Marine vegetation mapping in Sweden

In 2020 we embarked on a project together with the Swedish regional and national authorities, to combine Copernicus Sentinel-2 satellite imagery, novel machine learning techniques and advanced data processing to create the first spatial overview of the distribution of SAV at national scale in Sweden. A training dataset constructed with more than 30 000 manually labelled polygons was used to build the classification model, with which more than 3800 km2 of the underwater habitats, of the entire shallow coastal zone of Sweden, was mapped. Applying Sentinel-2 imagery from 2019-2020, the resulting classification, in 10-meter spatial resolution, provides nationwide coverage and detailed insight into the status of aquatic vegetation in the shallow water areas of Sweden.

Moreover, in close coordination with the key stakeholders, we have developed a cloud-based web-application that enables individual counties along the Swedish coast to perform their own SAV mapping without prior specialist knowledge. This ties into a long-term objective of the counties in Sweden to easier comply with national monitoring and reporting requirements, and thereby for Sweden to better comply with international requirements at EU level.

The project outcome and methodologies are further described in the following peer-reviewed article:

Huber, Silvia. et al. (2021), Novel approach to large-scale monitoring of submerged aquatic vegetation: A nationwide example from Sweden. Integrated Environmental Assessment and Management.