The Internet of Animals:
How we can benefit from the collective wisdom of life
Abstract:
The collective wisdom of the Earth´s animals provides an immense bio-treasure of unprecedented information for humankind. Learning from animals in the ´Internet of Animals´ can help us predict natural catastrophes, forecast global zoonotic disease spreads or safeguard food resources while monitoring in situ every corner of the planet. The evolved senses of animals as well as technical sensors on animal-borne tracking tags enables local earth observations at highest spatial and temporal resolution. To protect and understand the ecosystem services provided by animals, we need to monitor individual animals seamlessly on a global scale. At the same time, these unprecedented life-history data of individual wild animals provide deep, novel insight into fundamental biological processes.
The ICARUS initiative, an international bottom-up, science-driven technology development of small, cheap and autonomous IoT (Internet of Things) sensing devices for animal movement and behavior is aiming towards this: wearables for wildlife. The resulting big data available in the open-source data base Movebank help understand, monitor, predict and protect life on our planet.
Bio:
Martin Wikelski is the Director of the Max Planck Institute of Animal Behavior (formerly Ornithology) in Radolfzell (Germany), Professor in Biology at the University of Konstanz and member of the German Academy of Sciences Leopoldina. Previously, he held positions at the University of Washington, Seattle, WA; Smithsonian Tropical Research Institute, Panama; University of Illinois Urbana-Champaign; Princeton University. His specialization is the study of global animal movement.
Summary
Focus:
Understanding ecosystems and the roles animals play
Leverage sensor data from the animals themselves
Are assembling datasets of the flow of animal across the planet and various environmental data around these animals
Internet of animals:
Animals can communicate with us using wearables
Have been placing wearable tags on animals since 1960’s, from bees to rhinos
124 GPS points per sqkm
~15000 animals feeding live data
Animal Tracker app: public access to a subset of the data
Ocean tracking also available, difference hardware/communication challenges
ICARUS system
Global satellite-based monitoring system
First: Antenna on the ISS
Currently developing: fleet of microsatellites
Tags are a few grams/centimeters, small enough for birds
Tracking local and migration patterns
Whole paths and intermediate stopping points
What can we do?
Change weather/climate predictions
Sensors can be taken in many places, measure temperature, wind, gases, etc.
Can use the choices that animals make for travel, feeding, breeding to infer what must be going on with the climate (can detect complex patterns, need correlation analysis to leverage data)
Detect disease spread
Animal health affected by different diseases
Measure animal temperature, motion (e.g. ear shaking), etc.
Key for ecosystem health, food safety
Natural catastrophes
Animals are often sensitive to earthquakes, tsunami
E.g. Banda Aceh earthquake: fewer casualties in region where people knew how to interpret animals’ reactions to earthquakes
E.g. Monitoring goat motion on Mt Etna, observing that their paths are predictive of eruptions (more activity -> eruption soon)
E.g. Monitoring farm animal behavior around earthquakes in Apenine area, Italy: higher activity before earthquakes
Leverage local knowledge to choose species that are sensitive to different phenomena
Ecosystem protection
Goal: protect endangered animals from hunting or snares that target other species
Placing tags on animals to track position/acceleration to capture behavior
Can detect when animals are caught in traps
Can detect when and where animals are killed/hunted
Less effective at preventing poaching of game animals
Hunters still hunt
More effective if placed on animals that are not hunted but respond to hunters (as opposed to tourists or rangers)
Tags can detect when cats start to hunt birds, can emit alarm sound to alert birds
Financialization of conservation
Animals across the planet sequester 6 GTons CO2e
Many other economic and societal benefits of ecosystems
The “Interspecies Bank” concept
Pay money to people who protect an ecosystem as long as they’re effective
Need daily verification that an animal is ok
Umbrella animals (e.g. large mammals) are best at indicating the overall health of the ecosystem
Don’t need a full understanding of the ecosystem to coarsely track health from observations of the key species
Ecosystem services
Straw-colored fruit bats: plant billion trees every night by dropping seeds from the fruit they eat
Musk Ox climate service: graze bushes, reduce their dark albedo and keeps the ground temperature colder, thus keeping methane in permafrost
MoveApps
Docker that puts together data analysis systems
Gets real-time alerts
Sensors
IOT networks enable us to collect data from sensors
On land there are many networks (e.g. cell towers)
City: narrowband IOT
Country: 200km (can use SigFox)
Remote areas: satellite
Can put multiple comms chips on same device
Oceans are still very challenging to
Energy:
Batteries, capacitors
Energy harvesting
Solar panels
Ground vs space communication (different comms protocols)
Cost dropping to $20
Shrinking to few grams (1-18g)
Larger tags can host a solar panel, $100/tag for prototypes, target is $30-40
Commercial: Tractive tags for dogs/cats, 1.5 million tags (https://tractive.com/)
Ways to apply tags
Guns that shoot nettles that stick to fur
Deer put their heads through hole to lick salt, get collared
In many cases, don’t need to capture animals
Ecosystem analysis:
Digital for nature: bring many sensors into a single system
Camera traps, audio sensors, environmental DNA (scoop up samples from streams), individual animals
Mystery: 30% of North American songbirds have disappeared over the past few decades. Need to understand when, where and how an animal dies. Monitor target animals and who else is out there.