Dr. David Bahr
Regis University, Department of Computer Science
Time-Lapse Video Assembly
The videos are constructed from pictures taken at 11:00am every day from December 2005 through April 2006 at Mary Jane Ski Resort. The ski run is RifleSight Notch as seen from the Utah Junction parking lot. Pictures that were obscured by poor visibility due to snowstorms or fog were not included. Pictures taken at different times of day cannot be included because the sun casts different shadows at different hours, obscuring the movement. Each day in the movie lasts approximately 1/100th of a second. When a day is missing, the previous frame is scaled to last longer and cover the gap. Particularly in December and March there are 4 day gaps, and in these cases the frames last roughly 1/25th of a second (a noticeable delay in the movie).
A typical photo from March 13, 2006 is shown below.
The camera is mounted outside on a building, and therefore, despite best efforts, the waterproof camera enclosure shifts positions with wind and snow. Thermal expansion of the siding on the building also shifts the camera. The camera (a Nikon Coolpix) has an annoyingly inconsistent zoom, sometimes varying by as much as 6% of the total image size. As a result, the frames are co-registered (aligned) to three permanent survey markers that were placed on the ski run, but the field of view can vary dramatically. In many cases one or more of the survey markers drifted from the field of view and the alignment is a "best guess" based on nearby trees. Most of the videos are cropped so that the changing field of view is not visible.
Dave Bahr is shown below, aligning and servicing the camera.
Incidentally, when watching the movies, did you notice the snow rising and falling against the trunks of the trees in the upper right? The snow accumulates and melts in cycles throughout the season. You can also see some of the trees bowing under heavy snow and wind. With everything else moving, the only stable locations for survey markers are near the base of large trees or on fixed resort trail signs.
Tad Pfeffer is shown below surveying the targets mounted on the trees and trail signs.
A number of Regis College students participated in the field work. Joe Stachecki ('07) and Byron Schwab ('07) are shown below measuring the distance between bumps.
And Regis students CJ Kummer ('09) and Jesse Stephens ('06) take notes on the dimensions of individual moguls.
Allison Zech (Douglass Elementary School, Boulder, CO) also assisted by measuring the the slope of the ski run and the distance between bumps. Incidentally, in a very clever 5th grade science fair experiment (2009), Allison reconstructed a ski slope with sand in an inclined box. By repeatedly moving a toy skier down the sand, and by turning only on the backside of bumps in the sand, she showed that the sand moguls move uphill. She took photos and made measurements to back up her results. Very cool!
Please contact me directly with any questions, comments, revisions, corrections, rants, or raves.
Contact: David Bahr at dbahr at regis.edu