We are working to make underwater acoustic data, vessel tracking data, and related software products more easily findable and improve their accessibility for researchers in Canada and around the world. Learn about our metadata standards, visit our data discovery portal, or explore the FishSounds portal.
We are creating visualization tools to support environmental management and decision making in the context of marine ecosystems. Use our Ocean Soundscape Atlas to explore how ship traffic creates underwater noise pollution or learn about our decision making tool that helps groups of people choose among a competing set of alternatives based on their individual preferences.
We are developing novel machine-learning algorithms to detect and classify calls made my marine mammals and fish. The algorithms are being used by researchers as part of passive acoustic monitoring systems, helping them protect endangered species such as the North Atlantic right whale. Try out our Python package Ketos, which provides tools that help you develop and train your own deep learning models at detecting and classifying underwater sounds.
We are developing a Python package named Kadlu that streamlines and automates the data pipeline for ocean ambient noise modelling. Learn more about Kadlu or try out one of several tutorials.
We are creating machine-learning tools for algae bloom detection using satellite imagery from commercial nanosatellites and long-running publicly available programs like Landsat and Sentinel. By combining the frequent return time of nanosatellites with the historical data and high spectral resolution of satellite programs run by the CSA, NASA and the ESA, we hope to combine the benefits of both systems to create tools that aid in freshwater and ocean monitoring, public health, and ecosystem protection. Learn about such tools and the significance of satellite data.