Buffering and overlay are the two most used operations in all of GIS. A buffer builds zones of specified distance around features — 500 m around schools, 30 m around streams. Overlay combines layers geometrically — intersecting, unioning, clipping — so that one geometry carries the attributes of many themes. Between them they answer the bread-and-butter questions: what is near what, and what coincides with what.
Buffering: proximity made explicit
Buffers translate policy language into geometry: setback rules, protection zones, service radii, impact corridors. Variants matter — variable-distance buffers driven by attributes (pipeline class, road category), one-sided buffers along lines, dissolved versus separate rings.
The honest caveat: straight-line distance is often the wrong measure. Where movement follows networks, travel-time service areas (network analysis) should replace circular buffers — a 1 km buffer and a 15-minute walk are different worlds.
Overlay: combining themes
Intersect keeps only shared areas, union keeps everything with combined attributes, clip cuts one layer by another’s extent, erase removes it, identity blends. Behind each stands the same machinery: geometric intersection plus attribute inheritance, producing the composite polygons on which criteria can be evaluated.
Classic applications are everywhere: land suitable AND available AND outside hazard zones; parcels within the flood extent; habitat lost inside the project footprint.
The practical pitfalls
Overlay multiplies sliver polygons wherever near-coincident boundaries meet — the topology problem in action. Mixed projections silently distort distances and areas; buffering in geographic coordinates is the canonical rookie error. And overlays inherit the worst positional accuracy among their inputs, a fact suitability maps rarely confess.
Disciplined practice cleans topology first, works in appropriate projected CRS, and carries input accuracy into the interpretation of results.