Particle composition-based water classification method for estimating chlorophyll-a in coastal waters from OLCI images

Abstract

The complex composition of seawater presents significant challenges for accurately estimating biogeochemical data through optical measurements, both in situ and via satellite data. To address the regional applicability of single bio-optical or remote sensing algorithms caused by these challenges, we explored a water optical classification method based on inherent optical properties and particle composition. The ratio of organic particulate matter to total suspended particulate matter concentration (POM/SPM) serves as an optical discriminator of water bodies based on the proportions of organic and mineral particles. The boundary value is determined by the relationship between the particulate backscattering coefficient bbp(λ) and POM/SPM. By analyzing in situ data collected from the coastal waters of Qinhuangdao in the Bohai Sea, China, we developed empirical algorithms to estimate both the POM/SPM ratio and chlorophyll-a (Chl-a) concentration, the latter being a key parameter derived from current ocean remote sensing that indicates phytoplankton abundance. The evaluation of our algorithms demonstrates that accounting for POM/SPM variations significantly improves Chl-a estimate accuracy across the optically-complex coastal waters near Qinhuangdao compared to algorithms that do not consider changes in particle composition, such as the well-known OC4Me algorithm. Furthermore, we determined the distribution of monthly averaged Chl-a concentration and POM/SPM ratio on the coast of Qinhuangdao, Bohai Sea, in 2023. Our results show, for the first time, that the monthly average variations of the POM/SPM ratio in the Bohai Sea and Chl-a concentrations exhibit pronounced seasonal fluctuations.