Analysis of the Data

There were a lot of questions to answer. Where? How high? How accurate? Where was the camera pointing during each photo? I thought I was going to hit 91,000'... did I hit that mark?


Google Maps Trajectory


The GPS data retrieved from instamapper was essentially useless for modeling the trajectory. The only other available data was the photographs, so photogrammetry was utilized to identify the camera location. This was done by identifying distinguishable locations in the photographs and locating them in Google Earth. X,Y, and Z coordinates for those distinguishable locations were obtained from Google Earth and the distances between those points were calculated. Next, the distances between the points were measured on each photograph. Given enough points, the comparative distances between the different data sources, the height, location, and camera tilt were then obtained.

Photogrammetry Images

Above is a side-by-side comparison of point analysis in Google Earth and on the photograph.
(Note: More points than just the shown few were actually utilized for the photographic model.)

In all, 16 photographs were analyzed in detail to map the travel time, trajectory and elevation. The total flight time was 2hr 57mins 39sec based on the time stamp on the photographs, or 43 minutes longer than anticipated. Based on the data obtained from the photos, the balloon burst 2hr 27min 28sec into the flight at an elevation of 93,032 feet. As depicted below, burst occurred 25 miles East of the predicted location, 59 minutes later than anticipated, and 9,518 feet higher than calculated.


Actual Trajectory vs Predicted Trajectory

Above is a depiction of the actual trajectory vs. the "lighter" predicted trajectory. Clicking the image will download the google earth kml file for both trajectories.



The image above was roughly stitched together as an aid when trying to identify photo orientations. Left is looking Northwest towards Lancaster, off-center right is looking Southeast towards the Chesapeake Bay, and the right portion is again looking towards the Northwest. The image itself is rough, but was a good quick reference tool. I'll eventually go back and reblend these images.