An improved algorithm for remotely sensing marine dissolved organic carbon: Climatology in the northern South China Sea shelf-sea and adjacent waters

By using field in situ observations, data from the NOMAD database, and remotely sensed data from MODIS-Aqua, an improved algorithm has been developed for estimating the surface concentrations of dissolved organic carbon (DOC) in the Northern South China Sea Shelf-sea (NoSoCS) and adjacent waters through a combination of the remotely sensed sea-surface temperature (SST) and the absorption coefficient of colored dissolved organic matter (CDOM). Unlike previously reported algorithms, which were based on only one of these two parameters, this algorithm does not require seasonal tuning and is applicable to both coastal and pelagic waters. The uncertainty in the remotely sensed concentrations of DOC is about ±4–10µM. The climatological distributions of the satellite-derived surface DOC in the NoSoCS and adjacent waters between 2002 and 2012 followed a general trend of increasing concentrations landward, from about 58 and 73μM in the open South China Sea (SCS) in January and July, respectively, to >100μM at the mouth of the Pearl River. The intra-annual changes in the monthly average concentration of DOC in the NoSoCS follow a distinct seasonal cycle, with the lowest and highest concentrations, 56μM and 75μM, in February and July, respectively. The seasonal cycle of DOC was in phase with that of SST but out of phase with that of primary production, suggesting that biological activity was not likely its primary control. Instead, it was consistent with the entrainment of the cold and less DOC-rich sub-surface water in the surface water by enhanced vertical mixing during winter, and the high inflow of organic-rich river water to the NoSoCS in summer. Sub-regionally, lower concentrations of DOC were found in the known upwelling zones in the NoSoCS, where vertical mixing was enhanced. In the inner shelf, the effect of riverine input was indicated by higher concentrations of DOC in the plume of water at the mouth of the Pearl River. This work represents the first attempt to characterize the regional distribution of surface concentration of DOC in the NoSoCS and adjacent waters.