Claustre¹, Hervé, Julia Uitz¹, André Morel¹, David Antoine¹ and Stanford Hooker²

¹Laboratoire d’Océanographie de Villefranche, CNRS / INSU & Université Pierre et Marie Curie, Quai de la Darse BP 8, 06 238 Villefranche sur mer, cedex, France, Tel: 33 4 93 76 37 29, E-mail: claustre@obs-vlfr.fr and ²NASA/GSFC/Code 970.2, Bldg 28 Room W126, Greenbelt, MD 20771

Combining satellite ocean color and the analysis of a HPLC pigment data base to infer phytoplankton (functional groups) distribution at the global scale.

The analysis of phytoplankton pigments by HPLC being a JGOFS core measurement, numerous pigment data have been collected during the various JGOFS in situ investigations of the last decade. In this study, these pigment data are gathered together with data from other sources into a data base which comprises ~ 2000 profiles from different contrasting oceanic areas (from subtropical gyres to upwellings). This database is analyzed with the aim of assessing the (expected) relationship between the size of the vegetal stock (chlorophyll concentration) and its composition [phytoplankton communities, depicted from pigment biomarkers, or phytoplankton size proxies, as proposed by Vidussi et al. (2001): JGR 106, 19,939-19,956]. We demonstrate that the concentration and depth distribution of pigment biomarkers and size proxies are related, in a statistical sense, to, and thus can be parameterized from, surface chlorophyll a concentration, [Chla_s]. The depth distribution of these biomarkers and size proxies is modelled from [Chla_s], following the approach of Morel and Berthon (1989, L&O, 34, 1545-1562) that was developed to infer the depth distribution of Chla concentration from [Chla_s]. These parameterizations, when combined with synoptical fields of [Chla_s] derived from satellite ocean color, allow to derive the fields of these biomarkers and size proxies. The usefulness of such fields towards the initialization and /or validation of biogeochemical and biooptical models is discussed.