Skip to content
/ tcplotter Public

Create plots of predicted thermochronometer ages and closure temperatures for various cooling histories and mineral parameters

License

Notifications You must be signed in to change notification settings

HUGG/tcplotter

Repository files navigation

Tcplotter code description and instructions for use

Binder Documentation Status Anaconda-Server Badge License: MIT DOI

Tcplotter is a Python package for creating and customizing thermochronometer age and closure temperature plots presented in the article Short communication: Modelling competing effects of cooling rate, grain size and radiation damage in low temperature thermochronometers by D. Whipp, D. Kellett, I. Coutand, and R. Ketcham. The code is designed to be easy to use to either reproduce plots from the article or customize the plots for your own use. Below you will find some essential details about using the code and detailed documentation can be found on the tcplotter documentation site.

Getting started

The easiest way to get started using Tcplotter is by using Binder either with the link here or by clicking on the Binder button above. This will open an interactive web interface to the code where you can see how it works and even create your own plots.

Installation

Currently, we recommend using the code via Binder as the easiest option, as we do not yet have a Python package available for Tcplotter. If you would like to install the software for your own use, you can find detailed instructions on the Tcplotter documentation page.

Usage

Tcplotter can be used either as a function in a Python script or interpreter, or from the command line. The four main Tcplotter functions/command-line tools are:

  • time_vs_temp
  • eu_vs_radius
  • rate_vs_radius_eu
  • rate_vs_age_tc

Brief examples of possible usage for both cases can be found below.

Usage in a Python script or interpreter

Assuming the base tcplotter directory is your working directory, functions available in Tcplotter can be imported as follows:

from tcplotter import time_vs_temp, eu_vs_radius, rate_vs_radius_eu, rate_vs_age_tc

Once imported, you can use functions as shown below:

eu_versus_radius(save_plot=True)

You can find more information about the function parameters using the help() function:

help(eu_vs_radius)

Command-line usage

Command-line usage is similar to that for use in a Python script, except that the underscores in the function names have been replaced by hyphens. For example, assuming you are in a terminal in the tcplotter/tcplotter directory, you can type the following to use the eu_vs_radius() function:

./eu-vs-radius --save-plot

To find more information about options available for command-line use you can include the --help or -h flags.

./eu-vs-radius -h

Attribution

How to cite the code

When including plots generated using Tcplotter in publications or presentations, please cite the following:

  • Whipp, D. M. and Ketcham, R. A.: tcplotter: a Python package for creating and customizing thermochronometer age and closure temperature plots, Zenodo [code], https://doi.org/10.5281/zenodo.5958939, 2022.

How to cite the paper

You're also welcome to cite the manuscript related to Tcplotter when referencing ideas from the article:

  • Whipp, D. M., Kellett, D. A., Coutand, I., and Ketcham, R. A.: Short communication: Modeling competing effects of cooling rate, grain size, and radiation damage in low-temperature thermochronometers, Geochronology, 4, 143–152, https://doi.org/10.5194/gchron-4-143-2022, 2022.

How to cite related articles

The age prediction software used for calculating apatite and zircon (U-Th)/He and apatite fission-track ages was written by Richard Ketcham at the University of Texas, USA. Results published using this software should also cite the articles below:

  • Ketcham, R. A., Donelick, R. A., & Carlson, W. D.: Variability of apatite fission-track annealing kinetics III: Extrapolation to geological time scales. American Mineralogist, 84, 1235-1255, doi: 10.2138/am-1999-0903, 1999.

  • Ketcham, R. A., Mora, A., and Parra, M.: Deciphering exhumation and burial history with multi-sample down-well thermochronometric inverse modelling, Basin Res., 30, 48-64, 10.1111/bre.12207, 2018.

License

The Tcplotter software is licensed under an MIT License: Tcplotter software license

About

Create plots of predicted thermochronometer ages and closure temperatures for various cooling histories and mineral parameters

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published