Sound sensing sound fencing for humans and bison

A solution by Modulab submitted to Real-time wildlife monitoring and live behavioural analytics to reduce human wildlife conflict

We want to create an automated system that senses the presence of the bison approaching the human habitat and induces their return in nature by dispatching loud human civilisation-like noises. We will also build a complex library of bison sounds, being able to catalogue their behavior through sound. This is a novel and soft approach to mitigate the human wildlife conflict, with development perspectives.

(Pitched: 16/04/2018)

One Page Summary

To resolve the issue of damages caused by bison to human properties we propose to develop sound repellents placed around conflict hot spots in the bison rewilding area (orchards, fields, hay stacks and access ways to these properties). Acoustic monitoring will be used as the primary means of analysis of the effectiveness of acoustic repellents and impact on the communication and well-being of other species. We seek to test a wide variety of sounds (ultra sonic, audible, infrasonic, urban, antropogenic, abiotic) and playback patterns to identify if and what may be disturbing to the the bison. We will evaluate against indirect behavioral observations from camera trap footage placed on site.
Acoustic sensors as well as sound generators are small, increasingly affordable and non-invasive, and can be deployed in the field for extended periods of time to monitor wildlife and their acoustic surroundings. We will install sound recording equipment, work with experts to conduct baseline and ongoing acoustic monitoring and assemble/build custom sound repellents to deter bison from conflict areas.
The project will commence with an initial study and inventory of human wildlife conflict areas (historical as well as potential) and habitat use, based on existing data and support from the WWF team. This will provide the foundation of a deployment map for 10-12 repellents. Repellents will only be installed following comprehensive acoustic monitoring to estimate species occupancy, abundance, population density and community composition, monitoring spatial and temporal trends in animal behavior. After installation of the repellents acoustic proxies will be used to measure the effectiveness of the sound repellents.
Our initial research shows that bison do not like loud human noises. Few studies that are present are on the American bison, describe that behavior is changed when hearing loud noises like airplane or chopper overfly, drilling, and vehicles. It seems that human noise disturbance itself is already a driver to keep bison away. We expect that sounds like a chainsaw, gunshot, and metal banging would yield the same results. During the cold season when food availability is low, disturbance from logging intensifies in the bison home range. The damage in the area so far is debarking orchards and feeding on haystacks.
Studies of terrestrial mammals have shown that noise levels of 120 dBA can damage mammals' ears. Levels at 95 dBA can cause temporary loss of hearing sensitivity. Noise from aircraft has affected other large carnivores by causing changes in home ranges, foraging patterns, breeding behavior. Wild ungulates (American bison, caribou, bighorn sheep) appear to be much more sensitive to noise disturbance than domestic livestock (Weisenberger, et al. 1996). Behavioral reactions may be related to the past history of disturbances by such things as humans and aircraft. Noise, lights and disturbance caused by winter drilling has likely altered behavior patterns, particularly of females and younger individuals, who will avoid areas where winter drilling is ongoing. Ronald Lake herd individuals will sometimes avoid regions with noise disturbance produced during forestry activities (Mikisew Cree First Nation 2015).
One negative effect of the running and avoidance behavior is increased energy use. For a 90-kg animal, the calculated energy usage due to aircraft harassment is 64 calories/minute when running. It is 20 calories/minute when walking. When conditions are favorable, this expenditure can be counteracted with increased feeding. However, during harsh winter conditions, this may not be possible.
We will measure if aversive conditioning (where unpleasant stimuli are applied to unwanted behavior, such as using noises to scare bison away) is effective in deterring bison in the Tarcu Mountains rewilding context as well the impact, positive or negative on other wildlife. The acoustic indices method will be used to quantify global spectral and temporal characteristics of sound recordings in the habitat, once a bison has approached a repellent area. It will be possible to determine changes/reactions of fauna in the area focusing on the bison reaction. The method is useful because it enables quantitative analysis without the time-intensive process of extracting individual species calls, however they also have drawbacks such as sensitivity to non-biotic noise (Sueur et al, 2014), but we will look at the time-frames which exclude the moments when the sound repellents emit. We will evaluate the effectiveness of a repellent sound in location/climate condition by correlating behavioral observation from camera trap footage with spectral and temporal characteristics the effectiveness of a respective sound.