24 hour sun trajectory

Text and pictures © 2005-2024 Guillaume Dargaud
Last updated on 2021/11/05

"Photography is the recording of strangeness and beauty with beguiling precision"    — Sebastian Smee.

The winterover is now long finished and I've been home for the last few months, busy getting reacquainted with the world, but also sorting slides, scanning images and organizing my archive of images of the year in Antarctica. I took about 50 rolls during the summers and 120 during the winter. Of those 6000 or so slides I selected no more than 500 images for scanning and further work. Most images stand on their own and I could add them to the 16 pages already written of my winterover blog, but what would be the point, you've seen similar images taken by the much lower quality digital camera I was also using during the winter. No, on those few pages of epilogue I'll add the images that truly stand out from what you just saw.

Above: The trajectory of the sun above Concordia over a 24 hour period. Scroll to the right if you don't have a large screen.

What, only a single image on this page ? Well, I must say that this is most likely the most difficult image I've ever taken. Some statistics: 15 years in the planning, 5 attempts, 24 hours to perform a full sequence of shots, several hours to scan slides and clean up the resulting images, a day of testing to figure out the best way to assemble them, a day to perform the assembly in a panorama program, 20 minutes of computation on an AMD 64 dual core to create the base image and then half a day of tweaking to polish up the image.

The result is an image measuring 20000x5200 pixels, yeah, that's about a hundred megapixels, 50 times your average cell phone camera. But what is it ? This image shows the trajectory of the sun on the high antarctic plateau over a 24 hour period. With a wide angle I took 24 pictures at precisely one hour interval, each time turning the camera towards the sun but keeping the horizon in the middle of the frame. Images were then assembled carefully to produce this image.

Now if this image had been taken at the south pole itself the sun would stay horizontal, but since Dome C is at 75 south latitude, there's a 30 degree difference in elevation between the sun at noon (above the Concordia building) and at midnight. Now onto practicalities: I used a pole as a reference point and went out every hour to dutifully take the shots. This is the most delicate part of the process as there are many possibilities for failure such as the sky becoming cloudy (happened on two previous attempts), loss of reference (a tripod I used on a previous attempt was moved by a well intentioned person), arriving late for a shot (one failure due to pranksters who finished my roll by taking shots of their hairy butts), or even all the images being lost after successful processing !

Left: The same image, seen as a full fisheye projection pointing vertically down (also aptly named 'polar panorama'), showing nicely that the sun rotates following a circle around the earth, and not some kind of weird sine curve as could be imagined from the above image.

So after an attempt at DdU in 1994, another one at Dome C in 2000 and two during the 2004 summer campaign, I managed to take the proper sequence of shots 2 days before the arrival of my successor in november 2005. Hope you find it worth the effort. By the way, I've seen several similar images which are very obvious fakes, probably done with some kind of astronomy software. Here's some more information on how to combine images.