Appendix Q — Ocean: Bio-Optical Model
Implementation status
| Model | DITL | PACE V3 | PACE V4 | Evaluation | Planned |
|---|---|---|---|---|---|
| Bio-1 | x | x | x | — | — |
| Bio-2/3 | — | — | — | x | — |
Q.1 Overview
FastMAPOL retrieves ocean optical properties by coupling the aerosol retrieval with a forward radiative transfer model of ocean inherent optical properties (IOPs). The ocean optical properties are parameterized using bio-optical models that describe the spectral absorption and scattering of seawater constituents.
Three parameterizations are considered:
| Model | Parameters | Application |
|---|---|---|
| Bio-1 | 1 parameter | Open ocean (Case-1 waters) |
| Bio-2 | 5 parameters | Coastal and optically complex waters |
| Bio-3 | 3 parameters | Optimized coastal parameterization |
These models describe the spectral absorption and scattering properties of four main water constituents:
- pure seawater
- phytoplankton
- non-algal particles (NAP)
- colored dissolved organic matter (CDOM)
The total inherent optical properties are constructed from the contributions of these components. Bio-1 is used in default for the current released data product.
Q.2 Bio-1 Optical Model (Open Ocean)
The Bio-1 model is a bio-optical parameterization designed for open ocean waters (Case-1 waters) (Zhai et al. 2015, 2017). The model can be also derived from the generalized Bio-2 model (Gao et al. 2018) by imposing constraints on its parameters (Gao et al. 2019)
In this model, the optical properties are parameterized primarily as a function of chlorophyll concentration \([Chl\,a]\).
Q.2.1 Phytoplankton Absorption
The phytoplankton absorption coefficient \(a_{ph}\) follows the formulation defined in the generalized model (see Section 6.3).
Q.2.2 Particulate Absorption
For open ocean waters, the contribution of non-algal particles (NAP) is assumed negligible. Therefore the combined detrital and CDOM absorption coefficient \(a_{dg}\) depends only on phytoplankton.
The value at 440 nm is parameterized as
\[ a_{dg}(440) = p_2\, a_p(440,[Chl\,a]) \]
where
\[ p_2 = 0.3 + \frac{5.7\,R_2\,a_p(440,[Chl\,a])}{0.02 + a_p(440,[Chl\,a])}. \]
The parameter \(R_2\) is assumed to be
\[ R_2 = 0.5. \]
The exponential spectral slope is fixed as
\[ S_{dg} = 0.018. \]
Q.2.3 Particulate Backscattering
The particulate backscattering coefficient \(b_{bp}\) is also assumed to depend only on phytoplankton in open waters.
The magnitude at 660 nm is parameterized as
\[ b_{bp}(660) = 0.347\,[Chl\,a]^{0.766}. \]
The spectral slope of particulate backscattering is
\[ S_{bp} = -0.5\left(\log_{10}[Chl\,a] - 0.3\right), \]
valid for
\[ 0.02 < [Chl\,a] < 2 \; \text{mg m}^{-3}. \]
Outside this range, the slope is set to zero.
Q.2.4 Backscattering Fraction
The particulate backscattering fraction is parameterized as
\[ B_p = 0.002 + 0.01\left(0.5 - 0.25\log_{10}[Chl\,a]\right). \]
The parameter \(B_p\) is assumed to be spectrally independent (Huot et al. (2008)).
Q.2.5 Phase matrix (TBD)
- water
- NAP
- Phytoplankton (FF)
Q.3 Bio-2 Optical Model (Five-Parameter Model)
The Bio-2 model describes optically complex coastal waters using a five-parameter parameterization (Gao et al. 2018).
The ocean is assumed to be homogeneous with a depth of 200 m and composed of four optical components:
- pure seawater
- phytoplankton covariant particles
- non-algal particles (NAP)
- colored dissolved organic matter (CDOM)
CDOM contributes only to absorption, while the other components contribute to both absorption and scattering.
The absorption and scattering coefficients of pure seawater, \(a_w\) and \(b_w\), are taken from laboratory measurements. The backscattering fraction of pure water is assumed to be
\[ B_w = 0.5. \]
Q.3.1 Phytoplankton Absorption
Phytoplankton absorption is parameterized as
\[ a_{ph}(\lambda) = A_{ph}(\lambda)\,[Chl\,a]^{E_{ph}(\lambda)}. \]
Here
- \(A_{ph}(\lambda)\) and \(E_{ph}(\lambda)\) are empirical coefficients
- \([Chl\,a]\) is chlorophyll concentration in mg m\(^{-3}\).
Q.3.2 CDOM and Detrital Absorption
The combined absorption coefficient of CDOM and NAP is modeled as
\[ a_{dg}(\lambda) = a_{dg}(440)\exp\!\left[-S_{dg}(\lambda-440)\right]. \]
The parameter \(S_{dg}\) represents the exponential spectral slope.
Q.3.3 Particulate Backscattering
The particulate backscattering coefficient is modeled as
\[ b_{bp}(\lambda) = b_{bp}(660)\left(\frac{\lambda}{660}\right)^{-S_{bp}}. \]
The spectral slope \(S_{bp}\) controls the wavelength dependence.
Q.3.4 Backscattering Fraction
The particulate backscattering fraction is
\[ B_p(\lambda) = B_p(660)\left(\frac{\lambda}{660}\right)^{-S_{Bp}}. \]
The parameter \(S_{Bp}\) defines the spectral variation of the backscattering fraction.
Q.3.5 Total Inherent Optical Properties
The total absorption coefficient is
\[ a(\lambda) = a_w(\lambda) + a_{ph}(\lambda) + a_{dg}(\lambda). \]
The total backscattering coefficient is
\[ b_b(\lambda) = 0.5\,b_w(\lambda) + b_{bp}(\lambda). \]
These inherent optical properties are widely used to describe the ocean color spectrum in bio-optical models.
Q.4 Bio-3 Optical Model (Three-Parameter Model)
The Bio-3 model is a reduced-parameter version of the Bio-2 formulation designed to balance physical realism with retrieval stability.
In this model, a subset of the Bio-2 parameters is retrieved while the remaining parameters are constrained using empirical relationships derived from observations (Aryal et al. 2024, 2026; Hannadige et al. 2023).