It's been a while since a blog post and the work in this one was actually done in summer 2017. In this post we'll explore the testing of the quadcopter we performed in order to determine whether the drone is feasible in the field.
A little while ago we were asking if anyone was interested in collaborating in some field feasibility testing. Happily, we were contacted by a number of people and were able to work with the National Trust at their property - Scotney Castle, in Kent.
We want to give a big thanks at this point to Ross Wingfield who was the head ranger at Scotney and did an amazing job supporting us in a heap of different ways from driving us round the site to showing us his own drone sourced topography maps of the site. We couldn't have asked for more or have done it without him.
We initially visited the site in late June for an assessment of potential sites and whether the quadcopter or plane would be a better tool for surveying. We identified a number of sites, mostly along waterways, hedge and tree lines that we thought had good potential for bat commuting routes. It was also decided at this stage that the quadcopter would be a better tool for surveying due to the small survey sites, topography and the layout of the grounds.
The actual survey dates were set for late July and we planned on spending two nights on site. We decided on three survey routes and flew all of these during the day to ensure the quadcopter wasn't close to any trees or obstacles. With the waypoints set we were ready to get into the night time testing!
Aside from testing the quadcopter to see if we can make successful recordings we were also interested in whether the detector on the quad would record bats that aren't detectable on the ground. To test this we decided to run a simultaneous ground transect below the drone with a spare Peersonic detector identical to the one attached to the quad.
Here's the three survey routes we flew taken from the drone telemetry:
We got some really great recordings of Myotis bats. These were a real oppertunity to see how well the calls came out over the top of the noise of the drone. This shows that the noise from the drone is pretty low (visible as a horizontal yellow band in the figure below), and that calls in this frequency range are clearly visible over the top. Fab news!
This shows that the quadcopter can record bats with very little noise, good potential in the future we think.
The data collected from the trip to Scotney will form the base of a paper we're writing and hoping will be accepted for publication later this year. Looking forward to getting this out!