Dear Prof. Klemas,

one of the approaches is to apply an semi-emprirical bio-optical algorithm that can retrieve several water constituents simultaneously. Such algorithms employ RTM that account for spectral contribution to the satellite measured remote sensing reflectance  from several components, e.g. water per se, chlorophyl-a, suspended mineral matter, dissolved organic carbon. Among the most  known algorithms are QAA by Lee et all (http://oceancolor.gsfc.nasa.gov/WIKI/QAA.html) or GSM by Garver and Siegel, 1997. et al (Garver, S. A., and D. A. Siegel (1997), Inherent optical property inversion

of ocean color spectra and its biogeochemical interpretation: I.

Time series from the Sargasso Sea, J. Geophys. Res., 102, 18,607 –

18,625.)

We also have tried to develop and apply our algorithm (called BOREALI) in the Kara Sea which you may find interesting.

(https://www.researchgate.net/publication/241087744_Spaceborne_quantitative_assessment_of_dissolved_organic_carbon_fluxes_in_the_Kara_Sea)

With best regards!

Anton

Article Spaceborne quantitative assessment of dissolved organic carb...

One possibility with hyperspectral imaging is to use a modeling system such as for example, https://www.researchgate.net/publication/7736417_Interpretation_of_hyperspectral_remote-sensing_imagery_by_spectrum_matching_and_look-up_tables with a range of CDOM and Chl-a in the models. 

If there's enough Chl-a, you may get a signal from fluorescence (although high CDOM will tend to reduce this for obvious reasons).

On the other hand there are still features due to the phytoplankton near the red edge that are not fluorescence - but it requires a fair bit of phytoplankton which is likely not present if there's truly a lot of CDOM (enough CDOM, fewer photons in the blue/green due to much higher absorption, therefore worse conditions for phytoplankton). Given enough CDOM even this feature may be reduced. 

Again, it seems to me that modeling the hyperspectral signature with something like HydroLight may give the best tools to determine what kinds of estimates are possible.

Article Interpretation of hyperspectral remote-sensing imagery by sp...

Hi Vic,

Greetings!!

You might check in with Bruce Monger at Cornell:

Bruce Monger

Senior Research Associate

Department of Earth & Atmospheric Sciences

4124 Snee Hall

Cornell University

Ithaca, New York 14853

-----------------------

Phone: 607-227-2972

Fax: 607-254-4780

Email:  [email protected]

Cheers,

Al Strong

Hi Victor,

The best strategy is to test retrieval algorithms at longer wavelengths. You should check the band rations algorithms with spectral band combination Rrs(550)/Rrs(590). We have tested it in the Baltic Sea and it works. You could also try Rrs(550)/Rrs(620) combination. Another strategy is to retrieve the total absorption at 676 nm from semi-analytical algorithms and than relate it to chlorophyll a concentration. CDOM absorption at 676 is negligible even in most organic reach waters.

You can refer to the following publication:

Conference Paper Investigation the Behavior of MODIS Ocean Color Products und...