Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn spherical geometry processing techniques using GeometryOps.jl in this 13-minute conference talk from JuliaCon Global 2025. Discover how to perform accurate geometric operations directly on the sphere, achieving inherent accuracy that cannot be matched by projecting to planar space. Explore the fundamental principle that Earth can be approximated as a perfect sphere with only ~0.5% maximum error, making spherical calculations suitable for most applications. Master techniques for handling complex geometries that encircle poles or cross the antimeridian, where planar representations fail. Understand how to compute accurate geometric properties like area and arclength in latitude-longitude space, avoiding the inherent inaccuracies of naive calculations. Examine three key use cases driving this work: global grid systems and polar-encircling geometries in ocean simulations, accurate geometric property calculations that only spherical or geodesic computations can provide, and precise mapping between different grid representations in climate models. Learn how this approach enables coupling between Oceananigans.jl and SpeedyWeather.jl for complete climate simulation by computing intersection areas between grid polygons directly on the sphere. Gain insights into optimization techniques inspired by Google's s2 library and Charles Karney's GeographicLib while exploring Julia-specific implementations for spherical geometry processing.
