330
S. Andréfouët
physical environment. These products are then used to provide habitat inventories,
design better conservation plans, optimize sampling strategies for monitoring and
for biological species inventories, assess fishery stocks, model their sustainability,
and identify areas vulnerable to climate change (Andréfouët 2008, 2011).
We remind that the vast majority of biological coral reef information come from
field surveys and monitoring programs. All data for model parameterization and
validation, as well as data used to interpret remote sensing images, will come from the
field. Designing sampling strategy that represents exhaustively the diversity found
in an ecological system is difficult without remote sensing information and some
sort of clear protocol for allocation of efforts. However, to date, few monitoring
sampling strategies seem to have taken advantage of remote sensing data to optimize
the representativity of the sampling as advocated by Scopélitis et al. (2010). High
resolution images offer a powerful mean to select representative monitoring sites and
to design adaptive monitoring strategies but this remain underused.
In the Red Sea and Western Indian Ocean, spatial applications dedicated to coastal
management remain scarce, irrespective of the origin of the spatial data (remote sensing or interpolation of field data, as in Riegl and Piller 2000). No mentions of coral
reef fishery stock assessments conducted using remotely sensed habitat maps could
be found in the peer reviewed literature (Hamel and Andréfouët 2010). However,
finfish, sea cucumber and several other species important for subsistence and commercial fisheries are targeted throughout the region. Also, there are marine protected
areas in both the Red Sea and Western Indian Ocean, and traditional fisheries in
Kenya, Tanzania and Madagascar are under the scrutiny of scientists (McClanahan
et al. 2009; Barnes and Rawlinson 2009; McClanahan et al. 2011). Yet, no spatial resource management plans have been enhanced with remote sensing products to date,
but see work in progress in the French Iles Eparses in the Mozambique Channel
(Grellier et al. 2012).
Beyond what may be considered best as long term management perspectives, several direct remote sensing studies had immediate potential for enhanced management,
such as the demonstration of human impacts on reefs due to urban developments
and other activities. Change detection analysis makes a powerful diagnostic tool if
changes visible on images can be correlated to human activities. Moufaddal (2005)
demonstrated the human infringements on local coral reefs on the Egyptian coast
with a time series of Landsat TM and ETM +. Furthermore, Vanderstraete et al.
(2005) proposed reef risk indices after mapping changes with Landsat and potential
sources of disturbances.
Recently, to support the plans of the Government of Madagascar to increase marine
protected area coverage by over one million hectares, Allnut et al. (2012) compared
four methods for marine spatial planning of Madagascar’s west coast. Among the four
methods, two used spatial conservation target-based optimization software (Marxan
and Zonation), now a typical approach for marine conservation planning. Input
data was drawn from the following variables: fishing pressure, exposure to climate
change, and biodiversity (habitats, species distributions, biological richness, and
biodiversity value). Habitat data came from the geomorphological Landsat-derived
coral reef atlas provided by Andréfouët et al. (2009) and exposure to climate index
was draw from the Maina et al. (2008) database (mentioned in the indirect remote
S. Andréfouët
physical environment. These products are then used to provide habitat inventories,
design better conservation plans, optimize sampling strategies for monitoring and
for biological species inventories, assess fishery stocks, model their sustainability,
and identify areas vulnerable to climate change (Andréfouët 2008, 2011).
We remind that the vast majority of biological coral reef information come from
field surveys and monitoring programs. All data for model parameterization and
validation, as well as data used to interpret remote sensing images, will come from the
field. Designing sampling strategy that represents exhaustively the diversity found
in an ecological system is difficult without remote sensing information and some
sort of clear protocol for allocation of efforts. However, to date, few monitoring
sampling strategies seem to have taken advantage of remote sensing data to optimize
the representativity of the sampling as advocated by Scopélitis et al. (2010). High
resolution images offer a powerful mean to select representative monitoring sites and
to design adaptive monitoring strategies but this remain underused.
In the Red Sea and Western Indian Ocean, spatial applications dedicated to coastal
management remain scarce, irrespective of the origin of the spatial data (remote sensing or interpolation of field data, as in Riegl and Piller 2000). No mentions of coral
reef fishery stock assessments conducted using remotely sensed habitat maps could
be found in the peer reviewed literature (Hamel and Andréfouët 2010). However,
finfish, sea cucumber and several other species important for subsistence and commercial fisheries are targeted throughout the region. Also, there are marine protected
areas in both the Red Sea and Western Indian Ocean, and traditional fisheries in
Kenya, Tanzania and Madagascar are under the scrutiny of scientists (McClanahan
et al. 2009; Barnes and Rawlinson 2009; McClanahan et al. 2011). Yet, no spatial resource management plans have been enhanced with remote sensing products to date,
but see work in progress in the French Iles Eparses in the Mozambique Channel
(Grellier et al. 2012).
Beyond what may be considered best as long term management perspectives, several direct remote sensing studies had immediate potential for enhanced management,
such as the demonstration of human impacts on reefs due to urban developments
and other activities. Change detection analysis makes a powerful diagnostic tool if
changes visible on images can be correlated to human activities. Moufaddal (2005)
demonstrated the human infringements on local coral reefs on the Egyptian coast
with a time series of Landsat TM and ETM +. Furthermore, Vanderstraete et al.
(2005) proposed reef risk indices after mapping changes with Landsat and potential
sources of disturbances.
Recently, to support the plans of the Government of Madagascar to increase marine
protected area coverage by over one million hectares, Allnut et al. (2012) compared
four methods for marine spatial planning of Madagascar’s west coast. Among the four
methods, two used spatial conservation target-based optimization software (Marxan
and Zonation), now a typical approach for marine conservation planning. Input
data was drawn from the following variables: fishing pressure, exposure to climate
change, and biodiversity (habitats, species distributions, biological richness, and
biodiversity value). Habitat data came from the geomorphological Landsat-derived
coral reef atlas provided by Andréfouët et al. (2009) and exposure to climate index
was draw from the Maina et al. (2008) database (mentioned in the indirect remote