on the algal-dominated reef crest appears to be advected as detritus to backreef sinks.
This gradient in metabolic activity inward from the reef periphery implies a general
shift away from net autotrophy towards close mixotrophic balance as hydrodynamic
closure increases, and parallels the obvious biological and morphological zonation
seen on reefs. Moreover, a general correlation between calcification and gross
photosynthesis is embedded in this spatial pattern (Kinsey, 1977).
These broad geomorphic patterns in organic metabolism led Crossland et al. (1991)
to hypothesize that reef-perimeter zones are coupled to interior reef zones by
hydrodynamics and the feeding behaviour of organisms. Kinsey (1983) observed that
within reef systems that are in overall trophic balance, P/R is usually greater than one in
high-energy zones that are hydrodynamically open and act as sources of organic matter.
Quiescent backreef settings with higher water-residence times were recognized as sinks
for organic matter (Kinsey, 1985). Essentially, Kinsey (1983, 1985) proposed that the
positive excess production on reef slopes, crests, and flats is carried downstream, and
subsequently deposited, on low-energy backreef platforms, or, perhaps sporadically due
to severe storms, in lagoons. This functional association between neighboring producer
and consumer communities acts to retain production within the ecosystem (Hatcher,
1990; Smith and Buddemeier, 1995). A consequence for the modeling of whole-reef
organic-carbon metabolism is that it is not valid to generalize the plentiful
measurements acquired on the easily-worked reef-flats to the entire reef system.
The use of moderate resolution satellite remote sensing to segment reef systems
based on boundaries defined by hydrodynamic closure is a key element in integrating
in situ process measurements with remote sensing. If so organized, the development of
excess production models may proceed from a simple formulation for backreef areas,
where the advection of nutrients and organic detritus is comparatively steady and
predictable, toward more complex designs for fringing reefs that admit the effect of
their more open boundaries. As model complexity increases to allow trans-boundary
transfers, the required diversity of metabolism measurements in the field, the image
mapping of benthic community structure, and the sensing of environmental forcing all
scale up in unison. This is typically a valid paradigm for open ocean reef systems,
clearly established by the observed differing metabolic performance of reef-slopes,
flats, and lagoons (Hatcher, 1997a).
3.3 REMOTE SENSING OF REEF SYSTEM ZONES
Globally, coral reefs cover about 600 km
2
. Improved definition of zonation within
this total area would result in a first order improvement in reef carbon-flux estimates
(Crossland et al., 1991). Fortunately, remote sensing affords the capability to nonintrusively map large coastal areas repeatedly, synoptically, and at various levels of
detail dictated in part by the spatial, temporal, spectral, and radiometric resolutions of
the chosen sensor (Green et al., 1996). Using the interpretation of reef-flat, knoll, and
lagoon areas on aerial photographs, Atkinson and Grigg (1984) acquired representative
measurements of metabolic performance for each zone in an investigation of benthic
net community production at French Frigate Shoals, Hawaii. That study was an early
attempt to use a form of aerial remote sensing to discriminate net heterotrophic and net
autotrophic reef zones, and to model them as distinct internal entities in order to attain
valid holistic estimates of E for the atoll under investigation.
Two decades ago, the remote sensing capabilities for landscape definition that are
required to apply the approach of Atkinson and Grigg (1984) across entire reef systems
were not available. In contrast, at present a new generation of aircraft and satellite116
Brock, Yates and Halley
This gradient in metabolic activity inward from the reef periphery implies a general
shift away from net autotrophy towards close mixotrophic balance as hydrodynamic
closure increases, and parallels the obvious biological and morphological zonation
seen on reefs. Moreover, a general correlation between calcification and gross
photosynthesis is embedded in this spatial pattern (Kinsey, 1977).
These broad geomorphic patterns in organic metabolism led Crossland et al. (1991)
to hypothesize that reef-perimeter zones are coupled to interior reef zones by
hydrodynamics and the feeding behaviour of organisms. Kinsey (1983) observed that
within reef systems that are in overall trophic balance, P/R is usually greater than one in
high-energy zones that are hydrodynamically open and act as sources of organic matter.
Quiescent backreef settings with higher water-residence times were recognized as sinks
for organic matter (Kinsey, 1985). Essentially, Kinsey (1983, 1985) proposed that the
positive excess production on reef slopes, crests, and flats is carried downstream, and
subsequently deposited, on low-energy backreef platforms, or, perhaps sporadically due
to severe storms, in lagoons. This functional association between neighboring producer
and consumer communities acts to retain production within the ecosystem (Hatcher,
1990; Smith and Buddemeier, 1995). A consequence for the modeling of whole-reef
organic-carbon metabolism is that it is not valid to generalize the plentiful
measurements acquired on the easily-worked reef-flats to the entire reef system.
The use of moderate resolution satellite remote sensing to segment reef systems
based on boundaries defined by hydrodynamic closure is a key element in integrating
in situ process measurements with remote sensing. If so organized, the development of
excess production models may proceed from a simple formulation for backreef areas,
where the advection of nutrients and organic detritus is comparatively steady and
predictable, toward more complex designs for fringing reefs that admit the effect of
their more open boundaries. As model complexity increases to allow trans-boundary
transfers, the required diversity of metabolism measurements in the field, the image
mapping of benthic community structure, and the sensing of environmental forcing all
scale up in unison. This is typically a valid paradigm for open ocean reef systems,
clearly established by the observed differing metabolic performance of reef-slopes,
flats, and lagoons (Hatcher, 1997a).
3.3 REMOTE SENSING OF REEF SYSTEM ZONES
Globally, coral reefs cover about 600 km
2
. Improved definition of zonation within
this total area would result in a first order improvement in reef carbon-flux estimates
(Crossland et al., 1991). Fortunately, remote sensing affords the capability to nonintrusively map large coastal areas repeatedly, synoptically, and at various levels of
detail dictated in part by the spatial, temporal, spectral, and radiometric resolutions of
the chosen sensor (Green et al., 1996). Using the interpretation of reef-flat, knoll, and
lagoon areas on aerial photographs, Atkinson and Grigg (1984) acquired representative
measurements of metabolic performance for each zone in an investigation of benthic
net community production at French Frigate Shoals, Hawaii. That study was an early
attempt to use a form of aerial remote sensing to discriminate net heterotrophic and net
autotrophic reef zones, and to model them as distinct internal entities in order to attain
valid holistic estimates of E for the atoll under investigation.
Two decades ago, the remote sensing capabilities for landscape definition that are
required to apply the approach of Atkinson and Grigg (1984) across entire reef systems
were not available. In contrast, at present a new generation of aircraft and satellite116
Brock, Yates and Halley