• Free software: Apache 2
• Documentation: https://geoxarray.github.io/.

Geolocation utilities for xarray objects. Geoxarray is meant to bring together all of the features and conversions needed by various python packages working with geolocation xarray objects. This means being able to convert between various coordinate system implementations (rasterio, cartopy, pyresample, NetCDF CF grid mapping, etc). It also means providing basic access to properties of the geolocation information like bounding boxes.

The geoxarray library is available on PyPI and can be installed with pip:

pip install geoxarray

For the most recent development versions of geoxarray, it can be installed directly from the root of the source directory:

pip install -e .

Or to install into an existing conda-based environment:

.. code-block:: bash

conda install -c conda-forge geoxarray

Besides the xarray dependency, the geoxarray package uses CRS objects from the pyproj library. Additionally, geoxarray has a lot of optional dependencies when it comes to converting to other libraries' CRS or geolocation objects. These libraries include, but may not be limited to:

• rasterio
• cartopy
• pyresample

At the time of writing, rioxarray is an independent project whose features related to CRS and dimension handling are very similar if not exactly the same as geoxarray. Rioxarray existed first and paved the way to show how CRS information can be handled in an xarray-friendly way. Much of geoxarray is inspired by rioxarray, if not copied directly. Portions of code copied from rioxarray are noted in docstrings for that code and are under the Apache License of rioxarray which has been copied as LICENSE_rioxarray in the geoxarray package and repository.

Geoxarray is actively being developed as a side project. Additions and modifications are done as developers have time. If you would like to contribute, suggest features, or discuss anything else please file a bug on github.

See the documentation website for how-tos, concepts, and API documentation.

Parse various formats of storing Coordinate Reference System information:

import geoxarray

pyproj_crs = my_data_arr.geo.crs

Write CRS information in a CF-compatible way or add CRS information to an object that didn't have any:

import geoxarray

new_dataset = my_dataset.geo.write_crs("EPSG:4326", inplace=False)