Publications

Hu, MQ; Zhang, YC; Ma, RH; Xue, K; Cao, ZG; Chu, Q; Jing, YY (2021). Optimized remote sensing estimation of the lake algal biomass by considering the vertically heterogeneous chlorophyll distribution: Study case in Lake Chaohu of China. SCIENCE OF THE TOTAL ENVIRONMENT, 771, 144811.

Abstract
Due to the difference of vertical distribution of algae in lakes, it is necessary to carry out remote sensing estimation of algal biomass based on the vertically heterogeneous distribution of chlorophyll in order to improve the accuracy of biomass inversion. A new algorithm is proposed and validated to measure algal biomass in Lake Chaohu based on the Moderate Resolution Imaging Spectrometer (MODIS) images. The algal biomass index (ABI) is defined as the difference in remote-sensing reflectance (R-rs, sr(-1)) at 555 nm normalized against two baselines with one formed linearly between R-rs(859) and R-rs(469) and another formed linearly between R-rs(645) and R-rs(469). Both theory and model simulations show that ABI has a good relation with the algal biomass in the euphotic zone (R-2 = 0.88, p < 0.01, N = 50). Field data were further used to estimate the biomass outside the euphotic layer through an empirical algorithm. The ABI algorithm was applied to MODIS Rayleigh-corrected reflectance (R-rc) data after testing the sensitivity to sun glint and thickness of aerosols, which showed an acceptable precision (root mean square error < 21.31 mg and mean relative error < 16.08%). Spectral analyses showed that ABI algorithm was immune to concentration of colored dissolved organic matter (CDOM) but relatively sensitive to suspended particulate inorganic matter (SPIM), which can be solved by using Turbid Water Index (TWI) though in such a challenging environment. A long-term (2012-2017) estimation of algal biomass was further calculated based on the robust algorithm, which shows both seasonal and spatial variations in Lake Chaohu. Tests of ABI algorithm on Sentinel-3 OLCI demonstrates the potential for application in other remote sensors, which meets the need of observation using multi-sensor remote sensing in the future. (C) 2021 Elsevier B.V. All rights reserved.

DOI:
10.1016/j.scitotenv.2020.144811

ISSN:
0048-9697