HistMapR

Rapid digitization of historical land-use maps in R

Background

We created this package due to our need to classify large areas of historical land-use map quickly and easily. The package contains four functions for the classification of historical land-use maps in R, calling functions from the terra package by R. Hijmans. The devtools package was invaluable in turning our scripts into this package. A full description of the method is available <a href="https://doi.org/10.1111/2041-210X.12788" target="_blank">here.</a>

Installation

This package can be installed using the devtools package. HistMapR also requires terra, ggplot2 and gridExtra to run the functions. Even though installing through devtools installs dependent packages, there are issues in installing the packages that these are in-turn dependent upon, which can cause problems. This code adapted from <a href="https://github.com/BiologicalRecordsCentre/sparta" target="_blank">here</a>.

req.packages<-c("devtools", "terra", "ggplot2","gridExtra") # required packages
new.packages <- req.packages[!(req.packages %in% installed.packages()[,"Package"])] # which are not installed?
if(length(new.packages)) install.packages(new.packages, dep=TRUE) # install those as required.

library(devtools)
install_github("AGAuffret/HistMapR")
library(HistMapR)

Versions

Version 0.1 was released in 2017, calling functions from the raster package, and was updated now and then as issues were identified. In 2025, to keep up with the times, the package was re-worked to call from the terra package instead. This did not result in any major improvement in performance (in fact, smooth_map(), which calls the focal() function actually seems to be marginally slower), but the functionality is the same, and hopefully it will be more useful for people using terra in their R GIS workflows. If I have understood things correctly, the above code should install the most recent version of the package. If you want the older, raster-dependent version, you should write install_github("AGAuffret/HistMapR@v0.1").

Examples and tutorials

Very simple examples of each function can be seen using the examples function. The input maps in these examples are from the Swedish Agency Lantmäteriet (copyright expired). A more detailed tutorial with walkthrough scripts and examples can be found on Figshare (though not updated with package updates, so it's always good to check the help files), along with the historical land-use classification over southern Sweden for which this package was created. <a href="https://doi.org/10.17045/sthlmuni.4649854" target="_blank">Link.</a>. There is also a <a href="https://www.youtube.com/watch?v=5iD1pyDBBks" target="_blank">very basic 10 minute tutorial video </a> that we made to go with the published paper, where we walk through the process of digitizing maps with HistMapR (based on v0.1).

Contributors and maintenance

This package was created by <a href="mailto:alistair.auffret@slu.se">Alistair Auffret</a> and Adam Kimberley. It's our first package and first time on GitHub, so help and suggestions are welcome.

References

Auffret, A.G., Kimberley, A., Plue, J., Skånes, H., Jakobsson, S., Waldén, E., Wennbom, M., Wood, H., Bullock, J.M., Cousins, S.A.O., Hooftman, D.A.P., Gartz, M. & Tränk, L. (2017) HistMapR: Rapid digitization of historical land-use maps in R. Methods in Ecology and Evolution, doi: 10.1111/2041-210X.12788.

Auffret, A.G., Kimberley, A., Plue, J., Skånes, H., Jakobsson, S., Waldén, E., Wennbom, M., Wood, H., Bullock, J.M., Cousins, S.A.O., Hooftman, D.A.P., Gartz, M. & Tränk, L. (2017) Data from: HistMapR: Rapid digitization of historical land-use maps in R. figshare data repository, doi: 10.17045/sthlmuni.4649854.

Hijmans, R.J. (2016) raster: Geographic Data Analysis and Modeling. R package version 2.5-8, url: http://CRAN.R-project.org/package=raster.

Hijmans, R.J. (2023) terra: Spatial Data Analysis. R package version 1.7-29, url: http://CRAN.R-project.org/package=terra.

Licence

Released under the <a href="https://choosealicense.com/licenses/gpl-3.0/" target="_blank"> GNU GPLv3 license </a>.