derived products. We then review some of the major findings regarding global
ocean NPP and its variability and conclude with a discussion of new directions for
improving our retrieval and understanding of this critical ecosystem property.
8.2 Theoretical Basis
Field measurements of ocean primary production were made throughout the
twentieth century, but the modern measurement of NPP using radiolabeled carbon
(
14 C) can be traced to Steeman-Nielsen (1952). Following its introduction,
application of the
14
C technique proliferated in oceanographic field studies, owing
to its ease of use, high sensitivity, and ability to yield production estimates following a relatively short sample incubation period. Early
14 C studies provided
fundamental insights that were soon incorporated into NPP modeling efforts.
Simple empirical relationships between NPP and Chl or ambient light (PAR)
emerged as some of the first predictive expressions for aquatic NPP (Ryther and
Yentsch 1957; Talling 1957; Vollenweider 1966). Subsequent modeling efforts
have focused on a variety of additional factors, including detailed descriptions of
the underwater light field (Morel 1991; Smyth et al. 2005), improved characterization of physiology (Armstrong 2006; Westberry et al. 2008), and definition of
regionally-specific properties (Longhurst et al. 1995; Arrigo et al. 2008b).
Unfortunately, the increasing complexity of NPP models has often not translated
into improved predictive ability (see Sect. 8.4) and even the most complex satellite
NPP models remain necessarily crude representations of the photosynthetic variability revealed by genetic, biochemical, and physiological laboratory studies.
While remote sensing retrieval of NPP is challenging, its fundamental relationship is straight forward. By definition, NPP in a given water parcel is the
product of the extant phytoplankton biomass (expressed in the same currency as
NPP, carbon) and its specific growth rate (l),
NPP ¼ C phyto X l
ð8:1Þ
The two quantities, C phyto x l, encapsulate dependencies on several aspects of
the phytoplankton growth environment. For example, biomass (C phyto ) reflects a
balance between growth and loss processes, such as grazing by zooplankton. By
contrast, l is largely a function of light and nutrient availability.
Equation 8.1 represents the fundamental relationship for NPP, but it is not the
basis for most remote sensing NPP algorithms because both C phtyo and l are
grossly undersampled in the ocean, largely due to methodological difficulties. In
practice, chlorophyll concentration (Chl) has served as the central metric of
phytoplankton standing stock. Conversion of Chl into NPP thus requires a characterization of assimilation efficiency (i.e., net primary production per unit chlorophyll; P
b ). Much of the current error in NPP estimates results from unconstrained
variability in this ‘photosynthetic efficiency’ term (Milutinovic and Bertino 2011;
8 Oceanic Net Primary Production
207
ocean NPP and its variability and conclude with a discussion of new directions for
improving our retrieval and understanding of this critical ecosystem property.
8.2 Theoretical Basis
Field measurements of ocean primary production were made throughout the
twentieth century, but the modern measurement of NPP using radiolabeled carbon
(
14 C) can be traced to Steeman-Nielsen (1952). Following its introduction,
application of the
14
C technique proliferated in oceanographic field studies, owing
to its ease of use, high sensitivity, and ability to yield production estimates following a relatively short sample incubation period. Early
14 C studies provided
fundamental insights that were soon incorporated into NPP modeling efforts.
Simple empirical relationships between NPP and Chl or ambient light (PAR)
emerged as some of the first predictive expressions for aquatic NPP (Ryther and
Yentsch 1957; Talling 1957; Vollenweider 1966). Subsequent modeling efforts
have focused on a variety of additional factors, including detailed descriptions of
the underwater light field (Morel 1991; Smyth et al. 2005), improved characterization of physiology (Armstrong 2006; Westberry et al. 2008), and definition of
regionally-specific properties (Longhurst et al. 1995; Arrigo et al. 2008b).
Unfortunately, the increasing complexity of NPP models has often not translated
into improved predictive ability (see Sect. 8.4) and even the most complex satellite
NPP models remain necessarily crude representations of the photosynthetic variability revealed by genetic, biochemical, and physiological laboratory studies.
While remote sensing retrieval of NPP is challenging, its fundamental relationship is straight forward. By definition, NPP in a given water parcel is the
product of the extant phytoplankton biomass (expressed in the same currency as
NPP, carbon) and its specific growth rate (l),
NPP ¼ C phyto X l
ð8:1Þ
The two quantities, C phyto x l, encapsulate dependencies on several aspects of
the phytoplankton growth environment. For example, biomass (C phyto ) reflects a
balance between growth and loss processes, such as grazing by zooplankton. By
contrast, l is largely a function of light and nutrient availability.
Equation 8.1 represents the fundamental relationship for NPP, but it is not the
basis for most remote sensing NPP algorithms because both C phtyo and l are
grossly undersampled in the ocean, largely due to methodological difficulties. In
practice, chlorophyll concentration (Chl) has served as the central metric of
phytoplankton standing stock. Conversion of Chl into NPP thus requires a characterization of assimilation efficiency (i.e., net primary production per unit chlorophyll; P
b ). Much of the current error in NPP estimates results from unconstrained
variability in this ‘photosynthetic efficiency’ term (Milutinovic and Bertino 2011;
8 Oceanic Net Primary Production
207