Three meter grid index

It divides the globe into narrow latitude rows and sizes each row's longitude columns for an approximately three-meter footprint. The response provides the combined index, row, column, number of columns in that row, center coordinate, and geographic bounds. It is useful when a compact repeatable location key matters more than a street address, and it works without a network lookup or a proprietary word dictionary.

Convert a coordinate into a repeatable numeric address

Provide `lat` and `lon` as decimal degrees and the calculator assigns that point to exactly one global cell. The main `grid_index` value joins the latitude row and longitude column with a period, giving applications a compact key that can be stored, compared, logged, or placed in a URL. The separate `row` and `column` fields make sorting and database work easier, while `columns_in_row` explains the geometry of the selected latitude band. The same finite inputs always produce the same output because the algorithm has no network request, random choice, clock, or mutable reference table. Coordinates on ordinary boundaries follow a consistent half-open rule: the southern or western boundary is included and the northern or eastern boundary is excluded. The North Pole is assigned explicitly to the last row, while longitude 180 degrees is treated as the same antimeridian as -180 degrees. Those conventions remove ambiguous duplicate addresses at the edges of the coordinate system.

Understand what three meters means on a spherical grid

The calculation models Earth as a sphere with a fixed mean radius. It creates equal angular latitude rows whose north-to-south distance is no more than three meters. For each row, it measures the circumference of the parallel through the row center and chooses enough longitude columns that their centerline width is also no more than three meters. The number of columns therefore decreases toward the poles, where circles of latitude become shorter. This is more useful than applying one longitude step everywhere, which would make east-to-west cells progressively narrower at high latitudes. The returned bounds describe the exact angular rectangle used by this definition, and the returned center gives a convenient representative coordinate. Physical size remains approximate: Earth is not a perfect sphere, terrain adds surface distance, and positioning devices have their own uncertainty. Use the index for deterministic grouping and addressing, not as a substitute for cadastral boundaries, safety-critical navigation, or a professional geodetic survey.

Use the index safely in data systems

Store `grid_index` as text because it contains two potentially large integers separated by a period. Do not parse the combined value as a floating-point number, since that would discard the distinction between the row and column and can lose integer precision in some environments. For compound database keys, storing `row` and `column` in separate integer columns is equally valid; include a version or grid name in long-lived datasets so a future algorithm cannot be confused with this definition. Nearby coordinates usually share an index, which makes the key useful for deduplication, coarse presence records, cache partitions, and privacy-aware aggregation. However, two points separated by a cell boundary receive different keys even when they are centimeters apart. If proximity matters, compare the selected cell and its neighboring rows and columns or use an actual distance calculation. The scheme is inspired by the approximate scale associated with what3words, but it is numeric and independent: it neither converts official three-word addresses nor reproduces their proprietary cell identifiers or word assignments.