POSTER SESSION 3
15:30–17:30, Wednesday, September 16
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
ABSTRACT 1343 | POSTER 131
BIOMASS MAPPING OF AN INTERTIDAL EELGRASS BED IN THE ESTUARY AND GULF OF ST. LAWRENCE USING MULTISPECTRAL IMAGERY
Coastal vegetation, particularly eelgrass meadows, plays a fundamental role in supporting ecosystem functions and services, with biomass serving as a key indicator of ecological status. Remote sensing enables mapping of eelgrass meadows and assessing biomass distribution at larger spatial scales. This study aims to assess total aboveground biomass across major intertidal eelgrass meadows in the Estuary and Gulf of St. Lawrence (EGSL). We evaluate and adapt the method proposed by Hill et al. (2024) to quantify aboveground biomass (AGB) in intertidal eelgrass beds. In situ measurements of AGB, leaf area index (LAI), and multispectral imagery were collected across the EGSL. The methodology was first tested in Rimouski Bay using PlanetScope SuperDove and Sentinel-2 MSI imagery acquired at low and high tides and atmospherically corrected with ACOLITE. Eelgrass beds were mapped using a Support Vector Machine classification. For high-tide imagery, an inversion model retrieved bottom reflectance. Empirical relationships between reflectance and LAI were established. At high tide, the strongest relationship occurred in the green band (R² = 0.42; RMSE = 1.07), while at low tide, a stronger relationship was observed in the red band (R² = 0.60; RMSE = 0.40). Relationships between LAI and wet biomass, and between wet and dry biomass, were used to estimate carbon content. The low-tide biomass model explained a large proportion of variability (R² = 0.85) with low bias (11 g DWm–2), whereas the high-tide model performed less well, likely due to water column attenuation. These results highlight the potential of multispectral imagery for regional-scale eelgrass biomass and carbon stock estimation.
Brigitte Légaré*, Université du Québec à Rimouski (UQAR), [email protected], https://orcid.org/0000-0001-8746-0891
Simon Bélanger, Université du Québec à Rimouski, [email protected], https://orcid.org/0000-0002-9172-8376
Christian Nozais, Université du Québec à Rimouski, [email protected]
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
