Question 4 of 12: Vector versus Raster Data Structures

Question 4: Vector versus Raster Data Structures (12 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

The two models store space differently — vector as discrete geometric primitives (points, arcs, polygons) referenced by coordinates, raster as a regular grid of cells whose location is implicit — and each criterion favours one or the other.

(a) Data storage. Vector is compact for discrete features — only the boundary coordinates are stored — whereas raster stores a value for every cell, so full-coverage grids can be very large (though compression and run-length/quad-tree encoding help). Advantage: vector for discrete objects; raster is simpler but bulkier. (b) Point precision. Vector coordinates can be stored to arbitrary precision, giving sharp, exact geometry; raster precision is capped by the cell size, so positions are only good to roughly half a cell and boundaries look blocky. Advantage: vector. (c) Feature representation. Vector represents discrete, well-bounded objects (parcels, roads) with clean boundaries and identity; raster represents continuous fields (elevation, temperature, land cover) naturally. Advantage depends on the phenomenon — vector for discrete features, raster for continuous surfaces. (d) Feature searching. Vector carries explicit topology and indexed attribute tables, so querying "which parcels border this road" or attribute searches are efficient; raster has no inherent object identity, so feature-level searching is awkward and usually requires reclassification. Advantage: vector. (e) Change detection. Comparing two dates is straightforward in raster: overlay (subtract or cross-tabulate) co-registered grids cell-by-cell to map change directly; equivalent vector overlay requires topological intersection and is more complex. Advantage: raster. (f) Satellite digital image data. Satellite imagery is inherently a raster of pixels, so the raster model is the native structure for storing, displaying and analysing it (band math, classification). Advantage: raster.