Satellite remote sensing of a low-salinity water plume in the East China Sea

Abstract. With the aim to map and monitor a low-salinity water (LSW) plume in the East China Sea (ECS), we developed more robust and proper regional algorithms from large in-situ measurements of apparent and inherent optical properties (i.e. remote sensing reflectance, R_rs, and absorption coefficient of coloured dissolved organic matter, a_CDOM) determined in ECS and neighboring waters. Using the above data sets, we derived the following relationships between visible R_rs and absorption by CDOM, i.e. R_rs (412)/R_rs (555) vs. a_CDOM (400) (m⁻¹) and a_CDOM (412) (m⁻¹) with a correlation coefficient R² 0.67 greater than those noted for R_rs (443)/R_rs (555) and R_rs (490)/R_rs (555) vs. a_CDOM (400) (m⁻¹) and a_CDOM (412) (m⁻¹). Determination of a_CDOM (m⁻¹) at 400 nm and 412 nm is particularly necessary to describe its absorption as a function of wavelength λ using a single exponential model in which the spectral slope S as a proxy for CDOM composition is estimated by the ratio of a_CDOM at 412 nm and 400 nm and the reference is explained simply by a_CDOM at 412 nm. In order to derive salinity from the absorption coefficient of CDOM, in-situ measurements of salinity made in a wide range of water types from dense oceanic to light estuarine/coastal systems were used along with in-situ measurements of a_CDOM at 400 nm, 412 nm, 443 nm and 490 nm. The CDOM absorption at 400 nm was better inversely correlated (R²=0.86) with salinity than at 412 nm, 443 nm and 490 nm (R²=0.85–0.66), and this correlation corresponded best with an exponential (R²=0.98) rather than a linear function of salinity measured in a variety of water types from this and other regions. Validation against a discrete in-situ data set showed that empirical algorithms derived from the above relationships could be successfully applied to satellite data over the range of water types for which they have been developed. Thus, we applied these algorithms to a series of SeaWiFS images for the derivation of CDOM and salinity in the context of operational mapping and monitoring of the springtime evolution of LSW plume in the ECS. The results were very encouraging and showed interesting features in surface CDOM and salinity fields in the vicinity of the Yangtze River estuary and its offshore domains, when a regional atmospheric correction (SSMM) was employed instead of the standard (global) SeaWiFS algorithm (SAC) which revealed large errors around the edges of clouds/aerosols while masking out the nearshore areas. Nevertheless, there was good consistency between these two atmospheric correction algorithms over the relatively clear regions with a mean difference of 0.009 in a_CDOM (400) (m⁻¹) and 0.096 in salinity (psu). This study suggests the possible utilization of satellite remote sensing to assess CDOM and salinity and thus provides great potential in advancing our knowledge of the shelf-slope evolution and migration of the LSW plume properties in the ECS.