Raster data

Raster data represents space as a regular grid of cells, each holding a value. It is the counterpart to vector data, which represents things as points, lines and polygons. Where vector suits objects with edges, a building, a street, a district boundary, raster suits things that vary continuously and have no natural edges: temperature, elevation, rainfall, land cover.

Satellite imagery and sensor products arrive as raster, which is why raster work and remote sensing usually go together. The resolution, meaning the ground size of one cell, sets what you can see and what you cannot. A 1,000 metre cell describes a region, a 10 metre cell describes a neighbourhood, and no amount of processing recovers detail the resolution never captured.

Most analysis mixes the two. A raster of surface temperature becomes useful when it is summarised onto vector districts, or read at the locations of vector points. That step, moving between the grid and the objects, is where the assumptions sit: an average across a district can hide the hot street that prompted the question.