An assessment of optical properties and primary production derived from remote sensing in the Southern Ocean (SO GasEx)

Optical properties and primary production were measured during the Southern Ocean (SO) Gas Exchange Experiment (GasEx) (March-April 2008). To assess and evaluate these properties derived from remote sensing, absorption coefficients derived from remote sensing reflectance (R-rs) with the quasi-analytical algorithm were compared with those from in situ measurements from both an ac-9 optical instrument deployed on a profiling package and from discrete water samples analyzed using filter pad spectrophotometry. Total absorption coefficients from R-rs retrievals were found, on average, to be similar to 12% less than ac-9 measurements and similar to 15% less than filter pad measurements. Absorption coefficients of gelbstoff-detritus and phytoplankton pigments (at 443 nm) derived from R-rs were similar to 15% and similar to 25% less than ac-9 measurements, respectively. The difference can be well explained based on the determination methods and these results indicate general consistency between remote sensing retrievals and in situ measurements for these waters. Further, incorporating measured surface radiation data, water column primary production (PPeu) was estimated using chlorophyll concentration based models (Chl-PP) and a phytoplankton absorption based model (Aph-PP), where remote-sensing Chl was retrieved with an operational empirical algorithm. These estimated PPeu values were then compared with primary productivity measured using C-14 incubation techniques, and coefficient of determination (R-2, N = 13) of 0.74 were found for the Aph-PP results, while the R-2 of the Chl-PP results were less than 0.5. Such a contrast further highlights the importance of analytically retrieving phytoplankton absorption from measurement of ocean color and the advantage of using phytoplankton absorption to represent the role of phytoplankton in photosynthesis. Spatial distribution and contrast of PPeu in the greater SO GasEx region estimated from satellite data are also presented.