326
S. Andréfouët
Coral reefs and associated ecosystems such as seagrass meadows are at risk given
the cumulative stress induced by human activities and climate change. Understanding ecosystem resilience is now a priority and a challenge (Nyström et al. 2008).
Resilience capacity results from a combination of numerous processes related to
biodiversity, functional redundancy, connectivity, heterogeneity, and their dynamics
under human pressure (e.g. fisheries) and climate change. Mapping resilience is in its
infancy given our ignorance on many aspects of reef resilience, the high complexity
at stake and the amount of new data needed. However, we identified two relevant
remote sensing approaches to study resilience.
In its simpler form, resilience can be measured by multidate image analysis and
change detection. Across decades, it may be possible to observe cycles of loss of
important communities and their recovery which suggests resilience, as in Scopélitis
et al. (2009) for Saint Leu Reef in La Réunion Island. Long term observations
may also suggest steady degradation and continuous loss of critical habitats, due to
chronic disturbances, precluding the return to previous functional states, e.g. Toliara,
Madagascar (Bruggeman et al.). The temporal dynamics of coral reefs, seagrass and
mangroves habitats have been studied in the Western Indian Ocean with a variety
of image sensors (Makota et al. 2004; Wang et al. 2003; Gullstroem et al. 2006;
Ferreira et al. 2009; Knudby et al. 2010b), sometimes with a multi-sensor, very high
resolution, approach (Scopélitis et al. 2009; Andréfouët et al. 2013) (Fig. 16.5).
Using a more complex modelling design, resilience has been tentatively mapped
throughout Saudi Arabia west coast by Rowlands et al. (2012). They identified first a
number of key variables and indices required to characterize resilience. These indices
were then inferred and mapped from remote sensing using a variety of assumptions,
images and processing. Both “landscape” (related to coral, framework, depth) and
“stress” (related to fishery, temperature and distance to urban areas) factors and
indices were mapped to be merged into an integrated, spatially explicit, resilience
index. The result is a consistent modelled regional view of the capacities of individual
reefs to cope with climate change and human induced disturbances.
16.3 Indirect Remote Sensing of Coral Reefs
This section addresses remote sensing of meso-scale hydrodynamic and atmospheric
physical processes around coral reefs. The goal is to characterize how these physical
processes influence the functioning and structures of local communities and habitats.
This idea has been already expressed in the previous section. For instance, Hamylton (2011b) coupled a wave model with habitat maps to characterize community
patterning. The resilience study by Rowlands et al. (2012) also included meso-scale
forcing with an index of thermal stress computed with Moderate Resolution Imaging
Spectroradiometer (MODIS) 1 km Sea Surface Temperature (SST) data. It would be
possible to include here the land and riverine processes that also significantly affect
coastal reefs through run-offs, and even remote reefs at long distance through river
plumes dispersal.
S. Andréfouët
Coral reefs and associated ecosystems such as seagrass meadows are at risk given
the cumulative stress induced by human activities and climate change. Understanding ecosystem resilience is now a priority and a challenge (Nyström et al. 2008).
Resilience capacity results from a combination of numerous processes related to
biodiversity, functional redundancy, connectivity, heterogeneity, and their dynamics
under human pressure (e.g. fisheries) and climate change. Mapping resilience is in its
infancy given our ignorance on many aspects of reef resilience, the high complexity
at stake and the amount of new data needed. However, we identified two relevant
remote sensing approaches to study resilience.
In its simpler form, resilience can be measured by multidate image analysis and
change detection. Across decades, it may be possible to observe cycles of loss of
important communities and their recovery which suggests resilience, as in Scopélitis
et al. (2009) for Saint Leu Reef in La Réunion Island. Long term observations
may also suggest steady degradation and continuous loss of critical habitats, due to
chronic disturbances, precluding the return to previous functional states, e.g. Toliara,
Madagascar (Bruggeman et al.). The temporal dynamics of coral reefs, seagrass and
mangroves habitats have been studied in the Western Indian Ocean with a variety
of image sensors (Makota et al. 2004; Wang et al. 2003; Gullstroem et al. 2006;
Ferreira et al. 2009; Knudby et al. 2010b), sometimes with a multi-sensor, very high
resolution, approach (Scopélitis et al. 2009; Andréfouët et al. 2013) (Fig. 16.5).
Using a more complex modelling design, resilience has been tentatively mapped
throughout Saudi Arabia west coast by Rowlands et al. (2012). They identified first a
number of key variables and indices required to characterize resilience. These indices
were then inferred and mapped from remote sensing using a variety of assumptions,
images and processing. Both “landscape” (related to coral, framework, depth) and
“stress” (related to fishery, temperature and distance to urban areas) factors and
indices were mapped to be merged into an integrated, spatially explicit, resilience
index. The result is a consistent modelled regional view of the capacities of individual
reefs to cope with climate change and human induced disturbances.
16.3 Indirect Remote Sensing of Coral Reefs
This section addresses remote sensing of meso-scale hydrodynamic and atmospheric
physical processes around coral reefs. The goal is to characterize how these physical
processes influence the functioning and structures of local communities and habitats.
This idea has been already expressed in the previous section. For instance, Hamylton (2011b) coupled a wave model with habitat maps to characterize community
patterning. The resilience study by Rowlands et al. (2012) also included meso-scale
forcing with an index of thermal stress computed with Moderate Resolution Imaging
Spectroradiometer (MODIS) 1 km Sea Surface Temperature (SST) data. It would be
possible to include here the land and riverine processes that also significantly affect
coastal reefs through run-offs, and even remote reefs at long distance through river
plumes dispersal.