Impact of Spatial Scale on the Performance of a Photoacclimation Model

Abstract Due to the photoacclimation of phytoplankton chlorophyll pigments, it is critical to know the ratio of carbon‐to‐chlorophyll ( θ ) if one wants to estimate phytoplankton carbon from chlorophyll or vice versa. Behrenfeld et al. (2016, https://doi.org/10.1038/nclimate2838 ) developed a widely used model to estimate θ based on solar radiation within the mixed layer, which reproduced θ very well for the areas defined in the study, but there is a knowledge gap regarding whether this model is applicable to any spatial scale. Here, we evaluated the performance of this model for areas of different sizes and found that the agreement between model‐calculated θ and satellite‐estimated θ is highly dependent on the size of the interested area where the agreement is very poor ( R 2 is as low as ∼0.05) if the size is small. As the size of an interested area increases, the agreement between model estimates and satellite retrievals improves significantly ( R 2 is as high as ∼0.80). This trend or pattern persists regardless of whether the area is constrained within the same level or mixed levels of chlorophyll variation. These results highlight that the applicability of this popular model of carbon‐to‐chlorophyll ratio should be better applied to oceanic areas matching those used in the model development and underscore the need to develop a more universally applicable model for the photoacclimation of phytoplankton chlorophyll in the upper ocean.