[geo] Potential for perceived failure of stratospheric aerosol injection deployment

[Andrew Lockley]

https://www.pnas.org/doi/full/10.1073/pnas.2210036119

Potential for perceived failure of stratospheric aerosol injection
deployment
Patrick W. Keys https://orcid.org/0000-0002-7250-1563
patrick.k...@colostate.edu, Elizabeth A. Barnes, Noah S. Diffenbaugh
https://orcid.org/0000-0002-8856-4964, +1 , James W. Hurrell
https://orcid.org/0000-0002-3169-6384, and Curtis M. Bell-1Authors Info &
Affiliations
Edited by William Clark, Harvard University, Cambridge, MA; received June
16, 2022; accepted August 29, 2022
September 27, 2022
119 (40) e2210036119
https://doi.org/10.1073/pnas.2210036119
Data is empty
Vol. 119 | No. 40
Significance
Abstract
Results
Discussion
Conclusions
Methods
Data, Materials, and Software Availability
Acknowledgments
Supporting Information
References
Significance
Even if aggressive mitigation policies are implemented soon, climate change
impacts will worsen in the coming decades. One proposed response is
stratospheric aerosol injection (SAI), which would reflect a small amount
of the sun’s energy back to space, thereby cooling the planet. This
approach is broadly considered to be relatively inexpensive and
straightforward to deploy, and global cooling could occur rapidly. However,
on regional scales, internal climate variability is likely to dominate over
SAI forcing. This means that in the decade after SAI is deployed, many
regions of the world could locally experience even higher temperatures. Our
study provides conceptual insight for the possible perception of the
failure of SAI or other climate mitigation strategies.
Abstract
As anthropogenic activities warm the Earth, the fundamental solution of
reducing greenhouse gas emissions remains elusive. Given this mitigation
gap, global warming may lead to intolerable climate changes as adaptive
capacity is exceeded. Thus, there is emerging interest in solar radiation
modification, which is the process of deliberately increasing Earth’s
albedo to cool the planet. Stratospheric aerosol injection (SAI)—the
theoretical deployment of particles in the stratosphere to enhance
reflection of incoming solar radiation—is one strategy to slow, pause, or
reverse global warming. If SAI is ever pursued, it will likely be for a
specific aim, such as affording time to implement mitigation strategies,
lessening extremes, or reducing the odds of reaching a biogeophysical
tipping point. Using an ensemble climate model experiment that simulates
the deployment of SAI in the context of an intermediate greenhouse gas
trajectory, we quantified the probability that internal climate variability
masks the effectiveness of SAI deployment on regional temperatures. We
found that while global temperature was stabilized, substantial land areas
continued to experience warming. For example, in the SAI scenario we
explored, up to 55% of the global population experienced rising
temperatures over the decade following SAI deployment and large areas
exhibited high probability of extremely hot years. These conditions could
cause SAI to be perceived as a failure. Countries with the largest
economies experienced some of the largest probabilities of this perceived
failure. The potential for perceived failure could therefore have major
implications for policy decisions in the years immediately following SAI
deployment.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.

Anthropogenic climate change, primarily driven by increasing concentrations
of atmospheric greenhouse gasses, has caused Earth’s global mean
temperature to reach its warmest level in at least the last 2,000 y (1).
This global warming may exceed 1.5 °C above preindustrial temperatures
later this decade, at least for a short period of time, and most years are
likely to exceed the 1.5 °C threshold by 2040 across a range of emissions
scenarios (1). By the middle of this century (2041–2060), warming in excess
of 2.0 °C would be reached under intermediate, high, and very high emission
scenarios (1), and current policies have the world on track to warm by
roughly 3.0 °C by the end of the century (2). Moreover, emissions scenarios
that target global temperature stabilization at either 1.5 or 2.0 °C
require net-zero carbon emissions trajectories, which in practice will
necessitate new and enormously scaled-up carbon dioxide removal technology
(3).
In parallel with global policy shortfalls, current levels of warming are
driving substantial impacts on human and natural systems (4). For example,
climate change is already leading to intensification of extreme events such
as extreme heat, heavy rainfall, intense droughts, extreme wildfire
weather, and marine heat waves (4). These and other climate changes are
leading to a broad suite of impacts, such as migration of ecological niches
(5), increases in global tree mortality (6), increases in financial losses
from extremes (e.g., 7), and amplification of existing economic inequality
(8) and social injustices (9). Furthermore, there is the possibility that
biogeophysical tipping points may lead to new states in key Earth systems,
such as irreversible Antarctic ice loss, tropical rainforest dieback, and
slowing ocean circulations (10). These so-called tipping points are highly
uncertain—in terms of whether, when, and how they may occur (1). Despite
this uncertainty, there is paleoclimate evidence that tipping points have
been crossed in the past, and emerging evidence suggests that they could be
crossed as a result of anthropogenic change (11–13).
To possibly grant humanity additional time to sufficiently reduce
greenhouse gas emissions, lessen the existing negative impacts of climate
change, and avoid transgression of irreversible tipping points, there is
renewed interest in developing an international research agenda on solar
radiation modification (SRM)—a speculative form of climate change response
that has the potential to offset human-induced warming by reflecting a
small amount of solar energy back to space before it enters and warms the
planetary environment (14).
There are numerous challenges for advancing SRM science and research.
First, there are substantial ethical questions concerned with committing
future generations to an uncertain technology and the potential burden of
continuing climate intervention well into the future (15) or deciding when
and how to ramp down SRM deployment (16–19). Second, there are important
concerns related to how climate intervention may drive changes in essential
Earth system processes (20, 21). Third, there are concerns that the
negative consequences arising from SRM would disproportionately burden
populations that are systematically already burdened by climate change
impacts, global dispossession of resources, and wealth inequality (22, 23).
Research investigating public opinion has found considerable heterogeneity
in attitudes toward either research or use of climate intervention (24).
In addition to these social challenges, there exist basic scientific
questions about how to distinguish the climate effects of SRM from
anomalies driven by internal variability of the Earth system (25, 26). This
variability can lead to substantial short-term variation in socially
relevant climate phenomena, such as the frequency of extreme hot and cold
spells (27), the severity of drought (28), the path of the midlatitude
storm tracks (29), changes in regional temperature and precipitation (30),
the state of Arctic Sea ice (31), or the strength of tropical modes of
variability such as the El Niño Southern Oscillation (32) or the
Madden-Julian Oscillation (33). Research on the interaction between
human-induced climate impacts, or “signals,” and internal climate
variability, or “noise,” is a critical area of climate change science, not
least for supporting policymakers and the public in navigating the
expectations of climate change action against a backdrop of an internally
varying climate system (34).
Stratospheric aerosol injection (SAI) is the SRM strategy of releasing
particles into the stratosphere to slow, pause, or reverse global warming
(35). While climate simulations provide evidence that the long-term result
of SAI could lead to stabilized global temperatures (17), the impacts of
SAI may be regionally heterogeneous, with temperature and precipitation
varying considerably (36–39). Moreover, internal climatic variability may
mask the short-term perceived effectiveness of SAI; that is, it is possible
that while SAI could successfully stabilize mean global temperatures, the
perceived effectiveness on regional scales may be overwhelmed by local
climatic variability over the short term. Psychologically, a climate
change–related event connects to people’s perceptions most clearly when it
is directly and locally relevant (40, 41). Moreover, people who are
residents of a specific location may tacitly incorporate 10-y trends in
their perception of changes in climate (42). Hence, local changes in
climate—such as continued warming or the occurrence of extreme events—may
cause climate interventions such as SAI to be perceived as a failure. Given
the potential for SAI to abruptly cease and the likelihood of rapid climate
change following such cessation (e.g., 19, 43), the perception of failure
carries particular risks.
If SRM is ever pursued, it will likely be for a specific social or
geophysical aim (22). This may include halting an anticipated geophysical
tipping point [such as accelerated Antarctic ice loss (44), permafrost
melting, or forest die-off] or lessening the impacts of extremes such as
deadly heat waves in large population centers (45). Yet, if climate
variability were to mask the short-term perceived effectiveness of climate
intervention, it could undermine coordinated, international policy action
to address climate change broadly (46). Understanding the masking effects
of climate variability on regional scales will thus be critical for
interpreting the potential perceived success of any SRM strategy in the
immediate years following deployment.
To systematically distinguish the different possible outcomes associated
with the masking effect of internal climate variability, we introduce a set
of archetypal regional responses that could unfold under SAI. These
archetypes are motivated by the fact that in the period prior to SAI
deployment, a given region could be warming or not due to internal climate
variability, even in the context of global-scale warming (47). Similarly,
following deployment, that region could either experience warming or not,
even if the global temperature is stabilized. Thus, we defined four
archetypes of perceived success of climate intervention based on four
categories of pre- and postdeployment experience: 1) Rebound Warming (i.e.,
no warming followed by warming); 2) Continued Warming (i.e., warming
followed by more warming); 3) Stabilization (i.e., no warming either before
or after deployment); and 4) Recovery (i.e., warming followed by no
warming). The phenomena Rebound Warming and Continued Warming could both be
locally perceived as a failure of SAI to deliver on its intended purpose;
hence, throughout the rest of this work, the phrase “perceived failure”
refers to the combination of these two archetypes.
Past research into global SRM strategies employed climate or Earth system
models to simulate how the natural system might respond to different
intervention approaches (48). Here, we leveraged just one of them: the
Assessing Responses and Impacts of SRM on the Earth system with
Stratospheric Aerosol Injection (ARISE-SAI) ensemble carried out with the
Community Earth System Model, version 2 (CESM2) (49). ARISE-SAI simulates a
plausible deployment of SAI, designed to hold global mean temperature at
1.5 °C above preindustrial conditions in the context of the Shared
Socioeconomic Pathway 2 (SSP2)-4.5 future emissions scenario (Fig. 1A)
(49). Extending out to the year 2069, ARISE-SAI includes 10 ensemble
members, each initiated from slightly different initial conditions to
enable quantification of the irreducible uncertainty arising from internal
climate variability (e.g., 50). The 1.5 °C threshold is relevant for global
policy discourse in part because this is a global mean temperature increase
that is considered both an important Earth system threshold as well as a
key focus of global climate policy negotiations enshrined in the United
Nations’ Paris Agreement (51). The fact that ARISE-SAI simulates SAI
deployment that stabilizes global temperature at 1.5 °C while also
representing the effect of internal variability via a substantial number of
ensemble members makes ARISE-SAI a useful testbed for probing the
possibility of perceived failure of climate intervention.
Fig. 1.

Surface temperature trends. (A) Global mean surface temperature. Gray lines
denote individual ensemble members, and the black line denotes the ensemble
mean. (B and C) Ensemble-mean trends over years 2015 to 2034 under SSP2-4.5
(B) and 2035 to 2069 (C) with ARISE-SAI deployment. (D and E) Trends over
the predeployment decade (D) and postdeployment decade (E) for ensemble
member #9. (B–D) The percentage in the bottom of the maps denotes the
percentage of land area that exhibited warming trends as defined in the
text.
Results
Increases in greenhouse gas concentrations and other anthropogenic forcings
under the SSP2-4.5 scenario drove increases in temperatures globally (Fig.
1A), as seen in the forced (ensemble-mean) response during the 2015 to 2034
predeployment period of ARISE-SAI (Fig. 1B). Visualizing the ensemble mean
reduced many of the effects of internal climate variability, even though an
ensemble of more than 10 members is likely needed to fully remove such
effects regionally (e.g., 47, 52). Over the longer postdeployment period of
2035 to 2069, the ensemble mean exhibited a clear picture of temperatures
generally holding steady throughout the rest of the simulation (Fig. 1A),
indicative of SAI acting to stabilize temperatures even regionally (Fig.
1C). In reality, however, any area’s actual climate trajectory will be a
combination of both the forced response and internal climate variability,
which would be analogous to a single ensemble member (Fig. 1 D and E)
rather than the ensemble mean.
Focusing on the decade prior to SAI deployment (“predeployment decade”;
2025 to 2034), any ensemble member (e.g., member #9) will exhibit a large
range of temperature trends regionally under SSP2-4.5 (Fig. 1D), even
though the forced response is overwhelmingly warming. This is because
internal climate variability can drive short-term trends in temperature
that can partially mask (or augment) the longer-term, forced trend. What is
perhaps less appreciated is that internal climate variability can similarly
mask the effects of SAI on a regional scale. In the decade following
continuous SAI deployment (“postdeployment decade”; 2035 to 2044), ensemble
member #9 exhibited warming temperatures over 49% of the land surface (Fig.
1E), where warming is defined as decadal temperature trends larger than 0.1
°C/decade. This trend threshold was chosen to reflect the approximate
warming over the observational record (53); temperature trends less than
this are referred to here as “not warming,” since they capture both cooling
as well as small positive trends. Thus, the effects of internal climate
variability can cause the magnitude of regional warming trends in the
postdeployment decade to far exceed the forced trend from SAI.
Beijing, China, provides an example of how a single region can experience
each of the four archetypal responses under different individual
realizations of the ARISE-SAI experiment (Fig. 2). Ensemble member #1
exhibited the Recovery archetype (Fig. 2D), where SAI would potentially be
labeled a success in that the perception of temperature change would swing
from an increase in local temperature prior to deployment to a
stabilization or decrease in temperature after deployment. However, in
member #4, Beijing experienced Rebound Warming (Fig. 2A), with cooling over
the predeployment period followed by warming over the postdeployment
period. Likewise, in member #7, Beijing experienced Continued Warming (Fig.
2B), with substantial warming during both the pre- and postdeployment
decades.
Fig. 2.

Predeployment and postdeployment surface temperature trends for Beijing,
China. (A–D) Each panel highlights a different ensemble member denoted in
each panel by the thick black line, with the other nine members shown as
thin gray lines. SAI deployment was initiated in the year 2035 (teal
shading). Ten-year linear best-fit lines are shown for 2025 to 2034
(orange) and 2035 to 2044 (teal).
All four archetypal regional responses can be found across the globe, with
varying percentages of the ARISE-SAI ensemble (Fig. 3). While some regions,
notably Australia and parts of Africa, exhibited high probability of the
Recovery archetype (Fig. 3D), substantial parts of the land surface
experienced high probability of either Rebound Warming or Continued
Warming. Repeated occurrence of perceived failure in the same location
across multiple ensemble members can be largely understood as internal
climate variability persistently masking the effect of SAI deployment
(although more than 10 ensemble members would be required to completely
rule out the possibility of a weak, short-term forced response to SAI
itself; Fig. 1C).
Fig. 3.

Archetypal regional responses to ARISE-SAI. (A–D) The percentage of
ensemble members that exhibited specific archetypal responses over the 10 y
pre- and postdeployment: (A) Rebound Warming (not warming followed by
warming), (B) Continued Warming (warming followed by warming), (C)
Stabilization (not warming followed by not warming), and (D) Recovery
(warming followed by not warming).
Aggregating the occurrence of Rebound Warming and Continued Warming across
all ensemble members yielded the probability (computed as the percentage of
the 10 ensemble members) of internal variability leading to perceived
failure of SAI in the ARISE-SAI experiment (Fig. 4 A and B). While some
regions of the planet experienced near-zero probability of perceived
failure under ARISE-SAI deployment, there were other regions that
experienced greater than 50% probability of perceived failure. East
Antarctica—a region of global importance and priority with respect to the
potential for substantial changes in sea level (54)—appeared particularly
prone to climate variability masking the effectiveness of climate
intervention. Likewise, much of northern Eurasia and the western half of
North America experienced a very high probability of perceived failure in
the decade following deployment. For the case of North America, Pacific
Decadal Variability—which CESM is known to simulate with high fidelity
(55)—could be a key factor confounding the effects of climate intervention
(SI Appendix, Fig. S3).
Fig. 4.

Perceived failure over the 10 y following SAI deployment under ARISE. (A)
Probability of perceived failure over the postdeployment period, where the
probability was computed as the fraction of ensemble members exhibiting
warming trends. (B) Probability of a location exceeding its 2015 to 2034
(predeployment) maximum annual-mean temperature in the decade following SAI
deployment (2035 to 2046). (C) Projected number of people at each location
experiencing perceived failure of SAI over the postdeployment period in
ensemble member #9 using projected populations for 2040. Gray denotes
regions not experiencing perceived failure in that particular ensemble
member. (D) Percentage of members with 10% or more of a country’s projected
2040 population (see SI Appendix, Fig. S5 for alternative population
thresholds) experiencing perceived failure following SAI deployment versus
the country’s projected 2040 GDP in units of PPP. Circled area corresponds
to the projected 2040 population experiencing perceived failure averaged
across ensemble members.
We emphasize that these results are specific to ARISE-SAI deployment, which
is only one of many possible SAI deployment scenarios (e.g., 56).
Regardless, they suggest that internal variability in the climate system,
whether arising from random noise in the atmosphere or oceans (57) or from
potentially predictable coupled ocean-atmosphere modes of variability, can
effectively mask SAI deployment.
Our perceived failure metric relies on quantifying decadal temperature
trends. However, given the myriad impacts of extreme heat on natural and
human systems (27, 58), an alternative metric for the perceived
effectiveness of SAI could instead be a measure of the experience of
temperature extremes following deployment. We found that although the
forced response in ARISE-SAI resulted in a stabilization of global
temperatures (Fig. 1 A and C), it is still very likely that record hot
temperatures will occur following deployment (Fig. 4B). For example, for
broad areas of Africa, Eurasia, North America, South America, and
Antarctica, at least 1 y in the decade after SAI deployment was hotter than
the hottest year that occurred in 2015 to 2034. Moreover, the regions
experiencing persistently high perceived failure of SAI (Fig. 4A) did not
directly correspond to the regions experiencing extremely high mean annual
temperatures (Fig. 4B). This finding underlines that multiple climate
metrics are necessary when considering the perceived effectiveness of SAI.
Given the importance of local experiences for informing perceptions of
climate change (40), we next explored the populations exposed to perceived
failure of SAI in the specific ARISE-SAI deployment scenario examined here.
Using gridded population data projected for 2040 in SSP2 (59, 60), we found
that between 10% and 55% of the global population experienced perceived
failure across the 10-member ARISE-SAI ensemble (SI Appendix, Fig. S4). The
most severe example is shown in Fig. 4C for ensemble member #9, where
substantial populations in India, Southeast Asia, the eastern United
States, and West Africa were exposed to the potential of perceived failure
over the decade following ARISE-SAI deployment.
Perceptions of climate change–related phenomena can be related to both
individual local experiences as well as collective sociocultural
experiences (40, 61, 62). Thus, to further explore the socioeconomic
reality of perceived failure of SAI at the national level, we compared the
probability of country-level perceived failure against country-level gross
domestic product (GDP) in 2040 (in units of purchasing power parity; PPP)
(63). All of the largest economies in the world experienced substantial
probability of perceived failure in the postdeployment decade of ARISE-SAI
(Fig. 4D). The implication is that the countries with the most geopolitical
and global economic power—and perhaps those with the most financial
capacity to deploy continuous SAI to manage global temperatures
(64)—experienced at least a 50% probability of large populations being
exposed to the potential of perceived failure of SAI. These countries also
cover substantial land areas, potentially increasing the odds that internal
climatic variability could mask the benefits of SAI. Yet, the fact remains
that the countries that are apparently most prone to high potential of
perceived failure are those with the largest populations and the largest
economies.
Discussion
The “fast” dimension of climate intervention is a notable advantage of SAI
relative to other climate intervention approaches (14, 24). However, we
found that substantial areas of the world could experience warming trends
and extremely hot years, even after 10 y of continuous deployment in the
ARISE-SAI scenario—raising the possibility that SAI may not be perceived
locally as effective. Given the potential social, political, and economic
costs associated with climate intervention and increasing stakes associated
with a warming planet, this gap in time between deployment and local
perceived effectiveness could serve to undermine the fast dimension of SAI
intervention. Moreover, SAI is a technology that could potentially be
deployed quickly by a small group of actors (or a single actor), owing to
its relatively low cost and ease of deployment from a single location on
the planet (e.g., within the borders of a single country) (35, 64).
In light of our findings, several priorities emerge for a forward-thinking
SAI research agenda. First, the prevalence of perceived failure suggests
countries should expect public doubt in the short-term effectiveness of
SAI. The expectation of precise manipulation would be markedly inaccurate
(65). Moreover, different types of SAI deployment scenarios could lead to
different levels of masking (both more and less) of internal climate
variability. However, this issue will also emerge in the midst of more
general mitigation efforts (66), as internal climate variability will
likely produce continued warming in some regions in the years following
aggressive policies aimed at reducing greenhouse gas emissions—potentially
leading to similar perceptions of failure in the climate policy itself
(67). Thus, whether or not SAI is pursued, countries must recognize that
internal climate variability will need to be anticipated and
well-articulated if continued public support is desired. Furthermore, this
articulation must occur amid a communication environment that is already
fraught with climate-related misinformation (68).
To further explore the relevance of the perceived failure archetypes, we
performed a similar analysis using data from the Geoengineering
Large-Ensemble SAI experiment (69). The results provide complementary
insights into SAI deployed under a much higher emissions scenario
(Representative Concentration Pathway 8.5) and different stabilization
targets and deployment year (deployment in the year 2020 with the main aim
to keep global temperatures around 1 °C above preindustrial values).
Because of this, GLENS-SAI represents a much more aggressive SAI scenario
than ARISE-SAI. The GLENS-SAI results (see SI Appendix) again illustrate
the regional significance of internal climate variability and thus further
indicate that the potential for perceived failure will exist across many
different SAI deployment strategies.
Given that specific regions of the planet are predisposed to the effects of
large internal climate variability, such as that produced by the El Niño
Southern Oscillation or the Pacific Decadal Oscillation (70), it is likely
that these regions will also experience persistent masking of SAI
effectiveness. Such understanding of regionally persistent masking of SAI
effectiveness will complement and contribute to the growing literature on
detection and attribution of deployment of climate intervention (25, 26).
Further, because the possibility of perceived failure extends beyond SAI,
knowledge of specific regionally persistent internal variability will
benefit other climate mitigation policies, especially those contingent on
public support (71).
Conclusions
Our results highlight the need for continued research and understanding of
how climate variability may mask climate intervention in the years
immediately following deployment. If climate intervention is ever pursued,
it will likely be for a specific social or geophysical aim. Internal
climate variability, however, may mask the short-term perceived
effectiveness of that intervention, including in the targeted geographical
areas, ecosystems, or economic sectors for which the intervention was
deployed in the first place. Our results thus suggest that the scientific
community must better frame what the success of SAI—and climate
intervention more broadly—looks like in the context of internal climate
variability. Specifically, it will be important to understand how key
global drivers of variability, such as coupled ocean-atmosphere modes
operating on decadal timescales, may mask the intended results of climate
intervention strategies and to what extent this masking will be predictable
or detectable. Our analysis provides a foundation for that understanding
and motivation for improving the ability of global policy and scientific
organizations to better frame the stakes associated with the deployment of
climate intervention in the future.
Methods
ARISE Data.
Gridded, monthly near–surface air temperature fields (variable name TREFHT)
were obtained from the ensemble of simulations performed for the ARISE-SAI
(49). The ARISE ensemble was simulated with the CESM, version 2 (72) using
WACCM6 (Whole Atmosphere Community Climate Model Version 6) (73). We
averaged together the gridded, monthly fields to produce annual-mean
fields, with each field having a grid resolution of 0.94240838 degrees
latitude by 1.25 degrees longitude.
The ARISE dataset includes two sets of simulations composed of 10 ensemble
members each. The first set follows the SSP2-4.5 emissions scenario, while
the second is identical to the first but with the inclusion of SAI
beginning in the year 2035. The location and amount of aerosols released
into the stratosphere each year is determined by a controller algorithm
that works to keep global mean temperature, the north-south temperature
gradient, and the equator-to-pole temperature gradient at values based on
the 2020 to 2039 mean of the SSP2-4.5 simulations with CESM2 (WACCM6) (73).
Further details about the ARISE-SAI configuration and aerosol injection
strategy are provided in ref. 49.
Probability of Perceived Failure.
Decadal trends of annual mean temperature at each grid point were computed
using linear, least-squares regression over two 10-y periods: 1) the
predeployment decade (2025 to 2034) and 2) the postdeployment decade (2035
to 2044). Since SAI under ARISE is designed to stabilize global-mean
temperature (not to reverse the warming trend and induce cooling), we
defined “warming” as any decadal trend that exceeded 0.1 °C per decade. A
warming threshold of 0.1 °C per decade was chosen to reflect the
approximate warming we have thus far experienced over the observational
record (53). All trend magnitudes less than this were considered “not
warming.” We thus classified each of the ensemble members, for each
location, as falling into one of the four archetypes of perceived success
of climate intervention based on the pre- and/or postdeployment trends: 1)
Rebound Warming (i.e., no warming followed by warming); 2) Continued
Warming (i.e., warming followed by more warming); 3) Stabilization (i.e.,
no warming either before or after deployment); and 4) Recovery (i.e.,
warming followed by no warming). The combination of Rebound Warming and
Continued Warming represented the experience of potential perceived
failure, as both exhibited warming trends over the postdeployment decade
that exceeded 0.1 °C per decade. The probability of perceived failure was
then computed as the percentage of ensemble members (out of 10) that
experienced perceived failure at each location.
Populations and Country-Level Statistics for Those Experiencing Perceived
Failure.
Projected, gridded population data for the year 2040 were downloaded from
the Socioeconomic Data and Applications Center (SEDAC) for SSP2 (
https://sedac.ciesin.columbia.edu/data/collection/popdynamics/maps/services).
The SEDAC data were downloaded in netcdf format at a resolution of
one-eighth of a degree and then regridded to the ARISE/CESM2 grid using the
sum function. The global population was perfectly conserved in this
regridding process. The population experiencing perceived failure was then
computed as the sum of the populations at each grid point where the
postdeployment decade exhibited warming trends greater than 0.1 °C.
Projected GDP (in units of PPP) data for the year 2040 under SSP2 were
downloaded as shapefiles from the International Institute for Applied
Systems Analysis at the country level (
https://tntcat.iiasa.ac.at/SspDb/dsd?Action=htmlpage&page=10). Temperature
trends, projected population, and projected GDP were then calculated within
each country boundary using the python packages regionmask and geopandas.
Fig. 4D includes the percentage of members with 10% or more of a country’s
projected 2040 population experiencing perceived failure following SAI
deployment. SI Appendix, Fig. S5 displays results for the same analysis
using alternative population thresholds (i.e., 5%, 10%, 25%, and 50%).
Probability of Exceeding Predeployment Maximum Temperature.
For each grid point, we computed the maximum annual-mean temperature across
all available years prior to SAI deployment (2015 to 2034). This was done
for each ensemble member separately to simulate perceptions within each
individual realization of the climate system. The probability of exceeding
the predeployment maximum temperature was then defined as the number of
ensemble members (out of 10) that exceeded their predeployment maximum in
the decade following deployment (2035 to 2044).
Data, Materials, and Software Availability
All study data are included in the article and/or SI Appendix. The
manuscript will be submitted in parallel to the EarthArXiv preprint server,
under a CC BY 4.0 license. All ARISE and GLENS data are publicly available
(see information for access):
https://www.cesm.ucar.edu/projects/community-projects/ARISE-SAI/ and
http://www.cesm.ucar.edu/projects/community-projects/GLENS/. Population and
GDP data can be downloaded at
https://sedac.ciesin.columbia.edu/data/collection/popdynamics/maps/services
and https://tntcat.iiasa.ac.at/SspDb/dsd?Action=htmlpage&page=10. Code is
available on GitHub at
https://github.com/eabarnes1010/arise_perceived_failure (74) and is
archived on Zenodo at the following DOI:
https://doi.org/10.5281/zenodo.7072436 (75).
Acknowledgments
The views expressed here do not necessarily reflect the positions of the
U.S. Naval War College and the US government. P.W.K., E.A.B., and J.W.H.
were funded by the Defense Advanced Research Projects Agency Grant No.
HR00112290071. N.S.D. was supported by Stanford University.
Supporting Information
Appendix 01 (PDF)
DOWNLOAD
7.92 MB
References
1
IPCC, “Summary for policymakers” in Global Warming of 1.5 °C: An IPCC
Special Report on the Impacts of Global Warming of 1.5 °C above
Pre-industrial Levels and Related Global Greenhouse Gas Emission Pathways,
in the Context of Strengthening the Global Response to the Threat of
Climate Change, Sustainable Development, and Efforts to Eradicate Poverty,
V. Masson-Delmotte et al., Eds. (Cambridge University Press, 2021), pp.3–24.

Google Scholar
2
UNEP, Emissions Gap Report 2021: The Heat Is on—A World of Climate Promises
Not Yet Delivered (United Nations Environment Programme Nairobi, Kenya,
2021).
Go to reference
Google Scholar
3
NASEM, Negative Emissions Technologies and Reliable Sequestration: A
Research Agenda (National Academies Press, 2019).
Go to reference
Google Scholar
4
IPCC, Summary for Policymakers (Cambridge University Press, 2022).

Google Scholar
5
K. S. Sheldon, Climate change in the tropics: Ecological and evolutionary
responses at low latitudes. Annu. Rev. Ecol. Evol. Syst. 50, 303–333 (2019).
Go to reference
Google Scholar
6
H. Hartmann et al., Climate change risks to global forest health: Emergence
of unexpected events of elevated tree mortality worldwide. Annu. Rev. Plant
Biol. 73, 673–702 (2022).
Go to reference
Crossref
PubMed
Google Scholar
7
F. V. Davenport, M. Burke, N. S. Diffenbaugh, Contribution of historical
precipitation change to US flood damages. Proc. Natl. Acad. Sci. U.S.A.
118, e2017524118 (2021).
Go to reference
Crossref
PubMed
Google Scholar
8
N. S. Diffenbaugh, M. Burke, Global warming has increased global economic
inequality. Proc. Natl. Acad. Sci. U.S.A. 116, 9808–9813 (2019).
Go to reference
Crossref
PubMed
Google Scholar
9
T. Dietz, R. L. Shwom, C. T. Whitley, Climate change and society. Annu.
Rev. Sociol. 46, 135–158 (2020).
Go to reference
Google Scholar
10
T. M. Lenton et al., Climate tipping points - too risky to bet against.
Nature 575, 592–595 (2019).
Go to reference
Crossref
PubMed
Google Scholar
11
N. Boers, M. Rypdal, Critical slowing down suggests that the western
Greenland Ice Sheet is close to a tipping point. Proc. Natl. Acad. Sci.
U.S.A. 118, e2024192118 (2021).
Go to reference
Crossref
PubMed
Google Scholar
12
J. Lohmann, P. D. Ditlevsen, Risk of tipping the overturning circulation
due to increasing rates of ice melt. Proc. Natl. Acad. Sci. U.S.A. 118,
e2017989118 (2021).
Crossref
PubMed
Google Scholar
13
S. H. R. Rosier, et al., The tipping points and early warning indicators
for Pine Island Glacier, West Antarctica. Cryosphere 15, 1501–1516 (2021).
Go to reference
Google Scholar
14
NASEM, Reflecting Sunlight: Recommendations for Solar Geoengineering
Research and Research Governance (National Academies Press, 2021).

Google Scholar
15
D. W. Keith, Toward constructive disagreement about geoengineering. Science
374, 812–815 (2021).
Go to reference
Crossref
PubMed
Google Scholar
16
J. C. S. Long, J. G. Shepherd, “The strategic value of geoengineering
research” in Global Environmental Change, B. Freedman, Ed. (Springer
Netherlands, 2014), pp. 757–770.
Go to reference
Crossref
Google Scholar
17
D. G. MacMartin, K. L. Ricke, D. W. Keith, Solar geoengineering as part of
an overall strategy for meeting the 1.5 °C Paris target. Philos. Trans. A
Math. Phys. Eng. Sci. 376, 20160454 (2018).
Go to reference
Crossref
PubMed
Google Scholar
18
A. Jones et al., The impact of abrupt suspension of solar radiation
management (termination effect) in experiment G2 of the Geoengineering
Model Intercomparison Project (GeoMIP). J. Geophys. Res. 118, 9743–9752
(2013).
Crossref
Google Scholar
19
K. E. McCusker, K. C. Armour, C. M. Bitz, D. S. Battisti, Rapid and
extensive warming following cessation of solar radiation management.
Environ. Res. Lett. 9, 024005 (2014).

Crossref
Google Scholar
20
J. F. Tjiputra, A. Grini, H. Lee, Impact of idealized future stratospheric
aerosol injection on the large-scale ocean and land carbon cycles. J.
Geophys. Res. Biogeosci. 121, 2–27 (2016).
Go to reference
Crossref
Google Scholar
21
P. J. Irvine, B. Kravitz, M. G. Lawrence, H. Muri, An overview of the Earth
system science of solar geoengineering. Wiley Interdiscip. Rev. Clim.
Change 7, 815–833 (2016).
Go to reference
Crossref
Google Scholar
22
H. J. Buck, Geoengineering: Re-making climate for profit or humanitarian
intervention? Dev. Change 43, 253–270 (2012).

Crossref
PubMed
Google Scholar
23
J. A. Flegal, A. Gupta, Evoking equity as a rationale for solar
geoengineering research? Scrutinizing emerging expert visions of equity.
Int. Environ. Agreement Polit. Law Econ. 18, 45–61 (2018).
Go to reference
Crossref
Google Scholar
24
A. Mahajan, D. Tingley, G. Wagner, Fast, cheap, and imperfect? US public
opinion about solar geoengineering. Env. Polit. 28, 523–543 (2019).

Crossref
Google Scholar
25
D. G. MacMartin, P. J. Irvine, B. Kravitz, J. B. Horton, Technical
characteristics of a solar geoengineering deployment and implications for
governance. Clim. Policy 19, 1325–1339 (2019).

Crossref
Google Scholar
26
F. Fröb, S. Sonntag, J. Pongratz, H. Schmidt, T. Ilyina, Detectability of
artificial ocean alkalinization and stratospheric aerosol injection in
MPI‐ESM. Earths Futur. 8, e2020EF001634 (2020).

Crossref
Google Scholar
27
N. S. Diffenbaugh et al., Quantifying the influence of global warming on
unprecedented extreme climate events. Proc. Natl. Acad. Sci. U.S.A. 114,
4881–4886 (2017).

Crossref
PubMed
Google Scholar
28
N. S. Diffenbaugh, D. L. Swain, D. Touma, Anthropogenic warming has
increased drought risk in California. Proc. Natl. Acad. Sci. U.S.A. 112,
3931–3936 (2015).
Go to reference
Crossref
PubMed
Google Scholar
29
T. Woollings et al., Daily to decadal modulation of jet variability. J.
Clim. 31, 1297–1314 (2018).
Go to reference
Crossref
Google Scholar
30
C. Deser, Certain uncertainty: The role of internal climate variability in
projections of regional climate change and risk management. Earths Future
10.1029.2020EF001854 (2020).
Go to reference
Crossref
Google Scholar
31
Z. Labe, G. Magnusdottir, H. Stern, Variability of arctic sea ice thickness
using PIOMAS and the CESM large ensemble. J. Clim. 31, 3233–3247 (2018).
Go to reference
Crossref
Google Scholar
32
W. Cai et al., Changing El Niño–Southern oscillation in a warming climate.
Nat. Rev. Earth Environ. 2, 628–644 (2021).
Go to reference
Crossref
Google Scholar
33
Z. Martin et al., The influence of the quasi-biennial oscillation on the
Madden–Julian oscillation. Nat. Rev. Earth Environ. 2, 477–489 (2021).
Go to reference
Crossref
Google Scholar
34
J. S. Mankin, F. Lehner, S. Coats, K. A. McKinnon, The value of initial
condition large ensembles to robust adaptation decision‐making. Earths
Future 8, e01610 (2020).
Go to reference
Crossref
Google Scholar
35
D. W. Keith, “Geoengineering the climate: History and prospect 1” in The
Ethics of Nanotechnology, Geoengineering and Clean Energy, A. Maynard, J.
Stilgoe, Eds. (Routledge, 2020), pp. 207–246.

Crossref
Google Scholar
36
B. Kravitz et al., A multi-model assessment of regional climate disparities
caused by solar geoengineering. Environ. Res. Lett. 9, 074013 (2014).
Go to reference
Crossref
Google Scholar
37
K. L. Ricke, M. G. Morgan, M. R. Allen, Regional climate response to
solar-radiation management. Nat. Geosci. 3, 537–541 (2010).
Crossref
Google Scholar
38
G. A. Ban-Weiss, K. Caldeira, Geoengineering as an optimization problem.
Environ. Res. Lett. 5, 034009 (2010).
Crossref
Google Scholar
39
I. R. Simpson et al., The regional hydroclimate response to stratospheric
sulfate geoengineering and the role of stratospheric heating. J. Geophys.
Res. 124, 12587–12616 (2019).
Go to reference
Crossref
Google Scholar
40
A. Brügger, C. Demski, S. Capstick, How personal experience affects
perception of and decisions related to climate change: A psychological
view. Weather Clim. Soc. 13, 397–408 (2021).

Crossref
Google Scholar
41
C. P. Borick, B. G. Rabe, A reason to believe: Examining the factors that
determine individual views on global warming. Soc. Sci. Quarterly 91,
777–800 (2010).
Go to reference
Google Scholar
42
W. Shao, J. C. Garand, B. D. Keim, L. C. Hamilton, Science, scientists, and
local weather: Understanding mass perceptions of global warming. Soc. Sci.
Q. 97, 1023–1057 (2016).
Go to reference
Crossref
Google Scholar
43
A. Parker, P. J. Irvine, The risk of termination shock from solar
geoengineering. Earths Future 6, 456–467 (2018).
Go to reference
Crossref
Google Scholar
44
J. Garbe, T. Albrecht, A. Levermann, J. F. Donges, R. Winkelmann, The
hysteresis of the Antarctic Ice Sheet. Nature 585, 538–544 (2020).
Go to reference
Crossref
PubMed
Google Scholar
45
C. Mora et al., Global risk of deadly heat. Nat. Clim. Change 7, 501–506
(2017).
Go to reference
Crossref
Google Scholar
46
K. L. Ricke, K. Caldeira, Natural climate variability and future climate
policy. Nat. Clim. Chang. 4, 333–338 (2014).
Go to reference
Crossref
Google Scholar
47
C. Deser, A. Phillips, V. Bourdette, H. Teng, Uncertainty in climate change
projections: The role of internal variability. Clim. Dyn. 38, 527–546
(2012).

Crossref
Google Scholar
48
B. Kravitz et al., The geoengineering model intercomparison project
(GeoMIP). Atmos. Sci. Lett. 12, 162–167 (2011).
Go to reference
Crossref
Google Scholar
49
J. Richter, et al., Assessing responses and impacts of solar climate
intervention on the Earth system with stratospheric aerosol injection
(ARISE-SAI). EGUsphere, 1–35 [Preprint] (2022).

Google Scholar
50
J. E. Kay et al., The Community Earth System Model (CESM) large ensemble
project: A community resource for studying climate change in the presence
of internal climate variability. Bull. Am. Meteorol. Soc. 96, 1333–1349
(2015).
Go to reference
Crossref
Google Scholar
51
Y. Xu, V. Ramanathan, Well below 2 °C: Mitigation strategies for avoiding
dangerous to catastrophic climate changes. Proc. Natl. Acad. Sci. U.S.A.
114, 10315–10323 (2017).
Go to reference
Crossref
PubMed
Google Scholar
52
S. Milinski, N. Maher, D. Olonscheck, How large does a large ensemble need
to be? Earth Syst. Dyn. 11, 885–901 (2020).
Go to reference
Crossref
Google Scholar
53
NOAA National Centers for Environmental Information, State of the Climate:
Monthly Global Climate Report for Annual 2020 (National Oceanographic and
Atmospheric Administration, 2021).

Google Scholar
54
E. Rignot, J. Mouginot, B. Scheuchl, Four decades of Antarctic Ice Sheet
mass balance from 1979–2017. Proc. Natl. Acad. Sci. U.S.A. 116, 1095–1103
(2019).
Go to reference
Crossref
PubMed
Google Scholar
55
A. Capotondi, C. Deser, A. S. Phillips, Y. Okumura, S. M. Larson, ENSO and
Pacific decadal variability in the Community Earth System Model version 2.
J. Adv. Model. Earth Syst. 12, e2019MS002022 (2020).
Go to reference
Crossref
Google Scholar
56
M. J. Mills et al., Radiative and chemical response to interactive
stratospheric sulfate aerosols in fully coupled CESM1(WACCM). J. Geophys.
Res. 122, 13061–13078 (2017).
Go to reference
Crossref
Google Scholar
57
C. Deser, J. W. Hurrell, A. S. Phillips, The role of the North Atlantic
Oscillation in European climate projections. Clim. Dyn. 49, 3141–3157
(2017).
Go to reference
Crossref
Google Scholar
58
K. L. Ebi et al., Extreme weather and climate change: Population health and
health system implications. Annu. Rev. Public Health 42, 293–315 (2021).
Go to reference
Crossref
PubMed
Google Scholar
59
J. Gao, Downscaling Global Spatial Population Projections from 1/8-Degree
to 1-km Grid Cells (National Center for Atmospheric Research, 2017).
Go to reference
Google Scholar
60
J. Gao, Global 1-km downscaled population base year and projection grids
based on the shared socioeconomic pathways. Revision,
https://doi.org/10.7910/DVN/TLJ99B (2020).
Go to reference
Google Scholar
61
O. Renn, The social amplification/attenuation of risk framework:
Application to climate change. Wiley Interdiscip. Rev. Clim. Chang. 2,
154–169 (2011).
Go to reference
Crossref
Google Scholar
62
K. Sambrook, E. Konstantinidis, S. Russell, Y. Okan, The role of personal
experience and prior beliefs in shaping climate change perceptions: A
narrative review. Front. Psychol. 12, 669911 (2021).
Go to reference
Crossref
PubMed
Google Scholar
63
J. Crespo Cuaresma, Income projections for climate change research: A
framework based on human capital dynamics. Glob. Environ. Chang. 42,
226–236 (2017).
Go to reference
Crossref
Google Scholar
64
W. Smith, G. Wagner, Stratospheric aerosol injection tactics and costs in
the first 15 years of deployment. Environ. Res. Lett. 13, 124001 (2018).

Crossref
Google Scholar
65
National Research Council, National Research Council, (U.S.). Division on
Earth and Life Studies, National Research Council (U.S.). Ocean Studies
Board, National Research Council (U.S.). Board on Atmospheric Sciences and
Climate, Committee on Geoengineering Climate: Technical Evaluation and
Discussion of Impacts, Climate Intervention: Reflecting Sunlight to Cool
Earth (National Academies Press, 2015).
Go to reference
Google Scholar
66
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 (Cambridge University Press, 2022).
Go to reference
Google Scholar
67
N. S. Diffenbaugh, E. Barnes, P. Keys, Probability of continued local-scale
warming and extreme events during and after decarbonization (2022).
https://doi.org/10.21203/rs.3.rs-1902791/v1. Accessed 29 July 2022.
Go to reference
Google Scholar
68
S. Lewandowsky, L. Whitmarsh, Climate communication for biologists: When a
picture can tell a thousand words. PLoS Biol. 16, e2006004 (2018).
Go to reference
Crossref
PubMed
Google Scholar
69
S. Tilmes et al., CESM1(WACCM) stratospheric aerosol geoengineering large
ensemble project. Bull. Am. Meteorol. Soc. 99, 2361–2371 (2018).
Go to reference
Crossref
Google Scholar
70
M. Newman, M. A. Alexander, T. R. Ault, The Pacific decadal oscillation,
revisited. J. Clim. 29, 4399–4427 (2016).
Go to reference
Crossref
Google Scholar
71
S. Fankhauser et al., The meaning of net zero and how to get it right. Nat.
Clim. Chang. 12, 15–21 (2021).
Go to reference
Crossref
Google Scholar
72
G. Danabasoglu et al., The Community Earth System Model version 2 (CESM2).
J. Adv. Model. Earth Syst. 12, e2019MS001916 (2020).
Go to reference
Crossref
Google Scholar
73
G. A. Meehl et al., Characteristics of future warmer base states in CESM2.
Earth Space Sci. 7, e2020EA001296 (2020).

Crossref
Google Scholar
74
E. A. Barnes, Data from “Potential for perceived failure of stratospheric
aerosol injection deployment.” GitHub.
https://github.com/eabarnes1010/arise_perceived_failure. Deposited 12
September 2022.

Google Scholar
75
E. A. Barnes, Data from “Potential for perceived failure of stratospheric
aerosol injection deployment.” Zenodo.
https://zenodo.org/record/7072436#.YyOKwHbMLIU. Deposited 12 September 2022.

Google Scholar
VIEW FULL TEXTDOWNLOAD PDF
Further reading in this issue
RESEARCH ARTICLESEPTEMBER 26, 2022
Leptin treatment has vasculo-protective effects in lipodystrophic mice
Paulina Elena Stürzebecher, Susan Kralisch, [...]Jes-Niels Boeckel,
RESEARCH ARTICLESEPTEMBER 26, 2022
Landscape-scale concordance between local ecological knowledge for tropical
wild species and remote sensing of land cover
Yoshito Takasaki, Oliver T. Coomes, [...]Margaret Kalacska,
RESEARCH ARTICLESEPTEMBER 26, 2022
The contributions of positive outgroup and negative ingroup evaluation to
implicit bias favoring outgroups
Jimmy Calanchini, Kathleen Schmidt, [...]Samuel A. W. Klein,
Most Read
Most Cited
RESEARCH ARTICLEDECEMBER 30, 2013
Bodily maps of emotions
Emotions coordinate our behavior and physiological states during
survival-salient events and pleasurable interactions. Even though we are
often consciously aware of our current emotional state, such as anger or
happiness, the mechanisms giving ...Emotions are often felt in the body,
and somatosensory feedback has been proposed to trigger conscious emotional
experiences. Here we reveal maps of bodily sensations associated with
different emotions using a unique topographical self-report method. In ...
Lauri Nummenmaa, Enrico Glerean, [...]Jari K. Hietanen
RESEARCH ARTICLEAUGUST 14, 2017
Oxytocin-enforced norm compliance reduces xenophobic outgroup rejection
In the midst of rapid globalization, the peaceful coexistence of cultures
requires a deeper understanding of the forces that compel prosocial
behavior and thwart xenophobia. Yet, the conditions promoting such
outgroup-directed altruism have ...Never before have individuals had to
adapt to social environments defined by such magnitudes of ethnic diversity
and cultural differentiation. However, neurobiological evidence informing
about strategies to reduce xenophobic sentiment and foster ...
Nina Marsh, Dirk Scheele, [...]René Hurlemann
RESEARCH ARTICLEJUNE 11, 2018
Neural network retuning and neural predictors of learning success
associated with cello training
In sophisticated auditory–motor learning such as musical instrument
learning, little is understood about how brain plasticity develops over
time and how the related individual variability is reflected in the neural
architecture. In a ...The auditory and motor neural systems are closely
intertwined, enabling people to carry out tasks such as playing a musical
instrument whose mapping between action and sound is extremely
sophisticated. While the dorsal auditory stream has been shown to ...
Indiana Wollman, Virginia Penhune, [...]Robert J. Zatorre
Sign up for the
PNAS Highlights newsletter
Get in-depth science stories sent to your inbox twice a month.
n...@example.com
SUBSCRIBE
PNAS logo

BROWSE
CURRENT ISSUE
PNAS NEXUS
SPECIAL FEATURES
COLLOQUIA
LIST OF ISSUES
COLLECTED PAPERS
FRONT MATTER
JOURNAL CLUB
PODCASTS
INFORMATION
ABOUT
DIVERSITY AND INCLUSION
AUTHORS
REVIEWERS
SUBSCRIBERS
LIBRARIANS
PRESS
COZZARELLI PRIZE
PNAS UPDATES
Copyright © 2022 National Academy of Science. All rights reserved. | Online
ISSN 1091-6490

PNAS is a partner of CHORUS, CLOCKSS, COPE, CrossRef, ORCID, and
Research4Life.

Contact  Site Map  Terms & Privacy Policy  Accessibility
Reference #1

-- 
You received this message because you are subscribed to the Google Groups 
"geoengineering" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to geoengineering+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/geoengineering/CAJ3C-07JvXmV952cYtsPFN2NcO2ivE7G9uCke4FvWj7rnimaxw%40mail.gmail.com.