Remote sensing of surface [nitrite + nitrate] in river-influenced shelf-seas: The northern South China Sea Shelf-sea

Sea surface concentrations of [nitrite + nitrate], [N + N], have been assessed successfully by remote sensing in the open oceans by utilizing its relationship to sea surface temperature (SST). A similar approach was not met with similar success in the river-influenced shelf waters in the northern South China Sea Shelf-sea (NoSoCS) where riverine inputs, in which the relationship between the concentration of [N + N] and SST was not straight forward, are significant. By considering river-influenced shelf waters as a mixture of open ocean water and riverine inputs and then combining SST with the absorption coefficient of colored dissolved organic matter (CDOM) at 412 nm, ag(412), which serves as an indicator of the riverine influence, an algorithm has been developed for remotely sensing the surface concentration of [N + N] in the NoSoCS. This satellite-derived concentration of [N + N] was validated by direct observations. In 16 match-up comparisons within a time window of ±24 h and concentrations of [N + N] ranging between 0.2 and 74 μM, the uncertainty was ±40%. The climatological distributions of the derived concentration of [N + N] in the NoSoCS and surrounding waters between 2002 and 2014 are consistent with the reported distributions based on ship-board observations. Thus, as a result of riverine inputs, the concentration of [N + N] generally increases towards the coasts, varying by more than two orders of magnitude from about 0.5 μM and near zero in the open northern South China Sea (NSCS) in January and July respectively to about 50 and 100 μM at the mouth of the Pearl River. In the open NSCS where nutrient availability is controlled primarily by the enhanced winter convective overturn, the intra-annual variations in the concentration of [N + N] follow a distinct seasonal cycle, reaching a minimum close to zero in the summer and a maximum, 0.7 μM, in the winter. In the NoSoCS, as a result of the high riverine input of nutrients in the summer, in addition to the maximum in the winter, there is a secondary summer maximum, which becomes increasingly prominent towards the coast. This work represents a first attempt to characterize the distributions of the surface concentration of [N + N] in river-influenced shelf-seas from space. With appropriate regional tuning, a similar approach may be applicable to other river-influenced shelf-seas such as the Texas-Louisiana Shelf.