Water column structure and phytoplankton biomass profiles in the Gulf of Mexico

Remote sensors provide information on the average photosynthetic pigment concentrations for the upper 22% of the euphotic zone. To model primary production in the water column from satellite-derived photosynthetic pigments, estimates of the vertical distribution of pigment concentration are required. A Gaussian distribution function was used to represent the pigment vertical profiles with four parameters. We used 352 chlorophyll concentration (Chl(z)) profiles generated during oceanographic cruises to the Gulf of Mexico to estimate these parameters for seasons and regions within the gulf. Cluster analyses of data on surface temperature and chlorophyll, Chl(z) at the deer) chlorophyll maximum (DCM), and the depth of the DCM (Z(m)) were used to divide the Gulf of Mexico into three regions and the year into two seasons (cool and warm). Regression models were developed for each region and season to estimate Z(m) as a function of the mixed layer depth or an index of stratification. We were not able to generate useful regression models to predict the other three Gaussian parameters. Thus, representative means of these parameters were calculated for each region and season. Assuming a homogeneous biomass profile resulted in underestimation of integrated primary production by as much as 43% relative to the values derived from the nonhomogeneous profiles.