In defense of fences
HUMAN-DRIVEN HABITAT fragmentation
reduces global biodiversity and ecosystem
functioning (1). R. Woodroffe et al. (“To
fence or not to fence,” Perspective, 4 April,
p. 46) claim that fencing, despite some
positive outcomes (2, 3), overwhelmingly
exacerbates fragmentation and negatively
affects wildlife conservation. They suggest
that fencing should only be considered as a
last resort and that fence removal is important for climate change preparedness.
Woodroffe et al. underplay the critical
role fences play in habitat conservation
and protection of livelihoods in tropical Africa, where scattered islands of
natural habitat persist amidst a sea of
agricultural encroachment, spared often
through physical demarcation of protected
area boundaries (4). In Africa, biomass
extraction and subsistence/smallholder
agriculture remain the dominant drivers of
degradation (5). Although fencing can be
problematic, especially for gene-flow [but
see (6)] and large-scale mammal migration, it successfully arrests the gradual
erosion of habitats, combats poaching,
and can facilitate wildlife tolerance among
Woodroffe et al. cite growing
populations of unfenced carnivores/
megaherbivores in North America as a
model for other regions. Yet in Africa, the
notion of rural communities enthusiasti-
cally sharing dwindling environmental
space with wildlife is an ideal for which
both wildlife and the rural poor suffer
considerable costs (8). While it may be
tempting to generalize across biogeo-
graphic realms, the billion-strong African
population is expected to quadruple this
century (9), with rising demands for land
and increased potential for human-wildlife
conflict. There is little evidence that
large, sometimes dangerous, animals can
successfully move through agricultural
landscapes in the absence of fences, and
it would be unwise to assume that islands
of irreplaceable biodiversity would remain
intact should fencing be removed.
Fences should be recognized as a
fundamental conservation tool that may
often be the best option for a specific set of
circumstances. Decisions on fencing must
be based on context-dependent evaluation
of all alternatives, rather than dismissed as
a last resort.
M. Pfeifer,1,2 C. Packer,3 A. C. Burton,4,5
S. T. Garnett,6 A. J. Loveridge,7
D. MacNulty,8 P. J. Platts2
1Forest Ecology and Conservation Lab, Department
of Life Sciences, Imperial College London, Ascot,
SL5 7PY, UK. 2 York Institute for Tropical Ecosystems,
Environment Department, University of York,
York, YO10 5DD, UK. 3Department of Ecology,
Evolution and Behavior, University of Minnesota,
St. Paul, MN 55108, USA. 4Alberta Innovates
Technology Futures, Victoria, BC V8Z 7X8, Canada.
5Department of Biology, University of Victoria,
Victoria, BC V8W 2 Y2, Canada. 6Research Institute
for the Environment and Livelihoods, Charles
Darwin University, Darwin, N T 0909, Australia.
7Wildlife Conservation Research Unit, Department
of Zoology, Oxford University, Oxford, OX13 5QL,
UK. 8Department of Wildland Resources, Utah State
University, Logan, UT 84322, USA.
1. S. H. M. Butchart et al ., Science328, 1164 (2010).
2. C. Packer et al. , Ecol. Lett. 16, 635 (2013).
3. C. Packer et al., Ecol. Lett. 16, 1414-e4 (2013).
4. A. G. Bruner, R. E. Gullison, R. E. Rice, G. A. B. da Fonseca,
Science 291, 125 (2001).
5. B. Fisher, Nat. Geosci. 3, 375 (2010).
6. M. Trinkel etal.,
7. M.J.Somers,M.Hayward, Fencing for Conservation
(Springer, New York, 2012).
8. A.Balmford,T.Whitten,Oryx 37,238(2003).
9. United Nations, ESA/P/WP.227 (2013); http://esa.un.org/.
THE CONCLUSION OF PFEIFER et al.—that
wildlife fencing should be context-
dependent—echoes our own call for
fencing decisions to be based on realistic
assessments of the costs and benefits. We
did not, as Pfeifer et al. suggest, state that
fencing impacts were invariably nega-
tive, nor did we express a view that fence
removal was imperative.
Pfeifer et al. emphasize circumstances
in which fences encircle isolated wildlife
areas embedded in a matrix of human
activity. However, as stated in our
Perspective, many fences are constructed
within contiguous wildlife habitat. Some of
these fences are constructed for conser-
vation purposes (e.g., to contain rhinos
within a well-guarded area) and some
serve other purposes (e.g., to delineate
private property). Whatever their purpose,
the resulting barriers to wildlife move-
ment will have environmental impacts
that should be considered when deciding
whether to construct or remove fences.
We agree with Pfeifer et al. that appropri-
ately designed and well-maintained fences
may contribute to wildlife conservation in
small areas that are irretrievably isolated
by human development. However, even
in these circumstances, the likely benefits
and costs need to be assessed carefully. As
Edited by Jennifer Sills