Geohashes covering an area

Model: The geohash grid at the precision (cells 360° / 2^⌈5p/2⌉ wide and 180° / 2^⌊5p/2⌋ tall), kept where a cell overlaps the box or polygon, or where its center is inside. Polygon edges are straight in latitude and longitude, like the cells

Show your work

1. Cell size

   360° / 2^⌈5p/2⌉ by 180° / 2^⌊5p/2⌋

   precision 6 = 0.01099° × 0.005493°

2. Geohashes

   cells that overlap the area

   20 cells in the area's bounds = 20

The same steps an agent gets from the MCP server with explain: true.

Accuracy: Exact on the latitude-longitude grid; a polygon may not cross the antimeridian

When to use this: Use this to turn an area into a set of geohash prefixes you can index or query with: the cells that touch a bounding box or polygon, or only those whose centres fall inside. It is the usual way to make a spatial query out of a string prefix match.

Limitations: Pick the mode for the job. Overlap keeps every cell the area touches, so nothing inside is missed and some outside is included -- the right choice for a query. Centre keeps only cells centred inside, so cells are more or less contained, and a polygon smaller than one cell covers nothing at all, which is correct and surprising. Geohash cells are not square and alternate between wide and tall as precision grows, because each character adds five bits split unevenly between longitude and latitude. Polygon edges are treated as straight in latitude and longitude, the same space the cells live in.

Worked example: Downtown Pittsburgh at precision 6. Source: Geohash cell arithmetic (Niemeyer 2008). It is golden test vector v001, and every build checks the tool still gives its answer within its tolerance.

Review: Not yet independently reviewed by a GIS professional.

Last verified: 2026-09-18, when a maintainer last confirmed this tool's sources at the issuer. See the sources ledger.

Status: version 1.0.0, core 0.1.0. See this tool in the verification report.

Checked against: 22 golden test vectors (download the test vectors, each with its source and tolerance). See how results are checked and every source.

Sources

• Geohash (public domain algorithm), Niemeyer, G., geohash.org, Original description, as archived (the site is gone). Base-32 alphabet 0123456789bcdefghjkmnpqrstuvwxyz, longitude bit first.