The integration of bioacoustic indicators and artificial fear cues for the strategic management of kangaroo herbivory following fire and mining

Abstract

Western grey kangaroos (Macropus fuliginosus) have not previously been subject to tests for susceptibility to auditory based deterrents. This study presented a mob of western grey kangaroos with a series of treatments to determine behavioural responses towards artificial and biologically-significant acoustic deterrents. I observed and quantified nine common behaviours, including stablestate, feeding, vigilance, and alarm behaviour before, during and after presentation of each stimulus through three experiments.In the first experiment, four sounds were tested individually (each pre-recorded sound played once) to evaluate potentially effective deterring signals. Two sounds were artificial (an aerosol can hiss and a bullwhip crack) and two were natural (a kangaroo alarm footstomp and a raven call, a presumed benign control). Despite the promising literature on the potential for natural alarm stomps as non-invasive deterrents, artificial sounds were at least as evocative as bioacoustic sounds. A whip crack was more efficient deterring western grey kangaroos than their alarm stomp at generating flight and vigilant behaviours. Foraging behaviour was reduced from 71% to 1% within one minute of play back, whereas the foot stomp reduced feeding from 57% to 22%. An artificial hiss reduced feeding effort from 89% to 65% and the raven call dropped percentage of time spent foraging from 70% to 47% in the first minute post- treatment.In a second experiment, I tried to artificially enhance the rate of habituation to ascertain the likelihood of habituation to acoustic signals, by playing back the most fear-inducing sounds (the whip crack and the alarm foot stomp) at 5 s intervals for two full minutes. Feeding behaviour did not return to normal levels following either signal.In the third experiment, to evaluate if rate of applications has an effect, I then focused on the most aversive signal (whip) and varied the rate at 3 s, 15 s and 30 s intervals. More animals left the area overall, with the highest rate of stimulus, though not significantly so. There were no other effects of rate of playback on behaviour.The outcomes of these experiments suggest that artificial sounds may be at least as effective as bioacoustic sounds in generating alarm and flight among kangaroos while the rate of playback may influence a sound's ability to deter animals from a targeted area. Animals did not fully habituate to either signal during the timeframe of these experiments (three weeks; up to twice daily, but often less as dependent on weather; two minutes duration) despite my having replayed the signal repetitively at close intervals without reinforcing fear with any other effects.Despite the failures of commercial ultrasonic kangaroo deterrents, I was unable to find any reason that auditory deterrents cannot successfully form part of astrategic repellent program for the non-lethal management of kangaroos, ifmanaged appropriately. I have characterised and quantified changes in behaviourin response to several sounds, some that may be effective in fashioning anauditory based repellent. My results, mainly in regard to the whip crack, arepromising, but the overall efficacy of auditory based deterrents requires furtherresearch, especially in regard to rate of habituation, alternation of different signalsand to intensity, rate, frequency and duration of the signal.