package.

There several functios which import the text files output by the thermal history modelling software HeFTy at the end of an inverse model run and plots them as a single panel (for one model) or a faceted grid (for multiple models)

This can also make make date-eU plots from the AHe data and plot these as a shuffled grid with HeFTy models to show data and models side by side.

Installation

This package is only available on github. To install this please install devtools package within R. Then:

devtools::install_github('jstanley26/thermochronplotr')

Example

This is a basic example which shows you how to solve a common problem:

library(thermochronplotr)
#> Loading required package: rlang
library(tidyverse)
#> ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
#> ✔ ggplot2 3.3.6     ✔ purrr   0.3.4
#> ✔ tibble  3.1.7     ✔ dplyr   1.0.9
#> ✔ tidyr   1.2.0     ✔ stringr 1.4.0
#> ✔ readr   2.1.2     ✔ forcats 0.5.1
#> ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
#> ✖ purrr::%@%()         masks rlang::%@%()
#> ✖ purrr::as_function() masks rlang::as_function()
#> ✖ dplyr::filter()      masks stats::filter()
#> ✖ purrr::flatten()     masks rlang::flatten()
#> ✖ purrr::flatten_chr() masks rlang::flatten_chr()
#> ✖ purrr::flatten_dbl() masks rlang::flatten_dbl()
#> ✖ purrr::flatten_int() masks rlang::flatten_int()
#> ✖ purrr::flatten_lgl() masks rlang::flatten_lgl()
#> ✖ purrr::flatten_raw() masks rlang::flatten_raw()
#> ✖ purrr::invoke()      masks rlang::invoke()
#> ✖ dplyr::lag()         masks stats::lag()
#> ✖ purrr::splice()      masks rlang::splice()

Single HeFTy inversion plot

For a single inversion model using the example file (replace with your own when ready)

fn=system.file("extdata","HeFTyOut-Sample1.txt",package="thermochronplotr")
samplename = 'Sample 1'
cons = readconstraints(fn,samplename)
dfplot = readpaths(fn,samplename)
plot1 = plot_hefty_output(dfplot, cons)
plot1

plot1 is a ggplot2 opject, and can be customized accordingly. For example, if you want to change the x axis limits

library(ggplot2)
plot1+coord_cartesian(xlim=c(200,0))
#> Coordinate system already present. Adding new coordinate system, which will replace the existing one.

Multiple HeFTy inversion facet

To plot the results from multple inversions, for example from more than one sample, you need to create a vector of the file names, and a coresponding vector of the sample names (in the same order)

fn=system.file("extdata","HeFTyOut-Sample1.txt",package="thermochronplotr")
samplename = 'Sample1'
fn2=system.file("extdata","HeFTyOut-Sample2.txt",package="thermochronplotr")
samplename2 = 'Sample2'

fns = c(fn, fn2)
sns = c(samplename,samplename2)

cons2 = readconstraints(fns,sns)
dfplot2 = readpaths(fns,sns)
plot2 = plot_hefty_output(dfplot2, cons2)
plot2

Again, plot2 is a ggplot object and can be customized

date-eU plots for multiple samples

To use this function you must first create or read in a dataframe with (U-Th)/He data that has columns labeled “eU”, “Date”, “Unc”,“Color”, and “Sample”, where eU is the effective Uranium, Date is the (U-Th)/He date, Unc is the uncertainty (for error bars), and Sample is the sample name (only truly required if you have more than one sample you are trying to plot). “Color” is a column by which you can color the data points, for example you might have most of the data in black, but a few data points that are outliers to be plotted in grey that info is contained in the “Color” column.

fn=system.file("extdata","AHe_data.csv",package="thermochronplotr")
hedf<-readr::read_csv(fn, show_col_types = FALSE)
plot3 = plot_date_eu(hedf)
plot3

If you have split your data into bins to model in heFTy, you can denote where these are separated with vertical dashed lines by inputting the eUbin value(s) for example if your low eU bin was \<15ppm, you would put in eUbin=15. If you had low, medium, and high eU bins split at \<15ppm, 15-30 ppm, and >30ppm, you would put in eUbin=c(15,30). Alternately if you binned or averaged your data for HeFTy modeling you can also display this synthetic or aggregate data on the plot as large open diamonds by adding another dataframe, synthetic. It must have column names “Date”, “eU”, and “Unc”, plus “Sample” if there are multiple samples.

If you had forward models that you want to plot predicted date-eU paths you can plot these by adding a second dataframe, bestfitdf, with those predictions. Again this dataframe needs to have columns “Date” and “eU” (and “Sample” if there are multiple samples)

An exaple with eU bins denoted both by dashed lines and with the synthetic grains and with the best fit forward models plotted:

fn3=system.file("extdata","SyntheticGrains.csv",package="thermochronplotr")
synthetic <- readr::read_csv(fn3)
#> Rows: 4 Columns: 5
#> ── Column specification ────────────────────────────────────────────────────────
#> Delimiter: ","
#> chr (2): Sample, Bin
#> dbl (3): eU, Date, Unc
#> 
#> ℹ Use `spec()` to retrieve the full column specification for this data.
#> ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
fn4=system.file("extdata","ForwardModels.csv",package="thermochronplotr")
bestfitdf <- readr::read_csv(fn4)
#> New names:
#> Rows: 26 Columns: 7
#> ── Column specification
#> ──────────────────────────────────────────────────────── Delimiter: "," chr
#> (4): Sample, Description, Elevation_m, Sample_cat dbl (3): ...1, eU, Date
#> ℹ Use `spec()` to retrieve the full column specification for this data. ℹ
#> Specify the column types or set `show_col_types = FALSE` to quiet this message.
#> • `` -> `...1`
plot4 = plot_date_eu(hedf,bestfitdf,synthetic,eUbin=c(15,30))
plot4

Again, plots 3 and 4 are ggplot objects that can be modified

Shuffling HeFTy and Date-eU plots

Often it is nice to look at the data and the models together, so we can create a plot that places them side by side for the same sample. The inputs are the same as for the individual HeFTy or date-eU plots, so we can use the dataframes loaded earlier in the examples. Note that the best fit forward model bestfitdf, synthetic grains synthetic and lines denoting eU bins eUbin are optional arguments.

plot5 = shuffle_plots(hedf,bestfitdf,heftydf=dfplot2,constraints=cons2,synthetic=synthetic)
plot5

Axis limits are the same across the shuffled grid and can be input manually or are calculated from the total dataset. You can also change the order in which the samples are plotted by making the “Sample” column of hedf into a factor with custom levels. For example:

hedf2 = hedf %>%  mutate(Sample = factor(Sample,levels=c('Sample2', 'Sample1')))
plot6 = shuffle_plots(hedf2,bestfitdf,dfplot2,cons2)
plot6

jstanley26/thermochronplotr documentation built on June 4, 2022, 11:05 a.m.