It seems impossible: tracking every single bug in a big, buzzing swarm of bugs. But we can do it now. We have the technology.
Researchers at MIT and Yale released a bunch of midges (up to 100 of them) into a container, with three high-speed (125 FPS) 1-megapixel cameras focused on them. The cameras were synchronized to take three dimensional snapshots that tracked each insect along with its speed and orientation, and over time (about 5,400 frames), the system was able to generate complex volumetric maps of swarm behavior:
This is an interesting experiment, albeit a gross one because it involves bugs and bugs are gross. Which begs the question: why? Well, here are some things that nobody knew about swarms of midges until just now:
- Swarm size can vary, but the shape is similar: generally egg-shaped, symmetric, taller than wider, and thicker at the bottom than at the top.
- Individual midges don't really seem to align with their neighbors, unlike flocking birds and schooling fish.
- Individuals tend to fly faster horizontally than vertically.
- On average, all midges demonstrate a tendency toward the center of the swarm.
- Within the swarm there are hierarchical sub-groups.
So that's kinda neat, if you're in to midges (ick), but what's most interesting about all this is that analyzing swarm behavior with this level of detail just hasn't been done yet because of the technical problems involved. A swarm of insects is just one sort of swarm, but you could use the same basic technique on (say) a swarm of fish, or swarm of wildebeest, or a swarm of preteens at a Justin Bieber concert. In each case, the ability to track individual members of the swarm makes it possible to spot trends and correlations that would have been nearly impossible to spot in the group as a whole.
And finally, we should note that the Army Research Office provided support for the project, and we have no idea why, but feel free to use your imagination.