Machine learning powers new tool to protect North Atlantic right whales

Data and AI leader SAS is helping protect endangered North Atlantic right whales through a pioneering collaboration with Fathom Science Inc. The partnership validates WhaleCast, an innovative whale prediction model that creates heatmaps showing the likelihood of whale activity along the East Coast. This technology integration allows vessels to reduce speeds in high-risk areas, potentially saving the critically endangered species while demonstrating how machine learning can transform marine conservation efforts.

The big picture: Fathom Science, a North Carolina State University tech spin-off building digital twins of the ocean, partnered with SAS to validate their whale location prediction model that helps vessels avoid collisions with endangered right whales.

How it works: WhaleCast combines historical whale sighting analysis with ocean modeling to create heatmap projections of probable right whale locations that integrate with vessels’ existing touchscreen systems.

• The system enables mariners to identify high-probability whale areas and reduce speeds accordingly, directly addressing a key threat to the endangered species.
• Taylor Shropshire, Head of Marine Resiliency at Fathom Science, created the model to empower timely and reliable marine conservation decision-making.

Tech implementation: SAS’ Data for Good program provided advanced validation through sophisticated machine learning methodologies using SAS Viya.

• The validation process required substantial data expansion, with SAS Data Maker generating synthetic data to create nearly 500,000 data points matching the characteristics of the original dataset.
• Seven different machine learning models were developed to test and validate WhaleCast’s approach, dividing the expanded dataset into groups for more robust training and testing.

Advanced capabilities: Beyond validating the basic whale location model, SAS helped solve a secondary problem by calculating the probability of whales’ distance from shore.

• SAS Viya Workbench, a standalone programming environment, enabled quick development of additional calculations that enhanced the model’s accuracy.
• “It was interesting to see SAS create multiple models very quickly,” noted Shropshire, highlighting how the collaboration expanded from simple models to complex neural networks.

Why it matters: This collaboration demonstrates how machine learning can transform conservation efforts by providing mariners with actionable intelligence that helps protect critically endangered marine species.