GDAL
| This site replaces the former Compute Canada documentation site, and is now being managed by the Digital Research Alliance of Canada. Ce site remplace l'ancien site de documentation de Calcul Canada et est maintenant géré par l'Alliance de recherche numérique du Canada. |
GDAL is an open source translator library for raster geospatial data formats. It can be used as a a library, as which presents a single abstract data model to the calling application for all supported formats. It also comes with a variety of useful command line utilities for data translation and processing.
GDAL is used by a long list of software packages and its functionality can be used in scripts written in Python or R.
Using GDAL from Python
GDAL functionality can be used via the osgeo package, which we install as an extension to the GDAL module. In order to use it, you need to load a compatible Python module alongside the GDAL module.
Using osgeo under StdEnv/2020
Check which Python modules are compatible with e.g. gdal/3.5.1:
[name@server ~]$ module whatis gdal/3.5.1
gdal/3.5.1 : Description: GDAL is a translator library for raster geospatial data formats...
gdal/3.5.1 : Homepage: https://www.gdal.org/
gdal/3.5.1 : URL: https://www.gdal.org/
gdal/3.5.1 : Compatible modules: python/3.8, python/3.9, python/3.10
We have the choice between Python 3.8, 3.9 and 3.10. Let's choose python/3.10 for this.
[name@server ~]$ module load StdEnv/2020 gcc/9.3.0 python/3.10 gdal/3.5.1
#!/usr/bin/env python3
from osgeo import gdal
print("osgeo.gdal version:", gdal.__version__)
# osgeo.gdal version: 3.5.1
Using osgeo under StdEnv/2023
Check which Python modules are compatible with e.g. gdal/3.7.2:
[name@server ~]$ module whatis gdal/3.7.2
gdal/3.7.2 : Description: GDAL is a translator library for raster geospatial data formats...
data translation and processing.
gdal/3.7.2 : Homepage: https://www.gdal.org/
gdal/3.7.2 : URL: https://www.gdal.org/
gdal/3.7.2 : Compatible modules: python/3.10, python/3.11
gdal/3.7.2 : Extensions: osgeo-3.7.2
We have the choice between Python 3.10 and 3.11. Let's choose python/3.11 for this.
[name@server ~]$ module load StdEnv/2023 gcc/12.3 python/3.11 gdal/3.7.2
#!/usr/bin/env python3
from osgeo import gdal
print("osgeo.gdal version:", gdal.__version__)
# osgeo.gdal version: 3.7.2
Using GDAL from R
Several R-packages for Analysis of Spatial Data directly depend on GDAL as a System dependency. For example:
The older package rgdal has been discontinued in favor of sf and terra.
Installing sf and terra under StdEnv/2020
Installing these packages not only requires loading a gdal module, but also udunits
which is required by units.
# load required modules:
module load StdEnv/2020 gcc/9.3.0 udunits/2.2.28 gdal/3.5.1 r/4.2.2
# create a local R library in $HOME:
mkdir -p $HOME/R/x86_64-pc-linux-gnu-library/4.2
export R_LIBS="$HOME/R/x86_64-pc-linux-gnu-library/4.2:$R_LIBS"
# install sf and terra from a Canadian CRAN mirror:
R -e "install.packages(c('sf', 'terra'), repos='https://mirror.csclub.uwaterloo.ca/CRAN/', dep=TRUE)"
Installing sf and terra under StdEnv/2023
Note that under StdEnv/2023, in addition to modules gdal and udunits
also hdf/4.3.1 is required.
# load required modules:
module load StdEnv/2023 gcc/12.3 udunits/2.2.28 hdf/4.2.16 gdal/3.7.2 r/4.3.1
# create a local R library in $HOME:
mkdir -p $HOME/R/x86_64-pc-linux-gnu-library/4.3
export R_LIBS="$HOME/R/x86_64-pc-linux-gnu-library/4.3:$R_LIBS"
# install sf and terra from a Canadian CRAN mirror:
R -e "install.packages(c('sf', 'terra'), repos='https://mirror.csclub.uwaterloo.ca/CRAN/', dep=TRUE)"