Régis Gallon Garcia1,2, Mathieu Hélias1,2, Pascal Bailly du Bois1,2, Emmanuel Poizot1,2, Antoine Collin3
1Conservatoire National des Arts et Métiers, INTECHMER, France - 2 Normandie Université, France - 3Ecole Pratique des Hautes Etudes,France
Coastal zones, at the interface of the atmosphere, continent, and ocean, rank among the most productive and diverse ecosystems globally. These dynamic environments, shaped by numerous factors across spatial and temporal scales, support diverse habitats and organisms, including primary producers. Understanding the spatial distribution, biodiversity, community structure, and vulnerability of coastal habitats is vital for sustainable management under growing anthropogenic pressures.
Remote sensing has become indispensable for large-scale habitat mapping. Recent advancements in UAVs and aircrafts equipped with high-precision sensors, such as LiDAR and hyperspectral cameras, offer powerful alternatives to overcome limitations of traditional methods.
This study focuses on twelve sites along the Norman coastline, characterized by diverse geological substrates (granite, schist, calcareous rock). Field data and remote sensing outputs were used to characterize the vertical distribution and structure of macroalgal communities. Our innovant approach integrates the Tidal Height Normalization (THN) protocol to neutralize tidal effects, allowing precise analysis of altimetric distribution, geomorphological parameters derived from LiDAR data and spectral signatures from hyperspectral imagery to classify macroalgal belts. Geological information was also incorporated to account for substrate diversity. The methodology employs ensemble modeling to combine multiple predictive models, significantly enhancing accuracy and robustness.
This innovant and integrated approach provides a detailed and reliable representation of macroalgal communities, accounting for environmental variability across the Norman coastline.
Biography
Régis Gallon Garcia is a associate lecturer at Cnam-Intechmer, affiliated with the LUSAC (EA 4253, Normandie Univ.). His research focuses on the spatiotemporal dynamics of intertidal macroalgal communities. He combines field surveys with remote sensing data acquired using drones equipped with LiDAR, multispectral, and hyperspectral sensors. These datasets are integrated into machine learning models to predict the spatiotemporal distribution of macroalgae in the context of global change.