Hydrography-Driven Optical Domains in the Vaigat-Disko Bay and Godthabsfjord: Effects of Glacial Meltwater Discharge

Veloisa J. Mascarenhas, Oliver Zielinski

In: Frontiers in Marine Science (FMarS) 6 335 Pages 1-19 Frontiers Media SA Lausanne Switzerland 6/2019.


We present here an assessment of in situ hyperspectral bio-optical variability in the Vaigat-Disko Bay and Godthabsfjord along the southwest coast of Greenland. The dataset consists of state-of-the-art profiler measurements of hyperspectral apparent and inherent optical properties of water complemented by traditional observations of Secchi disk and Forel Ule scale in the context of ocean color. Water samples were collected and analyzed for concentration of optically active constituents (OACs). Near-surface observations of hydrographic parameters revealed three different water masses in the Bay: meltwater plume, frontal zone, and Atlantic water mass. Underwater spectral light availability reveals three different spectral types. Low salinity, increased temperature, deep euphotic depths, and case-1 water type remote sensing reflectance spectra with tabletop peaks in the 400–500 nm wavelength range characterize the glacial meltwater plume in the Vaigat-Disko Bay. The conservative relationship between salinity and chromophoric dissolved organic matter (CDOM) commonly observed in estuarine and shelf seas is weaker in the Godthabsfjord and reverses in the Vaigat-Disko Bay. Efficiency of machine learning techniques such as cluster analysis is tested in delineating water masses in the bay w.r.t. hydrographic and bio-optical parameters. Tests of optical closure yield low root mean square error at longer wavelengths. The study provides strong evidence that despite similar geographic setting, fjord ecosystems exhibit contrasting bio-optical properties which necessitate fjord-specific investigations.

Weitere Links

mascarenhas-zielinski-hydrography-driven-optical-domains-vaigat-disko-and-godthabsfjord-fmars-06-00335-201906.pdf (pdf, 4 MB)

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz