First test of digital lures
We have started testing digital lures with a camera to observe the different rates of animal interactions. These tests are aimed at possums for the simple reason that our property is infested with them. Our initial impressions are:
- Sounds, particularly of other possum noises, seem to be quite effective lures
- Light seems to work as a lure as well
- Sound can be used to prompt a response, confirming a pest is in the area
- Existing traps seem to have very poor kill rates
- Lots more testing needs to be done
Each night the camera was placed in front of a trap or traps with different sound or audio lures played. The following day the number of animal interactions was recorded (note: not actual kills just the number of times an animal was caught on camera). An example of the data collected is shown in the table below.
Below are a few examples of the video footage we are getting.
Video 1: Possum investigating trap without kill (no kills yet with more than 40 possum interactions)
Video 2: Possums investigating flashing light
Video 3: First time we saw female and baby possum – attracted to possum sound
We also tried using a possum booming and the response from a possum down the gully started within a minute of playing the sound. This sound acted as a repellent as it was one of the only nights there were no possum visits. This is still pretty encouraging as this could be used to detect if there is a possum in a particular area.
The funny thing with this test was that two hours later, while our speaker was still going loud and strong, the real possum was still responding but sounding like it had a sore throat.
Why is any of this important?
The device we are developing in parallel will have the ability to listen for predators (possums, rats, stoats, etc.) and then play the most appropriate lure. The device will ultimately be able to identify and kill the predator with a much higher kill rate than existing traps. All of this will be solar powered and connected to the cloud so it can improve as we learn more effective methods. Rather than manual testing, all tests will be automatically measured so we learn from the different pests, during different seasons, and in different weather, the best way to attract and remove them.
We will post updates on our blog of the progress on the development of the Cacophonometer and effectiveness of various traps over the next few weeks.
How can you help?
Below is a table that shows how we want to progress on the digital lures. As you can see there is a lot we still don’t know but if you want to buy a trial camera and put it in front of your traps you can help us find out what does and doesn’t work. Seems a bit unfair that we are having all the fun… so feel free to join in and help bring the cacophony back.
|Listen to hear if they are out there||P||P||P|
|Sound call to get a response to confirm if they are there||V|
|Digital sound to draw into the trap area (caught on video)||E||P||P||E|
|Digital image to draw into trap area (video records them)||E||E|
|Sound/image to hold the pest for long enough for AI identification||E|
|Sound/image hold the pest in area enough for a identification and kill||P|
|Sound that can be used as a deterrent||V|
|Digital lures summary|
|P = Theoretically possible|
|E = some evidence it works (literature and/or our testing)|
|V = looks very useful and promising|