22.5 Examples of the Use of Ecological Land Classifications in Ecological Assessments
331
22.5 Examples of the Use of
Ecological Land
Classifications in Ecological
Assessments and
Performance Evaluation
In this section, several examples of uses of ELCs
are presented to illustrate the wide variety of specific purposes and spatial scales for which ELCs
have been applied. Section 22.3 included a discussion of the use of ELCs in the ICBEMP for defining the areas needed to implement this ecoregionbased assessment. In addition, selected results from
the ICBEMP assessment are described in Chapter
34. The examples presented here are derived from
four areas of application of ELCs, which cover the
range of spatial scales defined in Table 22.2. ELCs
have often been used to address particular management practices (e.g., silviculture) at fine scales.
For example, the ecosystem classification developed in British Columbia (Banner et aI., 1996) was
designed to guide the management of small units
within a forested landscape (MacKinnon et aI.,
1992), among other goals. Such uses are not discussed further in this section; examples of ELCbased management practices are described in Sims
et aI. (1996). Finally, examples of ELC performance evaluation are discussed.
The first example concerns the use of higherlevel ELUs to stratify areas on a continental scale
for conservation. As part of a recent trend in conducting biodiversity assessments at continental
andlor regional scales with ELC frameworks (TNC,
1996; Noss, 1999; Soule and Terborgh, 1999), a
conservation assessment ofthe status and threats to
biodiversity in North America used 110 terrestrial
ecoregions covering the continental United States
and Canada (Ricketts et aI., 1999). The ecoregions
are based on three established ecoregional frameworks (ESWG, 1995; Gallant et aI., 1995; Omernik, 1995). Information on various aspects of biodiversity (distribution of major habitat types,
number of species, endemic species, and rare ecological elements) and on conservation status (habitat loss, remaining habitats, landscape fragmentation, and protection) was considered to set
broad-scale conservation priorities. Other aspects
of the use of ELCs for conservation are discussed
in Chapter 20.
A second group of examples comprises the use
of ELCs to identify restoration targets and strategies for restoration of terrestrial and wetland
ecosystems. Palik et aI. (2000) used the ELUs of
an ELC at the site-landscape scale to predict PV
of disturbed terrestrial ecosystems in a 11,400-ha
landscape in southwestern Georgia, USA. Higherlevel ELUs were combined with physical (geomorphology, soil characteristics) and land-use data
to predict PV distribution within lower-level ELUs
and to assess their disturbance levels. Disturbance
level was combined with information on the conservation status of each ELU to prioritize restoration efforts. At the regional scale, Detenbeck et aI.
(1999) applied an ecoregional framework to evaluate perturbations in wetland ecosystems and to develop restoration strategies for these wetlands in
the Great Lakes region, USA. Wetland distribution,
land-use pattern, and disturbance-level variation
were determined; ecoregions exhibited characteristic patterns of variation in these attributes. General
strategies and goals for wetland restoration at the
ecoregional scale were derived from information
on historic wetland distribution patterns and the
distributions of species of special concern, which
depend on specific wetland types or mosaics of
habitat types.
EAs often require precise information on the distribution of species for many analyses. However,
although general species range maps may exist,
precise species distribution maps are seldom available. In the third set of examples, ELCs were used
in innovative ways to predict the distributions of
species at the regional scale. In the United Kingdom, Cherrill et aI. (1995) combined a landscape
classification with land cover and plant community
maps to predict the probability of occurrence of 579
species. The model performed well for common native species, but relatively poorly for rare and introduced species. He et aI. (1998) used a combination of abiotic ELC stratification, a satellite-derived
land cover map, and plot-specific information to
predict the distribution of selected tree species and
age classes in northern Wisconsin, USA. Within
each ecoregion defined by the abiotic ELC, fieldinventory data were aggregated to provide information on secondary and subcanopy tree species
occurrence and age distributions. A probabilistic algorithm was derived to assign information from a
point coverage (forest inventory sampling points)
and a polygon coverage (ecoregion boundaries) to
a raster map (satellite land cover classification).
This method was the basis for assessment of forest
patterns across regional landscapes and the potential for restoration at the ecoregion scale.
The fourth set of application examples incorporates the use of ELCs for nonbiological purposes.
Bhat et al. (1998) employed an ecoregional approach to estimate the economic value of land- and
331
22.5 Examples of the Use of
Ecological Land
Classifications in Ecological
Assessments and
Performance Evaluation
In this section, several examples of uses of ELCs
are presented to illustrate the wide variety of specific purposes and spatial scales for which ELCs
have been applied. Section 22.3 included a discussion of the use of ELCs in the ICBEMP for defining the areas needed to implement this ecoregionbased assessment. In addition, selected results from
the ICBEMP assessment are described in Chapter
34. The examples presented here are derived from
four areas of application of ELCs, which cover the
range of spatial scales defined in Table 22.2. ELCs
have often been used to address particular management practices (e.g., silviculture) at fine scales.
For example, the ecosystem classification developed in British Columbia (Banner et aI., 1996) was
designed to guide the management of small units
within a forested landscape (MacKinnon et aI.,
1992), among other goals. Such uses are not discussed further in this section; examples of ELCbased management practices are described in Sims
et aI. (1996). Finally, examples of ELC performance evaluation are discussed.
The first example concerns the use of higherlevel ELUs to stratify areas on a continental scale
for conservation. As part of a recent trend in conducting biodiversity assessments at continental
andlor regional scales with ELC frameworks (TNC,
1996; Noss, 1999; Soule and Terborgh, 1999), a
conservation assessment ofthe status and threats to
biodiversity in North America used 110 terrestrial
ecoregions covering the continental United States
and Canada (Ricketts et aI., 1999). The ecoregions
are based on three established ecoregional frameworks (ESWG, 1995; Gallant et aI., 1995; Omernik, 1995). Information on various aspects of biodiversity (distribution of major habitat types,
number of species, endemic species, and rare ecological elements) and on conservation status (habitat loss, remaining habitats, landscape fragmentation, and protection) was considered to set
broad-scale conservation priorities. Other aspects
of the use of ELCs for conservation are discussed
in Chapter 20.
A second group of examples comprises the use
of ELCs to identify restoration targets and strategies for restoration of terrestrial and wetland
ecosystems. Palik et aI. (2000) used the ELUs of
an ELC at the site-landscape scale to predict PV
of disturbed terrestrial ecosystems in a 11,400-ha
landscape in southwestern Georgia, USA. Higherlevel ELUs were combined with physical (geomorphology, soil characteristics) and land-use data
to predict PV distribution within lower-level ELUs
and to assess their disturbance levels. Disturbance
level was combined with information on the conservation status of each ELU to prioritize restoration efforts. At the regional scale, Detenbeck et aI.
(1999) applied an ecoregional framework to evaluate perturbations in wetland ecosystems and to develop restoration strategies for these wetlands in
the Great Lakes region, USA. Wetland distribution,
land-use pattern, and disturbance-level variation
were determined; ecoregions exhibited characteristic patterns of variation in these attributes. General
strategies and goals for wetland restoration at the
ecoregional scale were derived from information
on historic wetland distribution patterns and the
distributions of species of special concern, which
depend on specific wetland types or mosaics of
habitat types.
EAs often require precise information on the distribution of species for many analyses. However,
although general species range maps may exist,
precise species distribution maps are seldom available. In the third set of examples, ELCs were used
in innovative ways to predict the distributions of
species at the regional scale. In the United Kingdom, Cherrill et aI. (1995) combined a landscape
classification with land cover and plant community
maps to predict the probability of occurrence of 579
species. The model performed well for common native species, but relatively poorly for rare and introduced species. He et aI. (1998) used a combination of abiotic ELC stratification, a satellite-derived
land cover map, and plot-specific information to
predict the distribution of selected tree species and
age classes in northern Wisconsin, USA. Within
each ecoregion defined by the abiotic ELC, fieldinventory data were aggregated to provide information on secondary and subcanopy tree species
occurrence and age distributions. A probabilistic algorithm was derived to assign information from a
point coverage (forest inventory sampling points)
and a polygon coverage (ecoregion boundaries) to
a raster map (satellite land cover classification).
This method was the basis for assessment of forest
patterns across regional landscapes and the potential for restoration at the ecoregion scale.
The fourth set of application examples incorporates the use of ELCs for nonbiological purposes.
Bhat et al. (1998) employed an ecoregional approach to estimate the economic value of land- and