Phytoplankton and iron: validation of a global three-dimensional ocean biogeochemical model

The JGOFS program and NASA ocean-color satellites have provided a wealth of data that can be used to test and validate models of ocean biogeochemistry. A coupled three-dimensional general circulation, biogeochemical, and radiative model of the global oceans was validated using these in situ data sources and satellite data sets. Biogeochemical processes in the model were determined from the influences of circulation and turbulence dynamics, irradiance availability, and the interactions among four phytoplankton functional groups (diatoms, chlorophytes, cyanobacteria, and coccolithophores) and four nutrients (nitrate, ammonium, silica, and dissolved iron).Annual mean log-transformed dissolved iron concentrations in the model were statistically positively correlated on basin scale with observations (P<0.05) over the eight (out of 12) major oceanographic basins where data were available. The model tended to overestimate in situ observations, except in the Antarctic where a large underestimate occurred. Inadequate scavenging and excessive remineralization and/or regeneration were possible reasons for the overestimation.Basin scale model chlorophyll seasonal distributions were positively correlated with SeaWiFS chlorophyll in each of the 12 oceanographic basins (P<0.05). The global mean difference was 3.9% (model higher than SeaWiFS).The four phytoplankton groups were initialized as homogeneous and equal distributions throughout the model domain. After 26 years of simulation, they arrived at reasonable distributions throughout the global oceans: diatoms predominated high latitudes, coastal, and equatorial upwelling areas, cyanobacteria predominated the mid-ocean gyres, and chlorophytes and coccolithophores represented transitional assemblages. Seasonal patterns exhibited a range of relative responses: from a seasonal succession in the North Atlantic with coccolithophores replacing diatoms as the dominant group in mid-summer, to successional patterns with cyanobacteria replacing diatoms in mid-summer in the central North Pacific. Diatoms were associated with regions where nutrient availability was high. Cyanobacteria predominated in quiescent regions with low nutrients.While the overall patterns of phytoplankton functional group distributions exhibited broad qualitative agreement with in situ data, quantitative comparisons were mixed. Three of the four phytoplankton groups exhibited statistically significant correspondence across basins. Diatoms did not. Some basins exhibited excellent correspondence, while most showed moderate agreement, with two functional groups in agreement with data and the other two in disagreement. The results are encouraging for a first attempt at simulating functional groups in a global coupled three-dimensional model but many issues remain.