Introduction to rpaleoclim

PaleoClim1 is a set of free, high resolution paleoclimate surfaces covering the whole globe. It includes data on surface temperature, precipitation and the standard bioclimatic variables commonly used in ecological modelling, derived from the HadCM3 general circulation model and downscaled to a spatial resolution of up to 2.5 minutes. Simulations are available for key time periods from the Late Holocene to mid-Pliocene. Data on current and Last Glacial Maximum climate is derived from CHELSA2 and reprocessed by PaleoClim to match their format; it is available at up to 30 seconds resolution.

This package provides a simple interface for downloading PaleoClim data in R, with support for caching and filtering retrieved data by period, resolution, and geographic extent.

Installation

You can install the development version of rpaleoclim from GitHub using the remotes package:

remotes::install_github("joeroe/rpaleoclim")

It depends on terra, which in turn requires a recent version of the the system libraries GDAL and PROJ. These are included in the binary releases of terra for Windows and MacOS, but if you build from source (e.g. on Linux) you might need to install them first; see the terra README for instructions for different systems.

Data available

rpaleoclim provides an R interface to download all the data listed at http://www.paleoclim.org/ and currently mirrored at http://www.sdmtoolbox.org. The tables below were last updated 2022-02-16; please refer to the PaleoClim website for the authoritative version.

If you notice a change or update to the PaleoClim data structure that isn’t supported by the current version of the package, please report it at: https://github.com/joeroe/rpaleoclim/issues.

Periods

PaleoClim includes climate reconstructions from simulations of the following time intervals, supplemented with two additional datasets from CHELSA:

code period bp source
cur Current (1979 – 2013) CHELSA
lh Late Holocene: Meghalayan 4.2-0.3 ka Fordham  et al. 2017
mh Mid Holocene: Northgrippian 8.326-4.2 ka Fordham  et al. 2017
eh Early Holocene: Greenlandian 11.7-8.326 ka Fordham  et al. 2017
yds Pleistocene: Younger Dryas Stadial 12.9-11.7 ka Fordham  et al. 2017
ba Pleistocene: Bølling-Allerød 14.7-12.9 ka Fordham  et al. 2017
hs1 Pleistocene: Heinrich Stadial 1 17.0-14.7 ka Fordham  et al. 2017
lgm Pleistocene: Last Glacial Maximum ca. 21 ka CHELSA
lig Pleistocene: Last Interglacial ca. 130 ka Otto-Bliesner et al. 2006
mis19 Pleistocene: MIS19 ca. 787 ka Brown et al. 2018
mpwp Pliocene: Mid-Pliocene warm period 3.205 Ma Hill 2015
m2 Pliocene: M2 ca. 3.3 Ma Dolan et al. 2015

Bioclimatic variables

PaleoClim uses the 16 standard “bioclimatic variables” with their conventional coding:

biovar definition formula
bio_1 Mean annual temperature
bio_2 Mean diurnal range
bio_3 Isothermality (bio2 / bio7) * 100
bio_4 Temperature seasonality sd(temp) * 100
bio_5 Maximum temperature of warmest month
bio_6 Minimum temperature of coldest month
bio_7 Annual temperature range bio5 - bio6
bio_8 Mean temperature of wettest quarter
bio_9 Mean temperature of driest quarter
bio_10 Mean temperature of warmest quarter
bio_11 Mean temperature of coldest quarter
bio_12 Total annual precipitation
bio_13 Precipitation of wettest month
bio_14 Precipitation of driest month
bio_15 Precipitation seasonality
bio_16 Precipitation of wettest quarter
bio_17 Precipitation of driest quarter
bio_18 Precipitation of warmest quarter
bio_19 Precipitation of coldest quarter

Other options

The options for resolution are:

Retrieving data

library(rpaleoclim)
library(terra)
#> terra 1.7.39

Use paleoclim() to download paleoclimate data from PaleoClim, specifying the desired time period (see above) and resolution. For example, to download data for the Late Holocene ("lh") at 10 min resolution:

paleoclim("lh", "10m")
#> class       : SpatRaster 
#> dimensions  : 1044, 2160, 19  (nrow, ncol, nlyr)
#> resolution  : 0.1666667, 0.1666667  (x, y)
#> extent      : -180.0001, 179.9999, -90.00014, 83.99986  (xmin, xmax, ymin, ymax)
#> coord. ref. : lon/lat WGS 84 (EPSG:4326) 
#> sources     : bio_1.tif  
#>               bio_10.tif  
#>               bio_11.tif  
#>               ... and 16 more source(s)
#> names       : bio_1, bio_10, bio_11, bio_12, bio_13, bio_14, ... 
#> min values  :  -526,   -334,   -656,      0,      0,      0, ... 
#> max values  :   314,    385,    288,   9696,   2399,    651, ...

The result is a SpatRaster object with up to 19 layers containing reconstructed values of the standard “bioclimatic variables” (see above) for this period. Consult the terra documentation for information on working with spatial raster data in R.

By default, paleoclim() loads the entire downloaded raster into R, i.e. the whole globe. You can conveniently crop it to a specific (rectangular) region of interest with the region argument. This can be specified with a SpatExtent object or anything coercible to one (see ?terra::ext), which includes most spatial data types, or simply a vector of coordinates (xmin, xmax, ymin, ymax):

europe <- c(-15, 45, 30, 90)
europe_lh <- paleoclim("lh", "10m", region = europe)

plot(europe_lh[["bio_12"]], main = "Late Holocene annual precipitation, Europe")

If you have already downloaded data from PaleoClim and simply want to read it into R, you can do so with load_paleoclim(), passing it the path to the .zip archive:

zipfile <- system.file("testdata", "LH_v1_10m_cropped.zip",
                       package = "rpaleoclim")
load_paleoclim(zipfile)
#> class       : SpatRaster 
#> dimensions  : 6, 6, 19  (nrow, ncol, nlyr)
#> resolution  : 0.1666667, 0.1666667  (x, y)
#> extent      : 0, 1, 0, 1  (xmin, xmax, ymin, ymax)
#> coord. ref. : lon/lat WGS 84 (EPSG:4326) 
#> sources     : bio_1.tif  
#>               bio_10.tif  
#>               bio_11.tif  
#>               ... and 16 more source(s)
#> names       : bio_1, bio_10, bio_11, bio_12, bio_13, bio_14, ... 
#> min values  :  -526,   -334,   -656,      0,      0,      0, ... 
#> max values  :   314,    385,    288,   9696,   2399,    651, ...

Caching

Note that in the examples above, we only downloaded data from the PaleoClim servers once. Repeated calls to paleoclim() that ask for the same time period and resolution reuse cached versions of the previously downloaded file. By default, these files are stored in R’s temporary directory, so that you only download the files once per session. The cached files are never modified; subsequent cropping, warping, etc. is either done in-memory or creates new temporary files.

The cache_path argument controls the directory that paleoclim() downloads files to and tries to read cached data from. It can be useful to change this to somewhere within the working directory to reuse the same files between sessions and ensure your analysis can be reproduced in the future, even if the remote PaleoClim data changes or disappears.

skip_cache = TRUE will force paleoclim() to download files from PaleoClim even if cached data exists in cache_path (in which case it will overwrite it).

quiet = TRUE suppresses messages about whether the data is being downloaded or read from a cached file.

Backwards compatibility with raster

Since version 0.9.1, rpaleoclim has depended on the terra package for reading and manipulating spatial data. Previous versions used raster and rgdal, which are now deprecated. To use the old raster types, you will need to install the optional dependency raster:

install.packages(c("raster"))

Then pass as = "raster" to paleoclim() or load_paleoclim() to return the data as a RasterStack object instead of a SpatRaster.

paleoclim("lh", "10m", as = "raster")
#> The legacy packages maptools, rgdal, and rgeos, underpinning the sp package,
#> which was just loaded, will retire in October 2023.
#> Please refer to R-spatial evolution reports for details, especially
#> https://r-spatial.org/r/2023/05/15/evolution4.html.
#> It may be desirable to make the sf package available;
#> package maintainers should consider adding sf to Suggests:.
#> The sp package is now running under evolution status 2
#>      (status 2 uses the sf package in place of rgdal)
#> Warning: `as = "raster"` is deprecated and will be removed in future versions of rpaleoclim
#> This warning is displayed once per session.
#> class      : RasterStack 
#> dimensions : 1044, 2160, 2255040, 19  (nrow, ncol, ncell, nlayers)
#> resolution : 0.1666667, 0.1666667  (x, y)
#> extent     : -180.0001, 179.9999, -90.00014, 83.99986  (xmin, xmax, ymin, ymax)
#> crs        : +proj=longlat +datum=WGS84 +no_defs 
#> names      : bio_1, bio_10, bio_11, bio_12, bio_13, bio_14, bio_15, bio_16, bio_17, bio_18, bio_19, bio_2, bio_3, bio_4, bio_5, ... 
#> min values :  -526,   -334,   -656,      0,      0,      0,      0,      0,      0,      0,      0,     6,     3,   116,  -292, ... 
#> max values :   314,    385,    288,   9696,   2399,    651,    215,   6321,   2025,   5128,   3860,   169,    90, 24243,   481, ...

Please note that raster support will be removed in a future version of rpaleoclim.

Citing data

Please follow the instructions from the authors when citing PaleoClim data. At time of writing, this includes a citation to the paper the describing the PaleoClim database:

As well as the original papers for the individual original datasets used.

Use citation("paleoclim") for more details and the references in BibTeX format.


  1. Brown, J., Hill, D., Dolan, A. et al. PaleoClim, high spatial resolution paleoclimate surfaces for global land areas. Sci Data 5, 180254 (2018). https://doi.org/10.1038/sdata.2018.254↩︎

  2. Karger, D., Conrad, O., Böhner, J. et al. Climatologies at high resolution for the earth’s land surface areas. Sci Data 4, 170122 (2017). https://doi.org/10.1038/sdata.2017.122↩︎