Ask A Cartographer

Best projection

February 07 2008 | 0 comments

I am working with data predominantly within Edmonton, AB Canada (although I do use the data sometimes for all of Alberta). WHich is the best projection to use that will preserve distance and shape? (distance is very important)


Mapping Center Answer:

The best projection to use if you are interested in distances is an equidistant projection – the azimuthal equidistant projection centered on distances and directions to all places will be true, but only from center point of the projection (Edmonton in this case). While distances will be correct between points along straight lines through center, all other distances incorrect. Distortion of areas and shapes increases away from center point.
It is impossible to preserve both distance and direction – one will have to be sacrificed to preserve the other. But there are compromises. If you want to have both shape and distances to be less distorted, you next best bet is the Equidistant Conic projection. While this is not conformal, perspective, or equal area (it is really a compromise between Lambert Conformal Conic and Albers Equal Area Conic) it is a good choice for mid-latitudes (like where Edmonton is). Distances are true only along all meridians and along one or two standard parallels. Directions, shapes and areas are reasonably accurate, but distortion increases away from standard parallels.

To help, you can take a look at the USGS Map Projections poster – you will find this on the walls of many workplaces that deal with making map projection decisions. You can also access it online at: The beauty of this poster is that it gives you the advantages and disadvantages of some of the most commonly used map projections. At the bottom of the page are tables summarizing map projection by their distortion properties as well as the extents at which they can be used and their general uses. Also, there is a glossary at the very bottom if you are unsure of some of the terminology that relates to map projections.

If you would like to post a comment, please login.

Contact Us | Legal | Privacy |