Quality control based Chlorophyll-a estimation with two-band and three-band algorithms using Sentinel-2 MSI data in a complex inland lake, China

The estimation of Chlorophyll-a concentration remains challenging for inland waters owing to the optical complexity. Hence, a quality control-based Chlorophyll-a retrieval approach was proposed for the Sentinel-2 multispectral instrument. Two algorithms were selected: 2-band algorithm, which is suitable for extremely turbid waters, and improved 3-band algorithm designed for moderately turbid waters. Two scatter plots between Rrs560/Rrs490 and Rrs412/Rrs443 and between Rrs560/Rrs490 and Rrs560 were generated to determine boundary lines of the quality control test for the two selected algorithms. The quality control test was calibrated using 188 groups of field data collected from Daihai Lake. Results show that the two selected algorithms were considerably improved after the quality control test. The retrieval of Chlorophyll-a concentration was further enhanced by integrating the improved algorithms. The quality control test was also evaluated against twelve Sentinel-2 images. The integrated algorithm was found to be suitable for most areas in Daihai Lake.