POSTER SESSION 1
15:30–17:30, Monday, September 14
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
ABSTRACT 1257 | POSTER 49
INTEGRATING SPECTRAL ANGLE MAPPER INTO OPTICAL WATER CLASSIFICATION FOR ENHANCED ENVIRONMENTAL MONITORING: A CASE STUDY ON THE COASTAL WATERS OFF SAGRES (SOUTHERN PORTUGAL)
Water classification is a major tool employed for simplifying the interpretation of water optical properties. However, the well-known Case 1/Case 2 classification is relatively coarse, potentially failing to capture the spectral variability of water coastal areas. This work presents a preliminary investigation on the spatio-temporal variability of optical water types off Sagres (SW Portugal), an area under the influence of the western coastal upwelling and where phytoplankton bloom events occur, by analyzing the spectral shape variations of remote sensing reflectance (Rrs).
To address this limitation, we employed a continuous classification approach based on the Spectral Angle Mapper (SAM) distance. SAM distance was calculated for each sample versus both a local reference Case 1 spectrum and standardized Optical Water Types. The methodology integrates an in situ hyperspectral dataset from a TACCS radiometer and a multispectral satellite dataset from MERIS (2008–2012), both acquired under MERIS validation campaigns in SW Portuguese coasts, spanning a three-station transect from coastal to oceanic waters off Sagres.
Preliminary results indicate that SAM distance varies systematically with distance from the coast, reflecting the expected gradient from coastal optically complex waters to oceanic waters. Temporal variability in the SAM signal appears to coincide with bloom events, suggesting sensitivity to episodic biogeochemical changes. By shifting from discrete classes to a continuous spectral metric, this approach enhances the sensitivity of environmental monitoring in coastal ecosystems, beyond simple concentration-based proxies. This continuous metric offers a robust framework for assessing water classification, providing a more physically-grounded characterization of optically complex coastal ecosystems.
Cristiano Ciccarelli, Università Parthenope di Napoli, [email protected], https://orcid.org/0009-0006-1502-324X
Massimiliano Lega, Università Parthenope di Napoli, [email protected]
Sónia Cristina, CIMA – Centro de Investigaçao Marinha e Ambiental/ARNET – Aquatic Research Network, [email protected]
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
