A model based on stacked-constraints approach for partitioning the light absorption coefficient of seawater into phytoplankton and non-phytoplankton components

Partitioning of the total non-water absorption coefficient of seawater, anw() (i.e., the light absorption coefficient after subtraction of pure water contribution), into phytoplankton, aph(), and non-phytoplankton, adg(), components is important in the areas of ocean optics, biology, and biogeochemistry. We propose a partitioning model based on stacked-constraints approach, which requires input of anw() at a minimum of six specific light wavelengths. Compared with existing models, our approach requires much less restrictive assumptions about the spectral slope of adg() and the spectral shape of aph(). Our model is based on several inequality constraints determined from an extensive, quality-verified set of field data covering oceanic and coastal waters from low to high-latitudes. With these constraints, the model first derives a wide range of speculative solutions for adg() and aph() and then identifies feasible solutions. Final model outputs include the optimal solution and a range of feasible solutions for adg() and aph(). The optimal solutions agree well with measurements. For example, the median ratio of the model-derived optimal solutions to measured adg() and aph() at 443nm is very close to 1, i.e., 1.004 and 0.988, respectively. The median absolute percent difference between the optimal solutions and measured values of adg(443) and aph(443) is 6.5% and 12%, respectively. The range of feasible solutions encompasses the measured adg() and aph() with a probability >90% at most wavelengths. Our results support the prospect for the applications of the partitioning model using the input data of anw() collected from various oceanographic and remote-sensing platforms.