pastclim

This R library is designed to provide an easy way to extract and manipulate palaeoclimate reconstructions for ecological and anthropological analyses.

A paper describing the functionality of pastclim can be found on bioRxiv. Please cite it if you use pastclim in your research.

Install the library

You will need to install the library from GitHub. For this step, you will need to use devtools (if you haven’t done so already, install it from CRAN with install.packages("devtools"). Once you have devtools, simply use:

devtools::install_github("EvolEcolGroup/pastclim")

Overview of functionality

On its dedicated website, you can find Articles giving you a step-by-step overview of the package, and a cheatsheet.

Pastclim currently includes data from Beyer et al 2020, a reconstruction of climate based on the HadCM3 model for the last 120k years, and Krapp et al 2021, which covers the last 800k years. The reconstructions are bias-corrected and downscaled to 0.5 degree. More details on these datasets can be found here. There are also instructions on how to build and use custom datasets.

You can also build the vignettes when installing pastclim (note that you will need to have the necessary tools to build vignettes already installed; requirements depend on your OS):

devtools::install_github("EvolEcolGroup/pastclim", build_vignette = TRUE)

If you built the vignettes, you can read them directly in R. For example, the overview can be obtained with:

vignette("pastclim_overview", package = "pastclim")

Current issues

If something does not work, check the issues on GitHub to see whether the problem has already been reported. If not, feel free to create an new issue. Please make sure you provide a reproducible example for the developers to investigate the problem.

Error in x$.self$finalize()

pastclim relies on terra to process rasters. There is a known bug in terra that leads to the occasional message:

"Error in x$.self$finalize() : attempt to apply non-function"

This is an error related to garbage collection, which does not affect the script being correctly executed, so it can be ignored. More discussion of this issue can be found on stackoverflow

terra without NETCDF driver for macOS

A number of versions of terra available as binaries for macOS X86-64 on CRAN (including the latest one) have been compiled without a NETCDF driver (the latest ARM library is OK). This prevents pastclim, which relies on terra, from correctly reading files. Other packages that rely on terra are similarly affected (e.g. stars; see this bug)

When loaded, pastclim checks if the driver is available; in case of a missing driver, you will get the error:

Error: The installed version of terra lacks support for reading netcdf files.
pastclim needs netcdf support: you will need to reinstall terra,
possibly from source, if there isn't a version with netcdf support
on CRAN. Alternatively, try the latest development version from R-universe:
install.packages('terra', repos='https://rspatial.r-universe.dev')

The easiest solution is probably to install the dev version of terra from R-universe with:

install.packages('terra', repos='https://rspatial.r-universe.dev')

Alternatively, if you want to install a specific version of terra from source, see instructions here.