rastergrid

A raster grid, often written as rastergrid, is a grid of regularly spaced cells used to store spatially distributed values in raster data models. It is a foundational concept in geographic information systems (GIS), remote sensing, and image processing.

A raster grid is defined by its extent, cell size (resolution), and coordinate reference system. Each cell holds a value representing a property at the corresponding location, with special nodata values used to indicate missing data. Grids may be single-band or multi-band, storing different variables per cell.

Structure and indexing: grids are typically modeled as two-dimensional arrays with row and column indices. In raster data, the top-left corner is often the origin, with value retrieval based on row, column, or geographic coordinates via a transformation from world coordinates to grid indices.

Applications: raster grids are used for digital elevation models, land cover maps, climate and vegetation variables, and satellite imagery. Operations include resampling to different resolutions, reprojection to another CRS, and map algebra or local/statistical analyses using neighborhood or focal operations.

Storage and formats: common formats include GeoTIFF, HDF5, NetCDF, and IMG/GDAL-based formats; grids can be tiled and compressed to improve access speed. Pyramids or overviews are generated to speed up rendering at varying zoom levels.

Relation to other models: raster grids contrast with vector data, which represent features as discrete geometries. The choice of grid resolution involves trade-offs between detail and storage. Tools such as GDAL and libraries like rasterio support creation, manipulation, and analysis of raster grids.