Detection of Phytoplankton Groups in European Waters using Satellite images

Mati Kahru,


“Standard” ocean color images (e.g. O’Reilly et al., 1998) of chlorophyll-a are indispensable in providing spatial distributions of phytoplankton biomass in Case 1 waters (by definition Case 1 waters are those where the optical properties are dominated by phytoplankton and its associated products). Most coastal waters, e.g. in European shelf seas, are not Case 1 waters and the standard ocean color algorithms have large errors due to problems in atmospheric correction and interference by suspended sediments and colored dissolved organic matter (CDOM). It may seem contradictory but blooms of specific phytoplankton groups also produce anomalous optical properties that deviate from Case 1 waters. It has been demonstrated that these anomalous optical properties can be used to distinguish between blooms of different phytoplankton groups or even species.  The following analysis used software from Wimsoft ( The time series of satellite data with information on phytoplankton groups can provide unprecedented information on the space-time dynamics of the marine ecosystem.

Currently the following phytoplankton taxa can be distinguished from satellite data:

·        Nodularia spumigena (and other cyanobacteria accumulations in the Baltic Sea, Kahru et al., 1994, Kahru, 2000)

·        Cyanobacteria Trichodesmium spp. in the global ocean (Subramaniam et al., 2002)

·        Coccolithophores (Brown et al., 1994)

·        Dinoflagellates with Mycosporine-like amino acids (MAAs) (Kahru and Mitchell, 1998). This method involves detection of UV reflectance that was potentially available from the Japanese sensor GLI.

·        Diatom blooms (Sathyendranath et al., 2004)

·        Detection of other phytoplankton groups in specific conditions is quite possible but needs further development.


·        The example above is a SeaWiFS image of nLw555 (normalized water-leaving radiance at 555 nm) on July 9, 1999. It shows increased reflectance due to at least two separate blooms: a coccolithophore bloom in the Skagerrak and the Nodularia bloom in the Baltic Sea.




·        Classification result of the previous image: coccolithophore bloom shown in black and the Nodularia bloom in dark gray.


While classification of phytoplankton blooms at 1 km resolution looks great compared to the resolution obtained from a typical oceanographic cruise, it has a number of limitations. In general case it is probably impossible to invert of the remote sensing reflectance spectrum and derive the concentrations of phytoplankton groups. The detection of phytoplankton groups is only possible under certain conditions:

·        The bloom has a relatively high biomass and occurs near the surface.

·        The bloom involves certain conspicuous phytoplankton species or groups with specific optical characteristics (e.g. Nodularia, coccolithophores, Trichodesmium, large diatoms with strong pigment packaging, etc).

·        Knowledge of the local phytoplankton is crucial: certain species or groups are known to bloom in certain areas and this helps their detection.



·        The example above (SeaWiFS nLw555) from March 15, 2002 shows the spring bloom consisting of different phytoplankton assemblages. Knowledge of the local phytoplankton is essential for classifying this image. A very tentative classification without knowledge of the local conditions might be three assemblages with the associated with the dominant group: (1) coccolithophores, (2) diatoms, (3) eukaryotic picoplankton. The distribution of the assemblages is shown by their respective number on the image.


The study of phytoplankton optical characteristics is an area of active research and new results are expected. It is certain that much more studies are needed in this very promising field. Currently there is only one satellite-derived time series of a phytoplankton group that is more than a decade long and consistent over the whole period – the time series of the cyanobacteria (Nodularia spumigena) blooms in the Baltic Sea (Kahru et al., 1994; Kahru, 1997; Kahru et al., 2000).




Brown, C.W. and J.A. Yoder. Coccolithophorid blooms in the global ocean. J. Geophys. Res., 99, 7467-7482.

Kahru, M., U. Horstmann, O. Rud, Satellite detection of increased cyanobacteria blooms in the Baltic Sea: Natural fluctuation or ecosystem change? Ambio, 23 (8):  469-472, 1994.

Kahru, M.  Using satellites to monitor large-scale environmental change:  A case study of cyanobacteria blooms in the Baltic Sea.  In:  Monitoring algal blooms:  New techniques for detecting large-scale environmental change.  M. Kahru and Ch. W. Brown (Eds.).  43-61, 1997.

Kahru, M., B.G. Mitchell.  Spectral reflectance and absorption of a massive red tide off Southern California.  J. Geophys. Res. Vol. 103, No. C10, 21,601-21,609, 1998.

Kahru, M., J.-M. Leppänen, O.Rud, O.P. Savchuk. Cyanobacteria blooms in the Gulf of Finland triggered by saltwater inflow into the Baltic Sea.  Marine Ecology Progress Series, 207:13-18, 2000.

O'Reilly, J.E., S. Maritorena, B.G. Mitchell, D.A. Siegel, K.L. Carder, S.A. Garver, M. Kahru and C.R. McClain.  Ocean color chlorophyll algorithms for SeaWiFS.  J. Geophys. Res. Vol. 103, No.C11, p. 24,937-24,953, 1998.

Sathyendranath, S., L. Watts, E.Devred, T. Platt, C. Caverhill, and H. Maass. Disrcimination of diatoms from other phytoplankton using ocean color data. Mar. Ecol. Prog. Ser., 272, 59-68, 2004.

Subramaniam, A., C.W. Brown, R.R. Hood, E.J. Carpenter, and D.G. Capone. Detecting Trichodesmium blooms in SeaWiFS imagery, Deep-Sea Res. II, 49, 107-121, 2002,