Inversion of Particle Size Distribution and Abundance of Phytoplankton Species From Their Backscattering Coefficients in Visible Light

This work presents a systematic inversion approach to estimate the abundances and parameters of particle size distribution (PSD) of multiple phytoplankton species and non-algal particles (NAPs) from their backscattering coefficients in visible light. Each phytoplankton cell is approximated as a three-layered sphere composed of membrane, chloroplast, and cytosol compartments. A first particle swarm optimization algorithm is developed to estimate the ingredient volume fractions in each cellular compartment from measured absorption and backscattering coefficients. A second optimization algorithm is developed to invert the abundances and PSD parameters of different species under random refractive index perturbation and measurement noise. A parameter averaging method is used to estimate the PSD parameters of single species with reasonable accuracy. A two-stage method is proposed to estimate the abundances and PSD parameters of multiple bloom-forming species in the subpolar North Atlantic. The efficacy of the proposed methods has been verified by simulations, and these methods are well organized to include more measurement data in the future to further improve their estimation accuracy.