Partitioning particulate scattering and absorption into contributions of phytoplankton and non-algal particles in winter in Lake Taihu (China)

Light scattering, backscattering, and absorption coefficients of particles were observed at 62 locations in Lake Taihu (China) in November 2008. A method using a priori knowledge and the measured data was proposed to partition particulate scattering and absorption into contributions of phytoplankton and non-algal particles. The results showed that phytoplankton weakly contributed to the particulate scattering and backscattering with the mean b (ph)/b (p) values usually below 10% and b (bph)/b (bt) values of 0.3-3.9% in the whole visible light spectrum, and an approximate relationship of b (bt) a parts per thousand b (bp) a parts per thousand b (bnap) was regarded as reasonable in Lake Taihu. In contrast with scattering and backscattering, phytoplankton made more contributions to the particulate absorption with the mean a (ph)/a (p) values varying in a wide range of about 20-70%. Both the scattering and absorption spectra of non-algal particles can be modeled well by corresponding methods. A power function model was used to simulate the scattering spectra, which presented high predictive accuracies with MAPE values usually below 5% and RMSE values below 1.5 m(-1), while the spectral absorption model also performed well with mean S (nap) being 0.0052 nm(-1) (standard deviation, SD = 0.0010 nm(-1)). As to the phytoplankton absorption, a quadratic function model used was considered to have a good performance with corresponding parameters being supported at each wavelength in the spectral range of 400-700 nm. Additionally, two basic bio-optical parameters were determined, that is, b (nap) (*) (550) = 0.604 m(2) g(-1) and a (ph) (*) (675) = 0.0288 m(2) mg(-1). Overall, these results obtained in the present study supply us with new knowledge about optical properties of suspended particulates in an inland and highly turbid lake (Lake Taihu), which are beneficial to the development of analytical models of water color remote sensing.