NATURE CLIMATE CHANGE | VOL 5 | DECEMBER 2015 | www.nature.com/natureclimatechange 1023
opinion & comment
5PgCyr−1 from the atmosphere, phased in
linearly over the next 10years, and, second,
fossil fuel emissions were held constant for
the next 10years and then reduced linearly
to a level equal to 20% of 2014 emissions by
2050 before further linear reduction to zero
by 2100. In this latter case, the cumulative
reduction in net emissions over the next
50years from fossil fuel use versus from
forest management would be roughly equal.
is estimate for the potential of forest
management to remove carbon from the
atmosphere is based on the gross sources
and sinks of carbon associated with three
types of forest management in the tropics7,10.
First, deforestation and degradation of
forests in tropical regions currently release
about 1PgCyr−1 (and by some estimates,
twice that amount). If deforestation
and degradation were stopped, those
emissions would cease. Second, carbon
is accumulating in secondary forests
recovering from harvests and from swidden
agriculture at rates as high as 3PgCyr−1.
is gross rate of uptake is taking place
now. If the associated gross emissions
from harvests and re-clearing of fallows
were stopped, the accumulation of up
to 3PgCyr−1 in growing forests would
continue for decades before declining as
the forests mature. A third activity, more
challenging than the rst two because
of higher costs per hectare, would be
the re-establishment of forests on lands
previously forested but not currently
used productively. Estimates of the areas
available for reforestation vary11,12, but re-
forestation of 500million hectares could
sequester at least 1PgCyr−1 for decades.
Together, the three activities would reduce
total emissions by as much as 5PgCyr−1
(a reduced source of 1PgCyr−1 and
increased sinks of 4PgCyr−1). e rate of
carbon accumulation in forests diminishes
as forests mature, however, and thus the
4PgCyr−1 sink would last, conservatively
in our analysis, for approximately 50years
(to 2065) before declining linearly to
zero by 2095 (Fig.1). As the large trees
containing most of the above-ground
forest carbon tend to be absent in degraded
forest (as they are sought aer in selective
logging), and large trees can take well over
a century to mature fully, the absorption is
likely to continue at a high level well beyond
a 50-year period13. Even if the re-forestation
of 500million hectares was not undertaken,
less-conservative assumptions on how
long the remaining 3PgCyr−1 sink would
continue yield similar results.
We recognize that implementation of
these strategies has political and economic
challenges, not least as current management
of tropical forests is responsible for net
emissions of carbon, not a net uptake.
We focus on tropical forest management
because they are currently the dominant
net source of terrestrial carbon to the
atmosphere and hold large stores of carbon
in their biomass. Indeed, estimates of the
sum of above- and below-ground carbon
(that is, peat) within tropical forests are
greater than 500PgC (ref.10), about half
the estimates of the carbon within fossil
fuel reserves2,14. Managing tropical forests
to help stabilize and then reduce CO2
concentrations could be implemented
more quickly than the phasing out of fossil
fuels, not needing as much development
(technical, economic, marketing,
infrastructure, and manufacturing and
installation capacity) as the expansion
of renewables to scale. Furthermore,
many tropical forests are already growing
(accumulating carbon) and will continue
for some decades as long as they are not
selectively logged, harvested, burned, or
cleared for agriculture or other uses. e
biophysical eects of tropical forests (for
example, albedo and evapotranspiration)
do not oset the biogeochemical eects
(carbon sequestration), as they may in
boreal forests15. And, nally, the recent
United Nations Framework Convention
on Climate Change (UNFCCC) eorts
to proceed with REDD+ (‘reduce
emissions from deforestation and forest
degradation in developing countries’;
political willingness to manage forests for
climate change mitigation16.
Our focus on tropical forests does not
preclude management of forests and other
ecosystems outside the tropics for additional
removal of carbon from the atmosphere.
Eorts in temperate zone forests and in
agricultural lands, grasslands and wetlands
could increase the amount of carbon
withdrawn from the atmosphere each year,
and the huge stores of carbon in arctic and
boreal systems highlight the need to keep
permafrost from thawing17. In this context,
it is worth noting that, in addition to the
sources and sinks of carbon attributable
to the land management suggested above,
other natural processes on land remove
approximately 25% of the carbon emitted
each year6. e calculations here are based
on the assumption that these natural sinks
on land will continue.
Above-ground carbon in forests
represents a vulnerable pool of carbon,
subject to droughts, res, insects and other
disturbances. us, the management of
forests to accumulate carbon must not delay
or dilute the phasing-out fossil fuel use. On
the contrary, the deliberate accumulation of
carbon on land may be of little long-term
benet if climate change proceeds because
of unrestrained use of fossil fuels, and if
forests, as a consequence, return to being
sources of carbon on a warming and
Despite reference here to forest
management, we note that the potential for
increasing carbon sinks in tropical forests
is also related to changing traditional
agricultural practices, including eliminating
swidden. ere is potential in modifying
agricultural methods to enhance above- and
below-ground storage of carbon within
Used strategically, the removal of
carbon from the atmosphere by tropical
forest conservation and restoration could
help stabilize and then reduce the CO2
concentration in the atmosphere during the
decades needed for an orderly transition
away from fossil fuels, and could play
a role equal to that of timely fossil fuel
elimination in avoiding dangerous climate
change. e Bonn Challenge to re-forest
150million hectares of degraded lands and
the goal of the New York Declaration on
Forests to eliminate deforestation by 2030
are good rst steps. But the timing and
extent of action is critical. Not only should
the restoration of the biosphere happen in
concert with the phasing out of fossil fuels,
but the longer we wait, and the higher the
rate of fossil fuel emissions, the smaller the
potential role of tropical forests in osetting
those emissions. ❐
R.A.Houghton* and Alexander A.Nassikas are at the
Woods Hole Research Center, 149 Woods Hole Road,
Falmouth, Massachusetts 02540-1644, USA.
Brett Byers is at Million Acre Pledge, 310 Hillside
Avenue, Piedmont, California 94611, USA, and
Rainforest Trust, 7078 Airlie Road, Warrenton,
Virginia 20187, USA.
1. Jacobson, M.Z. & Delucchi, M.A. Energy Policy
39, 1154–1169 (2011).
2. Meinshausen, M. etal. Nature 458, 1158–1163 (2009).
3. Pacala, S. & Socolow, R. Science 305, 968–972 (2004).
4. Mackey, B. etal. Nature Clim. Change 3, 552–557 (2013).
5. Houghton, R.A. in Recarbonization of the Biosphere: Ecosystems
and the Global Carbon Cycle (eds Lal, R. etal.) 59–82
6. Le Quéré, C. etal. Earth Syst. Sci. Data 7, 47–85 (2015).
7. Houghton, R.A. Carbon Manage. 4, 539–546 (2013).
8. Berenguer, E. etal. Global Change Biol. 20, 3713–3726 (2014).
9. Richter, D. deB. & Houghton, R.A. Carbon Manage
2, 41–47 (2011).
10. Grace, J., Mitchard, E. & Gloor, E. Global Change Biol.
20, 3238–3255 (2014).
11. Laestadius, L. etal. Unasyl va 238, 47–48 (2011).
12. Dinerstein, E. etal. Conservation Lett. 8, 262–271 (2014).
13. Stephenson, N.L. etal. Nature 507, 90–93 (2014).
14. McGlade, C. & Ekins, P. Nature 517, 187–190 (2015).
15. Pongratz, J., Reick, C.H., Raddatz, T. & Claussen, M.
Geophys. Res. Lett. 37, L08702 (2010).
17. Schuur, E.A. G. etal. Nature 520, 171–179 (2015).
18. Gatti, L.V. etal. Nature 506, 76–80 (2014).
© 2015 Macmillan Publishers Limited. All rights reserved