RESEARCH ARTICLE SUMMARY
Full article and list of author affiliations:
science.adr2118 Science 5 March 2026 1009
CLIMATE
Democratizing climate change mitigation pathway using modernized stabilization wedges
Nathan Johnson and Iain Staffell*
INTRODUCTION: Mitigating climate change is arguably society’s greatest challenge. Deep-decarbonization pathways envision radical transformations in how we produce and consume energy, goods, and services. Integrated assessment models have produced thousands of cost-optimal pathways, underpinned by millions of assumptions. Enacting any pathway requires broad societal buy-in; however, the barriers to producing and interpreting these pathways exclude most people from the conversation, sidelining
societal preferences and debate. In this work, we complementhese models with a simple, inclusive framework for comparingdiverse mitigation strategies and constructing decarbonizationpathways that reflect personal priorities and values.
RATIONALE: In 2004, Pacala and Socolow introduced the stabilization
wedges, a seminal framework for constructing, comparing,
and communicating decarbonization pathways. Since then,
the climate problem has shifted: Global greenhouse gas (GHG)
emissions have continued to rise, targets to limit warming have
been strengthened, and new climate solutions have emerged.
What is typically considered a mitigation strategy must expand
beyond technological fixes to include behavioral change and
nature-based
solutions, which are more difficult to represent
within cost-optimizing
frameworks.
RESULTS: We define a wedge as any activity that can scale linearly
over 30 years to avoid 2 gigatonnes of CO2 equivalent (GtCO2e) per
year by 2050 (~4% of global GHG emissions). Wedges provide a
standard unit to compare mitigation strategies and link deployment
to temperature outcomes. Limiting warming to 1.5°C
requires around 20 wedges in addition to the 17 wedges that
current policies are expected to deliver.
We identified 36 strategies that span electricity generation,
industry, transport, buildings, land, and food, each capable of
achieving at least one wedge, and quantified the deployment
needed globally by 2050. Technological solutions are central and
include building wind, solar, or nuclear power (~7% of global
electricity); deploying electric vehicles (~20% of passenger land
transport); installing heat pumps (~40% of buildings); and
capturing carbon (~90% of cement plants). Less examined options
address unsustainable consumption, such as reducing meat in
diets (~30%), food waste (~50%), and air travel (~70%). Natural
carbon sinks provide many options, including the expansion of
forests (~7% of tropical or ~20% of temperate), planting trees on
croplands (~40 or 80%) or pastures (~30 or 60%), and managing
agricultural soils (~60% of global cropland). Many strategies can
achieve multiple wedges, but all are constrained by technical,
biophysical, and/or socioeconomic limits. Even so, 20 wedges can
be delivered in ~6.9 trillion combinations, allowing pathways to
prioritize social acceptance and cobenefits, alongside cost.
To reveal where consensus exists in mainstream thinking, and
where society might wish to rebalance effort, we translated
hundreds of decarbonization pathways from integrated assessment
models into wedges. The exact mix of strategies varies widely, but
mitigation is generally concentrated in electricity generation (38%)
and industry (26%), relying heavily on renewables (~6 wedges)
and, to a lesser extent, on carbon capture (~2 wedges), whereas
nature-based
and behavioral strategies play limited roles.
CONCLUSION: Climate wedges complement existing tools by
turning a sprawling solution space into a clear list of options
without prescribing a single route to decarbonization. They
provide an accessible planning toolkit: Set a temperature target
and select strategies, weighing up their trade-offs.
The framework
could be downscaled to countries and institutions, and its
revealed preferences could inform future modeling to align
cost-optimal
scenarios with actions that people are more likely
to support.
*Corresponding author. Email: i. staffell@ imperial. ac. uk Cite this article as
N. Johnson, I. Staffell, Science 391, eadr2118 (2026). DOI: 10.1126/science.adr2118
Global decarbonization can be divided into
climate wedges, with 36 strategies able
to deliver at least one wedge. A wedge saves
2 billion tonnes of CO2 equivalent emissions
per year by 2050. Deploying 20 wedges
reduces global emissions by ~80% over
30 years, consistent with the Paris Agreement
goal. Shaded wedges (left) show the median
share of emissions reductions across
integrated assessment model pathways,
colored according to economic sectors (right).
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 1 of 9
CLIMATE
Democratizing climate change
mitigation pathways using
modernized stabilization wedges
Nathan Johnson and Iain Staffell*
Mitigating climate change requires broad societal buy-in.
Integrated assessment models (IAMs) produce cost-optimal
pathways, but these are complex and not easily customized
to reflect individuals’ preferences. Twenty years ago, the
stabilization wedge framework introduced a simpler way to
discuss decarbonization. Here, we modernized this framework,
identifying 36 strategies, each with the potential to mitigate
4% of global emissions by 2050, and quantified their
required scale of deployment. People can build personalized
decarbonization pathways by choosing a portfolio of these
strategies, with more than 6 trillion combinations that are able
to limit global warming to 1.5°C. We assessed which strategies
IAMs favor and found that they prioritize technological over
behavioral and nature-based
solutions, with limited agreement.
This framework empowers a general audience to construct
and debate pathways, by making informed choices that reflect
objectives beyond cost-optimization.
Climate change is an existential threat (1) that requires holistic action
across society to mitigate risks to ecosystems and livelihoods.
Constructive debate around decarbonization must reflect people’s contrasting
priorities. Normal scientific approaches address well-defined
problems, whereas multifaceted and divisive issues such as climate
change require postnormal science (2, 3). This recognizes deep uncertainties
and value conflicts and advocates for an inclusive approach
that extends beyond scientific expertise.
Integrated assessment models (IAMs) are frameworks of interacting
models of the global economy, energy, land, and climate systems, which
produce internally consistent cost-minimizing
technology and policy
scenarios for meeting human needs while limiting greenhouse gas
(GHG) emissions. IAM scenarios heavily influence thinking around
decarbonization because they are precise and holistic, accounting for
many factors and feedbacks that influence mitigation costs. However,
the thousands of scenarios generated offer much complexity and few
actionable insights—features of any one scenario (e.g., building more
nuclear reactors or carbon capture plants) are countered by other,
equally valid, scenarios that favor different approaches (4). Their focus
on cost-minimization
also means that various nonproductive mitigation
strategies (e.g., reducing meat consumption or air travel) are
overlooked (5, 6) and that scenarios are misaligned with citizens’ perspectives
(7, 8). Open-sourcing
models and comprehensive intercomparisons
have increased transparency in IAM research (9), following
criticisms of their “black-box”
nature (9, 10). However, their technology-dominated
focus, inherent complexity, and high barriers to public
usage limit broader engagement (10).
People have diverse perspectives on climate change mitigation (11),
which hinders progress despite strong global support for decarbonization
(12). Although affordability is key, many other criteria (e.g., convenience,
cobenefits, and perceived risks) must be balanced against
cost for wider social acceptance (13). The multidimensional trade-offs
and consequences of any given strategy require widespread discussion
to create societal buy-in,
prompting calls for more flexible and inclusive
approaches to complement IAMs (10). Project Drawdown (14) and
online simulators such as the 2050 Calculators (15) and EN-ROADS
(16) provide alternatives, but these have their own limitations. Project
Drawdown is passive and does not allow users to construct personalized
pathways, whereas the system dynamics approach of online simulators
offers precise personalization of pathways, but interactions are
complex and sometimes counterintuitive and thus difficult for nonexpert
users to explain and understand.
This work bridges gaps between existing tools by providing a simple
language and interactive framework for comparing mitigation strategies,
building on the stabilization wedges (17). It embodies postnormal
scientific values, prioritizing simplicity, transparency, and inclusivity
to allow the broadest possible audience to produce and scrutinize
decarbonization pathways. The stabilization wedges have been used
worldwide to engage students in climate change discourse (18); however,
much has changed in the two decades since their publication.
Progress in deploying most strategies has been slower than required
(19), new climate change mitigation strategies have emerged (20), and
greater effort is needed to limit warming to international targets (21).
Here, we modernized the framework to fit today’s context, expanded
the portfolio of mitigation strategies, and explored which strategies
IAMs select to drive decarbonization for context.
A wedge approach to mitigation
No single strategy can reduce global GHG emissions to zero, so we
first divided the problem into smaller mitigation wedges. We define a
“wedge” as an activity that reduces emissions relative to a baseline,
with additional effort that scales up linearly from 2020 to save 2 gigatonnes
of CO2 equivalent (GtCO2e) per year by 2050, thus reducing
cumulative emissions by 30 GtCO2e over 30 years (Fig. 1). Our estimates
of emissions saved by displacing activities account for all GHGs,
including upstream emissions from fuel production and electricity
generation, but exclude embodied emissions from capital infrastructure
(see materials and methods).
The wedges framework can be used to build decarbonization scenarios
as follows. Users first decide how many wedges to achieve,
which depends on their target level of warming or emissions, and how
these would evolve in their baseline scenario without further intervention.
Both are contentious. The 1.5°C target may prove infeasible (22),
implying that less stringent targets should be explorable despite the
complex ethical trade-offs
they create. Baseline scenarios are deeply
uncertain owing to unforeseeable technology shifts and policy developments
(23, 24). Existing climate policies will drive the deployment of
several mitigation strategies (e.g., electric vehicle mandates and renewable
energy targets) and are expected to stem future emissions growth.
If delivered in full, “current policies” contribute 17 wedges relative to
a “no policy” counterfactual, leading to ~2.5°C of warming by 2100. A
further 20 wedges are required to limit warming to 1.5°C, implying
that the number of wedges must be roughly doubled. These additional
wedges will likely be more difficult and costly to deploy because the remaining
sources of emissions become harder to abate (25, 26). Furthermore,
if policies are reversed or fail to deliver their stated impact
(e.g., because incentives are too weak), then baseline emissions will increase
and more wedges will be required.
The user then chooses a portfolio of strategies to deliver the desired
number of mitigation wedges. We identifed 36 strategies with the
potential to achieve a minimum of one wedge and quantified how
widely each must be deployed by 2050 to do so (Fig. 2). Wedges can
be achieved by a broad range of actions: using energy more efficiently,
using cleaner fuels and technologies, changing consumer behaviors,
capturing and storing CO2 emissions, or adopting sustainable and
regenerative land management. For example, reforesting 104 million
ha of tropical forests, which sequester 19.3 tonnes of CO2 (tCO2) per
Centre for Environmental Policy, Imperial College London, London, UK. *Corresponding
author. Email: i. staffell@ imperial. ac. uk
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 2 of 9
ha per year, would save 2 GtCO2 per year and achieve a wedge. Installing
renewable or nuclear power to displace 2800 terrawatt-hours
(TWh) of electricity, which would otherwise have been generated from
the global average fossil-fuel
mix and emitted ~700 gCO2e per kWh,
also achieves a wedge. See materials and methods for more worked
examples.
For all strategies, deployment must be additional to any that occurs
in the chosen baseline scenario. Some strategies address large sources
of emissions and can achieve multiple wedges. Solar and wind power
are widely identified as having the potential to achieve >4 wedges each
(see table S3). Many other strategies address smaller sources, allowing
for only one wedge. There are also constraints on the maximum number
of wedges possible within each sector, and thus some strategies
compete to reduce emissions (e.g., active travel and public transport).
Industry and fuel production produce the largest share of emissions
across 2050 baseline scenarios (see sectors in Fig. 2) and thus can accommodate
the largest mitigation efforts, followed by electricity generation,
transport, and buildings. The land and food sector can be a
net source (as it is today) or sink of emissions, with the potential to
equal industry in achieving up to 18 wedges (27). Nature-based
and
engineered carbon dioxide removal technologies (e.g., reforestation or
direct air capture) remove CO2 from the atmosphere, meaning that they
are not constrained by the scale of addressable emissions. However,
other factors constrain the potential of all strategies, including biophysical,
technical, and economic limits, which Fig. 2 summarizes as
general upper bounds to mitigation potential. Despite the various
constraints on wedge selection, there is considerable flexibility: A target
of 20 wedges could be delivered by 6.9 trillion possible combinations
of strategies.
Interactions between strategies can diminish their impact. Deploying
two related strategies that improve an activity’s efficiency and also
displace it would yield less than two wedges (e.g., combining building
insulation with heat pumps). Many strategies involve electrification
and thus require additional clean electricity to avoid increasing fossil
fuel consumption. Each wedge of heat pumps, direct air capture, electric
vehicles, and clean hydrogen requires an additional 0.4, 0.5, 0.7,
and 1.8 wedges of nuclear or renewable electricity, respectively.
Strategies also influence the competitiveness of others through indirect
interactions; for example, a wedge of reforestation increases
competition for land and thus food and bioenergy costs (28). We scoped
strategies around best-practice
implementation to reduce the influence
of negative interactions. For example, building coal power plants
to charge electric vehicles or clearing rainforests for bioenergy crops
is incompatible with effective emissions reductions. A core strength
of IAMs is their systematic and endogenous handling of interactions.
This framework instead leaves users to evaluate the implications of
interactions to maintain tractability and facilitate individuals’ perspectives.
Background on each strategy’s interactions, strengths and weaknesses
is provided in the “Contextualizing strategies” section of the
supplementary materials.
The climate wedges are designed to inspire debate and are not a rigid
roadmap for decarbonization. They can provide a first-order
translation
of real-world
policies or targets. For example, the Intergovernmental
Panel on Climate Change (IPCC)’s 1.5°C pathways will require >80% of
global electricity to be supplied by renewables by 2050, up from 30%
historically (29) and in our 2050 no-policy pathway. To achieve a wedge,
wind or solar power must provide 6.6% of global electricity in 2050, thus
>7.5 wedges are required. Rather than seeking the precise combination
of mitigation strategies and policies that reduces emissions at the lowest
cost, the wedges allow people to discuss and explore which strategies
they wish to deploy.
Users might consider some strategies infeasible or
find that they disagree with their preconditions. They might oppose strategies
on ethical [e.g., carbon capture and storage (CCS) or nuclear] (30),
cultural (e.g., reducing meat consumption or car travel) (31), or technical
(e.g., hydrogen, direct air capture, or enhanced weathering) (32, 33)
grounds. Users can disregard strategies accordingly and choose alternatives
that reflect their personal preferences, providing a framework
to evaluate the social acceptability of decarbonization pathways.
Options that can achieve a wedge
The proposed strategies are at very different stages of development:
Some have clear precedent and many are scaling rapidly, whereas others
remain niche (Fig. 3). Electricity sector strategies are among the
most mature, reducing emissions by displacing fossil fuel power plants
with low-carbon
alternatives or by capturing their emissions. A wedge
is achieved by generating 2800 TWh of electricity from solar, wind, or
nuclear power in 2050. This equates to just 300 kWh per capita in 2050,
less than 3% of per capita consumption in the United States today (34).
Together, solar, wind, and nuclear power produced 7400 TWh in 2024,
equivalent to 2.6 wedges (34). A wedge of coal-to-
gas
fuel switching would
require increasing global gas generation by 60%, producing 4100 TWh
by 2050. This would require 600 billion cubic meters of natural gas,
roughly one-sixth
of present global consumption (35), and would be incompatible
with deep decarbonization. CCS must be retrofitted at one-fifth
of coal plants or half of natural gas plants (producing 2900 or
7100 TWh, respectively) to achieve a wedge, yet just three commercial
CCS power plants are in operation, producing <10 TWh per year (36).
Bioenergy with CCS (BECCS) plants must generate less electricity to
achieve a wedge (1600 to 2000 TWh in 2050 depending on the feedstock),
as capturing CO2 from combusting biomass actively lowers atmospheric
concentrations, but each wedge requires up to one-third
of present global
biomass supply, raising sustainability concerns (37).
Passenger cars account for almost half of global transport emissions
(38). Halving fuel consumption of gasoline and diesel cars by 2050
Fig. 1. Historical global GHG emissions and a spectrum of stylized future
pathways to 2050, showing the relationship between the number of wedges
deployed and global temperature outcomes. The inset shows the definition of one
wedge of mitigation effort. In the graph, the black solid line shows historical GHG
emissions (70), and the colored lines show stylized future emissions pathways, with
color indicative of mitigation effort. Each pathway reflects the achievement of one
additional wedge relative to the pathway above it. Pathways begin in 2020 (taken as
the average of 2019 and 2021 to remove short-term
impacts of COVID-19)
and run to
2050. The thick lines show three representative pathways abstracted from several
scenarios (65, 71–73): “no policy” is a counterfactual with no climate policies
introduced globally, “current policies” assumes that existing policies are delivered in
full, and “decarbonization” illustrates a trajectory that limits warming to 1.5°C. Labels
show the observed global temperature rise in 2020 (74) and the projected temperature
rise in 2050 and 2100, all relative to the preindustrial period (from 1850 to 1900).
Projected temperatures are approximated from cumulative emissions to 2050 (see
materials and methods), showing the median and 33rd to 67th percentile range.
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 3 of 9
Fig. 2. The 36 mitigation strategies that have the potential to achieve at least one wedge of mitigation and the scale of deployment required for each. Each strategy is
depicted by an icon with a sentence that quantifies the scale of deployment needed to achieve 2 GtCO2e of mitigation in 2050, expressed relative to the global scale in 2050
unless otherwise stated. Strategies are grouped by sector, which is indicated by colored backgrounds. The key on the right explains other elements of the figure. Indicative upper
bounds for how many wedges can be achieved collectively within each sector were calculated by translating sectoral emissions in 2050 from four baseline scenarios into
wedges, with the exception of land use, which was derived from (27). Upper bounds for individual strategies are derived from a meta-review
of mitigation potentials (see
materials and methods). Strategies within industry cannot collectively mitigate all of the sector’s emissions because emissions savings from avoided fuel extraction and
production are accounted for in end-use
sectors and many smaller actions are also needed (see main text). pax, passenger.
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 4 of 9
Fig. 3. Uncertain assumptions and projections of the future influence the effort required to achieve a wedge and by how much each strategy must scale relative to
today. Strategies are grouped according to the sectors in Fig. 2, indicated by vertical, colored headings. Effort refers to the additional deployment required in 2050 to achieve a
wedge, unless otherwise stated. Effort is measured in various units specific to strategies and sectors. Colored bars show the mean effort required across baseline pathways
(where multiple pathways are used) or the median effort across 10,000 Monte Carlo simulations (where a single baseline pathway with uncertain input parameters is used).
Error bars are shown where uncertainty can be quantified, giving the maximum or minimum effort required across baseline pathways or the 5th and 95th percentiles across
Monte Carlo simulations. Yellow circles indicate deployment of each strategy in the most recent year for which data were available, where this can be calculated. Where these
exceed the corresponding bar, the additional effort required is <100% of present deployment.
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 5 of 9
(matching the efficiency of a Toyota Prius) achieves a wedge (39), but
this fleet would still emit 3 to 6 GtCO2e per year. Alternatively, increasing
the share of electric vehicles, biofuels, public transport, or active
(avoided) travel to deliver 17 to 19% of all passenger land transport achieves
a wedge. Effort is similar across these strategies because all are similarly
low-carbon
with best-practice
deployment. As these strategies all displace
car travel (50 to 70% of passenger land transport in 2050), a maximum
of two to four wedges are possible. In the case of electric vehicles,
a wedge also requires ~2000 TWh of clean electricity (0.7 nuclear or renewable
wedges). Active and public transport are well established globally,
and thus per capita usage only needs to double from 2020 to 2050
to achieve a wedge, compared with a factor of 10 for biofuels. However,
these strategies are seeing limited or negative growth (40). Conversely,
the stock of electric vehicles increased sixfold from 2020 to 2024 (41)
and must increase a further 17-fold
to achieve a wedge. Freight transport
can achieve several wedges through a similar mix of options.
Reducing air travel can also achieve a wedge but requires around a 70%
reduction in demand in 2050, comparable to the restrictions introduced
during the COVID-19
pandemic (42). The use of sustainable aviation
fuels could contribute, but present blending must rise from a maximum
of 50 (43) to 75% and be applied to all flights.
Industry features a more diverse group of strategies, owing to the
variety of products and processes. Steel and cement are the most
emissions-intensive
commodities, accounting for half of direct sector
emissions (44). Other industries sum to three to five wedges in 2050
under baseline scenarios, but their diversity means no single measure
can achieve a wedge. CCS is the primary strategy for deep decarbonization
of cement production and needs to be applied to 93% of cement
plants by 2050 to achieve a wedge. CCS is also a candidate for decarbonizing
steel production, alongside clean hydrogen and electricity
(33). Respectively, these must replace 44 or 35% of steel production in
2050 to achieve a wedge, with hydrogen-electric
steel also requiring
~4000 TWh of clean electricity (1.5 nuclear or renewables wedges).
Hydrogen has many other potential uses across the economy (45), but
demand growth is uncertain, increasing from 100 Mt to 200 to 600 Mt
in 2050 (33). A wedge is achieved for each 150 Mt of clean hydrogen
produced in 2050, requiring ~5100 TWh of clean electricity (1.8 nuclear
or renewables wedges) for the 70% produced through electrolysis.
Extracting and producing fossil fuels adds a further four to seven
wedges in 2050 baseline scenarios, which are reduced indirectly by
(and accounted for within) other strategies that displace fossil fuels.
Within this, methane emissions contribute ~2 wedges by 2050, hence
a wedge could also be achieved by direct actions that halve upstream
leakage, venting, and flaring. Refrigerants for heat pumps, air conditioners,
and refrigeration produce up to 13,000 times the warming
effect of CO2 (46, 47), and thus their rapid phase-out
and careful destruction
can achieve a wedge. Using direct air capture to draw CO2 from
air could offset residual emissions from hard-to-
abate
sectors, but the
technology must scale 200,000-fold
from capturing <10 ktCO2 today
to achieve a wedge (48, 49). A wedge of direct air capture also requires
1500 TWh of clean electricity (0.5 nuclear or renewable wedges), which
is equivalent to present consumption by US households (50).
Most emissions from buildings arise from heating and cooling (51),
so improving insulation and the efficiency of heating offer large potential.
One wedge requires roughly doubling insulation levels, which
would halve heat transfer from 1.5 to 0.8 Wm–2 K–1 between 2020 and
2050, a value still five times greater than the Passivhaus standard of
0.15 Wm–2 K–1 (52). Heat pumps can produce negligible emissions, so
a wedge is possible if heat pumps deliver 25 to 60% of global heating
in 2050 using ~1300 TWh of clean electricity (0.4 nuclear or renewables
wedges). Across the developing world, almost a billion households
rely on biomass cookstoves, two-thirds
of which are inefficient
“traditional” stoves (53). Immediately replacing all traditional stoves
with improved stoves would yield 0.85 wedges, so achieving one wedge
requires that new stoves deployed through 2050 are also improved.
Reducing emissions from the land and food sector is a clear priority
because sources are large and interventions are potentially rapid (27).
A wedge can be achieved by halving food loss and waste to one-sixth
of all food produced or by reducing meat consumption across regions
that overconsume (everywhere except for South Asia and sub-Saharan
Africa) by 30%. The latter would lower the mean meat consumption
of these regions to 270 kcal per day, compared with ~90 kcal recommended
on health grounds (54). Reducing agricultural production also
relieves pressure to convert natural ecosystems to farmland (55), which
is another route to achieving wedges. For example, a wedge requires
reducing tropical forest losses (including deforestation) by 40% by
2050, equivalent to 75 Mha over 30 years (an area the size of Pakistan).
A similar reduction was delivered in the Brazilian Amazon by Luiz
Inácio Lula da Silva within 6 months of his presidency (56). A half-wedge
is achieved by phasing out tropical peatland drainage by 2050
with a second half-wedge
from rewetting 18 Mha (an area the size of
the state of Washington). The remaining strategies bolster land-based
carbon sinks to remove CO2 from the atmosphere, with the density of
carbon stored per hectare per year determining the land area required.
Achieving a wedge requires 104 Mha of forest (an area the size of
Colombia) to be reestablished in the tropics by 2050, compared with
904 Mha of croplands (an area the size of the United States) to adopt
soil carbon management. Reestablishing temperate forests and planting
trees in croplands and pastures require areas between these two
extremes. Several of these strategies can theoretically contribute multiple
wedges, but nature-based
carbon storage can increase pressure
on food systems and be reversed by future disturbances, such as fires
or deforestation (27).
IAM results in the language of wedges
IAMs have produced many scenarios for deep decarbonization, which
can provide a potential starting point for constructing pathways with
the climate wedges framework (Fig. 4). We grouped IAM pathways from
the IPCC Sixth Assessment Report (AR6) database (57) into baseline
and mitigation pairs that yield 2050 GHG emissions consistent with
the stylized current policies and decarbonization scenarios in Fig. 1.
Emissions between the paired scenarios differ by 40.4 ± 10.3 GtCO2e
year–1 in 2050, equivalent to 20.2 ± 5.2 wedges. Figure 4A disaggregates
this reduction in GHG emissions across the five broad sectors assessed
in Fig. 2, whereas Fig. 4, B to E, quantifies the additional deployment
of individual strategies (i.e., deployment in mitigation pathways minus
that in their corresponding baseline pathways). Sector-level
mitigation
is measured directly from GHG emissions reductions calculated within
IAMs, and thus reflects differences in scope and assumptions, and
accounts for interactions. Deployment of individual strategies is instead
measured by activity and converted to wedges using the definitions
from Fig. 3, and thus excludes these differences and interactions.
For example, a mean of 33.3 EJ (9250 TWh) of additional wind power
is deployed in decarbonization relative to baseline scenarios, translating
to 3.3 wedges.
Most mitigation effort is achieved within electricity generation
(38%) and industry (26%), compared with transport (12%), land (11%),
and buildings (5%). This sectoral split (Fig. 4A) agrees with the decomposition
of a single pair of IAM scenarios (58), with mean absolute
differences of 8 to 20%, yet is broader in representing 249 to 367 IAM
scenario pairs and incorporating land-use
change. Different subsets of
scenario pairs contribute to our evaluation of each sector and strategy
(hence the sample size varies between bars in Fig. 4), as IAMs do not
report all variables consistently, and so 3.4 GtCO2e year–1 (8%) cannot
be allocated to specific sectors. IAMs tend to prioritize the energy system
over the land system (58, 59), where coarse disaggregation of
land uses and demands for specific agricultural and forestry commodities
limit which strategies are included. Nature-based
strategies
contribute only 2.2 wedges across AR6 scenarios, despite offering
comparable mitigation potentials to those within the energy sector
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 6 of 9
(27, 60). Users of the climate wedges framework should therefore
consider the wider potential of nature-based
mitigation when choosing
strategies.
Notably, 8.8 wedges of clean electricity are deployed (Fig. 4B); however,
2.1 wedges are not additive and are required to power electrification
in industry and transport. The additional wedges required to power
heat pumps could not be calculated because the AR6 database does not
disaggregate mitigation efforts within the buildings sector. Wind and
solar power contribute 3.3 and 3.1 wedges, respectively, reinforcing thinking
from the IPCC (4) and others (61) that renewables are the largest
driver of decarbonization, despite challenges with electricity system integration
(62, 63). CCS totals 2.1 wedges across the five applications we
evaluated (1.3 in power and 0.8 in industry). Across all sectors, nuclear
power and electric vehicles are the only other strategies to contribute
more than a single wedge on average, with clean hydrogen and methane
reduction close behind at 0.95 each.
The AR6 database variables allowed us to evaluate 17 of our 36
strategies, and hence, 10, 21, 40, and 63% of sector-wide
mitigation
cannot be accounted for across electricity, transport, industry, and land
use, respectively. Technological strategies are favored, whereas naturebased
and behavioral strategies are mostly absent because decisions
in IAMs are driven by discounted long-run
system costs, making them
less well suited for modeling nonproductive activities (64). These mismatches
(colorless bars in Fig. 4, B to E) also include decarbonization
by actions with insufficient mitigation potential to be considered as
wedges (e.g., material recycling or solar water heating), those that are
not disaggregated in IAM reporting (e.g., public transport, active travel,
and vehicle efficiency), and interactions between strategies that cannot
be disentangled with available data.
Although IAM scenarios generally agree on the balance of emissions
reductions between sectors, deployment of specific strategies ranges
substantially, with the interquartile range greater than the mean for
10 of the 17 strategies we evaluated. For example, models select very
different combinations of wind, solar, and nuclear power, which respectively
contribute 1.5 to 4.6, 1.3 to 4.5, and 0.1 to 1.9 wedges at the
25th to 75th percentiles. Our disaggregation generally agrees with
A B
C D E
Fig. 4. Translating IAM outputs to wedges shows the aggregate mix of mitigation strategies selected by cost-optimizing
models. Shaded bars show mean differences
between pairs of baseline and decarbonization scenarios from the IPCC AR6 database (57), colored according to the sectors in Fig. 2. Error bars show the interquartile range of
the differences across pairs. n is the number of scenario pairs represented. (A) The reduction in GHG emissions in 2050 in decarbonization relative to baseline scenarios,
measured in terms of GtCO2e abated, attributed to sectors. (B) The additional deployment of individual power-sector
strategies in 2050 in decarbonization relative to baseline
scenarios, translated into mitigation wedges based on the required effort given in Fig. 3. Negative bars with dotted borders indicate the amount of clean electricity that must be
deployed to power hydrogen electrolyzers and electric vehicles, highlighting interactions between strategies. (C to E) The additional deployment of strategies within the
industry, transport, and land-use
sectors. Strategies within the buildings sector cannot be computed because the most granular data available are total demand per fuel.
Colorless bars show the discrepancy between IAM-calculated
sectoral emissions reductions from (A) and the sum of individual bottom-up
strategy calculations, which represent
only around half of all strategies. Error bars for afforestation in (E) indicate that the mean is above the 75th percentile. Details on the pairing of scenarios, the IAM variables used,
and individual pair-wise
results are provided in the materials and methods. HFC, hydrofluorocarbon.
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 7 of 9
other decompositions of annual emissions reductions in 2030 and
2050 and cumulative emissions reductions to 2050, noting differences
in scope (60, 65, 66) (see fig. S30). Renewable energy provides the largest
emissions reductions with smaller contributions from electrification,
hydrogen, and CCS, indicating general agreement across model types
and decomposition techniques. Our decomposition could be broader
and more precise if scenario outputs were reported more comprehensively
and consistently to public databases, and hence, future work could
extend our results accordingly.
The broad set of IAM scenarios we assessed here shows there are
many different cost-effective
pathways to decarbonization. In reality,
the possibility space around future emissions scenarios is even wider
than IAMs imply, as nature-based
and behavioral strategies and noneconomic
drivers of mitigation are frequently overlooked (10). Against
a backdrop of trillions of options, socially accepted decarbonization
pathways can be achieved only through inclusive dialogue, enabled by
frameworks such as the climate wedges.
Discussion and conclusions
An effective global response to climate change requires that emissions
are reduced rapidly across all aspects of society. We broke this
challenge up into more manageable discrete choices. To limit warming
to 1.5°C, 20 strategies must be deployed from the 36 that we
propose, at the scale we estimate. These require profound, often order-of-
magnitude,
changes to be made and must be additional to the 17 wedges
that current policies are set to achieve globally. Deep decarbonization
requires best-practice
deployment of strategies in supportive combinations
to avoid negative interactions. For example, to reduce rather
than displace heating emissions, heat pumps must run on low-carbon
electricity (supported by clean-electricity
wedges) and use climatefriendly
refrigerants (supported by the refrigerant pollution wedge).
The climate wedges framework challenges its user to think holistically
about the interactions, trade-offs,
and synergies that their pathways
create.
IAM scenarios offer many competing views on how to decarbonize
at the lowest cost (4). At the same time, disregarding human preferences
has produced scenarios that are not widely supported by the
public (7). Cost must be a feature, as decarbonization cannot degrade
living standards, but it cannot be the only consideration. More computer
modeling will not generate societal buy-in,
and so science must
now seek to engage society in decarbonization through alternative
means.
A first step is moving from normal scientific reasoning toward postnormal
science, which embodies the wills and wants of the global
public who must support and adopt mitigation strategies (2, 3). Many
of the strategies we identified require that individuals change their behavior
and all require general public support. Fostering informed opinions
in an empowered public relies on people possessing a firm understanding
of the options for decarbonization. Our framework can be understood
broadly and used to quickly construct and debate pathways for
mitigating climate change. Its accessibility and flexibility complement
the precision and complexity of IAMs but requires key simplifications:
stylizing strategies to linear uptake, excluding interactions between
strategies, and pragmatic treatment of uncertainty. Insights from the
wedges on people’s preferences for particular mitigation strategies
could potentially be used to constrain IAMs, combining the strengths
of both frameworks to produce consistent cost-optimal
scenarios that
incorporate society’s values, thus providing a greater chance of acceptance
and adoption.
Each proposed strategy carries many benefits and challenges. Some
are expensive, but others could save money. Some are perceived as
safe, and others as risky. The criteria for appraising strategies are
manifold, and their hierarchy of importance will vary between individuals
and nations. The framework can be applied to other strategies
for reducing emissions, including those less widely discussed. For
example, a wedge could be achieved by empowering women through
health and education, in turn reducing population growth by 0.1% per
year to 2050. We do not explicitly consider “degrowth” strategies (6),
but many of the proposed strategies target consumption (e.g., diet,
food loss, and travel). Likewise, geoengineering measures such as solar
radiation management could be converted into wedge-equivalents
through their impact on reducing global temperatures, raising very
different risks and ethical considerations.
As climate policy is enacted at the national level, producing country-scale
wedges would be a useful complement to our global framework.
This translation first requires resizing the wedge. Scaling wedges in
proportion to population implies equal effort per capita. Taking
Indonesia as an example, 284 million inhabitants scales each wedge
to 70 MtCO2e in 2050. Other indicators such as gross domestic product
could instead reflect the ability to finance mitigation. The second requirement
is context for how many wedges to achieve, which could
use existing emissions projections under current policies and net-zero
pathways. Climate Action Tracker projects that Indonesia’s 2050 emissions
must fall by ~1190 MtCO2e to become 1.5°C compliant, which
implies 17 country-scale
wedges (67). This compares to 20 wedges in
our global framework, reflecting Indonesia’s lower emissions per
capita. Third, the effort required for each strategy must be recalculated.
A first approximation would scale these linearly with the size of a
wedge, meaning, for example, 100 TWh of electricity from wind, 45 Mt
of clean steel, or reforesting 4 Mha. Estimates should be refined with
country-specific
data for displaced activities (e.g., electricity mix and
land carbon fluxes), where available. Finally, the upper bound for each
strategy’s mitigation potential should be adjusted according to the
nation’s resources (e.g., industrial output and land area) to ensusre that
pathways are physically and socioeconomically credible. For example,
Indonesia has lost 1.3 Mha of forest per year since 2000, emitting 1 GtCO2
per year (68), which, if avoided, would deliver ~14 country-scale
wedges.
Modernizing the climate-stabilization
wedges allows a new generation
to engage in contemporary debates about decarbonization and,
through informed dialogue, brings people from different nations and
cultures into closer agreement on how to address climate change.
Materials and methods summary
This study centers on three questions, which users of the climate wedges
framework must consider when constructing a mitigation pathway:
How many mitigation wedges should be deployed, what strategies and
scale of deployment are needed to achieve wedges, and what other factors
influence decisions on the portfolio of wedges to deploy?
Linking wedges to temperatures
We updated the wedge unit to deliver 2 GtCO2e of annual savings by
2050 (fig. S2), approximately preserving the annual increment in effort
of wedges from Pacala and Socolow (17). We represented no-policy,
current-policy,
and deep-decarbonization
pathways in terms of the
number of wedges deployed by simple linearization of projected emissions
from 2020 to 2050. We connected the deployment of wedges to
temperature outcomes using the IPCC AR6 scenario database (57), linking
cumulative GHG emissions to 2050 to temperature response in 2050
and 2100 (fig. S3). The latter shows a nonlinear response due to the path
dependency of post-2050
emissions.
Scoping strategies and their maximum potential
We formed a long list of strategies with wedge-scale
potential by reviewing
the portfolio of strategies from Pacala and Socolow (17) and
Project Drawdown (14) and mining the literature for additional strategies,
notably on carbon dioxide removal, freight transport, and industry.
We retained 36 strategies capable of at least one wedge (with aggregation
where appropriate) after excluding those constrained by policy
saturation or limited headroom (e.g., hydrofluorocarbon alternatives
and space-cooling
efficiency). We triangulated the maximum technical
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 8 of 9
potential for each strategy (table S3) by comparing emissions saved
from addressing the entire market, with integrated potentials from
Project Drawdown (14), Fuss et al. (32) for CO2 removal, Roe et al. (27)
for land use, and the upper-bound
of deployment within the AR6
database (57).
Calculating the effort required for a wedge
We created a framework of calculations for how much activity would
be required to mitigate 2 GtCO2e in 2050 across each strategy (eqs. S1
to S44). These draw on underpinning parameters taken from four
baseline scenarios (tables S9 to S15) emulated within the Global
Calculator (69), chosen for its transparency and granularity of reporting.
The literature was used to supplement these where needed, particularly
for food and land-use
strategies. We calculated the abatement
potential per unit for each strategy (e.g., the GHG saved by driving 1 km
in an electric vehicle powered by clean electricity versus a conventional
gasoline vehicle). These were inverted to give the scale of deployment
required to yield a wedge and normalized to a share of global
activity (e.g., percentage of total passenger kilometers in 2050) (table
S5). Effort was calculated across the four baselines, with the average
and range presented to represent deep uncertainty. For land-based
strategies, uncertainty was quantified, where possible, using Monte
Carlo sampling across distributions derived from the literature. We
quantified two first-order
interactions between strategies. First, the
competition between strategies within addressable markets (e.g., wind
and solar power displacing power generated from fossil fuels) informs
maximum limits on sectoral deployment. Second, we estimated the
amount of additional clean power required for electrification wedges
(electric vehicles, heat pumps, clean hydrogen, and direct air capture).
Other interactions are left to be considered by users of the framework
to preserve flexibility and tractability. Tables S43 to S73 qualitatively
summarize trade-offs,
synergies, saturation risks, costs, and maturity
of the strategies.
Contextualizing IAM deployment as wedges
We took IAM scenarios from the AR6 database (57), creating pairs of
baseline and mitigation runs by model, scenario family, and policy or
technology category. We focused on 381 vetted pairs that approximately
align with our current policy and decarbonization pathways (fig. S7).
We estimated mitigation per sector from the difference in sectoral
emissions within each pair (figs. S8 to S12), and the deployment of
individual strategies from the difference in activity within each pair
(e.g., EJ of electricity produced by renewables). The latter used our
bottom-up
effort definitions to translate differences in deployment to
wedges (figs. S13 to S29).
Materials and methods are available in the supplementary materials.
REFERENCES AND NOTES
1. IPCC, “Climate change 2022: Impacts, adaptation and vulnerability. Contribution of
Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on
Climate Change,” H.-O.
Pörtner et al., Eds. (Cambridge Univ. Press, 2022).
2. S. O. Funtowicz, J. R. Ravetz, Science for the post-normal
age. Futures 25, 739–755
(1993). doi: 10.1016/0016-3287(
93)90022-L
3. T. M. Saloranta, Post-normal
science and the global climate change issue. Clim. Change
50, 395–404 (2001). doi: 10.1023/A:1010636822581
4. IPCC, “Climate change 2022: Mitigation of climate change. Contribution of Working
Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate
Change,” P. R. Shukla et al., Eds. (Cambridge Univ. Press, 2022).
5. J. Fuhrman et al., Ambitious efforts on residual emissions can reduce CO2 removal and
lower peak temperatures in a net-zero
future. Environ. Res. Lett. 19, 064012 (2024).
doi: 10.1088/1748-9326/
ad456d
6. J. Hickel et al., Urgent need for post-growth
climate mitigation scenarios. Nat. Energy 6,
766–768 (2021). doi: 10.1038/s41560-021-
00884-
9
7. G. Xexakis, E. Trutnevyte, Model-based
scenarios of EU27 electricity supply are not
aligned with the perspectives of French, German, and Polish citizens. Renew. Sustain.
Energy Transit. 2, 100031 (2022). doi: 10.1016/j.rset.2022.100031
8. W. Peng et al., Climate policy models need to get real about people – here’s how. Nature
594, 174–176 (2021). doi: 10.1038/d41586-021-
01500-
2;
pmid: 34103720
9. A. Gambhir, I. Butnar, P.-H.
Li, P. Smith, N. Strachan, A review of criticisms of integrated
assessment models and proposed approaches to address these, through the lens of
BECCS. Energies 12, 1747 (2019). doi: 10.3390/en12091747
10. A. Gambhir, G. Ganguly, S. Mittal, Climate change mitigation scenario databases should
incorporate more non-IAM
pathways. Joule 6, 2663–2667 (2022). doi: 10.1016/
j.joule.2022.11.007
11. UN Development Programme, “The Peoples’ Climate Vote” (2021); https://www.undp.
org/publications/peoples-climate-
vote.
12. P. Andre, T. Boneva, F. Chopra, A. Falk, Globally representative evidence on the actual
and perceived support for climate action. Nat. Clim. Change 14, 253–259 (2024).
doi: 10.1038/s41558-024-
01925-
3
13. K. S. Nielsen et al., Improving climate change mitigation analysis: A framework for
examining feasibility. One Earth 3, 325–336 (2020). doi: 10.1016/j.oneear.2020.08.007
14. P. Hawken, Drawdown: The Most Comprehensive Plan Ever Proposed to Reverse Global
Warming (Penguin, 2017).
15. A. Strapasson et al., Modelling carbon mitigation pathways by 2050: Insights from the
Global Calculator. Energy Strategy Rev. 29, 100494 (2020). doi: 10.1016/
j.esr.2020.100494
16. Climate Interactive, “EN-ROADS
Climate Solutions Simulator” (2025); https://en-roads.
climateinteractive.org/.
17. S. Pacala, R. Socolow, Stabilization wedges: Solving the climate problem for the next 50
years with current technologies. Science 305, 968–972 (2004). doi: 10.1126/
science.1100103; pmid: 15310891
18. Carbon Mitigation Initiative, “The Wedges Game” (2015); https://cmi.princeton.edu/
resources/stabilization-wedges/
the-wedges-
game/.
19. N. J. Johnson, R. Gross, I. Staffell, Stabilisation wedges: Measuring progress towards
transforming the global energy and land use systems. Environ. Res. Lett. 16, 064011
(2021). doi: 10.1088/1748-9326/
abec06
20. S. Fawzy, A. I. Osman, J. Doran, D. W. Rooney, Strategies for mitigation of climate change:
A review. Environ. Chem. Lett. 18, 2069–2094 (2020). doi: 10.1007/s10311-020-
01059-
w
21. S. J. Davis, L. Cao, K. Caldeira, M. I. Hoffert, Rethinking wedges. Environ. Res. Lett. 8,
011001 (2013). doi: 10.1088/1748-9326/
8/1/011001
22. R. D. Lamboll et al., Assessing the size and uncertainty of remaining carbon budgets.
Nat. Clim. Change 13, 1360–1367 (2023). doi: 10.1038/s41558-023-
01848-
5
23. N. Grant, A. Hawkes, T. Napp, A. Gambhir, The appropriate use of reference scenarios in
mitigation analysis. Nat. Clim. Change 10, 605–610 (2020). doi: 10.1038/s41558-020-
0826-
9
24. Z. Hausfather, G. P. Peters, Emissions – the ‘business as usual’ story is misleading. Nature
577, 618–620 (2020). doi: 10.1038/d41586-020-
00177-
3;
pmid: 31996825
25. M. J. Kotchen, J. A. Rising, G. Wagner, The costs of “costless” climate mitigation. Science
382, 1001–1003 (2023). doi: 10.1126/science.adj2453; pmid: 38033066
26. T. Mai et al., Getting to 100%: Six strategies for the challenging last 10%. Joule 6,
1981–1994 (2022). doi: 10.1016/j.joule.2022.08.004
27. S. Roe et al., Contribution of the land sector to a 1.5°C world. Nat. Clim. Change 9,
817–828 (2019). doi: 10.1038/s41558-019-
0591-
9
28. S. Chiquier, A. Gurgel, J. Morris, Y. H. H. Chen, S. Paltsev, Integrated assessment of carbon
dioxide removal portfolios: Land, energy, and economic trade-offs
for climate policy.
Environ. Res. Lett. 20, 024002 (2025). doi: 10.1088/1748-9326/
ada4c0
29. IEA, “Renewables 2024” (2024); https://www.iea.org/reports/renewables-2024.
30. M. Poumadère, R. Bertoldo, J. Samadi, Public perceptions and governance of
controversial technologies to tackle climate change: Nuclear power, carbon capture and
storage, wind, and geoengineering. Wiley Interdiscip. Rev. Clim. Change 2, 712–727
(2011). doi: 10.1002/wcc.134
31. A. P. Mayer, E. K. Smith, Multidimensional partisanship shapes climate policy support and
behaviours. Nat. Clim. Change 13, 32–39 (2023). doi: 10.1038/s41558-022-
01548-
6
32. S. Fuss et al., Negative emissions – Part 2: Costs, potentials and side effects. Environ.
Res. Lett. 13, 063002 (2018). doi: 10.1088/1748-9326/
aabf9f
33. N. Johnson et al., Realistic roles for hydrogen in the future energy transition. Nat. Rev.
Clean Technol. 1, 351–371 (2025). doi: 10.1038/s44359-025-
00050-
4
34. Ember, “Electricity Data Explorer” (2025); https://ember-energy.
org/data/electricity-data-
explorer/.
35. Energy Institute, “Statistical Review of World Energy” (2025); https://www.energyinst.
org/statistical-review.
36. GCCSI, “CORE CCS Facilities Database” (2024); https://co2re.co/FacilityData.
37. M. Fajardy, N. Mac Dowell, Can BECCS deliver sustainable and resource efficient negative
emissions? Energy Environ. Sci. 10, 1389–1426 (2017). doi: 10.1039/C7EE00465F
38. IEA, “Net Zero by 2050” (2021); https://www.iea.org/reports/net-zero-
by-
2050.
39. Green Car Guide, “Toyota Prius Review” (2024); https://www.greencarguide.co.uk/
car-reviews-
and-
road-
tests/
toyota-prius/.
40. International Transport Forum, “ITF Transport Outlook 2023” (2023); https://www.
itf-oecd.
org/itf-transport-
outlook-
2023.
41. IEA, “Global EV Outlook 2024” (2024); https://www.iea.org/reports/global-ev-
outlook-
2024.
42. IATA, “Air Passenger Market Analysis: December 2020” (2020); https://www.iata.org/
en/iata-repository/publications/economic-reports/air-passenger-monthly-analysis—
december-2020/.
Downloaded from https://www.science.org on May 05, 2026
Research Article
Science 5 March 2026 9 of 9
43. F. Afonso et al., Strategies towards a more sustainable aviation: A systematic review.
Prog. Aerosp. Sci. 137, 100878 (2023). doi: 10.1016/j.paerosci.2022.100878
44. IEA, “Energy Technology Perspectives 2023” (2023); https://www.iea.org/reports/
energy-technology-
perspectives-
2023.
45. I. Staffell et al., The role of hydrogen and fuel cells in the global energy system.
Energy Environ. Sci. 12, 463–491 (2019). doi: 10.1039/C8EE01157E
46. WMO, “WMO Statement on the State of the Global Climate in 2017” (Geneva, 2018);
47. G. Dreyfus et al., “Assessment of climate and development benefits of efficient and
climate-friendly
cooling” (2020); https://www.ccacoalition.org/resources/assessment-climate-
and-
development-
benefits-
efficient-
and-
climate-
friendly-
cooling.
48. G. Realmonte et al., An inter-model
assessment of the role of direct air capture in deep
mitigation pathways. Nat. Commun. 10, 3277 (2019). doi: 10.1038/s41467-019-
10842-
5;
pmid: 31332176
49. IEA, “Direct Air Capture” (2025). https://www.iea.org/energy-system/
carbon-capture-
utilisation-
and-
storage/
direct-air-
capture.
50. US Energy Information Administration, “Monthly Energy Review” (2024). https://www.
eia.gov/totalenergy/data/monthly/.
51. I. Staffell, S. Pfenninger, N. Johnson, A global model of hourly space heating and cooling
demand at multiple spatial scales. Nat. Energy 8, 1328–1344 (2023). doi: 10.1038/
s41560-023-
01341-
5
52. Passivhaus Trust, “How to Build a Passivhaus: Rules of Thumb” (2015); https://www.
passivhaustrust.org.uk/UserFiles/File/Technical%20Papers/ROT/How%20to%20
build%20a%20Passivhaus_Chapters%201%20to%204(3).pdf.
53. ESMAP, “The State of Access to Modern Energy Cooking Services” (2020); https://
documents1.worldbank.org/curated/en/937141600195758792/pdf/The-State-
of-
Access-
to-
Modern-
Energy-
Cooking-
Services.
pdf.
54. W. Willett et al., Food in the Anthropocene: The EAT-Lancet
Commission on healthy diets
from sustainable food systems. Lancet 393, 447–492 (2019). doi: 10.1016/S0140-6736(
18)31788-4;
pmid: 30660336
55. P. G. Curtis, C. M. Slay, N. L. Harris, A. Tyukavina, M. C. Hansen, Classifying drivers of global
forest loss. Science 361, 1108–1111 (2018). doi: 10.1126/science.aau3445; pmid: 30213911
56. Mongabay, “Deforestation in the Amazon rainforest continues to plunge” (2023); https://
news.mongabay.com/2023/09/deforestation-in-
the-
amazon-
rainforest-
continues-
to-
plunge/.
57. E. Byers et al., AR6 Scenarios Database hosted by IIASA. Zenodo (2022); https://doi.
org/10.5281/zenodo.5886912.
58. S. Mathy, P. Menanteau, P. Criqui, After the Paris Agreement: Measuring the global
decarbonization wedges from national energy scenarios. Ecol. Econ. 150, 273–289
(2018). doi: 10.1016/j.ecolecon.2018.04.012
59. I. Keppo et al., Exploring the possibility space: Taking stock of the diverse capabilities and
gaps in integrated assessment models. Environ. Res. Lett. 16, 053006 (2021).
doi: 10.1088/1748-9326/
abe5d8
60. IPCC, “Summary for policymakers” in “Climate change 2022: Mitigation of climate
change. Contribution of Working Group III to the Sixth Assessment Report of the
Intergovernmental Panel on Climate Change,” P. R. Shukla et al., Eds. (Cambridge Univ.
Press, 2023), pp. 3–48.
61. N. M. Haegel et al., Photovoltaics at multi-terawatt
scale: Waiting is not an option. Science
380, 39–42 (2023). doi: 10.1126/science.adf6957; pmid: 37023188
62. P. J. Heptonstall, R. J. K. Gross, A systematic review of the costs and impacts of
integrating variable renewables into power grids. Nat. Energy 6, 72–83 (2020).
doi: 10.1038/s41560-020-
00695-
4
63. R. McKenna et al., System impacts of wind energy developments: Key research
challenges and opportunities. Joule 9, 101799 (2025). doi: 10.1016/j.joule.2024.11.016
64. C. Wilson et al., Evaluating process-based
integrated assessment models of climate
change mitigation. Clim. Change 166, 3 (2021). doi: 10.1007/s10584-021-
03099-
9
65. IEA, “World Energy Outlook 2023” (2023); https://www.iea.org/reports/world-energy-
outlook-
2023.
66. IEA, “Net Zero Roadmap: A Global Pathway to Keep the 1.5 °C Goal in Reach” (2024);
roadmap-
a-
global-
pathway-
to-
keep-
the-
15-
0c-
goal-
in-
reach.
67. Climate Action Tracker, “Country Assessments: Indonesia” (2025); https://
climateactiontracker.org/countries/indonesia/.
68. Global Forest Watch, “Indonesia deforestation rates and statistics” (2025); https://www.
globalforestwatch.org/dashboards/country/IDN/.
69. Global Calculator, “The Global Calculator” (2024); https://tool.globalcalculator.net/.
70. M. W. Jones et al., National contributions to climate change due to historical emissions of
carbon dioxide, methane, and nitrous oxide since 1850. Sci. Data 10, 155 (2023).
doi: 10.1038/s41597-023-
02041-
1;
pmid: 36991071
71. IPCC, “Climate change 2014: Mitigation of climate change. Working Group III Contribution
of the Fifth Assessment Report of the Intergovernmental Panel on Climate Change,”
O. Edenhofer et al., Eds. (Cambridge Univ. Press, 2014).
72. IEA, “Energy Technology Perspectives 2012” (2012); https://www.iea.org/reports/
energy-technology-
perspectives-
2012.
73. UNEP, “The Emissions Gap Report 2023” (2023); https://www.unep.org/resources/
emissions-gap-
report-
2023.
74. P. M. Forster et al., Indicators of Global Climate Change 2024: Annual update of key
indicators of the state of the climate system and human influence. Earth Syst. Sci. Data
17, 2641–2680 (2025). doi: 10.5194/essd-17-
2641-
2025
75. N. Johnson, I. Staffell, Democratising climate change mitigation pathways using
modernised Stabilisation Wedges (v1.0). Zenodo (2025); https://doi.org/10.5281/
zenodo.16637361.
AC KNOWLEDGMENTS
We thank R. Socolow, R. Gross, J. Skea, J. Rogelj, R. Lamboll, A. Foley, A. Hawkes, and
J. van den Heuvel for useful discussions. Funding: N.J. acknowledges funding from the
Engineering and Physical Sciences Research Council (EPSRC) through a doctoral studentship.
N.J. and I.S. acknowledge funding from the Engineering and Physical Sciences Research
Council, grant EP/R045518/1 Author contributions: Conceptualization: N.J., I.S.;
Methodology: N.J., I.S.; Investigation: N.J., I.S.; Visualization: N.J., I.S.; Funding acquisition:
I.S.; Project administration: N.J.; Supervision: I.S.; Writing – original draft: N.J.; Writing –
review & editing: N.J., I.S. Competing interests: The authors declare that they have no
competing interests. Data, code, and materials availability: All code and data needed to
reproduce the results reported in this paper can be found at Zenodo (75). An interactive web
application that implements the climate wedges framework is available at https://
climatewedges.com. License information: Copyright © 2026 the authors, some rights
reserved; exclusive licensee American Association for the Advancement of Science. No claim
to original US government works. https://www.science.org/about/science-licenses-
journal-
article-
reuse.
This research was funded in whole or in part by the UKRI’s Engineering and
Physical Sciences Research Council (EP/R045518/1), a cOAlition S organization; as required,
the author will make the Author Accepted Manuscript (AAM) version available under a CC BY
public copyright license.
SUPPLEMENTA RY MATE RIALS
science.org/doi/10.1126/science.adr2118
Materials and Methods; Supplementary Text; Figs. S1 to S38; Tables S1 to S73;
References (76–381)
Submitted 20 June 2024; accepted 14 October 2025
10.1126/science.adr2118
Downloaded from https://www.science.org on May 05, 2026
Democratizing climate change mitigation pathways using modernized stabilization
wedges
Nathan Johnson and Iain Staffell
Science 391 (6789), eadr2118. DOI: 10.1126/science.adr2118
Editor’s summary
The most effective strategy to decide how to mitigate anthropogenic climate change is to break the problem down into
pieces. A well-known example of this kind of deconstruction was developed in 2004 by Pacala and Socolow (10.1126/
science.1100103), who identified a collection of independent actions called “stabilization wedges” that used existing
technology to limit atmospheric carbon dioxide concentrations to below 500 parts per million. Johnson and Staffell
updated this scheme with an expanded portfolio of wedges that provide multiple pathways to limit global warming to
1.5°C above preindustrial level, as advocated by the Paris Agreement (see the Perspective by McJeon and Ou). —
Jesse Smith
View the article online
Permissions
Use of this article is subject to the Terms of service
Science (ISSN 1095-9203) is published by the American Association for the Advancement of Science. 1200 New York Avenue NW,
Washington, DC 20005. The title Science is a registered trademark of AAAS.
Copyright © 2026 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim
to original U.S. Government Works
Downloaded from https://www.science.org on May 05, 2026
Santa Cruz Climate Action Network 