I’ve been interested in small-scale variation in temperature for sometime, having worked on the impacts of thermal variation on dragonflies for my PhD. However, measuring temperature is a complicated task… Where do you measure? How often? What time of day? I have been thinking about this kind of thing when I started coming across Public Lab projects that were conducting aerial surveys using balloons. That got me thinking about flying, and before you know it I’ve pinched a colleague’s quadcopter and we’re flying (cautiously) around the University of Leeds campus:

Having successfully flown the copter around, we were interested in whether we would be able to take thermal images from the underside. At this point I borrowed another bit of kit from a colleague: a FLiR i3 camera that takes thermal pictures from which you can measure temperature, which we attached (fairly securely) to the undercarriage of the quadcopter:

What you can see there is the quadcopter with the camera strung across the legs. The camera takes photos when a trigger is pulled and so we rigged a paint brush and rubber band to keep the trigger constantly depressed leading to a photo every 2 seconds. Now it was time to survey a few sites, for which we used a set of ponds at Rodley Nature Reserve in Leeds. The result was a quite well-controlled, low-level fly-past of a series of ponds:

So we got lots of data, a Nobel Prize and everything was fine.  Actually, nothng in that previous sentence is true… We got zero useful data, I’m still waiting for the Nobel committee to call, and as a result the pilot study (pun not intended) didn’t really give anything useful.  I think the problem was with the camera: the camera wanted to recalibrate after every photo, but with the trigger constantly depressed there wasn’t enough time to calibrate and take an image.  What came out were a series of homogeneous images that gradually tended towards -40 degrees. It’s cold in Yorkshire, but not that cold. Still, it was an interesting experiment, and I am now pursuing an alternative set up with a company here in the UK.  I’ll blog more about that if/when I get it set up.

Many thanks to Duncan and Jonathan for the loan of the quadcopter, Bill (who still doesn’t know how imperilled his camera was) for the loan of the FLiR camera, and Nicole for enthusiastic assistance in the field.