Datagridded

Datagridded is a term used to describe data organized in a regular lattice of cells, typically represented as an N-dimensional array where each cell holds a value and coordinates map to discrete grid indices. This organization emphasizes fixed geometry and uniform spacing, enabling consistent indexing, slicing, and vectorized computation across the grid.

In practice, datagridded data is stored as multidimensional arrays on disk or in memory, with axes representing dimensions such as spatial coordinates, depth, time, or other modalities. Metadata records axis bounds, resolution, and units. Common formats for gridded data include NetCDF, HDF5, and raster-like TIFF, while database systems implement datagridded storage via array data types and chunked storage in array databases such as SciDB or Rasdaman.

Operations commonly supported on datagridded data include indexing by grid coordinates, subsetting with slices along axes, aggregations, regridding, resampling, and aligning grids from different sources. These features facilitate efficient visualization, exploration, and numerical processing.

Applications of datagridded data span geospatial and earth sciences, climate and weather modeling, remote sensing, digital elevation models, scientific simulations, and image data processing. The grid structure simplifies parallel computation and facilitates operations that assume uniform spacing, but it can lead to high memory usage for large grids and may require reprojection or interpolation when integrating disparate grids.

Datagridded data is related to, but distinct from, relational tabular data. It aligns closely with raster data, gridded datasets, and array-based data models, and is a core concept in array databases and scientific data formats.