Comments
Ecoregions in Ascendance: Reply to Jepson
and Whittaker
ERIC WIKRAMANAYAKE,* ERIC DINERSTEIN,* COLBY LOUCKS,* DAVID OLSON,*
JOHN MORRISON,* JOHN LAMOREUX,* MEGHAN MCKNIGHT,* AND
PRASHANT HEDAO†
*Conservation Science Program, World Wildlife Fund – United States, 1250 24th Street NW, Washington, D.C.
20037, U.S.A.
†Environmental Systems Research Institute, Inc., Redlands, CA 92373, U.S.A.
Introduction
In their essay, Jepson and Whittaker (2002 [this issue])
express three aims: (1) to place the World Wildlife Fund
( WWF ) ecoregions (Olson et al. 2001; Wikramanayake
et al. 2001) in context with alternative systems for setting conservation priorities; (2) to express concern about
the explicitness, transparency, and repeatability of the
methods employed to define and delineate ecoregions;
and (3) to ask whether the WWF ecoregions improve upon
existing schemes of representing biodiversity (which they
attempt to investigate by performing a qualitative test with
Indonesia as a case study).
Their essay contains a number of misrepresentations
of our ecoregion approach. In the interest of brevity, we
address only the major issues in this reply. We begin by
arguing for the value of ecoregions, outlining the rationale behind the ecoregion framework and the delineation process, and comparing these with other efforts. The
approach is detailed explicitly by Wikramanayake et al.
(2001) and others ( Ricketts et al. 1999; Abell et al. 2000;
Dinerstein et al. 2000).
Why Ecoregions?
Ecoregions may be a relatively new concept in conservation biology, but they are based on classical biogeography. Although a single institution like WWF may actually
produce an ecoregion map, the ecoregions themselves
Address correspondence to E. Dinerstein, email eric.dinerstein@
wwfus.org
Paper submitted August 13, 2001; revised manuscript accepted August 20, 2001.
238
Conservation Biology, Pages 238–243
Volume 16, No. 1, February 2002
are developed through extensive collaboration with biogeographers, taxonomists, conservation biologists, and
ecologists from around the world. Since 1999, three of
the leading international nongovernmental organizations—
WWF, The Nature Conservancy ( TNC ), and Conservation
International (CI )—and many other conservation partners
around the world have identified ecoregions or larger biogeographic units such as hotspots (CI ) as an appropriate
scale for conservation priority setting and planning. These
organizations have found that ecoregions provide the critical spatial link between global priority-setting efforts and
site-based assessments.
Perhaps the most attractive feature of ecoregions is
that they define biogeographic units that are most suitable for meeting a fundamental goal of biodiversity conservation: the representation of all habitats and distinct
biotas in networks of conservation areas. The use of ecoregions allows us to avoid redundancy and enhance complementarity in the design of reserve networks better
than do approaches that rely on political units to determine conservation priorities. Developing conservation
strategies within a framework of ecoregions also allows
us to address the ecological processes that maintain biodiversity and to protect populations of species with large
spatial needs that cannot be accommodated at the site
scale or that do not adhere to political boundaries ( Dinerstein et al. 2000; Groves et al. 2000; Mace et al. 2000).
The magnitude of the shift to ecoregions for conservation planning is worth noting for the benefit of conservation biologists unaware of this trend. The WWF has
targeted 189 of the planet’s 867 terrestrial ecoregions, 15
freshwater ecoregions, and 24 marine ecoregions for such
analyses. Within a few years, TNC plans to finish ecoregion
strategies for all 62 of its U.S. ecoregions, and CI plans to
complete similar analyses for all 25 hotspots ( Myers et al.
2000), many of which overlap with the Global 200 Ecoregions identified by WWF (Olson & Dinerstein 1998). The
Wikramanayake et al.
WWF is collaborating with both organizations on several of
these efforts. Regional analyses, such as those conducted
for the Nearctic, Neotropics, and Indo-Pacific ( Dinerstein et al. 1995; Ricketts et al. 1999; Abell et al. 2000;
Wikramanayake et al. 2001), constitute the foundation
for the WWF’s efforts and have received favorable reviews in the conservation biology literature ( Pimm 1999;
Terborgh 1999; Caicco 2000; Cassin 2000; Ebersole 2000;
Ruhren 2000). Other regional assessments in this series
are in press ( Burgess et al. 2002) or in preparation (Afrotropics freshwater, Palearctic, Neotropics). Given the
trend among conservation nongovernmental organizations, we predict that within 2 years, submissions on the
practice of ecoregion-based conservation will become increasingly prevalent in peer-reviewed journals such as
Conservation Biology.
Even Jepson (1999) acknowledges the usefulness of
our ecoregion framework to the conservation of biodiversity. In a rather critical review of BirdLife International’s
landmark publication, Endemic Bird Areas of the World:
Priorities for Biodiversity Conservation, he writes: “In
short, the meat in this book does not justify the 846 pages.
WWF–US has incorporated endemic bird areas (EBAs) in
the ecoregion approach which combines a representation and hotspot approach in the Global 200. Ecoregions
probably represent the culmination of the biologist’s
perspective on biodiversity priorities.”
Jepson confuses, as he has elsewhere ( Jepson & Canney 2001), representation with prioritization. The Dasmann-Udvardy and Omernik-Bailey assessments are representation systems, as are WWF’s ecoregions (Olson et
al. 2001). In contrast, strategic prioritization highlights
areas essential for the conservation of irreplaceable and
threatened biodiversity (Margules & Pressey 2000). Conservation International’s hotspots ( Myers et al. 2000),
WWF’s Global 200 Ecoregions (Olson & Dinerstein 1998),
BirdLife International’s EBAs (Stattersfield et al. 1998),
and the World Conservation Union ( IUCN) and WWF’s
centers of plant diversity ( IUCN & WWF 1995) exemplify prioritization systems that are highly congruent despite their differing scales of analysis ( da Fonseca et al.
2000).
Ecoregions in Context
Here we address Jepson and Whittaker’s (2002 [this issue]) suggestion that WWF ecoregions are inferior to existing systems of classification. Jepson’s specific criticism
of WWF’s ecoregions is that they conflict with the Dasmann-Udvardy system of biogeographic representation.
They do not. Rather, ecoregions represent continuous
updating and improvement of this system, with the goal
of reconciling the numerous idiosyncratic approaches
that have developed over the last few decades into a
standardized global scheme. They also introduce the con-
Ecoregions in Ascendance
239
cept to the current generation of conservation planners
and incorporate aquatic systems. Where the implementation of existing frameworks is weak, the delineation of
ecoregions will strengthen it; where implementation is
strong, as in Indonesia, it will reinforce it and address remaining gaps as demonstrated below.
Because of their biogeographic baseline, ecoregions
reflect the distribution of species and communities more
accurately than do units derived from models based on
combinations of coarse-scale biophysical features such
as rainfall and temperature, which Jepson and Whittaker
view as objective, transparent, and repeatable. Bailey’s
(1998) terrestrial ecoregions are derived primarily from
such models and are based on global-scale biophysical
data. The resulting ecoregion units are excessively large
for regional planning and do not effectively reflect biogeographic patterns determined largely by historical processes, particularly in island systems such as in the IndoPacific. Bailey’s system also ignores many tropical islands
with extraordinary levels of endemism but too small in
size to be classified in his approach.
Dasmann (1974) and Udvardy (1975) were the first to
map a global system of biogeographic units (referred to
as biotic provinces) within a framework of realms and
biomes. Their biotic provinces are based on known biogeographic patterns, rather than on biophysical models,
and are thus comparable to ecoregions in their derivation and intent (i.e., identifying distinct biotas). But the
biogeographic resolution of Dasmann and Udvardy’s biotic provinces is considerably coarser than our system of
terrestrial ecoregions. We have identified 867 terrestrial
ecoregions for the world (Olson et al. 2001), compared
with the 198 and 193 units of Dasmann’s and Udvardy’s
systems, respectively. The increased resolution is most
apparent in the tropics (between the Tropics of Cancer
and Capricorn), where Dasmann (1974) and Udvardy
(1975) identified 115 and 117 units, respectively, compared with 463 identified in the WWF terrestrial ecoregion map of the world. A finer-scale example that covers
part of Indonesia—Irian Jaya, the western part of the island of New Guinea, and the country of Papua New
Guinea—makes the point even more clearly. Dasmann
and Udvardy treat the island of New Guinea as a single
unit, whereas we distinguish 12 ecoregions consisting of
four lowland and four montane broadleaf forests, one alpine scrub ecoregion along the central cordillera, one
mangrove forest, one freshwater swamp forest, and one
savanna-grassland ecoregion.
In sum, the Bailey, Dasmann, and Udvardy models, however useful as a broad framework for understanding biogeography, are far too coarse for either representation or prioritization analyses (sensu Jepson). Thus, MacKinnon and
Wind (1981) identified a third, finer-scale tier, biounits,
which they nested within Udvardy’s biogeographic provinces. MacKinnon subsequently applied the biounit (and
subunit) concept to a broader regional area, from Pakistan
Conservation Biology
Volume 16, No. 1, February 2002
240
Ecoregions in Ascendance
to Papua New Guinea ( MacKinnon & MacKinnon 1986;
MacKinnon 1997 ). Although this step facilitates regional
planning, the biounit classification mixed distinct biomes,
making representation analyses difficult. For instance, a
single biounit can be comprised of mangroves, tropical
dry deciduous forests, thorn scrub, tropical semi-evergreen forests, and tropical moist forests. These habitat
types differ not only in characteristic species, but also in
ecological dynamics such as species assemblages, natural densities of large herbivores, vectors for seed dispersal, types of natural-disturbance regimes, response to
disturbance, and degree of resiliency. Mangroves are ecological systems vastly different from tropical dry thorn
scrub, and the conservation planning and representation
of each is improved by recognizing them as separate
ecoregions (if, for example, the area of mangroves is sufficiently extensive to meet the spatial criterion of an
ecoregion). Thus, in our analysis of terrestrial ecoregions
of the Indo-Pacific region ( Wikramanayake et al. 2001),
we used MacKinnon’s construct of the original habitat
to replace biounits with ecoregions within the Udvardy
framework.
To subdivide Udvardy’s biogeographic realms ( Fig. 1),
we identified bioregions. Bioregions are based on accepted
biogeographic demarcations and have been described in
detail, with maps, by Wikramanayake et al. (2001). Within
a given bioregion, ecoregions can be nested under different biomes. In other words, bioregions allow us to
achieve a further level of representation in a global or regional priority portfolio by capturing a minimum number of ecoregions within each biome occurring in each
bioregion. For example, the Western Ghats Mountains
along India’s western coast are clothed in rainforests, as
are the mountains of central Borneo. But the flora and
fauna of the Western Ghats are distinctly different from
those in central Bornean forests: the former has Nilgiri
Wikramanayake et al.
tahr (Hemitragus hylocriusI ), lion-tailed macaques (Macaca silenus), Nilgiri langurs (Presbytis johnii ), tigers, and
a diverse forest association, whereas the latter has orangutans (Pongo pygmaeus), gibbons, Sumatran rhinoceros
(Dicerorhinus sumatrensis), and forest associations dominated by Dipterocarpaceae spp. To conserve one or the
other as a representative rainforest habitat of Asia in a global or regional conservation strategy would be insufficient.
Our analysis carefully considered previous approaches,
such as EBAs (Stattersfield et al. 1998) and centers of plant
diversity ( IUCN & WWF 1995). Where these efforts distinguished distinct biotas at a high level of resolution, we
based our ecoregion delineation on them. For example, in
western Indonesia we used the maps of the original vegetation from MacKinnon (1997 ), but in eastern Indonesia we
used the EBAs (Stattersfield et al. 1998) because we considered them finer and more robust discriminators of
biodiversity for this system of small island complexes.
We did not use hotspots ( Mittermeier et al. 1999) because these units are regional in extent and are based on
a combination of high biodiversity and threat, rather
than strictly on biogeography.
Explicitness, Transparency, and Repeatability
We delineated ecoregions according to a set of rules based
on biogeography, habitat type, and elevation. Wikramanayake et al. (2001) explicitly describe these rules and define
the biogeographic units and all indices and thresholds that
were used to delineate and assess ecoregions for the IndoPacific region. To make the process of ecoregion delineation transparent and repeatable, we have made available all
the information we used from Wikramanayake et al.
(2001). We have also provided lists of all contributors and
Figure 1. Hierarchy of spatial
units used in the conservation
assessment of the Indo-Pacific.
Conservation Biology
Volume 16, No. 1, February 2002
Wikramanayake et al.
Ecoregions in Ascendance
reviewers for the respective ecoregions. As in any attempt
to represent nature in spatial units, we do not expect unanimous acceptance of the ecoregion boundaries as we have
presented them. We do hope that many regional biogeographers and conservation biologists, once they understand
how the global map incorporates published, widely used
classification schemes, will find it acceptable.
Ecoregions Misrepresented: Gestalt, Experts,
and Indonesia
If nature could be fit into a one-size-fits-all framework,
our task would be simple. Unfortunately, it does not. A
map of eastern Indonesia shows a plethora of lines that
have sprung up around Wallace’s now-famous line—all
in attempt to refine Wallace’s original effort and observations. This is because the biodiversity of archipelagos results from a complex suite of biogeographic variables
including species vagility and origin, geologic history, geomorphology, landform, size, and interisland distance. Compounding the problem of detecting patterns of biogeography is a lack of comprehensive and reliable biodiversity data
for many places on Earth. There are no biological databases
that are uniform in detail and resolution for the entire
planet. Thus, developing a global framework of ecological
units necessitates working with uneven information. At
times, this requires sifting through and combining the best
available information from several sources and using various biological and geophysical parameters. In his aforementioned review of EBAs, Jepson (1999) commends us
for incorporating them into the Global 200 Ecoregion
analysis. But in their current essay, Jepson and Whittaker
(2002, this issue) criticize it as part of a gestalt approach
to ecoregion delineation. Yet they defend the DasmanUdvardy-MacKinnon-Wind EBA approach that was used
to guide planning for Indonesia’s current protected-area
system—even though it uses different types of data and
data sources, and “zonal” and “azonal” features ( Jepson
and Whittaker’s words). It is unclear to us how such an
approach is less a gestalt synthesis than ours. We also believe that expert review and input strengthen the validity and accuracy of ecoregions and our approach, despite
Jepson and Whittaker’s skepticism.
Table 1.
241
The WWF Ecoregion Approach and the
Goal of Representation of Indonesia’s
Protected-Areas System
Much of what has been described above can be viewed
as a rather narrow set of academic disagreements best
suited for a biogeography journal or a debate among participants at a workshop on ecoregion conservation planning in Indonesia. But Jepson and Whittaker’s critique
assumes real-world significance when the issue becomes
the relationship between designing biogeographic frameworks and establishing representative networks of protected areas. They claim that “Given that the existing
schemes, despite flaws, have achieved their purpose in
Indonesia . . . we question the wisdom of introducing
the WWF ecoregion system.” This statement is the extent of their “test” of the efficacy of ecoregions against
the existing planning system. We strongly disagree with
their opinion, believe that the ecoregion approach improves on the previous framework, and present a quantitative test of our own below.
Ecoregions are ultimately tools for conservation planning. Although the ecoregion analyses point out many of
the strengths of the existing design of a protected-areas
system for Indonesia, it also highlights some significant
gaps. A recent assessment by Derek Holmes ( by Wikramanayake et al. 2001) projects that, at current rates of destruction, the remaining intact lowland forests of
Sumatra and Kalimantan will disappear within the next 5
and 9 years, respectively. Taking this analysis further, we
can compare current levels of protection in lowland and
montane forest ecoregions of Sumatra, Java, Borneo, and
Sulawesi. Almost 15% of the montane forests are contained within IUCN category I–IV protected areas, but
less than 6% of the remaining lowland forests are protected ( Table 1). Thus, protection is biased toward inclusion of montane forests, which are largely unproductive
for agriculture and too steep for logging, plantations, and
settlements, whereas the rich lowland rainforests of Indonesia remain underrepresented within the protectedareas system. An analysis based on ecoregions illustrates
this disparity in representation better than one based on
classification schemes promoted by Jepson and Whittaker that do not effectively distinguish montane and
lowland habitats.
Comparison of degree of protection (PA, protected area) in lowland and montane forests of Java, Sumatra, Borneo, and Sulawesi.
Protected areas (km2)*
Forest
Lowland
Montane
IUCN I-VIII
IUCN I-IV
Total area of lowland and
montane forests (km2)
I–VIII in PA (%)
I–IV in PA (%)
80,029
53,630
68,519
46,056
1,221,400
312,200
6.6
17.2
5.6
14.8
* This analysis uses World Conservation Monitoring Centre and Asian Bureau for Conservation (1997 ) protected areas and does not include
those designated as proposed.
Conservation Biology
Volume 16, No. 1, February 2002
242
Ecoregions in Ascendance
Jepson and Whittaker also suggest that ecoregions can
divert conservation efforts from biologically important
areas. They are concerned that treatment of the lowland
forests of Sulawesi as a single ecoregion may suggest to
planners that targeting resources to any peninsula in Sulawesi is equally good, thereby detracting from important and high-priority conservation areas such as the northern Minahasa peninsula. In reality, any intelligent,
competent conservation planner will look carefully at biogeographic patterns to ensure adequate representation
within a given ecoregion, typically by creating biogeographic subdivisions, a method used by the WWF, CI, and
TNC (see Dinerstein et al. 2000).
Jepson and his colleagues have over the past 2 years
criticized the prioritization efforts of BirdLife International’s EBAs, CI’s hotspots, and WWF’s terrestrial ecoregion analysis without offering the scientific community a
compelling alternative for representation of global biodiversity or priority setting. In their critiques of all three approaches they are literally and figuratively all over the
map, offering conflicting recommendations. For example:
in a critique of EBAs, Jepson (1999) proposes that WWF
ecoregions are better units for prioritization and representation than are targets based on range-restricted birds as
proposed by BirdLife; in a critique of hotspots, Jepson
and Canney (2001) evaluate hotspots using nonbiological
criteria and chastise CI for not doing a better job of representing biodiversity, when hotspots is not a representation scheme but a prioritization scheme. Finally, in the
current essay Jepson and Whittaker criticize the WWF
ecoregion approach in favor of more “transparent, less gestalt-like” approaches for setting priorities such as EBAs,
when the WWF ecoregions approach is first and foremost
an attempt to improve representation of distinct habitats
and communities.
Ecoregions have already helped improve conservation
planning by catalyzing debate on critical issues such as
representation of biodiversity and large-scale ecological
processes. In the case of Indonesia, and we suspect in
many other biologically important places on Earth, the
increased use of ecoregions or other biogeographic units
will assist conservation planners in making a compelling,
rigorous case to decision-makers and donors for greatly
expanded conservation efforts. The next phase of our
work is to use ecoregions as the basis for a “grand synthesis” of global priority-setting approaches so that we
can speak with one voice to decision-makers and donors
about the fate of the biota on our planet. We hope Jepson and his colleagues will join us in this important step
to save life on Earth.
Literature Cited
Abell, R. A., D. M. Olson, E. Dinerstein, P. T. Hurley, J. T. Diggs, W. Eichbaum, S. Walters, W. Wettengel, T. Allnutt, C. J. Loucks, and P. Hedao.
Conservation Biology
Volume 16, No. 1, February 2002
Wikramanayake et al.
2000. Freshwater ecoregions of North America: a conservation assessment. Island Press, Washington, D.C.
Bailey, R. G. 1998. Ecoregions: the ecosystem geography of the oceans
and continents. Springer Verlag, New York.
Burgess, N., E. Underwood-Russell, J. D’Amico, D. M. Olson, E. Dinerstein, and I. Itoua. 2002. Terrestrial ecoregions of Africa: a conservation assessment. Island Press, Washington, D.C.
Caicco, S. 2000. Review of terrestrial ecoregions of North America: a
conservation assessment. Natural Areas Journal 20(4):390.
Cassin, J. L. 2000. Filling the gap in conservation strategies: a mesoscale tool for biodiversity assessment and conservation. Conservation Biology 14:1555–1556.
da Fonseca, G. A. B., et al. 2000. Following Africa’s lead in setting priorities. Nature 405:393–394.
Dasmann, R. F. 1974. Biotic provinces of the world: further development of a system for defining and classifying natural regions for
purposes of conservation. Occasional paper 9. World Conservation
Union, Gland, Switzerland.
Dinerstein, E., D. M. Olson, D. Graham, A. Webster, S. Pimm, M. Bookbinder, and G. Ledec. 1995. A conservation assessment of the terrestrial ecoregions of Latin America and the Caribbean. World
Bank, New York.
Dinerstein, E., G. Powell, D. M. Olson, E. Wikramanayake, R. Abell,
C. Loucks, E. Underwood, T. Allnutt, W. Wettengel, T. Ricketts,
H. Strand, S. O’Connor, and N. Burgess. 2000. A workbook for conducting biological assessments and developing biodiversity visions
for ecoregion-based conservation. World Wildlife Fund, Washington, D.C.
Ebersole, J. P. 2000. Review of terrestrial ecoregions of North America:
a conservation assessment. BioScience 50:541–543.
Groves, C., L. Valutis, D. Vosick, B. Neely, K. Wheaton, J. Touval, and
B. Runnels. 2000. Designing a geography of hope: a practitioners
handbook for ecoregional conservation planning. The Nature Conservancy, Washington, D.C.
Jepson, P. 1999. [Review of] Endemic bird areas of the world: priorities for biodiversity conservation. Ibis 141:163.
Jepson, P. 2001. Global biodiversity plan needs to convince local policy-makers. Nature 409:12.
Jepson, P., and S. Canney. 2001. Biodiversity hotspots: hot for what?
Global Ecology & Biogeography 10:225–227.
Jepson, P., and R. J. Whittaker. 2002. Ecoregions in context: a critique with special reference to Indonesia. Conservation Biology
16:42–57.
Mace, G. M., et al. 2000. It’s time to work together and stop duplicating conservation efforts. Nature 405:393.
MacKinnon, J. 1997. Protected areas systems review of the IndoMalayan realm. The Asian Bureau for Conservation and The World
Conservation Monitoring Centre/World Bank Publication, Canterbury, United Kingdom.
MacKinnon, J., and K. MacKinnon. 1986. Review of the protected areas system in the Indo-Malayan Realm. World Conservation Union,
Gland, Switzerland, and United Nations Environmental Programme,
Cambridge, United Kingdom.
MacKinnon, J., and J. Wind. 1981. Birds of Indonesia. Special report
F0/INS/78/061. Available from the Food and Agriculture Organization of the United Nations, Rome.
Margules, C. R., and R. L. Pressey. 2000. Systematic conservation planning. Nature 405:243–253.
Mittermeier, R. A., N. Myers, P. Gil, and C. G. Mittermeier, editors.
1999. Hotspots, Earth’s biologically richest and most endangered
terrestrial ecosystems. CEMEX, Mexico City.
Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca,
and J. Kent. 2000. Biodiversity hotspots for conservation priorities.
Nature 403:853–858.
Olson, D. M., and E. Dinerstein. 1998. The Global 200: a representation approach to conserving the Earth’s distinctive ecoregions.
Conservation Biology 12:502–515.
Wikramanayake et al.
Olson, D. M., et al. 2001. Terrestrial ecoregions of the world: a new
map of life of Earth. BioScience 51:933–938.
Pimm, S. 1999. Seeing both the woods and the trees: a novel approach
to classifying biodiversity may aid attempts at conservation. Nature
402:853–854.
Ricketts, T. H., E. Dinerstein, D. M. Olson, C. J. Loucks, W. Eichbaum,
D. DellaSala, K. Kavanagh, P. Hedao, P. T. Hurley, K. M. Carney,
R. Abell, and S. Walters. 1999. Terrestrial ecoregions of North
America: a conservation assessment. Island Press, Washington, D.C.
Ruhren, S. 2000. Review of terrestrial ecoregions of North America: a
conservation assessment. World Birdwatch 22(4):23–24.
Stattersfield, A. J., M. J. Corsby, A. J. Long, and D. C. Wege. 1998. Global directory of endemic bird areas. BirdLife International, Cambridge, United Kingdom.
Terborgh, J. 1999. Requiem for nature. Island Press, Washington, D.C.
Ecoregions in Ascendance
243
Udvardy, M. D. F. 1975. A classification of the biogeographical provinces of the world. Occasional paper 18. World Conservation
Union, Gland, Switzerland.
Wikramanayake, E., E. Dinerstein, C. J. Loucks, D. M. Olson, J. Morrison, J. Lamoreux, M. McKnight, and P. Hedao. 2001. Terrestrial
ecoregions of the Indo-Pacific: a conservation assessment. Island
Press, Washington, D.C.
World Conservation Monitoring Centre ( WCMC) and Asian Bureau for
Conservation. 1997. Protected areas system review of the IndoMalayan realm. J. MacKinnon, editor. Asian Bureau for Conservation Limited, Canterbury, England.
World Conservation Union ( IUCN ) and World Wildlife Fund ( WWF ).
1995. Centres of plant diversity: a guide and strategy for their conservation. 2. Asia, Australasia and the Pacific. IUCN, Cambridge,
United Kingdom.
Conservation Biology
Volume 16, No. 1, February 2002