Public called on to help decode whale song
Marine scientists have launched an appeal asking wildlife enthusiasts for help in decoding the secrets of whale song in a global crowdsourcing experiment.
Experts in the UK and North America are asking “citizen scientists” to study and sift through about 15 000 recordings of calls by pilot whales and killer whales around the planet, to see if new phrases, meanings and dialects can be uncovered.
The Whale Project, launched on Tuesday by Scientific American and the online citizen science organisation Zooniverse, is similar to the first major attempt to use crowdsourcing by amateur astronomers to help discover new galaxies by studying images taken by the Hubble space telescope in July 2007.
Participants visiting Whale.fm will be asked to study and then compare the sound wave patterns, or spectograms, of calls made by whales in different pods and families of whales around the world.
They will be asked to identify identical or very similar sound wave patterns, and will be able to play back each sound excerpt to help them match segments. Every sound recording is linked to a specific location in the sea, or geotagged, allowing scientists to precisely place clusters of calls in the areas where specific families of whale are known to inhabit.
Professor Ian Boyd, one of the project’s collaborators from the University of St Andrews’ sea mammal research unit, said scientists had discovered that people were often naturally much more able than computers to see similarities in complex spectograms.
“The first thing we want them to do is compare the images because what the human brain is very, very good at doing is comparing images, and is much better than a computer,” Boyd said. “For someone like me who’s tone deaf, who isn’t very good at telling sounds apart, we’re very, very good at making distinctions between small changes in shapes and objects.”
Boyd said pilot and killer whales had very complex calls or repertoires.
Marine scientists now wanted to investigate the differences in each group’s calls, like a dialect, and whether they could discover different kinds of messages from analysing these calls.
“If these animals have some form of linguistics or language tradition, we’re wanting to try to find the words within that repertoire of sounds. We don’t know what they mean but what we do find is they have different lexicons; different groups have different types of sound, and they probably inherit these sounds from their parents,” he said.
“It’s like a dialect. We want to be able to compare them; both these species have such complex sorts of sounds, and some of these sounds are repeated again, again and again. So they are not random.”
Every matched group of sounds would be compared with the whales’ location and activities that the whales were involved in. “We want to try and take that back to the context where they’re produced, such as hunting or social situations.”
Scientific American has previously run citizen science projects to track dragonfly swarms, the Gulf oil spill and a “great sunflower project”, recording their observations of the natural world. Mariette DiChristina, the editor-in-chief, said: “One doesn’t need a science degree to be a citizen scientist. All you need is a curiosity about the world around you and an interest in observing, measuring and reporting what you hear and see.”—