Launched in 1978, the experimental satellite Nimbus-7 was aimed towards 
the study of marine resources. One of its radiometers, the Coastal Zone Color 
Scanner (CZCS) gathers data in distinct channels with defined wavelengths, most 
of them in the visible spectrum (Table I). Evaluation of phytoplankton distribu- 
tion is an interesting approach to analyze oceanic biomass. It can be achieved 
by remote sensing since phytoplankton content changes the optical ‘characteristics 
of the sea. As the phytoplankton concentration increases, the maximum transmis- 
sion shifts towards green (Yentsch, 1960). Indeed, phytoplankton contains the 
photosynthetically active pigment, chlorophyll a, which absorbs strongly near 
443nm. Thus, determination of the upwelling subsurface radiance, Lgg (W.m-2st71), 
from the signal, Lo, reaching the satellite sensor enables the evaluation of 
chlorophyll concentration in the sea. The phytoplanktonic biomass can then be 
derived approximately (Hjjerslev, 1980). 
The present study was carried out in order to gain further insight on 
the distribution of phytoplankton in the Northwestern Mediterranean (Gulf of 
Lions). Furthermore, we attempted to reveal the most characteristic boundaries 
of this biomass and to follow their variations throughout the year 1979. Indeed, 
using computerized treatment of CZCS data we were able to map phytoplankton dis- 
tribution and to visualize some characteristic oceanographic features such as 
mesoscale cyclonic eddy, outflow of fresh waters from the Rhóne river, and 
coastal upwellings. 
Table I: Characteristics of the "Coastal Zone Color Scanner" 
  
Channels 
1 2 3 4 5 6 
Wavelengths 0.443 0.520 0.550 0.670 0.750 11.5 
center 
(Micrometers) (blue) (green) (yellow) (red) (near infrared) (infrared) 
AX 0.02 0.02 0.02 0.02 1.00 2.00 
(Micrometers) 
  
Atmospheric effects 
  
The CZCS collects data through different channels (Table I) with de- 
fined wavelengths. One of them (n?1) is centered at 443nm i.e. the main absorp- 
tion peak of chlorophyll g (Hovis et al. 1980). Since the spectral signature of 
phytoplankton represents less than 5$ of the signal reaching the radiometer, it 
is most important to correct for atmospheric influence (Quentzel, Kaestner, 
1980). For this purpose, we used the algorithm described by Gordon and Clark 
(1980) with slight modifications. In the modified method (Wilson, Smith, 1980) 
the hypothesis that the sea is totally absorbent at 670nm is not supposed to be 
valid near the coast. Thus, it is only the starting hypothesis for an iterative 
procedure which derives the subsurface upwelling radiance Lgg. 
Phytoplankton concentration 
Algorithms to derive phytoplankton pigment concentration from the Lgg 
values obtained in the various spectral bands of the CZCS, have been empirically 
derived from several sea-truth measurements (Morel, Prieur, 1978; Austin, 1980; 
Gordon et al., 1980). The algorithms used in the present study are of the type: 
C-aRjb where C is the concentration of chlorophyll a and phaeopigments, Rij is 
690 
Pe N 
ans 
ues rA - 
iM LS — dora R^ as " » 
a — LL mm i o tamtn e at heme 
BR QO << OY 
Hub ~0 th ©