Whether you think algorithms are cool or creepy, they are here to stay. For a team of mathematicians, computer scientists and marine biologists, the predictions derived from algorithmic calculations are helping to predict when narwhals hunt.
Narwhals, notwithstanding their unicorn-like tusks, are a mysterious species. They live in distant Arctic regions and hunt as deep as 1,000 meters down.
They orient themselves using echolocation, making clicking sounds to explore their surroundings. When they hunt, the clicking sounds turn into buzzing sounds as the interval times shorten.
We have very limited information about when and where they hunt. "In a situation where narwhals are in deep water, in the middle of the Bay of Baffin during December, we currently have no way of finding out where or when they are foraging," said cetacean researcher Mads Peter Heide-Jørgensen, a professor at the Greenland Institute of Natural Resources and adjunct professor at the University of Copenhagen.
The traditional methods of gathering and analysing information—using recording devices attached to the narwhal’s bodies—are very data-intensive, time-consuming and impossible to collect in many places.
Using artificial intelligence
Using data collected from five narwhals in Scoresby Sound fjord in East Greenland, the researchers sought to find out whether it was possible to use artificial intelligence to detect patterns in the way the whales moved and the buzzing sounds they emitted.
"The major challenge was that these whales have very complex movement patterns, which can be tough to analyze. This becomes possible only with the use of deep learning, which could learn to recognize both the various swimming patterns of whales as well as their buzzing sounds. The algorithm then discovered connections between the two," said Assistant Professor Raghavendra Selvan of the Department of Computer Science at the University of Copenhagen.
It was found that narwhals did not appear to make sudden changes in acceleration when foraging and that their hunting patterns in that respect might differ from other marine mammals.
The findings of the study was published in Volume 62 of the Ecological Informatics journal.
The researchers hope to follow up by characterizing different types of buzzing sounds, so as to identify the precise buzzing sounds that lead to a catch. This is done by giving the narwhals a temperature pill that detects temperature drops in their stomachs as they consume cold fish or squid.