Publications

Xing, XG; Morel, A; Claustre, H; D'Ortenzio, F; Poteau, A (2012). Combined processing and mutual interpretation of radiometry and fluorometry from autonomous profiling Bio-Argo floats: 2. Colored dissolved organic matter absorption retrieval. JOURNAL OF GEOPHYSICAL RESEARCH-OCEANS, 117, C04022.

Abstract
Eight autonomous profiling "Bio-Argo" floats were deployed offshore during about 2 years (2008-2010) in Pacific, Atlantic, and Mediterranean zones. They were equipped with miniaturized bio-optical sensors, namely a radiometer measuring within the upper layer the downward irradiance at 412, 490, and 555 nm, and two fluorometers for detection of chlorophyll-a (Chla) and colored dissolved organic matter (CDOM; profiles from 400 m to surface). A first study dealt with the interpretation of the Chla fluorescence signal in terms of concentration, using for this purpose the diffuse attenuation coefficient for irradiance at 490 nm, K-d(490), taken as a proxy for the Chla absorption. The present study examines the possibility of similarly using the K-d(412) values combined with retrieved Chla profiles to convert the CDOM fluorometric qualitative information into a CDOM absorption coefficient (a(y)). The rationale is to take advantage of the fact that K-d is more sensitive to CDOM presence at 412 nm than at 490 nm. A validation of this method is tested through its application to field data, collected from a ship over a wide range of trophic conditions (Biogeochemistry and Optics South Pacific Experiment (BIOSOPE) cruise); these data include both in situ fluorescence profiles and CDOM absorption as measured on discrete samples. In addition, near-surface a(y) values retrieved from the floats agree with those derivable from ocean color imagery (Moderate Resolution Imaging Spectroradiometer (MODIS-A)). The low sensitivity of commercially available CDOM fluorometers presently raises difficulties when applying this technique to open ocean waters. It was nevertheless possible to derive from the floats records meaningful time series of CDOM vertical distribution.

DOI:
0148-0227

ISSN:
10.1029/2011JC007632