Hyperspectral and Lidar: Complementary Tools to Identify Benthic Features and Assess the Ecological Status of Sabellaria alveolata Reefs

Sabellaria alveolata is a sedentary gregarious tube-building species widely distributed from southwest Scotland to Morocco. This species builds what are currently considered the largest European biogenic reefs in the bay of Mont-Saint-Michel (France). As an ecosystem engineer, S. alveolata generates small to large scale topographic complexity, creating numerous spatial and trophic niches for other species to colonize. Sabellaria reefs are also under anthropogenic pressures, leading locally to massive degradation. However, stakeholders lack spatially explicit measures of reef ecological status, at adapted spatial resolution to provide key management information for this protected habitat. Traditional field surveys are extremely time-consuming and rely on expertise for visual ecological status assessment. The present study aims at using an automatic processing approach based on optical airborne data to (i) assess the potential of hyperspectral imagery to discriminate Sabellaria bioconstructions and its main ecosystem associated habitats, including different types of substrate as well as biological components and (ii) to use the combination of the hyperspectral and LiDAR signals to estimate the spatial structure of the different bioconstruction types (veneers vs hummocks and platforms) and ecological phases (retrograding and prograding). A reef from Mont-Saint-Michel was used as a test site. We built a processing chain based on supervised classification using Mahalanobis distance to generate an accurate distribution map (overall accuracy of 88% and a Kappa of 0.85) of ten Sabellaria-related benthic features, including large reef developing on sand and smaller veneers encrusting rocky shore areas. Specific spectral indices were used to define the spatial distribution of the main primary producers, in particular the microphytobenthos. Joining the hyperspectral and LiDAR data led characterizing the distribution of S. alvealata’s ecological status (prograding and retrograding phases) with an overall classification accuracy and Kappa coefficient that can respectively amount to up to 93% and 0.86. In our study site, the Sabellaria reef area (between 5.52 ha and 6.76 ha) was dominated by retrograding phases (between 53% and 58%). Our results showed that this automatic processing chain could be relevant for the spatial characterization of other Sabellaria reef sites. Study perspectives tend towards a quantitative estimation of their ecological status index.