the two data sets were highly correlated with
a r^ of 0.89. Moreover, the spatial pattern of
in situ values in the northeastern gulf bet-
ween 20 and 28 May 1981 corresponds closely
with two images taken during that time frame
bringing further confidence in the selected
atmospheric correction process.
signal
Variability of underwater
contamination
The IHS analysis showed that the areas conta-
minated principally by suspended matter in the
lower St. Lawrence estuary (LSLE) and western
GSL or by coccoliths in the eastern GSL were
relatively small and well defined in time and
space. They also appeared to covary positively
with the temporal pigment averages. The
pigment concentrations estimated by the
selected approach should thus be reasonably
accurate for most of the times and areas
covered by the database.
3. RESULTS AND DISCUSSION
In this paper, a powerful and useful remote
sensing tool was used to measure phytoplankton
pigment concentrations in the GSL and build
the first complete data set of satellite
images showing the pigment distribution in the
GSL (Fuentes-Yaco et al., 1995). This data set
will be now analyzed to evaluate both the
spatial and temporal variability of phyto-
plankton pigments in the GSL.
Spatial patterns
The major spatial feature seen in the seasonal
(Fig. 2) and annual (Fig. 3) composites is the
persistent west to east gradient between the
LSLE,” the northwest gulf and the central-
eastern gulf regions. This global distribution
of pigments in the GSL is closely linked to
major dynamical features known to exist. The
LSLE, Gaspé Current, and northwestern qulf
exhibited the highest concentrations of
chlorophyll-like pigments throughout the year
due to the presence of gyres and wind-induceq
upwellings. In this regard, these areas may be
considered as pigment pumps, helping to
increase concentration in the central and
southern regions (Magdalen Shallows) by
advection through the Honguedo Strait. In
shallower areas such as Baie des Chaleurs and
the Magdalen Islands, other factors, including
wind stress, contribute to the increase in
pigment concentrations through vertical mixing
and local ,nutrient enrichment. Downwelling
processes in the northeastern GSL appear to be
responsible for the low surface pigment
concentrations usually found throughout the
year in this region. However, some limited
production was detected in northern Esquiman
Channel, which may be linked to wind-driven
upwelling.
Seasonal variability
The calculated seasonal pigment composites
(Fig.2) showed high pigment concentrations in
spring and low ones in summer (Table 1).
However, the results also indicated that the
highest pigment concentrations occurred during
fall, which has not been reported previously.
The observed seasonal cycle of chlorophyll-
like pigments is characterized mostly by
changes in the areal extent of high pigment
values rather than in a general large-scale
rise and fall in concentrations. These areas
extend far ’éastward during August ‘end
September and cover even larger areas than
during spring.
Table 1. Statistics (mean and standard error) of phytoplankton pigment concentration (mg md)
by season and year.
Annual means
Region Statistic Spring Summer
LSLE mean 1e. 3 1.4
SE 0.2 0.2
GSL mean 0.6 0.6
SE 0:4 0.1
Interannual variability
Yearly means showed that 1980 had the highest
phytoplankton pigment concentrations of the
three years studied (Table 1), with the area
of strong pigment concentration in the western
GSL extending from the LSLE to Northumberland
Strait (Figure 3). During that year, high
pigment values were also observed along the
northern shore of the GSL and around the
Magdalen Islands. The year 1979 was
408
Seasonal means
Fall 1979 1980 1981 Average
tend lcd 1.6 1.2 1.4
0-2 0.3 0.2 0:2 0.1
0.8 0.6 0.8 0.5 0.6
Qoi 0.1 0.1 0.1 0.1
characterized by intermediate mean values,
with the areal pigment distribution being
close to the three-year average. Finally, 198
seemed to have the lowest concentrations of
the three years. However, the northwestern
Esquiman Channel showed elevated and
persistent pigment concentrations in 1981.
The observed temporal variations of the
spatial chlorophyll patterns parallel to some
extent the differences in salinity %
influenced by estuary runoff (Bugden et al
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996