Accelerometer-based detection of African swine fever infection in wild boar.

Infectious wildlife diseases that circulate at the interface with domestic animals pose significant threats worldwide and require early detection and warning. Although animal tracking technologies are used to discern behavioural changes, they are rarely used to monitor wildlife diseases. Common disease-induced behavioural changes include reduced activity and lethargy ('sickness behaviour'). Here, we investigated whether accelerometer sensors could detect the onset of African swine fever (ASF), a viral infection that induces high mortality in suids for which no vaccine is currently available. Taking advantage of an experiment designed to test an oral ASF vaccine, we equipped 12 wild boars with an accelerometer tag and quantified how ASF affects their activity pattern and behavioural fingerprint, using overall dynamic body acceleration. Wild boars showed a daily reduction in activity of 10-20% from the healthy to the viremia phase. Using change point statistics and comparing healthy individuals living in semi-free and free-ranging conditions, we show how the onset of disease-induced sickness can be detected and how such early detection could work in natural settings. Timely detection of infection in animals is crucial for disease surveillance and control, and accelerometer technology on sentinel animals provides a viable complementary tool to existing disease management approaches.

A multi-species evaluation of digital wildlife monitoring using the Sigfox IoT network.

Bio-telemetry from small tags attached to animals is one of the principal methods for studying the ecology and behaviour of wildlife. The field has constantly evolved over the last 80 years as technological improvement enabled a diversity of sensors to be integrated into the tags (e.g., GPS, accelerometers, etc.). However, retrieving data from tags on free-ranging animals remains a challenge since satellite and GSM networks are relatively expensive and or power hungry. Recently a new class of low-power communication networks have been developed and deployed worldwide to connect the internet of things (IoT). Here, we evaluated one of these, the Sigfox IoT network, for the potential as a real-time multi-sensor data retrieval and tag commanding system for studying fauna across a diversity of species and ecosystems. We tracked 312 individuals across 30 species (from 25 g bats to 3 t elephants) with seven different device concepts, resulting in more than 177,742 successful transmissions. We found a maximum line of sight communication distance of 280 km (on a flying cape vulture [Gyps coprotheres]), which sets a new documented record for animal-borne digital data transmission using terrestrial infrastructure. The average transmission success rate amounted to 68.3% (SD 22.1) on flying species and 54.1% (SD 27.4) on terrestrial species. In addition to GPS data, we also collected and transmitted data products from accelerometers, barometers, and thermometers. Further, we assessed the performance of Sigfox Atlas Native, a low-power method for positional estimates based on radio signal strengths and found a median accuracy of 12.89 km (MAD 5.17) on animals. We found that robust real-time communication (median message delay of 1.49 s), the extremely small size of the tags (starting at 1.28 g without GPS), and the low power demands (as low as 5.8 µAh per transmitted byte) unlock new possibilities for ecological data collection and global animal observation.