It is high-time conservation bodies started adopting high-tech strategies in their battles to protect animals. While some have been slow to the uptake, a lot of groups have seen the potential for new monitoring tools. With the help of groups such as Google more and more bodies are embracing the devices as a way to take on poaching head on. The WWF estimates illegal wildlife trade is worth around £16 billion a year and is one of the main factors of the devastating decline in some species. The Living Planet Index maintained by ZSL and WWF shows that 60% fewer vertebrates (mammals, reptiles, fish, etc) live in the wild now that half a century ago. The steepest drop has been recorded in the tropics.
While some falls in populations of species such as elephants and black rhinos has been stopped and, in some cases, even reversed through intensive conservation efforts, poachers are still killing them. Numbers for other species such as many monkeys are declining fast, however.
Poachers aren’t solely responsible for the decimation of biodiversity, with overfishing, urbanisation and climate change also playing key parts. However, for some species illegal hunting is the biggest factor in the decline, according to Andrew Terry, head of conservation at ZSL. “We have a particular focus on tackling the international wildlife trade but that is embedded in our broader conservation efforts,” says Mr Terry.
Zoologists have resorted to camera traps to photograph animals for decades but until recently they didn’t have wireless connections. Operators would have to physically visit them and remove film or later SD cards which were often full of useless images of moving branches or other wildlife that would have triggered them. Zoologists have used camera traps to photograph passing animals for decades but until recently these had no wireless connection, so their operators had to physically visit each one to remove its film and later its electronic SD card, which was often full of useless images of moving branches or other wildlife that had triggered the trap.
According to the director of wildlife and biodiversity at Resolve, Eric Dinerstein it was odd that conservationists were slow to take up technology. He went on to say “We jumped in around six years ago because we saw an opportunity to make a difference with a camera trap with intelligence and connectivity.”
Various other conservation bodies began to develop detection technology about the same time. They started working with tech companies that view wildlife protection as a showcase of their expertise. Systems like Resolve’s TrailGuard are now in the final stages of testing and will be soon be ready for field deployment.
Sam Seccombe Instant Detect project manager says “Conservation organisations don’t generally have the resources to recruit and employ expensive software engineers and developers, so we depend on collaboration with the tech industry.” For example, Resolve is working with Microsoft, Inmarsat, and Intel, while ZSL’s partners include Iridium and Google.
Google’s AutoML system, that enables people with limited technical experience to develop AI for purposes such as image recognition, is being deployed in Instant Detect. It is making it possible to recognise animals or people instantly from trap pictures.
“A successful business relies on being able to collect, analyse and interpret data rapidly to make the best business decisions,” Mr Seccombe says. He goes on to say that the same is true for conservationists and camera trap data. “By increasing the speed of image analysis, conservation impact can be made more quickly and be more effective,” he adds.
Mobile phone coverage is likely non-existent in remote wildlife parks so Instant Detect uses its own radio transmitters to send images to a based station and then on by satellite to headquarters. The first version of the system was trialled by ZSL when monitoring Australian night parrots, Kenyan elephants and rhinos, Antarctic penguins, and Canadian bears. It was found to suffer transmission problems in dense foliage.
A more reliable second version was developed which has had successful initial tests in Africa. Camera quality however was an issue, with nothing on the market meeting ZSL’s specifications. The group had to find a way to develop its own 5-megapixel Instant Detect camera that had a range of focal lengths and was triggered by either an external metal detector for poachers or an inbuilt infrared sensor to detect motion and heat of passing animals. My. Seccombe says “It seems ironic that most trail cameras being used by conservationists have been designed for the deer hunting market.”
Resolve’s TrailGuard incorporates Intel vision-processing chips in its cameras. It is also able to carry out AI image analysis so that only images of human intruders are transmitted. This helps with battery life and cutting down transmission costs. The first version of it operated in the Grumeti reserve in Tanzania in 2018. It managed to detect more than 50 intruders and enabled 30 arrests from 20 different poaching gangs. Mr Dinerseting saus that Resolve is manufacturing 1,000 updated units for installation in Africa and elsewhere.
The US foundation has proposed to protect over 100 wildlife parks and reserves over the next two years. They aim to do so by installing TrailGuards on the 10 trails that are most targeted by poachers in each place. The cost of installation would cost a park roughly £13,000 in the first year and slightly more in the second year, with operating expenses for data transmissions estimated to cost in the region of £150 a year. This is far less than alternative protection methods which include flying drones or employing additional staff.
Anthony Dancer, who manages ZSL’s tech programme, warns that new technology cannot stop illegal killing on its own. “Most protected sites around the world are terribly underfunded,” he says. “Even if we make the technology available, many places will not have enough resources to manage the technology or enough rangers for a large increase in enforcement.”
Aside from poaching for horns, teeth, meat or fur, etc people also kill animals to protect their crops or livestock. Resolve plans to take on this growing conservation problem by adapting its TrailGuard features and hardware. Their aim is to adapt it to identify animals instead of people, for a project called VillageGuard.
Camera traps would be installed along trails used by big animals that trample or eat crops or attack livestock. The initial five targets would be snow leopards and wolves in Nepal, grizzly bears in the U.S.A and lions and elephants in Africa. The system will have attached speakers that will frighten away unwanted animals by playing sounds such as human shouting.
Conservation bodies are also looking further than threats to wildlife from villagers or poachers, by resorting to technology to track elusive animals. They analyse the increasing volume of images from camera traps installed around the world. ZSL for examples uses both human volunteers and machine learning to do so. AI is currently being used in several projects to identify animals.
The biggest programme, Wildlife Insights, sits within Google Cloud and combines its machine learning expertise with a group of conservation groups. It has been taught to recognize 612 species with 8.7 million images supplied by member organisation. It is expected to expand quickly as conservations feed it more data.
Accuracy ranges from 80-98%, but this depends very much on the distinctiveness of the species as well as the quality of the image. “Wildlife Insights is essentially a massive open source system that will enable people around the world to manage and analyse biodiversity data,” says Mr Dancer.
As much as AI becomes a more powerful tool, humans will always be pivotal in wildlife identification, this includes experts zoologists and amateurs as well. ZSL’s free citizen science app, Instant Wild allows anyone with a smartphone to identify animals in camera images. So far it has been downloaded 130,000 times.
Technology is also helping the ones on the front line who work in the world’s reserves and parks (300,000-400,000 rangers and wardens), according to the International Ranger Federation. A system called Smart (Spatial Monitoring and Reporting Tool), developed by ZSL and other bodies allows rangers to collect and sort out data on mobile devices about locations of animals and humans. This includes illegal intruders, which allows the deployment of limited staff as efficiently as possible.
The system is already being used in 900 protected areas globally. Harvard computer science professor Milind Tambe has developed AI software that will be integrated into it next year. It will predict poacher behaviour so patrols can be directed to probable hotspots of illicit activity.
With wildlife under enormous pressure, technology has huge potential to enable conservationists to deploy limited resources more efficiently and battle poachers and the illegal trade. Mr. Dancer of ZSL says“We urgently need to innovate, and to create new partnerships with industry, governments and academia, to develop new solutions. This is where technology and tech partnerships have the potential to be transformational — by enabling us to target our resources more efficiently and more effectively, and to scale our impact.”