Written by Ilja J. Kocken Feel free to use this stuff, as long as you adhere to the file:LICENSE.md
NOTE: This is just the method I use, I'm not publishing the actual calibration dataset that I'm using here!
- Calculate bootstrapped York regression from clumped isotope calibration data. clumped_calib_boot().
- Calculate bootstrapped sample means for the age, d13C, d18O, and D47. bootstrap_means().
- Apply the bootstrapped calibration to calculate temperature and d18Osw. temp_d18Osw_calc().
- Summarize bootstrapped values and calculate 68 and 95% CIs. our_summary().
- Also comes with the classic d18Occ calibrations. See temp_calc().
- As well as the inverses, for when you have constraints on temperature and wish to calculate d18Osw. See d18Osw_calc().
How to calculate bootstrapped averages and apply a clumped-isotope specific temperature calibration.
To use this:
- Install this package
clumpedcalib
, see #installation. - Update the input calibration data
file:dat/example_calib.csv. Make sure the
column names include
X
,D47
,sd_X
,sd_D47
. - Add your sample data at the replicate level.
- Check to see if you like the function that calculates d18Osw. Currently, it defaults to the Kim & O'Neil 1997 calibration, as adjusted by Bemis et al., 1998.
- If you want to use the wrapper function: create your own definition of filter outliers so that it filters out bad measurements and NA's for your data.
- Use the wrapper function to do it all at once apply_calib_and_d18O_boot(), or better yet:
- Walk through the example workflow step-by-step or apply the bulletpoint functions from above one-by-one.
- Have a look at the functions to better understand what's happening.
The bootstrapped calibration idea is based on a Matlab script written by Alvaro Fernandez.
To install this development package:
remotes::install_github("japhir/clumpedcalib")
Or you can look at all the functions and only copy what you like.