Improving the monitoring of submerged aquatic vegetation using satellite remote sensing using machine learning and object-based image analysis

Submerged Aquatic Vegetation (SAV) are important in aquatic ecosystems as they represent hotspots of biodiversity and ecosystem functioning. Earth observation technologies (EOT) constitute a cost-effective monitoring tool for environmental variables at large scales but their use in freshwater environmental monitoring remains limited because of the optical complexity of inland waters. This research develops and builds upon an innovative open-source machine learning and geographic object-based image analysis approach to characterize and map SAV. Associated field work performed near the Port of Montreal’s expansion site at Contrecoeur allows for research into how construction disturbances impact nutrients at the landscape level within the project and supports future SAV-related EOT developments. This research benefits the partner organization, Hatfield Consultants, in its work to develop EOT to support environmental variable monitoring in fluvial systems.

Faculty Supervisor:

Andrea Bertolo

Student:

Partner:

Hatfield Consultants

Discipline:

Physics

Sector:

Professional, scientific and technical services

University:

Université du Québec à Trois-Rivières

Program: