sensors do not provide reliable quantitative estimates of in situ coral cover, nor can they
discriminate live coral from fleshy algae (Dustan et al., 2000; Green et al., 1996).
In contrast, regional landscape definition, i.e. the synoptic discrimination of
geomorphic zones using imagery from Landsat, SPOT, and similar satellites, is of great
value from a reef system E modeling perspective (Bouvet et al., 2003). This coarsescale mapping is the first step in the implementation of our proposed synoptic E model
strategy, because the landscape units thus identified form a mosaic of compartments
within the model domain. The required model complexity for each compartment may
then be matched to its openness to abiotic and biotic exchange processes.
A recent SPOT-based study of atoll rims in French Polynesia (Andréfouët et al.,
2001b) illustrates the role proposed by the authors for moderate resolution satellite
sensors within the suggested synoptic E model framework. Andréfouët et al. (2001b)
used SPOT HRV multi-spectral images to characterize the structure and degree of
hydrodynamic closure of atoll rims, distinguishing nine rim types based on the
dominance of vegetated, submerged, intertidal, and emerged domains. Knowledge of
rim structure and climatological ocean swell were combined to estimate the water
residence time for the lagoons, yielding a quantitative measure of closure that has major
consequences for internal biological processes (Andréfouët et al., 2001b). In closely
related studies, SPOT images were used to assess atoll reef-flat spillways bordering
lagoons in the Tuamotu Archipelago, and to thereby investigate the use of water
renewal time as a criterion for categorizing the adjacent lagoons (Andréfouët and Payri,
2001; Pages and Andréfouët, 2001). Potential flow rates and water-renewal times
under various swell conditions were calculated for the spillways observed by SPOT,
and a positive correlation between estimated water-renewal times and lagoonal
concentrations of dissolved organic matter (DOM) was identified (Pages and
Andréfouët, 2001). Similarly, Andréfouët and Payri (2001) found a significant
relationship between phytoplankton biomass and water-renewal time for lagoons
greater than 25 km
2 in surface area. Pages and Andréfouët (2001) concluded that SPOT
images are useful in the rapid regional assessment of lagoon hydrodynamics.
3.4 MAPPING INTRA-ZONE BENTHIC BIOTOPES
The use of habitat mapping based on remote sensing in the estimation of reef
system E assumes that metabolic performance is correlated with the composition of the
benthic communities within reef zones (Hatcher, 1997b). Aside from the traditional
descriptive interpretation of aerial photography, remote sensing studies of coral reefs
challenging, because high spatial detail is required (Maeder et al., 2002; Mumby et al.,
1997a), and because coral, algae, and seagrass all contain photosynthetic pigments and
exhibit similar spectral reflectance (Holden and LeDrew, 1998, 1999; Myers et al.,
1999; Hochberg and Atkinson, 2000; Schalles et al., 2000; Hedley and Mumby, 2002;
Hochberg et al., 2003; Kutser et al., 2003).
Aircraft-based sensing has significant advantages in the ecological mapping of reef
system components. Tropical coastal ecosystems are highly complex geographically,
and the high resolution images acquired from aircraft can capture intricate spatial detail.
High resolution aircraft imaging, either aerial photography or hyperspectral scanning
and imaging, can provide habitat information at the level of coral colonies and patch
reefs (Catt and Hopley, 1988). Moreover, aircraft can fly below the cloud cover that
typically obscures features on satellite images in the tropics (Chauvaud et al., 1998).
prior to the mid-1990s were mainly attempts to map geomorphic zones (Green
et al., 1996). The discrimination of ecologically meaningful assemblages is more
118
Brock, Yates and Halley
discriminate live coral from fleshy algae (Dustan et al., 2000; Green et al., 1996).
In contrast, regional landscape definition, i.e. the synoptic discrimination of
geomorphic zones using imagery from Landsat, SPOT, and similar satellites, is of great
value from a reef system E modeling perspective (Bouvet et al., 2003). This coarsescale mapping is the first step in the implementation of our proposed synoptic E model
strategy, because the landscape units thus identified form a mosaic of compartments
within the model domain. The required model complexity for each compartment may
then be matched to its openness to abiotic and biotic exchange processes.
A recent SPOT-based study of atoll rims in French Polynesia (Andréfouët et al.,
2001b) illustrates the role proposed by the authors for moderate resolution satellite
sensors within the suggested synoptic E model framework. Andréfouët et al. (2001b)
used SPOT HRV multi-spectral images to characterize the structure and degree of
hydrodynamic closure of atoll rims, distinguishing nine rim types based on the
dominance of vegetated, submerged, intertidal, and emerged domains. Knowledge of
rim structure and climatological ocean swell were combined to estimate the water
residence time for the lagoons, yielding a quantitative measure of closure that has major
consequences for internal biological processes (Andréfouët et al., 2001b). In closely
related studies, SPOT images were used to assess atoll reef-flat spillways bordering
lagoons in the Tuamotu Archipelago, and to thereby investigate the use of water
renewal time as a criterion for categorizing the adjacent lagoons (Andréfouët and Payri,
2001; Pages and Andréfouët, 2001). Potential flow rates and water-renewal times
under various swell conditions were calculated for the spillways observed by SPOT,
and a positive correlation between estimated water-renewal times and lagoonal
concentrations of dissolved organic matter (DOM) was identified (Pages and
Andréfouët, 2001). Similarly, Andréfouët and Payri (2001) found a significant
relationship between phytoplankton biomass and water-renewal time for lagoons
greater than 25 km
2 in surface area. Pages and Andréfouët (2001) concluded that SPOT
images are useful in the rapid regional assessment of lagoon hydrodynamics.
3.4 MAPPING INTRA-ZONE BENTHIC BIOTOPES
The use of habitat mapping based on remote sensing in the estimation of reef
system E assumes that metabolic performance is correlated with the composition of the
benthic communities within reef zones (Hatcher, 1997b). Aside from the traditional
descriptive interpretation of aerial photography, remote sensing studies of coral reefs
challenging, because high spatial detail is required (Maeder et al., 2002; Mumby et al.,
1997a), and because coral, algae, and seagrass all contain photosynthetic pigments and
exhibit similar spectral reflectance (Holden and LeDrew, 1998, 1999; Myers et al.,
1999; Hochberg and Atkinson, 2000; Schalles et al., 2000; Hedley and Mumby, 2002;
Hochberg et al., 2003; Kutser et al., 2003).
Aircraft-based sensing has significant advantages in the ecological mapping of reef
system components. Tropical coastal ecosystems are highly complex geographically,
and the high resolution images acquired from aircraft can capture intricate spatial detail.
High resolution aircraft imaging, either aerial photography or hyperspectral scanning
and imaging, can provide habitat information at the level of coral colonies and patch
reefs (Catt and Hopley, 1988). Moreover, aircraft can fly below the cloud cover that
typically obscures features on satellite images in the tropics (Chauvaud et al., 1998).
prior to the mid-1990s were mainly attempts to map geomorphic zones (Green
et al., 1996). The discrimination of ecologically meaningful assemblages is more
118
Brock, Yates and Halley