The Impact of Warming Seas on Beluga Whale Aggregations
Introduction to Studying Beluga Whales
The aggregation behavior of beluga whales has historically been challenging to study due to technological limitations. However, with the rapid rise in sea surface temperatures caused by climate change, it has become imperative to develop methods that can accurately capture and analyze this data.
A Research Initiative Using AI and Satellite Imagery
In a groundbreaking study, a team of researchers from Spain’s Universities of Cadiz, Alicante, and Leon decided to leverage deep convolutional neural networks (CNNs) to analyze open-access satellite images from Arctic regions, including Canada, Russia, and the U.S. state of Alaska. Their goal was to identify beluga whale aggregations, correlate sea surface temperatures (SSTs) with these patterns, and use data from the Intergovernmental Panel on Climate Change (IPCC) to forecast three scenarios for the 21st century.
The Importance of Beluga Whales’ Habitat
Beluga whales (Delphinapterus leucas) are currently classified under the Red List conservation category as of 2017. While they are considered of Least Concern by the IUCN, these whales rely on the protection and sustenance provided by sea ice in the Arctic. The algae growth within the ice attracts the fish that make up the belugas’ primary food source, highlighting the critical importance of their habitat.
The Future of Arctic Ice and Beluga Whales
By the end of this century, it is projected that Arctic ice could disappear during summer months due to the effects of climate change. Traditional methods of studying cetaceans through direct-observation aerial surveys are becoming less effective and reliable in rapidly changing environments. The accuracy and longevity of species-distribution models (SDMs) derived from aerial surveys are increasingly limited.
Innovative Solutions Using Remote Sensing and AI
In response to these challenges, researchers have turned to remote sensing imagery and AI technologies to gather more accurate and comprehensive data on beluga whales. By training a deep CNN model with high-resolution satellite images, the research team was able to identify beluga whales in a dataset obtained between 2007 and 2020. This approach allowed for the expansion and enhancement of training data by manipulating and augmenting the images to improve accuracy and reliability.
Conclusion
The study demonstrates the effectiveness of combining AI image analysis with satellite imagery to better understand beluga whale aggregations and their relationship to changing sea temperatures. This innovative approach provides valuable insights into the impact of climate change on these majestic marine mammals, highlighting the urgent need for conservation efforts to protect their vulnerable population.