On Projection Videos

I’ve been meaning for some time to share these videos that I produced last year to assist in teaching projections to my students. Specifically, I wanted to use them to emphasize the importance of choosing projection parameters carefully to reduce distortions in the subject area, and to show how two different-looking maps can really be the same projection.

The first video is of an Azimuthal Equidistant projection. The standard point moves around the map, beginning in the central US and ending near the southern end of Africa. I try to point out, when showing it, that the pattern of distortion remains the same because it’s the same projection, but that the location of those distortions on the earth changes as the standard point moves, and how the map at the beginning and the map at the end are appropriate for showing different locations.

The second is of an Albers Equal Area Conic. First the central meridian moves, then the two standard parallels. Here I point out that the areas of the land features never change throughout the movie. Their shapes shift around significantly, but area is always preserved. The angle distortion moves with the standard parallels, and we can choose a set of standard parallels to best depict each area. We begin with a projection best suited for India and end with one adjusted for Sweden.

By the time I show these videos, I’ve already gone over all these projection concepts — they’re just a nice way to reinforce what we’ve already discussed. Student responses suggest that the videos have been helpful in teaching distortions and the importance of choosing projection parameters. It can be a tough thing to get your head around, and I like to approach it from several different angles to make sure I’m reaching as many of them as I can.

I made these using GeoCart (and Tom Patterson’s lovely Natural Earth raster), in a painstaking process which consisted of: 1) adjust projection parameters by a small amount (I think it was .25 degrees), 2) export image, 3) repeat 1-2 several hundred times, 4) use some Photoshop automation to mark the standard point/central meridian (though I had to add the standard parallels manually), 5) stitch together with FrameByFrame

It took many hours. Soon thereafter daan Strebe, GeoCart’s author, pointed out at the 2010 NACIS meeting that he’d added an animation feature to the program, which probably would have saved me a lot of time.

If you’d like the originals (each a bit under 40 MB, in .mov format), drop me a line.