Morel, A, Huot, Y, Gentili, B, Werdell, PJ, Hooker, SB, Franz, BA (2007). Examining the consistency of products derived from various ocean color sensors in open ocean (Case 1) waters in the perspective of a multi-sensor approach. REMOTE SENSING OF ENVIRONMENT, 111(1), 69-88.
During its lifetime, a space-borne ocean color sensor provides world-wide information about important biogeochemical properties of the upper ocean every 2 to 4 days in cloudless regions. Merging simultaneous or complementary data from such sensors to obtain better spatial and temporal coverage is a recurring objective, but it can only be reached if the consistency of the sensor-specific products, as delivered by the various Space Agencies, has first been carefully examined. The goal of the present study is to provide a procedure for establishing a coherency of open ocean (Case-1 waters) data products, for which the various data processing methods are sufficiently similar. The development of the procedure includes a detailed comparison of the marine algorithms used (after atmospheric corrections) by space agencies for the production of standard products, such as the chlorophyll concentration, [Ch1], and the diffuse attenuation coefficient, K-d. The MODIS-Aqua, SeaWiFS and MERIS [Ch1] products agree over a wide range, between similar to 0.1 and 3 mg m(-3), whereas increasing divergences occur for oligotrophic waters ([Ch1] (from 0.02 to 0.09 mg m(-3)). For the K-d(490) coefficient, different algorithms are in use, with differing results. Based on a semi-analytical reflectance model and hyperspectral approach, the present work proposes a harmonization of the algorithms allowing the products of the various sensors to be comparable, and ultimately, meaningfully merged (the merging procedures themselves are not examined). Additional potential products, obtained by using [Ch1] as an intermediate tool, are also examined and proposed. These products include the thickness of the layer heated by the sun, the depth of the euphotic zone, and the Secchi disk depth. The physical limitations in the predictive skill of such downward extrapolations, made from information concerning only the upper layer, are stressed. (c) 2007 Elsevier Inc. All rights reserved.