Development of synthetic salinity from remote sensing for the Columbia River plume

The Columbia River plume (CRP) is an ecologically important source of nutrients, pollutants, and fresh water to the Oregon/Washington shelf. It is traditionally under-sampled, with observations constrained to ships or moorings. High-spatial- and temporal-resolution observations afforded by satellites would increase sampling if the plume could be quantitatively detected in the imagery. Two empirical algorithms are presented using data from the Moderate Resolution Imaging Spectroradiometer (MODIS) to estimate sea surface salinity in the region of CRP. Salinity cannot be detected directly, so a proxy for fresh, water is employed. Light absorption, by chromophoric dissolved organic matter ( ^aCDOM) is inversely proportional to salinity and linear because of conservative mixing of CDOM-rich terrestrial runoff with surrounding ocean water. To estimate synthetic salinity, simple linear (salinity versus ^aCDOM) and multiple linear (salinity and temperature versus ^aCDOM) algorithms were developed from in situ measurements of ^aCDOM collected on the Coastal Ocean Processes-River Influences on Shelf Ecosystems cruises. These algorithms were applied to MODIS 250 m resolution data layers of sea surface temperature and absorption by colored dissolved and detrital matter (^aCDM) estimated at 350 nm and 412 nm from the Garver-Siegel-Maritorena model version 1 algorithm. Validation of MODIS-derived synthetic salinity with coincident in situ measurements revealed significant correlation during both downwelling (simple, β₁ = 0.95 and r² = 0.89; multiple, β₁ = 0.92 and r ² = 0.89) and upwelling periods (simple, β¹ = 1.26 and r² = 0.85; multiple, β₁ = 1.10 and r² = 0.87) using the 412 nm data layer. Synthetic salinity estimated using the 350 nm data layer consistently overestimated salinity. These algorithms, when applied to ^aCDM at 412 nm, enable synoptic observations of CRP not permitted by ships or moorings alone.