Using various geospatial tools with Anaconda

Install Anaconda if it is not already available on your system (I recommend the lightweight Miniconda). On Arch Linux, it can be installed from the AUR:

yay -Syu miniconda3

To set up an Anaconda virtual environment with the required tools (QGIS, Python, or R), save the snippet under the relevant section as a YAML file (e.g. environment.yml). Then, run the following commands, replacing ${envname} with the name of the environment as defined in the first line of the YAML file:

conda env create --file environment.yml
conda activate ${envname}

Note
All of these packages can be installed in a single Anaconda virtual environment via the conda-forge channel (though I haven’t checked to see if there are potential conflicts).

QGIS

name: qgis
channels:
  - conda-forge
  - defaults
dependencies:
  - python=3
  - qgis

Python

name: py-geo
channels:
  - conda-forge
  - anaconda
dependencies:
  - python=3
  - geopandas
  - rioxarray
  - rasterstats
  - scipy
  - cfgrib
  - netcdf4
  - cdo  # unavailable on Windows; use Conda with WSL
  - dask  # parallel computing
  - pooch  # to download data
  - matplotlib  # plotting tools
  - matplotlib_scalebar
  - nc-time-axis
  - seaborn
  - cartopy
  - mapclassify
  - contextily
  - pyviz
  - hvplot
  - statsmodels
  - jupyterlab  # interactive notebooks
  - jupyter_bokeh
  - pytest  # testing and linting
  - flake8
  - pylint
  - black
  - black-jupyter

R

See this post for more information about setting up R to use with JupyterLab in an Anaconda environment.

name: r-geo
channels:
  - conda-forge
  - r
  - anaconda
dependencies:
  - r-base
  - r-essentials
  - libgit2  # to install packages from Git repositories
  - r-devtools
  - r-terra
  - r-rgdal
  - r-sf
  - r-geojsonio
  - r-rgeos
  - r-rastervis
  - jupyterlab