New insights into how turtles navigate hundreds of miles to Ascension Island
Scientists are a step closer to understanding how green sea turtles navigate hundreds of miles of ocean to nest on the beaches of Ascension Island.
Every year between December and June, thousands of turtles migrate from Brazil to the tiny south Atlantic island to mate, with the females crawling ashore to lay their eggs in the sand.
A team at Heriot-Watt University in Edinburgh has developed a model that simulates part of the turtles’ journey, uncovering more about the information they rely on to make the arduous trip.
They found the species uses a range of cues, with several “weak” cues typically more successful than fewer strong ones, which could lead them off-course.
An estimated 3,000-5,000 green turtles nest on Ascension Island each year.
Professor Kevin Painter, a mathematical biologist at the university, said: “Almost 150 years after Charles Darwin highlighted the migration, we still don’t understand exactly how sea turtles navigate there.
“The remote location and the fact the turtles only make the journey every three or four years makes it difficult to conduct fieldwork.
“That’s why modelling can be such a powerful tool; it’s so difficult to study these navigations in the real world that a verified model can provide theories and more insight.”
He explained: “Our model revealed that multi-modal strategies, when turtles use a range of cues to navigate, generally improves homing.
“It’s not just one cue, such as a reaction to the magnetic field, an inbuilt compass or following an odour. More weak cues are typically more successful than fewer strong cues.
“This makes sense, as relying on a single navigating cue could render a population sensitive to change, whether natural or otherwise, and lead a turtle into ‘blind spots’.
“Understanding how fish, animals and birds navigate means we understand how to protect those that are endangered, and what is having an impact on their behaviour.”
Professor Painter’s research was published in Ecological Modelling.