Evaluation of the Quasi-Analytical Algorithm (QAA) for Estimating Total Absorption Coefficient of Turbid Inland Waters in Northeast China

Total light absorption plays a vital role in aquatic ecosystem. In this research, the performance of quasi-analytical algorithm (QAA) was evaluated for deriving water total absorption coefficient a(λ), from remote sensing reflectance R
_rs
(λ), using in situ data sets collected in typically turbid waters including Chagan Lake, Songhua Lake, and Shitoukoumen and Xinlicheng reservoirs in Northeast China. A better performance of algorithm was observed when the calculation parameter of reference wavelength was shifted from 670 nm to 700 or 740 nm. In particular, the accuracy of estimated absorption coefficients at 550 and 675 nm for Songhua Lake in the summer was much improved when using 700 nm as reference band. In general, the QAA typically underestimated a
_nw
(λ) (nonwater absorption) for all the waters in different seasons with regression slope lower than one especially at 400 nm. However, longer reference wavelength leads to more uncertainties in deriving a
_nw
(λ) with higher root-mean-square error (RMSE) and absolute percentage difference (APD) at 440 nm. It was found that if the average contribution of chromophoric dissolved organic matter (CDOM) absorption at 440 nm dominates over the nonwater absorption, QAA achieved a better predicted result. Under the premise of that, the larger the contribution rate of particulate absorption at 440 nm, the poorer the predicted results were derived. The relationship could be established between a
_nw
(λ) and R
_rs
(λ) through QAA algorithm, but it is not an optimal algorithm for turbid case 2 waters and large amounts of in situ data sets in different seasons are needed to calibrate the algorithm to achieve better performances.