Fishnet grid from bounding box

Supply the western, southern, eastern, and northern edges in decimal degrees along with a square cell size. The result reports the grid dimensions and every cell's row, column, and exact bounds. Cells begin at the southwest corner, follow a predictable order, and are clipped along the far edges so the requested box is covered exactly even when its width or height is not evenly divisible by the chosen size.

Define the bounding box and cell size

Start with a conventional longitude and latitude bounding box. West and east describe longitudes from -180 to 180, while south and north describe latitudes from -90 to 90. East must be greater than west, and north must be greater than south. The capability intentionally does not interpret a west value greater than east as an antimeridian crossing, because wrapping would make column order and clipped edge behavior less obvious. Set cell_size in decimal degrees. The same value controls width and height, creating angular squares rather than squares of equal physical area. A one-degree longitude span becomes physically narrower toward the poles, so this grid is best for transparent geographic partitioning, tiling, sampling, and display preparation where angular boundaries are appropriate. If equal distances or equal areas matter, project the source extent into a suitable planar coordinate system and use a projected-grid workflow. Choose a size that produces enough detail for the task without creating unnecessary output. The request is rejected before allocation if it would exceed 10,000 cells. This explicit ceiling keeps browser, API, and downstream JSON handling predictable.

Understand coverage, clipping, and ordering

The fishnet is anchored at the exact southwest corner supplied in the request; it is not aligned to a hidden global origin. Row zero begins at south, and column zero begins at west. Each normal cell extends by cell_size toward the north and east. When the bounding-box width or height is not an exact multiple of that size, cells in the final column or row are shortened and clipped to east or north. No cell extends beyond the requested box, and no gap is left inside it. The response includes rows, columns, and cell_count so you can validate expected output volume without recounting the cells array. Cells are returned in row-major order: all columns of the southernmost row appear first, followed by the next row northward. Every record contains a zero-based row and column and four numeric bounds named west, south, east, and north. Coordinates are normalized to twelve decimal places to remove artifacts such as 0.30000000000000004 that can otherwise arise from repeated floating-point addition. Use the returned bounds directly when drawing rectangles, constructing polygons, labeling tiles, or splitting another dataset.

Use a deterministic fishnet in data workflows

A generated fishnet is useful whenever a continuous study area must become repeatable rectangular units. A mapping application can turn each bounds record into a polygon ring for visualization. A sampling workflow can attach one survey task to every row and column. An analytics pipeline can test points against returned cells, aggregate records per tile, or divide a large query into smaller geographic requests. Because the origin is the supplied bounding box, two calls produce matching cell identifiers only when west, south, east, north, and cell_size are identical; retain those inputs with any stored row and column. The operation is deterministic and uses no map provider, network request, random seed, clock, or shared state. Identical JSON input therefore gives identical grid geometry and ordering. This capability returns bounds rather than GeoJSON to keep the response compact and neutral between GIS formats. It does not calculate centroids, areas, neighbors, point membership, intersections, or coordinate transformations. Those steps can be performed downstream from the explicit bounds. Interactive browser execution is free, while each successful API request uses the published base price of $0.002.