{"id":18285,"date":"2017-12-05T12:00:00","date_gmt":"2017-12-05T12:00:00","guid":{"rendered":"https:\/\/dev.inrs.ca\/new-tool-for-the-early-detection-of-blue-green-algae\/"},"modified":"2020-11-10T12:10:03","modified_gmt":"2020-11-10T17:10:03","slug":"new-tool-for-the-early-detection-of-blue-green-algae","status":"publish","type":"post","link":"https:\/\/dev.inrs.ca\/en\/news\/new-tool-for-the-early-detection-of-blue-green-algae\/","title":{"rendered":"New tool for the early detection of blue-green algae"},"content":{"rendered":"\n

Algae and\/or cyanobacteria blooms are increasingly invading Quebec\u2019s waters, with some 500 bodies of water affected in the past ten years. Methods for tracking these microorganisms are costly and labour-intensive, but satellite imaging has shown real promise as a way to monitor their development. However, detecting blooms in fresh water the moment they appear is a challenge. A research team led by INRS professor\u00a0Karem Chokmani<\/a>\u00a0is working to overcome this challenge by developing an approach for monitoring algae blooms near the shores of freshwater lakes that pose the biggest threat of cyanobacteria contamination for the population.<\/strong><\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

Satellite and drone imaging<\/strong><\/h2>\n\n\n\n

The researchers will combine data from the Sentinel-2 satellite with the hyperspectral data acquired using INRS drones. They will also develop a methodology based on MODIS data and transfer it to the higher spatial resolution data captured by the Sentinal-2 satellite. This approach, based on ensemble methods applied to MODIS data, makes it possible to detect and track small algae blooms even in low concentrations. However, given the spatial resolution quality of MODIS data, it was impossible to monitor the status and evolution of the blooms at the edges of lakes.\u00a0<\/p>\n\n\n\n

An approach with multiple benefits<\/strong><\/h2>\n\n\n\n

This approach builds on existing methods to deliver a number of advantages:\u00a0<\/p>\n\n\n\n