communities resulted in a marked decrease in satellite-based chlorophyll fluorescence efficiency. Further evidence was given by Behrenfeld et al. (2009) who
highlighted the tight correspondence between regions of elevated MODIS fluorescence quantum yields and modeled regions of Fe-limitation and low dust deposition
(the primary mechanism for Fe input to the open ocean). These findings imply that
satellite-detected regions of elevated Chl fluorescence yields may provide an avenue
for time-resolved assessments of phytoplankton assimilation efficiencies that can
account for the unique physiological consequences iron stress. Furthermore, such an
approach could account for the highly dynamic nature of iron supply, which can be
linked to episodic upwelling or atmospheric deposition events.
8.6.3 Phytoplankton Community Composition
It has long been recognized that ocean NPP exhibits strong regional variability and
that some of this variability is tied to community taxonomic structure. Regionally
specific NPP algorithms have been employed to indirectly account for taxonomic
variability. For example, several studies have partitioned the ocean into distinct
‘biogeographical provinces’ that are empirically assigned unique photosynthetic
parameters (Longhurst et al. 1995; Sathyendranath et al. 1995). This approach
draws upon extensive field data sets of
14 C uptake data and avoids any necessity
for explicit predictive relationships. Alternatively, empirical predictive relationships may be derived for specific broad ocean regions, such as the Arctic or
Southern Ocean (Arrigo et al. 2008a, b).
In addition to influencing regional variability, taxonomic contributions to NPP
have relevance to understanding ecosystem carbon flow, export efficiency, and
fisheries production (Ryther 1969), for example. A variety of satellite ocean-color
based studies have aimed to directly decompose bulk emergent bio-optical signals
into contributions from different phytoplankton groups. While few models have
been successful at resolving species-level differences in satellite ocean color data
(Westberry and Siegel 2006; Balch et al. 2005; Alvain et al. 2008; Bracher et al.
2009), algorithms do exist for identifying broad phytoplankton size classes (Ciotti
et al. 2002; Devred et al. 2006; Uitz et al. 2006; Hirata et al. 2008). These techniques rely on the first order relationship between cell size and ecosystem function
(after Sieburth et al. 1978). However, the end-point of most of these studies has
been to assess different phytoplankton size-class contributions to pigment biomass
(Chl) only, rather than their contributions to NPP. For example, Uitz et al. (2006)
used a large in situ dataset of Chl and other diagnostic pigment markers to generate
empirical parameterizations between surface [satellite] Chl and relative dominance
of three size classes of phytoplankton; pico-, nano-, and micro-phytoplankton. The
link between size-fractionated Chl estimates and NPP was made in subsequent
work by Uitz and co-workers who associated class-specific photophysiological
variables with pigment-based size classes in field datasets (Uitz et al. 2008), then
applied these relationships to satellite data (Uitz et al. 2010).
8 Oceanic Net Primary Production
219
highlighted the tight correspondence between regions of elevated MODIS fluorescence quantum yields and modeled regions of Fe-limitation and low dust deposition
(the primary mechanism for Fe input to the open ocean). These findings imply that
satellite-detected regions of elevated Chl fluorescence yields may provide an avenue
for time-resolved assessments of phytoplankton assimilation efficiencies that can
account for the unique physiological consequences iron stress. Furthermore, such an
approach could account for the highly dynamic nature of iron supply, which can be
linked to episodic upwelling or atmospheric deposition events.
8.6.3 Phytoplankton Community Composition
It has long been recognized that ocean NPP exhibits strong regional variability and
that some of this variability is tied to community taxonomic structure. Regionally
specific NPP algorithms have been employed to indirectly account for taxonomic
variability. For example, several studies have partitioned the ocean into distinct
‘biogeographical provinces’ that are empirically assigned unique photosynthetic
parameters (Longhurst et al. 1995; Sathyendranath et al. 1995). This approach
draws upon extensive field data sets of
14 C uptake data and avoids any necessity
for explicit predictive relationships. Alternatively, empirical predictive relationships may be derived for specific broad ocean regions, such as the Arctic or
Southern Ocean (Arrigo et al. 2008a, b).
In addition to influencing regional variability, taxonomic contributions to NPP
have relevance to understanding ecosystem carbon flow, export efficiency, and
fisheries production (Ryther 1969), for example. A variety of satellite ocean-color
based studies have aimed to directly decompose bulk emergent bio-optical signals
into contributions from different phytoplankton groups. While few models have
been successful at resolving species-level differences in satellite ocean color data
(Westberry and Siegel 2006; Balch et al. 2005; Alvain et al. 2008; Bracher et al.
2009), algorithms do exist for identifying broad phytoplankton size classes (Ciotti
et al. 2002; Devred et al. 2006; Uitz et al. 2006; Hirata et al. 2008). These techniques rely on the first order relationship between cell size and ecosystem function
(after Sieburth et al. 1978). However, the end-point of most of these studies has
been to assess different phytoplankton size-class contributions to pigment biomass
(Chl) only, rather than their contributions to NPP. For example, Uitz et al. (2006)
used a large in situ dataset of Chl and other diagnostic pigment markers to generate
empirical parameterizations between surface [satellite] Chl and relative dominance
of three size classes of phytoplankton; pico-, nano-, and micro-phytoplankton. The
link between size-fractionated Chl estimates and NPP was made in subsequent
work by Uitz and co-workers who associated class-specific photophysiological
variables with pigment-based size classes in field datasets (Uitz et al. 2008), then
applied these relationships to satellite data (Uitz et al. 2010).
8 Oceanic Net Primary Production
219