Nothing confuses new GIS users faster than a map whose layers refuse to line up. One road layer sits 200 metres east of the satellite image, another appears in the ocean, and a distance measurement that should read 850 metres reads 0.007 instead. In almost every case the cause is the same: the layers use different coordinate reference systems, or CRS for short, and the software has not been told how to reconcile them.

A CRS is the rulebook that turns coordinates into places on Earth. It combines a datum, which models the shape of the Earth, a projection, which flattens that curved surface onto a flat map, and a unit. Geographic systems such as WGS84, EPSG:4326, store positions as latitude and longitude in degrees. Projected systems such as WGS84 UTM Zone 43N, EPSG:32643, store positions as eastings and northings in metres. Degrees are excellent for global sharing and GPS, but terrible for measuring area, because one degree of longitude is about 111 kilometres at the equator and shrinks to zero at the poles.

Projected systems also have national variants that can outperform generic UTM for a specific country. Many mapping agencies publish a national grid tuned to minimise distortion across that country, and cadastral or engineering work should use the agency system when the client requires it. The principle stays the same: measure in metres in a projection designed for your area, share in latitude and longitude for global compatibility, and record the EPSG code on every output. A map without its EPSG code is like a measurement without its unit — technically present, practically useless — so add the code to filenames, legends and metadata as a matter of routine.