A three-step semi analytical algorithm (3SAA) for estimating inherent optical properties over oceanic, coastal, and inland waters from remote sensing reflectance

We present a three-step inverse model (3SAA) for estimating the inherent optical properties (IOPs) of surface waters from remote sensing reflectance spectra, Rrs(?). The derived IOPs include total (a(?)), phytoplankton (aphy(?)), and colored detrital matter (acdm(?)), absorption coefficients, (bb(?)) particulate (bbp(?)) backscattering coefficients. first step uses an improved neural network approach to estimate diffuse attenuation coefficient downwelling irradiance Rrs. a(?) bbp(?) are then estimated using LS2 (Loisel et al., 2018), which does not require spectral assumptions on hence can assess bb(?) at any wavelength Rrs(?) is measured. Then, optimization algorithm combined with water class (OWC) aphy(?) acdm(?) anw(?).The proposed evaluated in situ dataset collected open oceanic, coastal, inland waters. Comparisons other standard semi-analytical algorithms (QAA GSM), as well match-up exercises, have also been performed. applicability OLCI observations was assessed through analysis global spatial patterns 3SAA GSM. good performance manifested by median absolute percentage differences (MAPD) 13%, 23%, 34% bbp(443), anw(443), aphy(443) acdm(443), respectively oceanic Due absence constraints inversion IOPs, adoption OWC-based approach, only slightly degraded bio-optical complex