GeoPython 2024: Discrete Global Grid Systems (DGGS) in Python and Xarray

Benoît Bovy
Discrete Global Grid Systems (DGGS) in Python and Xarray
XDGGS is an Xarray extension that makes it easy to work with data cubes referenced on discrete global grid systems (DGGS) in Python.
Discrete Global Grid Systems (DGGS) together form a family of spatial reference systems that each partition the whole globe into a finite number of cells or zones usually organized as a hierarchical structure and such that it facilitates spatial indexing. Those reference systems offer a powerful framework for representing geospatial features and/or coverages that overcomes some of the limitations of the more common representations such as projected rasters (distorsion) and vectors (arbitrary geometry). There are a variety of DGG systems for a wide range of applications, each system being defined by the shape of its cells (square, triangle, hexagon) among other properties. Despite the existence of mature libraries implementing DGGS in low-level languages (e.g., S2, H3, HEALPix, DGGRID), the Python geospatial ecosystem is still lacking a generic and high-level tool that allows working efficiently with DGGS data.
In this talk we will start with a brief explanation of the concept of DGGS and a review of the current libraries exposing DGGS capabilities to Python. Next we will introduce XDGGS, a new Xarray extension that allows interacting with data on various DGG systems (backends) via a unified, high-level and user-friendly API. Still at an early stage of development, it currently supports two backends: H3 and HEALPix. XDGGS nicely fills a gap in the Xarray ecosystem that is already widely adopted for handling geospatial data including global gridded datasets (built-in core features), rasters (Rioxarray) and more recently vector data cubes (Xvec) and unstructured meshes (UXarray).