Hexagonal Hierarchical Spatial Index (H3)
Overview
Hexagonal Hierarchical Spatial Index (H3) is a geospatial indexing system developed by Uber that uses hexagonal grids to partition the Earth’s surface for efficient spatial analysis and data representation. H3 provides a hierarchical system of increasingly fine hexagonal cells, enabling multi-resolution spatial analysis.
Key Concepts
Hexagonal grid is chosen because hexagons have uniform adjacency (6 neighbors) and approximate circles well. Resolution refers to the 16 levels (0-15) of increasingly smaller hexagons. Cell index is a 64-bit integer uniquely identifying each hexagon. Hierarchy allows each hexagon to contain approximately 7 child hexagons at the next resolution. Edge and vertex modes provide indexing for hexagon boundaries and corners.
Resolution Levels
| Resolution | Avg. Area | Edge Length | Use Case |
|---|---|---|---|
| 0 | 4.3M km² | 1,108 km | Continental |
| 4 | 1,770 km² | 22 km | Regional |
| 7 | 5.2 km² | 1.2 km | City district |
| 9 | 105,332 m² | 174 m | Neighborhood |
| 12 | 307 m² | 9.4 m | Building |
| 15 | 0.9 m² | 0.5 m | Sub-meter |
Advantages
- Uniform distance to neighbors (unlike square grids)
- Efficient spatial joins and aggregation
- Consistent global coverage
- Fast point-to-cell and cell-to-point operations
Appendix
Created: 2025-12-13 | Modified: 2025-12-13