This story was originally posted by Waterloo News.

Waterloo researchers’ tool to be part of the European Space Agency's Climate Change Initiative: New machine learning model predicts lake ice conditions with 94 percent accuracy.

Researchers at the University of Waterloo, led by Climate Institute member Claude Duguay have developed new tools using machine learning models to identify lake ice conditions with unprecedented accuracy. These tools will significantly advance climate change monitoring and public safety. By processing satellite radar altimetry data, the algorithm can differentiate between open water, thin ice, growing ice, or melting ice with 94% accuracy.

The innovative algorithm will be implemented in the European Space Agency's (ESA) Climate Change Initiative to assess lake ice trends over time. It is crucial for public safety and recreation, as lake ice conditions impact transportation, food security, recreational activities, and tourism in Northern communities.

space

Sentinel-3, the satellite that acquired the data used in the study (ESA/ATG Medialab)

Master's student Jaya Sree Mugunthan and Professor Claude Duguay from the Department of Geography and Environmental Management emphasize the importance of higher resolution satellite data combined with state-of-the-art machine learning algorithms for these advancements. The researchers analyzed large Northern Hemisphere lakes across various conditions, demonstrating the algorithm's robustness.

As data becomes available, the algorithm can quickly report on ice conditions, benefiting the public. Future work will explore new machine learning approaches using data from recent satellite launches and apply the model to smaller lakes, such as those in Ontario's cottage country.

Dr Duguay

Dr. Duguay conducting research during a field visit in Inuvik.

This research is in keeping with Waterloo’s commitment to being a leader in sustainability research and education as the institution has been a catalyst for environmental innovation, solutions and talent for 50 years.