Big Data in Conservation
What do conservation, the Chinese black market, and sharks all have in common?
People don’t often put ‘conservation’ and ‘Big Data’ together as concepts – when you mention saving species, often they’ll imagine hiking through remote forests, tracking animals in all sorts of weather. To be fair, there is a good amount of that sort of work in ecology. But increasingly, conservationists around the world are relying on technology to mine Big Data for information, to give them the answers they need to identify threats and protect species.
Case in point: three months ago, the Fu Yuan Yu Leng 999 (a Chinese ship) was caught red handed by Ecuadorian authorities, with one of the largest illegal shark hauls in history – over 300 tonnes of fish. Over half this haul comprised of baby sharks and endangered species, all headed for the Chinese black market. The crew were all tried and sentenced for poaching, but as it turned out, this particular boat didn’t catch the fish. So who did?
Enter Big Data. By using Global Fishing Watch, a platform that uses artificial intelligence to collect and analyse ship satellite data, SkyTruth (a non-profit company) tracked the FYY Leng 999 from China to the Galapagos. They saw how it steadily steamed across the Pacific for a month, until suddenly stopping, alone. But not for long. Four additional ships cruised to the reefer, suggesting a cargo transfer from these potential poachers to the cargo ship FYY Leng 999.
Chinese authorities insisted these ships were rogue Taiwanese ships, travelling unnoticed by turning off their satellite data. But one remaining question lingered. Before the FYY Leng 999 was apprehended by the Ecuadorian authorities, where was the ship headed? By examining past tracked activity of the FYY Leng 999, SkyTruth discovered it had a history of dabbling with unregulated fishing fleets, including a cluster of squid ships. Lo and behold, this same squid fleet was spotted just beyond Ecuadorian territory, very close to a critical hammerhead breeding ground.
This is an example where big data, properly processed and analysed, had powerful implications for a critically endangered species. With such a huge quantity of data out there in the world, growing every day, it can be difficult to spot patterns and predict trends, so it’s becoming increasingly important to marry conservation with technology, to catch threats before they happen and find out the most effective ways to address and even prevent them.
There is so much potential information out there. There are citizen recordings like eBird, which relies on app users to record sightings and uses Machine Learning to give real-time migration predictions. There are interdisciplinary projects like the Earthcube project, which combines data from all kinds of scientists to monitor and project coral reef declines. The wealth of information can be plundered by combining ecological know-how with machine learning and artificial intelligence. The potential applications are unlimited, and have been used already to help combat the bee crisis, deforestation in the Amazon, and using Twitter to monitor environmental conditions. As data sets become open and knowledge is being shared more and more, using Big Data to solve conservation’s biggest problems has never been more feasible.
1,751 total views, 2 views today