accumulation or export. Therefore, a significant absolute value of E implies non-steady
state conditions at the time-space scale under consideration (Gattuso et al., 1998;
Kinsey, 1985; Larkum, 1983). A community with positive E is defined as net
autotrophic, and if the reverse condition (negative E) exists, the community is called
net heterotrophic (Smith, 1988).
E is a measure of community metabolism that has been recognized by numerous
researchers as a key rate variable that describes the holistic trophic state of coral reef
ecosystems or their functional components (Hatcher, 1997a; Hatcher, 1997b).
Accordingly, E, in addition to net calcification (G), has been selected herein as a
currency for integrating process measurements and remote sensing within coral reef
ecosystem models. Estimating the excess production of reefs requires the evaluation of
basic ecosystem processes such as community photosynthesis, respiration,
remineralization, and organic carbon transfers. Together, these data can support a
holistic understanding of reef system ecological function (Grigg et al., 1984; Hatcher,
1997a).
Field investigations of reef metabolic processes using established methods are
limited by the requirement to track change in water chemistry in large controlled
volumes over defined time periods (Hatcher, 1997b). Consequently, biogeochemical
fluxes are determined only during brief expeditions, and the resulting data sets
generally do not provide spatial detail, diurnal variation, seasonality, or replicate
measurements (Kinsey, 1985). Crossland et al. (1991) noted that, given uncertainty in
flow respirometry measurements of reef metabolism, the E of whole reef systems must
be uncertain by at least +/- 0.3 g C m
-2 d
-1 . This degree of uncertainty casts doubt on
the conclusions of Smith and Buddemeier (1995) regarding E, and also those of
geographically specific studies. For example, the investigation of Grigg et al. (1984)
reported an E of 0.29 g C m
-2 d
-1 for French Frigate Shoals, Hawaii, which is about
equal to the uncertainty proposed by Crossland et al. (1991). However, after one-half
century of investigations (for the most part in the western Pacific and Atlantic Basins) it
seems reasonable to conclude that the global E of reef systems is not high, and that
reefs are generally about as net autotrophic as the oligotropic tropical oceans they
inhabit.
Flow or chamber respirometry can provide estimates of the P g and R of individual
reef communities, but not the E of interlinked community mosaics in near trophic
balance, the generic condition on coral reefs. The inability to determine if E differs
<4% from zero using these methods (Smith and Buddemeier, 1995) is a major
shortcoming, because it implies that even if practical restraints were eliminated, it
would still be impossible to quantify the actual net metabolism of entire reef systems
based solely on flow or chamber respirometry (Smith, 1983; Smith, 1985).
Furthermore, the metabolism of reef-flats, where most measurements have been made,
cannot be generalized to reef-slopes, lagoons, or entire reef systems (Hatcher, 1990).
Yet, even very minor deviation of E from zero has great significance in determining
trophic balance, nutrient import or export, and harvestable yield (Kinsey, 1985; Smith,
1983; Smith, 1985). Clearly, an alternate approach that admits variability in time-space
scales is required for estimation of the net community metabolism of entire reef
systems (Smith and Buddemeier, 1995).
Coral reef ecosystems have geographical extents in the tens of kilometers, and the
coral reef provinces within which these ecosystems nest have even larger space scales.
Aside from the measurement uncertainties cited above, the direct in situ measurement
of whole ecosystem properties such as excess production is an intractable problem due
to the breadth and geospatial variability of reef systems (Hatcher, 1997b). As an
113
Coral Reef Ecosystem Process Studies & Remote Sensing
state conditions at the time-space scale under consideration (Gattuso et al., 1998;
Kinsey, 1985; Larkum, 1983). A community with positive E is defined as net
autotrophic, and if the reverse condition (negative E) exists, the community is called
net heterotrophic (Smith, 1988).
E is a measure of community metabolism that has been recognized by numerous
researchers as a key rate variable that describes the holistic trophic state of coral reef
ecosystems or their functional components (Hatcher, 1997a; Hatcher, 1997b).
Accordingly, E, in addition to net calcification (G), has been selected herein as a
currency for integrating process measurements and remote sensing within coral reef
ecosystem models. Estimating the excess production of reefs requires the evaluation of
basic ecosystem processes such as community photosynthesis, respiration,
remineralization, and organic carbon transfers. Together, these data can support a
holistic understanding of reef system ecological function (Grigg et al., 1984; Hatcher,
1997a).
Field investigations of reef metabolic processes using established methods are
limited by the requirement to track change in water chemistry in large controlled
volumes over defined time periods (Hatcher, 1997b). Consequently, biogeochemical
fluxes are determined only during brief expeditions, and the resulting data sets
generally do not provide spatial detail, diurnal variation, seasonality, or replicate
measurements (Kinsey, 1985). Crossland et al. (1991) noted that, given uncertainty in
flow respirometry measurements of reef metabolism, the E of whole reef systems must
be uncertain by at least +/- 0.3 g C m
-2 d
-1 . This degree of uncertainty casts doubt on
the conclusions of Smith and Buddemeier (1995) regarding E, and also those of
geographically specific studies. For example, the investigation of Grigg et al. (1984)
reported an E of 0.29 g C m
-2 d
-1 for French Frigate Shoals, Hawaii, which is about
equal to the uncertainty proposed by Crossland et al. (1991). However, after one-half
century of investigations (for the most part in the western Pacific and Atlantic Basins) it
seems reasonable to conclude that the global E of reef systems is not high, and that
reefs are generally about as net autotrophic as the oligotropic tropical oceans they
inhabit.
Flow or chamber respirometry can provide estimates of the P g and R of individual
reef communities, but not the E of interlinked community mosaics in near trophic
balance, the generic condition on coral reefs. The inability to determine if E differs
<4% from zero using these methods (Smith and Buddemeier, 1995) is a major
shortcoming, because it implies that even if practical restraints were eliminated, it
would still be impossible to quantify the actual net metabolism of entire reef systems
based solely on flow or chamber respirometry (Smith, 1983; Smith, 1985).
Furthermore, the metabolism of reef-flats, where most measurements have been made,
cannot be generalized to reef-slopes, lagoons, or entire reef systems (Hatcher, 1990).
Yet, even very minor deviation of E from zero has great significance in determining
trophic balance, nutrient import or export, and harvestable yield (Kinsey, 1985; Smith,
1983; Smith, 1985). Clearly, an alternate approach that admits variability in time-space
scales is required for estimation of the net community metabolism of entire reef
systems (Smith and Buddemeier, 1995).
Coral reef ecosystems have geographical extents in the tens of kilometers, and the
coral reef provinces within which these ecosystems nest have even larger space scales.
Aside from the measurement uncertainties cited above, the direct in situ measurement
of whole ecosystem properties such as excess production is an intractable problem due
to the breadth and geospatial variability of reef systems (Hatcher, 1997b). As an
113
Coral Reef Ecosystem Process Studies & Remote Sensing