A powerful new way to test sound lures

We have created a new tool that allows us to test many different methods to lure and capture predators of any type, using sound and thermal vision. We are calling this tool the Cacophony Predator Lab.

In this blog article, we will discuss how sound lure experiments can be set up to test the effectiveness of various sounds for attracting possums. The sound lure software is configured from a web interface that allows you to upload any set of sounds and play them at any volume or time sequence.

Previous research has shown that sound lures perform better than food even if the sound is just a beep. This work took years because it involved large field trials and observation methods like chew cards and wax tags that are known to not be very sensitive (most possums just walk past them).

The Cacophony Predator Lab allows for much faster testing of lures because we have a more sensitive predator detector in the form of our thermal camera and a more effective trap so predators aren't allowed to learn to ignore lures (more on that below).

For the initial experiment we have 3 possum sounds that we think may be possible lures. We choose to play a different one every 10 minutes then observe the results. This test took place at a location where there had been about one possum observed every 4-5 days on average when no sound lure was in use. The results from a couple of nights of testing seem to indicate that possums are more likely to arrive after one of the sounds is played when compared to the other sounds.

Sound lure Date Time Animal Observed Last Sound Played Play-Detect Interval (mins)
No sound lures 29 May
29 May 00:23 Possum n/a n/a
29 May 00:23 Possum n/a n/a
30 May 21:34 Possum n/a n/a
31 May 06:04 Possum n/a n/a
7 June 00:07 Possum n/a n/a
12 June 23:21 Possum n/a n/a
1 July 21:32 Possum n/a n/a
2 July 00:32 Possum n/a n/a
2 July 02:37 Possum n/a n/a
7 July 03:00 Possum n/a n/a
9 July 23:05 Possum n/a n/a
3 possum sounds, every 10 minutes 12 July
14 July 19:28 Possum Possum sound 3 8
14 July 20:04 Possum Possum sound 2 4
14 July 21:35 Possum Possum sound 2 5
14 July 21:55 Possum Possum sound 3 5
14 July 22:04 Possum Possum sound 2 4
15 July 04:49 Possum Possum sound 1 9
Possum sound 2, every 20 minutes 15 July
16 July 21:53 Possum Possum sound 2 3
17 July 00:54 Possum Possum sound 2 4
No sound lures 19 July
Possum sound 2, every 20 minutes 22 July
24 July 23:29 Possum Possum sound 2 19
26 July 23:56 Possum Possum sound 2 6
29 July 23:27 Possum Possum sound 2 17
30 July 03:51 Possum Possum sound 2 1

The table below shows the number of days with and without sound lures playing along with the number of possum visits and then the rate of possum visits.

Days Visits Rate
No sound lure 45 11 0.24
Sound lure 15 12 0.80

The short summary of this is that with no sound lure the possums turned up on average 0.24 times per night but with the sound lure 0.8 times per night. Obviously, we would need a much larger set of results to verify that this is a statistically significant result but it does look promising.

The first part of the test seemed to show a preference for one of the possum sounds ("Possum sound 2" above) so we focused our testing on that sound after 15 July. The most promising sound was played every 20 minutes to see if the possum turns up just after the sound has been played (or not at all). Again, this looks very promising but we need a lot more data before we know if a sound lure if making a significant difference to animal behaviour.

It is very unlikely that one of the first three sounds that we tested so far happens to be the best lure for possums but the point of this article is to show how this new tool allows rapid testing of sound lures that can then be trialed on a much larger scale. It is likely that different lures will work best for males vs females, and juveniles at different times of the year - and that different lures will work best in different environments. The beauty of this system is that anything can be tested and shared so as soon as anyone finds a promising lure, it can be tested at scale.

Learning to ignore sound lures

One of the problems we learned from earlier testing of sound lures is that we suspect the local population of possums learned to ignore the sound. Part of what we need to effectively test sound lures is a trap with high removal rate so that a local population does not learn to ignore lures because of an empty reward once the sound is investigated. If animals are not removed then you may only be able to do one experiment per location, especially when the key metric is: how much time does it take for a possum to be detected with or without the sound lure.

Obviously, as possums are caught the detection rate will go down as there will be a lower density of possums in the immediate area. A potential way to counteract this could be to have a time period without sound lures after each possum caught to re-establish a baseline then try the playing sound lures again. In our initial test, two possums were caught during sound lure testing so the local population has been reduced. This means the higher detection rate when sounds are played could be even better than what the numbers suggest.

Lure volume

The louder the lure the more likely it is that a possum will hear it so in some ways louder is better. An observation we made that is obvious in hindsight is that a loud sound - played when a possum is near - may scare it off. We saw this during a couple of experiments but the good thing with our system is that we can both play lures at a loud volume to attract at a distance, and then turn down the volume or not play it at all when a predator is seen to be nearby by the thermal camera. Alternatively, we can switch to a different lure to get the possum to a trap.

Lure frequency

The goal of a sound lure is to get any predators in the area to investigate a trap. Playing a lure more often increases the chances that it will be heard by nearby predators. An interesting observation from our early experiments with possums is that they tend to turn up on camera 3-5 minutes after a sound lure has been played. This means that playing a sound every 5-10 minutes might be about right for luring possums that are wandering around. We can adjust this as we learn more about the time to catch after sounds lures are played. Again these tests are very early and the point of this blog is not the relevance of the data but to show that there is now a new tool to allow this sort of investigation.

Robust comparison

A future phase of testing could involve setting up the Cacophony Predator Lab in an area with a long, well documented history of trapping and then seeing how the catch rates are affected once sound lures are introduced into the mix. Ideally at the same time other commonly used possum traps can be set up away from the sound lure as a control. The positions of the classic traps and sound lure based traps would then swapped to see control for location differences.

We also plan to test a variety of sound lures and detection devices in more remote areas of bush to compare the effectiveness of various lures and detection approaches.


The Cacophony Predator Lab is a new tool that makes sound lure testing much faster and more flexible. Any combination of sounds, volume and schedule can be tested and the high sensitivity detection tool means small trials can show results much faster than previously. If you have any particular sound lure tests you want to run, please let us know and we will see if we can help.