Borfecchia, F.; Micheli, C.; De Cecco, L.; Sannino, G.; Struglia, M.V.; Di Sarra, A.G.; Gomez, C.; Mattiazzo, G. Satellite Multi/Hyper Spectral HR Sensors for Mapping the Posidonia oceanica in South Mediterranean Islands. Sustainability2021, 13, 13715.
Borfecchia, F.; Micheli, C.; De Cecco, L.; Sannino, G.; Struglia, M.V.; Di Sarra, A.G.; Gomez, C.; Mattiazzo, G. Satellite Multi/Hyper Spectral HR Sensors for Mapping the Posidonia oceanica in South Mediterranean Islands. Sustainability 2021, 13, 13715.
The Mediterranean basin is a hot spot of climate change where the Posidonia oceanica (L.) Delile (PO) and other seagrass are under stress due to its effect on marine habitats and the rising influence of anthropogenic activities (tourism, fishery). The PO and seabed ecosystems, in the coastal environments of Pantelleria and Lampedusa, suffer additional growing impacts from tourism in synergy with specific stress factors due to increasing vessel traffic for supplying potable water, fossil fuels for electrical power generation. Earth Observation (EO) data, provided by high resolution (HR) multi/hyperspectral operative satellite sensors of the last generation (i.e. Sentinel 2 MSI and PRISMA) have been successfully tested, using innovative calibration and sea truth collecting methods, for monitoring and mapping of PO meadows under stress, in the coastal waters of these islands, located in the Sicily Channel, to better support the sustainable management of these vulnerable ecosystems. The area of interest in Pantelleria was where the first prototype of the Italian Inertial Sea Wave Energy Converter (ISWEC) for renewable energy production was installed in 2015, and sea truth campaigns on the PO meadows were conducted. The PO of Lampedusa coastal areas, impacted by ship traffic linked to the previous factors and tropicalization effects of Italy southernmost climate change transitional zone, was mapped through a multi/hyper spectral EO-based approach, using training/testing data provided by side scan sonar data, previously acquired. Some advanced machine learning algorithms (MLA) were successfully evaluated with different supervised regression/classification models to map seabed and PO meadow classes and related Leaf Area Index (LAI) distributions in the areas of interest, using multi/hyperspectral data atmospherically corrected via different advanced approaches.
Posidonia oceanica (PO); LAI & density; PO health & Pergent model; sea truth sampling; Earth Observation; HR satellite multispectral/hyperspectral sensors; atmospheric correction; coastal monitoring; mapping shallow waters habitat seabed; Calibration/validation & training/test; Classification & regression Machine Learning; Model Performance & thematic Accuracy; Sentinel 2 MSI multispectral & PRISMA hyperspectral; ISWEC(Inertial Sea Wave Energy Converter)
EARTH SCIENCES, Environmental Sciences
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.