H3 + Ocean Data: Hexagonal Geospatial for Marine Science

Why hexagons are better than squares for ocean analysis.

The Problem

Ocean data comes from everywhere: satellites, buoys, ships, drones. Each source has different resolutions and coordinate systems. Joining them is painful.

Enter H3

H3 is Uber’s hexagonal hierarchical spatial index. Instead of lat/long points or square grids, it divides Earth into hexagons at multiple resolutions.

Why hexagons?

Uniform distance — neighbors are equidistant (squares have diagonal problems)
No edge effects — hexagons tile without gaps at any scale
Hierarchical — zoom in/out with consistent parent-child relationships

Ocean Use Cases

Multi-Dataset Joins

import h3

# Convert any lat/long to a hex cell
cell = h3.latlng_to_cell(lat, lng, resolution=5)

# Now join datasets by cell ID, not fuzzy coordinates
chlorophyll_df['h3'] = chlorophyll_df.apply(
    lambda r: h3.latlng_to_cell(r.lat, r.lng, 5), axis=1
)
zooplankton_df['h3'] = zooplankton_df.apply(
    lambda r: h3.latlng_to_cell(r.lat, r.lng, 5), axis=1
)

merged = chlorophyll_df.merge(zooplankton_df, on='h3')

Resolution Selection

Resolution	Hex Edge	Use Case
3	~70 km	Basin-scale patterns
5	~8 km	Regional analysis
7	~1 km	Coastal studies
9	~150 m	High-res satellite

Aggregation

# Aggregate to coarser resolution
fine_cells = df['h3_res7'].unique()
coarse_cells = [h3.cell_to_parent(c, 5) for c in fine_cells]

Why Not Squares?

At ocean scales, square grids distort near poles. H3’s icosahedral projection maintains consistent cell areas globally. For marine science spanning latitudes, this matters.

Hexagons: the bestagon for ocean data.

Source: C4IROcean-OceanDataPlatform