GeoStats.jl is a Julia framework for geospatial data science and geostatistical modeling. It’s fully implemented in Julia and designed to provide an extensible, high-performance stack that handles spatial domains, interpolation, simulation, learning, and visualization. The package is modular: it breaks out geometry, spatial domains, transforms, variograms, covariance models, and modeling into subpackages (e.g., GeoStatsBase, GeoStatsModels, GeoStatsTransforms). Users can represent georeferenced tables (points + attributes), define domains (grids, meshes, structured/unstructured), and then apply geostatistical operations such as kriging, interpolation, simulation, variogram estimation, and learning-based prediction. Visualization is supported via integration with Makie.jl to produce spatial renderings, mesh visualizations, and variable overlays.
Features
- Fully written in Julia with modular extensibility for geostatistical workflows
- Offers geometric processing and state-of-the-art spatial algorithms
- Includes visualization support and integrates with plotting tools like Makie.jl
- Part of a growing Julia ecosystem for geospatial modeling (JuliaEarth, JuliaGeo)
- Supports advanced features such as transiogram functions for spatial transitions
- Open source with documentation and community contributions
Project Samples
