<languages />
[[Category:Software]]
<translate>
<!--T:1-->
[https://www.gdal.org/ 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.

<!--T:2-->
GDAL is used by a [https://gdal.org/software_using_gdal.html#software-using-gdal long list of software packages]
and its functionality can be used in scripts written in [[Python]] or [[R]].

== Using GDAL from Python == <!--T:3-->
GDAL functionality can be used via the [https://gdal.org/api/python/osgeo.html 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 === <!--T:4-->
Check which Python modules are compatible with e.g. <code>gdal/3.5.1</code>:
{{Command| module whatis gdal/3.5.1
|result=
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
}}

<!--T:5-->
We have the choice between Python 3.8, 3.9 and 3.10. Let's choose <code>python/3.10</code> for this.
{{Command| module load StdEnv/2020 gcc/9.3.0 python/3.10 gdal/3.5.1}}
{{File
  |name=osgeo_gdal.py
  |lang="python"
  |contents=
#!/usr/bin/env python3
from osgeo import gdal

<!--T:6-->
print("osgeo.gdal version:", gdal.__version__)
# osgeo.gdal version: 3.5.1
}}

=== Using osgeo under StdEnv/2023 === <!--T:7-->
Check which Python modules are compatible with e.g. <code>gdal/3.7.2</code>:
{{Command| module whatis gdal/3.7.2 
|result=
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
}}

<!--T:8-->
We have the choice between Python 3.10 and 3.11. Let's choose <code>python/3.11</code> for this.
{{Command| module load StdEnv/2023 gcc/12.3 python/3.11 gdal/3.7.2}}
{{File
  |name=osgeo_gdal.py
  |lang="python"
  |contents=
#!/usr/bin/env python3
from osgeo import gdal

<!--T:9-->
print("osgeo.gdal version:", gdal.__version__)
# osgeo.gdal version: 3.7.2
}}

== Using GDAL from R == <!--T:10-->
Several [https://cran.r-project.org/web/views/Spatial.html R-packages for Analysis of Spatial Data] directly depend on GDAL
as a System dependency. For example:
* [https://CRAN.R-project.org/package=sf sf]: Simple Features for R
* [https://CRAN.R-project.org/package=terra terra]: Spatial Data Analysis 

<!--T:11-->
The older package [https://CRAN.R-project.org/package=rgdal rgdal] has been discontinued in favor of sf and terra.

=== Installing <code>sf</code> and <code>terra</code> under StdEnv/2020 === <!--T:12-->
Installing these packages not only requires loading a <code>gdal</code> module, but also <code>udunits</code>
which is required by [https://CRAN.R-project.org/package=units units].

<!--T:13-->
{{File
  |name=install_sf_terra_StdEnv2020.sh
  |lang="bash"
  |contents=
# load required modules:
module load  StdEnv/2020  gcc/9.3.0  udunits/2.2.28  gdal/3.5.1  r/4.2.2

<!--T:14-->
# 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"

<!--T:15-->
# 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 <code>sf</code> and <code>terra</code> under StdEnv/2023 === <!--T:16-->
Note that under StdEnv/2023, in addition to modules <code>gdal</code> and <code>udunits</code>
also <code>hdf/4.3.1</code> is required.


<!--T:17-->
{{File
  |name=install_sf_terra_StdEnv2020.sh
  |lang="bash"
  |contents=
# load required modules:
module load  StdEnv/2023  gcc/12.3  udunits/2.2.28  hdf/4.2.16  gdal/3.7.2  r/4.4.0

<!--T:18-->
# create a local R library in $HOME:
mkdir -p $HOME/R/x86_64-pc-linux-gnu-library/4.4
export R_LIBS="$HOME/R/x86_64-pc-linux-gnu-library/4.4:$R_LIBS"

<!--T:19-->
# 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)"
}}

</translate>