HEALPix is a 1997 sphere-pixelisation algorithm that divides a 2-sphere into hierarchical, equal-area, iso-latitude pixels, standard in cosmology and astronomical data pipelines.
Key Takeaways
The H=4, K=3 projection maps the sphere to 12 quadrilateral base facets, recursively subdivided so every pixel at a given resolution has identical area.
Pixel centers lie on discrete iso-latitude rings with equal angular spacing, making fast spherical harmonic transforms practical on large CMB datasets.
ESA’s Gaia mission uses HEALPix as the basis for source identification across its billion-star catalog, not just cosmology.
Approved as keyword HPX in the FITS World Coordinate System standard by the IAU in 2006, so HEALPix maps interoperate natively with astronomical file tooling.
Supported in C, C++, Fortran90, Python (healpy, astropy-healpix), Java, Julia, and JavaScript/TypeScript, with resolutions down to 0.4 milliarcseconds.
Hacker News Comment Review
Commenters immediately frame HEALPix in the context of geospatial indexing: Uber’s H3 and Google’s S2 solve the same core problem (assign stable cell IDs to points on a sphere) at terrestrial scale, making HEALPix a direct astronomical ancestor of those systems.
The geohash analogy surfaces more than once, suggesting HEALPix is practical for any hierarchical spatial index problem, not only CMB maps.
HiPS (Hierarchical Progressive Surveys) extends HEALPix into zoomable sky atlases, showing the scheme composes well into interactive map tiling layers beyond static storage.
Notable Comments
@ahurmazda: draws explicit parallel between HEALPix and Uber H3 as the same abstraction at different scales, useful framing for geospatial engineers.
@xioxox: points to HiPS maps at Aladin/CDS as a working example of HEALPix powering progressive-zoom sky imagery.
