CONSERVATION PLANNING AND ASSESSMENT OF IRREPLACEABILITY AND
VULNERABILITY OF CONSERVATION SITES IN THE HEART OF THE WEST REGION, MIDDLE
ROCKIES
Allison L.
Jones1, Katherine Daly2, Erik Molvar3 and
James Catlin1
KEY
WORDS: Conservation Planning, Wyoming Basins, Irreplaceability, Vulnerability,
SITES
ABSTRACT
We conducted a Geographic Information System (GIS) conservation
assessment of the “Heart of the West” region, which incorporates the Wyoming
Basins Ecoregion. This systematic assessment utilized three widely accepted
tracks of conservation planning: representing habitat variation across the
landscape, incorporating special elements, and securing important habitat for
focal species (Noss 2003). Only 4.1% of our study area is currently in some form
of protective federal status (i.e. GAP 1 or 2 status lands), and even these
holdings fall short of protecting much of the biological diversity of the study
area or representing all vegetative communities. Using a simulated annealing
site-selection algorithm, and with static focal species habitat models and
special element data (i.e. Natural Heritage species locations) as inputs, we
used quantitative conservation targets to identify unprotected areas (or
proposed core areas) within the region. If these core areas are protected as a
conservation network, it will serve to meet various goals such as: restoring
viable populations of native plants and animals; protecting sufficient amounts
of all habitat types from further degradation and loss; reducing habitat
fragmentation and restoring functional connectivity; and protecting and
restoring ecological and evolutionary processes. The final proposed network
included a total of 8,387,190 ha (or 44.9 % of
the study area) in the conservation network portfolio, which included all
existing protected areas. We conducted an irreplaceability and vulnerability
analysis of 28 large core areas in the proposed conservation network, based on
six criteria relating to quantitative conservation goals. This analysis will
aid decision-makers by identifying those sites that will contribute most to
explicit conservation goals.
INTRODUCTION
The past few decades have seen a marked increase in the number of
systematic, large-scale conservation planning endeavors meant to identify new
reserves (Bennett and Wit 2001, Noss et al. 2002, Groves 2003, Redford et al. 2003). Planning that is systematic in nature (i.e. the use of defined
planning units, explicit goals and reserve selection algorithms) may be
preferable to more traditional methods that tend to be more opportunistic and
biased in nature, and can result in a sub-optimal placement of reserves for
promoting long-term maintenance of biodiversity (Margules and Pressey 2000,
Scott et al. 2001). Among the key
elements of systematic conservation planning are explicit goals and objectives,
quantitative and substantiated conservation targets, well-documented and
replicable methods, rigorous peer review, and strong criteria for
implementation (Noss 2003).
The Wyoming Basins Ecoregion (Bailey 1995
as modified by The Nature Conservancy 2001a), with connections to both the Utah
High Plateaus Ecoregion and Southern Rockies Ecoregion, represents a region
that is currently not receiving adequate conservation attention (Freilich et
al. 2001). While many conservation
organizations and initiatives are focused on the greater Yellowstone Ecosystem
to the north and the Colorado Rockies to the south, this area in between has a
lower conservation profile and is increasingly suffering impacts from America’s
recent push for large-scale fossil fuel development. Much of the region is open for this use, as only 4.1% of our
study area is currently in some form of protective federal status (i.e. GAP 1
or 2 status lands). Although the
Wyoming Basins Ecoregion does not support an exceptionally high diversity of
species and is thus not considered to be a “hotspot” on a continental or global
level, it does represent not only an opportunity to conserve a still relatively
intact temperate ecosystem, but is also perhaps the most important stronghold
for sagebrush steppe species, including rare and declining species such as sage
grouse (Centrocercus urophasianus), white-tailed prairie dog (Cynomys
leucurus), black-footed ferret (Mustela nigripes), and burrowing owl
(Athene cunicularia).
We carried out a conservation
assessment of this region, which, combined with adjacent habitats in the Utah
High Plateaus Ecoregion and Southern Rockies Ecoregion, we refer to as the
“Heart of the West” study area. Our
assessment drew upon four conservation goals that have become commonplace with
large scale conservation planning (Noss and Cooperrider 1994): (1) viable
populations of all native plants and animals (including some that have been
extirpated) are protected and restored; (2) sufficient amounts of all habitat
types protected from further degradation and loss; (3) ecological and
evolutionary processes are protected and restored, and (4) land is protected
from further fragmentation so that functional connectivity can be restored,
resulting in a conservation network that is more resilient to environmental
change.
In order
to achieve these goals in our final product (a conservation network), we
incorporate three accepted approaches to developing a conservation assessment
(Noss et al. 1999, Foreman et al 2003, Miller et al. 2003, Rumsey et al. 2003):
(1) representation of all land cover, or vegetative community, types within a
network of core areas (Groves et al. 2000, Groves 2003); (2) identification and
protection of special elements such as rare species occurrences (Noss and
Coopeerrider 1994, Groves et al. 2000); and (3) protection and linking of key
habitat of focal species that serve critical ecosystem roles and/or whose
presence is indicative of healthy, functioning systems (Miller et al. 1998,
Soulé and Terborgh 1999). We chose this
three-track approach because there are weaknesses inherent in each of the three
methods (e.g. focal species analyses, see Bonn et al. 2002, Lindenmayer et al.
2002), and relying on only one or two of these approaches may not provide
sufficient protection for a large region (Noss et al. 2002, Carroll et al.
2003, Noss 2003). Together, the three
tracks of representation analysis, special element mapping, and focal species
analysis offer a comprehensive and effective approach to conservation planning.
We
identified core areas in the Heart of the West region that best represent the
selected set of conservation elements (i.e. Natural Heritage Program species
occurrences) at chosen target (goal) levels by using a simulated annealing
algorithm applied to the conservation elements. To identify priority core areas for immediate conservation
campaigns, we ranked each core area in terms of its irreplaceability and vulnerability
(Pressey and Cowling 2001, Noss et al. 2002).
The concept of irreplaceability provides a measure of the relative
contribution different core areas make to reaching overall conservation goals,
while a measure of vulnerability reveals the degree of current and future
threats to individual core areas and helps determine which cores are in urgent
need of immediate protection. We
recommend that regional conservationists and land-use planners give the highest
priority to those Heart of the West core areas that score high for both
irreplaceable biological value and high degree of vulnerability to human
threats.
Study Area
Our study area includes the
Wyoming Basins Ecoregion, as well as connecting lowland and upland habitat in
the adjacent Southern Rockies and Utah High Plateaus Ecoregions (Figure
1). We aimed to ensure some
landscape-scale connections to these ecoregions[1]
ensure long-term connectivity of wide-ranging focal species between the
northern Rockies and southern Rockies.
The landscapes and vegetation of the Heart of the West study area are
considerably diverse, as the region spans numerous life zones (Knight
1994). The bulk of our study area, the
Wyoming Basins Ecoregion, is dominated by sagebrush (Artemesia tridentata) steppe and various species of
bunchgrass. Moving into the uplands of
the Heart of the West that surround the basins, Rocky Mountain juniper (Juniperus scopulorum) delineates the
lower boundary between shrub zones and coniferous zones. Douglas fir (Pseudotsuga menziesii) transitions to lodgepole pine (Pinus contorta) at higher
elevations. Two thirds of the rare
plants endemic to Wyoming are found within the Heart of the West (Wyoming
Natural Diversity Database 2005) in part due to this impressive diversity of
life zones in the region.

FIGURE 1. Heart of
the West study area used for conservation assessment.
The SITES model
The enormity of the task of
delineating core areas and linkages (Harris 1984) given our numerous targeted
conservation elements made a manual approach to this conservation assessment
very difficult. We used SITES (version
1.0, Andelman et al. 1999) to delineate core areas and landscape linkages
within our study area by assembling and comparing alternative portfolios of
planning units. SITES allows the user
to identify an initial set of landscape patches that best represent a selected
set of conservation elements at chosen target levels. The SITES model attempts to minimize reserve design “cost” while
maximizing attainment of conservation goals in a compact set of core areas.
This set of objectives constitutes
the “objective cost function,” in which:
Cost = Area + Penalty +
Boundary Length
where Cost is the objective (i.e.
for core areas to be minimized), Area is the number of hectares in all cores,
Penalty is a cost imposed for failing to meet conservation target goals, and
Boundary Length is a cost determined by the total boundary length of all core
areas (thus causing core areas to be more compact and maximizing core to
exterior ratios).
We used
15,642 hexagonal planning units of 1,250 hectares (ha) each. A hexagonal unit is preferable to other
shapes or entities (e.g. square cells or watershed boundaries). It provides a
relatively smooth output (as compared with square cells), approximates a circle
- which has a low edge-to-area ratio, and the unit size remains constant. This reduces the likelihood of the algorithm
rejecting larger units such as larger watersheds which might be interpreted as
having a greater cost than smaller units (Wilmer, in prep). We chose
the 1,250 ha. size based on a sensitivity analysis that compared SITES results
with different sized planning units (as described in Jones et al. 2004).
We set the
SITES penalty value for all conservation elements at 1.0 (the potential range
is from 0 to 1.0) so that each planning unit was equivalent in terms of cost
and the algorithm was unconstrained in selecting where to achieve its
goals. This ensured that only areas
with less targets would be viewed as having “higher costs.” This increased the chances that all conservation
elements had an equal chance of being represented in the final solution at
approximately the levels for which we targeted them.
Each time
the SITES model is run, it performs 1,000,000 iterative attempts to identify
the minimum cost solution per run. We
ran the model with different variations of the SITES input parameters, such as
varying the boundary-length modifier to achieve different degrees of planning
unit “clumping.” We also adjusted some
of our conservation target goals (such as amount of Natural Heritage species
locations we wanted in solution) in these test runs and assessed the
effectiveness of the final solution in terms of capturing all of our
representation, special element, and focal species target goals within the
Heart of the West (Noss et al. 2002, Miller et al. 2003). The final SITES solution adopted as the proposed
conservation network was the one that we believed to best meet all of our
conservation goals for the region.
Representation Analysis
The actual level of representation
necessary to ensure, when adequately protected, persistence of any given land
cover type depends on many different variables, including the overall area
occupied by each land cover type, and the degree of connectivity of the land
cover type. Noss and Cooperrider (1994)
observed that, “science cannot tell us precisely how many times or in what size
reserves each...ecosystem type must be represented to be viable.” We propose, therefore, that representation
percentages be used to identify elements that may be relatively
under-represented within the proposed network, and not to speculate what level
of representation would provide for viability and persistence of all
communities within the network. We
utilized a 25% representation goal for all landcover types as often recommended
by The Nature Conservancy (Groves et al. 2000) and used by Noss et al. (2002)
for the Utah-Wyoming Mountains Ecoregional Plan.
Special elements
We assembled Natural Heritage Program (NHP) element occurrence data for the study area from the state Heritage Programs in Wyoming, Montana, Colorado, Utah and Idaho (Figure 2). We included 161 plant and 42 animal species in our final list of target species. The animal targets included 13 mammals, 7 reptiles, 3 amphibians, 13 birds, and 6 species of fish. This list was derived from The Nature Conservancy’s target list for the Wyoming Basins Ecoregion, plus all S1 (critically imperiled at the state level) and S2 (imperiled at the state level) species within the Book Cliffs in Utah and the portion of the Southern Rockies in Colorado and Wyoming that fall within our study area. Using SITES, we targeted 100% of G1 (critically imperiled globally) and G2 (imperiled globally) occurrences, and 25%-75% of all species occurrences of lower rank to be included in cores and linkages. We selected these target levels based on those used by Noss et al. (2002) in the Utah-Wyoming mountains (i.e. 100% of G1 and G2 species), as well as the number of total occurrences in the study area, and target goals set for the same species by Frelilich et al. (2001) in the Wyoming Basins. Using The Nature Conservancy guidelines (Comer 2001, The Nature Conservancy 2001b), we corrected for unequal survey efforts (and tbus over-representation in the SITES solution) for NHP species by capping targets at 25 occurrences. This was important to do in light of some very high profile species such as federally listed species that are surveyed for far more often than most other species in the region.

FIGURE 2. Natural Heritage Program (NHP) element occurrence
data for the study area from state Heritage Programs.
Stream
reaches containing conservation populations of cutthroat trout were another
special element used in the SITES analysis.
We chose this element because they are indicator species (Behnke 2002)
and because of the importance of cutthroat trout to stream ecosystems, the
severe decline of native trout species in the region, and also because this
element was not likely to have been covered by the (terrestrial) focal species
analyses and land cover representation analysis. We targeted 100% of all occurrences of cutthroat trout stream
segments for inclusion in a conservation network.
Due to the
important conservation value of roadless areas (Hitt and Frissell 1999, Wilcove
et al. 2000, Strittholt and DellaSala 2001), we chose to include 100% of
remaining roadless areas in the final network.
These roadless areas are comprised of all citizen-inventoried and
agency-inventoried roadless areas and all lands with GAP 1 status (Figure 3).
Focal species analysis
The focal species utilized in this
conservation assessment is described in Jones et al. (2004). The suite was selected by an advisory
science team, with refinement of the focal species list following the completion
of natural history literature reviews for each species and an expert peer
review process. A subset of the larger
suite of species, namely the gray wolf (Canis lupus) and sage grouse
were used for modeling habitat affinities, needs, and threat susceptibility
that could be extrapolated to other species using similar habitats (wolf and
sage grouse represent the habitats that are most prevalent in the study
area). Using ESRI ArcInfo and ESRI
ArcView, we constructed raster-based (using
either 200m2 or 1 km2 cells) static habitat suitability
models for both sage grouse and wolf (Jones et al. 2004). Habitat data used in these models included
vegetation type, slope, aspect, elevation, important prey habitat, and proximity
to streams. Habitat threats we included
in the models were road density (for wolf model) and oil and gas development
(for sage grouse model).

FIGURE 3. Roadless areas in study area
The
results of the sage grouse and wolf habitat suitability models were used as
inputs into the SITES analysis. We
targeted 100% of all top-scoring sage grouse habitat, and 25% of areas that
scored in the second highest sage grouse habitat category to be included in
cores and linkages. We targeted 75% of all top-scoring wolf habitat, and 25% of
areas that scored in the second highest wolf habitat category to be included in
cores and linkages.
We set these particular targets based on trial SITES runs, expert
opinion and focal species target goals used in other conservation network
designs (i.e. Miller et al. 2003).
Finalizing Linkages
SITES also reported how often each planning unit was
included in the initial set of cores during the trial runs. This information helped us then delineate
linkages after the cores were chosen.
Even if these potential “connecting units” sometimes were not included
in the final solution model, if they were selected once that meant they likely
included one or more of our targets.
This justified using these planning units as building blocks to
construct linkages between cores, along with natural ungulate migration routes,
stream linkages, and lightly roaded, public land. When stream or river linkages were used as the basis for
linkages, they were used as the backbone of the linkage, with at least a 0.5 km
buffer on either side of the stream or river.
Expert Assessment
Quantitative data on which to
evaluate conservation priorities are always limited. Thus, we recognized that the SITES analysis would need to be
supplemented by expert opinion.
Practitioners, local scientists and conservationists can provide
valuable and often undocumented information on conservation elements, important
habitats, threats and feasibility of site protection. In addition to providing key information, involvement of experts
can simultaneously help develop strong partnerships, provide necessary peer
review, and generally help garner acceptance and credibility of the final
conservation network (Noss 2003). Types
of expert assessment utilized in this process included expert peer review by
regional scientists of focal species accounts, focal species habitat models,
and the final Heart of the West Conservation Plan (Jones et al. 2004), as well
as expert workshops. Workshops were
conducted with scientists, local conservationists, and Natural Heritage Program
and TNC staff to gather input on threats, additional information on
conservation elements and important habitats, and placement of cores and
linkages. The input and advice
generated through these peer review processes were used to fine-tune the final boundaries
of core areas and linkages, as well as improve the Conservation Plan.
Irreplaceability and
Vulnerability Analysis
A key
concept in conservation planning is irreplaceability (Pressey and Cowling
2001). Irreplaceability provides a
measure of the relative contribution different core areas make to reaching
overall conservation goals, thus helping planners prioritize protection for
various core areas in a conservation network.
Irreplaceability can be described in two ways: (1) the likelihood that a
particular area is needed to achieve an explicit conservation goal, or (2) the
extent to which the options for achieving an explicit conservation goal are
narrowed if an area is not conserved. A
core area that ranks high in terms of irreplaceability is essential to meeting
a particular goal; an example would be a core area that contains the only known
occurrence of a species in the region.
Conversely, a core with a very low irreplaceability value might have a
number of replacements.
We
assigned irreplaceability values to core areas based on their contribution to
the goal of (1) protecting Natural Heritage Program species in the study area,
(2) protecting 100% of stream segments with Conservation Populations of
cutthroat trout, (3) protecting 100% of all roadless areas, (4) representing at
least 25% of each land cover type, (5) protecting 75% of top scoring wolf
habitat, and 25% of second highest scoring wolf habitat, and (6) protecting
100% of best-scoring class of sage grouse habitat, and 25% of second
best-scoring class of sage grouse habitat in the study area.
To allow
for direct comparison of the ability of core areas to meet goals for multiple
targets, we first normalized the quantity of any particular by dividing the
amount by the standard deviation, and then calculated a standard Z-score for
each core area based on how well that core area did in capturing conservation
elements (Jeo 2002). The Z-score was
calculated for the six categories of targets (described above) within a core
area. This procedure allowed us to
directly compare the number of targets in each core using meaningful units,
since the mean Z-score for the entire study area is, by definition,
approximately 0, and 1 unit represents one standard deviation from the mean
value. Using Z-scores allows values to
be combined such that each target receives equal weight and with explicit
consideration of the relative rarity of any target. For example, in order to rank core areas based on NHP data, we combined
all NHP targets into a single index score.
This method was particularly helpful in determining irreplaceability of
small core areas compared to large core areas.
Another
key consideration in conservation planning is threat or vulnerability (Margules
and Pressey 2000). Understanding the
current and future threats to individual core areas helps determine which cores
are in urgent need of immediate protection, and can help conservationists
prioritize core areas for attention while also developing specific strategies
and conservation plans for cores. Based
on the significant threats to core areas (and the wildlife they support) posed
by oil and gas development (Comer 1982, Van Dyke and Klein 1996, Lyon
2000, Ingelfinger 2001, Weller et al. 2002), and considering the use of road
density as a secondary metric of human threats to core areas (Brattstrom and
Bondello 1983, Mech et al. 1988, Hobbs and Huenneke 1992, McIntyre and Lovoral
1994, Mace et al. 1996, Gelbard and Belnap 2003) we calculated Z-scores
for the vulnerability of each core based on road density in that core, current
oil and gas well density in the core, and degree of future oil and gas activity
threatening the core area.
Based on
Z-score analysis, we assigned a vulnerability score of 0-100 to each core
area. Core areas were then plotted on a
graph of irreplaceability (y axis) versus vulnerability (x-axis) and the graph
divided into four quadrants (Margules and Pressey 2000, and Noss et al. 2002).
The upper right quadrant, which includes core areas with high irreplaceability
and high vulnerability, comprises the highest priority core areas for
conservation.
The final conservation network
based on what we determined to be the final SITES run identified the most
efficient and compact portfolio of core areas and linkages (Figure 4)
containing the 247 targets (including all focal species targets, special
elements, representation analysis) at, above, or very close to the pre-assigned
target levels (Table 1). To increase
connectivity between core areas, we added a few planning units by hand to the
final solution in areas representing key linkages. Whenever possible these manually added planning units followed
perennial watercourses and known important migration linkages for pronghorn (Antilocapra
Americana), deer (Odocoileus hemionus),
and elk (Cervus elaphus).
The final
conservation network solution included a total of 8,387,190 ha (or 44.9 % of
the study area) in the network portfolio.
The final network included all existing protected areas (i.e. GAP1
lands) because all of these units are roadless and the model incorporated 100%
of roadless areas into the conservation network. As the SITES model did not distinguish private lands from public
lands when assembling the conservation portfolio, private lands are amply
represented (29% of the network, Figure 5).
For planning purposes, we identified and named 28 of the larger core
areas in the network (Figure 4). These
key core areas range from 32,500 to 1,367,000 hectares in size. Overall, this assemblage of proposed
conservation sites met our goals represented by our three conservation tracks
(representation, special elements, and focal species), with some conservation
elements being represented well above their specified target levels.

FIGURE 4. Private land holdings (blue)
within Heart of the West conservation network core areas and linkages.
TABLE 1. Target
Goals, and Percentages Achieved in the Heart of the West Conservation Network.
|
Conservation Element |
Total in study area (ha) |
Amount targeted (ha, and %) |
Amount achieved in
network (ha) |
Percent of total (in study area) in final network |
Percent of target achieved in final network |
|
Good wolf habitat |
6,281,519 |
1,570,377
(25%) |
2,348,820 |
37.4% |
149.60% |
|
Best wolf habitat |
188,615 |
141,461
(75%) |
145,948 |
77.4% |
103% |
|
Good sage grouse habitat |
7,604,780 |
1,901,195
(25%) |
3,306,773 |
43.5% |
174% |
|
Best sage grouse habitat |
476,308 |
476,308
(100%) |
442,426 |
93% |
93% |
|
Cutthroat trout population segments |
17,157
(linear ha) |
17,157 (100%) |
17,157 |
100% |
100% |
|
Roadless areas |
199,864 |
199,864
(100%) |
199,864 |
100% |
100% |
|
Land cover types |
variesa |
25% of
each type |
Described
in Jones et al. 2004 |
varies
for each type |
107% -
349% of the target |
|
NHP species, G1s and G2s |
variesa |
100% of
each species |
Described
in Jones et al. 2004 |
varies
for each species |
75% -
1896% of the targetb |
|
NHP species, G3s through G5s |
variesa |
25% to
75% of each speciesc |
Described
in Jones et al. 2004 |
varies
for each species |
70% -
608% of the targetb |
a Land cover totals and total number of NHP species in
study area are described in Jones et al. 2004.
b Some of the targets were significantly
over-represented due to a clear
sampling bias in our study area for federally listed and candidate species such
as bald eagle, mountain plover, and boreal toad. Again, it is possible to achieve well over 100% of a target of
100% when occurrences are capped at 25.
cG3 to G5
species target goals ranged from 25% to 75% of occurrences, based on the number
of occurrences in study area, and target goals set for these species in The
Nature Conservancy’s Ecoregional Plan for the Wyoming Basins. Individual target goals, and final results
achieved, for these species are described in Jones et al. 2004.

FIGURE 5. The final conservation network for
the Heart of the
West study area. Core areas are multi-colored if named and
green if unnamed. Linkage zones are in
orange.
The SITES
model generally achieved our target goals for (land cover) representation,
special element incorporation, and focal species habitat representation. Land cover targets were met at levels
between 107% and 349% of the original target goals for those land cover types
(individual target goals, and final results achieved, for each land cover type
are described in Jones et al. 2004).
Overshooting the target goal was usually the result of redundancy in
certain habitat types such as sagebrush that were represented in large
quantities to achieve focal species habitat goals. Focal species habitat targets were also successfully incorporated
in the final conservation network. Other
than the “best sage grouse habitat” category (represented at 93% of the target
goal), all focal species classifications were represented in the network at
levels of 100% of target or greater (Table 1).
Special element targets were variably met in the final portfolio. Of the 203 NHP species incorporated in
special element mapping, 21 did not achieve the 100% representation goal in the
final conservation network. G1 and G2 species
targets were met at least the 75% level, and G3 to G5 targets were met at least
the 50% level. Some of the targets were
significantly over-represented. This
was due to a clear sampling bias in our study area for federally listed and
candidate species such as bald eagle (Haliaetus leucocephalus), mountain plover (Charadrius montanus), and
boreal toad (Bufo boreas). These
species are surveyed far more frequently than others and thus appear to be very
abundant in the study area, and easily picked up in abundance in all SITES
runs.
The final
conservation network also helped meet other, tangential, planning goals that
were not direct inputs into the SITES model.
For example, our proposed set of cores and linkages included over 67 %
of all perennial streams, and 63% of The Nature Conservancy’s portfolio sites,
in the study area. The network also
captured existing suitable habitat for other focal species (such as bighorn
sheep, beaver, bison and grizzly bear) at the level of 50% or greater.
Irreplaceability and Vulnerability Analysis
Drawing upon the approach of
Margules and Pressey (2000) and Noss et al. (2002), we plotted key (named,
Figure 4) core areas on a graph of irreplaceability (y-axis) versus
vulnerability (x-axis) and divided the graph into four quadrants (Figure 6).
The upper right quadrant, which encompasses clusters with high irreplaceability
and high vulnerability, is generally the highest priority for
conservation. This top tier of core
areas is followed by the upper left and lower right quadrants (moderate
priority), and finally, by the lower left quadrant, encompassing cores that are
relatively replaceable and face less severe threats (Noss et al. 2002). One can also prioritize cores within a
quadrant; for example, core areas in the lower left quadrant that are higher on
the irreplaceability axis (y-axis) would warrant higher priority than cores
lower in irreplaceability.
We urge
regional conservationists and activists to give high priority to those core
areas in the upper left quadrant over the lower right quadrant (Figure 6).
Areas of high and irreplaceable biological value deserve conservation
action even if not highly threatened today, and protection of these areas while
they are relatively ecologically intact is more efficient than having to
restore them in the future (Noss et al. 2002).

FIGURE 6. Irreplaceability versus vulnerability
graph. The X axis measures Z-scores for
vulnerability. Moving along the X-axis
from left to right, cores are more vulnerable to degradation or loss of protection. The Y axis measures Z-scores for
irreplaceability; cores higher on this axis are more irreplaceable. Cores in the upper right hand quadrant have
the highest priority for protection, as they are both irreplaceable, and
vulnerable.
The degree
of current oil and gas production in core areas was a major factor in the
irreplaceability and vulnerability analysis.
Often, the SITES results included areas with significant levels of
energy production, usually because they were embedded deep within a core or
were essential for connectivity, or still housed important NHP targets. For the final conservation network, we refer
to these core area planning units with high numbers of oil and gas wells as
Core Recovery Areas and Linkage Recovery Areas. Long-term Core Recovery Areas are those hexes with more than 25
wells per planning unit - or more than one well per 50 acres - and Short-term
Core Recovery Areas are those with between 5 and 25 oil and gas wells per
planning unit. It is likely that the
biggest threat to Heart of the West core areas in the near future is additional
oil and gas extraction efforts (High County News 2003a, 2003b). This factor weighed heavily in the
vulnerability analysis. Figure 7
depicts the degree of the future oil and gas threat facing core areas.

FIGURE 7.
Potential energy development in study area.
The methods used in this exercise
combine various approaches (representation analysis, special element mapping,
and focal species habitat modeling) that are often applied separately in large
scale conservation planning. This
three-track approach to designating core areas and linkages within a proposed
network of reserves, combined with irreplaceability and vulnerability analysis
of core areas, gives practitioners the benefit of a balanced, yet systematic
and rigorous approach to conservation site selection, along with a useful tool
to set conservation priorities within the network. However, to be most helpful, additional vulnerability analyses
should be conducted in the future as the level of human threats to core areas
change. Conservation priorities on the
ground are always changing over time as areas are protected or are lost to
development. Therefore the best
exercises in conservation planning are iterative, to account for changes in
development pressures as well as collection of new and/or better data in areas
important for biodiversity.
Following
the precautionary principle, the core areas highest in both irreplaceability
and vulnerability (featured in upper-right quadrant of Figure 6) should have
the highest priority for protection.
However, all core areas should be protected if the opportunity
arises. A caveat with basing
conservation priorities on computer mapping exercises is that the data that
informs irreplaceability analyses and determines whether a core area is ranked
in the highest priority (1 and 2) quadrants, is itself sometimes based on
incomplete information or analysis. For
example, biases in NHP databases are difficult to avoid. Usually, a lack of element occurrences in a
given area reflects lack of surveys in this region, rather than a lack of rare
and endemic species. In the future, as
additional surveys are conducted in certain core areas, sites that are
currently lower priorities for conservation may move upwards on the irreplaceability
axis (Noss et al. 2002).
There are
similar cautions regarding representation and focal species analysis. For instance, the assumption that
representation of similar percentages of all land cover types adequately
protects the myriad of species that use those vegetation types (the coarse
filter hypothesis), cannot be seriously tested short of a full inventory of an
area’s biota. In terms of focal species
modeling, our individual core areas do not necessarily include the specific
waters or lands needed to maintain viable populations of each target in each
core. Rather, the overall network was
designed under the working assumption that, assuming cores are relatively
connected across the landscape, viable populations of all focal species could
be maintained across the Heart of the West.
Practitioners and scientists utilizing our conservation network may want
to validate this assumption.
Only 4.1 %
of the proposed Heart of the West Wildlands Network is already in some form of
protective federal status (i.e. GAP 1 or 2 status lands, which are primarily
managed for maintenance of biological diversity or natural values). A concerted effort by conservationists,
local communities, land managers and politicians will be required to increase
the amount of area in cores and linkages that are protected at the state and
federal level. However, while we
consistently use terms such as “reserves” and “protect,” we stress that goal
achievement in this conservation planning exercise can readily be met through
means other than what is traditionally thought to be the implementation tools
for reserve design: conservation easements; direct fee acquisition; and
congressional, administrative executive, or agency designations of special
management areas (ranging from Research Natural Areas to national monuments to
wilderness). In fact, the practitioners
and conservationists currently implementing the Heart of the West program are
increasingly turning to new public land implementation strategies that focus
chiefly on changes in agency management.
The chief way to bring about the necessary management is by both
proactively and legally affecting positive change in grazing prescriptions
(through the permit renewal process), designated ORV routes (through travel
plan revisions), and energy development and other use-zoning (through land
management plan revisions).
Currently,
29% of our proposed cores and linkages for the Heart of the West are comprised
of privately held lands. Private lands
offer different and innovative options for land protection, such as
“conservation ranches,” private nature reserves, stewardship assistance and
other management agreements with landowners, and sale to, or easements with,
organizations who carry out all of the above activities, such as The Nature
Conservancy and other land trusts.
Those working to implement the Heart of the West conservation network
intend to engage private landowners who own land in cores and linkages, and
work towards controlled road access, management for biodiversity conservation
purposes, and toleration of carnivores.
The three
track approach to conservation network design combined with irreplaceability/
vulnerability planning gives practitioners a sound basis to make conservation
decisions in an uncertain and quickly changing world. This is clearly evident in the Wyoming Basins Ecoregion of the
Heart of the West which is “Ground
zero” for the current push to develop new sources of oil and gas. The Heart of the West is currently facing
habitat fragmentation and ecosystem degradation never before witnessed in this
region. For this reason, it is imperative that the proposed conservation
network for the Heart of the West is integrated into land use policies, plans
and actions for this region of the middle Rockies. We urge Native Americans, conservation groups, local communities,
mineral extraction companies, and government land management agencies to unite
in working toward its implementation.
ACKNOWLEDGEMENTS
This manuscript was partly derived
from the full conservation plan for the Heart of the West (Jones et al. 2004),
authored by AJ, EM, KD and JC as well as T. Lind, J. Freilich, K. Robinson, L.
Flaherty, and J. Kessler. The full
conservation plan, and associated focal species natural history literature
reviews and habitat models was reviewed by 67 individuals representing
government agencies and academic institutions.
The research, design and mapping efforts behind the Heart of the West
conservation plan were funded in part by the following foundations: George and
Delores Dore Eccles, JEPS, Maki, Peradam, Switzer, Turner, Walbridge, and
Wilburforce.
Andelman, S., I. Ball, F. Davis, and D. Stoms.
1999. SITES V 1.0: An analytical toolbox for designing ecoregional conservation
portfolios. A manual prepared for the
Nature Conservancy.
Bailey, R.G.
1995. Descriptions of the
eco-regions of the United States. USDA
Forest Service, Misc. Publication #1391.
US Government printing Office: 1995-393-835.
Behnke, R.J. 2002. Trout and salmon of North America. New York: Chanticleer Press, Inc.
Bennett, G. and
P. Wit. 2001. The development and application of Ecological Networks: A review
of proposals, plans and programmes. AIDEnvironment and IUCN, Amsterdam. 137 pp.
Bonn, A., A.L
Rodrigues, and K.J. Gaston. 2002. Threatened and endemic species: are they
good indicators of patterns of biodiversity on a national scale? Ecology
Letters 5:733-741.
Brattstrom, B.H. and M.C. Bondello.
1983. Effects of off-road
vehicle noise on desert vertebrates.
Pp. 7206 in: (R.H. Webb and H.G. Wilshire, eds.) Environmental effects
of off-road vehicles: impacts and management
in arid regions. Springer-Verlag, New
York.
Carroll, C. R.F.
Noss, and P.C. Paquet. 2001. Carnivores as focal species for conservation
planning in the Rocky Mountain region. Ecological Applications 11:961-980.
Carroll, C., R.F. Noss, P.C. Paquet, and N. H. Schumaker. 2003. Use of
population viability analyses and reserve selection algorithms in regional
conservation plans. Ecological Applications 13(6):1773-1789.
Comer, P. 2001. Observations and recommendations for setting
conservation goals in ecoregional plans.
The Nature Conservancy, Boulder, CO.
Comer, R.D. 1982. Understanding secondary effects of
development on wildlife resources in mitigation planning. Pps. 16-31 in: (Comer et al., eds.)
Proceedings of the second symposium on issues and technology in the management
of impacted western wildlife. Thorne
Ecological Institute, Boulder, CO.
Foreman,
D., K. Daly, R.F. Noss, M. Clark, K. Menke, D.R. Parsons, and R. Howard. 2003. New Mexico highlands wildlands network
vision. Wildlands Project, Richmond, VT.
Freilich, J., B. Budd, T. Kohley, and B. Hayden. 2001.
The Wyoming Basins Ecoregional Plan.
The Nature Conservancy, Lander, WY.
Gelbard, J.L. and J. Belnap.
2003. Roads as conduits for
exotic plant invasions in a semiarid landscape. Conservation Biology 17:420-432.
Groves, C. 2003. Drafting a Conservation Blueprint: A practitioner’s
guide to planning for biodiversity. Island Press, Washington, D.C.
Groves, C., L. Valutis, D. Vosick, B. Neely, K. Wheaton, J. Touval, and
B. Runnels. 2000. Designing a geography
of hope: a practitioner’s handbook for ecoregional conservation planning. The Nature
Conservancy, Arlington, Virginia.
Harris, L.D. 1984. The fragmented forest: island biogeography theory
and the preservation of biotic diversity. University of Chicago Press, Chicago.
High Country News. 2003a. The Red Desert braces for a gas boom. http://www.hcn.org/servlets/hcn.Article?article_id=14170.
High Country News. 2003b. In the rush to get out the gas, wildlife gets
short shrift. http://www.hcn.org/servlets/hcn.Article?article_id=14171.
Hobbs, R.J. and L.F.
Huenneke. 1992. Disturbance, diversity and invasion:
implications for conservation.
Conservation Biology 6: 324337.
Jeo, R.M. 2002. Extending conservation network framework to
the North Coast region of British Columbia, Canada. Round River Conservation
Studies, Salt Lake City, UT.
Jones, A., J. Catlin, T. Lind, J. Freilich, K. Robinson, L. Flaherty,
E. Molvar, J. Kessler, and K. Daly.
2004. Heart of the West Conservation
Plan. Wild Utah Project, Salt Lake
City, UT.
Hitt, N.P. and C.A. Frissell. 2000. An evaluation of wilderness and
aquatic biointegrity in western Montana. Pg. 138-142 in :McCool, S.F., D.N.
Cole, W.T. Borrie, J. OLoughlin, comps. 2000. Wilderness science in a time of
change conference- Volume 2: Wildernes within the context of larger systems;
1999 May 23-27; Missoula, MT. Proceedings RMRS-P-15-VOL-2. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Rocky Mountain Research Station.
Hobbs, R.J. and L.F.
Huenneke. 1992. Disturbance, diversity and invasion:
implications for conservation.
Conservation Biology 6: 324337.
Ingelfinger, F.M. 2001. The effects of natural gas development on
sagebrush steppe passerines in Sublette County, Wyoming. M.S. Thesis, Univ. of
Wyoming, 110 pp.
Knight, R.L. 1994. Mountains
and plains: the ecology of Wyoming landscapes.
Yale University Press, New Haven, CT.
338 pp.
Lindenmayer, D.B., A.D. Manning, P.L. Smith, H.P. Possingham, I.
Oliver, and M.A. McCarthy. 2002. The focal species approach and landscape
restoration: a critique. Conservation
Biology. 16:338-345.