Can WE take to the sky too?

Newscientist are running a competition to win an amazing Parrot AR Drone helicopter as part of the Ig Noble awards. This brilliant device can be controlled from your iPhone and features 2 onboard cameras.

Naturally I knew straight away what I could use it for. They are after a creative, scientific and fun use. Well, for me, that's bumblebee research - here's my submission:

I would like to use the AR Drone to further our research of Bumblebees.  Here's a link to our current project

Through our current work we have established behaviour patterns of bumblebees in the wild, such as when they leave and return to their nest and what triggers them to do so. They exhibit some remarkable behaviour - for example, here is a video showing a new baby "memorising" the nest location on its first flight:

it only does this once and it has learned where the nest is - it can then fly off.

(here is a montage set to music of lots of them doing it: )

The problem is - we don't know what happens next. In some situations we are able to time how long their trips away from the nest are, and also the colour of the pollen they bring back may give an indication of where they go. But in truth, we don't really know how far they go and we don't know whether their initial flights differ from their "established" flights. How do new baby bumblebees learn what to do?

For example - is their first flight just a short test flight?

We need the Drone to track and follow bees and find where they forage. We would like to understand what distance they fly, what patterns are exhibited by the locations they choose (e.g. are they all disperse or clumped together) - and for this we need to locate, view and measure them in foraging locations. We believe the drone could be used for this - either to spot & count marked bees in known locations or to attempt to track (probably harder) individual bees.

Although that sounds far fetched, there will be occasions when that is very possible. For example, when Queen bumblebees are searching for nest sites or hiberation sites. It is notoriously difficult to track this behaviour, although during spring we manually tracked some nest-searching queens, manually running along fields and ditches to note locations. They search along embankments, close to the ground. One issue is that the Queen can just choose to fly over a ditch or fence or over dense undergrowth to look for her next spot and you have lost her. It might only be 20 feet away, but you cannot follow.

We would like to research the locations that queens choose and how many they look for before finding a successful nest site. This would help us research and gauge the ongoing loss of bumblebee nest sites. Here, for example, is a video of a "cuckoo" bumblebee searching for an established nest to take over & evict..

This type of behaviour is extremely hard to follow on foot due to terrain - the drone could solve this problem, allowing us to chart activity levels, preferred sites and success rates. This would contribute new data to the field of bumbleebee research.