305
In this paper the results of recent studies of coccolithophore blooms in the North Atlantic are
summarised. The remote sensing and in situ observations on light scattering by coccoliths are
considered in terms of providing new information about coccolithophore distributions and
about the effects of this group of phytoplankton on the general optical and biogeochemical
properties of the ocean-atmosphere system, rather than in relation to the complex optical
processes themselves (see Gordon et al., 1988). The possible causes of large scale changes
in ocean phytoplankton ecology from diatom dominance (cold periods) to coccolithophore
dominance (warm periods), and the implications of such changes in terms of feedback effects
on the global environment (see Lovelock, 1986), are discussed. Finally, potential applications
of new methods of individual particle analysis in studies of the ecology and biogeochemistry
of coccolithophores are briefly evaluated.
COCCOLITHOPHORE OPTICS
Conditions of high (10-25 %) water reflectance have now been observed frequently by satellite
during spring and summer months for temperate and sub-polar oceans and shelf seas,
including upwelling areas (e.g. Fukushima et al., 1987; Dupouy and Demercq, 1987;
GREPMA, 1988; Ackleson et al., 1988) and, in all cases for which contemporary plankton
data was obtained, E. huxleyi was abundant. Examples of E. huxteyi blooms in the North Sea
and North Atlantic are illustrated in Fig. 2. The first attempt to relate satellite and plankton
data quantitatively (Holligan et al., 1983) demonstrated a correlation between reflectance and
the log of cell density. However, numbers of detached coccoliths were higher by as much as
two orders of magnitude and, on the grounds both of theoretical considerations based on
particle size and shape and of subsequent optical studies (see below), it is now established that
the free coccoliths as opposed to calcified cells are the dominant cause of light scatter.
Backscattering of light by coccoliths (Balch et at., 1989) and other types of particles is an
important potential source of error in algorithms for estimating chlorophyll using satellite
ocean colour sensors such as the CZCS, so that knowledge of both absorption and scattering
properties of coccolithophores is important for the quantitative analysis of such satellite data.
Although CZCS reflectance spectra for coccolithophore blooms often exhibit maximum values
in the blue waveband (443nm), observations on the development of individual blooms indicate
that blue reflectance relative to green (520nm) or yellow (550nm) tends to be low initially,
probably as a result of higher pigment concentrations (or high cell to coccolith ratios) during
the growth phase of the populations (Groom and Holligan, 1987). In this context it should be
noted that a major accessory pigment in coccolithophores, hexanoylfucoxanthin, shows peak
absorption at 520nm (Haxo, 1985).
In this paper the results of recent studies of coccolithophore blooms in the North Atlantic are
summarised. The remote sensing and in situ observations on light scattering by coccoliths are
considered in terms of providing new information about coccolithophore distributions and
about the effects of this group of phytoplankton on the general optical and biogeochemical
properties of the ocean-atmosphere system, rather than in relation to the complex optical
processes themselves (see Gordon et al., 1988). The possible causes of large scale changes
in ocean phytoplankton ecology from diatom dominance (cold periods) to coccolithophore
dominance (warm periods), and the implications of such changes in terms of feedback effects
on the global environment (see Lovelock, 1986), are discussed. Finally, potential applications
of new methods of individual particle analysis in studies of the ecology and biogeochemistry
of coccolithophores are briefly evaluated.
COCCOLITHOPHORE OPTICS
Conditions of high (10-25 %) water reflectance have now been observed frequently by satellite
during spring and summer months for temperate and sub-polar oceans and shelf seas,
including upwelling areas (e.g. Fukushima et al., 1987; Dupouy and Demercq, 1987;
GREPMA, 1988; Ackleson et al., 1988) and, in all cases for which contemporary plankton
data was obtained, E. huxleyi was abundant. Examples of E. huxteyi blooms in the North Sea
and North Atlantic are illustrated in Fig. 2. The first attempt to relate satellite and plankton
data quantitatively (Holligan et al., 1983) demonstrated a correlation between reflectance and
the log of cell density. However, numbers of detached coccoliths were higher by as much as
two orders of magnitude and, on the grounds both of theoretical considerations based on
particle size and shape and of subsequent optical studies (see below), it is now established that
the free coccoliths as opposed to calcified cells are the dominant cause of light scatter.
Backscattering of light by coccoliths (Balch et at., 1989) and other types of particles is an
important potential source of error in algorithms for estimating chlorophyll using satellite
ocean colour sensors such as the CZCS, so that knowledge of both absorption and scattering
properties of coccolithophores is important for the quantitative analysis of such satellite data.
Although CZCS reflectance spectra for coccolithophore blooms often exhibit maximum values
in the blue waveband (443nm), observations on the development of individual blooms indicate
that blue reflectance relative to green (520nm) or yellow (550nm) tends to be low initially,
probably as a result of higher pigment concentrations (or high cell to coccolith ratios) during
the growth phase of the populations (Groom and Holligan, 1987). In this context it should be
noted that a major accessory pigment in coccolithophores, hexanoylfucoxanthin, shows peak
absorption at 520nm (Haxo, 1985).