Which Projection for Your Study Area Map

Most study area maps do not need you to think hard about projection. A province sized area near the equator looks fine in almost anything. The trouble starts at three points: when the map is wide, when it is far from the equator, and when somebody measures something off it.

The short answer

For a study area smaller than about sixty degrees across, sitting below about sixty six degrees latitude, use an equidistant cylindrical projection with the standard parallel set to the middle of your map. For anything wider, or anything polar, use Lambert azimuthal equal area. For a whole world locator, use Robinson. Use Mercator only when you have a specific reason.

That covers nearly every study area figure. The rest of this explains why, which matters when a reviewer asks.

Equidistant cylindrical, and the standard parallel

A plain equirectangular map plots longitude and latitude straight onto x and y. It is the simplest thing that can work, and away from the equator it stretches east to west, which is why quick figures of Scandinavia or Canada look subtly wrong.

Setting the standard parallel to the middle latitude of your own map cancels most of that. The horizontal axis is scaled by the cosine of that latitude, so a box a hundred kilometres across measures the same on screen whether you run the ruler along it or up it. At mid latitudes the two directions agree to better than one percent across a province sized area, which is well inside what anybody will measure off a printed figure.

This is the default for a reason: it is simple to state in a caption, the grid stays square so coordinates are easy to read off, and over a small area the distortion is negligible.

Mercator, and when it is the wrong instinct

Mercator preserves angles and makes north up everywhere, which is why every web map uses it. It also inflates area with latitude, badly. Greenland is the standard example, but the effect is already obvious across Canada or Scandinavia.

Use it when your figure sits next to a screenshot from a web map and you want the shapes to match, or when bearings matter. Do not use it because it is familiar. A study area map that exaggerates the size of the northern half of the frame invites a comment, and there is no upside for a regional figure.

Lambert azimuthal equal area, for wide and polar maps

Cylindrical projections fall apart near the poles: latitude lines crowd together or stretch apart, and above about sixty six degrees the map stops being usable. Lambert azimuthal equal area centres itself on your study area and keeps areas true everywhere, at the cost of a curved grid.

Use it for continent scale maps, for anything Arctic or Antarctic, and whenever you are comparing the size of regions on the map. It is the only one of the four here that keeps area, which is why an area figure quoted from an equal area projection is defensible and one from Mercator is not.

The curved grid is the honest trade. Some readers find it unfamiliar, and it is worth naming the projection in the caption for that reason alone.

Robinson, for world locators

Robinson is a compromise: it preserves nothing exactly and looks reasonable everywhere. That makes it a poor choice for measurement and a good choice for the one job it is usually given, which is a small world map showing which country you are talking about.

A world locator carries no scale bar and no measurements, so nothing is lost. Do not use it for the panel that carries your data.

What this means for your scale bar

A scale bar is only exactly right where the local scale matches the bar. On an equidistant cylindrical map of a province that is everywhere, to within a fraction of a percent. On a Lambert map of a continent it is not: the scale genuinely changes across the frame, and a bar measured at the map centre can be half a percent or more out at the edges.

The practical answer is to keep the bar on the detail panel and leave the locator without one. If a wide map must carry a bar, say in the caption that it applies at the map centre. A tool that measures the bar where it is drawn rather than at the map centre gets you closer, but no single straight bar is exactly right across a continent, and pretending otherwise is worse than stating the limitation.

UTM, which is a different question

People often mean UTM when they say projection, because that is what their coordinates are in. UTM zones are six degrees of longitude wide and are excellent for measurement inside a zone. They are a poor choice for a figure that spans several zones, because each zone has its own grid and they do not line up.

For a figure, name the zone your coordinates came from in the caption and draw the map in something that suits the extent. For measurement, convert to the local UTM zone with the coordinate converter and work in metres.

Writing it in the caption

One clause. Equidistant cylindrical projection, WGS 84, standard parallel 46 degrees north. Or for a wide map, Lambert azimuthal equal area centred on 10 degrees north, 8 degrees east, WGS 84. A reader who wants to measure knows what they are working with, and a reviewer who checks finds it already answered.

The study area map maker applies the rules above automatically and reports the centre coordinates and UTM zone for each panel, which is most of what the caption needs.

Related Calculators and Reference

• Study Area Map Maker, which picks the projection by extent and latitude and lets you override it.
• Coordinate Converter to move between latitude and longitude and the right UTM zone for measurement.
• Study area map rules for a thesis, where naming the projection is item four on the checklist.
• GPS Waypoint Logger for collecting coordinates with the zone recorded alongside them.
• USGS on map projections for a fuller treatment of what each family preserves.