An improved algorithm for the column-integrated algal biomass retrieval in Lake Chaohu, a large eutrophic lake

Abstract

Column-integrated algal biomass (CAB), represented by integrated Chlorophyll-a concentrations (Chla) throughout the water column, is indicative of overall algal biomass for the entire lake. The performance of biomass retrieval algorithms is limited since previous research on CAB underestimate the impact of subsurface information. To enhance the CAB estimation in shallow lakes, an upper-to-lower method and a bivariate model were combined and a novel algorithm was applied to Lake Chaohu. Maximum Chlorophyll Index (MCI) algorithm outperformed the other three surface Chla retrieval algorithms for surface Chla inversion (R2 = 0.73, RMSE = 10.06 μg/L, MAPE = 35.22 %). The upper-to-lower method was used for subsurface Chla retrieval, depending on the empirical regression between two adjacent layers. The performance of different surface–subsurface combinations was validated to determine the optimal one for algorithm development. The algorithm was tested in two scenarios (no-bloom scenario and no-bloom + bloom scenario), displaying superior performance compared to conventional ones (R2 = 0.83, RMSE = 16.77 μg/L, MAPE = 25.68 %, N = 36, in no-bloom scenario; R2 = 0.88, RMSE = 32.34 μg/L, MAPE = 38.15 %, N = 50, in no-bloom + bloom scenario). Furthermore, the application of the algorithm was further extended to OLCI imageries, exhibiting a spatial consistency with RGB composites. The enhanced stability and reliability of total biomass estimations can provide a more comprehensive understanding of aquatic ecosystem.

Keywords

• Column-integrated algal biomass
• OLCI
• Upper-to-lower method