Spectral Angle Minimization for the Retrieval of Optically Active Seawater Constituents from MODIS Data

Evaluation of empirical and semi-analytical chlorophyll algorithms in the Ligurian and North Tyrrhenian Seas

“…The SAM LT algorithm, recently proposed by Maselli et al, 36 is based on the simulation of remote sensing reflectance, Rrs sim through the following equation: The algorithm simulates a wide range of reflectances by varying the concentrations of the three optically active constituents within Eq. (2).…”

“…In this way, the algorithm is insensitive to amplitude variations of the measured reflectances, which may be due to the presence of seawater constituents with variable spectral properties and/or to inaccurate atmospheric correction of the satellite data. 36 Since all specific coefficients of absorption and backscattering coefficients used in Eq. (2) are obtained from a bio-optical survey of the Ligurian and North Tyrrhenian Seas, the algorithm has intrinsically a local validity.…”

“…In this case, [CHL] is particularly overestimated for the CTD13 and CTD22 samples, both belonging to the BP09 cruise. Table 4 summarizes the results obtained by applying the optical classification rules of Lee and Hu, 21 and Maselli et al, 36 to the samples taken during the three cruises. The mean chlorophyll concentrations measured and estimated by the four algorithms in Case 1 and 2 waters are summarized in the bar plot of Fig.…”

“…5. The Lee and Hu criterion identifies 15 Case 1 water 36 to the data of the three cruises (total number of stations: 33-BP09, 9-MOMAR, 21-REP10). Lee ).…”

“…25,31,32 OC5 algorithm, which, although more specific for Atlantic waters (Bay of Biscay and the English Channel), performs well also in other areas. [33][34][35] The semi-analytical SAM LT algorithm 36 is locally tuned for the coastal Ligurian and North Tyrrhenian Sea areas.…”