ArticlePDF Available

Rewilding North America

  • Colorado State University and the Wildlife Conservation Society
Vol. 436|18 August 2005
Rewilding North America
A plan to restore animals that disappeared 13,000 years ago from North America offers an alternative
conservation strategy for the twenty-first century argue Josh Donlan and colleagues.
North America lost most of its large ver-
tebrate species — its megafauna —
some 13,000 years ago at the end of the
Pleistocene. And now Africa’s large mammals
are dying, stranded on a continent where wars
are waging over scarce resources. However
much we would wish otherwise, humans will
continue to cause extinctions, change ecosys-
tems and alter the course of evolution. Here,
we outline a bold plan for preserving some of
our global megafaunal heritage — one that
aims to restore some of the evolutionary and
ecological potential that was lost 13,000 years
ago, and offers an alternative vision for twenty-
first century conservation biology.
Our vision begins immediately, spans the
coming century, and is justified on ecological,
evolutionary, economic, aesthetic, and ethical
grounds. The idea is to actively promote the
restoration of large wild vertebrates into North
America in preference to the ‘pests and weeds
(rats and dandelions) that will otherwis e come
to dominate the landscape. This ‘Pleistocene
rewilding’ would be achieved through a series
of carefully managed ecosystem manipula-
tions using closely related species as proxies for
extinct large vertebrates, and would change the
underlying premise of conservation biology
from managing extinction to act ively restoring
natural processes.
Historic vision
Our proposal is based on several observations.
First, Earth is nowhere pristine; our econom-
ics, politics, demographics and technology
pervade every ecosystem. Such human influ-
ences are unprecedented and show alarming
signs of worsening. Second, environmentalists
are easily caricatured as purveyors of doom
and gloom, to the detriment of conservation.
Third, although human land-use patterns are
dynamic and uncertain, in some areas, such as
parts of the Great Plains in the United States,
human populations are declining1— which
may offer future conservation opportunities.
Fourth, humans were probably at least partly
responsible for the late Pleistocene extinctions
in North America, and our subsequent activi-
ties have curtailed the evolutionary potential
of most remaining large vertebrates. We there-
fore bear an ethical responsibility to redress
these problems.
North American conservationists routinely
turn to the arrival of Columbus in 1492 as a
restoration benchmark. But the arrival of the
first Americans from Eurasia roughly 13,000
years ago constitutes a less arbitrary baseline.
Mammal body-size distributions were similar
across all continents before the late Pleis-
tocene, but subsequent extinction of most
large species drastically altered those distribu-
tions in favour of smaller species.
In the Americas, where large-vertebrate
losses were greatest, the subsequent changes
were undoubtedly ecologically and evolution-
arily significant. Large carnivores and herbi-
vores often play important roles in the
maintenance of biodiversity, and thus many
extinct mammals must have shaped the evolu-
tion of the species we know today2. For example,
the Pronghorn (Antilocapra americana)
evolved over four million years in
North American grasslands which
changed abruptly in the late Pleis-
tocene; the now-extinct American
cheetah (Acinonyx trumani), a key
predator, almost certainly shaped
the Pronghorn’s astonishing speed3.
Beasts of old
Although historical perspectives
have influenced modern conser-
vation planning, existing pro-
grammes do not adequately
address the evolutionary potential
and long-term processes involved
in restoring large-mammal diver-
sity. Africa and parts of Asia are
now the only places where
megafauna are relatively intact,
and the loss of many of these
species within this century seems
likely. Given this risk of further
extinction, rewilding of North
American sites carries global con-
servation implications.
Moreover, humans have emo-
tional relationships with large ver-
tebrates that reflect our own
Pleistocene heritage. More than 1.5
million people annually visit San
Diego’s Wild Animal Park to catch
a glimpse of large mammals —
more than than the number of vis-
itors to most US National Parks. So
an understanding of ecological and
evolutionary history, inspired by
visits to private or public reserves
containing free-roaming megafauna, could
strengthen support for conservation. Pleis-
tocene rewilding would probably increase the
appeal and economic value of both private and
public reserves, as evidenced by the restoration
of wolves to Yellowstone National Park4.
We foresee several phases to Pleistocene
rewilding, some of which are already under-
way. The 50-kg Bolson tortoise (Gopherus
flavomarginatus) was widely distributed across
the Chihuahuan desert until the late Pleis-
tocene. Today it survives only in a small part of
northern Mexico and is critically endangered.
A number of appropriate sites exist for rein-
troduction, including Big Bend National Park,
Texas. And repatriation of captive Bolson tor-
toises to a private ranch in New Mexico is cur-
Figure 1 |Pleistocene rewilding in North America. Symbols
represent condor, horses (Equus caballus and E. asinus in grey;
E. przewalskii and E. hemionus in black), Bolson tortoises,
Camelids, cheetah, Asian (grey) and African (black) elephants,
and lions. a, The likely timescale and area required to restore
proxies for extinct large vertebrates. b, Conservation value and
ecological role (interactivity with other species) on the landscape.
c, Potential economic-cultural value versus potential conflict.
COMMENTARY NATURE|Vol 436|18 August 2005
rently under study. Restoring North America’s
largest surviving temperate terrestrial reptile
to its prehistoric range could bring ecological,
evolutionary, economic, and cultural benefits,
with no apparent costs (Fig. 1).
Likewise, horses and camels originated in
North America, and many species were pre-
sent in the late Pleistocene. Feral horses (Equus
caballus) and asses (E. asinus), widely viewed
as pests in the United States, are plausible
proxies for extinct American species. Also,
given that most of the surviving Eurasian and
African species are now critically endangered,
establishing Asian asses (E. hemionus) and
Przewalski’s horse (E. przewalskii) in North
America might help prevent the extinction of
these endangered species and would restore
equid species to their evolutionary homeland.
Similarly, Bactrian
camels (Camelus bactri-
anus) in North America
could provide a modern
proxy for Camelops, a late
Pleistocene camelid.
Wild Bactrian camels are
on the verge of extinction, currently restricted
to the Gobi desert. Domesticated or captive
camels might benefit arid North American
ecosystems by browsing on woody plants that
today often dominate southwestern US land-
scapes. With proper management, camels
could provide economic benefits as well5. The
overall benefits and disadvantages of horses
and camels as proxies will depend on local
contexts, and possibly on the presence of
appropriate predators.
Free to roam
The second more controversial phase of Pleis-
tocene rewilding could also begin immediately,
with the maintenance of small numbers of
African cheetah (Acinonyx jubatus), Asian (Ele-
phas maximus) and African (Loxodonta
africana) elephants, and lions (Panthera leo) on
private property. Many of these animals are
already in captivity in the United States and the
primary challenge will be to provide them with
naturalistic settings, including large protected
areas of appropriate habitat and in the case of
carnivores, live prey.
The African cheetah, a close relative of the
American cheetah, has only a modest chance
of persisting in the wild in the next century.
Breeding programmes are not self-sustaining,
but some of the 1,000 captive animals could be
used in rewilding. Free-roaming, managed
cheetahs in the southwestern United States
could save the fastest carnivore from extinc-
tion, restore what must have been strong inter-
actions with Pronghorn, and facilitate
ecotourism as an economic alternative to
ranchers (Fig. 1).
Managed elephant populations could simi-
larly benefit ranchers through grassland main-
tenance and ecotourism (Fig. 1). Five species
of proboscidians (mammoths, mastadons and
gomphotheres) once roamed North America
in the late Pleistocene; today the remaining
African and Asian elephants are in grave dan-
ger. Elephants inhibit woodland regeneration
and promote grasslands, as Pleistocene pro-
boscidians probably once did. With appropri-
ate resources, captive US stock and some of the
16,000 domesticated elephants in Asia could
be introduced to North America, where they
might suppress the woody plants that threaten
western grasslands. Fencing, which can be
effective in reducing human–elephant conflict
in Africa, would be the main economic cost.
Lions, which play a pivotal ecological role in
the Serengeti, represent the ultimate in Pleis-
tocene rewilding for North America. They are
increasingly threatened, with populations in
Asia and some parts of Africa critically endan-
gered. Replacing the extinct American lion
(Panthera leo atrox), although
challenging, has clear aesthetic
and economic benefits (Fig. 1).
Among the objections to
Pleistocene rewilding is that
the proposed proxies are not
genetically identical to the ani-
mals formerly existing in North America. And
our vision might strike some as ‘playing God’.
Existing lions and cheetahs are somewhat
smaller than their extinct counterparts, for
example, and Camelus is different from
Camelops. ‘Same’ is relative, however, as illus-
trated by the highly successful reintroduction
of Peregrine falcons (Falco peregrinus) in
North America. Captive-bred birds from
seven subspecies on four continents were used,
yet there were no differences among the birds
in subsequent breeding success6and the sub-
species now serve as a collective proxy for the
extinct midwestern peregrine falcon.
More challenging objections to Pleistocene
rewilding include the possibility of disease
transmission, the fact that habitats have not
remained static over millennia, and the likeli-
hood of unexpected ecological and social con-
sequences of reintroductions. These issues
must be addressed by sound research, prescient
management plans, and unbiased public dis-
course on a case-by-case and locality-by-local-
ity basis. Well-designed, hypothesis-driven
experiments will be needed to assess the
impacts of potential introductions before
releases take place. Large tracts of private land
probably hold the best immediate potential for
such studies, wit h the fossil record and research
providing guideposts and safeguards. For
example, 77,000 large mammals (most of them
Asian and African ungulates, but also cheetahs,
camels and kangaroos) roam free on Texas
ranches7, although their significance for con-
servation remains largely unevaluated.
The third stage in our vision for Pleistocene
rewilding would entail one or more ‘ecological
history p arks, covering vast areas of economi-
cally depressed parts of the Great Plains. As is
the case today in Africa, perimeter fencing
would limit the movements of otherwise free-
living ungulates, elephants, and large carni-
vores, while surrounding towns would benefit
economically from manageme nt and tour ism-
related jobs. A system of similar reserves
across several continents offers the best hope
for long-term survival of large mammals.
Meeting the challenge
In the coming century, by default or design, we
will constrain the breadth and future evolu-
tionary complexity of life on Earth. The default
scenario will surely include ever more pest-
and-weed dominated landscapes, the extinc-
tion of most, if not all, large vertebrates, and a
continuing struggle to slow the loss of biodi-
Pleistocene rewilding is an optimistic alter-
native. We ask of those who find the objections
compelling, are you content with the negative
slant of current conservation philosophy. [ok?]
Will you settle for an American wilderness
emptier than it was just 100 centuries ago? Will
you risk the extinction of the world’s
megafauna should economic, political, and cli-
mate change prove catastrophic for those pop-
ulations remaining in Asia and Africa? The
obstacles are substantial and the risks are not
trivial, but we can no longer accept a hands-off
approach to wilderness preservation. Instead,
we want to reinvigorate wild places, as widely
and rapidly as is prudently possible.
C. Josh Donlan is in the Department of Ecology
and Evolutionary Biology, Cornell University,
Ithaca, New York NY 14853, USA.
Co-authors are Harry W. Greene of Cornell
University, Joel Berger who is at the Teton Field
Office, North American Program, Carl E. Bock and
Jane H. Bock of the University of Colorado,
Boulder, David A. Burney of Fordham University,
New York, James A. Estes of the US Geological
Survey, University of California, Santa Cruz, Dave
Foreman of the Rewilding Institute, Albuquerque,
New Mexico, Paul S. Martin of the Department of
Geosciences, University of Arizona, Tucson, Gary
W. Roemer of the Department of Fishery and
Wildlife Sciences, New Mexico State University,
Las Cruces, New Mexico, Felisa A. Smith of the
University of New Mexico, Albuquerque, and
Michael E. Soulé who is based at Hotchkiss,
1. Lonsdale, R. E. & Archer, J. C. J. Geogr. 97, 108–122 (1998).
2. Janzen, D. H. & Martin, P. S. Science 215, 19–27 (1982).
3. Byers, J. A. American Pronghorn: Social Adaptations and the
Ghosts of Predators Past (Chicago Univ. , Chicago, 1997).
4. Duffield, W. J. & Neher, C. J. Economics of Wolf Recovery in
Yellowstone National Park. Transactions of the 61st North
American Wildlife and Natural Resources Conference
285–292 (1996).
5. Phillips, A., Heucke, J., Dorgers, B. & O'Reilly, G. Co-grazing
cattle and camels. A report for the Rural Industries Research
and Development Corporation (Rural Industries Research
and Development Corporation, Kingston, 2001).
6. Tordoff, H. B. & Redig, P. T. Conserv. Biol. 15, 528–532
7. Schmidly, D. J. Texas Natural History: A Century of Change
(Texas Tech Univ. Press, Lubbock, 2002).
Acknowledgements: We thank the Environmental
Leadership Program, Lichen Foundation, Turner
Endangered Species Fund, New Mexico Agricultural
Station, Ladder Ranch, C. Buell., S. Dobrott, T. Gorton, M.
K. Phillips and J. C. Truett for support and encouragement.
“Rewilding of North
American sites carries
global conservation
... In instances where extinction prevents the reintroduction of megafauna, ecologists consider introducing near-relatives of the extinct species. Such paleo-rewilding was the subject of a controversial essay in Nature (Donlan et al. 2005). As discussed by Noss (2019), Donlan et al. (2005) used the term "rewilding" to refer to the experimental introduction to North America of elephants, which represented the closest living relatives of animals that became extinct in that locale during the late Pleistocene, approximately 13,000 years ago. ...
... Such paleo-rewilding was the subject of a controversial essay in Nature (Donlan et al. 2005). As discussed by Noss (2019), Donlan et al. (2005) used the term "rewilding" to refer to the experimental introduction to North America of elephants, which represented the closest living relatives of animals that became extinct in that locale during the late Pleistocene, approximately 13,000 years ago. Such rewilding then became associated with the restoration of prehistoric landscapes; for example through the introduction of the predomesticated ancestor of the dromedary camel (Camelus dromedarius), which was driven into extinction from the wild 4,000 to 5,000 years ago (Root-Bernstein and Svenning 2016). ...
Full-text available
The Dutch conservation area Oostvaardersplassen was initiated as a rewilding project within the Netherlands' protected area network. It came under the spotlight when management strategies and practices were criticized by scientists, conservation practitioners, and the public, from a number of perspectives-not all of which were compatible. This article reviews the origin, evolution, and application of the rewilding concept and examines the Oostvaardersplas-sen project as a case study. Our assessment demonstrates that the area was never an appropriate site for rewilding, and beset by rudderless management, and led to a situation that was ecologically and ethically untenable. The case study is used to illustrate humanity's evolving role in environmental protection where advances in the understanding of ecological complexity, animal behavior, and sentience, cannot be ignored when addressing environmental protection, problem-solving, and management. Finally, it lays out options for the future in the absence of the three Cs of rewilding, the Cores, Corridors, Carnivores, and introduces the concept of the fourth C, Compassion.
Full-text available
More than 30 years after it was first proposed as a biodiversity conservation strategy, rewilding remains a controversial concept. There is currently little agreement about what the goals of rewilding are, and how these are best achieved, limiting the utility of rewilding in mainstream conservation. Achieving consensus about rewilding requires agreeing about what “wild” means, but many different definitions exist, reflecting the diversity of values in conservation. There are three key debates that must be addressed to find a consensual definition of “wild”: (1) to which extent can people and “wild” nature co-exist?; (2) how much space does “wild” nature need? and (3) what kinds of “wild” nature do we value? Depending on the kinds of “wild” nature rewilding aims to create, rewilding policy will be faced with managing different opportunities and risks for biodiversity and people.
Full-text available
Socioecological information should be properly employed in the process of spatial analysis, planning, and management in order to respond to complex and multidirectional biodiversity issues. We conducted this study to map socioecological hotspots, where landscapes of social significance and wildlife habitats overlap, show to what extent and how the spatial distribution of social values (SVs) of people toward their landscapes interact with wildlife habitats in socioecological hotspots, simulate the potential for habitat degradation as a result of human activities linked to SVs, identify strategic areas for landscape restoration in socioecological hotspots, where both environmental conditions and SVs support the persistence and colonization of wildlife, and detect specific areas, where SVs of people may be contradictive leading to land-use disputes. We developed a model to show the potential for habitat degradation based on the spatial distribution of SVs associated with landscapes. We restricted our study to the Upper Missouri River Basin (UMRB) and focused on habitats of five keystone mammal species to assess the validity of our model. Habitat loss, habitat subdivision, habitat dispersion, and habitat shrinkage can be four consequences of human activities for biodiversity in socioecological hotspots of the UMRB, however, the magnitude of impacts varies among landscapes and mammal species. Spatially explicit models to properly map SVs in relation to wildlife habitats are still associated with some uncertainties and limitations, and therefore, require further development. Change in SVs and public attitudes toward land use is essential to avoid further biodiversity loss in this region.
This review considers how rapid environmental change, generated through both inhuman natural forces and human-induced impacts, affects landscape futures and decision-making processes. To do this, we start by defining ‘futures’, and, more specifically, the different kinds of futures at stake in changing landscapes. We discuss how rapid environmental change not only puts immediate pressure on identifying alternative futures for landscapes, but also threatens to unsettle patterns of attachment to the landscape. We then explore different ways of managing tensions and consider strategies that have been used for breaking down binary divisions that may stymie informed and integrated decision-making. We conclude by adapting a five-point framework that incorporates uncertainty and environmental change when making decisions about landscape futures.
There is an increased interest in the possibility to use large animals in the restoration of degraded ecosystems and to increase the capacity of natural areas to sustain richer biodiversity. We quantify the dietary differences and similarities between five moose and 22 red deer introduced in a 2100 ha fenced area under restoration (Lille Vildmose, Denmark) and their potential in the restoration process. Moose and red deer were selected based on their biogeographic affinity and existing knowledge of their diet and feeding behaviour, and are expected to counteract encroachment on open habitats. Using eDNA metabarcoding of dung collected in the field and species identification through saliva recovered from browsed twigs, we investigated the species’ diets and diet overlap after the moose were released into the area, and assessed how their diets match the expected ecosystem function. Despite the partial overlap of some dietary items, such as Betulaceae and Salicaceae, we report a differential use in height of tree species and distinctive use of other food resources between the two species, with red deer eating more herbaceous plants and the moose diet also containing aquatic plants. This study exemplifies a monitoring approach to restoration projects where large herbivores are expected to provide a key ecosystem function through their browsing and grazing.
Full-text available
This article adds to the “golden rules” of rewilding (the 3 Cs), that of Cores, Carnivores, and Corridors, a fourth C – Compassion – which would ensure that any (re)introduction must be in the interests of the individual animals involved. Using the history of the Oostvaardeplassen project as a case study, the scientific and ethical constraints and opportunities for rewilding are discussed. All four Cs are absent from Oostvaardeplassen, which can, therefore, be considered an example of how rewilding should not be undertaken. So what are the alternative ways forward?
Rewilding is an increasingly significant normative concept in biological conservation, environmental planning and urban greenspace studies. Originally developed in relation to ecological restoration theory and practice the term has developed over time and has come to be used in a range of environmental discourses. Along with other ecosystem services tourism provides a major economic justification for rewilding, although specific studies on rewilding and tourism are limited. In introducing this special issue of Journal of Ecotourism on tourism and rewilding attention is given to the main definitions and approaches to rewilding, their application, and some of the issues that emerge, including contestation over not only how rewilding is framed but also the implications of the (re)introduction of wild animals. Rewilding therefore provides new opportunities to examine the nature-culture relationship and the ethics, management and policies associated with tourism's embeddedness in ecological restoration practices and the framing of the wild.
Pronghorn antelope are the fastest runners in North America, clocked at speeds of up to 100 kilometers per hour. Yet none of their current predators can come close to running this fast. Pronghorn also gather in groups, a behavior commonly viewed as a "safety in numbers" defense. But again, none of their living predators are fearsome enough to merit such a response. In this elegantly written book, John A. Byers argues that these mystifying behaviors evolved in response to the dangerous predators with whom pronghorn shared their grassland home for nearly four million years: among them fleet hyenas, lions, and cheetahs. Although these predators died out ten thousand years ago, pronghorn still behave as if they were present—as if they were living with the ghosts of predators past. Byers's provocative hypothesis will stimulate behavioral ecologists and mammalogists to consider whether other species' adaptations are also haunted by selective pressures from predators past. The book will also find a ready audience among evolutionary biologists and paleontologists.
Economics of Wolf Recovery in Yellowstone National Park
  • W J Duffield
  • C J Neher
Duffield, W. J. & Neher, C. J. Economics of Wolf Recovery in Yellowstone National Park. Transactions of the 61st North American Wildlife and Natural Resources Conference 285–292 (1996).
  • D H Janzen
  • P S Martin
Janzen, D. H. & Martin, P. S. Science 215, 19–27 (1982).
  • H B Tordoff
  • P T Redig
Tordoff, H. B. & Redig, P. T. Conserv. Biol. 15, 528–532 (2001).