field data and models from Mumby et al. (2004b) and the mapped habitat proximities to mangroves, to evaluate options for MPA design in Belize.
In much of the developing world, field data are often limited, acquisition of
high spatial resolution satellite data may be cost-prohibitive, and technical
capacity of agency staff for processing images may be low. However, governments
often have archives of aerial photographs available for free or low cost which can
be locally interpreted. For instance, Aswani and Lauer (2006) used local fisherfolk
to trace representative habitats on print-outs of aerial photographs covering the
Roviana and Vonavona regions of the western Solomon Islands. The researchers
found that local knowledge of fish distributions within each habitat was robust
compared with field data and therefore could be used to build MPAs that are
sensitive to local needs (Aswani and Lauer 2006). In fact, in cases where there is
strong local tenure over marine resources, community remote sensing approaches
to resource inventories that match local perceptions may be more readily used for
management applications than maps produced by external agencies.
15.3.2 Predictive Mapping of Fish Assemblages
A relatively new application of coral reef remote sensing data has emerged using
remote sensing-derived information about habitat type and structure to predict
spatial characteristics of reef community fish assemblages. Given the high costs
associated with collecting field data and the difficulty of accessing certain remote
reefs (Mumby et al. 1999), the ability to confidently predict relative differences in
abundance, biomass, species richness and diversity of coral reef fish across broad
spatial scales has obvious advantages for fisheries management and conservation.
Such advantages include improved information for prioritizing locations for
conservation and management through marine spatial planning and the ability to
assess potential changes to fish communities following large disturbances that alter
reef habitat structure.
New advances in passive and active sensor technology with very high spatial
resolution allow quantification of reef habitat and structural variables at a scale to
which fish may respond. Factors that contribute to site-specific composition of
coral reef fish assemblages include: biogeographic patterns (Thresher 1991);
habitat area (Bellwood and Hughes 2001); ecological processes such as recruitment, competition and predation (Carr et al. 2002); environmental stochasticity
(Connolly et al. 2005); and interactions between fish and habitat across a range of
scales (Friedlander and Parrish 1998; Lara and Gonzalez 1998). Coral reef remote
sensing scientists have developed techniques (discussed in previous chapters) to
quantify at least five habitat variables from remote sensing data which are shown
to be related to fish community structure, including: (1) depth; (2) structural
complexity; (3) substrate type; (4) habitat diversity; and (5) live coral cover.
Multiple studies have used bathymetry and topographic complexity derived
from LiDAR to develop predictive relationships of coral reef rugosity with fish
416
S. Jupiter et al.
In much of the developing world, field data are often limited, acquisition of
high spatial resolution satellite data may be cost-prohibitive, and technical
capacity of agency staff for processing images may be low. However, governments
often have archives of aerial photographs available for free or low cost which can
be locally interpreted. For instance, Aswani and Lauer (2006) used local fisherfolk
to trace representative habitats on print-outs of aerial photographs covering the
Roviana and Vonavona regions of the western Solomon Islands. The researchers
found that local knowledge of fish distributions within each habitat was robust
compared with field data and therefore could be used to build MPAs that are
sensitive to local needs (Aswani and Lauer 2006). In fact, in cases where there is
strong local tenure over marine resources, community remote sensing approaches
to resource inventories that match local perceptions may be more readily used for
management applications than maps produced by external agencies.
15.3.2 Predictive Mapping of Fish Assemblages
A relatively new application of coral reef remote sensing data has emerged using
remote sensing-derived information about habitat type and structure to predict
spatial characteristics of reef community fish assemblages. Given the high costs
associated with collecting field data and the difficulty of accessing certain remote
reefs (Mumby et al. 1999), the ability to confidently predict relative differences in
abundance, biomass, species richness and diversity of coral reef fish across broad
spatial scales has obvious advantages for fisheries management and conservation.
Such advantages include improved information for prioritizing locations for
conservation and management through marine spatial planning and the ability to
assess potential changes to fish communities following large disturbances that alter
reef habitat structure.
New advances in passive and active sensor technology with very high spatial
resolution allow quantification of reef habitat and structural variables at a scale to
which fish may respond. Factors that contribute to site-specific composition of
coral reef fish assemblages include: biogeographic patterns (Thresher 1991);
habitat area (Bellwood and Hughes 2001); ecological processes such as recruitment, competition and predation (Carr et al. 2002); environmental stochasticity
(Connolly et al. 2005); and interactions between fish and habitat across a range of
scales (Friedlander and Parrish 1998; Lara and Gonzalez 1998). Coral reef remote
sensing scientists have developed techniques (discussed in previous chapters) to
quantify at least five habitat variables from remote sensing data which are shown
to be related to fish community structure, including: (1) depth; (2) structural
complexity; (3) substrate type; (4) habitat diversity; and (5) live coral cover.
Multiple studies have used bathymetry and topographic complexity derived
from LiDAR to develop predictive relationships of coral reef rugosity with fish
416
S. Jupiter et al.