2.3 WORLD WILDLIFE FUND GLOBAL ECOREGIONS
Conservation planning requires information on the complex distribution of
ecological communities for the identification of areas of unique biodiversity. Digital
datasets classifying biogeographic realms or biomes down to a local ecosystem scale
have been developed for terrestrial ecosystems by the World Wildlife Fund (WWF)
(Olson et al., 2001) with online descriptions and browsing and GIS shapefiles available
for download (http://www.worldwildlife.org/ecoregions). The ecoregions identified
include areas of coastal mangroves, but not salt marshes, beaches, or other major
coastal habitats. WWF’s Global 200 (Olson et al., 2000) is an identified subset of 200
WWF Global Ecoregions classes that are most important for conservation of global
biodiversity. This data set includes freshwater and marine habitats in addition to
terrestrial ecoregions. However, the much more limited knowledge about marine
biodiversity limited the scope of the analysis to five broad habitat types (polar,
temperate shelf and seas, temperate upwelling, tropical upwelling, and tropical coral;
Olson et al., 2000).
One potential problem with global ecoregion datasets such as the WWF products is
that they are by necessity derived from a synthesis of other data sources, and include a
significant number of assumptions that must be considered before the dataset is used in
an “outside” analysis (one for which it was not designed). Often, once a product is
publicly available, it begins to be assimilated into other studies - whether limitations
and assumptions are appropriate to subsequent uses or not. Methodological details and
documentation of these assumptions are frequently not readily available to outside
users. For example, as a biogeographic framework the defined ecoregions of the WWF
Global Ecoregions set are more descriptive of habitat for some taxa (plants, insects)
than for other others. The use of smooth mapped ecoregions oversimplifies boundaries
across ecotones, mosaic habitats and unique embedded habitats (Olson et al., 2001).
Beyond the assumptions implicit in biogeographic approaches, additional assumptions
are inherent in the protocols used in mapping the regions in the first place. Expert
opinions were heavily used in developing the WWF Global 200 and have a large weight
in the decisions taken to prioritize areas for conservation, but expert opinion may be
biased toward better-known sites.
3. Global Datasets Derived from Satellite Remote Sensing
In summarizing the different remote sensing datasets available with global or
nearly global coverage, we group the datasets by their relative spatial resolution.
Subjective terms for spatial resolution often differ among investigators depending on
whether satellite or airborne platforms are the norm. In this chapter, we try to use a
manager’s perspective which would consider high spatial resolution studies as those
that have fine enough grain to distinguish coastal habitats. Thus, we define high spatial
resolution as ≤ 10 meter mapping units (pixels), moderate spatial resolution as 30-100
meter mapping units, and low spatial resolution as ≥ 500 meter mapping units.
The general use of satellite remote sensing sensors in tropical coastal management
was reviewed by Green et al. (1996, 2000). Sensors and techniques for mangrove
mapping (Green et al., 1997, 1998) have also been reviewed. The most recent reviews
of the many advances in remote sensing of coral reefs are by Mumby et al. (2004),
Andréfouët and Riegl (2004), and Andréfouët et al. (2004). Here, we summarize only
287
Data Synthesis for Management
Conservation planning requires information on the complex distribution of
ecological communities for the identification of areas of unique biodiversity. Digital
datasets classifying biogeographic realms or biomes down to a local ecosystem scale
have been developed for terrestrial ecosystems by the World Wildlife Fund (WWF)
(Olson et al., 2001) with online descriptions and browsing and GIS shapefiles available
for download (http://www.worldwildlife.org/ecoregions). The ecoregions identified
include areas of coastal mangroves, but not salt marshes, beaches, or other major
coastal habitats. WWF’s Global 200 (Olson et al., 2000) is an identified subset of 200
WWF Global Ecoregions classes that are most important for conservation of global
biodiversity. This data set includes freshwater and marine habitats in addition to
terrestrial ecoregions. However, the much more limited knowledge about marine
biodiversity limited the scope of the analysis to five broad habitat types (polar,
temperate shelf and seas, temperate upwelling, tropical upwelling, and tropical coral;
Olson et al., 2000).
One potential problem with global ecoregion datasets such as the WWF products is
that they are by necessity derived from a synthesis of other data sources, and include a
significant number of assumptions that must be considered before the dataset is used in
an “outside” analysis (one for which it was not designed). Often, once a product is
publicly available, it begins to be assimilated into other studies - whether limitations
and assumptions are appropriate to subsequent uses or not. Methodological details and
documentation of these assumptions are frequently not readily available to outside
users. For example, as a biogeographic framework the defined ecoregions of the WWF
Global Ecoregions set are more descriptive of habitat for some taxa (plants, insects)
than for other others. The use of smooth mapped ecoregions oversimplifies boundaries
across ecotones, mosaic habitats and unique embedded habitats (Olson et al., 2001).
Beyond the assumptions implicit in biogeographic approaches, additional assumptions
are inherent in the protocols used in mapping the regions in the first place. Expert
opinions were heavily used in developing the WWF Global 200 and have a large weight
in the decisions taken to prioritize areas for conservation, but expert opinion may be
biased toward better-known sites.
3. Global Datasets Derived from Satellite Remote Sensing
In summarizing the different remote sensing datasets available with global or
nearly global coverage, we group the datasets by their relative spatial resolution.
Subjective terms for spatial resolution often differ among investigators depending on
whether satellite or airborne platforms are the norm. In this chapter, we try to use a
manager’s perspective which would consider high spatial resolution studies as those
that have fine enough grain to distinguish coastal habitats. Thus, we define high spatial
resolution as ≤ 10 meter mapping units (pixels), moderate spatial resolution as 30-100
meter mapping units, and low spatial resolution as ≥ 500 meter mapping units.
The general use of satellite remote sensing sensors in tropical coastal management
was reviewed by Green et al. (1996, 2000). Sensors and techniques for mangrove
mapping (Green et al., 1997, 1998) have also been reviewed. The most recent reviews
of the many advances in remote sensing of coral reefs are by Mumby et al. (2004),
Andréfouët and Riegl (2004), and Andréfouët et al. (2004). Here, we summarize only
287
Data Synthesis for Management