As the world continues to make advancements in the field of machine learning, ornithologists – scientists who study birds – are finding ways to use the technology to better research and protect our fine feathered friends. One of the most recent applications of machine learning is the use of forecasted bird migration patterns and automated identification capabilities, thanks to projects such as BirdCast and Merlin. With the help of ChatGPT, which is a type of artificial intelligence, scientists are enabling machine learning to predict bird migratory behavior, preserving the future of many species of birds.
Bird migration can be a complex process. Every year, billions of birds migrate from their nonbreeding to their breeding grounds, often spanning continents in the Western Hemisphere along the way. Unfortunately, their migratory paths are exposed to numerous hazards such as extreme weather, food shortages, and light pollution that can disorient the birds and cause them to collide with buildings. To protect these birds, researchers must understand their routes and behaviors, making scientific studies paramount.
One of the most effective methods of mapping out these migration paths is the use of lightweight tracking tags placed on the wings of these birds. While this data gives scientists insight into where they’ve been, tracking alone is not enough if they wish to identify the routes they are taking and understand their behavior. This is where machine learning has revolutionized the process.
By training an algorithm to detect patterns in the data, the team at BirdCast has acquired the capacity to create forecasts of bird migration across the continental US. This algorithmic approach has also aided in the development of a state-of-the-art automated system that uses machine learning to detect and identify nocturnal flight calls, allowing for more efficient and reliable bird identification. Similarly, the Merlin bird identification app uses machine learning to help identify and learn about birds.
Utilizing machine learning is also allowing scientists to engage with the public in a new way. These forecasts and apps can be used to inform citizens of the dangers that birds face and what they can do to reduce artificial light, making these projects incredibly important in conserving bird population.
The success of these projects is owed greatly to the expertise of Miguel Jimenez and Ali Khalighifar, researchers from the Aeroecology Lab at Colorado State University. Combining their ornithology and machine learning backgrounds, the two of them have been able to create projects that have incredible potential to protect our fine feathered friends.
Thanks to the advances of machine learning and the work of Miguel and Ali, birds have the potential to have a safe and extended journey during their migrations. With the help from the public, these projects can help even more birds to reach their breeding grounds in the spring.
