How SAS is Improving Whale Protection with Machine Learning Techniques
In a groundbreaking initiative, SAS, a recognized leader in data and artificial intelligence, has teamed up with Fathom Science Inc., a spin-off from North Carolina State University, to enhance ocean conservation efforts, particularly for the critically endangered North Atlantic right whales. By leveraging advanced machine learning algorithms, SAS is assisting in the validation of a state-of-the-art whale location prediction model, which aims to prevent ship strikes on these majestic marine mammals.
The Importance of Whale Protection
The North Atlantic right whale, one of the world's most endangered marine species, faces numerous threats, primarily from human activities such as ship strikes and fishing gear entanglements. With an estimated population of less than 350 individuals, effective protection measures are critical to their survival. The collaboration between SAS and Fathom Science focuses on harnessing technology to safeguard these whales from potential dangers posed by maritime operations.
Introducing WhaleCast
Under the guidance of Taylor Shropshire, the Head of Marine Resiliency at Fathom Science, the team developed WhaleCast, an innovative tool that combines historical whale sighting data with advanced ocean models. This heatmap projection identifies areas along the East Coast where the likelihood of right whale activity is heightened. The goal is to integrate this data directly into existing onboard navigation systems, empowering maritime operators to adjust their courses and speeds based on real-time whale activity insights.
Using Machine Learning for Validation
To validate the WhaleCast model, Shropshire employed robust statistical analysis techniques alongside machine learning methods. To perform this validation, SAS' Data for Good program provided essential support. Using SAS® Viya®, a powerful analytics platform, the team generated synthetic data that closely resembles real-world data. This synthetic dataset enabled more effective validation processes, ensuring that WhaleCast could predict whale locations with greater accuracy.
Armed with nearly 500,000 data points, SAS volunteers efficiently categorized the data for training and testing seven distinct machine learning models. This rigorous approach ensures that the WhaleCast model is both reliable and actionable, providing crucial data for maritime safety.
The Role of SAS Viya Workbench
In addition to WhaleCast's predictive capabilities, the SAS Viya Workbench played a pivotal role in further refining these models. The Workbench allowed volunteers to efficiently calculate the probability of whale proximity to shore, providing even deeper insights into potential risks. The ability to quickly program and adjust these calculations emphasized the flexibility and power of the SAS platform in marine conservation efforts.
A Message of Hope
Shropshire remarked on the rapid advancements achieved through SAS's machine learning models. The juxtaposition of simple models to intricate neural networks provided a comprehensive understanding of the benefits and limitations of each method. This blend of technology not only enhances the reliability of WhaleCast but also optimizes decision-making for maritime operators.
Conclusion
The collaborative effort between SAS and Fathom Science is a shining example of how technology can foster significant advancements in wildlife conservation. By integrating data analytics and machine learning into maritime operations, they are paving the way for more responsible practices that resonate with marine life preservation goals. The public can look forward to a future where technology and ecology work hand in hand to protect our oceans and their inhabitants.
For more information on how SAS is advancing marine conservation efforts, visit
SAS.