Impact of improved light calculations on predicted phytoplankton growth and heating in an idealized upwelling-downwelling channel geometry

Ocean ecosystem models require accurate calculations of both hydrodynamics and biology; those calculations in turn require accurate calculation of in-water irradiance. Ecosystem models now achieve great accuracy in their hydrodynamical predictions, and the biological modules are becoming correspondingly sophisticated. The optical calculations are however often oversimplified, to the possible detriment of the physical and biological predictions. We used a recently developed, extremely fast radiative transfer code, EcoLight-S, to study differences in ecosystem and thermal development in an idealized upwelling-downwelling system when simple versus accurate irradiance calculations are used. The use of accurate irradiances gave up to 57% differences in chlorophyll concentrations after two weeks of simulated time, compared to predictions based on irradiances obtained using a simple exponential attenuation formula. Accurate irradiance calculations increased sea surface temperatures and decreased temperatures at depth, leading to increased stratification. Use of EcoLight-S couples the physical and biological calculations so that biology feeds back to physics, and vice versa. EcoLight-S outputs ancillary quantities such as remote sensing reflectance and in-water spectral irradiance, which can be used to validate ecosystem predictions using remotely sensed ocean color imagery or optical measurements from buoys or gliders, without the need to convert such measurements to chlorophyll values. After optimization, the ecosystem model total run times with EcoLight-S were less than 20% more than for the analytical irradiance models. We also found that the use of 24 h average irradiances gave factor-of-two differences in chlorophyll concentrations compared to the use of a diel irradiance pattern with the same 24 h average value.