par Neukermans, Griet;Loisel, Hubert;Mériaux, Xavier;McKee, David;Astoreca, Rosa ;Doxaran, D.
Référence Ocean Optics XX : 20th Ocean Optics Conference 2010(27 September-1 October 2010: Anchorage, Alaska, USA), Proceedings of the XX Ocean Optics Conference, 27 September-1 October 2010,Anchorage, Alaska, USA
Publication Publié, 2010
Publication dans des actes
Résumé : We investigate the relationship between various scattering properties (total, back and side scattering) and Total Suspended Matter (TSM), the dry mass concentration of marine particles in suspension, for coastal and offshore waters in various regions (Coastal Atlantic, Southern North Sea, Mediterranean Sea and French Guyana waters). We quantify the uncertainty on TSM concentration estimation from each scattering property. We further quantify the variability of the total (bp), side (bs) and back scattering (bbp) to mass concentration ratio and explain this variability in terms of the physical (size) and chemical (apparent density, refractive index) properties of the particles. We show that bbp, bp and bs correlate well with TSM (correlation coefficients higher than 0.92), though with considerable scatter along the regression line. We find that 75% of the predicted TSM concentrations from a model based on bs are within 29% of the measured TSM concentration and within 38% and 48% for bbp and bp based models, respectively. The variability of mass specific total scattering (bp*) is mainly explained by total geometric cross section (PSA) and backscattering ratio ( bbp ~ ), with observations above (below) the regression line having significantly higher (lower) bbp ~ and lower (higher) PSA. Variability of mass specific backscattering (bbp*) was smaller and could not be explained by a single parameter. We do find that points below the regression line have significantly lower densities (ρa), PSA, refractive index (n), bbp ~ and chlorophyll a:TSM ratio than points near or above the line. After classification of observations in Case 1 and 2 water types, we find that Case 2 waters (N=149) show significantly higher values of bbp*, ρa, n, bbp ~ and PSA and significantly lower values for bp* and chla:TSM than for Case 1 waters (N=123).