Estimation of Seasonal Primary Production From Moored Optical Sensors in the Sargasso Sea

A bio-optical mooring, which was deployed for 240 days during 1987 in the western Sargasso Sea (-35øN, 70øW), provided among other things a detailed record of the seasonal distribution of chlorophyll a fluorescence and the scalar irradiance of photosynthetically available radiation. This data base was augmented by shipboard measurements of chlorophyll a concentration, chlorophyll a fluorescence, scalar irradiance, and net photosynthetic rate during four seasonal cruises to the mooring. The combined data base clearly shows a spring bloom in the surface mixed layer associated with initial stabilization of the water column, formation of a subsurface chlorophyll maximum caused by enhanced stratification of the water column, and disappearance of this feature in the winter caused by deepening of the surface mixed layer. The moored and shipboard data were applied to a detailed description of the seasonal variability in the vertical distribution of chlorophyll a and photosynthetic rate. Specifically, photosynthetic rate at a given depth was calculated as the product of scalar irradiance, chlorophyll concentration, the specific absorption coefficient of the phytoplankton crop, and the quantum yield of photosynthesis. Values of scalar irradiance and chlorophyll a concentration were obtained directly from the moored sensors, values for the specific absorption coefficient of the phytoplankton were obtained by linear interpolation of shipboard measurements, and the quantum yield of photosynthesis was calculated as a function of scalar irradiance. Comparisons of photosynthetic rate calculated from such a formulation with measured rates of carbon assimilation indicated good agreement, with no apparent or easily explained biases in the predictions. Surprisingly, daily values in both the crop of phytoplankton and gross photosynthetic rate varied by only a factor of 2 during the year. The annual rate of gross photosynthetic production at the mooring was 12 tool C m-2 An analysis of errors suggests that the precision of this estimate is about ___20%.