A new algorithm for discriminating water sources from space: A case study for the southern Beaufort Sea using MODIS ocean color and SMOS salinity data

Identification of surface water sources in the Arctic Ocean is a key factor to better understand physical and biogeochemical processes. It is however restricted both geographically and temporally when using field observations. In this proof-of-concept study, we propose a new algorithm for discriminating surface water sources using satellite remote sensing data alone. The algorithm uses salinity and the light absorption coefficient of colored dissolved organic matter at 443nm [aCDOM(443), m−1] derived from SMOS/MIRAS and Aqua/MODIS satellite sensors, respectively, to identify the fraction of three end-members (i.e., seawater, ice melt water, and river water including precipitation) through the mass balance equations. An uncertainty analysis showed that fractions of river water can be derived reasonably, with caution of fractions for ice melt water and seawater. Application of this algorithm may lead to the discrimination of water sources in the surface layer of the Arctic Ocean in various environments where seawater, ice melt water, and river water are intermingled, which might be useful to improve our understanding of physical and biogeochemical processes related to each water source. •A new algorithm for discriminating water sources using satellite data is presented for Arctic Ocean.•Fractions of water sources can be discriminated with the associated uncertainties from space.•The satellite approach is complimentary to in situ discrimination.