498 NATURE CLIMATE CHANGE | VOL 5 | JUNE 2015 | www.nature.com/natureclimatechange
opinion & comment
Investing in negative emissions
Guy Lomax, Timothy M. Lenton, Adepeju Adeosun and Mark Workman
Methods of removing CO2 from the atmosphere add vital ﬂexibility to eorts to tackle climate change.
They must be brought into mainstream climate policy as soon as possible to open up the landscape for
innovation and development, and to discover which approaches work at scale.
To achieve the widely held policy target
of limiting average temperature change
to 2°C, integrated assessment models
(IAM) increasingly depend on massive-scale
‘negative emissions’ through biomass energy
with carbon capture and storage (BECCS),
deployed in the second half of this century1–6.
Yet this key technology is technically
immature today, and it is far from clear
whether such large-scale deployment several
decades in the future would be either feasible
or desirable1. Hence a recent Commentary by
Fuss etal. has branded BECCS a potentially
“dangerous distraction”1. But before anyone
dismisses what doesn’t yet exist, we argue
that the best way to determine “how safe
it is to bet on negative emissions in the
second half of this century”1 is to instigate
a policy framework for greenhouse-gas
removal (GGR) and invest in research and
development innovation now.
Two dimensions of ﬂexibility
Scalable GGR approaches bring unique
exibility to the mitigation toolbox. BECCS,
along with other methods for removing
greenhouse gases from the atmosphere,
oer two key dimensions of exibility; they
decouple abatement opportunities from
emissions sources in both space and time7–9.
Decoupling in space allows GGR to
indirectly mitigate emissions from areas of
the energy system that are most dicult or
expensive to decarbonize. Such ‘project-
level’ negative emissions can in principle
bring many benets as a complement to
conventional mitigation eorts, depending
on the direction and ecacy of climate policy
and GGR deployment. Possibilities include
(i) buying time for the development of clean
technologies, the replacement of locked-in
sources, and changes in societal attitudes;
(ii) reducing the total costs of meeting climate
targets by displacing the most challenging
and expensive emissions sources; (iii) making
more aggressive emissions cuts feasible by
simply adding new mitigation options; or
(iv) allowing continuing use of fossil fuels in
certain key sectors such as aviation7,9.
Decoupling in time raises the idea that
GGR theoretically could be deployed at
massive scale to generate global ‘net-negative’
emissions later this century, allowing us to
recover from emitting too much earlier this
century and overshooting CO2 concentration
targets1,4,9. e negative emissions capacity
outlined in the IPCC’s Fih Assessment
Report 3 implies BECCS input of up to
10gigatonnes of CO2 abatement per year
with global net negative emissions from
It is this second dimension of
exibility —decoupling in time — that
Fuss etal. rightly caution against as “betting”
on negative emissions1. ey argue that our
ability to reach or even assess the feasibility of
a late-century, massive-scale BECCS scenario
is severely constrained by (at least) four
main groups of uncertainties surrounding:
(i) access to sucient biomass supply and
storage space for captured CO2; (ii) the
uncertain response of the global carbon cycle;
(iii) relative costs and viability of untested
technology; and (iv) social and political
factors. Similar uncertainties face other GGR
technologies. For example, estimates of the
cost of direct air capture range from below
US$100 to more than US$1,000 per tonne
of CO2 abated10, and approaches based on
interactions with natural systems such as
soils and ocean alkalinity raise concerns over
potential environmental impacts that are not
yet fully understood11–13.
Responses to uncertainty
In the face of such uncertainties it can seem
premature to commit to long-term policy
support. A natural, scientic response is
to call for a substantial interdisciplinary
research agenda to explore and try to
constrain the uncertainties1, so that we
can best assess the future potential of GGR
and guide policy through the remaining
uncertainties. But at such vast scales of global
deployment, and over such long timescales
of technological, political and societal
development, many of the uncertainties
are inherent, and can only ever be loosely
constrained by modelling and research9. is
is well illustrated by, for example, existing
estimates of global sustainable biomass
resource in 2050 to 2100, which range
from around 30×1018J (30EJ) per year
to over 600EJyr-1 depending on assumed
trends in diet, crop yields, land use and
population14. e call to try to constrain the
unconstrainable instead may lead to ‘analysis
paralysis’, losing valuable time and helping to
self-full the prophecy that GGR cannot be
realized at scale.
A central problem is the framing of GGR
as a large-scale, late-century approach that
would inevitably entail major environmental
and social consequences7. is presents
multiple issues for policy, and immediately
polarizes rather than nuances the debate.
It risks both over-emphasizing the need
10. Hegglin, M.I. etal. Nature Geosci. 7, 768–778 (2014).
11. Hansen, J. et al. Rev. Geophys. 48, RG4004 (2010).
12. Morgenstern, O. etal. J.Geophys. Res. 115, D00M02 (2010).
13. Eyring, V. etal. Atmos. Chem. Phys. 10, 9451–9472 (2010).
is Correspondence was inspired by discussions with
third-year Mathematics undergraduates at the University
of Exeter: C. Serjeant, R. Illingworth, G. Beresford,
D. Mehta, M. Stanton and A. Clements. We acknowledge
the modelling groups for making their simulations
available for this analysis, the Chemistry-Climate Model
Validation (CCMVal) Activity for WCRP’s (World
Climate Research Programme) SPARC (Stratospheric
Processes and their Role in Climate) project for
organizing and coordinating the model data analysis
activity, and the British Atmospheric Data Centre
(BADC) for collecting and archiving the CCMVal model
output. We thank T. Shepherd for helpful comments on
the manuscript. is research was supported by the NERC
PROBEC project NE/K016016/1.
A. J. Ferraro*, M. Collins and F. H. Lambert
College of Engineering, Mathematics and
Physical Sciences, University of Exeter, Exeter,
EX4 4QF, UK. *e-mail: email@example.com
© 2015 Macmillan Publishers Limited. All rights reserved
NATURE CLIMATE CHANGE | VOL 5 | JUNE 2015 | www.nature.com/natureclimatechange 499
opinion & comment
for precaution and regulation over the
possible advantages, and portraying
reliance on these highly uncertain scenarios
as an economically optimal policy. By
concentrating on events more than 50years
in the future, it takes the debate away
from current scientic knowledge, global
experience and policy planning horizons,
giving the false impression that eective
policy engagement with GGR in the near
term is of little value or urgency. is
distracts attention from the nearer-term
value that BECCS and other GGR could oer
in supplementing ongoing mitigation eorts
at more modest scales, and the urgency of
the early technical and policy groundwork
necessary to enable future scale-up7.
us, although research to try to constrain
long-term uncertainties is undoubtedly
important, these uncertainties should not
be used to justify inaction on more pressing
near-term technology development and
policy support needs. Indeed, an alternative
response to such uncertainty is to start
learning by doing. BECCS and its component
technologies are at a relatively early stage of
technical development, as are many other
GGR options (Fig.1). Individually, bioenergy
and CCS technology and industries
are themselves at an early stage, and
integrating them poses further challenges
to technical viability and achieving
Advancing from the current state
of technical readiness to maturity and
widespread deployment is a process that
takes many decades. For example, one o-
cited example of successful scale-up of a new
energy technology is the United Kingdom’s
‘Dash for Gas’, the development and
nationwide roll-out of combined-cycle gas
turbine power plants in the 1990s. Even with
heavy, sustained R&D programmes by both
industry and government, it took 30years to
move from the rst plants to a competitive
energy technology17. Given the widespread
remaining research and development
challenges, and the large-scale need for
GGR anticipated several decades from now,
timely research and demonstration of the
technologies are themselves priorities.
Roadblocks to policy engagement
To support this learning-by-doing approach,
early policy engagement is vital, but it is also
confronted by several potential roadblocks.
e task of accounting for the removed
greenhouse gases poses a considerable
challenge to practical policy integration.
Unlike emissions from fossil fuel combustion,
the ows of greenhouse gases involved with
GGR approaches are much more diverse
and less well understood. Especially with
approaches based on ecosystems, soils and
biomass, the greenhouse-gas storage varies
with time and external factors18, making it
dicult to accurately measure the amount
of carbon stored. Risks of emissions through
direct and indirect land-use change also
threaten the eectiveness of biomass-based
GGR, requiring eective ways of quantifying
or minimizing such eects through policy19.
Aquatic crops (micro-/macro-algae)
Land-based energy crops
Direct air capture:
Direct air capture:
GGR technology components
Technology Readiness Level (TRL)
Figure 1 | Schematic diagram showing the Technology Readiness Levels (TRLs) of key science and technology components relevant to leading GGR approaches
of aorestation, BECCS, biochar (from biomass pyrolysis) and direct air capture, according to the authors’ assessment. IGCC, integrated gasiﬁcation combined
cycle. TRLs are a method of characterizing technological maturity from the most basic research (TRL 1) through to full-scale real-world operation (TRL 9).
Many important elements of all GGR technologies are still in research and early demonstration. Technologies often take decades to advance from this stage to
commercial deployment (TRL 9) and widespread scale-up, even with continuous R&D support (see text).
© 2015 Macmillan Publishers Limited. All rights reserved
500 NATURE CLIMATE CHANGE | VOL 5 | JUNE 2015 | www.nature.com/natureclimatechange
opinion & comment
Furthermore, GGR approaches do not
completely separate the greenhouse gases
from the natural carbon cycle, calling
into question the permanence of the
sequestration11,20. is problem ranges from
gradual decay of biochar in soils21, to diuse
leakage of CO2 from geological storage22,
to catastrophic release of forest carbon in a
wildre23. e risks or mechanisms of this
happening are oen poorly understood
and, as with storage itself, monitoring or
quantication of any loss is oen dicult.
ese issues create a challenge to
integrating GGR into international
accounting and accreditation schemes as
well as developing eective policy support
for them, and these challenges need to be
addressed early if the potential of BECCS
and other GGR is to be realized. Indeed,
developing eective and sustainable policy
is likely to require co-evolution and iterative
renement of policies as GGR eorts scale
up over decades, as is currently being seen in
the bioenergy sector24.
The risks of delaying policy engagement
Policy and technology development
undoubtedly take time, but delaying GGR
policy engagement also carries risks. First,
and most practically, it risks missing out
on the near-term and smaller-scale value
of some more mature and economically
attractive GGR options, potentially
including co-ring of biomass in fossil-
fuel CCS plants, sequestration through
biochar production, and carbonation of
mineral wastes15,25,26. Second, excluding
GGR from near-term policy attention
would reduce any incentives for businesses
and research organizations to expend
eort and investment on advancement
of GGR technology, and to engage with
policy to develop suitable support for
GGR-oriented businesses. Enabling such
innovation is essential to realizing the
A nal risk arises from the fact that
policy decisions made today will dene
the context in which the high rates of GGR
deployment anticipated by modelling
will occur in several decades’ time7.
Infrastructure, assets and technology
choices in the energy system, in particular,
can have a lifetime of many decades, and
ongoing development of the bioenergy and
the CCS sectors now with no thought for
their future integration could make roll-out
of BECCS dicult and costly. An eective
policy approach must aim to strike a balance
between the urgent need for policy support
on these key issues and the high level of
current uncertainty, taking low- or no-
regrets steps towards integration of GGR
into policy and near-term development.
A way forward
A rst step forward can come from noting
that the practical and conceptual diculties
in accounting, and to some extent the
uncertainties, are shared to varying degrees
by several emissions reduction technologies
that are currently the focus of policy eorts.
Life-cycle assessment methodologies,
developing guidelines for carbon accounting
in forestry and land-use change, approaches
for reducing risks of indirect emissions
from bioenergy and accounting, monitoring
and liability mechanisms for geological
storage are all transferable to dierent GGR
methods, and these mechanisms can form the
basis for policy integration. ese ongoing
overhauls of emissions accounting across
all sectors represent a good opportunity to
Based on the principles of integration with
mitigation policy and building exibility, we
therefore propose four principles for a high-
level strategy that can be applied in order to
begin to make progress towards successful
• Fund research, development and
demonstration of GGR systems, focusing
on constraining uncertainties, developing
practical accounting methods and bridging
any other gaps between technology
maturity and policy needs. Given the
value of GGR in tackling the most dicult
emissions sources, diverting some funding
from more advanced and speculative clean
energy research may pay o.
• Build up support for low-cost, early
opportunities through existing or new
bottom-up policy mechanisms. Examples
might include subsidies for electricity
generated from early BECCS opportunities
such as biomass co-ring in coal CCS
plants, or inclusion of soil carbon
enhancement or biochar in agricultural
policies. is will help to capture early
opportunities as well as stimulating
development and innovation.
• Commit to full integration of GGR into
emissions accounting, accreditation and
overall policy strategy in the longer term,
including any carbon pricing mechanisms.
is process will undoubtedly be complex,
but the commitment will stimulate
investment, research and long-term
planning for GGR.
• Develop steps to lay the broader
groundwork for future GGR and to
keep the GGR option open, avoiding
lock-out of valuable opportunities. e
rst three principles will go some way
towards achieving this, but there may be
further steps that can be taken that are
specic to each technology and must be
identied through close engagement with
stakeholders. An example of this might be
‘capture-ready’ requirements for bioenergy
plants to ensure that they can be retrotted
with CCS when this option becomes viable.
e challenge of meeting climate targets is
huge, and we will need to make use of every
tool at our disposal. GGR methods that can
extract CO2 from the atmosphere itself can
add vital exibility to the eorts and must
be brought into mainstream climate policy
as soon as possible to open up the landscape
for innovation and development. Eectively
integrating such diverse approaches into
policy will be challenging and complex, and
the principles proposed here only point to the
rst stages of the process. But they represent
essential steps that must be taken if we are not
to miss the opportunity that GGR provides. ❐
Guy Lomax1*, Timothy M. Lenton2,
Adepeju Adeosun3 and Mark Workman4 are at
1Energy Futures Lab, Imperial College London,
Exhibition Road, London SW7 2AZ, UK. 2Earth
System Science Group, College of Life and
Environmental Sciences, University of Exeter,
North Park Road, Exeter EX4 4QE, UK. 3Virgin
Earth Challenge, Virgin Management, 179 Harrow
Road, London W2 6NB, UK. 4Grantham Institute
for Climate Change, Imperial College London,
Exhibition Road, London SW7 2AZ, UK.
1. Fuss, S. et al. Nature Clim. Change 4, 850–853 (2014).
2. e Emissions Gap Report 2013 (UNEP, 2013).
3. IPCC Climate Change 2014: Mitigation of Climate Change
(eds Edenhofer, O. et al.) (Cambridge Univ. Press, 2014).
4. Azar, C., Johansson, D.J. A. & Mattsson, N. Environ. Res. Lett.
8, 034004 (2013).
5. Kriegler, E., Edenhofer, O., Reuster, L., Luderer, G. & Klein, D.
Climatic Change 118, 45–57 (2013).
6. Van Vuuren, D.P. et al. Climatic Change 118, 15–27 (2013).
7. Lomax, G., Workman, M., Lenton, T. & Shah, N. Energy Policy
78, 125–136 (2015).
8. Keith, D.W. Science 325, 1654–1655 (2009).
9. Meadowcro, J. Climatic Change 118, 137–149 (2013).
10. Goeppert, A., Czaun, M., Surya Prakash, G.K. & Olah, G. A.
Energy Environ. Sci. 5, 7833 (2012).
11. McLaren, D. Process Saf. Env iron. Prot. 90, 489–500 (2012).
12. Renforth, P., Jenkins, B.G. & Kruger, T. Energy 60, 442–452 (2013).
13. Hartmann, J. et al. Rev. Geophys. 51, 113–149 (2013).
14. Slade, R., Bauen, A. & Gross, R. Nature Clim. Change
4, 99–105 (2014).
15. Koornneef, J. et al. Int. J.Greenh. Gas Control 11, 117–132 (2012).
16. Gough, C. & Upham, P. Greenh. Gases Sci. Technol.
1, 324–334 (2011).
17. Watson, J., Kern, F. & Markusson, N. Technol. Forecast. Soc. Change
81, 192–204 (2014).
18. Stockmann, U. et al. Agric. Ecosyst. Environ. 164, 80–99 (2013).
19. Khanna, M., Crago, C.L. & Black, M. Interface Focus
1, 233–247 (2011).
20. Boucher, O. et al. WIREs Clim. Change 5, 23–35 (2013).
21. Lehmann, J., Czimczik, C., Laird, D. & Sohi, S. in Biochar for
Environmental Management (eds Lehmann, J. & Joseph, S.) Ch. 11
22. IPCC Special Report on Carbon Dioxide Capture and Storage
(eds Metz, B. et al.) (Cambridge Univ. Press, 2005).
23. Canadell, J.G. & Raupach, M.R. Science 320, 1456–1457 (2008).
24. UK Bioenergy Strategy (Department of Energy and Climate
25. Pratt, K. & Moran, D. Biomass Bioenergy 34, 1149–1158 (2010).
26. Renforth, P., Washbourne, C-L., Taylder, J. & Manning, D.A. C.
Environ. Sci. Technol. 45, 2035–2041 (2011).
© 2015 Macmillan Publishers Limited. All rights reserved