Operational Visibility

English Channel and Bay of Biscay Demonstrator

Click here to access a presentation of VISIOPE in PDF format

Optical transparency of ocean waters flowing in the upper layer of the English Channel and the Bay of Biscay is presented. Displayed images come from either satellite remote sensing or digital simulation (analysis and forecast). The parameters describing water transparency are the following:

Horizontal Visibility

expressed in meters, is the sighting range of a black disk target moved horizontally away from a diving observer (Duntley 1963).

Z Secchi

expressed in meters, is the vertical sighting range of a traditional Secchi disk lowered from a surface observer down to the water column (Preisendorfer 1986). The upper water column is assumed homogeneous on the vertical.

Values of these parameters are displayed every day. The turbidity is assumed constant during the day. Dates of observations and simulations are labeled according to a YYYYMMDD format (for instance May 19th, 2014 is labeled 20140519).

Satellite images (MODIS et VIIRS sensors) displayed are those acquired by the satellite one day before, because the night is necessary for data receiving, processing and delivering. Simulations of transparency (labeled SIMUL) are for the present day, one day before (analysis) and 5 days in advance (forecast).


Only transparency of surface waters are displayed. For instance waters flowing below the surface waters of river plumes (Adour, Gironde, Loire, Seine) can be clearer than at the surface. On the opposite, in areas where mud or sand resuspension occurs (Pertuis charentais, Noirmoutier) waters underneath can be more turbid than at the surface.

Simulation of coastal turbidity is based on a statistical model using the SHOM tidal coefficient and the significant wave height. This model does not take into account temporal variations of river plumes. Hence simulations in the vicinity of river plumes can be less reliable.

Users can compare recent simulations with satellite observations before using forecast in order to evaluate the quality of the simulation in case of period of abnormal statistical dynamics.

Satellite observations

Observations come from satellite remote sensing of surface waters, in the visible domain. Missing data due to the presence of clouds are displayed in black areas. Images are acquired by the following sensors:

MODIS : Moderate Resolution Imaging Spectroradiometer, on board AQUA satellite (USA).

VIIRS : Visible and Infrared Imager/Radiometer Suite), on board NPP satellite (USA).

Processing algorithms are from ACRI-ST and LOV Villefranche (Doron et al 2011). The relative precision on observed parameters has been estimated to be around 30% (2 σ).


Opposite to satellite observations, simulations are “free from clouds”. The simulated turbidity is assumed to have two main origins: mineral and organic. Mineral turbidity mainly comes from sand and mud resuspension while organic turbidity comes from phytoplankton growth. Detrital material and colored dissolved organic matter are not explicitly modeled.

Highly turbid waters are generally of mineral origin. These waters can be found near the coast where sediments are present on the sea floor. River plumes can be turbid also. Off the shelf, turbidity is mainly organic.

Mineral turbidity is computed using SHOM tidal coefficients and local values of the significant wave height modeled with WaveWatch III (Ardhuin et al 2010). The relation between turbidity and its forcing terms (waves and tides) has been settled by statistical analysis of series of MODIS and MERIS (ESA/ENVISAT) satellite images spanning six years (2008 to 2013) in the area, according to Rivier et al (2012). Satellite processing algorithms were from IFREMER (Gohin et al 2005).

Organic turbidity is not simulated but simply composed of recent satellite measurements of chlorophyll a (CHLA composite L4 from IFREMER).

Optical backscattering, absorption and attenuation coefficients are function of the level of organic and mineral contents, according to empirical relations deduced from the literature and adapted to our satellite data algorithms (see SHOM 2014 presentation).

Finally Secchi depths and horizontal visibility ranges are computed by applying the contrast reduction equation to the previously computed optical coefficients (Duntley 1963).

The overall precision of a simulation of the visibility ranges has not yet been evaluated.

Brest, May 21st, 2014
Frédéric Jourdin
SHOM, France.


Ardhuin, Fabrice, and Coauthors, 2010: Semiempirical Dissipation Source Functions for Ocean Waves. Part I: Definition, Calibration, and Validation. J. Phys. Oceanogr., 40, 1917–1941.

S.Q. Duntley 1963 « Light in the Sea » J Opt Soc Am 53, 214-233

Maéva Doron, Marcel Babin, Odile Hembise, Antoine Mangin, Philippe Garnesson 2011 « Ocean transparency from space: Validation of algorithms estimating Secchi depth using MERIS, MODIS and SeaWiFS data » Remote Sensing of Environment 115: 2986–3001

Francis Gohin, Sophie Loyer, Michel Lunven, Claire Labry, Jean-Marie Froidefond, Daniel Delmas, Martin Huret, Alain Herbland 2005 « Satellite-derived parameters for biological modelling in coastal waters: Illustration over the eastern continental shelf of the Bay of Biscay » Remote Sensing of Environment 95: 29–46

Aurélie Rivier, Francis Gohin, Philippe Bryère, Caroline Petus, Nicolas Guillou, Georges Chapalain 2012 « Observed vs. predicted variability in non-algal suspended particulate matter concentration in the English Channel in relation to tides and waves » Geo-Mar Lett 32:139–151