Content uploaded by David G McGrath
Author content
All content in this area was uploaded by David G McGrath
Content may be subject to copyright.
4 DECEMBER 2009 VOL 326 SCIENCE www.sciencemag.org
1350
POLICYFORUM
Brazil has two major opportunities to
end the clearing of its Amazon for-
est and to reduce global greenhouse
gas emissions substantially. The fi rst is its for-
mal announcement within United Nations cli-
mate treaty negotiations in 2008 of an Amazon
deforestation reduction target, which prompted
Norway to commit $1 billion if it sustains prog-
ress toward this target ( 1). The second is a wide-
spread marketplace transition within the beef
and soy industries, the main drivers of defor-
estation, to exclude Amazon deforesters from
their supply chains ( 2) [supplementary online
material (SOM), section (§) 4]. According to
our analysis, these recent developments fi nally
make feasible the end of deforestation in the
Brazilian Amazon, which could result in a 2 to
5% reduction in global carbon emissions. The
$7 to $18 billion beyond Brazil’s current bud-
get outlays that may be needed to stop the clear-
ing [a range intermediate to previous cost esti-
mates ( 3, 4)] could be provided by the REDD
(Reducing Emissions from Deforestation and
Forest Degradation) mechanism for compen-
sating deforestation reduction that is under
negotiation within the UN climate treaty ( 5), or
by payments for tropical forest carbon credits
under a U.S. cap-and-trade system ( 6).
Deforestation History
Brazil has been the world leader in tropical
deforestation, clearing an average of 19,500
km2/year from 1996 to 2005. This forest con-
version to pasture and farmland released 0.7 to
1.4 GtCO2e (billion tons of CO2 equivalents)
per year to the atmosphere ( 7) (SOM, § 1). In
2008, the Brazilian government committed to
reducing deforestation to 20% of the histori-
cal (1996–2005) rate by 2020 ( 8) (SOM, § 2),
motivated by plummeting rates of forest clear-
ing. From July 2005 to July 2009, deforesta-
tion declined to 36% of its historical levels
(see fi gure, above). To help achieve this reduc-
tion, Brazil expanded the network of Amazon
protected areas from 1.26 to 1.82 million km2;
the network now contains 51% of the region’s
remaining forest area ( 9) (table S4). Federal
campaigns to publicize and cancel credit for
illegal land holdings, to pressure buyers of
Amazon products, and to imprison illegal
operators may have contributed to the decline,
as did a retraction of the region’s cattle and soy
industries (SOM, § 3, and fi g. S1).
Steps to End Deforestation
For Brazil to build upon its success and end
deforestation, even if the profi tability of Ama-
zon cattle ranching and soy farming soar in
the coming years, it must support low-defor-
estation livelihoods for forest peoples and
smallholder farmers, expand the law-abiding
“responsible” fraction of the cattle and soy
sectors, improve law enforcement, and effec-
tively manage protected areas.
Indigenous groups and traditional forest
communities, totaling 420,000 people, have
defended their perimeters from incursions by
deforesters ( 9, 10), but have never received
compensation for this enforcement service.
There are also 400,000 smallholder farms
(up to 100 ha) ( 11) established in forested or
marginal lands that could shift to low-defor-
estation production systems.
Cattle ranching, associated with four-fi fths
of Amazon deforestation, must stabilize and
intensify on a diminishing area of pasture-
land, ceding space to a modest expansion of
relatively lucrative soy production (SOM, §
3). Support within the cattle and soy sectors
for declining deforestation could be strength-
ened by identifying, rewarding, and expanding
The End of Deforestation in the
Brazilian Amazon
ENVIRONMENT
Daniel Nepstad,
1 ,2 ‡ Britaldo S. Soares-Filho,
3 ‡ Frank Merry,
1 ,2 * André Lima,
2 Paulo Moutinho,
1, 2
John Carter,
4 Maria Bowman,
1, 2 † Andrea Cattaneo,
1 Hermann Rodrigues,
3 Stephan Schwartzman,
5
David G. McGrath,
1 ,2,6‡ Claudia M. Stickler,
1 ,2,7 Ruben Lubowski,
5 Pedro Piris-Cabezas,
5 ,8
Sergio Rivero,
6 Ane Alencar,
2 ,7 Oriana Almeida,
2 ,6 Osvaldo Stella
2
Government commitments and market
transitions lay the foundation for an effort
to save the forest and reduce carbon emission.
PHOTO CREDIT: ISTOCKPHOTOS
*Present address: Gordon and Betty Moore Foundation,
Palo Alto, CA 94304, USA.
†Present address: University of California, Berkeley,
Berkeley, CA 94720, USA.
‡Authors for correspondence. E-mail: dnepstad@whrc.org
and britaldo@csr.ufmg.br
1Woods Hole Research Center, Falmouth, MA 02540, USA.
2Instituto de Pesquisa Ambiental da Amazonia, Brasília-DF
71.503-505, Brazil. 3Universidade Federal de Minas Gerais,
Belo Horizonte, MG 31270-901, Brazil. 4Aliança da Terra,
Goiânia, GO 74.670-600, Brazil. 5Environmental Defense
Fund, Washington, DC 20009, USA. 6University of Florida,
Gainesville, FL 32611–7315, USA. 7Universidad Rey Juan
Carlos, Tulipán s/n, 28933, Móstoles, Spain. 8Universidade
Federal do Pará, 66.075-110, Belém, Pará, Brazil.
Average annual deforestation, 1996–2005
Projected deforestation
(Business-as-usual)
Brazilian government target
Target to end deforestation
Historical deforestation
40
35
30
25
20
15
10
5
0
2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020
Yea r
Area (103 km2/year)
Historical deforestation in the Brazilian Amazon and future deforestation under three scenarios.
The fi rst scenario simulates deforestation from 2005 into the future under business-as-usual conditions that
assume economic trends and governance levels through 2003 ( 14). The intermediate curve is the current
deforestation reduction target of the Brazilian government ( 8), and the lower curve, which ends deforestation
in 2020, is the scenario analyzed here (SOM, § 2).
Published by AAAS
on December 3, 2009 www.sciencemag.orgDownloaded from
www.sciencemag.org SCIENCE VOL326 4 DECEMBER 2009 1351
POLICYFORUM
the pool of “responsible” producers striving to
comply with the law and to practice good land
stewardship. Legal compliance could be facili-
tated through approval and implementation of
land-use zoning plans, which lower the legal
forest reserve requirement on private proper-
ties in farming and ranching regions ( 12). This
requirement was abruptly raised from 50 to
80% of each property in 1996 without effec-
tive mechanisms for facilitating compliance
( 2) (SOM, § 6). The substantial fl ow of fed-
eral farm credit could be redirected toward the
intensifi cation of cattle production and sup-
port for forest-based economies (SOM, § 7).
Market exclusion of deforesters ( 2) could be
strengthened through government measures
that penalize companies and banks that indis-
criminately do business with Amazon farmers
and cattle ranchers.
Some farmers and ranchers will need
compensation for the opportunity costs
incurred in maintaining private forests. Five
landholder compensation qualifi cation cri-
teria could be used, including forest cover
beyond 50% of the property (SOM, § 6).
What Will It Cost and Who Will Pay?
We estimated the potential cost of a 10-year
program for ending deforestation (see fi g-
ure, page1350). Using spatially explicit eco-
nomic models and programmatic estimates,
we assess budgetary costs of ending defores-
tation assuming that the benefi ts of reduced
deforestation outweigh the opportunity costs
to society. These benefi ts include reduced for-
est fi re, air pollution, fl ooding, biodiversity
loss, soil erosion, and, perhaps, rainfall inhi-
bition ( 3, 13). They are diffi cult to quantify
and are largely untreated in most economic
models ( 4), even though they lower the net
costs of reducing deforestation.
Annual investments in community forest-
based economic activities, health, education,
and cultural preservation for the region’s indig-
enous and traditional forest peoples and small-
holder farmers would total $3.6 to $7.2 bil-
lion from 2010 to 2020 (see table, below and
SOM, § 5). The total opportunity cost poten-
tially incurred by landholders is estimated at
$14 billion (table S3 and fi g. S9), or $26 billion
if a minimum forest cover of 60% is imposed
for each Amazon state to avoid rainfall inhi-
bition ( 13) (SOM § 8). However, our estimate
includes only those private forests that would
qualify for compensation, which represent
only 10 to 15% of potential opportunity costs
(see table, below, and SOM, § 6).
Combining these costs with additional
investments in law enforcement and protected
area management gives a total budget of $7 to
$18 billion (see table, below, and SOM, § 9).
Already initiated by the Norway commitment,
this investment could reduce carbon emis-
sions from 2010 to 2020 by ~6 GtCO2e below
the historical baseline and by 12 GtCO2e
below projected emissions (see fi gure, page
1350) ( 14), culminating in annual emissions
reductions that are 2 to 5% of global emis-
sions rates in 2000–2006 (SOM, § 2). Under
a REDD system, as designed in the American
Clean Energy Security Act passed by the U.S.
House of Representatives, reductions under
Brazil’s deforestation target could generate
revenues valued from $37 billion to $111 bil-
lion between 2013 and 2020 ( 6) (SOM, § 10),
providing a margin for expanding the program
to end deforestation.
Ending deforestation in the Brazilian Ama-
zon in 2020 with less than 20% of the forest
cleared (table S4) would be an extraordinary
and extremely difficult achievement, per-
haps unique in the history of frontier expan-
sion. The likelihood of success, however, is
greatly enhanced by state-level programs that
link zoning and property registries with state-
wide deforestation reduction targets (SOM §
11). The Governors’ Climate and Forests Task
Force is working to connect these Amazon
state programs with international emissions
offset programs under development for Cali-
fornia and other U.S. states ( 15). State-level
programs must also eventually link up with
the federal “Amazon Fund,” where the Nor-
wegian commitment resides ( 1). Most tropi-
cal nations will require time to develop Bra-
zil’s institutional capacity, civil society orga-
nization, and legal framework ( 16). Ending
deforestation in the Brazilian Amazon and
reducing it elsewhere in the tropics is a cost-
effective approach to climate change mitiga-
tion with multiple benefi ts ( 13, 16).
References and Notes
1. J. Tollefson, Nature 460, 936 (2009).
2. D. C. Nepstad et al., Conserv. Biol. 20, 1595 (2006).
3. D. C. Nepstad et al., The Costs and Benefi ts of Reducing
Carbon Emissions from Deforestation and Forest Deg-
radation in the Brazilian Amazon (Woods Hole Research
Center, Falmouth, MA, 2007); http://whrc.org/policy/
BaliReports/index.htm.
4. G. E. Kindermann et al., Proc. Natl. Acad. Sci. U.S.A.
105, 10302 (2008).
5. R. E. Gullison et al., Science 316, 985 (2007).
6. P. Piris-Cabezas, R. Lubowski, The Brazilian National Plan
on Climate Change: Potential Impacts in a U.S. Cap-and-
Trade System (Environmental Defense Fund, Washington,
DC, 2009); www.edf.org/documents/10563_Brazil-
ian_national_plan_on_climate_change.pdf.
7. R. A. Houghton, in Tropical Deforestation and Climate
Change, P. Moutinho, S. Schwartzman, Eds. [Amazon
Institute for Environmental Research (IPAM), Belém, Pará,
Brazil, 2005] pp. 13–21.
8. Government of Brazil, Presidential decree 6.263, 21
November 2007, National Climate Change Plan
(Government of Brazil, Brasília, 2008); www.mma.gov.br/
estruturas/smcq_climaticas/_arquivos/plano_nacional_
mudanca_clima.pdf.
9. B. S. Soares-Filho et al., Reducing Carbon Emissions from
Deforestation: The Role of ARPA’s Protected Areas in the
Brazilian Amazon (IPAM, Belém, Pará, Brazil, 2008).
10. D. Nepstad et al., Conserv. Biol. 20, 65 (2006).
11. Instituto Brasileiro de Geografi a e Estatística (IBGE),
Censo agropecuário de 2006; www.ibge.gov.br.
12. A. Lima, Zoneamento ecologico-economico: A luz dos dire-
itos socioambientais (Jurus Editora, Curitiba, Brazil, 2005).
13. C. M. Stickler et al., Glob. Change Biol. 15, 2803 (2009).
14. B. S. Soares-Filho et al., Nature 440, 520 (2006).
15. Governors’ Climate and Forests Task Force, www.
climatechange.ca.gov/forestry_task_force/index.html.
16. A. Angelsen et al., Reducing Emissions from Deforesta-
tion and Forest Degradation (REDD): An Options Assess-
ment Report (Meridian Institute, Washington, DC, 2009);
www.redd-oar.org/ (accessed 14 September 2009).
17. Supported by the National Science Foundation, the
National Aeronautics and Space Administration
(LBA-ECO), the Gordon and Betty Moore Foundation, the
David and Lucille Packard Foundation, the Electric Power
Research Institute, Linden Conservation Trust, the Blue
Moon Foundation, and the Global Opportunities Fund,
government of the United Kingdom. K. Schwalbe assisted
with editing and graphics; P. Moreira, Foster Brown, and
three anonymous reviewers provided comments on an
earlier draft.
Region or state
Forest peoples‘
fund
(106 U.S. $)
Enforcement and
landholder compensation
(106 U.S. $)
Estimated costs of a program to end deforestation in the Brazilian Amazon
Protected area
management
(106 U.S. $)
Total cost
(106 U.S. $)
Brazilian Amazon
Acre
Amapá
Amazonas
Maranhão
Mato Grosso
Pará
Rondonia
^
Roraima
Tocantins
3,606
252
68
565
189
335
1,357
580
116
147
7,213
503
135
1,129
377
669
2,715
1,159
231
293
Low High Low High Low High Low High
1,459
106
13
229
13
693
280
94
27
4
6,502
147
12
116
248
4,135
639
1,127
19
60
1,456
54
56
546
10
80
488
79
90
51
4,368
163
168
1,639
31
240
1,464
238
271
154
6,521
412
136
1,340
212
1,107
2,125
752
233
202
18,082
813
315
2,884
656
5,044
4,818
2,524
522
507 Supporting Online Material
www.sciencemag.org/cgi/content/full/325/5945/1350/DC1
10.1126/science.1182108
Ending deforestation in the Brazilian Amazon by 2020. These estimates for costs incurred from 2010 to
2020 assume that current budgetary outlays from the Brazilian government continue. (SOM § 9)
Published by AAAS
on December 3, 2009 www.sciencemag.orgDownloaded from