This is the accessible text file for GAO report number GAO-11-71 
entitled 'Technology Assessment: Climate engineering: Technical 
status, future directions, and potential responses' which was released 
on August 25, 2011. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office Center for Science, 
Technology, and Engineering: 

Report to Congressional Requester: 

July 2011: 

Technology Assessment: 

Climate engineering: Technical status, future directions, and 
potential responses: 

1. What is the current state of climate engineering technology and 
related sciences? 

2. What are potential future directions for research on climate 
engineering technologies, 2010 through 2039, and possible consequences? 

3. With respect to gauging potential responses: What is the level of 
awareness of climate engineering technologies, and what are attitudes 
toward researching and implementing them? 

GA0-11-71: 

[Cover image from GAO represents gaseous carbon dioxide (CO2) 
molecules in ambient air, currently measured at around 390 parts per 
million. Carbon dioxide consists of a central carbon atom doubly 
bonded with two oxygen atoms (0=C=0). Carbon dioxide is colorless and 
odorless at room temperature. Plants consume CO2 by photosynthesis, 
which converts CO2 into nutrients using energy from the Sun. Many 
scientists believe that the increased atmospheric CO2 concentration 
has increased the acidity of ocean waters and is the primary cause of 
increased global average surface temperature. Global management of CO2 
and related risks underlies current ideas about engineering the global 
climate system.] 

GAO Highlights: 

Highlights of GAO-11-71, a report to the Ranking Member, Committee on 
Science, Space, and Technology, House of Representatives. 

Why GAO did this study: 

Reports of rising global temperatures have raised questions about 
responses to climate change, including efforts to (1) reduce carbon 
dioxide (CO2) emissions, (2) adapt to climate change, and (3) design 
and develop climate engineering technologies for deliberate, large-
scale intervention in Earth’s climate. 

Reporting earlier that the nation lacks a coordinated climate-change 
strategy that includes climate engineering, GAO now assesses climate 
engineering technologies, focusing on their technical status, future 
directions for research on them, and potential responses. 

To perform this technology assessment, GAO reviewed the peer-reviewed 
scientific literature and government reports, consulted experts with a 
wide variety of backgrounds and viewpoints, and surveyed 1,006 adults 
across the United States. Experts convened with the assistance of the 
National Academy of Sciences advised GAO, and several reviewed a draft 
of this report. GAO incorporated their technical and other comments in 
the final report as appropriate. 

What GAO found: 

Climate engineering technologies do not now offer a viable response to 
global climate change. Experts advocating research to develop and 
evaluate the technologies believe that research on these technologies 
is urgently needed or would provide an insurance policy against worst 
case climate scenarios—but caution that the misuse of research could 
bring new risks. Government reports and the literature suggest that 
research progress will require not only technology studies but also 
efforts to improve climate models and data. 

The technologies being proposed have been categorized as carbon 
dioxide removal (CDR) and solar radiation management (SRM). CDR would 
reduce the atmospheric concentration of CO2, allowing more heat to 
escape and thus cooling the Earth. For example, proposed CDR 
technologies include enhancing the uptake of CO2 in oceans and forests 
and capturing CO2 from air chemically for storage underground. SRM 
technologies would place reflective material in space or in Earth’s 
atmosphere to scatter or reflect sunlight (for example, by injecting 
sulfate aerosols into the stratosphere to scatter incoming solar 
radiation or brightening clouds) or would increase the planet’s 
reflectivity (for example, by painting roofs and pavements in light 
colors). (See figure.) 

Figure: Examples of climate engineering technologies: 

[Refer to PSD for image: illustration] 

Depicted on the illustration: 

Cloud brightening at sea; 
Capturing CO2 from the air; 
Iron fertilization of the ocean; 
Pumping liquid CO2 into geological formations; 
Aerosols in stratosphere; 
Growing trees; 
Light-colored roofs and pavements. 

Source: GAO. 

[End of figure] 

GAO found these technologies currently immature, many with potentially 
negative consequences. Some studies say, for example, that 
stratospheric aerosols might greatly reduce summer
precipitation in places such as India and northern China. 

Many experts advocated research because of its potential benefits but 
also recognized its risks. For example, a country might unilaterally 
deploy a technology with a transboundary effect. Research advocates 
emphasized the need for risk management, envisioning a federal 
research effort that would (1) focus internationally on transparency 
and cooperation, given transboundary effects; (2) enable the public 
and national leaders to consider issues before they become crises; and 
(3) anticipate opportunities and risks. A small number of those we 
consulted opposed research; they anticipated major technology risks or 
limited future climate change. 

Based on GAO’s survey, a majority of U.S. adults are not familiar with 
climate engineering. When given information on the technologies, they 
tend to be open to research but concerned about safety. 

View [hyperlink, http://www.gao.gov/products/GAO-11-71] or key 
components at www.gao.gov. For more information, contact Timothy 
Persons at (202) 512-6412 or personst@gao.gov. 

Report multimedia: 

Depiction of the global carbon cycle changes over time: 
[hyperlink, http://www.gao.gov/multimedia/interactive/GAO-11-71a] 

Global average energy budget of the Earth’s atmosphere: 
[hyperlink, http://www.gao.gov/multimedia/interactive/GAO-11-71b] 

[End of section] 

GAO: 
Accountability--Integrity--Reliability: 

July 28, 2011: 

The Honorable Eddie Bernice Johnson: 
Ranking Member: 
Committee on Science, Space, and Technology: 
House of Representatives: 

Dear Ms. Johnson: 

In response to committee reports accompanying the legislative branch 
fiscal year 2008 appropriations bill, the U.S. Government Accountability 
Office established a permanent operational technology assessment group 
within GAO's Applied Research and Methods team: the Center for 
Science, Technology, and Engineering. Responding to your request that 
we conduct a technology assessment on proposed technological 
approaches toward engineering the climate, we examined the current 
state of climate engineering science and technology, experts' views of 
the future of U.S. climate engineering research, and potential public 
responses to climate engineering. We also discuss in this report key 
considerations for the use of climate engineering technologies and 
their policy implications. 

As agreed with your office, we plan no distribution of this report 
until 14 days after its issue date unless you publicly announce it 
contents earlier. We will then send copies of this report to 
interested congressional committees; the Secretaries of Agriculture, 
Commerce, Defense, Energy, and State; the Administrator of the 
National Aeronautics and Space Administration; the Administrator of 
the Environmental Protection Agency; and the Director of the National 
Science Foundation. We will provide copies to others on request. In 
addition, the report will be available at no charge on the GAO website 
at [hyperlink, http://www.gao.gov]. 

If you have any questions concerning this report, you may contact me 
at (202) 512-6412 or personst@gao.gov. Contact points for our Offices 
of Congressional Relations and Public Affairs may be found on the last 
page of this report. Major contributors to this report are listed on 
page 117. 

Sincerely yours, 

Signed by: 

Timothy M. Persons, Ph.D. 
Chief Scientist: 

[End of section] 

Summary: 

Reports of rising global average surface temperature have raised 
interest in the potential for engineering Earth's climate, 
supplementary to ongoing efforts to reduce greenhouse gas emissions 
and prepare for climate change through adaptation. Proposed climate 
engineering technologies, or direct, deliberate, large-scale 
interventions in Earth's climate, generally aim at either carbon 
dioxide removal (CDR) or solar radiation management (SRM). Whereas CDR 
would reduce the atmospheric concentration of carbon dioxide (CO), 
thus reducing greenhouse warming, SRM would either deflect sunlight 
before it reaches Earth or otherwise cool Earth by increasing the 
reflectivity of its surface or atmosphere. 

In conducting this technology assessment, we focused primarily on the 
technical status of climate engineering and the views of a wide range 
of experts on the future of research.[Footnote 1] Our findings 
indicate that: 

* climate engineering technologies are not now an option for 
addressing global climate change, given our assessment of their 
maturity, potential effectiveness, cost factors, and potential 
consequences. Experts told us that gaps in collecting and modeling 
climate data, identified in government and scientific reports, are 
likely to limit progress in future climate engineering research. 

* the majority of the experts we consulted supported starting 
significant climate engineering research now. Advocates and opponents 
of research described concerns about its risks and the possible misuse 
of its results. Research advocates supported balancing such concerns 
against the potential for reducing risks from climate change. They 
further envisioned a future federal research effort that would 
emphasize risk management, have an international focus, engage the 
public and national leaders, and anticipate new trends and 
developments. 

* a survey of the public suggests that the public is open to climate 
engineering research but is concerned about its possible harm and 
supports reducing CO2 emissions. 

Technical status: 

To assess the current state of climate engineering technology, we 
rated each technology for its maturity on a scale of 1 to 9, using 
technology readiness levels (TRL)—a standard tool for assessing the 
readiness of emerging technologies before full-fledged production or 
incorporation into an existing technology or system. We found that 
climate engineering technologies are currently immature, based on the 
TRL scores we calculated, and may face challenges with respect to 
potential effectiveness, cost factors, and potential consequences. (We 
characterized a technology with a TRL score lower than 6 as immature.) 

CDR technologies are designed to do one of the following: (1) 
chemically scrub CO2 from the atmosphere by direct air capture, 
followed by geologic sequestration of the removed CO2; (2) use biochar 
and biomass approaches to capture and sequester CO2; (3) manage land 
use to enhance the natural uptake and storage of CO2; (4) accelerate 
CO2 transfer from the atmosphere to the deep ocean for sequestration. 
We scored all but one CDR technology at a maturity of TRL 2. This 
means that we found that scientific or government publications have 
reported: 

* observation of the technology's basic scientific principles through 
theoretical research or mathematical models and; 

* conceptualization of an application of the technology in the context 
of addressing global climate change—but not an analytic and 
experimental proof of concept. 

The highest-scoring CDR technology (at TRL 3) was direct air capture 
of CO2, which has had laboratory demonstrations using a prototype and 
field demonstrations of underground sequestration of CO2. However, 
direct air capture is believed to be decades away from large-scale 
commercialization. Additionally, for each of the currently proposed 
CDR technologies, we found that implementation on a scale that could 
affect global climate change may be impractical, either because vast 
areas of land would be required or because of inefficient processes, 
high cost, or unrealistically challenging logistics. 

SRM technologies would inject aerosols into Earth's stratosphere to 
scatter a fraction of incoming sunlight, artificially brighten clouds, 
place solar radiation scatterers or reflectors in space, or increase 
the reflectivity of Earth's surface. All SRM technologies' maturity 
measured TRL 2 or less. That is, none had an analytical and 
experimental proof of concept. Additionally, we found that the SRM 
technologies that we rated "potentially fully effective" have not, 
thus far, been shown to be without possibly serious consequences. 
Further, each SRM technology must be maintained to sustain its effects 
on Earth's temperature; discontinuing the technology for any reason 
would result in Earth's temperature rising to a level dictated by 
other changes, such as an increased concentration of CO2 in Earth's 
atmosphere. 

A key challenge in climate engineering research is safely evaluating 
the technologies' potential risks in advance of large-scale field 
tests or deployment. Climate modeling would be a helpful evaluative 
tool, but a number of both federal agency and scientific reports have 
identified limitations in climate models and their underlying bases. 
Expanded scientific knowledge, enhanced precision and accuracy of 
tools for measuring key climate variables, and the development of 
dedicated high-performance computing would help fill the gaps and make 
future research more effective. 

Future directions: 

To determine how experts view the future of climate engineering 
research, we consulted 45 experts with a wide range of backgrounds and 
professional affiliations. We used future scenarios developed by one 
set of experts as a foresight tool to help elicit other experts' 
views. We found that the majority of those we consulted advocated 
starting significant research now or soon and believed that such 
research would have the potential to help reduce future risks from 
climate change. However, some conditioned their advocacy on the 
continuation of efforts to reduce emissions. Additionally, some 
pointed to new risks that the research or technologies developed from 
it might introduce. 

Many of the experts we consulted advocated research now because of 
their anticipation that substantial progress toward effective 
technologies might require two or more decades. Others said that 
climate engineering research is needed, even if future climate trends 
(such as the pace of change) are currently uncertain, because such 
research represents "an insurance policy against the worst case 
[climate change] scenarios." Many of those who called for research now 
saw the situation as urgent, reflecting foresight literature that 
warns against falling behind a potentially damaging trend—with 
possibly irreversible and very costly consequences. Their view was 
that climate engineering research now would constitute timely 
preparation for action and thus may help minimize the possibility of 
negative outcomes. 

A small number of those we consulted opposed future research on 
climate engineering. Research opponents reasoned that future climate 
change will not be great enough to warrant climate engineering or that 
alternatives such as pursuing emissions reductions (without climate 
engineering) would be preferable. However, the reason for opposing 
climate engineering research that was most strongly expressed 
concerned the risks associated with the research itself or the 
technologies' deployment. 

Both research advocates and opponents cautioned that climate 
engineering research carries risks either in conducting certain kinds 
of research or in using the results (for example, deploying 
potentially risky technologies that were developed on the basis of the 
research). Some also noted that other nations are conducting research 
and warned that, in the future, a single nation might unilaterally 
deploy a technology with transboundary effects. The research advocates 
suggested managing risks from climate engineering by, for example, 
conducting interdisciplinary risk assessments, developing norms and 
best practice guidelines for open and safe research, evaluating 
deployment risks in advance—and, potentially, as we discuss below, 
conducting joint research with other countries. Some advocates also 
indicated that rigorous research could help reduce risks from the 
uninformed use of risky technologies (as, for example, might occur in 
a perceived emergency) or emphasized the need to weigh potential risks 
from climate engineering against risks from climate change. 

Research advocates envisioned federal research that would foster 
developing and evaluating technologies like CDR and SRM and emphasize 
risk management. The majority of research advocates supported research 
that would include: 

* an international focus, sponsoring, for example, joint research with 
other nations (to foster cooperation and shared norms) and the study 
of how one nation's deployment might affect others, including those 
that might respond negatively or be especially vulnerable; 

* engagement with the public and U.S. decision-makers that might entail 
conducting studies to address concerns and support decisions (for 
example, studies of economic, ethical, legal, and social issues and 
studies of systemic risks); and; 

* foresight activities to help anticipate emerging research 
developments, key trends, and their implications for climate 
engineering research—notably, the new or emerging opportunities and 
risks that such changes might bring. 

Such features are broadly relevant to risk management in that they 
might (1) reduce risks of international tensions or even conflict 
resulting from climate engineering, (2) help prepare the nation in 
advance of possible crises, and (3) anticipate new risks that might be 
associated with future technologies. 

The United States does not now have a coordinated federal approach to 
climate engineering research, and we earlier recommended that such an 
approach be developed in the context of a federal strategy to address 
climate change (GAO 2010a). Other approaches to addressing climate 
change include efforts to (1) reduce CO2 emissions and (2) adapt to 
climate change. 

Potential responses: 

To understand public opinion, we analyzed survey data from 1,006 
adults 18 years old and older selected to represent the U.S. population. 
We provided them with basic materials on climate engineering—-that is, 
information similar in amount and type to what they might receive in 
the news media. The materials included a definition and examples of 
climate engineering technologies. Our survey revealed that a majority 
of the U.S. population is not familiar with climate engineering but 
may be open to research. 

Once provided with explanatory material, about 50-70 percent of the 
respondents across a range of demographic groups would be open to 
research on climate engineering and about 45 percent would be somewhat 
to extremely optimistic about its benefits. Such optimism would be 
tempered by caution, as we estimate that about 50-75 percent of the 
U.S. adult public would be concerned about the technologies' safety. 
Our survey results also indicate that support for reducing CO2 
emissions is more widespread than support for climate engineering. 
About 65-75 percent of the public would support the involvement of 
multiple organizations and interests in decision-making on these 
technologies. They included the scientific community, a coalition of 
national governments, individual national governments, the general 
public, private foundations, and not-for-profit organizations. 

[End of section] 

Contents: 

Highlights: 

Letter: 

Summary: 

1. Introduction: 

2. Background: 

3. The current state of climate engineering science and technology: 
3.1. Selected CDR technologies: 
3.1.1. Direct air capture of CO2 with geologic sequestration: 
3.1.2 Bioenergy with CO2 capture and sequestration: 
3.1.3 Biochar and biomass: 
3.1.4. Land-use management: 
3.1.5. Enhanced weathering: 
3.1.6. Ocean fertilization: 
3.2. Selected SRM technologies: 
3.2.1. Stratospheric aerosols: 
3.2.2. Cloud brightening: 
3.2.3. Scatterers or reflectors in space 36 
3.2.4 Reflective deserts, flora, and habitats 39 
3.3 Status of knowledge and tools for understanding climate engineering: 
3.3.1 Better models would help in evaluating climate engineering 
proposals: 
3.3.2 Key advancements in scientific knowledge could help improve 
climate models: 
3.3.3 Better observational networks could help resolve uncertainties 
in climate engineering science: 
3.3.4. High-performance computing resources could help advance climate 
engineering science: 

4. Experts' views of the future of climate engineering research: 
4.1. A majority of experts called for research now: 
4.2. Some experts opposed starting research: 
4.3. A majority of experts envisioned federal research with specific 
features: 
4.4. Some experts thought that uncertain trends might affect future 
research: 

5. Potential responses to climate engineering research: 
5.1. Unfamiliarity with geoengineering: 
5.2. Concern about harm and openness to research: 
5.3. Views on geoengineering in the context of climate and energy 
policy: 
5.4. Support for national and international cooperation on 
geoengineering: 

6. Conclusions: 

7. Experts' review of a draft of this report: 
7.1. Our framing of the topic: 
7.2. Our assessment of the technologies: 
7.3. Our assessment of knowledge and tools for understanding climate 
engineering: 
7.4. Our foresight and survey methodologies: 

8. Appendices: 
8.1. Objectives, scope, and methodology: 
8.2. Experts we consulted on climate engineering technologies: 
8.3. Foresight scenarios: 
8.4. The six external experts who participated in building the 
scenarios: 
8.5. Experts who commented in response to the scenarios: 
8.6. Experts who participated in our meeting on climate engineering: 
8.7. Reviewers of the report draft: 

9. References: 
GAO contacts and staff acknowledgments: 
Related GAO products: 
Other GAO technology assessments: 

Figures: 

Figure 1.1. The Keeling curve, 1960-2010: 

Figure 1.2. Earth's carbon cycle: 

Figure 2.1. Global average energy budget of Earth's atmosphere: 

Figure 3.1. Capturing and absorbing CO2 from air: 

Figure 3.2. CO2-absorbing synthetic tree: 

Figure 4.1. Taking early action to avoid potentially damaging trends: 

Illustration from foresight literature: 

Figure 5.1. U.S. public support for actions on climate and energy, 
August 2010: 

Figure 5.2. U.S. public views on who should decide geoengineering 
technology's use, August 2010: 

Figure 8.1. Four scenarios defining possible futures: 

Tables and boxes: 

Table 1.1. Selected climate engineering proposals, 1877-1992: 

Table 3.1. Selected CDR technologies: Their maturity and a summary of 
available information: 

Table 3.2. Selected SRM technologies: Their maturity and a summary of 
available information: 

Table 5.1. Geoengineering types and examples given to survey 
respondents: 

Table 8.1. Nine technology readiness levels described: 

Box 4.1. Climate engineering research: Risk mitigation strategies from 
the literature: 

Abbreviations: 

AOGCM: atmosphere-ocean general circulation model: 

BECS: bioenergy with CO2 capture and sequestration: 

CCS: carbon capture and sequestration: 

CDR: carbon dioxide removal: 

CLARREO: Climate Absolute Radiance and Refractivity Observatory: 

DOE: U.S. Department of Energy: 

FOR: enhanced oil recovery: 
ESM: Earth system model: 

GCM: general circulation model: 

GPU: graphics processing unit: 

IPCC: Intergovernmental Panel on Climate Change: 

NAS: National Academy of Sciences: 

NASA: National Aeronautics and Space Administration: 

NETL: National Energy Technology Laboratory: 

NIST: National Institute of Standards and Technology: 

NOAA: National Oceanic and Atmospheric Administration: 

NRC: National Research Council: 

NSF: National Science Foundation: 

PNNL: Pacific Northwest National Laboratory: 

R&D: research and development: 

SRM: solar radiation management: 

TRL: technology readiness level: 

USDA: U.S. Department of Agriculture: 

[End of section] 

1: Introduction: 

Every day, millions of tons of carbon-rich compounds called fossil 
fuels are extracted, refined or processed, and combusted to supply the 
world with energy, releasing as a byproduct millions of tons of carbon 
dioxide gas (CO2).[Footnote 2] From 1900 to 2007, annual global CO2 
emissions from fossil fuel consumption increased, on average, at a 
rate of about 2.6 percent per year (Boden, Marland, and Andres 2010). 
[Footnote 3] 

As emissions increased, the atmospheric concentration of CO2 rose. 
Figure 1.1 shows the rise in the concentration of CO2 between 1960 and 
2010 (Ralph Keeling 2011).[Footnote 4] 

C.D. Keeling (1960), noting that CO2 levels at observation stations 
were increasing over time, attributed this increase to fossil fuel 
combustion.[Footnote 5] Although CO2 is not the most abundant 
greenhouse gas, many scientists have concluded that CO2 emitted by 
human activities is the principal cause of the enhanced greenhouse 
effect (Lacis et al. 2010).[Footnote 6] 

Figure 1.1: The Keeling curve, 1960-2010: 

[Refer to PDF for image: line graph] 

Year: 1960; 
Approximate preindustrial concentration CO2 (ppm): 317. 

Year: 1961; 
Approximate preindustrial concentration CO2 (ppm): 318. 

Year: 1962; 
Approximate preindustrial concentration CO2 (ppm): 318. 

Year: 1963; 
Approximate preindustrial concentration CO2 (ppm): 319. 

Year: 1964; 
Approximate preindustrial concentration CO2 (ppm): 320. 

Year: 1965; 
Approximate preindustrial concentration CO2 (ppm): 320. 

Year: 1966; 
Approximate preindustrial concentration CO2 (ppm): 321. 

Year: 1967; 
Approximate preindustrial concentration CO2 (ppm): 322. 

Year: 1968; 
Approximate preindustrial concentration CO2 (ppm): 323. 

Year: 1969; 
Approximate preindustrial concentration CO2 (ppm): 325. 

Year: 1970; 
Approximate preindustrial concentration CO2 (ppm): 326. 

Year: 1971; 
Approximate preindustrial concentration CO2 (ppm): 326. 

Year: 1972; 
Approximate preindustrial concentration CO2 (ppm): 327. 

Year: 1973; 
Approximate preindustrial concentration CO2 (ppm): 330. 

Year: 1974; 
Approximate preindustrial concentration CO2 (ppm): 330. 

Year: 1975; 
Approximate preindustrial concentration CO2 (ppm): 331. 

Year: 1976; 
Approximate preindustrial concentration CO2 (ppm): 332. 

Year: 1977; 
Approximate preindustrial concentration CO2 (ppm): 334. 

Year: 1978; 
Approximate preindustrial concentration CO2 (ppm): 336. 

Year: 1979; 
Approximate preindustrial concentration CO2 (ppm): 337. 

Year: 1980; 
Approximate preindustrial concentration CO2 (ppm): 339. 

Year: 1981; 
Approximate preindustrial concentration CO2 (ppm): 340. 

Year: 1982; 
Approximate preindustrial concentration CO2 (ppm): 341. 

Year: 1983; 
Approximate preindustrial concentration CO2 (ppm): 343. 

Year: 1984; 
Approximate preindustrial concentration CO2 (ppm): 344. 

Year: 1985; 
Approximate preindustrial concentration CO2 (ppm): 346. 

Year: 1986; 
Approximate preindustrial concentration CO2 (ppm): 347. 

Year: 1987; 
Approximate preindustrial concentration CO2 (ppm): 349. 

Year: 1988; 
Approximate preindustrial concentration CO2 (ppm): 351. 

Year: 1989; 
Approximate preindustrial concentration CO2 (ppm): 353. 

Year: 1990; 
Approximate preindustrial concentration CO2 (ppm): 354. 

Year: 1991; 
Approximate preindustrial concentration CO2 (ppm): 356. 

Year: 1992; 
Approximate preindustrial concentration CO2 (ppm): 356. 

Year: 1993; 
Approximate preindustrial concentration CO2 (ppm): 357. 

Year: 1994; 
Approximate preindustrial concentration CO2 (ppm): 359. 

Year: 1995; 
Approximate preindustrial concentration CO2 (ppm): 361. 

Year: 1996; 
Approximate preindustrial concentration CO2 (ppm): 363. 

Year: 1997; 
Approximate preindustrial concentration CO2 (ppm): 364. 

Year: 1998; 
Approximate preindustrial concentration CO2 (ppm): 367. 

Year: 1999; 
Approximate preindustrial concentration CO2 (ppm): 368. 

Year: 2000; 
Approximate preindustrial concentration CO2 (ppm): 369. 

Year: 2001; 
Approximate preindustrial concentration CO2 (ppm): 371. 

Year: 2002; 
Approximate preindustrial concentration CO2 (ppm): 373. 

Year: 2003; 
Approximate preindustrial concentration CO2 (ppm): 376. 

Year: 2004; 
Approximate preindustrial concentration CO2 (ppm): 377. 

Year: 2005; 
Approximate preindustrial concentration CO2 (ppm): 380. 

Year: 2006; 
Approximate preindustrial concentration CO2 (ppm): 382. 

Year: 2007; 
Approximate preindustrial concentration CO2 (ppm): 384. 

Year: 2008; 
Approximate preindustrial concentration CO2 (ppm): 385. 

Year: 2009; 
Approximate preindustrial concentration CO2 (ppm): 387. 

Year: 2010; 
Approximate preindustrial concentration CO2 (ppm): 390. 

The orange line indicates the annual average atmospheric concentration 
of CO2 derived from monthly in situ air measurements at Mauna Loa 
Observatory, Hawaii. The cyclical pattern of the monthly measurements 
shown in light gray indicates seasonal fluctuations. The approximate 
preindustrial concentration of 280 parts per million (ppm) indicates 
the estimated atmospheric abundance of CO2 around the year 1750. In 
1960, the atmospheric concentration of CO2 was about 317 ppm; by 2010, 
it had risen to about 390 ppm. 

Source: GAO, adapted from Ralph Keeling (2011). 

[End of figure] 

Over the past century, global mean surface temperature increased by 
about 0.75 degrees Celsius, and many scientists expect the rise to 
continue in coming decades (NRC 2010a; Solomon et al. 2007), as we 
describe in the background section of this report.[Footnote 7] A few 
scientists have argued that a doubling of the atmospheric CO2 
concentration, by itself, would increase the global average 
temperature by only about 1 degree Celsius and that the models 
predicting rising temperatures in the coming decades are incomplete 
and are therefore considerably uncertain (Lindzen 2010; Lindzen and 
Choi 2009).[Footnote 8] Nevertheless, there is a consensus of many 
authoritative scientific bodies, which have conveyed a sense of 
urgency on the climate change issue; hence the following discussion on 
climate engineering, or direct, deliberate large-scale interventions in
Earth’s climate.[Footnote 9] 

The future effects of warming are uncertain. The National Research 
Council (NRC) recently examined potential consequences of rising 
temperatures over the next century, such as changes in vegetation, 
precipitation, and the rate of sea level rise (NRC 2010a). NRC’s 
report suggests an overall potential for negative effects on people, 
infrastructures, and ecosystems. For example, the projected rise in 
sea level could threaten several large ports and urban centers in the 
United States, such as Miami, New York, and Norfolk, as well as low-
lying island groups, such as the Maldives.[Footnote 10] Some 
researchers have suggested that climate change could have even more 
extreme adverse consequences.[Footnote 11] Others have proposed that 
rising temperatures might benefit certain geographic areas or economic 
sectors; for example, agricultural productivity might increase in some 
areas, although researchers caution that how climate change affects 
agriculture is complex and uncertain (Gornall et al. 2010). 
Additionally, while global surface temperature is increasing on 
average, it is not increasing uniformly (Solomon et al. 2007). For 
example, scientists have observed that temperatures have risen more in 
areas that are relatively colder, and the observed change in 
temperature is greater in winter than in summer and greater at night 
than in the day (Solomon et al. 2007). Disproportionate warming of 
cold temperatures could have important implications for human health 
and mortality, if exposure to heat is less dangerous than exposure to 
cold (NRC 2010a). 

Two broad strategies to meet the challenges of climate change through 
public policy are mitigation and adaptation. Mitigation aims to limit 
climate change, usually by decreasing greenhouse gas emissions (GAO 
2008a). For example, mitigation might replace high-carbon fuels, such 
as coal, with fuels that emit less CO2 per unit of energy, such as 
natural gas. Mitigation might also enhance the capacity of sinks, 
which reabsorb CO2 from the atmosphere and store it on Earth (GAO 
2008a). For example, incremental changes in land use could increase 
the amount of carbon stored as cellulosic fiber in forests and other 
vegetation that removes CO2 from the atmosphere by photosynthesis. 
Adaptation aims to adjust Earth’s systems, infrastructures, or social 
programs in response to actual or expected changes in the climate.
For example, adaptation can make systems more robust in the face of 
climatic extremes, exploit new opportunities, or cope with adversity
(GAO 2009a). 

Success in mitigating climate change or adapting to it can depend on 
technological progress. For example, the cost of mitigation is likely 
to be lower if alternatives to fossil fuels are less expensive (Popp 
2006). Adaptation can also be affected by the technology, as happens 
in predicting the weather, controlling indoor temperatures with 
heating and air conditioning, or managing a sea level rise, as in 
building harbor gates in Venice, Italy (Spencer et al. 2005). However, 
neither mitigation nor adaptation has progressed sufficiently to 
moderate current climate projections or diminish the seriousness of 
their effects. For example, the relative expense of low-carbon energy 
technology presently tends to limit its use. And requirements to 
reduce emissions can be difficult to enforce, as the Kyoto Protocol 
demonstrates, or can fail to encourage advances in low-carbon energy 
technology (Barrett 2008; Barrett 1998). 

Even if deep emissions cuts were to stabilize the atmospheric 
concentration of CO2 at the current level, scientific models predict 
that average global surface temperature is likely to rise 0.3 to 0.9 
degrees Celsius by 2100 (Backlund et al. 2008). Some scientists 
suggest that climatic perturbation from anthropogenic CO2 emissions is 
nearing a tipping point beyond which it will be difficult or 
impossible to remediate changes in Earth’s climate. Figure 1.2 
illustrates Earth’s carbon cycle, which regulates the flow of carbon 
between the atmosphere and land-based and oceanic sinks. 

These and other possible challenges to the success of mitigation and 
adaptation have helped stimulate public policy interest in climate 
engineering, which would develop and use technology to moderate Earth’
s climate by controlling the radiation balance and, thus, average 
global temperature. The United Kingdom’s Royal Society (the oldest 
scientific academy in continuous existence) has identified other 
distinguishing characteristics of this strategy as well, highlighting 
the “deliberate, large-scale intervention in the Earth’s climate system”
in its definition of geoengineering (Royal Society 2009, ix).[Footnote 12] 

Figure 1.2 Earth's carbon cycle: 

[Refer to PDF for image: illustration] 

Depicted on the illustration: 

Oceanic: 38,700 (black) plus 137 (red); 
Atmospheric: 590 (black) plus 187 (red) (.8/yr. net to terrestrial) 
(1.3/yr. net to oceanic); 
Terrestrial: 3,800 (black) minus 20 (red) (.5/yr. net to oceanic); 
Geological: 7,180 (black); 
Industrialization: 6.4/yr. 

Source: GAO, adapted from Sarmiento and Gruber (2002), updated using 
Field, Sarmiento, and Hales (2007). 

Note: All numeric values are in gigatons (GtC), or billions of metric 
tons, of carbon. In Earth's carbon cycle, preindustrial reservoir 
sizes are represented by black numbers. Cumulative postindustrial 
reservoir transfers are represented by red numbers. Current fluxes 
between reservoirs are shown in smaller type; the largest flux is 6.4 
GtC per year from industrialization. This ongoing carbon imbalance is 
causing ocean water to become more acidic and is believed to be the 
primary cause of increased global average surface temperature. (An 
animated depiction of changes in the global carbon cycle over time 
may be accessed at [hyperlink, 
http://www.gao.gov/multimedia/interaction/GAO-11-71a]) 

[End of figure] 

In its 2009 report, the Royal Society described two major approaches 
to climate engineering: accelerating the movement of carbon from the 
atmosphere to terrestrial and oceanic carbon sinks, or carbon dioxide 
removal (CDR), and controlling net incoming radiation from the Sun, or 
solar radiation management (SRM). As CDR reduces the atmospheric 
concentration of CO2, the enhanced greenhouse effect is weakened, and 
thermal radiation more easily escapes into space.[Footnote 13] SRM, in 
contrast, attempts to reduce net incoming solar radiation by 
deflecting sunlight or by increasing the reflectivity of the 
atmosphere, clouds, or Earth’s surface.[Footnote 14] 

The concept of engineering the climate is not new (Fleming 2010). 
Table 1.1 shows examples of climate engineering proposals dating from 
1877. Today, policymakers and scientists are examining climate 
engineering as a way to manage potential catastrophic risks from
climate change. 

We designed this report to complement our September 2010 report on 
geoengineering (GAO 2010a). In this context, we conducted this 
technology assessment of climate engineering.[Footnote 15] Our 
objectives for this report were to examine (1) the current state of 
climate engineering science and technology, (2) expert views of the 
future of U.S. climate engineering research, and (3) public 
perceptions of climate engineering (we describe our methodology in 
section 8.1). 

To determine the current state of the science and technology of 
climate engineering, we reviewed a broad range of scientific and 
engineering literature, including proceedings from conferences such as 
the 2010 Asilomar International Conference on Climate Intervention 
Technologies (Asilomar Scientific Organizing Committee 2010). We 
revisited GAO-10-903, a complementary report on climate engineering we 
issued in September 2010 (GAO 2010a). We reviewed relevant 
congressional testimony. We interviewed a broad range of experts and 
officials working on climate engineering and proponents of specific 
climate engineering technologies. This report is an assessment of 
technologies to engineer the climate and the quality of information 
available to assess these technologies. We did not independently 
assess whether climate change is occurring or what is causing any 
climate change if it is occurring or whether current scientific 
knowledge supports the occurrence of climate change or its causes. We 
did not assess whether climate changes are or will be sufficient to 
warrant using these technologies. 

To ensure a balance of views and information, we analyzed and 
synthesized information from an array of experts with diverse views on 
our subject. We used the Royal Society’s classification of climate 
engineering approaches to focus our analysis on CDR and SRM technologies
(Royal Society 2009, 1). From the information we found in the 
literature and our interviews with experts, we assessed climate 
engineering technologies along four key dimensions: (1) maturity, (2) 
potential effectiveness, (3) cost factors, and (4) potential 
consequences. We did not independently assess the accuracy of the cost 
estimates, but we report estimates we found in the literature. 

To assess how experts view the future of climate engineering research, 
we (1) conducted a foresight exercise in which experts developed 
alternative future scenarios; (2) elicited comments, stimulated by the 
scenarios, from a broad array of experts; and (3) asked other experts 
to respond to the preliminary synthesis we developed from the 
scenarios and earlier comments. 

Table 1.1 Selected climate engineering proposals, 1877-1992: 

Date: 1877; 
Who: Nathaniel Shaler, American scientist; 
Proposal: Suggested rerouting the Pacific's warm Kuroshio Current 
through the Bering Strait to raise Arctic temperatures as much as 30 
degrees Fahrenheit. 

Date: 1912; 
Who: Carroll Livingston Riker, American engineer, and William M. 
Calder, U.S. Senator; 
Proposal: Proposed building a 200-mile jetty into the Atlantic Ocean 
to divert the warm Gulf Stream over the colder Labrador current to 
change the climate of North America's Atlantic Coast; Calder 
introduced a bill to study its feasibility. 

Date: 1929; 
Who: Hermann Oberth, German-Hungarian physicist and engineer; 
Proposal: Proposed building giant mirrors on a space station to focus 
the Sun's radiation on Earth's surface, making the far North habitable 
and freeing sea lanes to Siberian harbors. 

Date: 1945; 
Who: Julian Huxley, biologist and Secretary-General of UNESCO 1946-48; 
Proposal: Proposed exploding atomic bombs at an appropriate height 
above the polar regions to raise the temperature of the Arctic Ocean 
and warm the entire climate of the northern temperate zones 

Date: c. 1958; 
Who: Arkady Markin, Soviet engineer; 
Proposal: Proposed that the United States and Soviet Union build a 
gigantic dam across the Bering Strait and use nuclear power–driven 
propeller pumps to push the warm Pacific current into the Atlantic by 
way of the Arctic Sea. Arctic ice would melt, and the Siberian and 
North American frozen areas would become temperate and productive. 

Date: 1958; 
Who: M. Gorodsky, Soviet engineer and mathematician, and Valentin 
Cherenkov, Soviet meteorologist; 
Proposal: Proposed placing a ring of metallic potassium particles into 
Earth’s polar orbit to diffuse light reaching Earth and increase solar 
radiation to thaw the permanently frozen soil of Russia, Canada, and 
Alaska and melt polar ice. 

Date: 1965; 
Who: President’s Science Advisory Committee, United States; 
Proposal: Investigated injecting condensation or freezing nuclei into 
the atmosphere to counteract the effects of increasing carbon dioxide. 

Date: 1977; 
Who: Cesare Marchetti, Italian industrial physicist; 
Proposal: Coined the term “geoengineering” and proposed sequestering 
CO2 in the deep ocean. 

Date: 1983; 
Who: Stanford Penner, A. M. Schneider, and E. M. Kennedy, American 
physicists; 
Proposal: Suggested introducing small particles into the atmosphere to 
reflect more sunlight back into space. 

Date: 1988; 
Who: John H. Martin, American oceanographer; 
Proposal: Proposed dispersing a relatively small amount of iron into 
appropriate areas of the ocean to create large algae blooms that could 
take in enough atmospheric carbon to reverse the greenhouse effect and 
cool Earth. 

Date: 1989; 
Who: James T. Early, American climatologist; 
Proposal: Suggested deflecting sunlight by 2 percent with a $1 
trillion to $10 trillion “space shade” placed in Earth orbit. 

Date: 1990; 
Who: John Latham, British cloud physicist; 
Proposal: Proposed seeding marine stratocumulus clouds with seawater 
droplets to increase their reflectivity and longevity. 

Date: 1992; 
Who: NAS Committee on Science, Engineering, and Public Policy; 
Proposal: Proposed adding more dust to naturally occurring 
stratospheric dust to increase the net reflection of sunlight. 

Source: GAO. 

Note: Table 1.1 is based in part on an outline provided by James R. 
Fleming. We selected proposals beginning in the 19th century to 
illustrate a variety of climate engineering technologies and points in 
Earth’s climate system where interventions could occur. The table 
excludes numerous proposals to generate rain or alter hurricanes, 
which are not intended to cause long-term change. 

[End of table] 

We also conducted focus groups and a web-based survey of the U.S. 
adult population. We surveyed a representative sample of U.S. 
residents 18 years old and older from July 19 to August 5, 2010, 
receiving usable responses from 1,006 respondents. We used the term 
“geoengineering” in the information we gave the focus group and 
survey participants, given that we and others, such as the Royal 
Society, had used this term earlier. 

We convened a meeting of scientists, engineers, and other experts, 
with the assistance of the National Academy of Sciences (NAS), that we 
called the Meeting on Climate Engineering. 

We helped NAS select a diverse and balanced group of participants with 
expertise in climate engineering, climate science, measurement 
science, foresight studies, emerging technologies, research 
strategies, and the international, public opinion, and public 
engagement dimensions of climate engineering. We provided them with 
the preliminary results of our work, and the meeting served as a forum 
in which the participants expressed general reactions to and gave 
advice and suggestions on our preliminary findings. Their comments led 
us to review additional published and unpublished literature. 

Following the meeting, we contacted the participants in person or by 
telephone or e-mail to clarify and expand what we had heard. We used 
what we learned from the meeting participants to update and clarify 
our exposition of the current state of climate engineering technology, 
expert views of the future of U.S. climate engineering research, and 
public perceptions of climate engineering. We then sent a complete 
draft of our report to the participants in the Meeting on Climate 
Engineering who had agreed to review it. 

We conducted our work for this technology assessment from January 2010 
through July 2011 in accordance with GAO’s quality standards as they 
pertain to technology assessments. Those standards require that we 
plan and perform the technology assessment to obtain sufficient and 
appropriate evidence to provide a reasonable basis for our findings 
and conclusions, based on our technology assessment objectives. We 
believe that the evidence we obtained provides a reasonable basis for 
our findings and conclusions, based on our technology assessment 
objectives. 

[End of section] 

2. Background: 

Global temperature increases such as those measured on Earth have been 
attributed to a gradual change in the balance of energy flowing into 
and away from Earth’s surface. Earth’s system maintains a constant 
average temperature only if the same amount of energy leaves the 
system as enters it. If more energy enters than leaves, the difference 
manifests as a temperature increase. Figure 2.1 shows current 
estimates of the equilibrium transfer of energy. 

Figure 2.1 Global average energy budget of Earth's atmosphere: 

[Refer to PDF for image: illustration] 

Incoming sunlight: 342 W/m2: 
Reflected from atmosphere and clouds: 77 W/m2; 
Reflected from Earth's surface: 30 W/m2; 
Absorbed by atmosphere and clouds: 67 W/m2; 
Absorbed by Earth's surface: 168 W/m2. 

Atmosphere and clouds: 
Emitted from atmosphere and clouds: 195 W/m2; 
Emitted directly to space: 40 W/m2; 
Absorbed by atmosphere and clouds: 350 W/m2; 

Transferred from Earth's surface by evaporation and convection: 102 
W/m2; 
Emitted from atmosphere and clouds: 324 W/m2; 
Emitted from Earth's surface: 390 W/m2. 

Source: GAO, adapted from Kiehl and Trenberth 1997. 

Note: All numeric values are in watts per square meter (VV/m2). 
Incoming sunlight is both reflected from and absorbed by the 
atmosphere, clouds, and Earth's surface. Some of the energy absorbed 
by Earth's surface is transferred to the atmosphere by evaporation and 
convection, and the remainder is emitted as heat energy. The majority 
of the heat energy is absorbed by the atmosphere and clouds, with some 
escaping directly to space. Energy absorbed by the atmosphere and 
clouds is reradiated as heat energy back to Earth's surface as well as 
directly to space. Based on the composition of the atmosphere and 
clouds, the heat energy they absorb can accumulate by the greenhouse 
effect in which energy emitted from Earth's surface is trapped by 
gases in the atmosphere and clouds. For this reason, greenhouse gases 
in Earth's atmosphere can affect global average surface temperature. 
(An animated depiction of the global average energy budget of Earth's 
atmosphere may be viewed at [hyperlink, 
www.gao.gov/multimedia/interactive/GAO-11-71b].) 

[End of figure] 

Solar radiation is the predominant source of energy entering Earth’s 
system. It has an average global power of approximately 342 watts per 
square meter (W/m2). The system, including Earth’s surface and the 
atmosphere, absorbs about 69 percent of incoming solar radiation and 
reflects the remaining 31 percent back into space. That is, Earth’s 
surface absorbs about 49 percent of incoming radiation, and the 
atmosphere absorbs about 20 percent. Earth’s atmosphere and clouds 
reflect approximately 23 percent into space, while Earth’s surface 
(land, vegetation, water, and ice) reflects approximately 9 percent. 
Energy absorbed by the atmosphere affects the planet’s climate system 
through subsequent energy transfers (Solomon et al. 2007). 

The energy Earth’s surface and atmosphere absorb warms the planet. An 
inflow of energy to Earth without an equivalent outflow would result 
in continually increasing temperatures. However, Earth reemits energy 
from the surface to the atmosphere in the form of thermal radiation 
(long wavelength or infrared radiation) (Solomon et al. 2007). 

Approximately 10 percent of the thermal radiation reemitted by Earth 
passes through the atmosphere into space, and 90 percent is absorbed 
in the atmosphere, primarily in greenhouse gases, which efficiently 
absorb long-wave radiation. The atmospheric concentration of 
greenhouse gases is very low. Water vapor (H2O) is the most important 
greenhouse gas and is highly variable but typically makes up about
1 percent of the atmosphere (Solomon et al. 2010; Kiehl and Trenberth 
1997). Carbon dioxide is the second most important greenhouse gas; the 
current atmospheric concentration of CO2 is approximately 390 ppm (R. 
F. Keeling et al. 2009; Kiehl and Trenberth 1997; C. D. Keeling et al. 
2001).[Footnote 16] 

Just as the planet must maintain a balance of incoming and outgoing 
energy, the atmosphere and clouds must emit as much energy as they 
absorb to maintain a constant temperature. Therefore, the atmosphere 
and clouds emit long-wave radiation at approximately the same rate as 
they absorb energy from the Sun and Earth. This is manifested as 
additional thermal emissions both into space and toward Earth. The 
planet’s surface absorbs the Earth-bound thermal radiation, which 
raises Earth’s surface temperature, which increases thermal radiation 
from Earth’s surface, and so on, until this feedback achieves stable 
temperatures. 

The relationship between temperature and thermal radiation emitted 
from Earth is approximately described by the Stefan-Boltzmann law: 

F = o T4 

where F is the thermal radiation emitted from Earth’s surface in watts 
per square meter (W/m2), 0is the Stefan-Boltzmann constant, and T is 
the temperature of Earth’s surface in Kelvin (K).[Footnote 17] The 
Stefan-Boltzmann law provides evidence for atmospheric greenhouse gas 
feedback in Earth’s energy system. If Earth’s radiation, absorbed and 
reemitted, were only 235 W/m2 (342 W/m2 minus 107 W/m2 of reflected 
solar radiation), its average surface temperature would be about
254 K (–19 degrees Celsius). But Hansen and colleagues have estimated 
that Earth’s actual average surface air temperature between 1951 and 
1980 was approximately 287 K (14 degrees Celsius) (Hansen et al. 
2010). The difference in temperature is attributed to greenhouse gases 
that trap thermal radiation, warming Earth as depicted in figure 2.1. 
Thermal radiation emitted by Earth’s surface at 287 K is 385 W/m2, 
which compares favorably with the 390 W/m2 in the figure, 
corresponding to a temperature of 288 K. 

Climate scientists infer that accumulations of anthropogenic 
greenhouse gases are gradually adding to Earth’s natural greenhouse 
process. These accumulations absorb more thermal radiation emitted by 
Earth’s surface and reduce thermal radiation that escapes into space. 
The additional thermal radiation the greenhouse gases absorb is 
reradiated to space and back toward Earth. The planet’s surface 
absorbs the additional Earth-bound thermal radiation, which raises 
Earth’s surface temperature, which increases thermal radiation from 
Earth’s surface, and so on, until this feedback achieves a new, higher 
stable temperature. The magnitude and effect of this change in Earth’s 
global energy system are important subjects of climate science studies 
today (Solomon et al. 2007; NRC 2010a). 

[End of section] 

3: The current state of climate engineering science and technology: 

Most climate engineering proposals would aim to remediate the climate 
by affecting Earth’s energy balance, using either CDR to reduce the 
atmospheric concentration of CO2 or SRM to reduce incoming solar 
radiation. These two approaches differ significantly in their 
technical challenges and potential consequences (Royal Society 2009). 
The literature and our interviews with experts suggested four key 
dimensions on which we assessed these technologies, to the extent 
possible, given their current development: (1) maturity, (2) potential 
effectiveness, (3) cost factors, and (4) potential consequences (see 
section 8.1). Since developing many of the technologies we examined 
would require advances in new scientific data and analyses, we 
identified the climate’s representative physical, chemical, and 
biological algorithms; the geographic, temporal, and technical sensors 
of essential climate mechanisms; and next-generation, high-performance 
computing resources dedicated to climate science as areas that 
represent current shortfalls in knowledge and infrastructure. 

CDR technologies may be characterized as predominantly land-based or 
predominantly ocean-based (NRC 2010a; Royal Society 2009). Land-based 
technologies include direct air capture, bioenergy with CO2 capture 
and sequestration, biochar and other biomass-related methods, land-use 
management, and enhanced weathering. Direct air-capture systems 
attempt to capture CO2 from air directly and then store it in deep 
subsurface geologic formations. Bioenergy with CO2 capture and 
sequestration would also store CO2 underground, and biochar and other 
biomass-related methods would sequester carbon in soil or bury it. 
Land-use management practices we reviewed would enhance natural 
sequestration of CO2 in forests. Enhanced weathering would fix 
atmospheric CO2 in silicate rocks in a chemical reaction and then 
store it as either carbonate rock or dissolved bicarbonate in the 
ocean. Ocean-based technologies would fertilize the ocean to promote 
the growth of phytoplankton to sequester CO2. 

Seven SRM technologies have been reported in sufficient detail for us 
to assess them as candidates for climate engineering. Two would be 
deployed in the atmosphere—one scattering solar radiation back into 
space using stratospheric aerosols, the other reflecting solar 
radiation by brightening marine clouds. Two would be deployed in space—
one scattering or reflecting solar radiation from Earth orbit, the 
other scattering or reflecting solar radiation at a stable position 
between Earth and the Sun. The three remaining technologies would 
artificially reflect additional solar radiation from Earth’s surfaces—
covered deserts, more reflective flora, or more reflective settled 
areas. 

We found that since most climate engineering technologies are in early 
stages of development, none could be used to engineer the climate on a 
large scale at this time. We used technology readiness levels to rate 
the maturity of each technology on a scale from 1 to 9, with scores 
lower than TRL 6 indicating an immature technology. No CDR technology 
scored higher than TRL 3, and no SRM technology scored higher than TRL 
2.[Footnote 18] 

Considerable uncertainty surrounds the potential effectiveness of the 
technologies we reviewed, in part because they are immature. 
Additionally, for several proposed CDR technologies, the amount of CO2 
removed may be difficult to verify through modeling or direct 
measurements. 

The technologies’ cost factors we report represent, for CDR, resources 
used to remove CO2 from the atmosphere and store it. For SRM, they 
represent resources required to counteract global warming caused by 
doubling the preindustrial atmospheric concentration of CO2 or, for 
technologies that are potentially not fully effective, resources 
required to counteract global warming to the maximum extent possible. 
Some of the studies we reviewed indicate possible cost levels; we 
report these for illustration, but we did not evaluate them 
independently. Some studies described cost levels qualitatively (Royal 
Society 2009). 

Using many of the CDR and SRM technologies we reviewed would pose 
risks, some of which might not yet be known. Although minimal risks 
have been reported for air capture, some risks are related to the 
geologic sequestration of CO2. Land-use management approaches to 
capture and store CO2 are not generally regarded as risky. Enhanced 
weathering would pose environmental risks from the large-scale mining 
activities that would be needed to support it. The short-term and long-
term ecological, economic, and climatologic risks from ocean 
fertilization remain uncertain. Using SRM technologies could
affect temperatures but would not abate ocean acidification. Potential 
effects on precipitation are varied. Failing to sustain SRM 
technologies, once deployed, could result in a potentially rapid 
temperature rise. 

In sections 3.1 and 3.2, we present our assessment of the CDR and SRM 
technologies. In section 3.3, we describe the status of scientific 
knowledge and infrastructure related to climate engineering 
technologies. 

3.1: Selected CDR technologies: 

Table 3.1 summarizes our assessment of the maturity of six CDR 
technologies and presents information from published reports on their 
potential effectiveness, cost factors, and potential consequences. TRL 
ratings assess the maturity of each technology. Potential 
effectiveness is described in terms of an overall qualitative rating, 
where possible, and quantitative estimates of (1) the maximum capacity 
to reduce the global atmospheric concentration of CO2 (ppm) from its 
projected level of 500 ppm in 2100 and (2) the annual capacity to 
remove CO2 from Earth’s atmosphere (gigatons of CO2 or CO2-C 
equivalent per year), which we compared to annual anthropogenic emissions 
of 33 gigatons of CO2.[Footnote 19] Cost factors represent the resources 
used to remove CO2 from the atmosphere and store it. Potential 
consequences associated with each technology include reported negative 
consequences, risks, and cobenefits. 

Table 3.1 Selected CDR technologies: Their maturity and a summary of 
available information: 

Technology: Direct air capture of CO2 with geologic sequestration; 
Maturity[A]: Low (TRL 3):
* Basic principles understood and reported; 
* System concept formulated; 
* Experimental proof of concept demonstrated with a prototype unit in 
a laboratory environment; 
* Models of CO2 injection and transport developed and used for risk 
analysis and for simulating fate of injected CO2; 
* Basic technological components not demonstrated as working together; 
* No plans or prototypes for large-scale industrial implementation; 
* Geological sequestration of CO2 is more mature but not practiced on 
a scale to potentially affect climate. 
Potential effectiveness[B]: Not rated: 
* No “obvious limit” to the amount of CO2 reduction by year 2100; 
* Could theoretically counter all global anthropogenic CO2 emissions 
at 33 gigatons per year; 
* Large energy penalty: net increase in CO2 emissions if fossil fuel 
used (electricity from fossil fuels would release more CO2 than an air 
capture unit would remove); 
* Uncertainty around technical scalability. 
Cost factors[C]: 
* Viability may depend on nature and extent of a carbon market; 
* Process energy requirements for currently inefficient technologies 
for directly separating CO2 from air in very dilute concentration; 
* Transportation and logistics for sequestration of captured CO2; 
* Construction and management of geologic CO2 sequestration sites 
(e.g., CO2 injection, measuring, monitoring, and verification); 
* Greatly varied estimates in the scientific literature: $27 to $630 
or more per ton of CO2 removed (excluding transportation, 
sequestration, and other costs). 
Potential consequences[D]: 
* Aspects associated with handling process materials or chemicals; 
* May have sequestration risks such as potential for CO2 to escape 
from underground storage in the event of reservoir fracture or fissure 
from built-up pressure. 

Technology: Bioenergy with CO2 capture and sequestration; 
Maturity[A]: Low (TRL 2): 
* Basic principles understood and reported; 
* System concept formulated; 
* No experimental demonstration of proof of concept (no laboratory 
scale experiments that indicate CO2 reducing potential); 
* Emerging technology leverages what is known about CO2 capture and 
geologic sequestration; 
Potential effectiveness[B]: Low to medium: 
* Maximum ability to reduce atmospheric CO2: 50–150 ppm by 2100; 
* Net carbon negative under ideal conditions; 
* Depends on plant productivity and land area cultivated; 
Cost factors[C]: 
* Viability may depend on nature and extent of a carbon market; 
* Value of land in other uses; 
* Potentially large land area for growing and harvesting biomass; 
* Type of biomass feedstock (e.g., switchgrass); 
* Process energy requirements for bioenergy production (e.g., pyrolysis); 
* Construction and management of geologic CO2 sequestration sites 
(e.g., CO2 injection, measuring, monitoring, and verification); 
* Transportation and logistics for sequestering captured CO2; 
* Greatly varied estimates in the scientific literature: $150–$500 per 
ton of CO2 removed (excluding transportation and sequestration costs); 
Potential consequences[D]: 
* Potential land-use trade-offs; related impacts on food prices, water 
resources, fertilizer use; 
* CO2 sequestration risks same as direct air capture. 

Technology: Biochar and biomass methods; 
Maturity[A]: Low (TRL 2):
* Basic principles understood and reported; 
* System concept formulated; 
* Proof of concept shown in modeling and experimental results 
demonstrating its CO2 capturing ability–but CO2 sequestration aspects 
uncertain; 
* Not practiced on a scale to affect climate. No plans or prototypes 
for large-scale implementation; 
* Substantial uncertainties about capacity to reduce net emissions of 
CO2; 
Potential effectiveness[B]: Low:
* Maximum ability to reduce atmospheric CO2: 10–50 ppm by 2100; 
* Maximum annual sustainable reduction: 1–2 gigatons CO2-C equivalent 
of CO2, CH4, and N2O; 
* Net carbon negative under ideal conditions (comparable to bioenergy 
with CO2 capture and sequestration); 
Cost factors[C]: 
* Viability may depend on nature and extent of a carbon market; 
* Soil fertility outcomes; 
* Type of pyrolysis feedstock and related factors; 
* Process energy requirements for bioenergy production (e.g., pyrolysis); 
* Greatly varied estimates in the scientific literature: $2–$62 per 
ton of CO2 removed; 
Potential consequences[D]: 
* Potential land-use trade-offs; 
* Long-term effects on soil uncertain; 
* Health and safety of pyrolysis and biochar handling; 
* Local benefits to soil enhance crop yield. 

Technology: Land-use management (reforestation, afforestation, or 
reductions in deforestation; 
Maturity[A]: Low (TRL 2): 
* Basic principles understood and reported; 
* Techniques well established; 
* System concept formulated and estimates of its carbon mitigation 
potential reported based on modeling studies; 
* No experimental demonstration or proof of systemwide concept of CO2 
capture and sequestration by land-use activities; 
* Not practiced on a scale to affect climate. No plans for large-scale 
implementation; 
Potential effectiveness[B]: Low to medium: 
* Potential removal of 1.3–13.8 gigatons CO2 annually; 
* 0.4–14.2 metric tons of CO2 sequestered per acre per year; 
* Possible rerelease of sequestered CO2; 
Cost factors[C]: 
* Viability may depend on nature and extent of a carbon market; 
* Value of land in other uses; 
* Potentially large land area for growing or preserving forests; 
* Type of flora planted or preserved; 
* Natural resource requirements for maintenance and management of 
forests (e.g., water); 
* Measuring, monitoring, and verification; 
Potential consequences[D]: 
* Potential land-use trade-offs; 
* Possible cobenefits such as reduced water runoff. 

Technology: Enhanced weathering; 
Maturity[A]: Low (TRL 2): 
* Basic principles understood and reported; 
* System concept formulated; 
* No experimental demonstration of proof of system-wide concept; 
* Not practiced on a scale to affect climate. No plans or prototypes 
for large-scale implementation; 
Potential effectiveness[B]: Not rated: 
* Limited studies in literature; 
* Some estimates based on models but varied conclusions about levels 
of effectiveness; 
Cost factors[C]: 
* Viability may depend on nature and extent of a carbon market; 
* Design and implementation of silicate-based weathering scheme, 
including distribution and delivery of material; 
* Mining and transportation of silicate rock, and logistics; 
* Greatly varied estimates in the scientific literature: $4–$100 per 
ton of CO2 removed; 
Potential consequences[D]: 
* Potentially undesirable environmental and other consequences from 
large-scale mining and transportation. 

Technology: Ocean fertilization; 
Maturity[A]: Low (TRL 2): 
* Basic principles understood and reported; 
* System concept formulated; 
* Limited small-scale field experiments conducted but results unclear; 
* Published research mainly theoretical; 
* Not practiced on a scale to affect climate. No plans or prototypes 
for large-scale implementation; 
Potential effectiveness[B]: Low: 
* Maximum ability to reduce atmospheric CO2: 10–30 ppm by 2100; 
* Scientific uncertainty surrounding (1) duration of carbon 
sequestered in the ocean, (2) how ecological impacts might limit 
effectiveness, and (3) how often iron would need to be added; 
* Outcomes from limited experiments not understood or well documented; 
Cost factors[C]: 
* Viability may depend on nature and extent of a carbon market; 
* Design and implementation of ocean fertilization scheme, including 
distribution and delivery of material; 
* Mining and transportation of iron ore, and logistics; 
* Greatly varied estimates in the scientific literature: $8–$80 per 
ton of CO2 removed; 
Potential consequences[D]: 
* Ecological effect on ocean not well understood; 
* Risk of algal blooms causing anoxic zones in the ocean. 

Source: GAO. 

[A] In this report, we considered each technology’s maturity in terms 
of its readiness for application in a system designed to address 
global climate change. To do this, we used technology readiness levels 
(TRL), a standard tool that some federal agencies use to assess the 
maturity of emerging technologies. We characterized technologies with 
TRL scores lower than 6 as “immature” (section 8.1). The TRL rating 
methodology considers the maturity level of the whole integrated 
system rather than individual components of a particular technology. 

[B] We assessed potential effectiveness by considering the qualitative 
judgments of the Royal Society and reported estimates of two 
quantitative measures: (1) maximum ability to reduce the atmospheric 
CO2 (ppm) projected for 2100 and (2) annual capacity to remove CO2 
from Earth’s atmosphere (gigatons of CO2 or CO2 -C equivalent per 
year). Additionally, we reviewed scientific literature with respect to 
these measures of effectiveness and for assessments indicating the 
feasibility of implementing CDR technologies on a global scale to 
achieve a net reduction of atmospheric CO2 concentration. A technology 
was not assigned an overall qualitative rating when there were 
substantial uncertainties in the literature about its effectiveness 
(see section 8.1). 

[C] Cost factors are resources a system uses to remove CO2 from the 
atmosphere and store it. Some of the studies we reviewed indicated 
possible cost levels, which we provide here for illustration. We did 
not evaluate this information independently. 

[D] Includes potential consequences, risks, and cobenefits. 

[End of table] 

3.1.1: Direct air capture of CO2 with geologic sequestration: 

3.1.1.1: What it is: 

Direct air capture would chemically scrub CO2 directly from the 
atmosphere. In some conceptual designs, air is brought into contact 
with a CO2-absorbing liquid solution containing sodium hydroxide or 
with a solid sorbent in the form of a synthetic ion-exchange resin 
that selectively absorbs CO2 gas.[Footnote 20] Figures 3.1 and 3.2 
illustrate two different air-capture units. Figure 3.1 shows an 
artist’s rendering of the air-contactor design, and figure 3.2 
illustrates a CO2-absorbing synthetic tree made from a proprietary 
resin. A CO2-absorbing resin (sorbent material) could be shaped as 
a tree or as packing material placed inside a large column where it 
would be brought into contact with air. The CO2-rich solution or 
synthetic resin would be sent to a regenerator, where the CO2 would 
be separated from the liquid by thermal cycling or by exposure to 
humid air. The resulting concentrated stream of CO2 could be 
compressed to liquid form and delivered (by trucks, ships, or 
pipelines) to a sequestration site.[Footnote 21] The sorbent would 
be recycled to capture additional CO2. 

Experts have proposed the compression and transportation of captured 
CO2 for sequestration in deep underground geologic or saline 
formations. Most candidate geologic formations consist of layers of 
porous underground rock capped by layers of nonporous rock that would 
keep the injected fluids trapped in the lower pore spaces. The CO2 
would be compressed under elevated pressure (greater than 2,000 psi, 
or 13 megapascals (MPa)) and sequestered at the capture site, on 
shore, or in the deep ocean, where the hydrostatic head of the sea 
water above would keep the CO2 from rising to the surface (DOE 2006). 
[Footnote 22] 

Figure 3.1: Capturing and absorbing CO2 from air: 

[Refer to PDF for image: illustration] 

Source: Carbon Engineering Ltd. 

Note: This is a virtual rendering of an air-capture unit designed by 
Carbon Engineering Ltd. Each such unit would capture about 100,000 
tons of CO2 per ear. A battery of such units is intended to work with 
a chemical recovery plant to produce high-purity CO2. 

[End of figure] 

Figure 3.2: CO2-absorbing synthetic tree: 

[Refer to PDF for image: illustration] 

Source: Columbia University. 

Note: This is a synthetic tree made from a proprietary resin that can 
absorb CO2 from air. 

[End of figure] 

3.1.1.2: Maturity and potential effectiveness: 

We assessed the maturity of direct air capture of CO2 with geologic 
sequestration at TRL 3, given that the basic principles have been 
observed and reported, a system concept has been formulated, and the 
literature shows proof of concept—that is, the technology has had 
laboratory demonstrations using a prototype unit. Direct air capture 
of CO2 is probably decades away from commercialization, even though 
its fundamental chemistry and processes are well understood and 
laboratory-scale direct air-capture demonstrations are supported at 
two universities. According to the literature, direct air capture 
could theoretically remove total annual global anthropogenic CO2 
emissions, estimated at approximately 33 gigatons. The Royal Society 
reported that this technology had no “obvious limit” to the amount of 
CO2 it could capture from the atmosphere. Large-scale implementation, 
however, is currently neither cost-effective nor thermodynamically 
efficient. The main difficulty with direct air capture is in the 
removal of atmospheric CO2 in its extremely low concentration 
(approximately 390 ppm), which lowers the thermodynamic efficiency of 
the process (Ranjan 2010).[Footnote 23] This would make air capture 
even more challenging than, for example, capturing CO2 from a flue 
stack where the thermodynamic efficiencies were comparatively much 
higher (approximately 20 percent), mainly because of the higher 
concentration of CO2 in the flue gas (about 12 percent or 
approximately 120,000 ppm). 

The low atmospheric CO2 concentration presents other difficulties such 
as a significantly large energy penalty associated with the CO2 
absorption system for air capture (Herzog 2003). The total energy 
required to capture a unit of CO2 from air is such that if carbon-
based fuels such as coal were used as the energy source, more CO2 
would be released to the environment than removed (Zeman 2007). The 
energy process requirements for the direct air capture of CO2 would 
thus have to come from noncarbon or low carbon energy sources. Hence, 
substantial uncertainties surround the scalability of air capture. 

Our interviews with National Energy Technology Laboratory (NETL) 
engineers revealed that the capacity for sequestering CO2 in deep 
underground saline formations is vast enough to store essentially all 
CO2 emissions from coal-fired power plants within the United States. 
[Footnote 24] Carbon dioxide injection in subsurface geologic 
formations has been used for decades in enhanced oil recovery (EOR) to 
extract additional oil from depleted oil reservoirs. EOR’s history has 
made the overall challenges of the permanent sequestration of fluids 
well understood. The oil industry uses well-developed reservoir 
simulation models with computer programs that have sufficiently 
sophisticated computational power to routinely characterize subsurface 
oil reservoirs. It uses these tools extensively for oil production 
forecasting and to predict the state of fluids in the reservoirs, such 
as pressure distribution profiles and fluid flow characteristics. Oil 
exploration companies often conduct seismic surveys to determine the 
size and shape of subsurface reservoirs. They use well logging and 
sampling to determine the porosity, permeability, and resistivity of 
reservoirs and the hydrocarbons they contain.[Footnote 25] Recently 
published reports show that the private sector, universities, and 
national laboratories are developing and using computational 
techniques to model and simulate CO2 injection, transport, and storage 
(CMI 2010; Grimstad et al. 2009; Hao et al. 2009; MacMinn and Juanes 
2009; Stauffer et al. 2009). 

While advances in this area are notable, further research is needed to 
improve the existing technologies. What is known about CO2 injection 
for enhanced oil recovery could help in identifying deep underground 
saline formations suitable for permanent CO2 sequestration. Carbon 
dioxide sequestration is being researched for its feasibility in large-
scale demonstrations. Several worldwide projects are sequestering CO2 
in underground reservoirs to accelerate mainstream CO2 mitigation. 
[Footnote 26] 

While the technology behind CO2 injection is well developed, an 
integrated direct air capture and sequestration system has not been 
demonstrated. Furthermore, geologic sequestration of CO2 has not been 
practiced on the large scale envisioned by climate engineering. 

3.1.1.3: Cost factors: 

Cost estimates for direct air capture are based largely on theoretical 
calculations or assumptions, with some studies making qualitative cost 
comparisons (Royal Society 2009). Direct air capture’s relatively high 
cost results from the extremely low concentration of CO2 in the 
atmosphere (about 390 ppm) compared to a coal-fired stack (about 
120,000 ppm).[Footnote 27] Studies have reported that the steps in 
selective CO2 capture and release from a solvent consume more energy-—
and therefore account for the majority of the costs-—than transportation 
and underground sequestration. Besides the energy costs, other factors 
include transportation and logistics for sequestration of captured CO2 
and the long-term management of the sequestration site-—for example, 
CO2 injection, measuring, monitoring, and verification. 

Cost estimates for air capture in the literature vary substantially, 
from a low range of $27–$135 per ton of CO2 removed (Pielke 2009) to a 
higher range of $420–$630 or more per ton of CO2 removed (Ranjan and 
Herzog 2010).[Footnote 28] The cost estimate from Ranjan and Herzog 
took thermodynamics into account, concluding that direct air capture 
is unlikely to be a serious option in the absence of a carbon market. 
The literature estimates costs related to CO2 injection and monitoring 
of $0.20–$30 per ton of CO2 sequestered, reflecting a wide range of 
geologic parameters that could affect cost at specific locations (Metz 
et al. 2005). The potentially high cost of direct air capture of CO2 
and the lack of a carbon market could impede its large-scale adoption. 

3.1.1.4: Potential consequences: 

While direct air capture has minimally undesirable consequences 
(except those associated with handling process materials or 
chemicals), risks have been postulated for injecting large amounts of 
CO2 in deep underground saline formations (Oruganti and Bryant 2009; 
Ehlig-Economides and Economides 2010). Experience with geologic 
storage is limited, and the effectiveness of risk management methods 
still needs to be demonstrated for use with CO2 storage. Although CO2 
has been injected in oil reservoirs for decades, saline formations 
have not been proven safe or permanent. Leakage from underground 
sequestration sites could contaminate groundwater or cause CO2 to 
escape into the atmosphere. One technical paper expressing doubt about 
mitigation by underground geologic CO2 storage based its theoretical 
analysis on established reservoir models and assumptions of a closed 
form of reservoir that would render underground geologic CO2 storage 
impractical and unsuitable (Ehlig-Economides and Economides 2010). 
These assumptions and analyses were subsequently challenged by the 
U.S. Department of Energy’s (DOE) Pacific Northwest National 
Laboratory (PNNL) (Dooley and Davidson 2010). 

Other studies have reported that sealing faults or fissures in an 
underground reservoir could cause local pressure build-up with 
potential rock fractures at the weakest point, in the neighborhood of 
a fault, and cascading problems such as well failure, CO2 seepage, 
atmospheric CO2 release, and groundwater contamination (Oruganti and 
Bryant 2009).[Footnote 29] Unknown or undocumented preexisting wells 
in the reservoir provide another way for CO2 to escape to the 
atmosphere: industry experts we interviewed generally agreed that 
these concerns merit further analysis and a thorough characterization 
of geologic reservoirs. 

However, studies and simulations by industry, academia, and national 
laboratories suggest that such risk is generally small and manageable. 
For example, sites are chosen for sequestration only after the 
thorough characterization of a reservoir and its geology. Promising 
sites are assessed in detail to ensure minimal or no risk. NETL’s 
recent report advocated robust simulation to accurately model the 
transport and fate of CO2 for identifying, estimating, and mitigating 
risks arising from CO2 injection into the subsurface formation 
(Sullivan et al. 2011). Thus, CO2 sequestration in deep underground 
geologic formations might be safe, provided the risks were managed 
adequately. Our interviews and literature review suggest that careful 
site characterization and appropriate monitoring and verification 
during injection are key to avoiding hazards, steps DOE has pursued at 
American Electric Power’s West Virginia plant. 

3.1.2: Bioenergy with CO2 capture and sequestration: 

3.1.2.1: What it is: 

Bioenergy with CO2 capture and sequestration (BECS) would harvest a 
biomass crop such as switchgrass for biofuel production and capture 
and sequester the CO2 in geologic formations as it is released in the 
conversion of biofuel to electricity. Analogous to carbon capture and 
sequestration (CCS), this leverages what is known about bioenergy for 
fuels and CCS (Royal Society 2009).[Footnote 30] As vegetation grows, 
photosynthesis removes large quantities of carbon from the atmosphere. 
A harvested crop could be used to produce biofuel or simply as a fuel 
to generate electricity. The CO2 that would be released could be 
captured and sequestered in geologic formations. Since BECS actively 
absorbs CO2 from the atmosphere over the entire life of a growing 
plant, this approach could, on a large scale, reduce atmospheric CO2 
(Read 2008). 

3.1.2.2: Maturity and potential effectiveness: 

We assessed the maturity of BECS at TRL 2. Although it has been 
recognized that BECS can remove CO2 from the atmosphere, it has not 
been applied on a scale that would affect climate change (Carbo et al. 
2010). This is an emerging technology that leverages what is already 
known about CO2 capture and geologic sequestration. For example, the 
Energy Research Center of the Netherlands has a multidisciplinary 
research program dedicated to BECS. BECS potentially leads to negative 
CO2 emissions-—that is, to CO2 uptake from the atmosphere through 
natural sequestration of CO2 in biomass (Carbo et al. 2010). Ranjan 
and Herzog (2010) concluded that BECS could result in negative net 
emissions if the biomass were harvested sustainably. 

While the concept is simple, no instances of BECS are in operation. 
For example, BECS has not been demonstrated at any electric power 
generation facility. BECS is limited by the rate of growth of 
vegetation and conflicts with other uses of land, such as agriculture. 
For example, sequestering 1 gigaton of CO2 through BECS would require 
more than 200,000 square miles of land for plant growth (Ranjan 2010). 
While BECS could benefit local environments on a small scale, the 
Royal Society views it as having a low to medium capacity to remove 
CO2 from the atmosphere (Royal Society 2009; Royal Society 2001). 
According to the Royal Society, it can reduce the atmospheric CO2 
concentration by at most 50–150 ppm by the end of this century 
compared to a projected CO2 concentration of 500 ppm by 2100 (Royal 
Society 2009). 

3.1.2.3: Cost factors and potential consequences: 

BECS’s implementation costs are variable and depend on the 
availability of land for harvesting biomass, unintended emissions, the 
targeted amount by which atmospheric CO2 concentration would be 
reduced, and a carbon market, among other things (Azar et al. 2006). 
Other cost factors include transportation and logistics for 
sequestration, including the long-term management of the sequestration 
sites (as with direct air capture). An article by the Energy Research 
Center of the Netherlands concluded that incremental costs for CO2 
capture and storage are relatively low for biofuel production and are 
competitive with carbon capture and sequestration in fossil-fired 
power plants (Carbo et al. 2010). Another study reported BECS cost 
estimates of $150–$500 per ton of CO2 removed and suggested that BECS 
looked more promising than air capture from a cost perspective, 
although land requirements could potentially be large (Ranjan 2010). 
The literature describes BECS’s technical feasibility and potential as 
a negative-emissions energy system that is benign and free of risks 
associated with some other climate engineering approaches (Read and
Lermit 2005). As with direct air capture, however, the CO2 
sequestration aspects may pose risks. Furthermore, diverting resources 
to large-scale BECS activities could pose land-use trade-offs or 
affect food prices, water resources, and fertilizer use. 

3.1.3: Biochar and biomass: 

3.1.3.1: What it is: 

Biochar is a carbon-rich organic material that results from heating 
biomass, or terrestrial vegetation, in the absence of or in a limited 
supply of oxygen (Whitman et al. 2010).[Footnote 31] Biochar and 
biomass methods begin with the uptake of CO2 in photosynthesis 
(Lehmann 2007). The carbon locked in plants during their growth would 
be converted to charcoal instead of being released to the atmosphere. 
Biochar differs from charcoal in that its primary use is for 
biosequestration rather than fuel. That is, after plants die, biochar 
can be buried underground or stored in soil to keep carbon from being 
released to the atmosphere as CO2. 

3.1.3.2: Maturity and potential effectiveness: 

We rated the maturity of biochar and biomass at TRL 2. Ongoing and 
published research is available on the sustainability of biochar to 
mitigate global climate change (Woolf et al. 2010). While its proof of 
concept has been demonstrated in published modeling and experimental 
results, we found uncertainties in experimental data demonstrating the 
efficacy of biochar as a net carbon sink. For example, how long the 
captured CO2 in biochar will remain sequestered is uncertain. Similar 
to BECS, biochar production by pyrolysis is considered to be a carbon-
negative process. Reports show its benefits to soil, but the current 
immaturity of biochar sequestration technology precludes it from being 
practiced on a scale large enough to affect the climate. Its maximum 
sustainable potential for reducing net CO2, CH4, and N2O emissions has 
been estimated at 1–2 gigatons of CO2–C equivalent per year, compared 
to annual anthropogenic emissions of these greenhouse gases of 15 
gigatons of CO2–C equivalent (Laird et al. 2009; Woolf et al. 2010). 
[Footnote 32] Lehmann and colleagues (2006) quoted a higher future 
potential of biochar as a carbon sink of 5.5–9.5 gigatons of carbon per 
year by 2100. The Royal Society views biochar as low in effectiveness 
because its maximum anticipated reduction in atmospheric CO2 
concentration would be only 10–50 ppm by the end of this century 
compared to a projected atmospheric CO2 concentration of 500 ppm in 
2100 (Royal Society 2009). Therefore, biochar could be viewed as a 
small-scale contributor to a climate engineering approach to enhancing 
the global terrestrial carbon sink (Royal Society 2009). 

Although producing biochar and storing it in soil have been suggested 
as a way to abate climate change, provide energy, and increase crop 
yields, scientists have expressed uncertainty about its global effect 
and sustainability (Woolf et al. 2010). Its emission balance is highly 
variable and largely depends on the feedstock available, the existing 
soil fertility, and the local energy needs (Woolf et al. 2010). While 
biochar and biomass sequestration methods currently represent a 
trivial carbon sink, experts are researching them as a means of 
abating climate change and improving soil fertility. 

3.1.3.3: Cost factors and potential consequences: 

The costs of biochar and biomass are uncertain and inherently 
variable, depending on factors such as the type of feedstock used, the 
cost of pyrolysis, and carbon markets. According to one scientist, 
cost might depend more significantly on soil fertility outcomes. 
Roberts and colleagues found break-even prices of about $2–$62 per ton 
of CO2 removed, depending on the pyrolysis feedstock used (Roberts et 
al. 2010). While the literature has reported no negative consequences 
of biochar or biomass in soil, their handling and application might 
pose safety and health risks not yet adequately managed and captured 
in an overall cost structure of biochar systems. Pyrolysis could also 
affect health and safety. Biochar’s effects on emissions of N2O, CH4, 
and CO2 from soil are poorly characterized and need to be further 
researched (Whitman et al. 2010). Land-use trade-offs are possible 
(food versus the growth of biomass for fuel), but it is unclear 
whether they would be a factor for biochar. For example, the 
sustainable potential for biochar calculated by Woolf et al (2010) 
assumed no land-use trade-offs. 

3.1.4: Land-use management: 

3.1.4.1: What it is: 

Land-use management would enhance CO2 uptake in trees, soils, and 
biomass to increase their sequestration of carbon (DOE 2006). Although 
it could involve a variety of activities, we restricted our review to 
practices related to forestry, including reforestation, afforestation, 
and reductions in deforestation. Reforestation would plant trees where 
forests were previously cleared or burned; afforestation would plant 
trees where they had not historically grown. Reductions in 
deforestation would conserve existing forests. 

3.1.4.2: Maturity and potential effectiveness: 

We assessed the maturity of land-use management for climate 
engineering at TRL 2 because of the absence of experiments 
demonstrating its effectiveness at the scale required to affect the 
climate, despite the existence of technologies and knowledge
required to sequester carbon through land-use management for 
mitigation.[Footnote 33] Bottom-up regional studies and global top-
down models yield estimates of the potential for CO2 uptake through 
land-use management of 1.3–13.8 gigatons of CO2 per year in 2030 
(Nabuurs et al. 2007).[Footnote 34] 

The effectiveness of land-use management would depend on many factors, 
such as the vegetation’s species, location, and growth phase. For 
example, in the United States, afforestation could potentially 
sequester 2.2–9.5 metric tons of CO2 per acre per year, reforestation 
1.1–7.7 metric tons of CO2 per acre per year, depending on the types 
of trees and where they were planted (Murray et al. 2005). Nabuurs 
and colleagues reported a range for both of 0.4–14.2 tons of CO2 
per acre per Year worldwide.[Footnote 35] The rate of carbon 
accumulation also varies over a tree’s life cycle, starting out 
slowly when a tree is first planted, then increasing. Although 
land-use management practices are well understood and well 
established, their sequestration potential could be enhanced if 
scientists were to improve the understanding of carbon uptake and 
transfer in plants and soils. 

The capacity for sequestration through afforestation or reforestation 
also depends on the amount of land available. The estimates of 
sequestration potential reported by Nabuurs and colleagues suggest 
that the land area required to store a gigaton of CO2 per year could 
range from about 100,000 to 3.9 million square miles. Other potential 
challenges to land-use management for climate engineering include 
threats to permanence, such as fire, insect outbreaks, drought, or 
harvesting and problems in reliably measuring, monitoring, and 
verifying the amount of carbon stored, although progress has been 
made in this area, and costs may decline further as new methods are 
developed (Royal Society 2009; Sohngen 2009; Canadell and Raupach 
2008; Tavoni et al. 2007; Royal Society 2001).[Footnote 36] Climate 
change itself could also affect the capacity for sequestration through 
land-use management, but it is unclear whether capacity would be 
enhanced or diminished (Nabuurs et al. 2007). 

3.1.4.3: Cost factors and potential consequences: 

The costs of sequestration through land-use management would depend on 
a number of factors, most importantly the value of land in other uses 
(Sohngen 2009; Jepma 2008; Nabuurs et al. 2007; Sohngen and Sedjo 
2006). Costs would also arise from implementing and managing forestry 
practices (such as planting seedlings or harvesting); measuring, 
monitoring, and verification; engaging in other transactions (for 
example, developing and implementing long-term sequestration 
contracts); and system-wide adjustments (for example, changes in the 
price of land) (Sohngen 2009). Although land-use management is not 
generally regarded as risky, some practices could affect other systems 
as well as climate—for example, afforestation could reduce water 
runoff and affect the ecology. 

3.1.5: Enhanced weathering: 

3.1.5.1: What it is: 

Weathering refers to the physical or chemical breakdown of Earth’s 
minerals in direct contact with the atmosphere. Thousands of years 
of the weathering of silicate rocks, for example, have removed CO2 
naturally from the atmosphere, as the CO2 has reacted chemically 
with silicate rocks to form solid carbonates. The reaction can be 
written: 

CaSiO3 + CO2 yields CaCO3 + SiO2. 

This natural weathering of rocks could be enhanced by chemically 
reacting the silicate or carbonate rocks with CO2 in the presence of 
sea water to produce a carbonic acid solution that could be spread in 
the ocean (Rau et al. 2007; Royal Society 2009).[Footnote 37] 

3.1.5.2: Maturity and potential effectiveness: 

We assessed the maturity of enhanced weathering at TRL 2. While the 
basic principles of enhanced weathering have been observed and a 
concept proposed, we did not find published experimental results 
describing this approach as a CO2 reducing strategy. Neither enhanced 
weathering’s potential nor its technological elements have been 
clarified. The chemical reaction that facilitates it sometimes 
converts silicate rocks to carbonates by reaction with CO2. The 
carbonate materials resulting from enhanced weathering can be stored 
in the deep ocean or in soil. Similarly, the CO2 could react with 
carbonate rocks in seawater for conversion and storage as bicarbonate 
ions in the ocean where a large pool of such ions is already present. 
Since Earth’s silicate minerals are abundant, fixation in carbonate 
rocks could remove large amounts of CO2 from the atmosphere. 
Scientists have made a number of proposals to hasten natural 
weathering.[Footnote 38] For example, Rau and colleagues have reported 
its potential effectiveness based on models (Rau et al. 2007). While a 
very large potential for carbon storage in soils and oceans has been 
reported for this technology, its effectiveness remains uncertain. 
Enhanced weathering has not been practiced on a scale that would 
affect climate. 

3.1.5.3: Cost factors and potential consequences: 

Enhanced weathering’s costs are uncertain but are likely to be driven 
by mining and transportation costs (Royal Society 2009). Cost factors 
would include, for example, the design and implementation of a 
silicate-based weathering scheme and the distribution and delivery of 
raw materials. Rau and colleagues reported variability in cost 
estimates of $4–$65 per ton of CO2 removed under various assumptions, 
whereas IPCC’s estimate was $50–$100 per ton of CO2 captured (Rau et 
al. 2007; Metz et al. 2005). Overall, this technology is expected to 
be relatively simple and low in cost. Enhanced weathering that 
entailed large-scale mining and transportation could require 
additional energy and water and might adversely affect air and water 
quality (consistent with mining activities) and aquatic life in the 
long term (Royal Society 2009). Viability would depend on carbon 
markets. 

3.1.6: Ocean fertilization: 

3.1.6.1: What it is: 

Ocean fertilization releases iron to certain areas of the ocean 
surface to increase phytoplankton growth and promote CO2 fixation 
(Buesseler et al. 2008a). Oceans act as a large sink of CO2. 
Atmospheric CO2 is exchanged at the surface and slowly transferred to 
deeper waters with the capacity to store about 35,000 gigatons of 
carbon (Royal Society 2009).39 Phytoplankton, algae, and other 
microscopic plants on the ocean surface absorb CO2 in photosynthesis 
and recycle it to the bottom as organic matter. 

As the material settles into the deep ocean bottom, the microorganisms 
residing there use it for food, transferring CO2 back to the ocean as 
they breathe. The combined phytoplankton photosynthesis at the surface 
and respiration removes CO2 at the surface and releases it at greater 
depths. This is called the biological pump; studies suggest 
manipulating this pump to expedite CO2 sequestration. 

3.1.6.2: Maturity and potential effectiveness: 

We assessed the maturity of ocean fertilization at TRL 2. Basic 
principles have been observed and reported, and the concept has been 
formulated, with multiple studies proposing iron fertilization as an 
option for reducing CO2 in the atmosphere. Oceans are the largest 
natural absorbers of CO2 on the planet (at about 337 gigatons of CO2 
per year) and the largest natural reservoir of excess carbon (Rau 2009). 
However, most of the CO2 the oceans absorb is released back to the 
atmosphere in a continuous exchange while only a small portion of it 
is transferred to and sequestered in the deep ocean. 

The large number of theoretical studies attempting to understand 
fertilization’s complexities with sophisticated ocean models—as many 
as 12 between 1993 and 2008—have been complemented with only a few 
small-scale field experiments, whose results were uncertain and not 
well documented. Ocean fertilization studies suggest that 30,000–
110,000 tons of carbon could be sequestered from air by adding 1 ton 
of iron to certain parts of the ocean, but verifying this technology’s 
effectiveness is difficult and uncertain (Buesseler et al. 2008b). 
[Footnote 40] For example, modeling simulations suggest a cumulative 
storage potential of 26–70 gigatons of carbon (equivalent to 95–255 
gigatons of CO2) for large-scale ocean fertilization—relatively low 
compared to terrestrial sequestration potential in vegetation (200 
gigatons of carbon) or in deep geological formation (several hundred 
gigatons of carbon) (Bertram 2009). 

Another study based on models reported that large-scale sustained iron 
fertilization (30 percent of the global ocean area) could store at 
most 0.5 gigatons of carbon (equivalent to about 2 gigatons of CO2) per 
year. This amount is small compared to anthropogenic emissions of 
approximately 8–9 gigatons of carbon (equivalent to about 30–33 
gigatons of CO2) per year. According to the Royal Society, ocean 
fertilization could reduce the atmospheric CO2 concentration by a 
maximum of 10–30 ppm by the end of this century, which would be 
considered to be low in effectiveness. While these estimates have 
not been substantiated experimentally, these studies show that even 
sustained fertilization of oceans would have only a minor effect on 
the increasing atmospheric CO2 concentration (Secretariat of the 
Convention on Biological Diversity 2009). 

Ocean fertilization as a long-term carbon storage strategy has not 
been demonstrated (Buesseler et al. 2008b). The literature 
characterizes its effectiveness as highly uncertain, the models 
governing biochemical cycling of nutrients and the circulation of 
ocean currents as poorly understood or uncertain, and the strategy for 
mitigating CO2 as risky. For example, the science is unclear regarding 
ecological consequences, the duration of carbon sequestered in the 
oceans, and the frequency with which iron should be added (Buesseler 
et al. 2008b). Scientists are researching the ocean’s biochemical 
processes and the effects and efficacy of iron fertilization to better 
understand them. 

3.1.6.3: Cost factors and potential consequences: 

Ocean fertilization could be cost-effective at capturing and 
sequestering atmospheric CO2 in the deep ocean, but relatively little 
is known about its efficacy.[Footnote 41] The design and 
implementation of any ocean fertilization scheme, including mining, 
distribution, and delivery of materials, would affect its success. The 
literature has reported significant uncertainty with respect to cost. 
Some ocean fertilization modeling has helped determine its efficiency 
at removing carbon from the atmosphere but estimating a cost range is 
difficult. One estimate put the minimum cost at approximately $8 per 
ton of CO2 removed (Buesseler et al. 2008b). An evaluation by Boyd 
characterized ocean fertilization as a medium-risk strategy with costs 
of $8–$80 per ton of CO2 removed (Boyd 2008). 

Because ocean fertilization is not well understood and is largely 
theoretical, it could pose ecological risks (Royal Society 2009). A 
report from the Woods Hole Oceanographic Institution indicated that 
iron-fertilized phytoplankton blooms could eventually prevent oceans 
from sustaining life. An image in that report showed bloom and anoxic 
(or dead) zones stretching for hundreds of kilometers (Buesseler et 
al. 2008b).[Footnote 42] The Royal Society and the U.K. House of 
Commons Science and Technology Committee reported that ecosystem-based 
methods—=whether fertilizing the ocean or blocking sunlight=—would be 
subject to unknown risks if implemented on a large scale. Other 
studies have also presented images of the unintended consequences of 
manipulating ecosystems—-dead zones in the sea resulting from 
phytoplankton boom are an example (Buesseler et al. 2008b). Other 
potential risks of ocean fertilization are greater ocean 
acidification, additional emissions of greenhouse gases, and the 
reduction of oxygen in the ocean to levels some species cannot 
tolerate (Buesseler et al. 2008b). 

3.2: Selected SRM Technologies: 

In this section, we summarize our assessment of the maturity of 
selected SRM technologies and present information from peer-reviewed 
literature on their potential effectiveness, cost factors, and 
potential consequences. TRL ratings indicate the maturity of each 
technology. Potential effectiveness is described in terms of the 
anticipated ability to counteract warming caused by doubling the 
preindustrial atmospheric concentration of CO2. In calculating our 
ratings, we relied on reported results from: 

* climate engineering modeling studies using general circulation 
models (GCM) and; 

* energy balance studies of the effects of increasing reflectivities. 

Cost factors represent resources required to counteract global warming 
from doubling the preindustrial atmospheric concentration of CO2 or, 
for technologies that are not anticipated to be fully effective, the 
resources required to counteract warming to the maximum extent 
possible. Potential consequences associated with each technology 
include reported negative consequences and cobenefits. (See table 3.2.) 

Table 3.2 Selected SRM technologies: Their maturity and a summary of 
available information: 

Technology: Stratospheric aerosols; 
Maturity[A]: Low (TRL 1): 
* Basic principles understood and reported; 
* No system concept proposed; 
Potential effectiveness[B]: Potentially fully effective: 
* Aerosols must be continuously replaced; 
Cost factors[C]: 
* Design, fabrication, testing, acquisition, and deployment of aerosol 
delivery scheme, including distribution and delivery mechanisms, 
fabrication of aerosol dispersal equipment, and all associated 
infrastructure; 
* Literature-based estimates vary significantly: $35 billion to $65 
billion in the first year; $13 billion to $25 billion in operating 
cost each year thereafter; 
Potential consequences[D]: 
* Little change in global average annual precipitation; 
* Disruption of Asian and African summer monsoons with accompanying 
reduction in precipitation; 
* Delayed ozone layer recovery in southern hemisphere and about a 30-
year delay in recovery of Antarctic ozone hole; 
* Scattering interference with terrestrial astronomy; 
* Efficiency of solar-collector power plants reduced by increased 
diffuse radiation. 

Technology: Marine cloud brightening; 
Maturity[A]: Low (TRL 2): 
* Basic principles understood and reported; 
* System concept proposed; 
* Proof of concept not demonstrated; 
Potential effectiveness[B]: Potentially fully effective: 
* Model-dependent estimates of effectiveness vary; 
* Clouds must be continuously brightened; 
Cost factors[C]: 
* Design, fabrication, testing, acquisition, and deployment of a fleet 
of 1,500 wind-driven spray vessels; 
* Fleet infrastructure and operation; 
* Estimates in the scientific literature vary significantly at $42 
million for development, $47 million for production tooling, $2.3 
billion to $4.7 billion for 1,500-vessel fleet acquisition; 
Potential consequences[D]: 
* Small changes in global average temperature, regional temperatures, 
and global precipitation; 
* Large regional changes in precipitation, evaporation, and runoff; 
both precipitation and runoff increase, and the net result might not 
“dry out” the continents. 

Technology: Scatterers or reflectors in space: 
* Earth orbit; 
* Deep space; 
Maturity[A]: Low (TRL 2): 
* Basic principles understood and reported; 
* System concepts proposed, but proof of concept not demonstrated; 
Potential effectiveness[B]: Potentially fully effective: 
* Spacecraft’s limited lifetime; 
Cost factors[C]: 
* Design, fabrication, testing, acquisition, and deployment of a fleet 
of millions to trillions of reflecting or scattering spacecraft; 
* Launch vehicle; 
* Infrastructure and operation; 
* Estimates in the scientific literature vary significantly: an 
estimate of $1.3 trillion and an estimate of less than $5 trillion; 
Potential consequences[D]: 
Earth-orbit technologies: 
* A cool band in the tropics with unknown effects on ocean currents, 
temperature, precipitation, and wind; 
* A multitude of bright “stars” in the morning and evening that would 
interfere with terrestrial astronomy; 
Deep-space technologies:
* Annual average tropical temperatures a little cooler; 
* Annual average higher latitude temperatures a little warmer; 
* Small reduction of annual global precipitation. 

Technology: Terrestrial reflectivity: 
* Deserts; 
* Flora; 
* Urban or settled areas; 
Maturity[A]: Low (Up to TRL 2): 
* Basic principles understood and reported; 
* One technology proposed a system concept but without demonstrated 
proof of concept; 
Potential effectiveness[B]: Potential effectiveness of 0.21 (urban 
areas) to more than 57 percent (deserts); 
* Sustainability issues: maintaining reflectivity and missing 
information on reflective flora; 
Cost factors[C]: 
* Design, fabrication, testing, acquisition, and deployment of 
reflective material or flora; 
* Infrastructure and maintenance; 
* Estimates in the scientific literature to maintain reflectivity vary 
greatly from $78 billion (urban areas) to $3 trillion per year (deserts); 
Potential consequences[D]: 
* Cool deserts might change large-scale patterns of atmospheric 
circulation; 
* Reflective crops would probably not significantly affect global 
average temperature but might reduce regional summer temperatures; 
* Reflective urban areas would probably not affect global average 
temperature but might reduce air-conditioning costs. 

Source: GAO. 

[A] In this report, we considered each technology’s maturity in terms 
of its readiness for application in a system designed to address 
global climate change. To do this, we used technology readiness levels 
(TRL), a standard tool that some federal agencies use to assess the 
maturity of emerging technologies. We characterized technologies with 
TRL scores lower than 6 as “immature” (see section 8.1). The TRL 
rating methodology considers the maturity level of the whole 
integrated system rather than individual components of a particular 
technology. 

[B] We assessed potential effectiveness in terms of a technology’s 
potential ability to counteract global warming caused by doubling the 
preindustrial CO2 concentration. 

[C] Cost factors are resources a system uses to counteract global 
warming caused by doubled preindustrial atmospheric CO2 concentration, 
or for technologies that are potentially not fully effective, 
resources required to counteract global warming to the maximum extent 
possible. Some of the studies we reviewed indicate possible cost 
levels, which we provide here for illustration. We did not evaluate 
this information independently. 

[D] Includes potential consequences, risks, and cobenefits. 

[End of table] 

3.2.1: Stratospheric aerosols: 

3.2.1.1: What it is: 

Deploying aerosols would use knowledge gained from volcanic eruptions 
that inject aerosols into the stratosphere, cooling Earth for short 
periods. Aerosols smaller than 1 micrometer in diameter (1 millionth 
of a meter) would cool Earth primarily by scattering a fraction of the 
solar radiation. While enough solar radiation would be scattered back 
into space to cool Earth, a larger fraction would be scattered toward 
Earth, increasing diffuse radiation (Robock 2000). Larger aerosols 
would scatter solar radiation less efficiently and absorb both solar 
and thermal radiation, acting somewhat like a greenhouse gas (Rasch, 
Crutzen, and Coleman 2008; Rasch, Tilmes et al. 2008). If the volcanic 
sulfate aerosols were sufficient to cool Earth, the sulfates would 
accumulate in size and remain in the stratosphere for about 1 year. 
[Footnote 43] 

3.2.1.2: Maturity and potential effectiveness: 

We assessed stratospheric aerosol technology at TRL 1 because only 
basic principles have been reported. We could not rate this technology 
at TRL 2 because we found no system concepts reported in the 
literature. Recent estimates using complex coupled atmosphere-ocean 
general circulation models indicated that about 3 million tons of 
sulfur injected per year into the stratosphere and forming volcanic-sized 
sulfate aerosols would compensate for the doubled CO2 concentration 
(Rasch, Crutzen, and Coleman 2008). In a recent investigation using a 
chemistry climate model, Heckendorn and colleagues found that sulfates 
from continuous injection of sulfur gas formed larger aerosols that 
would be less effective than volcanic sized aerosols (Heckendorn et 
al. 2009). Because sulfate aerosols have a lifetime of about a year 
in the stratosphere, they must be replenished to sustain their cooling 
effect (Rasch, Crutzen, and Coleman 2008). 

3.2.1.3: Cost factors and potential consequences: 

It could cost $35 billion to $65 billion in the first year and $13 billion 
to $25 billion in each subsequent year to inject sufficient sulfate 
aerosols into the stratosphere to counteract global warming caused by 
doubling preindustrial CO2 concentration. Robock and colleagues 
estimated the cost of injecting 1 million tons of a sulfur gas 
(that will become sulfate aerosols) per year into the stratosphere 
(Robock et al. 2009). Since about 3 million tons of sulfur might be 
required to counteract global warming caused by doubling preindustrial 
CO2 concentration, we scaled Robock and colleagues’ cost estimate, 
assuming no economy of scale, to 3.2 million and 6 million tons per year 
of hydrogen sulfide and sulfur dioxide, respectively (gases containing 
3 million tons of sulfur). The scaled cost estimate is $35 billion to 
$65 billion in the first year (the cost of the airplanes used to inject 
the aerosols plus 1 year of operations) and $13 billion to $25 billion 
in operating costs in each subsequent year to sustain the effort. 
Robock and colleagues considered several potential aerosol injection 
systems, including KC-135 aircraft-refueling tankers and F-15 aircraft. 
They found that the total cost of using the aircraft-refueling tankers 
would be lower than the total cost of the alternatives, but the tankers 
do not fly high enough (Robock et al. 2009). 

Using the F-15s was the least expensive among the remaining 
alternatives. Robock and colleagues’ estimated operating cost for the 
F-15s was an upper bound based on the hourly cost of the tankers; the 
authors expected that the hourly cost of operating F-15s would be 
lower because they use less fuel and fewer pilots than the tankers. 
However, because the F-15s are smaller than the tankers, they would 
require more than ten times the number of trips that the tankers would 
require to inject the same quantity of aerosols. Other alternatives 
considered by the authors, including injection systems based on 
artillery or balloons, would be significantly more expensive than the 
fighter aircraft. The scaled estimates do not include system design, 
fabricating aerosol dispersal equipment, or infrastructure. 

Volcanic stratospheric sulfate aerosols increase diffuse solar 
radiation, which can increase the growth of terrestrial vegetation 
(Robock et al. 2009). Cooling by these aerosols can interfere with the 
hydrological cycle (Trenberth and Dai 2007). The surface area of these 
aerosols can lead to reactions that deplete stratospheric ozone 
(Tilmes et al. 2008; Solomon 1999). Robock and colleagues reported 
performing a modeling study using an IPCC “business-as-usual” scenario 
with an increase in greenhouse gases and sufficient stratospheric 
sulfate aerosols to significantly cool Earth. They found little annual 
average change in global precipitation but significantly reduced 
precipitation in India, with large reductions in summer monsoon 
precipitation in India and northern China that could threaten food and 
water supplies. They found a similar reduction in the Sahel in Africa. 
They also found that abruptly stopping the injection of aerosols would 
raise temperature rapidly and be difficult to adapt to. 

In another modeling study using the same greenhouse gas scenario, 
Tilmes and colleagues found that changes in stratospheric dynamics and 
chemistry delayed the recovery of the ozone layer in middle and high 
latitudes in the southern hemisphere and reduced the ozone layer in 
high latitudes in the northern hemisphere (Tilmes et al. 2009). The 
recovery of the Antarctic ozone hole would be delayed by about 30 
years. They stated that the increase in ultraviolet radiation of up to 
10 percent observed in the middle and high latitudes in the 1980s and 
1990s would probably worsen. 

Using an aerosol-chemistry climate model, Heckendorn and colleagues 
found larger sulfate aerosols, which increased stratospheric water 
vapor and reduced stratospheric ozone (Heckendorn et al. 2009). 
Additional water vapor (a greenhouse gas) would reduce effectiveness 
but reduced ozone (another greenhouse gas) would increase 
effectiveness. The net effect is not known because detailed radiation 
forcing calculations were beyond the scope of the 2009 study. 

Other collateral consequences of stratospheric aerosols would include 
negative effects on astronomy and on solar energy power plants. 
Suspended above all terrestrial telescopes, stratospheric aerosols 
would interfere with terrestrial optical astronomy. Scattering from 
stratospheric aerosols would also reduce the efficiency of power 
plants that concentrate solar radiation to generate electricity. 
Although solar radiation scattered from aerosols would result in 
significant diffuse radiation, the concentrators in these power plants 
cannot use it. For example, the peak power output of Solar Electric 
Generating Stations in California fell up to 20 percent after Mount 
Pinatubo erupted, even though total solar radiation was reduced by 
less than 3 percent (Murphy 2009). Aerosol effects in the stratosphere 
could be reversed by stopping their injection because sulfate aerosols 
remain in the stratosphere for approximately 1 year. 

3.2.2: Cloud brightening: 

3.2.2.1: What it is: 

Reflectivity in clouds generally increases as the number of water 
droplets in them increases (Twomey 1977). Latham and colleagues 
proposed to increase the reflectivity of marine clouds by increasing 
the number of water droplets (Latham et al. 2008). They proposed to 
loft droplets of sea water micrometers in diameter that would shrink 
by evaporation as they rose into the base of the clouds, where 
moisture would condense, and increase their number (Latham et al. 
2008). In designing wind-driven spray vessel-based cloud brightening 
equipment, Salter, Sortino, and Latham (2008) proposed to avoid the 
problems of remotely operating and maintaining sails, ropes, and 
reefing gear by using Flettner rotors—vertical spinning cylinders that 
produce forces perpendicular to the wind direction—instead of sails. 

3.2.2.2: Maturity and potential effectiveness: 

We assessed cloud brightening technology at TRL 2. Basic principles 
have been reported, allowing at least TRL 1. Demonstration of proof of 
concept has not been reported (Salter, Sortino, and Latham 2008), 
ruling out TRL 3. A system concept has been proposed, and there is 
encouraging evidence that this technology might work: Ship tracks 
(which are white streaks observed in satellite images of the oceans 
that are attributed to sulfate aerosols in the exhaust trails from 
ships) indicate that adding aerosols to the marine environment can 
make clouds, but they fall short of proof of concept that lofting 
droplets of sea water into marine clouds will brighten them as assumed 
in the analyses discussed below. Having a system concept does not 
automatically qualify this technology for TRL 2 but it cannot be ruled 
out, given the information available in Salter, Sortino, and Latham 
(2008). 

Four recent investigations of cloud brightening reported effectiveness 
ranging from fully effective to fully effective with a significant 
margin. Latham and colleagues used two different atmosphere-only 
general circulation models and calculated the increased reflectivity 
of brightened clouds. They found full effectiveness with significant 
margin for one when they brightened all marine clouds and full 
effectiveness for the other when they brightened clouds over 35 to
45 percent of the ocean area (Latham et al. 2008). Using analytical 
methods, Lenton and Vaughan found full effectiveness but warned that
conversion of droplets reaching the base of the clouds into droplets 
in the clouds is not well understood and, if the conversion is 
insufficient, this technology would not be effective (Lenton and 
Vaughan 2009). Rasch and colleagues used a fully coupled atmosphere-
ocean general circulation model and found full effectiveness if clouds 
were brightened over between 40 percent and 70 percent of the oceans 
(Rasch et al. 2009). Bala and colleagues used a similar atmosphere 
model coupled to a simple slab-ocean/sea-ice general circulation model 
and found full effectiveness when they reduced water droplet size in 
all marine clouds (Bala et al. 2010). Brightened clouds have a 
lifetime of a few days and must be continuously brightened to sustain 
cooling (Latham et al. 2008). 

3.2.2.3: Cost factors and potential consequences: 

It could cost $2.4 billion to $4.8 billion to brighten enough marine 
clouds to compensate for a doubling of the concentration of CO2 in the 
atmosphere. Salter, Sortino, and Latham (2008) estimated that full 
effectiveness would require a fleet of 1,500 of their wind-driven 
spray vessels.[Footnote 44] Their cost estimate did not include system 
testing, acquisition, deployment, infrastructure, and operation. 

The investigations using coupled atmosphere-ocean general circulation 
models predicted climate changes. Rasch, Latham, and Chen (2009), 
using a fully coupled ocean-atmosphere model, found that as they 
brightened increasing fractions of clouds, they not only could 
counteract global warming caused by doubling atmospheric CO2 
concentrations but could also counteract the effects of this warming 
on sea ice and precipitation—but not all at the same time. For 
example, when they counteracted global warming, they overcompensated 
for the loss of south polar sea ice and the change in global 
precipitation and undercompensated for the loss of north polar sea ice 
(Rasch, Latham, and Chen 2009). Bala and colleagues used an atmosphere 
coupled to a simple slab-ocean/sea-ice model and found that: 

* changes in global and regional annual average temperatures were 
small, 

* changes in global annual precipitation were small, and, 

* regional changes in precipitation, evaporation, and runoff were 
large. Precipitation and runoff increased over land, particularly over 
Central America, the Amazon, India, and the Sahel, suggesting that 
this technology might not dry the continents (Bala et al. 2010). 

The brightness of clouds could be returned to normal within a few days 
of ceasing to deploy the cloud brightening technology (Latham et al. 
2008). 

3.2.3: Scatterers or reflectors in space: 

3.2.3.1: What it is: 

Proposals have been made to reduce the solar radiation that reaches 
Earth by placing scatterers or reflectors in Earth orbit or in deeper 
space at a stable position between Earth and the Sun called the inner 
Lagrange point (or L1)-—approximately 1 percent of the distance from 
Earth toward the Sun-—where gravitational and orbital forces are 
balanced. 

Proposed technologies include scatterers or reflectors in Earth orbit. 
NAS dismissed the use of 55,000 110-ton 100-square kilometer 
reflective solar “sails” in orbit that would reflect 1 percent of 
solar radiation as “a very difficult if not unmanageable control 
problem” (NAS 1992).[Footnote 45] 

Pearson, Oldson, and Levin (2006) proposed Saturn-like rings of space 
dust or parasol spacecraft. To be practical, the space dust option 
would require the ability to fabricate in space. The ring of 
spacecraft would consist of 5 million parasol spacecraft, each 
measuring 5 km long by 200 m wide (1 square km) and having mass of 
1,000 kg. They would be electromagnetically tethered in Earth’s 
equatorial plane at altitudes between 1,300 km and 3,200 km. The 
spacecraft’s parasols would point at the Sun and shade the tropics of 
the winter hemisphere. 

The proposed options also included scatterers or reflectors at L1: 

* a 3,400-ton, 1,800-km diameter diaphanous scattering screen 
fabricated in low Earth orbit (Teller et al. 1997); 

* a 100-million ton, 2,000-km diameter, 10 micrometer thick opaque 
disc or transparent prism made from moon glass (Early 1989); 

* a 420-million ton, 3,600-km diameter, 5.1 micrometer thick iron 
mirror made from asteroids (McInnes 2002); 

* 16 trillion spacecraft (a total of 19 million tons), each 0.6 meters 
in diameter and 5 micrometers thick, covering an ellipse 6,200 km by 
7,200 km (Angel 2006). 

The first three technologies are impractical at this time because they 
require manufacturing capabilities in space. The fourth technology 
would consist of 16 trillion autonomous fliers, manufactured on Earth, 
launched electromagnetically into orbit, and moved into position with 
ion propulsion. Once in position, they would use a system analogous to 
the global positioning system and radiation pressure motive power with 
tilting mirrors for station-keeping. 

3.2.3.2: Maturity and potential effectiveness: 

We assessed scattering or reflecting technologies in space at TRL 2. 
Basic principles have been reported and system concepts have been 
proposed, allowing at least TRL 1, but demonstration of proofs of 
concept have not been reported (Angel 2006; Pearson et al. 2006), 
ruling out TRL 3. Having system concepts does not automatically 
qualify these technologies for TRL 2 but it cannot be ruled out, given 
the information available in Angel (2006) and in Pearson, Oldson, and 
Levin (2006). 

Pearson, Oldson, and Levin (2006) used a simplified one-dimensional 
energy balance model to design a system of parasol spacecraft to 
reduce solar radiation to compensate for doubled preindustrial CO2 
concentrations in the atmosphere. Their design study indicated that 
this could be accomplished by shading about 36 percent of a Saturn-
like equatorial ring with their parasol spacecraft. 

Angel’s autonomous spacecraft fliers were designed to reduce solar 
radiation by the 1.8 percent required by general circulation models 
(in this case, an atmospheric general circulation model coupled to 
slab ocean and sea-ice models) (Govindasamy and Caldeira 2000) to 
compensate for global warming caused by doubling the preindustrial CO2 
concentration. 

None of these space-based SRM technologies would be a realistic 
contributor in the short term. They should not be dismissed from 
future consideration, particularly if climate engineering were to be 
employed for as long as a century (Royal Society 2009). However, the 
spacecraft would have to be replaced when they reached the end of 
their service life to sustain cooling. 

3.2.3.3: Cost factors and potential consequences: 

Following NAS’s assertion that the cost of establishing space-based 
climate engineering projects would be dominated by launch costs (NAS 
1992), Pearson, Oldson, and Levin (2006) estimated a cost of $1.3 
trillion for their equatorial Saturn-ring-like collection of 
reflectors. Their launch cost was based on a proposed ram accelerator 
and an orbiting tether, achieving a low Earth orbit launch cost of 
$250 per kg. This cost estimate did not include design, fabrication, 
testing, acquisition, deployment, infrastructure, or operation. They 
did not provide an explicit projected lifetime for their spacecraft. 
However, they did explore the consequences of a 100-year lifetime. 
Following on their launch costs as discussed, the replacement cost 
estimate would be $13 billion per year. 

It could cost less than $5 trillion for Angel’s fliers at L1. 
Fabrication costs were estimated at $50 per kg, which Angel rounded up 
to $1 trillion. Estimates of launch costs were based on 20 
electromagnetic launchers each launching 800,000 fliers into orbit 
every 5 minutes for 10 years. The electromagnetic launchers would put 
the fliers into orbit, and ion propulsion would move them to L1, where 
the fliers would use mirrors to adjust radiation pressure from solar 
radiation to maintain position. The cost estimate for the launchers 
was $600 billion and the estimated cost of electrical energy for the 
launchers was $150 billion; Angel rounded the sum to $1 trillion, 
corresponding to a launch cost of $50 per kg. Angel stated that a total 
project cost, including development and operations, of less than $5 
trillion seemed possible but gave insufficient detail to evaluate 
development and operation costs. Also, he did not explicitly mention 
testing, acquisition, deployment, and infrastructure. The projected 
lifetime for the fliers is 50 years, which means that 320 billion 
fliers would have to be replaced every year, but Angel did not 
provide an estimated cost for replacement. 

Orbital equatorial Saturn-ring-like disposition of reflectors is a 
regional technology that would shade and cool the winter portion of 
the tropics. The design study used a simplified energy balance model 
of Earth’s climate system, not a general circulation model (GCM). 
Therefore, climate responses other than a set of average temperatures 
for bands of latitudes are not available. The effects on the ocean 
currents, ocean temperature, precipitation, and wind are unknown. 
However, a multitude of bright “stars” at morning and evening would 
interfere with terrestrial astronomy. 

Uniformly reducing solar radiation with reflectors or scatterers at L1 
enough to counteract the warming effect of doubling the concentration 
of CO2 might not significantly reduce CO2 fertilization from doubling 
CO2. Govindasamy and colleagues modeled this effect with normal and 
uniformly reduced solar radiation at both the concentration of CO2 in 
1991 and double the concentration of CO2 in 1991 (Govindasamy et al. 
2002). In their modeling study, they chose a reduction in solar 
radiation that could nearly counteract the warming effect of doubling 
the concentration of CO2. They found that doubling CO2 resulted in CO2 
fertilization—-that is, plant productivity increased by 76–77 percent 
and biomass increased by 87–92 percent.[Footnote 46] When they 
uniformly reduced solar radiation to counteract the warming effect of 
this doubling of the concentration of CO2, they found that plant 
productivity fell by 2.3–3 percent and biomass fell by 1.9–4.7 
percent. Govindasamy and colleagues indicated that in reality, CO2-
fertilized ecosystems might encounter nutrient limitations, 
diminishing the magnitude but not changing the direction of the CO2 
fertilization. Furthermore, they indicated that CO2 fertilization 
might affect ecosystems in ways not represented in the model through 
species abundance and competition, habitat loss, biodiversity, and 
other disturbances. This investigation applies directly to reflectors 
or scatterers at L1 that uniformly reduce solar radiation without 
otherwise affecting the Earth system. Therefore, this modeling study 
indicated that CO2 fertilization would outweigh reduction in plant 
productivity because of uniformly reduced solar radiation from 
reflectors or scatterers at L1. 

Modeling studies indicate that SRM technologies that counteract the 
greenhouse effect of a doubled preindustrial concentration of CO2 by 
uniformly reducing solar radiation also indicate that the globally 
averaged engineered climate is very similar to the globally averaged 
preindustrial climate (Caldeira and Wood 2008; Govindasamy et al. 
2002; Govindasamy and Caldeira 2000). These studies indicated that 
annual average tropical temperatures would be a little cooler,
the higher latitudes might be a little warmer, and the reduction of 
annual global precipitation would be small. 

Since the spacecraft in Earth orbit and at L1 would be controlled, it 
should be possible to reverse these technologies. It is assumed that 
parasol spacecraft in Earth orbit, which are controlled to maximize 
shading, could be reversed by commanding the parasols to minimize 
shading (Pearson et al. 2006). Fliers at L1 could be reversed by 
commanding the fliers to go into halo orbits (Angel 2006). 

3.2.4: Reflective deserts, flora, and habitats: 

3.2.4.1: What it is: 

Increasing Earth’s surface reflectivity in deserts, flora, and settled 
areas has been proposed. Gaskill would double the reflectivity of 
deserts by covering them with white polyethylene, estimating that up 
to 12 trillion square meters of Earth’s deserts (about 2 percent of 
Earth’s surface) would be suitable for reflectivity enhancement 
(Gaskill 2004; Gaskill n.d.). 

Similarly, Ridgwell and colleagues proposed increasing the 
reflectivity of crops by selecting varieties that are glossy or have 
reflective shapes and structure (Ridgwell et al. 2009). Hamwey 
proposed to increase the reflectivity of open shrubland, grasslands, 
and savannah and to double the reflectivity of all human settlements, 
excluding agricultural land (Hamwey 2007). Akbari, Menon, and 
Rosenfeld (2009) proposed to increase the reflectivity of urban 
roofs and pavement. 

3.2.4.2: Maturity and potential effectiveness: 

We assessed increased reflectivity of desert technology at TRL 2. 
Basic principles have been reported and a system concept has been 
proposed, allowing at least TRL 1, but demonstration of proof of 
concept has not been reported (Gaskill 2004; Gaskill n.d.), ruling out 
TRL 3. Having a system concept does not automatically qualify this 
technology for TRL 2 but it cannot be ruled out given the information 
available in Gaskill (2004) and Gaskill (n.d.). We assessed 
technologies for increasing the reflectivity of flora and settled 
areas at TRL 1 because only basic principles have been reported; the 
absence of system concepts precluded a rating of TRL 2 (Ridgwell et 
al. 2009; Hamwey 2007). 

Technologies for increasing the reflectivity of deserts could 
potentially be more than 57 percent effective in compensating for 
global warming from doubled preindustrial CO2. Gaskill proposed to 
increase reflectivity from 36 to 80 percent over 10 trillion square 
meters of the 12 trillion square meters of desert areas that he deemed 
suitable (Gaskill 2004; Gaskill n.d.). The Royal Society’s (2009) and 
Lenton and Vaughan’s (2009) interpretation of Gaskill corresponded to 
an effectiveness of 74 percent. Lenton and Vaughan’s refinement of 
Gaskill’s proposal corresponded to 57 percent effectiveness, 
accounting for lower average intensity of solar radiation over land 
and absorption in the atmosphere. However, they also stated that 
deserts have higher-than-average solar radiation because they are 
generally in the lower latitudes, so that increased reflectivity would 
be somewhat more effective (Lenton and Vaughan 2009). Sustaining 
reflective deserts would require maintenance. 

Increasing the reflectivity of flora could be up to about 25 percent 
effective. Ridgwell and colleagues investigated the effect of 
increasing the reflectivity of crops with a fully coupled climate 
model (Ridgwell et al. 2009). They focused on an increase of 20 
percent, asserting that an increase of 35 percent observed after 
coating plants with a white chalky suspension provided a first-order 
guide as to the possible upper limit of reflectivity increase. They 
found a global average cooling of only 0.11 degrees Celsius. Hamwey 
investigated increasing the reflectivity of open shrubland, 
grasslands, and savannah with a static two-dimensional radiative 
transfer model (Hamwey 2007). His preliminary estimate was that an 
increase in reflectance of 25 percent corresponded to about 16 percent 
effectiveness. Lenton and Vaughan interpreted these results with 
energy balance analyses (Lenton and Vaughan 2009). Following Ridgwell 
and colleagues, their estimate-—using a larger area estimate and a
40 percent increase in reflectance-—corresponded to an upper limit of 
about 9 percent effectiveness. Their interpretation of Hamwey’s data 
corresponded to essentially the same effectiveness as Hamwey’s-—about 
16 percent. Thus the total effectiveness of reflective flora—cropland, 
open shrubland, grasslands, and savannah combined, using Lenton and 
Vaughan’s reinterpretations based on energy balance-—would be up to 
about 25 percent. 

Because crops are customarily replanted annually, no additional effort 
should be required to maintain their reflectivity (Ridgwell et al. 
2009). Hamwey provided no information on the effort required to 
maintain the reflectivity of open shrubland, grasslands, and savannah 
(Hamwey 2007). 

Increasing the reflectivity of settled areas could be about 4.3 
percent effective. Akbari, Menon, and Rosenfeld’s (2009) estimate for 
urban area equal to 1 percent of Earth’s land surface and a net 
increase for urban reflectivity by 10 percent corresponded to an 
effectiveness of only about 1.2 percent. However, Lenton and Vaughan 
(2009) suggested that the urban area Akbari, Menon, and Rosenfeld 
(2009) used, could have been 5.6 times overestimated, in which case 
increasing the reflectivity of urban areas would be only about 0.21 
percent effective. Hamwey’s (2007) estimate for doubling reflectivity 
for areas of human settlement (not including agricultural land) 
corresponded to an estimated overall effectiveness of about 4.6 
percent. Lenton and Vaughan’s correction to Hamwey’s estimate 
accounting for absorption in the atmosphere and an underestimate in 
solar radiation corresponded to an effectiveness of about 4.3 percent. 
Maintaining high reflectivity would be the sustainability issue for 
these technologies. 

3.2.4.3: Cost factors and potential consequences: 

The maintenance cost for reflective deserts that could potentially 
compensate for more than 57 percent of the doubling of the 
concentration of CO2 in the atmosphere could be about $3 trillion per 
year. Gaskill proposed to increase reflectivity of 10 trillion square 
meters of the deserts (Lenton and Vaughan 2009; Royal Society 2009). 
The Royal Society provided the following cost estimate for reflective 
deserts (Royal Society 2009): if the cost of reflective sheeting, with 
an allowance for routine replacement from damage, were somewhat 
similar to that of painting, it would be several trillion dollars per 
year. The Royal Society’s method would yield an annual maintenance 
cost for reflective deserts of about $3 trillion (Royal Society 2009). 
The estimates did not include design, fabrication, testing, 
acquisition, installation, or infrastructure costs. 

We found no cost estimates for increasing the reflectivity of flora in 
the peer-reviewed literature (Royal Society 2009). We found no cost 
estimates for increasing the reflectivity of areas of human settlement 
in the peer-reviewed literature. However, the estimated maintenance 
cost for urban areas that would compensate for 0.21 to 1.2 percent of 
the doubled concentration of CO2 in the atmosphere was from about $78 
billion to about $440 billion per year. The Royal Society (2009) made 
a rough estimate of the costs of painting urban surfaces and structures 
white using standard costs for domestic and industrial painting. 
Assuming repainting once every 10 years, it estimated combined paint 
and manpower costs on the order of $0.30 per square meter per year. 
The urban area Akbari, Menon, and Rosenfeld (2009) studied was 1 
percent of Earth’s land area-—that is, about 1.47 trillion square 
meters. Using the Royal Society’s (2009) cost estimation method, 
maintenance would cost about $440 billion per year. Lenton and Vaughan 
(2009) suggested that the global urban area might be only about 
260 billion square meters, in which case maintenance would cost 
about $78 billion per year. These estimates did not include design, 
fabrication, testing, acquisition, installation, or infrastructure. 

Desert reflectivity is regional. The Royal Society (2009) stated that 
as with other very localized SRM technologies, this approach could 
change large-scale patterns of atmospheric circulation, like the East 
African monsoon that brings rain to sub-Saharan Africa. The technology 
could be reversed by removing the reflective material. 

A 2009 modeling study by Ridgwell and colleagues indicated that 
increasing the reflectivity of crops by 20 percent would not create a 
significant effect on global average temperature but that reflective 
crops could have an appreciable cooling effect regionally. This study 
indicated that reflective crops could depress temperatures by more 
than 1 degree Celsius during summer months in a pattern broadly 
corresponding to the densest cropland coverage in the model. 

Hamwey’s 2007 investigation of increasing the reflectivity of open 
shrubland, grassland, and savannah used a radiative transfer model, 
and Lenton and Vaughn’s 2009 investigation of reflective crops and 
open shrubland, grassland, and savannah used an analytical approach 
based on energy balance considerations, so neither investigation can 
be used to evaluate climate consequences other than global average 
temperature. Hamwey did not discuss ecological issues associated with 
such a massive change to natural flora. Increasing the reflectivity of 
flora could reduce overall photosynthesis, which could reduce net 
carbon uptake by vegetation and crop yields. However, this is judged 
to be of relatively low risk, since photosynthesis tends to be light-
saturated during most of the growing season (Royal Society 2009). This 
technology could be reversed by replanting original flora. 

Since the analyses of reflective urban areas (Akbari et al. 2009; 
Lenton and Vaughan 2009) and human habitats (Lenton and Vaughan 2009; 
Hamwey 2007) were based on analytic estimates of radiative forcing, 
radiative transfer, or energy balance, their results cannot be used to 
evaluate climate consequences other than global average temperatures. 
However, reflective surfaces could reduce air-conditioning costs 
(Levinson and Akbari 2010). Effects would be reversible by returning 
reflective surfaces to their original condition. 

3.3: Status of knowledge and tools for understanding climate 
engineering: 

Gordon (2010, 7–8) identified 26 examples of areas of climate research 
that are important to understanding climate engineering and 8 examples 
of climate engineering research tools. The report described resources 
at several federal agencies that could help advance climate 
engineering research and gave examples of a number of their 
achievements in these areas (Gordon 2010, 8–37). 

Further efforts to improve scientific understanding related to climate 
engineering are under way, but reports from DOE, the National 
Aeronautics and Space Administration (NASA), National Institute of 
Standards and Technology (NIST), National Oceanic and Atmospheric 
Administration (NOAA), peer-reviewed scientific publications, and 
interviews with scientists indicate that the science is characterized 
by significant uncertainties. These gaps are related to the 
measurement of climate variables and models of the climate system that 
can simulate the effects of climate engineering on outcomes such as 
temperature or precipitation. The reports we reviewed described key 
limitations related to climate engineering science and three key 
challenges to improving them: (1) resolving uncertainties in 
scientific knowledge; (2) improving the coverage, continuity, and 
accuracy of observational networks used to measure essential climate 
mechanisms; and (3) developing greater high-performance computational 
resources and dedicating them to climate modeling. 

3.3.1: Better models would help in evaluating climate engineering 
proposals: 

Best practices in technology development recommend thoroughly testing 
new technologies before employing them in essential systems (GAO 
1999). Tests usually involve controlled experiments to understand how 
a technology being developed works and to assess its performance. 
However, large-scale field testing of climate engineering technologies 
is difficult (Gordon 2010, 3-4, 20, 27, and 32). For example, 
according to NIST scientists we interviewed, estimations of or 
assumptions about relevant chemical, physical, and optical properties 
that are acceptable for many common applications would introduce 
unacceptable risk in large-scale climate engineering experiments that 
could permanently alter the chemistry of the atmosphere. 

Complex climate models such as general circulation models (GCM) can be 
used to simulate the effects of large-scale climate engineering 
proposals and evaluate them without deploying them. However, the 
models are only as good as the data and the scientists’ understanding 
of how the climate system works (Meehl and Hibbard 2007; GAO 1995). 
Scientists attending the Aspen Global Change Institute’s 2006 session 
on Earth System Models said that gaps in climate models or inadequate 
data could affect the outcomes of numerical simulations designed to 
test climate engineering proposals (Meehl and Hibbard 2007). 

General circulation models of Earth’s climate evolved from short-term 
weather forecasting models first developed almost half a century ago 
(Slingo et al. 2009; McGuffie and Henderson-Sellers 2001). Advances in 
computing power and scientists’ understanding of the climate system 
have helped improve the models’ simulation capabilities (Slingo et al. 
2009), but according to a NOAA official these improvements are still 
not sophisticated enough to rely on for climate engineering. 
Atmosphere-ocean general circulation models (AOGCM) are today’s 
standard in climate models; they typically account for a number of 
factors that can influence the climate, such as oceans, land surface, 
and sea ice (Bader et al. 2008; Meehl and Hibbard 2007). Since 2000, 
AOGCM simulations have included aerosol effects, terrestrial 
processes, ocean mixing, and sea ice movement, but reports show that 
these models have important limitations with implications for 
simulations of the effects of climate engineering technologies. 

For example, simulations of aerosol-based SRM technologies require not 
only a thorough understanding of how aerosols behave in the atmosphere 
but also a computationally intensive representation of this behavior 
in a climate model. At present, aerosol treatment is not standardized 
across GCMs, and the models generate different results in terms of 
predicted temperature changes and precipitation patterns (Kravitz et 
al. 2011). Climate engineering researchers are beginning to 
standardize modeling scenarios that describe actions to manipulate the 
climate. This standardization would allow researchers to compare the 
robustness of the models’ responses to engineered inputs and to 
investigate how simplifying assumptions and structures used in the 
models can influence these outcomes (Kravitz et al. 2011). One 
scientist noted that climate chemistry models focusing on atmospheric 
processes can also contribute to scientific understanding of aerosols 
but can be computationally intensive. 

Earth systems models (ESM) representing the forefront in climate 
models aim to account for biological and chemical processes, such as 
the carbon cycle, that are not typically present in AOGCMs (Bader et 
al. 2008; Meehl and Hibbard 2007; Washington 2006). Climate models 
that included these additional processes could help scientists 
discover consequences of climate engineering proposals that are not 
predicted by the current generation of models (Meehl and Hibbard 
2007). For example, simulations of CDR-based proposals could be 
influenced by improving the representation of the carbon cycle in 
climate models (Bader et al. 2008). 

Scientists have identified several potential advancements related to 
ESMs that could improve their use in evaluating climate
engineering proposals: 

* scientific knowledge that would facilitate improvements in 
computational algorithms that represent physical, chemical, or
biological processes; 

* improvements to observational networks that measure essential 
climate mechanisms;[Footnote 47] and; 

* greater high-performance computing resources dedicated to climate
engineering-related science. 

3.3.2: Key advancements in scientific knowledge could help improve 
climate models: 

Although scientific knowledge of Earth’s physical, chemical, and 
biological processes has increased over time, it remains characterized 
by substantial gaps that can affect measures of climate sensitivity 
simulated by climate models (NRC 2010a; Bader et al. 2008; Solomon et 
al. 2007; Meehl and Hibbard 2007).[Footnote 48] Increased scientific 
knowledge about a number of environmental processes could improve 
scientific confidence in estimates of climate sensitivity. For 
example, Bader and colleagues (2008) highlighted the importance of 
improving representations of terrestrial, oceanic, and atmospheric 
carbon-feedback processes for more reliable estimates of future 
climate change. About half of all anthropogenic carbon emissions are 
sequestered in terrestrial or oceanic sinks whose mechanisms and 
capacities are not adequately revealed by observations (NRC 2007). 
Similarly, the relative magnitude of Earth’s energy reservoirs and the 
exchanges between them are not fully understood (Trenberth and Fasullo 
2010).[Footnote 49] Scientists’ limited understanding of how aerosols 
and clouds affect Earth’s energy budget and hydrological cycle is the 
most important source of uncertainty in climate models (NRC 2007). 
Aerosols may affect climate to the same degree as CO2 at current 
levels, but uncertainty about the effect of aerosols is about five 
times greater than the corresponding uncertainty about CO2 (NRC 2007). 
Experts at a NASA workshop reported that using climate models to 
simulate and evaluate aerosol-based SRM proposals to modify the 
climate is limited by the lack of models that explore how these 
aerosols would affect stratospheric ozone and the biosphere (Lane et 
al. 2007). 

3.3.3: Better observational networks: 

could help resolve uncertainties in climate engineering science
Observational sensing systems such as satellites and ground-based 
stations collect data that help scientists track climate trends and 
model climate mechanisms. Scientists have expressed several concerns 
about the coverage, continuity, and accuracy of observational networks 
that gather data related to climate mechanisms that are central to 
climate engineering technologies. 

Observational network abilities depend in part on where sensors are 
placed and the density of their distribution (OSTP 2010; Ohring 2007). 
Climate engineering scientists have expressed concern about the 
adequacy of observational networks in the atmosphere (Gordon 2010, 
23). For example, some scientists have criticized the sparse 
distribution and output of sensors in the upper atmosphere, where a 
number of processes have implications for CDR and SRM technologies 
(NRC 2010a). In particular, scientists from NOAA and Oak Ridge 
National Laboratory said that CO2 measurements from these sensors may 
be insufficient to permit conclusive statements about the effects of a 
given CDR technology. Upper atmosphere observations of the types of 
aerosols under consideration in some SRM proposals are also rare. 
Moreover, instruments that measure the optical properties of aerosols 
were recently eliminated from two satellites in the Joint Polar 
Satellite System (Gordon 2010, 15). 

Scientists have also expressed concern about the continuity of 
measurements by observational networks. Scientists have noted that 
deferring the implementation of adequate observational networks could 
miss opportunities to collect data on infrequent and unpredictable 
natural events, such as large volcanic eruptions, that could help 
scientists understand mechanisms related to climate engineering 
(Asilomar Scientific Organizing Committee 2010; Gordon 2010, 23). 
[Footnote 50] Scientists have also criticized the lack of redundancy 
in observational networks, which could create a gap in the measurement 
record if a single satellite or sensor were to fail (OSTP 2010; Ohring 
2007; NRC 2007). For example, NASA’s Glory Climate Satellite, intended 
to collect data on aerosols and solar energy in the atmosphere, 
recently failed to reach orbit at its launch. 

The continuity of measurements can also be affected if programs to 
collect data are not sustained over a long period of time. For 
example, federal budget cuts in the past decade have canceled, 
delayed, or degraded the collection of data from NASA’s Earth 
Observing System satellites, whose instruments and sensors measure 
essential climate variables such as Earth’s radiation budget, the 
global distribution of CO2, concentrations of methane and other 
greenhouse gases, air temperature and moisture content, cloud cover, 
and sea surface temperatures (NRC 2010a). Further, a 2007 NRC report 
predicted that the nation’s system of environmental satellites could 
decline dramatically and the number of operating sensors and 
instruments on NASA’s spacecraft would decrease by about 40 percent by 
2010 (NRC 2007). Scientists have also noted the difficulty of 
comparing continuous observations measured by different satellites or 
sensors without any overlap in their observation periods. 

Scientists have expressed concern about the accuracy of data collected 
from existing sensing devices (NRC 2007). For example, according to 
NIST scientists, unknown drifts in instrument data can cause 
measurements to show misleading evidence of change or false trends. 
Additionally, because most operational or weather satellite-based 
sensors share a common heritage, an artificial trend in a reading from 
one sensor is likely to exist in similar readings from other versions 
of the sensor, which would bias the measurements if the drift remained 
undetected. Moreover, large variations exist in solar radiation 
measurements even over small geographic areas and the causes are 
uncertain.[Footnote 51] 

Satellite programs developed to monitor and track local weather 
patterns might not be accurate or precise enough to measure long-term 
global climate change (Fraser et al. 2008). Climate-relevant signals 
are extremely small compared to fluctuations in weather and 
temperature observed daily, seasonally, or annually (Ohring 2007). For 
example, a decade’s anticipated average global temperature change is 
about 0.2 degrees Celsius, or about 1/50th of the temperature change 
that accompanies typical weather events. It is similarly difficult to 
accurately measure small variations in incoming or outgoing solar 
radiation on the order of 0.01 percent over decades without adequate 
optical instruments. Measuring radiation with accurate sensors is 
critical to advancing climate science: IPCC has reported that most 
climate change uncertainty derives from changes in Earth’s outgoing 
broadband radiation (J. T. Houghton et al. 2001). 

Within the federal government, steps are being taken to resolve some 
of these concerns, which might improve the ability to assess climate 
engineering technologies or proposals. Various agencies have proposed 
a long-term measurement strategy in Achieving Satellite Instrument 
Calibration for Climate Change (Ohring 2007). Satellite missions are 
being designed to help calibrate and reconcile some of the data 
received from existing climate measuring devices. For example, NASA’s 
Climate Absolute Radiance and Refractivity Observatory (CLARREO) 
mission is intended to yield a benchmark data record for detecting, 
projecting, and attributing change in the climate system.[Footnote 52] 

CLARREO would constitute a major effort to correct systematic biases 
and discontinuities in satellite-based climate measurements and to 
provide a robust climate reference point for future sensors that is 
traceable to accepted physics-based standards, called the International 
System of Units. Traceability ensures that environmental measurements 
are comparable, independent of the organization or country making them. 
The National Science Foundation (NSF) is sponsoring the construction 
of an integrated, Earth-based observation system called the National 
Ecological Observatory Network that will collect data across the 
United States on climate and land use changes and the effect of 
invasive species on natural resources and biodiversity.[Footnote 53] 
According to NSF, it will be the first observatory network that can 
both detect and forecast ecological change on a continental scale 
over multiple decades. Gordon notes that the network could inform 
research on several climate engineering technologies, including 
land-use management and biochar (Gordon 2010, 10). 

3.3.4: High-performance computing resources could help advance climate 
engineering science: 

Advances in computing could help scientists improve models used to 
simulate essential climate mechanisms and outcomes related to climate 
engineering. Limits in computational resources demand that existing 
climate models simplify certain processes essential to climate 
engineering instead of computing them numerically (Bader et al. 2008). 
Unlike short-term weather modelers, climate modelers have not moved to 
higher resolutions.[Footnote 54] Instead, they have used modern 
computational power to include additional physical components in the 
calculations.[Footnote 55] This is particularly important for climate 
engineering where, for example, stratospheric chemistry (to treat 
stratospheric aerosol processes) and the hydrological cycle (to treat 
cloud brightening) are important. A typical climate model represents 
Earth’s system as a grid of boxes anywhere from 100 to 300 kilometers 
on a side, which is larger than a typical cumulus cloud of about 1 
square kilometer (Slingo et al. 2009; Bader et al. 2008). Other 
significant climate features like oceanic eddies also act on a much 
smaller scale than the resolution current climate models support. 
Finer resolution that could be supported by increased computing power 
would help improve climate models’ representations of atmospheric and 
oceanic circulation (Bader et al. 2008). 

Computing advances could also facilitate the use of climate models to 
predict outcomes for geographic regions or across shorter time 
intervals. Scientists at NOAA’s Earth Systems Research Laboratory said 
that finer resolution could improve climate models’ predictions of 
regional changes that could be useful in evaluating climate 
engineering proposals. A climate scientist and a systems engineer also 
noted the potential value for climate engineering of greater precision 
in predicting regional changes in temperature and hydrological 
processes than existing models provide. They also observed that 
studies of climate engineering could benefit from simulations over 
shorter time intervals than are used in existing models.[Footnote 56] 
Officials at NOAA have predicted that at the historical rate of 
increase in computing power, supercomputers able to run cloud-
resolving ESMs with a grid size of a few kilometers should be 
available by 2025. 

Some federal agencies are already developing tools that would take 
advantage of anticipated, massively parallel, fine-grain computational 
architectures based on thousands of graphics processing units (GPU). 
[Footnote 57] For example, NOAA is encoding an ESM to operate with 
small amounts of local memory that will allow it to run on these GPUs. 
Additionally, DOE, NOAA, and the National Center for Atmospheric 
Research purchased supercomputers such as the Cray XT6 and Cray Baker 
computers with funding from the American Recovery and Reinvestment Act 
of 2009, and scientists at these agencies hope to develop an ESM by 
2025. 

As table 3.1 and table 3.2 show, climate engineering technologies are 
in early stages of development and have variable and uncertain cost 
factors, while uncertainty surrounds their potential effectiveness and 
potential consequences. Moreover, gaps in scientific knowledge, data, 
and computing resources challenge the models of climate mechanisms 
related to climate engineering. 

[End of section] 

4: Experts’ views of the future of climate engineering research: 

Because climate engineering technologies are currently immature, we 
explored future prospects for climate engineering research by 
obtaining a wide range of expert views. We obtained expert views on 
the future of research in three stages: (1) 6 experts met with our 
Chief Scientist to construct alternative future scenarios for climate 
engineering research, (2) 28 additional experts representing a wide 
range of professional disciplines and organizational affiliations 
shared their views about the future in response to the scenarios, and 
(3) some of the experts participating in the Meeting on Climate 
Engineering, which we convened with the help of NAS, volunteered their 
thoughts about the future.[Footnote 58] (In the appendices to this 
report, sections 8.3 and 8.4 present the scenarios and list the 
experts who served as scenario-builders, section 8.5 lists the experts 
who provided comments in response to the scenarios, and section 8.6 
lists the experts who participated in the meeting we convened with the 
assistance of NAS.) Altogether, 45 experts contributed views on the 
prospects for climate engineering research across the next 20 years. 
[Footnote 59] 

Briefly, we found the following: 

* The majority of the experts we consulted advocated starting 
significant climate engineering research now or in the very near 
future.[Footnote 60] Among the reasons they gave for starting research 
now is the anticipation that two decades or more of research will be 
needed to make substantial progress toward developing and evaluating 
climate engineering technologies with the potential to reduce emerging 
or future risks from climate change. Research advocates also 
envisioned safeguards to protect against potential adverse 
consequences or risks arising from the research. 

* A small number of those we consulted opposed starting such research, 
in part to prevent negative consequences either from the research or 
from deploying technologies developed from it.[Footnote 61] 

* The majority envisioned a federal effort that would direct and 
support research on climate engineering with specific features such as 
(1) an international focus, (2) engagement of both the public and 
decision-makers, and (3) foresight considerations to help anticipate 
emerging research developments and their opportunities and risks. 
[Footnote 62] 

* Experts identified many trends as potentially affecting research, 
including the pace of climate change, emissions-reduction 
developments, and scientific breakthroughs. 

4.1: A majority of experts called for research now: 

The majority of the experts we consulted about the future advocated 
starting significant research now or in the very near future, largely 
from concern about future climate change and the need to reduce its 
risks.[Footnote 63] In this report, we define “starting significant 
climate engineering research” as increasing research beyond that now 
being conducted. We had reported earlier that a relatively small 
amount of federal research is directly focused on climate engineering 
(GAO 2010a, 19).[Footnote 64] The advocates of research now—and some 
experts who did not indicate whether they advocated starting research 
now—anticipated that research will produce technologies or evaluative 
information or both that might help reduce risks associated with 
climate change or uninformed responses to it. Risks from climate 
change might include, for example, potential breakdowns in food and 
water supply chains (as climate change brings precipitation changes 
and rises in sea level), mass migration, and international conflict. 
More than half of these proponents anticipated that substantial 
research progress will take time-—perhaps two decades or more-—or 
stated that we cannot and should not wait for a crisis. Additionally, 
research advocates indicated that a cautious risk management approach 
could help reduce research-related risks. 

Those who advocated research now either did so urgently because they 
anticipated a definite need for climate engineering or viewed research 
as an insurance policy. For example, some warned against (1) losing 
the ability to prevent what they perceived as potentially irreversible 
changes or (2) being unprepared for a crisis. With respect to the 
latter, the report’s scenarios (1) describe how leaders who are 
unprepared while under heightened pressure to act quickly in a crisis 
might decide to deploy inadequately understood, risky technologies and 
(2) present the view that informed leaders might decide not to deploy 
risky technologies.[Footnote 65] 

Others who called for research now recognized the uncertainty of the 
future and viewed climate engineering research “as an insurance policy 
against the worst case scenarios” in the longer-term future. One said 
that the nation should: 

“make investments [in] ... fundamental research ... to be able to 
react quickly [if needed] ... [Spending to limit risk] on climate, 
terrorism, national defense, nonproliferation, should be viewed 
identically.”[Footnote 66] 

Overall, those calling for research now reflected foresight literature 
that warns against falling behind a potentially damaging trend, as 
illustrated in figure 4.1. Those advocating research now recognized 
important cautions, discussing two types of risk associated with 
research: (1) risks from conducting certain kinds of research (for 
example, large-scale field trials of potentially risky technologies) 
and (2) risks from using or misusing research results (for example, 
deploying risky technologies developed from the research).[Footnote 67] 
Various research advocates therefore suggested potentially 
complementary remedies such as: 

* managing risks (from research and using its results) with strategies 
like those outlined in box 4.1, which have been highlighted in the 
literature; 

* evaluating the risk of deploying specific technologies, in advance, 
which could lead to taking some risky technologies off the table; and; 

* setting international research limitations or guidelines.[Footnote 
68] 

Figure 4.1: Taking early action to avoid potentially damaging trends: 
Illustration from foresight literature: 

[Refer to PDF for image: line graph] 

The graph plots: 

Potentially Damaging Trends: 
Minimal; 
Reversible; 
Irreversible and very costly. 

against: 

Time Period: 
Current; 
Near-term future (somewhat uncertain); 
Longer-term future (very uncertain). 

The following are plotted: 
Current, potentially damaging trend; 
Projected, with early intervention; 
Projected, without intervention; 
Projected, with late intervention. 

Source: GAO adapted from Rejeski (2003). 

[End of figure] 

Box 4.1: Climate engineering research: Risk mitigation strategies from 
the literature: 

* The research community’s voluntary self-governance and development of
norms and best practice guidelines for open and safe research; 

* Required examinations of ethical, legal, and social implications in 
federally funded research projects; 

* Interventions that bring social scientists, ethicists, or trained 
risk assessors directly into laboratories to ensure early accounting 
for risks and social and ethical issues; 

* Application of an institutional review board concept to climate 
engineering research[A]; 

* Commissioned and independently conducted interdisciplinary risk 
assessments; 

* A multistage approach in which initial research (for example, 
computer modeling and laboratory studies) investigates risks before 
progressing to small-scale field studies, which in turn provide added 
information on risks before progressing to large-scale field 
experiments; 

* Study of risk trade-offs and analysis of options for reducing 
overall risks; 

* Developing norms for deployment decisions, facilitated by research 
activity. 

Source: Olson forthcoming; Morgan and Ricke 2010; Victor 2008; Graham 
and Wiener 1995. 

[A] Institutional review boards typically review research projects 
that use humans so as to protect their rights and welfare. However, 
the concept could be expanded to require an institutional review of 
research for field experiments that use Earth as a subject. IRB review 
requirements could be linked directly to federal research grants given 
for climate engineering. 

[End of Box 4.1] 

As we discussed above, research advocates suggested the study of 
climate engineering risks, and we earlier reported (GAO 2010a) that 
experts had told us that potentially “unintended consequences ... 
require further study.”[Footnote 69] 

Some research advocates qualified their positive view of climate 
engineering research with the proviso that emissions reduction efforts 
be continued. They warned that if the concentration of CO2 continues 
to rise into the long-term future, deploying SRM-—and increasing it 
over time to maintain acceptable temperatures—would lead to serious 
risk. That is, should extensive SRM deployments fail or be 
discontinued for any reason, “the bounce-back effect [of sudden 
warming] would be staggering.” 

Some research advocates also indicated that reducing emissions (and 
using apparently safer CDR technologies, such as direct air capture 
and sequestration, if needed to reduce build-up) would logically 
reduce the need for potentially risky SRM deployment. 

Finally, a research advocate who reviewed this report emphasized the 
need to consider the net effect of (1) the risks of climate change 
without climate engineering, (2) the potential reduction of climate 
risks through climate engineering, and (3) the introduction of 
possible new risks through climate engineering. 

4.2. Some experts opposed starting research: 

We noted above that a small number of the experts we consulted opposed 
starting significant research on climate engineering.[Footnote 70] 
Some thought that pursuing further research would, in the words of one 
commenter, open a Pandora’s box better left unopened. Opponents of 
research viewed climate engineering as technological hubris or as 
likely to be ineffective or not needed. Those who said that climate 
engineering is not needed either believed that climate change has been 
exaggerated or preferred other approaches, such as “building ecosystem 
and community resilience to respond to climate change,” “adopting more 
sustainable agricultural policies,” or “making [a] massive investment 
in energy efficiency.” However, the most strongly expressed opposition 
to climate engineering research concerned risks. One research opponent 
envisioned situations in which stratospheric aerosols would produce 
conflicts or catastrophic results in some parts of the world: 

“There are wars waged over the position and density of the clouds, 
rainfall patterns, ocean alkalinity, and volcanic eruptions as 
confusion prevails over what phenomena are natural and which are 
manmade. Different, often conflicting experiments are sponsored by 
different countries .... 

“... the precipitation patterns over large parts of Africa and Asia, 
which are already suffering from drought and food insecurity ... [are 
disturbed by the SRM deployment and eventually] hundreds of millions 
of people die because of crop failures and chaotic weather events .... 
[A] very small number of people control the climate levers, [and ...] 
global tensions rise.”[Footnote 71] 

The research opponents in our study did not envision varied strategies 
for managing, reducing, or avoiding risks from research or technologies 
developed from it. One opponent of research endorsed international 
moratoriums and said that: 

“It is illogical to assert that the best risk-avoidance strategy is to 
increase research. The best way to avoid responses that are extremely 
high risk is not to research them more; it is to make sure, through 
legally binding agreements, that they are prohibited.” 

While research advocates suggested evaluating technologies in advance 
of deployment, some opponents thought that, as one said, 

“the effects of human intervention are impossible to predict with a 
high degree of certainty. Any large-scale attempt to tame the climate 
system ... has a high probability of backfiring.” 

Finally, some research opponents were concerned about moral hazard—-
that is, the possibility that the results of climate engineering 
research would: 

“undermine the political will to reduce emissions.” 

Some research opponents feared that climate engineering would be 
substituted for, rather than used to complement, emissions reduction 
efforts. One research opponent suggested that: 
 
“Political leaders ... faced with the choice of politically difficult 
unilateral reductions in carbon emissions and the illusion of a techno-
fix, [will] go for the latter.” 

As we discussed in the previous section, some advocates warned of 
negative outcomes if climate engineering, particularly SRM, were 
pursued in the absence of emissions reduction. 

4.3. A majority of experts envisioned federal research with specific 
features: 

We reported earlier that the United States does not have a “
coordinated federal strategy for geoengineering, including guidance on 
how to define ... geoengineering activities or efforts to identify and 
track ... funding related to geoengineering” (GAO 2010a, 23). In that 
report, we recommended that: 

“the appropriate entities within the Executive Office of the President 
(EOP), such as the Office of Science and Technology Policy (OSTP), in 
consultation with relevant federal agencies, develop a clear, defined, 
and coordinated approach to geoengineering research in the context of 
a federal strategy to address climate change that (1) defines 
geoengineering for federal agencies; (2) leverages existing resources 
by having federal agencies collect information and coordinate federal 
research related to geoengineering in a transparent manner; and if the 
administration decides to establish a formal geoengineering research 
program, (3) sets clear research priorities to inform decision-making 
and future governance efforts.” (GAO 2010a, 39) 

OSTP neither agreed nor disagreed with our recommendation but provided 
technical and other comments. With respect to the context of a federal 
strategy to address climate change, we note that other approaches to 
addressing climate change include efforts to (1) reduce CO2 emissions 
and (2) adapt to climate change. 

In our work for this report, we found that experts who advocated 
starting significant research now generally also advocate or envision 
a federal research effort with specific features.[Footnote 72] That 
is, they envision federal research that would foster the development 
of technologies like CDR and SRM, rigorously evaluate related risks, 
and include specific features such as: 

* an international focus, 

* engagement of the public and national decision-makers, and, 

* incorporation of foresight considerations aimed at identifying new 
opportunities, anticipating new risks, and adapting research to 
emerging trends and developments. 

As outlined in this report’s overview of the technologies, knowledge 
is currently limited on proposed CDR and SRM technologies, and experts’ 
comments to us further suggested that planning might need to precede 
the first phase of any federal research effort. 

Some research advocates who envision a U.S. climate engineering 
research effort explained that other nations, the United Kingdom among 
them, are already studying these technologies or establishing 
programs.[Footnote 73] They said that the absence of a U.S. research 
effort could leave the United States “without a seat at the table,” 
unprepared to play a leading role, or unable to respond to other 
nations’ actions. One advocate of a U.S. research effort said that: 

“If it ever becomes necessary to deploy geoengineering techniques, 
doing so will be a momentous decision for humanity. The United States 
should be prepared to play a leading role in the decision, and it 
should be unthinkable that the decision could be made without 
substantial input from the U.S. scientific and technical community.” 

Others imagined a future in which individual nations would 
unilaterally engage in SRM; one motivation might be to resolve local 
or regional problems caused by climate change (Morgan and Ricke 2010). 
On one hand, such actions could have transboundary or global SRM 
effects, conceivably raising issues of national security, stimulating 
other nations to respond, or requiring a U.S. response. On the other 
hand, the risk of unilateral action might be reduced with cooperative 
international research that fostered trust and cooperation among 
nations on issues pertaining to climate engineering.[Footnote 74] 

For reasons such as these, many of the research advocates in our study 
suggested an international approach to federally sponsored research. 
[Footnote 75] Suggested activities included: 

* studying strategies for responding to situations arising from 
insufficient international cooperation in the use of climate 
engineering; 

*sponsoring or encouraging joint research with other nations 
(including developing or emerging industrial nations) because this 
might (1) help the United States keep pace with other nations’ 
research; (2) facilitate rigorous, transparent evaluation of new 
technologies that others develop; and (3) foster cooperation and 
consensus—-or an evolving set of norms about conducting research, 
which might, in turn, foster support for guidelines;[Footnote 76] and; 

* studying how the responsibilities of nations that deploy these 
technologies could impinge on others’ geopolitical equity, human 
rights, and justice—which would logically be most important for 
vulnerable or poor populations. 

Other international issues suggested for research included (1) 
studying how to define climate emergencies and achieve international 
agreement on responses to them and (2) exploring issues concerning 
military engagement in climate engineering research. 

The possibility of U.S. leadership in internationally focused research 
was suggested, as was cooperation: 

What the U.S. can do ... is to lead the process of framing the 
[climate engineering] issue as one requiring global collaboration and 
evidence-based decision-making processes that focus not only on macro 
results but also on fairness in distributional aspects of action 
versus inaction.” 

One expert said that in a substantial, internationally focused U.S. 
research effort, the United States could lead by example, emphasizing 
values such as transparency and attention to risk issues. 

The engagement of researchers with the public and with U.S. decision-
makers (and possibly international leaders) was another desirable 
focus for a federal research effort, according to research advocates. 
[Footnote 77] Their views included statements that: 

* engagement can foster shared learning across national leadership, 
the general public, and the research community; help ensure 
transparency; build shared norms; and bring an informed: 

“democratic process [to] ... decisions that ... broadly affect society;”
 
* engagement results might help frame research agendas to reflect the 
concerns and needs of the public and decision-makers; and; 

* information provided to the public and decision-makers might address 
(1) the systemic risks of the various climate engineering approaches, 
(2) trade-offs in pursuing alternative strategies, and (3) analyses of 
ethical, economic, legal, and social issues. 

A broad, multidisciplinary research agenda consistent with these views 
is discussed in the foresight scenarios we developed for this report. 

Our discussion of risks (earlier in this section) indicated that 
uninformed national or global leaders who were under heightened 
pressure in a perceived climate crisis might make hasty choices, 
whereas informed leaders might make a more measured response. 
Logically, the same might be true of the general public. That is, 
public engagement in advance of a crisis could help ensure that public 
concern about harm from technologies is addressed in advance (through 
research on benefits and risks), that research results are 
appropriately conveyed to the public, and that public expectations are 
consistent with likely real-world consequences. In sum, communication 
among researchers, the public, and decision-makers might help prepare 
the nation for a measured response to a future crisis. 

Some advocating a federal research effort also envisioned its 
incorporation of foresight activities designed to (1) anticipate 
emerging directions in developing climate engineering technologies and 
new or changing risks associated with such directions and (2) help 
research keep pace with other developing trends that could affect the 
research agenda and support.[Footnote 78] Some examples of foresight 
activities are communicating with other researchers in related areas, 
monitoring or surveying research, and using horizon scans and other 
futures methods that could help anticipate and track relevant 
developments and potential new risks.[Footnote 79] One research 
advocate also suggested exploring low-probability, high-impact events 
(described by Taleb 2007) with game theory or scenario planning. 

Other examples include iteratively monitoring a variety of 
developments and trends and, where appropriate, supporting studies 
that obtain better evidence on them. This could help guide decisions 
about forward directions (GAO 2008b, 67–68) and is compatible with 
other suggestions for adaptively managing climate engineering 
research. (With respect to the latter, experts in our study endorsed 
adaptive management to better achieve continuous improvement, based on 
(1) changing practices over time in response to experience and 
performance assessment and (2) learning how to intervene in a complex, 
imperfectly understood climate system.[Footnote 80]) 

Finally, the overall results of our communications with experts 
indicated uncertainty, or at least a diversity of views, on what 
technical research and evaluation will be needed for specific CDR and 
SRM technologies. Some experts noted the lack of any map or forward-
looking plan showing how climate engineering research might progress 
along various paths. We also found that experts expressed widely 
different views on: 

* the scope of whatever global climate engineering efforts might 
eventually be implemented or deployed and; 

* the level of effort or funding needed for research.[Footnote 81] 

For example, experts variously characterized the scope and scale of 
the deployment of stratospheric aerosols in terms of (1) operations 
that “rogue” actors might carry out unilaterally or, in contrast, (2) 
huge operations that might amount to “the largest engineering project 
in the history of people.” 

Some experts we consulted suggested that research funding might start 
with as little as a few million dollars. The scenarios developed for 
this report suggest that more effective research would have a 
considerably higher budget but they do not specify an amount. We 
reported earlier that 13 federal agencies had identified at least 52 
research activities relevant to climate engineering in fiscal years 
2009 and 2010 (GAO 2010a)—with funding of $1.9 million to investigate 
specific climate engineering approaches. Much larger amounts have 
funded activities related to conventional mitigation strategies or 
basic science that could be applied to improving scientific 
understanding of climate engineering.[Footnote 82] 

Because information about climate engineering and related research is 
limited, one of the experts who advocated federal research suggested 
that a federal effort begin with initial developmental work to 
delineate scale and cost. (Further research might then be planned and 
potential research costs estimated (GAO 2009b).) According to another 
expert, planning efforts would benefit from the development of an 
overall research strategy, including, for example, a: 
 
“multidisciplinary framework for integrated systems analysis ... and 
risk assessment tailored to designing and evaluating geoengineering 
technologies and their potential deployment as subscale experiments.” 

4.4. Some experts thought that uncertain trends might affect future 
research: 

When the experts we consulted envisioned research with a foresight 
component, they saw the following as relevant and potentially critical 
to track. First, signals of impending climate-related events would be 
relevant because these could potentially heighten the urgency and 
priority of the research. An example might be a collapse of ocean 
fisheries attributed to global warming and ocean acidification, with 
depletion of food supplies in vulnerable areas. Second, trends in 
policies for or new approaches to emissions reduction could affect 
prospects for CDR’s implementation. Our scenarios illustrate the view 
that establishing carbon constraints would encourage an anticipation 
of the use of CDR research, creating an incentive for research and 
innovation.83 Experts differed in assessing how CDR research might 
develop in the absence of significant carbon constraints. Some said it 
would be difficult to sustain research or deploy CDR technology 
without carbon constraints, while others disagreed, citing the 
possibility of deployment through a major public works program (Parson 
2006). Also relevant would be developments in sequestration related to 
advances in carbon capture and storage. 

Other potentially important areas to track are nanotechnology and 
synthetic biology breakthroughs (Rejeski 2010; Shetty et al. 2008); 
advances in these areas might bring new developments in climate 
engineering technologies. Examples include future “programmable plants”
that would sequester more carbon than natural plants and airborne 
microbes that would consume greenhouse gases. However, such 
developments might entail new risks. 

Future research breakthroughs might lead to or create new low-cost, 
low-carbon technologies and thus speed emissions reduction. One expert 
envisioned no-carbon energy sources like solar power costing less than 
carbon-based energy. Developments such as these could have important 
implications for the future role of CDR. 

Additionally, experts thought trends in public opinion on climate 
engineering research might affect support for research or specific 
projects. Monitoring trends in public opinion could be a key element 
of public engagement; for example, it might signal a need to study the 
safety implications of certain kinds of studies. 

Finally, experts (1) suggested links between future developments in 
climate engineering and possible international tensions or conflicts 
that might develop from economic issues, cultural changes, or 
demographic shifts and (2) indicated that low-probability, high-impact 
events might affect future research. They suggested examples of the 
latter, such as an SRM experiment’s coinciding with a natural volcanic 
eruption and producing unprecedented cooling; abrupt changes in ocean 
currents sharpening climate differentials; catastrophic alterations in 
weather patterns; geopolitical instability caused by widespread and 
prolonged famine in Africa or the Indian subcontinent attributable to 
global warming; a biotechnology disaster’s leading to strong public 
sentiment against technological interventions; the low-cost 
distribution of locally affordable technology’s reducing shipping-
related carbon emissions; or sudden cooling from an asteroid hit. 

If research planners believe that some low-probability events 
represent sufficient risks or opportunities, they might decide either 
on contingency planning or on hedging—that is, selecting a strategy 
that works reasonably well across a variety of outcomes, including 
certain low-probability, high-impact events (Popper et al. 2005). 

[End of section] 

5: Potential responses to climate engineering research: 

Because climate engineering technologies are potentially risky and 
could affect a large number of people and because experts have noted 
the importance of public engagement on this issue, we collected 
baseline measures of public opinion on climate engineering research 
among U.S. adults today. We analyzed survey responses from 1,006 U.S. 
adults 18 years old and older (representing the U.S. public) to address 
our third objective concerning the extent of awareness of geoengineering 
among the U.S. public and how the public views potential research into 
and implementation of geoengineering technologies.[Footnote 84] 

We found that the majority of the U.S. public is not familiar with 
geoengineering. Because public understanding of geoengineering is not 
well developed and public opinion in this area may be influenced by a 
variety of factors that may change over time, it is important to note 
that the results we report are not intended to predict future U.S. 
public views. Rather, our results provide valuable baseline 
information about current awareness of geoengineering and how the U.S. 
public might respond if it learned more about geoengineering. 

Because the public lacked familiarity with geoengineering, we provided 
survey respondents with basic information about geoengineering 
technologies before asking questions about them. 

Our key findings are that if the public were given the same type of 
information that we gave our survey respondents, then: 

* about 50–70 percent of the U.S. public across a range of demographic 
groups would be open to research on geoengineering.[Footnote 85] Many 
survey respondents expressed concern about the potential for harm from 
geoengineering technologies, but a majority also said they believe 
research should be done to determine whether these technologies are 
practical; 

* about half of the U.S. public would support developing 
geoengineering technologies. At the same time, about 75 percent would 
support reducing CO2 emissions and increasing reliance on solar and 
wind power; 

* about 65–75 percent of the U.S. public would support a great deal, a 
lot, or a moderate amount of involvement by the scientific community, 
a coalition of national governments, individual national governments, 
the general public, and private foundations and not-for-profit 
organizations in making decisions related to geoengineering. 

5.1. Unfamiliarity with geoengineering: 

Many people in the United States believe that Earth is warming but are 
not certain that this can be changed, while others do not believe that 
global warming is happening (Leiserowitz et al. 2010, 7; Maibach et 
al. 2009, 1 and 13; Nisbet and Myers 2007, 451). National surveys of 
U.S. public opinion have found broad public support for a variety of 
measures to increase energy efficiency, diversify the energy supply, and
reduce CO2 emissions (Pew 2010, 3; Bittle et al. 2009, 11), but 
geoengineering has not yet received widespread attention. 

Given the diversity of views on climate change, our survey asked 
respondents to consider their own views on climate change and how 
serious climate change might be and to indicate whether they thought 
any action should be taken. From the responses to this question, we 
estimate that about 40 percent of the U.S. public thinks that 
immediate action on climate change is necessary, about 35 percent 
thinks that action should be taken only after further research, about
10 percent thinks that no action should be taken, and about 15 percent 
is unsure. Among those who do not believe the climate is changing, we 
estimate that about 50 percent thinks no action should be taken and 
about 40 percent thinks that action should be taken only after further 
research. In other words, members of the U.S. public who do not 
believe the climate is changing do not necessarily oppose research on 
climate change.[Footnote 86] 

To ensure that our survey respondents had a basic understanding of 
geoengineering, we gave them a brief definition of geoengineering and 
examples of CDR and SRM technologies before we asked them questions 
about geoengineering. The information we gave them was similar in 
amount and type to information they might receive in the nightly news 
or in a short news article. 

Immediately after we defined geoengineering for our respondents and 
gave them examples of CDR and SRM technologies, the survey asked them 
whether they had ever heard or read anything about geoengineering 
technologies before they began the survey. From the results, we 
estimate that if provided with information about geoengineering 
similar to that given our survey respondents, about 65 percent of the 
U.S. adult public would not have recalled hearing or reading anything 
about geoengineering technologies at the time of our survey. The 
results of our survey pretest interviews, which included follow-up 
questions, indicated that some members of the public recall reading or 
hearing about technology proposals such as sequestration of carbon in 
the ocean or other geoengineering-type technologies in science and 
technology literature. 

5.2. Concern about harm and openness to research: 

As identified above, climate engineering includes a number of 
technologies, and different technologies may have different risks and 
benefits. To assess whether information about the potential for harm 
from different technologies affects public reaction to climate 
engineering, we decided to conduct a split-ballot survey in which we 
gave half the sample information about technologies that had been 
identified as relatively safe and the other half information about 
technologies that had been identified as less safe.[Footnote 87] 

This allowed us to examine whether receiving information about the 
less safe or the more safe technologies is associated with greater 
concern about harm from geoengineering. It also allowed us to assess 
whether public opinion on research and decision making depends on the 
information members of the public are given about experts’ assessments 
of a technology’s relative safety. 

We differentiated technologies by the experts’ assessments of safety 
as described in the Royal Society report (Royal Society 2009, 6). The 
two relatively safe technologies in our survey were (1) increasing 
reflection from Earth’s surface (by painting roofs, roads, and 
pavement white, for example) and (2) capturing CO2 from the air (in 
the information we gave the respondents, we also called this CO2 air 
capture and capturing CO2 from the air). The two less safe technologies 
were (1) putting sulfates, or tiny mirror-like particles, into the 
stratosphere and (2) seeding large ocean areas with fertilizer. Table 
5.1 shows the information the respondents received about technology by 
the ballot group they were assigned to—-506 respondents received 
information about increasing reflection from Earth’s surface and CO2 
air capture, and 500 received information about stratospheric sulfates 
and ocean fertilization. 

We randomly assigned survey respondents to receive information about 
the relatively safe and the less safe technologies. At the outset of 
receiving the information about geoengineering, survey respondents 
were told that: 

“Some scientists believe it might be possible to deliberately change 
Earth’s temperature and cool down the planet by changing some of the 
things that seem to be causing global warming. Using technologies to 
do this is known as ‘geoengineering.’ 

“There are two different types of geoengineering. The first type 
involves reflecting some of the light and heat of the sun’s radiation 
back into space. The second involves reducing the level of carbon 
dioxide in the atmosphere.” 

The respondents were not aware that the survey had two different sets 
of examples of geoengineering. All the survey questions were identical. 

Table 5.1 Geoengineering types and examples given to survey 
respondents: 

Example: Reflecting back into space some light and heat from the Sun’s 
radiation; 
Technology type: Relatively safe (506 respondents): Increasing 
reflection from the surface of Earth: 
Increasing reflection from the surface of Earth involves lightening 
and brightening the surface of the earth, to reflect some of the 
sunlight back into space. By reflecting sunlight into space, the 
temperature would be reduced. Increasing reflection from the surface 
of Earth could involve painting roofs, roads, and pavement. Although 
this should reduce the temperature at least some, there are doubts 
whether reflecting the surface of Earth could have a substantial 
effect on global temperatures. Unlike some geoengineering techniques, 
however, there is little risk of negative consequences. So the 
technique of reflecting the surface of Earth is not very effective, 
but it is safe.
Technology type: Less safe: (500 respondents): Putting sulfates into 
the stratosphere: 
Putting sulfates, which are tiny mirror-like particles, into the 
stratosphere. This would re-create what happens when large volcanoes 
erupt and shoot sulfates high into the atmosphere. The sulfates 
circulate in the stratosphere and reflect some sunlight before it 
reaches Earth. Research has shown that this technique would probably 
be very effective at reducing the global temperature. The extent and 
type of consequences from stratospheric sulfates is unknown, however. 
For example, there could be increased damage to the ozone layer or 
altered rainfall patterns around the world. So the technique of 
stratospheric sulfates is likely to be very effective, but there is 
also risk of serious negative consequences. 

Example: Reducing carbon dioxide in the atmosphere; 
Technology type: Relatively safe (506 respondents): carbon dioxide
from the air: 
CO2 air capture would chemically remove CO2 directly from the air. The 
CO2 could be turned into a liquid and piped underground for storage in 
geologic structures. This technique directly treats the cause of 
climate change—greenhouse gases—and research has shown that CO2 air 
capture would be very effective. Unlike some geoengineering 
techniques, it would not directly affect complex natural systems and 
is believed to be safe. So the technique of CO2 air capture is likely 
to be very effective, and it is safe.
Technology type: Less safe: (500 respondents): Seeding large ocean areas
with fertilizer: 
Ocean fertilization involves adding nutrients such as iron to some 
areas of the open ocean where they are in short supply. This promotes 
the growth of small plants called phytoplankton, and as the plants 
grow, they soak up CO2 from the atmosphere. This technique directly 
treats the cause of climate change—greenhouse gas such as CO2. It is 
not yet known how much carbon would be removed for longer than a few 
years; we need to learn more about the effectiveness of ocean 
fertilization. The extent and type of consequences from fertilizing 
the oceans are also largely unknown. For example, there may be harmful 
side effects if ocean fertilization were attempted on a large scale. 
So the technique of ocean fertilization may not be very effective and 
there is also the risk of serious negative consequences. 

Source: GAO. 

Note: The information provided to respondents was based on a report 
from the Royal Society (Royal Society 2009). 

[End of table] 

The survey results indicated that some 50 percent or more of both 
survey ballot groups were somewhat to extremely concerned that 
geoengineering could be harmful. More specifically, we estimate that: 

* about 50 percent of the U.S. adult public would be somewhat to 
extremely concerned that geoengineering technologies could be harmful 
if they were given information similar to what we gave the respondents 
about relatively safe technologies (increasing reflection from Earth’s 
surface and capturing CO2 from the air) and; 

* about 75 percent of the public would be somewhat to extremely 
concerned that geoengineering technologies could be harmful if given 
information similar to what we gave respondents about less safe 
technologies (stratospheric sulfates and ocean fertilization). 

These results suggest that many people would be concerned about the 
safety of even technologies that experts have identified as relatively 
safe. For technologies experts deemed less safe, a substantial 
majority would express concern. 

Despite these differences in respondents’ concerns, they did not 
differ greatly in their responses to other questions about 
geoengineering research and decision making. Consequently, we report 
the results from all other survey questions for all survey respondents 
combined. 

In addition to the issue of the technologies’ harm, the survey asked 
respondents how optimistic they were that geoengineering technologies 
could be beneficial. From the results, we estimate that about 45 
percent of the public would be somewhat to extremely optimistic, about 
40 percent would be slightly to not at all optimistic, and about 15 
percent would be unsure whether geoengineering technologies could be 
beneficial. As reflected in responses to an open-ended question in our 
survey, public optimism about geoengineering is likely to be tempered 
by concern that the technologies’ effects are not fully known. As one 
survey respondent put it: 

“Since the outcome is uncertain, more research needs to be done to 
find out how much of any one thing is enough or too much.” 

Given that research may be seen as a way to assess whether specific 
technologies might work and to identify harmful consequences, we used 
the survey to identify a baseline estimate of support for research on 
geoengineering among the U.S. public. From the results, we estimate 
that about 65 percent of the public, exposed to the same type of 
information as in our survey, would say they believe that research 
should be done to determine whether geoengineering technologies that 
deliberately modify the climate are practical. Further, respondents 
who received information about less safe technologies were just as 
likely to support research to determine whether geoengineering is 
practical as were respondents who received information about safer 
technologies; moreover, about 60 percent of those who said they were 
extremely concerned that geoengineering could be harmful indicated 
that research should be done. 

The survey respondents’ comments in response to an open-ended question 
in our survey illustrate that research and small-scale testing are 
seen as ways to determine whether technologies can be safely and 
effectively deployed. In other words, respondents identified research 
and small-scale testing as ways to assess the potential for harm from 
climate engineering technologies and to allow for more informed 
decisions about their use. 

The survey results also indicate that while approximately 65 percent 
of the public overall would support research on geoengineering, about 
half or more of the U.S. public across a range of demographic and 
political groups, including age, gender, race, ethnicity, education, 
and partisanship, would say that research should be done to determine 
whether geoengineering technologies are practical. In other words, 
support for research on geoengineering would not be limited to 
specific demographic groups. 

To explore potential public support for government-sponsored research 
on geoengineering, our survey also asked respondents two separate 
questions about whether they would support or oppose the U.S. government
’s paying for research on CDR or SRM technologies. The responses to 
these questions suggest that if public information were similar to 
that in our survey, about half the public would support the U.S. 
government’s paying for research on CDR technologies and about 45 
percent would support its paying for research on SRM technologies. 

As we remarked previously, public understanding of geoengineering is 
not well developed and our survey results do not necessarily predict 
future views. Furthermore, we did not ask respondents to consider the 
trade-offs between federal financing of geoengineering research and 
other possible spending priorities, including tax cuts or deficit 
reduction. We also did not ask respondents whether they supported 
private companies’ or other entities’ paying for research on climate 
engineering. Support for the government’s paying for research on 
geoengineering technologies could have been less or more had we asked 
respondents to choose alternative policy options or alternative 
funding sources. Research funded by a corporation or foreign 
government, for example, might yield different public support. 

5.3. Views on climate engineering in the context of climate and energy 
policy: 

National surveys of U.S. public opinion have found broad public 
support for a variety of measures to increase energy efficiency and 
diversify the energy supply (Pew 2010, 3; Bittle et al. 2009, 11). To 
place the public’s view of climate engineering in the broader context 
of public opinion on climate and energy policy, we asked survey 
questions about reducing CO2 emissions by increasing reliance on 
noncarbon-based energy sources and other methods in addition to 
climate engineering. From the results, we estimate that about three-
quarters of the public support (strongly support or somewhat support) 
developing more fuel-efficient cars, power plants, and other such 
technologies; encouraging businesses to reduce their CO2 emissions; 
and relying more on wind and solar power (figure 5.1). About 65 
percent of the public strongly or somewhat supports actions to 
encourage people to reduce CO2 emissions–for example, by driving less 
or renovating their homes. At the same time, our results indicate that 
if the public were given the same type of information as in our 
survey, about half would strongly or somewhat support developing 
geoengineering technologies. About 45 percent strongly or somewhat 
support relying more on nuclear power. 

Figure 5.1: U.S. public support for actions on climate and energy, 
August 2010: 

[Refer to PDF for image: survey results] 

Estimated percentages of support or opposition are rounded to the 
nearest 5%. 

Survey question: How much, if at all, would you support or oppose each 
of the following actions? 

1. Developing more fuel-efficient cars, power plants, and 
manufacturing processes to reduce carbon dioxide emissions: 
Strongly support: 55%; 
Somewhat support: 25%; 
Neither support nor oppose: 10%; 
Somewhat oppose: less than 5%; 
Strongly oppose: less than 5%; 
Don’t know: 10%. 

2. Relying more on solar power: 
Strongly support: 50%; 
Somewhat support: 30%; 
Neither support nor oppose: 10%; 
Somewhat oppose: less than 5%; 
Strongly oppose: less than 5%; 
Don’t know: 10%. 

3. Encouraging businesses to reduce their carbon dioxide emissions: 
Strongly support: 50%; 
Somewhat support: 30%; 
Neither support nor oppose: 10%; 
Somewhat oppose: less than 5%; 
Strongly oppose: less than 5%; 
Don’t know: 10%. 

4. Relying more on wind power: 
Strongly support: 50%; 
Somewhat support: 30%; 
Neither support nor oppose: 10%; 
Somewhat oppose: less than 5%; 
Strongly oppose: less than 5%; 
Don’t know: 10%. 

5. Encouraging people to drive less, renovate their houses, and take 
other actions to reduce their carbon dioxide emissions: 
Strongly support: 40%; 
Somewhat support: 30%; 
Neither support nor oppose: 20%; 
Somewhat oppose: less than 5%; 
Strongly oppose: less than 5%; 
Don’t know: 10%. 

6. Developing geoengineering technologies that could cool the climate 
or absorb carbon dioxide from the atmosphere: 
Strongly support: 20%; 
Somewhat support: 30%; 
Neither support nor oppose: 25%; 
Somewhat oppose: 5%; 
Strongly oppose: 10%; 
Don’t know: 10%. 

7. Relying more on nuclear power: 
Strongly support: 15%; 
Somewhat support: 30%; 
Neither support nor oppose: 20%; 
Somewhat oppose: 10%; 
Strongly oppose: 10%; 
Don’t know: 10%. 

Source: GAO. 

Note: Estimates have 95 percent confidence intervals of within plus or 
minus 4 percentage points. 

[End of figure] 

As the Royal Society reported, concern has been raised that 
geoengineering proposals could reduce public support for mitigating 
the effects of CO2 emissions and could divert resources from 
adaptation (Royal Society 2009). This is referred to as the “moral 
hazard” problem. Given low public awareness of geoengineering, it is 
difficult to determine with any confidence whether the U.S. public 
would reduce support for mitigation as it learned more about 
geoengineering or how concerned the public would be about this moral 
hazard. Our survey results suggest that if the public were given the 
same type of information about geoengineering as our survey 
respondents, it might support a range of approaches to climate and 
energy policy, including climate engineering, rather than viewing 
different approaches as trade-offs. 

As with the results of qualitative research that found U.K. public 
support for combining geoengineering with mitigation efforts (Ipsos 
MORI 2010, 1–2), we found that at least some of the U.S. public views 
geoengineering as an additional method of addressing climate change 
rather than as an alternative to mitigation and adaptation. In open-
ended comments, for example, some respondents expressed support for 
using other recognizable means to address climate change, such as 
reducing CO2 emissions, and using geoengineering as a last resort. 

5.4. Support for national and international cooperation on 
geoengineering: 

To obtain baseline information on U.S. public views on the extent to 
which different groups should be involved in deciding to use a 
geoengineering technology, our survey asked respondents how much 
involvement different public and private sector groups should have in 
making these decisions. From the results of our survey, we estimate 
that if the public were given the same type of information as in our 
survey, a total of about 75 percent would support a great deal, a lot, 
or a moderate amount of involvement by the scientific community in 
making decisions related to geoengineering (figure 5.2). At the same 
time, a total of about 70 percent would support a great deal, a lot, 
or a moderate amount of involvement by a coalition of national 
governments; about 65 percent would support this level of involvement 
by individual national governments, the general public, and private 
foundations and not-for-profit organizations; and about half would 
support this level of involvement by private companies. 

Figure 5.2: U.S. public views on who should decide geoengineering 
technology’s use, August 2010: 

[Refer to PDF for image: survey results] 

Estimated percentages of support or opposition are rounded to the 
nearest 5%. 

Survey question: How much, if any, involvement in decisions to 
actually use a geoengineering technology on a broad scale should each 
of the following groups have? 

1. The scientific community (for example, universities): 
A great deal: 30%; 
A lot: 30%; 
A moderate amount: 20%; 
A little: 10%; 
None: 5%; 
Don’t know: 10%. 

2. A coalition of national governments: 
A great deal: 25%; 
A lot: 20%; 
A moderate amount: 25%; 
A little: 10%; 
None: 10%; 
Don’t know: 10%. 

3. Individual national governments: 
A great deal: 20%; 
A lot: 25%; 
A moderate amount: 25%; 
A little: 10%; 
None: 10%; 
Don’t know: 10%. 

4. The general public: 
A great deal: 20%; 
A lot: 20%; 
A moderate amount: 30%; 
A little: 15%; 
None: 10%; 
Don’t know: 10%. 

5. Private foundations and not-for-profit organizations: 
A great deal: 15%; 
A lot: 20%; 
A moderate amount: 30%; 
A little: 15%; 
None: 10%; 
Don’t know: 10%. 

6. Private, for-profit companies: 
A great deal: 10%; 
A lot: 15%; 
A moderate amount: 30%; 
A little: 20%; 
None: 20%; 
Don’t know: 10%. 

Source: GAO. 

Note: Estimates have 95 percent confidence intervals of within plus or 
minus 4 percentage points. 

[End of figure] 

To provide additional insight into the U.S. public’s initial views on 
actions related to geoengineering, we asked survey respondents whether 
they supported or opposed the U.S. government’s coordinating more 
closely with other countries on geoengineering issues. We estimate 
that about 55 percent of the U.S. public would support the government’
s coordinating more closely with other countries on geoengineering 
issues, about 15 percent would oppose closer coordination, and about
30 percent would be unsure. Overall, the findings from our survey 
suggest that if the public were given similar information about 
geoengineering, it would be open to the involvement of multiple 
national and international groups. In addition to expressing support 
for involvement by a range of groups in response to closed-ended 
questions, survey respondents noted the importance of involving the 
scientific community, governments, the public, and the private sector 
in making decisions about geoengineering in their answers to open-
ended questions. In the words of one respondent, “national governments, 
along with the scientific community, should determine under what 
circumstances it would be okay to actually use geoengineering 
technologies.” 

[End of section] 

6: Conclusions: 

In this technology assessment, we have evaluated climate engineering 
technologies that could be part of a portfolio of climate policy 
options, along with mitigation and adaptation. We found that the 
technologies we reviewed are all in early stages of development. It is 
likely that significant improvements in climate engineering technology 
and related information will take decades of research because (1) 
today’s technologies are not mature and (2) data collection and modeling 
capabilities related to climate engineering research are marked by 
important gaps. Experts have warned that a delay in starting 
significant climate engineering research could mean falling behind in 
our capacity to address a potentially damaging climate trend. We have 
previously reported that the United States does not have a coordinated 
strategy for climate engineering research. 

We cannot ignore the possibility of new risks from either climate 
engineering research or its use or misuse. We found in our survey of 
U.S. adults that a majority would be open to climate engineering 
research but expressed concern about possible harm. Likewise, experts 
who advocate research emphasized that conducting significant climate 
engineering research and using the results could bring new risks, such 
as the possibility of international conflict arising from one nation’s 
unilaterally deploying climate engineering technologies that adversely 
affect other nations. Additionally, future technological developments 
may bring new and currently unknown risks. 

Experts we consulted suggested facilitating climate engineering 
researchers’ interactions with the U.S. public, national decision-
makers, and the international research community. They also said that 
international research could (1) help ensure that the nation is aware 
of and keeps pace with others’ research and (2) give the United States 
an opportunity to lead by example by emphasizing transparency in and 
risk management for the research. Foresight efforts concerning 
emerging trends and technological developments could help the nation 
better anticipate future risks and opportunities. 

[End of section] 

7: Experts’ review of a draft of this report: 

The fifteen experts listed in section 8.7 reviewed a draft of this 
report, at our request, and submitted comments to us. In this section, 
we summarize how we addressed the technical and other comments 
requiring a response. We also received a number of positive comments 
that do not require a response. 

7.1. Our framing of the topic: 

Some comments pertained to the presentation of anthropogenic climate 
change or climate policy in the introduction to this report. These 
comments ranged from objecting to the presentation of alternative 
views of climate change to suggesting that we highlight scientific 
consensus on anthropogenic climate change. We retained information on 
the range of views as key introductory content but added a 
clarification acknowledging the endorsement of IPCC’s view by numerous 
scientific bodies. We also added a statement linking the large 
consensus among authoritative scientific bodies to the sense of 
urgency that has contributed to discussions of engineering the 
climate. In response to other comments on the need to emphasize the 
potential role of climate engineering as a complement to mitigation 
and adaptation, we highlighted a GAO recommendation that the federal 
government develop a coordinated approach to geoengineering research 
in the context of a federal strategy to address climate change (GAO 
2010a). We also incorporated suggestions to balance the risks 
introduced by engineering the climate against the risks of climate 
change without climate engineering. 

7.2. Our assessment of technologies: 

Several comments surrounded the scope of the criteria we used to 
assess climate engineering technologies. Many of these comments 
concerned the appropriateness of TRLs to measure the readiness of soft 
climate engineering technologies for deployment, as opposed to devices 
or hard technologies. Given these comments, we discussed a key 
limitation of TRLs—that is, their sensitivity to certain criteria, 
such as the definition of a system concept or concrete plan. 
Developing an alternative way to measure the maturity of technologies 
was beyond our scope. In response to comments on the other key 
measurements, we revised the draft to emphasize the potential 
effectiveness and potential consequences of the technologies we 
assessed. To address concerns about the precision of cost estimates 
from the scientific literature, we focused on cost factors, or 
resources required to develop or deploy climate engineering 
technologies. Finally, we revised tables 3.1 and 3.2 to reflect these 
amplifications and clarifications. 

7.3. Our assessment of knowledge and tools for understanding climate 
engineering: 

To incorporate comments on the status of knowledge and tools for 
understanding climate engineering, we reemphasized our focus on the 
value of research to help improve climate science, observational 
systems, or computing power. We replaced generalizations with examples 
of areas that scientists have targeted for improvements, and we 
strengthened our citations. We added climate chemistry models to our 
taxonomy of existing climate models. We updated some examples, such as 
NASA’s CLARREO mission. We also accepted editorial comments clarifying 
certain ideas. For example, we characterized scientists’ concerns 
about the reliability of observational networks in terms of the 
continuity of the observational record, and we revised the text to 
highlight the potential value of developing high-performance computing 
resources that could be dedicated to resolving uncertainties about 
regional climate variables. Although some comments noted that various 
observations could apply to other areas of climate science, we 
considered these comments to be beyond the scope of our report. 
Comments on decision-making under uncertainty were also beyond our 
scope. 

7.4. Our foresight and survey methodologies: 

Comments on foresight and survey methodologies centered on the 
rationale for the content of the events described in the scenarios and 
the survey questions. In the foresight section of this report, the 
experts’ views on the future are not based primarily on events 
described in the scenarios. Rather, the scenarios led a wide range of 
experts to share their views on the future of climate engineering 
research over the next 20 years (section 8.1.2), thus allowing a broad 
thematic discussion of these views, which sometimes differed sharply. 
The scenarios reflect the views of the experts who helped build them, 
but our overall foresight approach gave considerable latitude to the 
expression of views by all experts we consulted. Additionally, we 
included experts with many different kinds of expertise and varied 
views on climate engineering. For these reasons, we are confident 
that our overall results would have been similar had the scenarios 
differed or been produced on the basis of an explicit underlying rationale. 

Objectives for the survey included developing baseline information on 
public awareness of climate engineering technologies, views about 
research on them, and opinions on who should be involved in decisions 
related to climate engineering. Our focus groups and pretest 
interviews indicated that members of the public were unlikely to have 
either detailed knowledge or established opinions about climate 
engineering and that public views on climate engineering depended on 
the technology. Therefore, we developed and pretested (1) a basic 
definition of climate engineering with examples of different 
technologies, in both audio-visual and written formats so respondents 
could choose between the two, and (2) basic survey questions about
each respondent’s awareness of and views on research and groups that 
should be involved in decisions. Our pretest results led us to believe 
that the respondents understood the basic questions and that these 
were unbiased and provided the baseline information we needed to meet 
our objectives. 

We did not incorporate other suggestions that were beyond the scope of 
our report. It was, for example, as much beyond our scope to develop a 
detailed strategy for deploying climate engineering as to compare a 
climate-engineered world with one lacking any deliberate climate 
intervention. 

[End of section] 

8: Appendices: 

8.1. Objectives, scope, and methodology: 

In this appendix, we describe the several targeted, coordinated 
methods we used to report on: 

* the current state of climate engineering technology, 

* experts’ views of the future of climate engineering research, and, 

* public perceptions of climate engineering. 

In addition to the separate methods we used to address each objective, 
with the assistance of the National Academy of Sciences (NAS) we 
convened a meeting of scientists, engineers, and other experts that we 
called the Meeting on Climate Engineering. Because climate engineering 
is complex, NAS selected, with our assistance, a diverse and balanced 
group of experts on climate engineering, climate science, measurement 
sciences, foresight studies, emerging technologies, research 
strategies, and the international, public opinion, and public 
engagement dimensions of climate engineering. Experts participating in 
our Meeting on Climate Engineering are listed in section 8.6. 

Before meeting in Washington, D.C., on October 6–7, 2010, the 
participants were provided with a written summary of our progress on 
this technology assessment. We explained to the participants that the 
summary was a working document showing what we had developed up to 
that point and that it did not fully describe our methodology. 

The participants were organized into subgroups to focus on the major 
topics of our technology assessment, including carbon dioxide removal 
(CDR) technologies, solar radiation management (SRM) technologies, the 
future of climate engineering, and public perceptions. The 
participants in each subgroup presented a 5-minute summary of their 
views on our preliminary findings, and then the entire group discussed 
the feedback. The meeting ended with general reactions to and advice 
and suggestions on our preliminary findings. 

The participants’ comments led us to review additional literature and 
unpublished studies that they suggested. Following the meeting, we 
also contacted the participants in person or by telephone or e-mail to 
clarify and expand what we had heard. We used what we learned from 
this meeting of experts to update, clarify, and correct where 
appropriate our information on the current state of climate 
engineering technology, expert views of the future of climate 
engineering research, and public perceptions of climate engineering. 
We incorporated in our draft report the lessons we learned from the 
meeting to give it greater accuracy and contextual sophistication. 

8.1.1. Our method for assessing the state of climate engineering 
technology: 

To determine the current state of the science and technology of 
climate engineering, we reviewed a broad range of scientific and 
engineering literature. We started with the literature the Royal 
Society report referenced (Royal Society 2009, 63–68), and then we 
reviewed the literature we found in NAS, National Research Council 
(NRC), and U.S. government reports on climate change. We identified 
other literature from scientific and climate-related organizations 
such as the National Aeronautics and Space Administration (NASA) and 
National Oceanic and Atmospheric Administration (NOAA), and we 
reviewed proceedings from conferences such as the 2010 Asilomar 
International Conference on Climate Intervention Technologies 
(Asilomar Scientific Organizing Committee 2010). We revisited the 
report on climate engineering that we issued in September 2010 (GAO 
2010a), which is complementary to this report. We reviewed relevant 
congressional testimony. We sought additional literature from the 
experts we spoke with. 

We identified experts on climate engineering and proponents of 
specific climate change technologies from our review of the literature 
and conference proceedings. To ensure balance across the views and 
information we obtained, we interviewed a broad range of experts and 
officials working in climate science research and climate engineering 
whose track records had been proven through their peer-reviewed 
publications and presentations at conferences. We interviewed these 
experts to seek information that was not in their published work. We 
interviewed scientists, engineers, and knowledgeable officials with 
the Department of Energy’s (DOE) National Energy Technology Laboratory 
(NETL), Lawrence Livermore National Laboratory, and Pacific Northwest 
National Laboratory (PNNL) and during site visits conducted at: 

* the National Center for Atmospheric Research, 

* the National Oceanographic and Atmospheric Administration’s Earth 
System Research Laboratory, 

* the National Institute of Standards and Technology, 

* Oak Ridge National Laboratory, 

* American Electric Power’s Mountaineer Power Plant in West Virginia, 

* the Institute for Advanced Study at Princeton University, 

* the Marine Biological Laboratory at Woods Hole, 

* Scripps Institution of Oceanography, and, 

* Woods Hole Oceanographic Institution. 

We interviewed selected attendees at the 2010 Asilomar International 
Conference on Climate Intervention Technologies. 

We reviewed records of earlier interviews we had conducted on topics 
relevant to this technology assessment. We analyzed interviews with 
high-level private-sector officials from: 

* Alstom, which develops carbon capture technology and equipment; 

* Dow, which conducts research on and development and manufacture of 
solvents or sorbents needed for CO2 capture; and; 

* Schlumberger Carbon Services, which engages in geological mapping 
and the characterization of subterranean structures for storing CO2. 

We interviewed experts at academic institutions such as Columbia 
University, the Massachusetts Institute of Technology, and Stanford 
University. 

Because climate science and climate engineering are interdisciplinary 
and extremely complex, with cross-cutting issues that may be beyond 
any one expert’s realm, we synthesized information from an array of 
experts with diverse views on these subjects. We did not try to 
interview an equal number with alternative perspectives on all issues 
or technologies, because we were not evaluating the information we 
gathered by the number of experts who mentioned a topic or stated a 
particular view. Our objectives were to identify experts’ (1) general 
understanding of the current state of climate science and engineering 
and (2) their major uncertainties and outstanding issues on these 
subjects. We did not attempt to determine the independence of 
individual experts, but we did try to obtain a balanced set of views. 
We wanted to obtain a broad perspective on the current state of 
climate science and engineering and objectively report this 
information. The experts we spoke with are listed in section 8.2. 

We used the Royal Society’s classification of climate engineering 
approaches to focus our review on CDR and SRM technologies (Royal 
Society 2009, l). We did not include climate engineering approaches 
that address other, non-CO2 greenhouse gas emissions such as nitrous 
oxide. After consulting with experts, we limited our assessment of 
climate engineering technologies to those the Royal Society addressed 
in its 2009 report. Of those technologies, we did not assess ocean 
reflectivity or ocean upwelling or downwelling because we found 
limited information on them in the peer-reviewed literature. We also 
did not assess research on the possible causes of climate change. 

We assessed and described the current status of climate engineering 
technologies along four key dimensions: (1) maturity, (2) potential 
effectiveness, (3) cost factors, and (4) potential consequences. We 
assessed the maturity of climate engineering technologies by their 
technology readiness levels (TRL) (table 8.1). TRLs are a standard 
tool for assessing the readiness of an emerging technology for 
production or incorporation into an existing technology or system. The 
Department of Defense and NASA use TRLs, as does the European Space 
Agency. 

We used the AFRL (Air Force Research Laboratory) Technology Readiness 
Level Calculator Version 2.2 (Nolte 2004) to determine technology 
readiness levels for the climate engineering technologies we reviewed. 
Table 8.1 summarizes key features of TRL ratings. The first column 
presents definitions of TRL levels used as “Top Level Views” in the 
TRL calculator. The calculator operates conditionally: to achieve a 
rating at any level, a technology must satisfy the requirements for 
all lower levels as well. For example, to achieve a rating of TRL 2, a 
technology must also satisfy the requirements for a rating of TRL 1. 
To achieve a rating of TRL 3, a technology must also satisfy the 
requirements for a rating of TRL 2, and thus must also satisfy the 
requirements for a rating of TRL 1. 

We developed criteria to rate climate engineering technologies using 
the TRL calculator. For the top level view of TRL 1, requiring that 
basic principles be observed and reported, we asked whether the 
technology had been described as a climate engineering technology in 
peer-reviewed scientific literature. All the climate engineering 
technologies we reviewed met this condition. 

For the top level view of TRL 2, requiring the formulation of a 
technology concept or application, we asked whether a system concept 
identifying key elements of the technology or a concrete plan existed 
for implementation on a global scale. Some technologies failed to meet 
this condition for climate engineering even though they would be fully 
mature in other applications. For example, increasing the reflectivity 
of settled areas by painting rooftops white would be mature on a small 
scale but lacked a system concept and a concrete plan for 
implementation on a global scale. Since this technology failed to meet 
the condition for TRL 2, it was rated at TRL 1. The sensitivity of the 
TRL ratings to the definition of a system concept or a concrete plan 
for climate engineering is a key limitation of using TRLs to evaluate 
technologies that are otherwise mature. 

Table 8.1 Nine technology readiness levels described: 

Level: 1. Basic principles have been observed and reported; 
Description: The lowest level of technology readiness. Scientific 
research begins translation into applied research and development; 
Example: Paper studies of the technology’s basic properties. 

Level: 2. Technology concept or application has been formulated; 
Description: Invention begins. Once basic principles are observed, 
practical applications can be invented. The application is 
speculative, and no proof or detailed analysis supports the assumption; 
Example: Limited to paper studies. 

Level: 3. Analytical and experimental critical function or 
characteristic proof of concept has been defined; 
Description: Active research and development (R&D) begins. Includes 
analytical and laboratory studies to physically validate analytical 
predictions of separate elements of the technology; 
Example: Components that are not yet integrated or representative 

Level: 4, Component or breadboard validation has been made in 
laboratory environment; 
Description: Basic technological components are integrated to 
establish that the pieces will work together. This is relatively “low 
fidelity” compared to the eventual system; 
Example: Ad hoc hardware integrated in a laboratory. 

Level: 5. Component or breadboard validation has been made in relevant 
environment; 
of components
Description: Fidelity of breadboard technology increases 
significantly. The basic technological components are integrated with 
reasonably realistic supporting elements so the technology can be 
tested in a simulated environment; 
Example: “High fidelity” laboratory integration. 

Level: 6. System and subsystem model or prototype has been 
demonstrated in a relevant environment; 
Description: Representative model or prototype system is well beyond 
level 5 testing in a relevant environment. Represents a major step up 
in the technology’s demonstrated readiness; 
Example: Prototype tested in a high-fidelity laboratory or simulated 
operational environment. 

Level: 7. System prototype has been demonstrated in an operational 
environment; 
Description: A prototype is operational or nearly operational. 
Represents a major step up from level 6, requiring the demonstration 
of an actual system prototype in an operational environment, such as 
in an aircraft, vehicle, or space; 
Example: Prototype tested in a test bed aircraft; 

Level: 8. Actual system is complete and has been qualified in testing 
and demonstration; 
Description: Technology has been proven to work in its final form and 
under expected conditions. In almost all cases, this level represents 
the end of true system development; 
Example: Developmental test and evaluation of the system to determine 
if it meets design specifications. 

Level: 9. Actual system has been proven in successful mission 
operations; 
Description: The technology is applied in its final form and under 
mission conditions, such as those encountered in operational test and 
evaluation. In almost all cases, this is the end of the last “bug 
fixing” aspects of true system development; 
Example: The system is used in operational mission conditions. 

Source: GAO based on Nolte (2004). 

Note: A breadboard is a representation of a system that can be used to 
determine concept feasibility and develop technical data. It is 
typically configured for laboratory use only. It may resemble the 
system in function only. 

[End of table] 

For the top level view of TRL 3, requiring analytical and experimental 
demonstration of proof of concept, we looked for significant 
experimental data on elements of the technology. For example, a 
technology designed to reduce solar radiation by placing scatterers at 
L1 fulfilled the basic requirements for TRL 2 but not TRL 3 because 
the supporting literature was theoretical and did not provide 
experimental data. Finally, for the top level view of TRL 4, requiring 
technological demonstration, we looked for evidence of system 
demonstration with a breadboard unit (a representation of the system, 
in function only, used to determine feasibility and to develop data, 
configured for laboratory use). Because none of the technologies that 
we reviewed had system data with breadboard units, none could be rated 
at TRL 4 or higher. 

We had earlier recommended that a technology should be at level 7-—
that is, a prototype has been demonstrated in an operational 
environment-—before being moved to engineering and manufacturing 
development. We had recommended further that a technology be at level 
6 before starting program definition and risk reduction (GAO 1999). We 
characterize technologies whose TRL scores are below 6 as “immature.” 

Two factors that affect global temperature are (1) the level of CO2 
and other greenhouse gases in the atmosphere and (2) the amount of 
solar radiation that Earth and its atmosphere absorb. 

Because CDR and SRM affect temperature in different ways, their 
effects are measured differently. CDR removes CO2 from the atmosphere 
while SRM reduces the amount of solar radiation that Earth and its 
atmosphere absorb—by reflecting the radiation into space before it 
reaches Earth’s atmosphere, when it reaches Earth’s atmosphere, or 
when it reaches Earth’s surface. 

To describe the effectiveness of proposed CDR technologies, we 
examined the Royal Society’s qualitative ratings of various 
technologies’ effectiveness (high, medium, and low). We also examined two 
quantitative measures reported in the literature: the estimated (1) 
maximum reduction of the atmospheric concentration of CO2 (ppm) from 
its projected level of 500 ppm in 2100 and (2) annual ability to 
remove CO2 from Earth’s atmosphere (gigatons of CO2 per year) when 
compared to annual anthropogenic emissions of 33 gigatons of CO2.
[Footnote 88] We assessed the qualitative ratings primarily by making: 

* a check for reasonableness. For example, for bioenergy with CO2 
capture and sequestration (BECS) the Royal Society reported an 
anticipated maximum CO2 reduction ability of between 50 ppm and 150 
ppm and rated BECS as low to medium in effectiveness. We confirmed the 
reasonableness of rating a reduction of 150 ppm as having medium 
effectiveness by noting that this level of reduction would put the 
concentration of CO2 in the year 2100 at 350 ppm—-which is below 
the current 390 ppm but does not approach the preindustrial 280 ppm. 

* comparisons to other scientific sources. We reviewed scientific 
literature for other assessments indicating the overall feasibility
of implementing individual CDR technologies on a global scale to 
achieve a net reduction of atmospheric CO2 concentration. For two 
technologies-—direct air capture of CO2 with geologic sequestration 
and enhanced weathering—-sources in the peer-reviewed literature 
provided views or information that differed substantially from the 
Royal Society’s ratings.[Footnote 89] 

Overall, for three of the six CDR technologies, our assessments 
confirmed the specific Royal Society qualitative effectiveness 
ratings. We included these three Royal Society ratings in the “
potential effectiveness” column in table 3.1. For one other technology 
(land use management), which the Royal Society rated as low, other 
scientific literature suggested a low to medium rating, which is 
reflected in table 3.1. For the remaining two CDR technologies (direct 
air capture of CO2 with sequestration, and enhanced weathering), we 
did not report an overall qualitative rating for potential 
effectiveness; that is, we indicated “not rated” because sources in 
the scientific literature provided information that differed 
considerably from the Royal Society’s ratings. However, where 
possible, we provided other relevant information. 

To describe the potential effectiveness of SRM technologies, we used 
the generally accepted benchmark of the climate change community (such 
as in the work of the Intergovernmental Panel on Climate Change 
(IPCC)) called equilibrium climate sensitivity.[Footnote 90] Climate 
modeling studies use equilibrium climate sensitivity as a benchmark to 
indicate the effect of greenhouse gases on the climate. Equilibrium 
climate sensitivity is defined as the change in global mean surface 
temperature following warming caused by a doubling of preindustrial 
CO2 levels (Solomon et al. 2007). The doubling of preindustrial CO2 
levels is also used in modeling studies as a standard condition for 
evaluating climate effects other than an increase in global average 
temperature. Following this approach, the climate engineering 
community evaluates the effects of SRM technologies against double 
preindustrial CO2 levels. We described the potential effectiveness of 
SRM technologies when fully implemented on a global scale, based on 
the extent to which they are estimated to reduce global average 
surface temperature compared to the benchmark. We categorized the 
potential effectiveness of each climate engineering technology as a 
percentage, where 100 percent is anticipated to lower global mean 
temperature from the benchmark to the preindustrial value and is 
termed “fully effective.” 

We did not assess the effectiveness of either deploying multiple 
climate engineering technologies simultaneously or combining them with 
reductions in carbon emissions and advances in energy technology. We 
did not assess the effectiveness of deploying a technology in any 
specific place. Because we focused on global mean surface temperature, 
we did not assess specific geographic temperatures or climate changes.
We did not independently determine the costs of implementing the 
technologies. Instead, we report cost factors and estimates from the 
literature we reviewed; these are based on ideas of what the 
technologies might be, not on detailed design and schedule data. For 
CDR, the cost factors represent resources used to remove CO2 from the 
atmosphere and store it; when quantified, these are presented on a per 
ton basis. For SRM, the cost factors represent resources required to 
counteract global warming from doubling the preindustrial atmospheric 
concentration of CO2, or, for technologies that are not anticipated to 
be fully effective, the resources required to counteract warming to 
the maximum extent possible. We were not able to determine the 
reliability of estimated costs in the literature because of 
insufficient data or inadequate descriptions of how costs were 
determined. 

We assessed the potential consequences of each technology by 
summarizing risks or consequences identified in the literature, 
modeling studies, and our interviews with experts. We also reviewed 
congressional hearings for the testimony of experts who presented 
risks of implementing specific technologies. We reviewed the ability 
of existing climate models to represent climate processes expected to 
result from climate engineering technologies, including altered wind 
currents, rain patterns, and ocean temperatures. We considered the 
ability to reverse a technology’s deployment as a type of consequence. 

To report on the status of knowledge and tools for understanding 
climate engineering, we reviewed relevant literature and interviewed 
scientists and other experts about climate science, observational 
networks, and computing resources. Our literature review included GAO 
publications as well as reports from the Department of Energy, 
Intergovernmental Panel on Climate Change, National Aeronautics and 
Space Administration, National Institute of Standards and Technology, 
National Oceanic and Atmospheric Administration, National Research 
Council, United States Climate Change Science Program, and World 
Climate Change Programme, in addition to peer-reviewed scientific 
literature. Because we found few studies focusing on climate 
engineering modeling or research, we included in our review some 
studies of climate models and science that are relevant to climate 
engineering. Our research on observational systems focused on the 
coverage, continuity, and accuracy of networks collecting measurements 
related to substances or processes that are important to climate 
engineering. Similarly, our examination of computing resources focused 
on current limitations or potential improvements that could affect 
climate engineering research, such as the spatial resolution of 
computations in current models. We did not independently evaluate 
whether scientific knowledge or tools are sufficiently well understood 
or developed for making decisions about the possible development or 
use of climate engineering technologies. We also did not assess 
whether climate change is occurring or what is causing any climate 
change if it is occurring or whether current scientific knowledge 
supports the occurrence of climate change or its causes. 

8.1.2. Our method for eliciting experts’ views of the future of 
climate engineering research: 

To assess how climate engineering research might develop in the 
future, we used the following three sources: (1) a foresight exercise 
in which experts developed alternative scenarios, (2) the comments of 
a broad array of experts stimulated by the scenarios, and (3) 
additional views of other experts in response to our preliminary 
synthesis developed from the scenarios and earlier comments. Sections 
8.4 to 8.6 list the experts we consulted in developing each of these 
sources.[Footnote 91] We present our summary of the three sources in 
the body of our report to suggest some possibilities for climate 
engineering research over the next 20 years. 

All experts we selected to participate in the foresight exercise and 
to comment in response to the scenarios met at least one of the 
following criteria: they (1) held a position in a university or other 
well-known organization relevant to climate engineering, climate 
change, or related topics; (2) had participated in academic or 
professional panels addressing climate engineering, climate change, or 
related topics; or (3) had authored peer-reviewed publications on 
climate engineering, climate change, or related topics. 

8.1.2.1. Scenario-building process: 

A meeting to build scenarios held on
July 27, 2010, at GAO headquarters was facilitated by a professional 
from the Institute for Alternative Futures. The overall goal of the 
exercise was to develop four scenarios to illustrate alternative 
possible futures for climate engineering research, including the 
amounts and kinds of research that might be conducted on CDR and SRM 
and whether significant progress was expected. 

The scenario-builders (listed in section 8.4) were selected to 
constitute, as a group, 

* expertise on specific technologies for engineering the climate, 
including CDR and SRM, and experience in the research or development 
of relevant technologies; 

* knowledge on climate engineering as well as the development of 
future-oriented scenarios, including foresight about emerging 
technologies and national and international approaches to them; and; 

* collective backgrounds in private industry, government (including 
the military), and other organizations such as those in academia. 

We selected six external scenario-builders. Each of the six was a 
leading expert in one or more key fields or had been recommended to us 
by other experts. The group’s knowledge and expertise represented a 
balance across the items bulleted above and spanned energy policy, 
climate change, oceanography, atmospheric science, and biotechnology, 
as well as research on CDR and SRM and other areas, such as foresight
and public engagement. Timothy Persons, GAO’s Chief Scientist, served 
as the host and ex-officio member of the group to help guide the 
discussion. 

To build the four scenarios, we began by reviewing scientific and 
engineering literature and interviewing scientists and engineers to 
help us identify what were likely to be the key factors in the future 
scope and direction of climate engineering research in the United 
States. We used this information to construct a questionnaire that we 
sent by e-mail before the meeting to the six external experts and 
GAO’s Chief Scientist. 

Before the scenario-builders met on July 27, 2010, they responded 
individually to our e-mail questionnaire. The questionnaire asked for 
their opinions on the goals of climate engineering research, the 
importance of making substantial progress toward those goals by 2030, 
the promise of different approaches toward reaching the goals, the 
research that might appropriately be supported by private or 
government funds, any leadership the federal government should take on 
climate engineering research, the need for international cooperation, 
the likelihood of future climate changes, and the moral hazard if 
climate engineering research looked as if it were headed on an 
efficient and effective course. We also asked for separate answers to 
these questions as they related expressly to CDR and SRM technologies. 
The questionnaire also listed various factors that might affect 
climate engineering research and asked for the scenario-builders’ 
opinions on these and other relevant factors. The answers we received 
suggested the importance of factors subsequently selected for the 
meeting’s discussion. For example, five of the six scenario-builders 
responded that government incentives to industry would make the 
prospect of achieving some or all CDR research goals by 2030 “highly 
promising.” We provided them with a summary of their answers to the 
questionnaire at the outset of the scenario-building meeting. 

During the meeting, the scenario-builders identified and discussed 
many kinds of factors important for future U.S. climate engineering 
research in a global context. They selected two policy-related factors 
as potentially most significant. One was whether a federal research 
program on CDR and SRM would be established and, if so, at what level 
(the scenario-builders did not focus on low-risk SRM methods such as 
whitening roofs and roads). The scenario-builders discussed a broad 
definition of a research program that might include related 
activities, such as engaging the public or encouraging industry to 
implement technology-related results (including improving 
opportunities for dissemination). The other policy factor was whether 
carbon constraints would be established and, if so, at what level in 
the United States and internationally. 

The scenario-builders discussed how carbon constraints can take the 
form of either emissions pricing or regulations designed to reduce 
carbon emissions. After selecting the two factors—a federal research 
program and carbon constraints—the scenario-builders specified three 
levels for each one, defining nine combinations, each of which might 
serve as the basis for a scenario (figure 8.1). From the nine possible 
combinations, the scenario-builders selected for further consideration 
the four combinations labeled on figure 8.1. The four resulting 
scenarios define a range of futures within the bounds set by variation 
across the two selected factors. 

Each scenario was developed separately for a specific combination of 
factors. However, a logical inference is that more pathways are 
possible within the range defined by the two factors because of the 
possibility of transitions from one scenario to another. Scenario II, 
for example, could overlap with Scenario IV. The purpose of the 
scenarios was to stimulate thinking about the future, not to limit 
anticipation to any one cell. 

We asked the scenario-builders to identify low-probability high-impact 
events such as “black swans” and “black pearls.” We defined black swan 
as an extremely unlikely event able to produce catastrophic or 
otherwise large effects. We defined black pearl as a black swan with 
positive effects. We generated this list to help identify wild cards 
or conditions that could drastically change the future as related to 
the climate and climate engineering research. 

Toward the end of the scenario-building meeting, we asked the six 
external scenario-builders to look ahead to 2030 and beyond and to 
consider possible outcomes linked to research on CDR and SRM. We asked 
them to assess, subjectively and qualitatively, three potential future 
situations that might occur in or after 2030: (1) an emergency in 
which decision-makers might consider using SRM, (2) continued global 
warming, and (3) a future with no further warming. 

Figure 8.1 Four scenarios defining alternative possible futures: 

[Refer to PDF for image: scenario matrix] 

Scenario I: Status quo on both: 
Relative level of federal research program (climate engineering) and 
related activities: low or none; 
Level of constraint on carbon: low or none. 

Scenario II: Some action on both; 
Relative level of federal research program (climate engineering) and 
related activities: medium; 
Level of constraint on carbon: medium. 

Scenario III: Action on research only; 
Relative level of federal research program (climate engineering) and 
related activities: high; 
Level of constraint on carbon: low or none. 

Scenario IV: More action on both; 
Relative level of federal research program (climate engineering) and 
related activities: high; 
Level of constraint on carbon: high. 

Source: GAO. 

[End of figure] 

8.1.2.2. Experts’ comments stimulated by scenarios: 

We used the scenarios to elicit additional views about the future from 
28 experts (listed in section 8.5) who represented a wider range of 
backgrounds and perspectives. To help ensure balance in the wider 
group of experts who would review and respond to the scenarios, we 
specifically selected some experts with competing views and different 
backgrounds. These experts were thus characterized by: 

* varied backgrounds (including, for example, economics, ethics, the 
humanities, and international relations); 

* a range of organizational affiliations (including universities, the 
public sector, the private sector, and advocacy groups or other 
organizations associated with a viewpoint); and; 

* differing perspectives (including some known to favor or oppose the 
development of climate engineering technologies or to have expressed 
uncertainty about climate change trends). 

In August 2010, we e-mailed the four scenarios to the selected 28 
experts along with a brief questionnaire on their reactions to the 
scenarios. We invited them to provide alternative mini scenarios or 
other statements of their views about the future. We asked them to 
identify black swans and black pearls. We also asked them for any 
message about the future of climate engineering research and its 
consequent risks that they believed would be important for 
policymakers to consider. Not all expressed views on all issues. We 
followed up with e-mail questions for clarification, as needed. In a 
few instances, we followed up with telephone conversations or met in 
person with experts who were available in the Washington, D.C., area. 
We synthesized the varied responses we received from the experts. 

8.1.2.3. Experts’ views of our initial synthesis and preliminary 
findings: 

As we described above, we convened with NAS’s help a meeting of 
scientists, engineers, and other experts. For this meeting, we 
presented information about the scenarios and asked the experts to 
discuss our preliminary findings about views expressed regarding the 
future and to share their own views about the future. Some experts
did not express views on the scenarios or all topics discussed. 

8.1.2.4. Our analysis: A qualitative foresight synthesis: 

We call our summary of the combined results of the exercises we have 
described a qualitative foresight synthesis. The summary is primarily 
based not on how many experts made specific comments or any number of 
votes taken of the experts but, rather, on a qualitative approach in 
which we identified recurring, prominent themes and used professional 
judgment. The summary is a synthesis of views from a diverse range of 
experts and from three interconnected foresight exercises. It is the 
result of an iterative process whereby one set of experts developed 
scenarios, another set commented on those scenarios, and a third set 
reviewed our initial synthesis of the first two exercises. In areas 
where either a clear majority of the experts we consulted agreed or 
only a small number took a specific position, we say that a “majority” 
expressed the position or that a small number stated a concern. 
However, for transparency, footnotes provide information on specific 
counts of experts we consulted who voiced key opinions. 

Although the experts we consulted do not necessarily represent the 
views of all those with similar expertise in the area of climate 
engineering, because of the three-stage process and the breadth of 
experts we consulted, we believe that the resulting overall set of 
views about the future that we present in section 4 of this report (“
Experts’ Views of the Future of Climate Engineering Research”) would 
be similar even if we had used a different set of scenarios or if we 
had consulted with a different but still diverse set of experts. 

8.1.3. Our method for assessing potential responses to climate 
engineering: 

To gather information about public awareness of and views on 
geoengineering technologies, we reviewed selected survey research on 
public opinion on climate change, conducted focus groups, and 
contracted with Knowledge Networks Inc. to use its online research 
panel to field a survey we developed. The survey was fielded from July 
19 to August 5, 2010. Of a total sample of 1,623 U.S. residents 18 
years old and older, 1,006 completed the survey. 

From our review of the research on climate engineering and survey 
research on climate change, we did not expect the focus group 
participants or survey respondents to know very much about climate 
engineering technologies. Therefore, before asking questions about 
geoengineering, we gave the focus group participants and the survey 
respondents a basic definition of geoengineering, described the 
differences between CDR and SRM, and provided examples of both. The 
level of information we gave the focus group participants and survey 
respondents was comparable to what average adults exposed to news 
media descriptions of these technologies might be expected to receive. 

To help us develop the protocol for the focus group with members of 
the public and to increase our understanding of public perceptions of 
climate change and climate engineering, we conducted four focus groups 
with GAO employees. We used what we learned from these focus groups to 
make changes to the protocol for the public focus group.[Footnote 92] 
We selected the 11 members of the public focus group for their 
diversity in age, gender, race, ethnicity, and education. Some 
participants spoke both English and Spanish; they translated for one 
participant who was fluent only in Spanish. A GAO analyst fluent in 
English and Spanish observed the focus group. 

We first asked the focus group participants to discuss their beliefs 
about climate change, including whether they believed the climate is 
changing and, if so, what the cause is. We then asked them if they 
thought there was anything they personally could do to affect climate 
change and what, if anything, the public, industry, government, or 
scientists and engineers should do with respect to climate change. 
Participants identified personal actions such as driving less, using 
alternative fuels, and writing letters to influence elected 
representatives. With respect to government, industry, and scientists 
and engineers, participants thought greater enforcement of existing 
laws, the provision of government incentives to address climate 
change, and increased public education about climate change were ways 
to address climate change. When asked whether they were aware of any 
scientific or engineering solutions to climate change, focus group 
participants did not identify any specific solutions. One participant 
stated that scientists and engineers might develop solutions to 
climate change but that money is not being directed to this. 

After asking focus group participants if they were aware of any 
scientific or technological solutions to climate change, we explained 
what geoengineering is and gave them information about three different 
technologies, including CDR and SRM technologies. We asked 
participants to discuss their reactions to each technology and whether 
they supported or opposed it. In addition, we asked them to discuss 
how the federal government, industry, and individuals should fund and 
make decisions about geoengineering. 

We chose to use “geoengineering” in the information we gave the focus 
group and survey participants, given that we and others, such as the 
Royal Society, had used this term earlier. In our focus groups, we 
found that participants raised concerns about the potential for harm 
from geoengineering technologies and reacted differently to different 
technologies. For example, one participant, asked to react to 
information about stratospheric sulfates, expressed the view that 
dinosaurs had become extinct by the Sun’s having been blocked. 
Another, reacting to the concept of direct air capture, expressed 
concern about the long-term storage of CO2. 

To assess whether these differences in reaction to different 
technologies exist also in the larger population, we administered a 
split-ballot survey. Using experts’ assessments of safety described in 
the Royal Society report on geoengineering, we gave half the 
respondents information about technologies (one CDR and one SRM) that 
experts identified as relatively safe, and we gave the other half 
information about technologies (one CDR and one SRM) that experts 
identified as relatively less safe. We included a question in the 
survey to assess whether this difference in information about experts’ 
assessment of safety affected participants’ perceptions of potential 
harm from CDR and SRM technologies. We also examined whether views 
about geoengineering research, development, and decision-making were 
affected by learning about more or less safe technologies. Our survey 
results indicated that respondents differed in their level of concern 
about harm from geoengineering, depending on whether they received 
information about more or less safe technologies, but they did not 
differ greatly in their responses to other questions about 
geoengineering research and decision-making. Consequently, we report 
the results from all other survey questions combined. 

The respondents could choose one or more of three ways to receive 
information about different types of geoengineering technologies: they 
could (1) view a video and listen to a narration, (2) listen to the 
narration, or (3) read printed information. All survey questions were 
identical in the two survey ballot groups. 

Every survey introduces sampling and nonsampling errors, including 
errors of processing, measurement, coverage, and nonresponse. We took 
steps to reduce such errors. To reduce processing error, we verified 
all computer programming and analyses independently. To reduce 
measurement error, we conducted 11 pretests with persons of varied 
education, income, English proficiency, age, gender, and race. 
[Footnote 93] The pretests included face-to-face interviews using the 
draft written survey as well as telephone interviews with those 
completing the web version of the survey. From the pretest results, we 
made a number of changes to reduce the likelihood of measurement error 
from respondents’ misunderstanding or misinterpreting the survey 
questions. We also asked all pretest respondents whether any specific 
questions or the survey overall was biased in any way, and we made 
changes to address the concerns they raised. Knowledge Networks’ 
online research panel was designed to minimize errors of coverage of 
the target population of U.S. adults. The sample frame was based on 
probability sampling that covered both people who had home access to 
the Internet and those who did not. Knowledge Networks also used a 
dual sampling frame that included both households that had telephones 
(including only cell phones) and households that did not, as well as 
households with listed and those with unlisted telephone numbers. 
Knowledge Networks recruited panel members randomly. Households were 
provided with access to the Internet and the necessary hardware if 
they needed it. For a specific survey like ours, Knowledge Networks 
selects panelists randomly, and no one not selected may respond. 

To calculate the survey’s response rate, we used RR4, a method 
described by the American Association for Public Opinion Research. The 
RR4 method is based on multiplying the recruitment rate (18.3 
percent), the profile rate (58.4 percent), and the completion rate 
62.0 percent) to yield an overall response rate of 7 percent. To 
reduce the potential for nonresponse error, we weighted the survey 
data using Knowledge Networks’ study-specific post-stratification 
weight. From our assessment of Knowledge Networks’ probability 
sampling methods and weighting methodologies and the results of our 
nonresponse bias analysis, we determined that the sample selected for 
our study was statistically representative of the U.S. adult 
population. 

Sampling error is a measure of the likely variation introduced in a 
survey’s results by using a probability procedure based on random 
selections. In terms of the margin of error at the 95 percent 
confidence level, the sampling error for survey estimates from the 
total sample is plus or minus 4 percentage points, unless otherwise 
noted. In terms of the margin of error at the 95 percent confidence 
level, the sampling error for estimates based on subgroups of the 
sample is plus or minus 9 percentage points, unless otherwise noted. 
Because the overall response rate was low and because sources of 
nonsampling error such as differences in survey results from panel 
attrition and panel conditioning might be present, nonsampling error 
may also have contributed to the total survey error of the results. To 
avoid false precision, therefore, we rounded the survey results we 
report in the text to the nearest 5 percentage points. 

The public perceptions elicited by this survey are based on limited 
information about geoengineering and do not necessarily predict U.S. 
public views. We found that about 65 percent of the respondents had 
not heard about geoengineering before reading the survey; therefore, 
responses to the survey are likely to reflect reactions to information 
about geoengineering that we provided in the survey. 

If the respondents had been provided with different information about 
geoengineering, the survey responses could also differ. Also, climatic 
or other events might change public views of geoengineering. When we 
asked respondents about their support for geoengineering research or 
for government funding of geoengineering research, we did not present 
them with competing programs to choose from (programs for cancer 
treatment, for example) or with alternatives, such as using government 
funding for national defense or cutting taxes. These kinds of choices 
might have produced different results. 

The initial version of the survey included a question designed to help 
assess whether the respondents thought that exploring geoengineering 
solutions could distract from other potential solutions to climate 
change, such as reducing CO2 emissions by driving less or developing 
more fuel-efficient technologies. Because the question included more 
than one policy option on which respondents could hold different views 
and focused on what respondents would expect to happen in the future 
but could not yield direct information about how members of the public 
might actually behave, we revised the survey to include a separate 
series of questions to assess where initial support for geoengineering 
might fall relative to other policy options. 

8.1.4. External review: 

We invited all participants in the Meeting on Climate Engineering to 
review our draft report. We sent the draft report to the 16 
participants who agreed to review and help revise the report. While we 
asked the 16 reviewers to focus on the sections most relevant to their 
expertise, we also told them that we welcomed any comments on the 
entire draft. One of the 16 did not participate in the review because 
of schedule conflicts. Fifteen reviewers (see section 8.7) provided 
technical or other comments that we incorporated as appropriate. These 
15 reviewers were meeting participants who collectively represented 
expertise relevant to each of the three major areas of our report, 
including the current state of climate engineering technology, expert 
views of the future of climate engineering research, and public 
perceptions of climate engineering. The external review was conducted 
in February 2011. 

We conducted our work for this technology assessment from January 2010 
through July 2011 in accordance with GAO’s quality standards as they 
pertain to technology assessments. Those standards require that we 
plan and perform the technology assessment to obtain sufficient and 
appropriate evidence to provide a reasonable basis for our findings 
and conclusions based on our technology assessment objectives and that 
we discuss limitations of our work. We believe that the evidence we 
obtained provides a reasonable basis for our findings and conclusions, 
based on our technology assessment objectives. 

8.2. Experts we consulted on climate engineering technologies: 

Barrett, Scott, Lenfest-Earth Institute Professor of Natural Resource 
Economics, School of International and Public Affairs and Earth 
Institute, Columbia University, New York. 

Benford, Gregory, Professor of Physics, Department of Physics and 
Astronomy, University of California, Irvine. 

Caldeira, Ken, Physicist and Environmental Scientist, Energy and 
Environmental Sciences Directorate, Lawrence Livermore National 
Laboratory, Livermore, California. 

Crutzen, Paul J., Emeritus, Max Planck Institute for Chemistry, Mainz, 
Germany; Institute Scholar, International Institute for Applied 
Systems Analysis, Laxenburg, Austria; Emeritus Professor, Scripps 
Institution of Oceanography, University of California at San Diego, La 
Jolla. 

Doney, Scott C., Senior Scientist, Department of Marine Chemistry and 
Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, 
Massachusetts. 

Ducklow, Hugh W., Director and Senior Scientist, The Ecosystems 
Center, Marine Biological Laboratory, Woods Hole, Massachusetts; 
Professor, Department of Ecology and Evolutionary Biology, Brown 
University, Providence, Rhode Island. 

Dyson, Freeman, Professor Emeritus, School of Natural Sciences, 
Institute for Advanced Study, Princeton, New Jersey. 

Fahey, David W., Research Physicist, Atmospheric and Chemical 
Processes, Chemical Sciences Division, Earth System Research 
Laboratory, National Oceanic and Atmospheric Administration, Boulder, 
Colorado. 

Garten, Jr., Charles T., Senior Research Staff Member, Nutrient 
Biogeochemistry Group, Environmental Sciences Division, Oak Ridge 
National Laboratory, Oak Ridge, Tennessee. 

Gibbons, John H. (Jack), President, Resource Strategies, The Plains, 
Virginia; Consultant, Lawrence Livermore National Research Laboratory, 
Livermore, California; Division Advisor, Division on Engineering and
Physical Sciences, The National Academies, Washington, D.C. 

Hack, James J., Director, National Center for Computational Sciences, 
Oak Ridge National Laboratory, Oak Ridge, Tennessee. 

Keeling, Ralph, Professor, Scripps Institution of Oceanography, 
University of California at San Diego, La Jolla. 

Keith, David, Director, ISEEE Energy and Environmental Systems Group; 
Professor and Canada Research Chair of Energy and the Environment; 
Professor, Department of Chemical and Petroleum Engineering, 
University of Calgary, Calgary, Alberta, Canada; Adjunct Professor, 
Department of Engineering and Public Policy, Carnegie Mellon 
University, Pittsburgh, Pennsylvania. 

Lackner, Klaus, Department Chair, Ewing and J. Lamar Worzel Professor 
of Geophysics, Earth and Environmental Engineering and Director, 
Lenfest Center for Sustainable Energy, The Earth Institute, Columbia 
University, New York. 

Latham, John, Emeritus Professor of Physics, University of Manchester, 
United Kingdom; Visiting Professor, University of Leeds, United 
Kingdom; and Senior Research Associate, National Center for 
Atmospheric Research, Boulder, Colorado. 

Lindzen, Richard S., Alfred P. Sloan Professor of Meteorology, 
Department of Earth, Atmospheric, and Planetary Sciences, 
Massachusetts Institute of Technology, Cambridge, Massachusetts; 
Distinguished Visiting Scientist, Jet Propulsion Laboratory, 
California Institute of Technology, Pasadena, California. 

Long, Jane C. S., Associate Director, Energy and Environment 
Directorate, Lawrence Livermore National Laboratory, Livermore, 
California. 

MacCracken, Michael, Chief Scientist for Climate Change Programs, 
Climate Institute, Washington, D.C. 

MacDonald, Alexander E. “Sandy,” Deputy Assistant Administrator for 
Laboratories and Cooperative Institutes, Office of Oceanic and 
Atmospheric Research; Director, Earth System Research Laboratory, 
National Oceanic and Atmospheric Administration, Boulder, Colorado. 

Marland, Gregg, Distinguished R&D Staff, Environmental Sciences 
Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee; Guest 
Professor, Ecotechnology Program, Mid Sweden University, Östersund, 
Sweden. 

Melillo, Jerry M., Distinguished Scientist, The Ecosystems Center, 
Marine Biological Laboratory, Woods Hole, Massachusetts; Professor 
(MBL) of Ecology and Evolutionary Biology, Department of Ecology and 
Evolutionary Biology, Division of Biology and Medicine, Brown 
University, Providence, Rhode Island. 

Morgan, M. Granger, University Professor, Lord Chair Professor of 
Engineering, Head, Department of Engineering and Public Policy, 
Professor of Engineering and Public Policy and of Electrical and 
Computer Engineering, and Professor, The H. John Heinz III School of 
Public Policy and Management, Carnegie Mellon University, Pittsburgh, 
Pennsylvania. 

Parson, Edward A. “Ted,” Joseph L. Sax Collegiate Professor of Law and 
Professor of Natural Resources and Environment, University of 
Michigan, Ann Arbor, Michigan; Senior Research Associate, Centre for 
Global Studies, University of Victoria, British Columbia, Canada. 

Rasch, Philip, Chief Scientist for Climate Science and Laboratory 
Fellow, Pacific Northwest National Laboratory, Richland, Washington. 

Ravishankara, A. R., Director, Chemical Sciences Division, Earth 
System Research Laboratory, National Oceanic and Atmospheric 
Administration, Boulder, Colorado; Assistant Professor, Department of 
Chemistry and Biochemistry, and Affiliate, Cooperative Institute for 
Research in Environmental Sciences, University of Colorado, Boulder, 
Colorado. 

Robock, Alan, Distinguished Professor (Professor II), Department of 
Environmental Sciences; Associate Director, Center for Environmental 
Prediction; Director, Meteorology Undergraduate Program; Member, 
Graduate Program in Atmospheric Science, Rutgers University, New 
Brunswick, New Jersey. 

Rothstein, Lewis M., Professor of Oceanography, Graduate School of 
Oceanography and Treasurer, Metcalf Institute Advisory Board, Metcalf 
Institute for Marine and Environmental Reporting, University of Rhode 
Island, Narragansett, Rhode Island. 

Schneider, Stephen H., Melvin and Joan Lane Professor for 
Interdisciplinary Environmental Studies and Professor, Department of 
Biology, Stanford University; Senior Fellow, Woods Institute for the 
Environment; Professor, by courtesy, Civil and Environmental 
Engineering, Stanford University, Stanford, California (deceased July 
19, 2010). 

Shepherd, John, Professorial Research Fellow in Earth System Science, 
School of Ocean and Earth Science, National Oceanography Centre, 
University of Southampton, Southampton, United Kingdom. 

Socolow, Robert H., Professor, Department of Aerospace and Mechanical 
Engineering; Co-Director, Carbon Mitigation Initiative, Princeton 
University, Princeton, New Jersey. 

Somerville, Richard C. J., Distinguished Professor Emeritus and 
Research Professor, Scripps Institution of Oceanography, University of 
California at San Diego, La Jolla; Team Member, National Science 
Foundation Science and Technology Center for Multiscaling Modeling of 
Atmospheric Processes, Department of Atmospheric Science, Colorado 
State University, Fort Collins, Colorado. 

Strand, Stuart E., Research Professor, Department of Civil and 
Environmental Engineering and School of Forest Resources, University 
of Washington, Seattle, Washington. 

Tilmes, Simone, Project Scientist I, Middle-Upper Atmosphere and WACCM 
Group, The Earth and Sun Systems Laboratory, National Center for 
Atmospheric Research, Boulder, Colorado. 

Tombari, John, President, Schlumberger Carbon Services, Houston, Texas. 

8.3. Foresight scenarios: 

This appendix contains the four scenarios depicting alternative 
futures that six external experts worked with us to develop for use in
this technology assessment. Our purposes in developing these scenarios 
included (1) illustrating how some experts view alternative possible 
futures (2010 to 2030) and judge resulting risk levels (for 2030 and 
later years) and (2) stimulating other experts’ thinking about the 
future and eliciting their views.[Footnote 94] 

Developing these scenarios constituted the first of three steps we 
took to elicit a range of views about the future. In the second and 
third steps, we asked other experts to express views or comment in 
response to the scenarios. Specifically, 

* in step 2, 28 experts responded to the scenarios; and; 
 
* in step 3, 11 additional experts responded to a description of the 
scenarios and a summary that synthesized step-2 comments and views 
about the future. 

Although some commenters at both steps critiqued or suggested 
improvements to the scenarios (on points that concern, for example, 
the effect of carbon constraints, the dollar values associated with 
carbon constraints or research, and the specification of risk), this 
appendix presents the scenarios as they were when the 28 commenters 
first saw them.[Footnote 95] Our report’s methodology is detailed in 
section 8.1; the range of views experts expressed across our three-
step process is represented in the body of the report. 

It is important to keep in mind several characteristics of the four 
scenarios. First, one expert reviewing the scenarios drew our 
attention to a two-part explanation of how carbon constraints could 
affect CDR research (which the scenarios do not describe): (1) 
establishing and maintaining a federal research program that includes 
a significant CDR component is more likely when people are confident 
that CDR technologies will be used once successfully developed and (2) 
establishing carbon constraints could encourage the expectation that 
investing in CDR research is worthwhile. 

Next, whereas two of the scenarios (II and IV) specify a degree of 
carbon constraint that is equivalent to the effect of an international 
price on CO2 emissions applicable across all sectors of all major 
emitting nations, the effect of such a price is not comparable to the 
effect of prices established for limited sectors or regions. 

Further, Scenario II assumes “modest” research funding starting at 
“tens of millions of dollars” for a program involving several 
agencies. This assumption was intended to apply to a dedicated 
research effort for climate engineering that excluded large-scale 
testing and deployment of any of the technologies (which would be much 
more expensive). It was not intended to include relevant but separate 
research in a variety of federal agencies. (Scenarios III and IV, 
which describe greater research efforts, do not specify funding 
levels. We discuss uncertainties about funding in the body of the 
report.) 

Finally, all four scenarios give examples of risk for 2030 and later 
years. They present judgments about levels of risk across three 
potential developments—a future climate emergency and response (that 
could involve decision risks), continued future warming (that could be 
associated with risks from climate change), and no future warming 
(that might possibly be associated in some scenarios with having 
risked resources to prepare for a threat that did not occur). Risk 
levels vary across the scenarios and represent the combined effects of 
factors that are varied across the scenarios, including different 
levels of (1) climate engineering research 2010–30, and the 
technologies and information developed from it, and (2) other factors 
in the scenarios such as emissions reduction. 

We present two key caveats concerning risk levels. The scenarios 
present risk levels that represent (1) inexact qualitative judgments 
that may account for probability and potential severity and (2) 
judgments about degree of risk that are not necessarily comparable 
across the three potential developments.[Footnote 96] Nevertheless, 
comparisons can be made across scenarios. For example, high decision 
risk in one scenario and medium decision risk in another implies a 
judgment that decision risk is higher in one than in the other. 

Finally, we note that the scenarios’ diagrams of risk levels and CO2 
concentrations are not exact but are, instead, illustrative 
approximations. 

Scenario I: Status quo: 

Between 2010 and 2015, various efforts to jumpstart global agreement 
on carbon constraints have only token success.[Footnote 97] Subsequent 
global efforts to stem carbon dioxide (CO2) and other greenhouse gas 
emissions also fail. Individual countries that favor reducing 
emissions realize that they cannot “go it alone” economically. 
Adaptation and mitigation continue on paths set earlier. 

Americans have diverse views about climate change, and those who are 
aware of geoengineering approaches remain skeptical about their safety 
and utility. Debates on global carbon constraints and U.S. 
geoengineering research programs are limited to a small community of 
academics, interest groups, and national decision-makers. Proposed 
federal legislation to establish a research program states four main 
goals: (1) develop inexpensive, scalable carbon dioxide removal (CDR) 
and sequestration methods (using mechanical or biological approaches);
(2) understand and evaluate fast-acting methods like stratospheric 
aerosol injection, including the modeling of potential side effects 
and total cost analyses; (3) involve other nations’ governments and 
scientists in joint research and in setting international research 
guidelines and limitations; and (4) inform decision-makers about 
systemic risks and tradeoffs among various geoengineering technologies 
and between these and other climate change approaches. But 
congressional efforts to enact legislation fail, despite the support 
of nearly half the Congress. 

Without carbon constraints to stimulate private-sector research and 
development (R&D) and without a federal research program dedicated to 
geoengineering, U.S. scientists focus their efforts on other areas. 
The United States makes rapid advances in emerging areas such as 
synthetic biology and nanotechnology, but applications to 
geoengineering are limited. Various other nations (and some private 
sector organizations) develop fast-acting technologies for use in a 
climate emergency, but they do not always focus on identifying side 
effects or share their results with the global scientific community. 
Efforts to develop international guidelines that limit field tests and 
deployment fail. 

Risks across three potential developments (2030 and later years): 

* Climate emergency and response: Immediate decision risks would be 
very high if a sudden acceleration of the warming trend occurred 
spontaneously. World leaders would be under pressure to make decisions 
quickly-—and might opt to use fast-acting, risky geoengineering 
technologies—-despite inadequate information on their effectiveness 
and side effects.[Footnote 98] (See red bar in illustration.) 

Figure: Anticipated future risks: 

[Refer to PDF for image: vertical bar graph] 

Emergency (red bar): High; 
Warming (orange bar): Moderate to high; 
Waste (gray bar): Low. 

[End of figure] 

* Continued warming: Future global climate risks would be high. As of 
2030, this scenario sets the United States and, indeed, the world, on 
a path of increasing CO2 emissions and rising atmospheric 
concentrations. Decades of CDR research starting in 2030 would be 
needed prior to deployment aimed at decreasing future CO2 build-up. 
Prospects thus include temperature increase and far future sea level 
rise that might engulf vulnerable areas, naval installations, and so 
forth; such a future might also bring other very serious negative 
consequences on a global scale. (See the line chart and the orange bar 
in the bar chart.) 

Figure: Line chart: 

[Refer to PDF for image: line chart] 

CO2 build-up plotted against Past to future. 

[End of figure] 

* Resources wasted on geoengineering, if no future warming: Risks of 
having wasted efforts and expenditures would be near zero. This 
scenario commits few, if any, new resources. (See gray bar in the bar 
chart.) 

Scenario II: Some action: 

By 2015 or soon thereafter, somewhat improved data and models of 
climate change reduce uncertainty and appear to validate earlier 
conclusions about global warming caused by human activity. A series of 
extreme weather developments causes widespread concern. As a result, 
major emitting nations agree to new carbon constraints with strong 
enforcement, but the reduction goals and guidelines are limited 
(equivalent to a $10 to $15 price on a ton of CO2). These measures 
slightly increase both mitigation efforts and existing incentives for 
private-sector R&D on direct air capture and sequestration of CO2. 
Scientists expect these changes will not stabilize future accumulated 
levels of CO2 but may delay a far-off climate emergency by about 10 
years and represent a start.[Footnote 99] Additionally, U.S. legislation
establishes a modest geoengineering research program that involves 
several federal agencies 100 The funding level is tens of millions of 
dollars the first year, with plans for modest annual increases. Public 
acceptance of these developments is mixed. The modest research program 
has no public engagement or outreach component. 

Without adequate information on the general public’s views and 
concerns about geoengineering, the government and scientists do not 
craft the research program in a way that encourages public acceptance, 
and inadvertently they alienate some original supporters. Some years 
of benign weather intervene. While the research program continues to 
receive its original level of support, the planned annual expenditure 
increases are not put into effect. The research program leverages 
its limited funds by encouraging the private sector to develop new 
methods of direct CO2 air capture and sequestration that are somewhat 
less expensive than technologies developed through 2010. But both in 
the United States and around the globe, industries that emit significant 
CO2 are not eager to purchase the new technologies to offset emissions: 
the limited carbon constraints have not created a sufficient incentive. 
The research program also makes some advances in developing and 
evaluating fast-acting methods like stratospheric aerosol injection, 
but research by others outpaces the federal effort. Some new fast-acting, 
high-impact technologies are not rigorously evaluated for side effects. 
Results are not always shared with the global scientific community. 
Thus, we lack key information on some new methods and their implications. 

Risks across three potential developments (2030 and later years): 

* Climate emergency and response: Immediate decision risks would be 
moderate to high. By 2030, world leaders responding to an emergency 
would have some geoengineering information to guide them, but the 
information would be inadequate for some new technologies. [Footnote 
101] 

Figure: Anticipated future risks: 

[Refer to PDF for image: vertical bar graph] 

Emergency (red bar): Moderate to high; 
Warming (orange bar): Moderate to high; 
Waste (gray bar): Moderate. 

[End of figure] 

* Continued warming: Future global climate risks would be moderate to 
high. As of 2030, more is known about carbon emissions and controls 
than in Scenario I. Still, starting serious R&D on CDR in 2030 would 
mean years or decades of delay before deployment. The world would likely 
be on a path of continued build-up of CO2 concentrations-—although its 
trajectory would be slightly slower/lower than in Scenario I. The 
prospect of negative consequences like sea level rise would still 
loom, eventually, in the far future. 

Figure: Line chart: 

[Refer to PDF for image: line chart] 

CO2 build-up plotted against Past to future. 

[End of figure] 

* Resources wasted on geoengineering if no future warming: Risks of 
having wasted efforts and expenditures would be moderate. In the 
absence of warming, some new geoengineering technologies would not be 
useful, but others might serve other purposes, such as helping to 
reduce ocean acidification. 

Scenario III: Action on research but not carbon: 

By 2015 or soon thereafter, significantly improved data and models of 
climate change appear to validate earlier conclusions about 
anthropogenic global warming. Highly disruptive and extreme weather 
events affect the United States and many other nations, causing waves 
of concern and even, periodically, a crisis atmosphere. Other nations 
pursue geoengineering research, a fact that is widely reported. The 
balance of U.S. public opinion turns toward taking action on climate 
change, despite opposition from some at home and the lack of global 
agreement on carbon constraints. 

Although public opinion generally favors climate action, some opinion 
leaders believe that, economically, the United States cannot “go it 
alone” in legislating carbon constraints. Those who are opposed 
emphasize this point, and legislative measures to step up U.S. 
emission controls fail on a close vote. At the same time, the Congress 
and the president work together successfully to design and build 
support for legislation that establishes an aggressive federal 
geoengineering research program, starting with moderate resources but 
progressing toward a major funding commitment. The research program 
involves public engagement to build support in the years ahead (including 
years in which extreme climate events may not occur); establishes an 
adaptive strategy that entails periodic reviews by an external body 
such as the National Academies and horizon scans to identify new 
opportunities; promotes innovation through creative incentives, such 
as federal contests with cash awards, in addition to using more 
conventional approaches; and emphasizes international cooperation. The 
main goals of this research program are similar to those in the failed 
legislation outlined in Scenario I (points 1–4). 

As a result, major advances are made in developing, understanding, and 
evaluating fast-acting methods (like next-generation stratospheric 
aerosol and injection methods); understanding tradeoffs among 
different approaches; building new approaches that reduce the 
potential for side effects; and furthering basic science concerning 
climate change. Other advances are made in international cooperation 
on research limitations and guidelines for the use of geoengineering. 
Additionally, the research program helps develop potentially 
transformative methods of direct CO2 air capture and sequestration. 
These new technologies cost substantially less than 2010 technologies 
but, given the lack of carbon constraints, there are virtually no 
incentives for emitting industries to buy them. These technologies 
often fall into the “valley of death” between R&D and commercial 
success and large-scale deployment. Researchers and commercial firms 
become discouraged. The focus on direct air capture and sequestration 
suffers some loss of credibility (that is, the government is seen as 
investing in unused technologies), and it is significantly cut back. 

Risks across three potential developments (2030 and later years): 

* Climate emergency and response: Immediate decision risks would be 
moderate. By 2030, decision-makers have information to support 
decisions about the use (or nonuse) of fast-acting geoengineering 
technologies. Catastrophic risks are minimized.[Footnote 102] 

Figure: Anticipated future risks: 

[Refer to PDF for image: vertical bar graph] 

Emergency (red bar): Moderate; 
Warming (orange bar): High; 
Waste (gray bar): Moderate. 

[End of figure] 

* Continued warming: Future global climate risks would be high. 
Knowledge has increased somewhat but—-without utilization of CDR-—the 
world is still likely on a path of building up the concentration of 
CO2 in the atmosphere. This brings the prospect of higher temperatures 
which imply, in the far future, a sea level rise and the possible 
consequences in Scenario I. 

Figure: Line chart: 

[Refer to PDF for image: line chart] 

CO2 build-up plotted against Past to future. 

[End of figure] 

* Resources wasted on geoengineering if no future warming: Risks of 
having wasted efforts and expenditures would be moderate. The 
financial losses and efforts in a federal research program designed 
specifically to combat warming could be somewhat offset if some new 
technologies can be used to address ocean acidification or
to develop spin-off technologies to apply in other areas. 

Scenario IV: Major action: 

By 2015 or soon thereafter, significantly stronger climate-change data 
and models will have reduced uncertainty, deepened understanding, and 
validated earlier scientific conclusions. Also during this half 
decade, several unprecedented, highly disruptive, and extreme weather 
events will affect a number of nations (including the United States), 
causing mass deaths, migration, and devastating property damage. In a 
jarring development, one nation unilaterally stages a major real-world 
test of a fast-acting geoengineering technology in a remote area—
without first warning other nations. The test’s negative effects are 
limited, but there is a step jump increase in global recognition of 
the need for coordination and cooperation. 

In the United States, the balance of public opinion tips toward 
favoring an aggressive lowering of climate risks. Taking a leadership 
role, U.S. envoys help achieve a global agreement on relatively 
aggressive carbon constraints (equivalent to a carbon price of $30 per 
ton of CO2). The global carbon constraints create a worldwide 
incentive for the private sector to pursue mitigation strategies, such 
as alternative fuels and renewables, as well as geoengineering 
approaches like scalable, direct air capture and sequestration. A new 
presidential-congressional initiative establishes an aggressive, 
innovative, and adaptive geoengineering research program that cuts 
across multiple agencies. It includes strong international cooperation 
and other goals similar to those in the failed legislation outlined in 
Scenario I (points 1–4). Additionally, this initiative emphasizes 
adaptation, research innovation, and public engagement. 

In part because of this research program, new developments in areas 
such as synthetic biology and nanotechnology are applied to 
geoengineering (and to other areas such as energy production and 
conservation), resulting in a number of potentially game-changing 
breakthroughs. The new U.S. initiative sets in motion a range of 
programs and policies to ensure that new technologies will have 
opportunities to (1) transform energy sectors and help lower future 
emissions in the United States and around the globe, (2) reduce 
existing and continuing build-up of CO2 through air capture (because 
emissions reduction will not be complete), and (3) improve the U.S. 
economic and export profile. Measures to spur dissemination of new 
technologies include, for example, working with states and regions to 
develop targeted sector or regional plans, as well as international 
coordination. 

Additionally, the research program includes evaluations of side 
effects; analyses of economic, legal, and social implications; and 
analyses of tradeoffs and systemic risk—to help inform policymakers 
and the interested public. Overall, the program’s public engagement 
feature and its effectuation of economic gains and international 
cooperation help sustain support for this initiative through 2030. 

Risks across three potential developments (2030 and later years): 

* Climate emergency/response: Immediate decision risks are low to 
moderate. By 2030, U.S. decision-makers and the global community would 
have information that helps prepare them for responding to a climate 
emergency. Additionally, there would be international mechanisms in 
place to support global cooperation, and thus help avoid conflicts. 

Figure: Anticipated future risks: 

[Refer to PDF for image: vertical bar graph] 

Emergency (red bar): Low to moderate; 
Warming (orange bar): Low to moderate; 
Waste (gray bar): Moderate to high. 

[End of figure] 

* Continued warming: Future global climate risks would be low to 
moderate. By 2030, the world is on a path toward eventual 
stabilization and subsequent reduction of CO2 build-up—hence, less 
warming. Although some sea level rise may occur, the overall prospects 
for negative consequences in the far future would be substantially 
reduced relative to Scenarios I–III. 

Figure: Line chart: 

[Refer to PDF for image: line chart] 

CO2 build-up plotted against Past to future. 

[End of figure] 

* Resources wasted on geoengineering if no future warming: Risks would 
be moderate to high. In this scenario, very large investments (in 
terms of both financial resources and efforts that might have been 
used in other ways) would have been made, and unrecoverable losses 
could be significant. As in Scenarios II and III, if discoveries and 
technologies developed as a result of geoengineering research were 
able to be used in other ways, losses could be mitigated—for example, 
by helping to reduce ocean acidification. In the longer term,
some of the geoengineering technologies developed to combat warming 
might be used instead to help avoid other adverse affects that might 
be associated with extremely high concentrations of CO2. 

8.4. The six external experts who participated in building the 
scenarios: 

Cannizzaro, Christopher, Physical Science Officer/AAAS Science and 
Technology Policy Fellow, Office of Space and Advanced Technology 
(OES/SAT), U.S. Department of State, Washington, D.C. 

Gallaudet, Tim, Deputy Director, Navy’s Task Force Climate Change, 
Office of the Oceanographer of the Navy, Chief of Naval Operations 
Staff, Washington, D.C. 

Lackner, Klaus, Department Chair, Ewing and J. Lamar Worzel Professor 
of Geophysics, Earth and Environmental Engineering; Director, Lenfest 
Center for Sustainable Energy, The Earth Institute, Columbia 
University, New York. 

Patrinos, Aristides A. N., President, Synthetic Genomics, La Jolla, 
California and Washington, D.C. 

Rasch, Philip, Chief Scientist for Climate Science and Laboratory 
Fellow, Pacific Northwest National Laboratory, Richland, Washington. 

Rejeski, David, Director, Science and Technology Innovation Program, 
and Director, Project on Emerging Nanotechnologies, Woodrow Wilson 
International Center for Scholars, Washington, D.C. 

8.5. Experts who commented in response to the scenarios: 

Barrett, Scott, Lenfest-Earth Institute Professor of Natural Resource 
Economics, School of International and Public Affairs and Earth 
Institute, Columbia University, New York. 

Beck, Robert A., Executive Vice President and Chief Operating Officer, 
National Coal Council Inc., Washington, D.C. 

Bronson, Diana, Programme Manager and Researcher, ETC Group, Ottawa,
Ontario, Canada. 

Bunzl, Martin, Professor, Department of Philosophy, Rutgers 
University; Director, Rutgers Initiative on Climate and Social Policy, 
Rutgers University, New Brunswick, New Jersey. 

Carlson, Rob, Principal, Biodesic, Seattle, Washington. 

Cascio, Jamais, Senior Fellow, Institute for Emerging Ethics and 
Technologies, Hartford, Connecticut; Director of Impacts Analysis, 
Center for Responsible Nanotechnology, Menlo Park, California; 
Research Fellow, Institute for the Future, Palo Alto, California. 

Christy, John R., Distinguished Professor of Atmospheric Science and 
Director, Earth System Science Center, University of Alabama, 
Huntsville, Alabama; Alabama State Climatologist, The Alabama Office 
of the State Climatologist, Huntsville, Alabama. 

Fetter, Steve, Assistant Director At-Large, U.S. Office of Science and 
Technology Policy, Executive Office of the President of the United 
States, Washington, D.C. 

Fleming, James R., Professor and Director of Science, Technology, and 
Society Program, Colby College, Waterville, Maine. 

Hamilton, Clive, Professor of Public Ethics, Centre for Applied 
Philosophy and Public Ethics, a joint center of the Australian 
National University, Charles Sturt University, and the University of 
Melbourne, Melbourne, Australia; Vice-Chancellor’s Chair, Charles 
Sturt University, Sydney, Australia. 

Hawkins, David G., Director of Climate Programs, Natural Resources 
Defense Council, New York, New York. 

Hsing, Helen, Managing Director, Strategic Planning and External 
Liaison, U.S. Government Accountability Office, Washington, D.C. 

Johnson, Jean, Executive Vice President, Director of Education 
Insights and Director of Programs, Public Agenda, New York, New York. 

Khosla, Vinod, Partner, Khosla Ventures, Menlo Park, California. 

Lane, Lee, Visiting Fellow, Hudson Institute, Washington, D.C. 

Lomborg, Bjørn, Director, Copenhagen Consensus Center, Denmark. 

Long, Jane C. S., Associate Director, Energy and Environment 
Directorate, Lawrence Livermore National Laboratory, Livermore, 
California. 

MacCracken, Michael, Chief Scientist for Climate Change Programs, 
Climate Institute, Washington, D.C. 

Maynard, Andrew D., Director, University of Michigan Risk Science 
Center, and Professor, Environmental Health Sciences, School of
Public Health, University of Michigan, Ann Arbor, Michigan. 

Olson, Robert L., Senior Fellow, Institute for Alternative Futures, 
Alexandria, Virginia. 

Robock, Alan, Distinguished Professor (Professor II), Department of 
Environmental Sciences; Associate Director, Center for Environmental 
Prediction; Director, Meteorology Undergraduate Program; Member, 
Graduate Program in Atmospheric Science, Rutgers University, New 
Brunswick, New Jersey. 

Sarewitz, Daniel, Co-Director, Consortium for Science, Policy & 
Outcomes; Associate Director, Center for Nanotechnology in Society; 
Professor of Science and Society, College of Liberal Arts and 
Sciences; and Professor, School of Life Sciences and School of 
Sustainability, Arizona State University, Tempe, Arizona. 

Schneider, John P., Deputy Director for Research, Earth System 
Research Laboratory, National Oceanic and Atmospheric Administration, 
Boulder, Colorado. 

Seidel, Stephen, Vice President for Policy Analysis and General 
Counsel, Pew Center on Global Climate Change, Arlington, Virginia. 

Suarez, Pablo, Associate Director of Programmes, Red Cross/Red 
Crescent Climate Centre, The Hague, The Netherlands; Visiting Fellow, 
Frederick S. Pardee Center for the Study of the Longer-Range Future, 
Boston University, Boston, Massachusetts. 

Victor, David G., Professor, International School of International 
Relations and Pacific Studies; Director, Laboratory on International 
Law and Regulation, University of California at San Diego, San Diego. 

Wiener, Jonathan B., William R. and Thomas L. Perkins Professor of Law 
and Director, JD-LLM Program in International and Comparative Law, 
Duke Law School, Durham, North Carolina; Professor of Environmental 
Policy, Nicholas School of the Environment and Professor of Public 
Policy, Sanford School of Public Policy, Duke University, Durham, 
N.C.; Fellow, Resources for the Future, Wash., D.C. 

Wilcoxen, Peter J., Director, Center for Environmental Policy and 
Administration, and Associate Professor, Economics and Public 
Administration, The Maxwell School, Syracuse University, Syracuse, New 
York; Nonresident Senior Fellow, Economic Studies, The Brookings 
Institution, Washington, D.C. 

8.6. Experts who participated in our meeting on climate engineering: 

The 11 of these experts whose names are starred (*) both (1) commented 
on the future of climate engineering during the Meeting and (2) had 
not previously participated in either constructing the scenarios or 
commenting on them. We discussed the views of these experts in section 
4 of this report (Experts’ Views of the Future of Climate Engineering 
Research). John Latham was scheduled to attend this meeting but was 
unable to be there and provided written comments instead. 

*Berg, Robert J., Trustee, World Academy of Art and Science, 
Pittsburgh, Pennsylvania; Senior Advisor, World Federation of United 
Nations Associations, New York, New York. 

Bunzl, Martin, Professor, Department of Philosophy, Rutgers 
University; Director, Rutgers Initiative on Climate and Social Policy, 
Rutgers University, New Brunswick, New Jersey. 

*Duren, Riley, Chief Systems Engineer, Earth Science and Technology 
Directorate, Jet Propulsion Laboratory, California Institute of 
Technology, Pasadena, California. 

Espinal, Laura, Materials Scientist, Ceramics Division, Functional 
Properties Group, Material Measurement Laboratory, National Institute 
of Standards and Technology, Gaithersburg, Maryland. 

Fetter, Steve, Assistant Director At-Large, U.S. Office of Science and 
Technology Policy, Executive Office of the President of the United 
States, Washington, D.C. 

*Fraser, Gerald T., Chief, Optical Technology Division, Physical 
Measurement Laboratory, National Institute of Standards and 
Technology, Gaithersburg, Maryland. 

*Hunter, Kenneth W., Senior Fellow, Institute for Global Chinese 
Affairs; Senior Fellow, Joint Institute for Food Safety and Applied 
Nutrition, University of Maryland, College Park, Maryland. 

*Janetos, Anthony, Director, Joint Global Change Research Institute, 
Pacific Northwest National Laboratory and University of Maryland, 
College Park, Maryland. 

Johnson, Jean, Executive Vice President, Director of Education 
Insights and Director of Programs, Public Agenda, New York, New York.
*LaPorte, Todd R., Professor Emeritus of Political Science, University 
of California, Berkeley, California. 

*Lehmann, Christopher Johannes, Associate Professor, Department of 
Crop and Soil Sciences, Cornell University, Ithaca, New York. 

Long, Jane C. S., Associate Director, Energy and Environment 
Directorate, Lawrence Livermore National Laboratory, Livermore, 
California. 

MacCracken, Michael, Chief Scientist for Climate Change Programs, 
Climate Institute, Washington, D.C. 

*MacDonald, Alexander E. “Sandy,” Director, Earth System Research 
Laboratory; and Deputy Assistant Administrator for Laboratories and 
Cooperative Institutes, Office of Oceanic and Atmospheric Research, 
National Oceanic and Atmospheric Administration, Boulder, Colorado. 

Rejeski, David, Director, Science and Technology Innovation Program, 
and Director, Project on Emerging Nanotechnologies, Woodrow Wilson 
International Center for Scholars, Washington, D.C. 

*St. John, Courtney, Climate Change Affairs Officer, Navy’s Task Force 
Climate Change, Office of the Oceanographer of the Navy, Chief of 
Naval Operations Staff, Washington, D.C. 

Thernstrom, Samuel, Senior Policy Advisor, Geoengineering Task Force, 
Bipartisan Policy Center, Washington, D.C.; Senior Climate Policy 
Advisor, Clean Air Task Force, Boston, Massachusetts. 

*Tilmes, Simone, Project Scientist I, Chemistry Climate Liaison, 
Atmospheric Chemistry Division, National Center for Atmospheric 
Research, Boulder, Colorado. 

Tombari, John, President, Schlumberger Carbon Services, Houston, Texas.
*Toon, Owen B., Chair and Professor, Department of Atmospheric and 
Oceanic Sciences; Research Associate, Laboratory for Atmospheric and 
Space Physics; Director, Toon Aerosol Research Group, University of 
Colorado, Boulder. 

8.7. Reviewers of the report draft: 

Duren, Riley, Chief Systems Engineer, Earth Science and Technology 
Directorate, Jet Propulsion Laboratory, California Institute of 
Technology, Pasadena, California. 

Espinal, Laura, Materials Scientist, Ceramics Division, Functional 
Properties Group, Material Measurement Laboratory, National Institute 
of Standards and Technology, Gaithersburg, Maryland. 

Fraser, Gerald T., Chief, Optical Technology Division, Physical 
Measurement Laboratory, National Institute of Standards and 
Technology, Gaithersburg, Maryland. 

Hunter, Kenneth W., Senior Fellow, Institute for Global Chinese 
Affairs; Senior Fellow, Joint Institute for Food Safety and Applied 
Nutrition, University of Maryland, College Park, Maryland. 

Janetos, Anthony, Director, Joint Global Change Research Institute, 
Pacific Northwest National Laboratory, and University of Maryland, 
College Park, Maryland. 

Johnson, Jean, Executive Vice President, Director of Education 
Insights and Director of Programs, Public Agenda, New York, New York. 

LaPorte, Todd R., Professor Emeritus of Political Science, University 
of California, Berkeley. 

Latham, John, Emeritus Professor of Physics, University of Manchester, 
UK; Visiting Professor, University of Leeds, UK; Senior Research 
Associate, National Center for Atmospheric Research, Boulder, Colorado. 

Lehmann, Christopher Johannes, Associate Professor, Department of Crop 
and Soil Sciences, Cornell University, Ithaca, New York. 

MacCracken, Michael, Chief Scientist for Climate Change Programs, 
Climate Institute, Washington, D.C. 

MacDonald, Alexander E. “Sandy,” Director, Earth System Research 
Laboratory, and Deputy Assistant Administrator for Laboratories and 
Cooperative Institutes, Office of Oceanic and Atmospheric Research, 
National Oceanic and Atmospheric Administration, Boulder, Colorado. 

St. John, Courtney, Climate Change Affairs Officer, Navy’s Task Force 
Climate Change, Office of the Oceanographer of the Navy, Chief of 
Naval Operations Staff, Washington, D.C. 

Thernstrom, Samuel, Senior Policy Advisor, Geoengineering Task Force, 
Bipartisan Policy Center, Washington, D.C.; Senior Climate Policy 
Advisor, Clean Air Task Force, Boston, Massachusetts. 

Tilmes, Simone, Project Scientist I, Chemistry Climate Liaison, 
Atmospheric Chemistry Division, National Center for Atmospheric 
Research, Boulder, Colorado. 

Toon, Owen B., Chair and Professor, Department of Atmospheric and 
Oceanic Sciences; Research Associate, Laboratory for Atmospheric and 
Space Physics; Director, Toon Aerosol Research Group, University of 
Colorado, Boulder. 

[End of section] 

9. References: 

Akbari, Hashem, Surabi Menon, and Arthur Rosenfeld. 2009. “Global 
Cooling: Increasing World-Wide Urban Albedos to Offset CO2.” Climatic 
Change 94 (3–4): 275–86. Angel, Roger. 2006. “Feasibility of Cooling 
the Earth with a Cloud of Small Spacecraft Near the Inner Lagrange 
Point (L1).” Proceedings of the National Academy of Sciences of the 
USA 103 (46): 17184–89. 

Asilomar Scientific Organizing Committee. 2010. The Asilomar 
Conference Recommendations on Principles for Research into Climate 
Engineering Techniques: Conference Report. Washington, D.C.: Climate 
Institute, November. 

Azar, Christian, Kristian Lindgren, Eric Larson, and Kenneth 
Möllersten. 2006. “Carbon Capture and Storage from Fossil Fuels and 
Biomass—Costs and Potential Role in Stabilizing the Atmosphere.” 
Climatic Change 74 (1–3): 47–79. 

Backlund, P., A. Janetos, D. Schimel, J. Hatfield, K. Boote, P. Fay, 
L. Hahn, C. Izaurralde, B. A. Kimball, T. Mader, J. Morgan, D. Ort, W. 
Polley, A. Thomson, D. Wolfe, M. G. Ryan, S. R. Archer, R. Birdsey, C. 
Dahm, L. Heath, J. Hicke, D. Hollinger, T. Huxman, G. Okin, R. Oren, 
J. Randerson, W. Schlesinger, D. Lettenmaier, D. Major, L. Poff, S. 
Running, L. Hansen, D. Inouye, B. P. Kelly, L. Meyerson, B. Peterson, 
and R. Shaw. 2008. The Effects of Climate Change on Agriculture, Land 
Resources, Water Resources, and Biodiversity in the United States. 
Report by the U.S. Climate Change Science Program and the Subcommittee 
on Global Change Research. Climate Synthesis and Assessment Product 
4.3. Washington, D.C.: U.S. Department of Agriculture, May. 

Bader, David C., Curt Covey, William J. Gutowski Jr., Isaac M. Held, 
Kenneth E. Kunkel, Ronald L. Miller, Robin T. Tokmakian, and Minghua 
H. Zhang. 2008. Climate Models: An Assessment of Strengths and 
Limitations. Synthesis and Assessment Product 3.1. Report
by the U.S. Climate Change Science Program and the Subcommittee on 
Global Change Research. Washington, D.C.: U.S. Department of Energy, 
Office of Biological and Environmental Research, July. 

Baird, Colin. 1998. Environmental Chemistry. 2nd ed. New York: W. H. 
Freeman and Company. 

Bala, G., Ken Caldeira, Rama Nemani, Long Cao, George Ban-Weiss, and 
Ho-Jeong Shin. 2010. “Albedo Enhancement of Marine Clouds to 
Counteract Global Warming: Impacts on the Hydrological Cycle.” Climate 
Dynamics, Online First™, June 23. [hyperlink, 
www.springerlink.com/content/9569172415150486]. 

Barrett, Scott. 2008. “The Incredible Economics of Geoengineering.” 
Environmental and Resource Economics 39 (1): 45–54. 

Barrett, Scott. 1998. “Political Economy of the Kyoto Protocol.” 
Oxford Review of Economic Policy 14 (4): 20–39. 

Bertram, Christine. 2009. “Ocean Iron Fertilization: An Option for 
Mitigating Climate Change?” Kiel Policy Brief No. 3. Kiel, Germany: 
Kiel Institute for the World Economy, March. 

Bittle, Scott, Jonathan Rochkind, and Amber Ott. 2009. The Energy 
Learning Curve: Coming from Different Starting Points, the Public Sees 
Similar Solutions. New York: Public Agenda. 

Boden, T. A., G. Marland, and R. J. Andres. 2010. Global, Regional, 
and National Fossil-Fuel CO2 Emissions. Carbon Dioxide Information 
Analysis Center, Oak Ridge National Laboratory. Oak Ridge, Tenn.: U.S.
Department of Energy. 

Boyd, Philip W. 2008. “Introduction and Synthesis” (Theme Section: 
Implications of Large-Scale Iron Fertilization of the Oceans). Marine 
Ecology Progress Series 364: 213–18, July 29. 

Buesseler, Ken O., Scott C. Doney, David M. Karl, Philip W. Boyd, Ken 
Caldeira, Fei Chai, Kenneth H. Coale, Hein J. W. de Baar, Paul G. 
Falkowski, Kenneth S. Johnson, Richard S. Lampitt, Anthony F. 
Michaels, S. W. A. Naqvi, Victor Smetacek, Shigenobu Takeda, and 
Andrew J. Watson. 2008a. “Ocean Iron Fertilization—Moving Forward in a 
Sea of Uncertainty.” Science 319 (5860): 161. 

Buesseler, Ken, Scott Doney, and Hauke Kite-Powell, eds. 2008b. 
“Should We Fertilize the Ocean to Reduce Greenhouse Gases?”
Oceanus 46 (1). 

Caldeira, Ken, and Lowell Wood. 2008. “Global and Arctic Climate 
Engineering: Numerical Model Studies.” Philosophical Transactions of 
the Royal Society A 366: 4039–56. 

Canadell, Josep G., and Michael R. Raupach. 2008. “Managing Forests 
for Climate Change Mitigation.” Science 320 (5882): 1456–57. 

Carbo, Michiel C., Ruben Smith, Bram van der Drift, and Daniel Jansen. 
2010. “Opportunities for BioSNG Production with CCS.” 1st 
International Workshop on Biomass & CCS, Orléans, France, October 14–
15. 

CMI (Carbon Mitigation Initiative). 2010. Carbon Mitigation 
Initiative: Annual Report 2009. Princeton, N.J.: Princeton University. 

CNA Corporation. 2007. National Security and the Threat of Climate 
Change. Alexandria, Va.: The CNA Corporation. 

DOE (U.S. Department of Energy). 2006. U.S. Climate Change Technology 
Program Strategic Plan. DOE/PI-0005. Washington, D.C.: U.S. Climate 
Change Technology Program, September. 

Dooley, J. J., and C. L. Davidson. 2010. A Brief Technical Critique of 
Ehlig-Economides and Economides 2010 “Sequestering Carbon Dioxide in a 
Closed Underground Volume.” PNNL 19249. Richland, Wash.: Pacific 
Northwest National Laboratory, April. 

Dyer, Gwynne. 2010. Climate Wars: The Fight for Survival as the World 
Overheats. Oxford, Eng.: Oneworld Publications. 

Early, James T. 1989. “Space-Based Solar Shield to Offset Greenhouse 
Effect.” Journal of the British Interplanetary Society 42: 567–69.
Ehlig-Economides, Christine, and Michael J. Economides. 2010. 
“Sequestering Carbon Dioxide in a Closed Underground Volume.” Journal 
of Petroleum Science and Engineering 70 (1–2): 123–30. 

Eilperin, Juliet. 2010. “Maldives’ Fight on Carbon Status Is One of 
Survival.” Washington Post, October 11, p. A5. 

Field, Christopher B., Jorge Sarmiento, and Burke Hales. 2007. “The 
Carbon Cycle of North America in a Global Context.” In The First State 
of the Carbon Cycle Report (SOCCR): The North American Carbon Budget 
and Implications for the Global Carbon Cycle, Anthony W. King, Lisa 
Dilling, Gregory P. Zimmerman, David M. Fairman, Richard A. Houghton, 
Gregg Marland, Adam Z. Rose, and Thomas J. Wilbanks, eds., pp. 21–28. 
Synthesis and Assessment Product 2.2. Report by the U.S. Climate 
Change Science Program and the Subcommittee on Global Change Research. 
Asheville, N.C.: NOAA, National Climatic Data Center, November. 

Fleming, James Rodger. 2010. Fixing the Sky: The Checkered History of 
Weather and Climate Control. New York: Columbia University Press. 

Fraser, G. T., S. W. Brown, R. U. Datla, B. C. Johnson, K. R. Lykke, 
and J. P. Rice. 2008. “Measurement Science for Climate Remote Sensing.”
In Earth Observing Systems XIII, James J. Butler and Jack Xiong, eds. 
Proceedings of SPIE 7801, 708102 (Proceedings of SPIE Optics and 
Photonics Conference 2008). 

GAO (U.S. Government Accountability Office). 2010a. Climate Change: A 
Coordinated Strategy Could Focus Federal Geoengineering Research and 
Inform Governance Efforts. [hyperlink, 
http://www.gao.gov/products/GAO-10-903]. Washington, D.C.: September 
23. 

GAO (U.S. Government Accountability Office). 2010b. Climate Change: 
Preliminary Observations on Geoengineering Science, Federal Efforts, 
and Governance Issues. Testimony before the Committee on Science and 
Technology, U.S. House of Representatives, 111th Cong. [hyperlink, 
http://www.gao.gov/products/GAO-10-546T]. Washington, D.C.: March 18. 

GAO (U.S. Government Accountability Office). 2010c. Coal Power Plants: 
Opportunities Exist for DOE to Provide Better Information on the 
Maturity of Key Technologies to Reduce Carbon Dioxide Emissions. 
[hyperlink, http://www.gao.gov/products/GAO-10-675]. Washington, D.C.: 
June 16. 

GAO (U.S. Government Accountability Office). 2009a. Climate Change 
Adaptation: Strategic Federal Planning Could Help Government Officials 
Make More Informed Decisions. [hyperlink, 
http://www.gao.gov/products/GAO-10-113]. Washington, D.C.: October 7. 

GAO (U.S. Government Accountability Office). 2009b. GAO Cost 
Estimating and Assessment Guide: Best Practices for Developing and 
Managing Capital Program Costs. [hyperlink, 
http://www.gao.gov/products/GAO-09-3SP]. Washington, D.C.: March. 

GAO (U.S. Government Accountability Office). 2009c. International 
Space Station: Significant Challenges May Limit Onboard Research.
[hyperlink, http://www.gao.gov/products/GAO-10-9]. Washington, D.C.: 
November 25. 

GAO (U.S. Government Accountability Office). 2008a. Climate Change: 
Expert Opinion on the Economics of Policy Options to Address Climate 
Change. [hyperlink, http://www.gao.gov/products/GAO-08-605]. 
Washington, D.C.: May 9. 

GAO (U.S. Government Accountability Office). 2008b. Highway Safety: 
Foresight Issues Challenge DOT’s Efforts to Assess and Respond to New 
Technology-Based Trends. [hyperlink, 
http://www.gao.gov/products/GAO-09-56]. Washington, D.C.: October 3. 

GAO (U.S. Government Accountability Office). 1999. Best Practices: 
Better Management of Technology Development Can Improve Weapon System 
Outcomes. [hyperlink, http://www.gao.gov/products/GAO/NSIAD-99-162]. 
Washington, D.C.: July 30. 

GAO (U.S. Government Accountability Office). 1995. Global Warming: 
Limitations of General Circulation Models and Costs of Modeling
Efforts. [hyperlink, http://www.gao.gov/products/GAO/RCED-95-164]. 
Washington, D.C.: July 13. 

Gaskill, Alvia. 2004. DOE Meeting Summary: Summary of Meeting with 
U.S. DOE to Discuss Geoengineering Options to Prevent Abrupt Long-Term 
Climate Change. Held at the U.S. Department of Energy, Washington, 
D.C., June 16, 2004. [hyperlink, 
http://www.global-warming-geo-engineering.org/3/contents.html] 

Gaskill, Alvia, ed. n.d. Global Warming Albedo Enhancement. 
Sedlescombe East Sussex, Eng.: Library4Science. [hyperlink, 
http://www.library4science.com] (2000–11). 

Gordon, Bart. 2010. Engineering the Climate: Research Needs and 
Strategies for International Coordination. Report by the Chairman, 
Committee on Science and Technology, U.S. House of Representatives, 
111th Cong., 2nd Sess. Washington, D.C.: October. 

Gornall, Jemma, Richard Betts, Eleanor Burke, Robin Clark, Joanne 
Camp, Kate Willett, and Andrew Wiltshire. 2010. “Implications of 
Climate Change for Agricultural Productivity in the Early Twenty-First 
Century,” Philosophical Transactions of the Royal Society B 365: 2973–
89. 

Govindasamy, B., S. Thompson, P. B. Duffy, K. Caldeira, and C. Delire. 
2002. “Impact of Geoengineering Schemes on the Terrestrial Biosphere.” 
Geophysical Research Letters 29 (22): 2061. 

Govindasamy, Bala, and Ken Caldeira. 2000. “Geoengineering Earth’s 
Radiation Balance to Mitigate CO2-induced Climate Change.” Geophysical 
Research Letters 27 (14): 2141– 44. 

Graham, John D., and Jonathan Baert Wiener, eds. 1995. Risk versus 
Risk: Tradeoffs in Protecting Health and the Environment. Cambridge, 
Mass.: Harvard University Press. 

Grimstad, Alv-Arne, Sorin Georgescu, Erik Lindeberg, and Jean-Francois 
Vuillaume. 2009. “Modelling and Simulation of Mechanisms for Leakage 
of CO2 from Geological Storage.” Proceedings of the 9th International 
Conference on Greenhouse Gas Control Technologies (GHGT-9), 
Washington, D.C., November 16–20, 2008. Energy Procedia 1 (1): 2511-
2518. 

Hamwey, Robert M. 2007. “Active Amplification of the Terrestrial 
Albedo to Mitigate Climate Change: An Exploratory Study.” Mitigation 
and Adaptation Strategies for Global Change 12 (4): 419–39. 

Hansen, J., R. Ruedy, M. Sato, and K. Lo. 2010. “Global Surface 
Temperature Change.” Reviews of Geophysics, 48, RG4004: 1–29. 

Hao, Yue, Tom Wolery, and Susan Carroll. 2009. Preliminary Simulations 
of CO2 Transport in the Dolostone Formations in the Ordos Basin, 
China, LLNL-TR-412701. Report prepared under the auspices of the U.S. 
Department of Energy by Lawrence Livermore National Laboratory under 
Contract DE-AC52-07NA27344. Livermore, Calif.: Lawrence Livermore 
National Laboratory, May 1. 

Heckendorn, P., D. Weisenstein, S. Fueglistaler, B. P. Luo, E. 
Rozanov, M. Schraner, L. W. Thomason, and T. Peter. 2009. “The Impact 
of Geoengineering Aerosols on Stratospheric Temperature and Ozone.” 
Environmental Research Letters 4 (4):1-12. 

Herzog, Howard. 2003. Assessing the Feasibility of Capturing CO2 from 
the Air. Massachusetts Institute of Technology, Laboratory for Energy 
and the Environment, Pub.No. MIT LFEE 2003-02 WP. Cambridge, Mass.: 
Massachusetts Institute of Technology, October. 

Houghton, J. T., Y. Ding, D. J. Griggs, M. Noguer, P. J. van der 
Linden, X. Dai, K. Maskell, and C. A. Johnson, eds. 2001. Climate 
Change 2001: The Scientific Basis. Contribution of Working Group I to 
the Third Assessment Report of the Intergovernmental Panel on Climate 
Change. New York and Cambridge, Eng.: Cambridge University Press. 

Houghton, R. A., E. A. Davidson, and G. M. Woodwell. 1998. “Missing 
Sinks, Feedbacks, and Understanding the Role of Terrestrial Ecosystems 
in the Global Carbon Balance.” Global Biogeochemical Cycles 12 (1): 25–
34. Ipsos MORI. 2010. Experiment Earth? Report on a Public Dialogue on 
Geoengineering. London, Eng.: Ipsos MORI, August. 

Jepma, Catrinus J. 2008. “‘Biosphere Carbon Stock Management: 
Addressing the Threat of Abrupt Climate Change in the Next Few Decades.’
By Peter Read. An Editorial Comment.” Climatic Change 87 (3-4): 343–46. 

Keeling, C. D. 1960. “The Concentration and Isotopic Abundances of 
Carbon Dioxide in the Atmosphere,” Tellus 12 (2): 200–03. 

Keeling, C. D., S. C. Piper, R. B. Bacastow, M. Wahlen, T. P. Whorf, 
M. Heimann, and H. A. Meijer. 2001. Exchanges of Atmospheric CO2 and 
13CO2 with the Terrestrial Biosphere and Oceans from 1978 to 2000. I. 
Global Aspects, SIO Reference Series, No. 01-06. San Diego, Calif.: 
Scripps Institution of Oceanography. 

Keeling, R. F., S. C. Piper, A. F. Bollenbacher, and J. S. Walker. 
2009. “Atmospheric CO2 Records from Sites in the SIO Air Sampling 
Network.” In Trends: A Compendium of Data on Global Change. Carbon 
Dioxide Information Analysis Center, Oak Ridge National Laboratory. 
Oak Ridge, Tenn.: U.S. Department of Energy. 

Keeling, Ralph. 2011. Personal correspondence to Ana Ivelisse Aviles, 
Senior General Engineer, U.S. Government Accountability Office, 
Washington, D.C., February 1. 

Keith, David W. 2010. “Photophoretic Levitation of Engineered Aerosols 
for Geoengineering.” Proceedings of the National Academy of Sciences 
of the USA 107 (38): 16428–31. 

Kiehl, J. T., and Kevin E. Trenberth. 1997. “Earth’s Annual Global 
Mean Energy Budget.” Bulletin of the American Meteorological
Society 78 (2): 197–208. 

Kravitz, Ben, Alan Robock, Olivier Boucher, Hauke Schmidt, Karl E. 
Taylor, Georgiy Stenchikov, and Michael Schulz. 2011. “The 
Geoengineering Model Intercomparison Project (GeoMIP).” Atmospheric 
Science Letters 12 (2): 162–67. 

Lacis, Andrew A., Gavin A. Schmidt, David Rind, and Reto A. Ruedy. 
2010. “Atmospheric CO2: Principal Control Knob Governing Earth’s 
Temperature,” Science 330 (6002): 356–59. 

Laird, David A., Robert C. Brown, James E. Amonette, and Johannes 
Lehmann. 2009. “Review of the Pyrolysis Platform for Coproducing Bio-
Oil and Biochar.” Biofuels, Bioproducts and Biorefining 3 (5): 547–562. 

Lane, Lee, Ken Caldiera, Robert Chatfield, and Stephanie Langhoff, 
eds. 2007. Workshop Report on Managing Solar Radiation. Report of a 
workshop jointly sponsored by NASA/Ames Research Center and the 
Carnegie Institution of Washington Department of Global Ecology held 
at Ames Research Center, Moffett Field, California, November 18-19, 
2006. NASA/CP-2007-214558. Hanover, Md.: NASA Center for Aerospace 
Information, April. 

Latham, John, Philip Rasch, Chih-Chieh Chen, Laura Kettles, Alan 
Gadian, Andrew Gettelman, Hugh Morrison, Keith Bower, and Tom 
Choularton. 2008. “Global Temperature Stabilization via Controlled 
Albedo Enhancement of Low-Level Maritime Clouds.” Philosophical 
Transactions of the Royal Society A 366: 3969–87. 

Lehmann, Johannes. 2007. “A Handful of Carbon,” Nature 447 (7141): 143–
44. 

Lehmann, Johannes, John Gaunt, and Marco Rondon. 2006. “Bio-char 
Sequestration in Terrestrial Ecosystems—A Review.” Mitigation and 
Adaptation Strategies for Global Change 11 (2): 403-427. 

Leiserowitz, Anthony, Nicholas Smith, and Jennifer R. Marlon. 2010. 
Americans’ Knowledge of Climate Change. Yale University. New Haven, 
Conn.: Yale Project on Climate Change Communication. 

Lenton, T. M., and N. E. Vaughan. 2009. “The Radiative Forcing 
Potential of Different Climate Geoengineering Options.” Atmospheric 
Chemistry and Physics 9 (15): 5539–61. 

Levinson, Ronnen, and Hashem Akbari. 2010. “Potential Benefits of Cool 
Roofs on Commercial Buildings: Conserving Energy, Saving Money, and 
Reducing Emission of Greenhouse Gases and Air Pollutants.” Energy 
Efficiency 3 (1): 53–109. 

Lindzen, Richard S. 2010. “Global Warming: How to Approach the 
Science. (Climate Models and the Evidence?).” Testimony before the 
Subcommittee on Energy and the Environment, Committee on Science and 
Technology, U.S. House of Representatives, 111th Cong. Washington, 
D.C., November 17. 

Lindzen, Richard S., and Yong-Sang Choi. 2009. “On the Determination 
of Climate Feedbacks from ERBE Data.” Geophysical Research Letters 36, 
L16705. 

Long, Jane C. S. 2010. “Testimony to the House Science Committee: 
Geoengineering III Hearings.” Testimony before the Committee on 
Science and Technology, U.S. House of Representatives, 111th Cong. 
Washington, D.C., March 18. 

McCarl, Bruce A., Cordner Peacocke, Ray Chrisman, Chih-Chun Kung, and 
Ronald D. Sands. 2009. “Economics of Biochar Production, Utilization 
and Greenhouse Gas Offsets,” pp. 341–56 in Biochar for Environmental 
Management: Science and Technology, Johannes Lehmann and Stephen 
Joseph, eds. London, Eng.: Earthscan. 

McGuffie, K., and A. Henderson-Sellers. 2001. “Forty Years of Numerical 
Climate Modelling.” International Journal of Climatology 21 (9): 1067–109. 

McInnes, C. R. 2002. “Minimum Mass Solar Shield for Terrestrial 
Climate Control.” Journal of the British Interplanetary Society 55: 
307–11. 

MacMinn, Christopher W., and Ruben Juanes. 2009. “A Mathematical Model 
of the Footprint of the CO2 Plume during and after Injection in Deep 
Saline Aquifer Systems.” Proceedings of the 9th International 
Conference on Greenhouse Gas Control Technologies (GHGT-9), 
Washington, D.C., November 16–20, 2008. Energy Procedia 1 (1): 3429–36. 

Maibach, Edward, Connie Roser-Renouf, and Anthony Leiserowitz. 2009. 
Global Warming’s Six Americas 2009: An Audience Segmentation Analysis. 
Fairfax, Va.: George Mason University, Center for Climate Change 
Communication, June 19. 

Meehl, Gerald A., and Kathy Hibbard. 2007. A Summary Report: A 
Strategy for Climate Change Stabilization Experiments with AOGCMs and 
ESMs. WCRP Informal Report 3/2007, ICPO Pub. 112, and IGBP Report 57. 
Aspen Global Change Institute 2006 Session, Earth System Models: The 
Next Generation Aspen, Colorado, July 30–August 5, 2006 and Joint 
WGCM/Aimes Steering Committee Meeting September 27, 2006. 

Metz, Bert, Ogunlade Davidson, Heleen de Coninck, Manuela Loos, and 
Leo Meyer, eds. 2005. IPCC Special Report on Carbon Dioxide Capture 
and Storage: Summary for Policymakers and Technical Summary. Prepared 
by Working Group III of the Intergovernmental Panel on Climate Change. 
New York and Cambridge, Eng.: Cambridge University Press. 

MITRE. 2011. Risk Management Toolkit. McLean, Va. [hyperlink, 
http://www.mitre.org] (Our Work, Mission Areas, Enterprise Systems 
Engineering, Systems Engineering Practice Office, Toolkits, Risk 
Management. Visited January 24.) 

Morgan, M. Granger, and Katharine Ricke. 2010. Cooling the Earth 
through Solar Radiation Management: The Need for Research and an 
Approach to Its Governance. An Opinion Piece for International Risk 
Governance Council. Geneva, Switzerland: International Risk Governance 
Council. 

Murphy, Daniel M. 2009. “Effect of Stratospheric Aerosols on Direct 
Sunlight and Implications for Concentrating Solar Power.” 
Environmental Science and Technology 43 (8): 2784–86. 

Murray, Brian C., Brent Sohngen, Allan J. Sommer, Brooks Depro, Kelly 
Jones, Bruce McCarl, Dhazn Gillig, Benjamin DeAngelo, and Kenneth 
Andrasko. 2005. Greenhouse Gas Mitigation Potential in U.S. Forestry 
and Agriculture. Office of Atmospheric Programs (6207J). EPA 430-R-05-
006. Washington, D.C.: U.S. Environmental Protection Agency, November. 

Nabuurs, Gert Jan, Omar Masera, Kenneth Andrasko, Pablo Benitez-Ponce, 
Rizaldi Boer, Michael Dutschke, Elnour Elsiddig, Justin Ford-
Robertson, Peter Frumhoff, Timo Karjalainen, Olga Krankina, Werner A. 
Kurz, Mitsuo Matsumoto, Walter Oyhantcabal, N. H. Ravindranath, Maria 
José Sanz Sanchez, Xiaquan Zhang. 2007. “Forestry,” pp. 543–84, in 
Climate Change 2007: Mitigation, B. Metz, O. R. Davidson, P. R. Bosch, 
R. Dave, and L. A. Meyer, eds. Contribution of Working Group III to 
the Fourth Assessment Report of the Intergovernmental Panel on Climate 
Change. New York and Cambridge, Eng.: Cambridge University Press. 

NAS (National Academy of Sciences, Panel on Policy Implications of 
Greenhouse Warming). 1992. Policy Implications of Greenhouse Warming: 
Mitigation, Adaptation, and the Science Base. National Academy of 
Sciences, National Academy of Engineering, and Institute of Medicine. 
Washington, D.C.: National Academy Press. 

Nisbet, Matthew C., and Teresa Myers. 2007. “The Polls—Trends: Twenty 
Years of Public Opinion about Global Warming.” Public Opinion 
Quarterly 71 (3): 444–70. 

Nolte, William L. 2004. AFRL Hardware and Software Transition 
Readiness Level Calculator, Version 2.2. Air Force Research 
Laboratory, Wright-Patterson Air Force Base, Ohio. [hyperlink, 
http://www.acq.osd.mil/jctd/TRL/TRL%20Calc%20Ver%202_2.xls] 

NRC (National Research Council). 2010a. Advancing the Science of 
Climate Change. Washington, D.C.: National Academies Press.
 
NRC (National Research Council). 2010b. Ocean Acidification: A 
National Strategy to Meet the Challenges of a Changing Ocean. 
Washington, D.C.: National Academies Press. 

NRC (National Research Council). 2007. Earth Science and Applications 
from Space: National Imperatives for the Next Decade and Beyond. 
Washington, D.C.: National Academies Press. 

Ohring, George, ed. 2007. Achieving Satellite Instrument Calibration 
for Climate Change (ASIC3). Report of a Workshop Organized by National 
Oceanic and Atmospheric Administration, National Institute of 
Standards and Technology, National Aeronautics and Space 
Administration, National Polar-Orbiting Operational Environmental 
Satellite System-Integrated Program Office, and Space Dynamics 
Laboratory of Utah State University, Lansdowne, Virginia, May 16–18, 
2006. Washington, D.C.: National Oceanic and Atmospheric 
Administration. 

Olson, Robert L. Forthcoming. Geoengineering for Decision Makers. 
Washington, D.C.: Woodrow Wilson International Center for Scholars. 

Oruganti, YagnaDeepika, and Steven L. Bryant. 2009. “Pressure Build-Up 
during CO2 Storage in Partially Confined Aquifers.” Proceedings of the 
9th International Conference on Greenhouse Gas Control Technologies 
(GHGT-9). Washington, D.C., November 16–20, 2008. Energy Procedia 1 
(1): 3315–22. 

OSTP (Office of Science and Technology Policy). 2010. Achieving and 
Sustaining Earth Observations: A Preliminary Plan Based on a Strategic 
Assessment by the U.S. Group on Earth Observations. Washington, D.C.: 
Office of Science and Technology Policy, September. 

Parson, Edward A. 2006. “Reflections on Air Capture: The Political 
Economy of Active Intervention in the Global Environment: An Editorial 
Comment.” Climatic Change 74 (1–3): 5–15. 

Pearson, Jerome, John Oldson, and Eugene Levin. 2006. “Earth Rings for 
Planetary Environment Control.” Acta Astronautica 58 (1): 44–57. 

Pew (Pew Research Center for the People & the Press). 2010. “Public 
Remains of Two Minds on Energy Policy.” Congressional Connection. Pew 
Research Center/National Journal Poll sponsored by the Society for 
Human Resource Management. Washington, D.C.: June 14. 

Pielke Jr., Roger A. 2009. “An Idealized Assessment of the Economics 
of Air Capture of Carbon Dioxide in Mitigation Policy.” Environmental 
Science & Policy 12 (3): 216–25. 

Popp, David. 2006. “ENTICE-BR: The Effects of Backstop Technology R&D 
on Climate Policy Models.” Energy Economics 28 (2): 188–222.
Popper, Steven W., Robert J. Lempert, and Steven C. Bankes. 2005. 
“Shaping the Future.” Scientific American 292 (4): 66–71. 

Ranjan, Manya. 2010. “Feasibility of Air Capture.” M.S. thesis, 
Engineering Systems Division, Massachusetts Institute of Technology, 
Cambridge, Massachusetts. 

Ranjan, Manya, and Howard J. Herzog. 2010. Feasibility of Air Capture. 
GHGT-10. Energy Procedia. Elsevier Use Only 00 (2010) 000-000. 
[hyperlink, http://sequestration.mit.edu/pdf/GHGT10_Ranjan.pdf] 

Rasch, Philip J., Paul J. Crutzen, and Danielle B. Coleman. 2008. 
“Exploring the Geoengineering of Climate Using Stratospheric Sulfate 
Aerosols: The Role of Particle Size.” Geophysical Research Letters 35, 
L02809. 

Rasch, Philip, John Latham, and Chih-Chieh (Jack) Chen. 2009. 
“Geoengineering by Cloud Seeding: Influence on Sea Ice and Climate 
System.” Environmental Research Letters 4 (4):1-8. 

Rasch, Philip J., Simone Tilmes, Richard P. Turco, Alan Robock, Luke 
Oman, Chih-Chieh (Jack) Chen, Georgiy L. Stenchikov, and Rolando R. 
Garcia. 2008. “An Overview of Geoengineering of Climate Using 
Stratospheric Sulphate Aerosols.” Philosophical Transactions of the 
Royal Society A 366: 4007–37. 

Rau, Greg H. 2009. “Geoengineering via Chemical Enhancement of Ocean 
CO2 Uptake and Storage or Ignore Ocean Chemistry at Our Peril.” 
Response to NAS request for input on geoengineering concepts. Lawrence 
Livermore National Laboratory, Livermore, California: June. 

Rau, Greg H., Kevin G. Knauss, William H. Langer, and Ken Caldeira. 
2007. “Reducing Energy-Related CO2 Emissions Using Accelerated 
Weathering of Limestone.” Energy 32 (8): 1471–77. 

Read, Peter. 2008. “Biosphere Carbon Stock Management: Addressing the 
Threat of Abrupt Climate Change in the Next Few Decades: An Editorial 
Essay.” Climatic Change 87 (3-4): 305–20. 

Read, Peter, and Jonathan Lermit. 2005. “Bio-energy with Carbon 
Storage (BECS): A Sequential Decision Approach to the Threat
of Abrupt Climate Change.” Energy 30 (14): 2654–71. 

Rejeski, David. 2010. “The Molecular Economy.” The Environmental Forum 
27 (1): 36–41. 

Rejeski, David W. 2003. “S&T Challenges in the 21st Century: Strategy 
and Tempo,” pp. 47–57 in AAAS Science and Technology Policy Yearbook 
2003, Albert H. Teich, Stephen D. Nelson, Stephen J. Lita, and Amanda 
Hunt, eds. Washington, D.C.: American Association for the Advancement 
of Science. 

Ridgwell, Andy, Joy S. Singarayer, Alistair M. Hetherington, and Paul 
J. Valdes. 2009. “Tackling Regional Climate Change by Leaf Albedo Bio-
geoengineering.” Current Biology 19 (2): 146–50. 

Roberts, Kelli G., Brent A. Gloy, Stephen Joseph, Norman R. Scott, and 
Johannes Lehmann. 2010. “Life Cycle Assessment of Biochar Systems: 
Estimating the Energetic, Economic, and Climate Change Potential.” 
Environmental Science & Technology 44 (2): 827–33. 

Robock, Alan. 2000. “Volcanic Eruptions and Climate.” Reviews of 
Geophysics 38 (2): 191–219. 

Robock, Alan, Allison Marquardt, Ben Kravitz, and Georgiy Stenchikov. 
2009. “Benefits, Risks, and Costs of Stratospheric Geoengineering.” 
Geophysical Research Letters 36, L19703. 

Robock, Alan, Luke Oman, and Georgiy L. Stenchikov. 2008. “Regional 
Climate Responses to Geoengineering with Tropical and Arctic SO2 
Injections.” Journal of Geophysical Research 113, D16101. 

Royal Society. 2009. Geoengineering the Climate: Science, Governance 
and Uncertainty. London, Eng. 

Royal Society. 2001. The Role of Land Carbon Sinks in Mitigating 
Global Climate Change. Policy document 10/01. Prepared by the Royal 
Society Working Group on Land Carbon Sinks. London, Eng.: July. 

Sabine, Christopher L., Richard A. Feely, Nicolas Gruber, Robert M. 
Key, Kitack Lee, John L. Bullister, Rik Wanninkhof, C. S. Wong, 
Douglas W. R. Wallace, Bronte Tilbrook, Frank J. Millero, Tsung-Hung 
Peng, Alexander Kozyr, Tsueno Ono, and Aida F. Rios. 2004. “The 
Oceanic Sink for Anthropogenic CO2.” Science, New Series 305 (5682): 
367–71. 

Salter, Stephen, Graham Sortino, and John Latham. 2008. “Sea-Going 
Hardware for the Cloud Albedo Method of Reversing Global Warming.” 
Philosophical Transactions of the Royal Society A, 366: 3989–4006.
Sarmiento, Jorge L., and Nicolas Gruber. 2002. “Sinks for 
Anthropogenic Carbon.” Physics Today 55 (8): 30–36. 

Schuiling, R. D., and P. Krijgsman. 2006. “Enhanced Weathering: An 
Effective and Cheap Tool to Sequester CO2.” Climatic Change 74 (1–3): 
349–54. 

Secretariat of the Convention on Biological Diversity. 2009. 
Scientific Synthesis of the Impacts of Ocean Fertilization on Marine 
Biodiversity. CBD Technical Series No. 45. Montreal, Quebec. 

Shetty, Reshma P., Drew Endy, and Thomas F. Knight. 2008. “Engineering 
BioBrick Vectors from BioBrick Parts.” Journal of Biological 
Engineering 2: 5. 

Slingo, Julia, Kevin Bates, Nikos Nikiforakis, Matthew Piggott, 
Malcolm Roberts, Len Shaffrey, Ian Stevens, Pier Luigi Vidale, and 
Hilary Weller. 2009. “Developing the Next-Generation Climate System 
Models: Challenges and Achievements.” Philosophical Transactions of 
the Royal Society A 367: 815–31. 

Sohngen, Brent. 2009. An Analysis of Forestry Carbon Sequestration as 
a Response to Climate Change. Frederiksberg, Denmark: Copenhagen 
Consensus Center, Copenhagen Business School. 

Sohngen, Brent, and Roger Sedjo. 2006. “Carbon Sequestration in Global 
Forests under Different Carbon Price Regimes,” The Energy Journal. 
Multi-Greenhouse Gas Mitigation and Climate Policy Special Issue, 
Special Issue No. 3: 109–26. 

Solomon, Susan. 1999. “Stratospheric Ozone Depletion: A Review of 
Concepts and History.” Reviews of Geophysics 37 (3): 275–316.
Solomon, Susan, Dahe Qin, Martin Manning, Melinda Marquis, Kristen 
Averyt, Melinda M. B. Tignor, Henry LeRoy Miller, Jr., and Zhenlin 
Chen, eds. 2007. Climate Change 2007: The Physical Science Basis. 
Contribution of Working Group I to the Fourth Assessment Report of the 
Intergovernmental Panel on Climate Change. New York and Cambridge, 
Eng.: Cambridge University Press. 

Solomon, Susan, Karen H. Rosenlof, Robert W. Portmann, John S. Daniel, 
Sean M. Davis, Todd J. Sanford, and Gian-Kasper Plattner. 2010. 
“Contributions of Stratospheric Water Vapor to Decadal Changes in the 
Rate of Global Warming.” Science 327 (5970): 1219–23. 

Spencer, T., P. M. Guthrie, J. da Mosto, and C. A. Fletcher. 2005. 
“Introduction: Large-Scale Engineering Solutions to Storm Surge 
Flooding,” pp. 241–43 in Flooding and Environmental Challenges for 
Venice and Its Lagoon: State of Knowledge, C. A. Fletcher and T. 
Spencer, eds. Cambridge, Eng.: Cambridge University Press. 

Stauffer, Philip H., Ronald C. Surdam, Zunsheng Jiao, Terry A. Miller, 
and Ramsey D. Bentley. 2009. “Combining Geologic Data and Numerical 
Modeling to Improve Estimates of the CO2 Sequestration Potential of 
the Rock Springs Uplift, Wyoming.” Proceedings of the 9th 
International Conference on Greenhouse Gas Control Technologies (GHGT-
9), Washington, D.C., November 16–20, 2008. Energy Procedia 1 (1): 
2717–24. 

Sullivan, Charlotte, Scott Frailey, James Damico, Joel Sminchak, 
Charles Gorecki, Kimberly Sams, Brian J. McPherson, David Borns, and 
Christine Doughty. 2011. Risk Analysis and Simulation for Geologic 
Storage of CO2. DOE/NETL-2011/1459. Albany, Ore.: National Energy 
Technology Laboratory, March. 

Taleb, Nassim Nicholas. 2007. The Black Swan: The Impact of the Highly 
Improbable. New York: Random House. 

Tavoni, Massimo, Brent Sohngen, and Valentina Bosetti. 2007. “Forestry 
and the Carbon Market Response to Stabilize Climate,” Energy Policy 35 
(11): 5346–53. 

Teller, E., L. Wood, and R. Hyde. 1997. Global Warming and Ice Ages: 
I. Prospects for Physics-Based Modulation of Global Change. UCRL-JC-
128715. Preprint. Livermore, Calif.: Lawrence Livermore National 
Laboratory, August 15. 

Tilmes, Simone, Rolando R. Garcia, Douglas E. Kinnison, Andrew 
Gettelman, and Philip J. Rasch. 2009. “Impact of Geoengineered 
Aerosols on the Troposphere and Stratosphere.” Journal of Geophysical 
Research 114, D12305. 

Tilmes, Simone, Rolf Müller, and Ross Salawitch. 2008. “The 
Sensitivity of Polar Ozone Depletion to Proposed Geoengineering 
Schemes." Science Express, April 24. 

Trenberth, Kevin E., and Aiguo Dai. 2007. “Effects of Mount Pinatubo 
Volcanic Eruption on the Hydrological Cycle as an Analog of 
Geoengineering.” Geophysical Research Letters 34, L15702. 

Trenberth, Kevin E., and John T. Fasullo. 2010. “Tracking Earth’s 
Energy.” Science 328 (5976): 316–17. 

Twomey, S. 1977. “The Influence of Pollution on the Shortwave Albedo 
of Clouds.” Journal of the Atmospheric Sciences 34 (7): 1149–52. 

U.S. House of Representatives. 2007. Committee on Appropriations. 
Legislative Branch Appropriations Bill, 2008: Report to Accompany H.R. 
2771. H.R. Rep. No. 110-198. Washington, D.C.: U.S. Government 
Printing Office. 

United States Senate. 2007. Committee on Appropriations, Legislative 
Branch Appropriations, 2008, Report to Accompany S. 1686. Rep. No. 110-
89. Washington, D.C.: U.S. Government Printing Office. 

Victor, David G. 2008. “On the Regulation of Geoengineering.” Oxford 
Review of Economic Policy 24 (2): 322–36. 

Washington, Warren M. 2006. “Computer Modeling the Twentieth-and 
Twenty-First-Century Climate.” Proceedings of the American 
Philosophical Society 150 (3): 414–27. 

Whitman, Thea, Sebastian M. Scholz, and Johannes Lehmann. 2010. 
“Biochar Projects for Mitigating Climate Change: An Investigation of 
Critical Methodology Issues for Carbon Accounting.” Carbon Management 
1 (1): 89–107. 

Woolf, Dominic, James E. Amonette, F. Alayne Street-Perrott, Johannes 
Lehmann, and Stephen Joseph. 2010. “Sustainable Biochar to Mitigate 
Global Climate Change.” Nature Communications 1 (56): 1–9. 

Zeman, Frank. 2007. “Energy and Material Balance of CO2 Capture from 
Ambient Air.” Environmental Science & Technology 41 (21): 7558–63. 
Zhang, Yan, Panagiotis Vouzis, and Nick Sahinidis. 2010. “Risk 
Assessment for CO2 Geological Sequestration.” Presented at the 
conference on Energy, Sustainability and Climate Change 2010, Center 
for Applied Optimization, University of Florida. Gainesville, Fla., 
February 26–28. 

[End of section] 

GAO contacts and staff acknowledgments: 

GAO contact: 
Timothy M. Persons, Ph.D., Chief Scientist, at (202) 512-6412 or 
personst@gao.gov. 

Other leadership for this project was provided by: 

Judith A. Droitcour, Ph.D., Assistant Director, Applied Research and 
Methods (ARM), and Ana Ivelisse Aviles, Ph.D., Analyst-in-Charge and 
Senior General Engineer, also in ARM. 

Key contributors: 

Pille Anvelt, Visual Communications Analyst; 
Virginia A. Chanley, Ph.D., Senior Design Methodologist; 
Nirmal Chaudhary, Ph.D., Senior General Engineer; 
F. Kendall Childers, M.S., Senior Physical Scientist; 
Nancy J. Donovan, M.P.A., Senior Analyst; 
Gloria Hernandez-Saunders, Senior Information Technology Specialist; 
Eric M. Larson, Ph.D., Senior Analyst; 
Penny Pickett, Ph.D., Senior Communications Analyst; 
Ardith A. Spence, Ph.D., Senior Economist; 
Gregory H. Wilmoth, Ph.D., Assistant Director; 

In addition, Juanita A. Aiken provided engagement support, and Leah 
Anderson and Katrina E. Pekar-Carpenter contributed as intern and 
student trainee. Other ARM staff made contributions as did staff and 
specialists from GAO’s Natural Resources and Environment team and 
other GAO offices. 

[End of section] 

Related GAO products: 

Climate Change: A Coordinated Strategy Could Focus Federal 
Geoengineering Research and Inform Governance Efforts, [hyperlink, 
http://www.gao.gov/products/GAO-10-903], Washington, D.C.: September 
23, 2010. 

Coal Power Plants: Opportunities Exist for DOE to Provide Better 
Information on the Maturity of Key Technologies to Reduce Carbon 
Dioxide Emissions, [hyperlink, http://www.gao.gov/products/GAO-10-
675], Washington, D.C.: June 16, 2010. 

Climate Change: Preliminary Observations on Geoengineering Science, 
Federal Efforts, and Governance Issues. Testimony before the Committee 
on Science and Technology, U.S. House of Representatives, 111th Cong., 
[hyperlink, http://www.gao.gov/products/GAO-10-546T]. Washington, 
D.C.: March 18, 2010. 

Climate Change Adaptation: Strategic Federal Planning Could Help 
Government Officials Make More Informed Decisions, [hyperlink, 
http://www.gao.gov/products/GAO-10-13], Washington, D.C.: October 7, 
2009. 

Climate Change: Expert Opinion on the Economics of Policy Options to 
Address Climate Change, [hyperlink, 
http://www.gao.gov/products/GAO-08-605], Washington, D.C.: May 9, 2008. 

Global Warming: Limitations of General Circulation Models and Costs of 
Modeling Efforts. [hyperlink, 
http://www.gao.gov/products/GAO/RCED-95-164], Washington, D.C.: July 
13, 1995. 

Other GAO technology assessments: 

Technology Assessment: 
Explosives Detection Technology to Protect Passenger Rail, 
[hyperlink, http://www.gao.gov/products/GAO-10-898], Washington, D.C.: 
July 28, 2010. 

Technology Assessment: 
Securing the Transport of Cargo Containers, GAO-06-68SU, 
Washington, D.C., January 25, 2006 [Classification: For Official Use 
Only]. 

Technology Assessment: 
Protecting Structures and Improving Communications during Wildland 
Fires, [hyperlink, http://www.gao.gov/products/GAO-05-380], 
Washington, D.C.: April 26, 2005. 

Technology Assessment: 
Cybersecurity for Critical Infrastructure Protection. [hyperlink, 
http://www.gao.gov/products/GAO-04-321], Washington, D.C.: May 28, 
2004. 

Technology Assessment: 
Using Biometrics for Border Security, [hyperlink, 
http://www.gao.gov/products/GAO-03-174], Washington, D.C.: November 
14, 2002. 

[End of section] 

Footnotes: 

[1] The request for this assessment was originally made during the 
111th Congress by the Chairman of the House Committee on Science and 
Technology, who has since retired. 

[2] Fossil fuels, such as coal, oil, and methane, are natural organic 
compounds of mostly carbon (C) and hydrogen (H). These fuels are 
formed from dead plant and animal matter that has been subjected to 
intense pressure and heat over geologic time scales. 

[3] Compound annual growth rate calculated from available emissions 
estimates. 

[4] Over time, atmospheric CO2 can be reabsorbed as sediment on the 
ocean floor through the carbon cycle. 

[5] In the atmosphere, greenhouse gases absorb and reemit radiation 
within the thermal infrared range of the electromagnetic spectrum. 
This is the fundamental cause of the greenhouse effect, or the warming 
of Earth’s atmosphere. In order of their prevalence by volume, the 
primary greenhouse gases are water vapor (H2O), CO2, methane (CH4), 
nitrous oxide (N2O), and ozone (O3) (Baird 1998). 

[6] Water vapor (H2O) is the most abundant greenhouse gas and has a 
powerful effect on warming (Solomon et al. 2007; Kiehl and Trenberth 
1997). Scientists have shown that the tropospheric water vapor 
concentration significantly affects the global average surface 
temperature. The enhanced greenhouse effect caused by emissions from 
human activities is sometimes called anthropogenic climate change. The 
increased concentration of CO2 is also known to be the leading cause 
of another major environmental concern in addition to warming: ocean 
acidification, manifested by decreases in pH (hydrogen ion 
concentration), is caused by the oceans’ greater uptake of atmospheric 
CO2 as its abundance increases (Sabine et al. 2004). 

[7] Multiple, interrelated systems can influence the enhanced 
greenhouse effect. 

[8] These scientists argue that all current climate prediction models 
incorrectly project more warming, based on positive feedback from 
water vapor and clouds. Specifically, they argue that such feedback 
has a negative effect (Lindzen and Choi 2009). The Intergovernmental 
Panel on Climate Change (IPCC) has also noted the uncertainty 
surrounding such feedback (Solomon et al. 2007). 

[9] This report is an assessment of technologies to engineer the 
climate and the quality of information available to assess these 
technologies. In this report, we did not assess whether the climate is 
changing or what is causing any climate change that is occurring or 
whether current scientific knowledge supports the notion that the 
climate is changing or its causes. We did not assess whether climate 
change is or will be sufficient to warrant using these technologies. 

[10] Mohamed Nasheed, President of the Maldives, has said that sea 
level rise is already causing coastal erosion in his country, 
evidenced by salt intrusion in the water table and relocations 
affecting 16 islands (Eilperin 2010). 

[11] For example, climate change might lead to greater scarcity of 
food, water, or shelter and social upheaval in many countries in 
Africa, Asia, and the Middle East (CNA Corporation 2007, 44). Some 
have suggested that disrupted food and water supplies in certain 
regions might lead to mass migrations or international conflict (Dyer 
2010). 

[12] We use the term "geoengineering" in appropriate contexts, as when 
it refers to information we collected in a survey of U.S. adults and 
their attitudes toward technologies to address climate change. We 
described the alternative terms gates in Venice, Italy (Spencer et al. 
2005). "climate engineering," "climate remediation," and "climate 
However, neither mitigation nor adaptation intervention" in a 
September 2010 report (GAO 2010a, 3). 

[13] Although experts differ on which technologies to define as 
climate engineering (Gordon 2010, ii), in this report we limited our 
assessment to key climate engineering technologies among those 
reviewed by the Royal Society (Royal Society 2009). 

[14] Because SRM would not affect the atmospheric concentration of 
CO2, it would not abate increased ocean acidification. 

[15] In the Senate report accompanying the proposed bill for the 
legislative branch fiscal ear 2008 appropriation, the Senate Committee 
on Appropriations recommended the establishment of a permanent 
technology assessment function within GAO (United States Senate 2007, 
see S. Rep. No. 110-89, at 42–43 (2007)). The House Committee on 
Appropriations, in providing funding to GAO to perform technology 
assessment studies, noted that “it is necessary for the Congress to 
equip itself with effective means for securing competent, timely 
and unbiased information concerning the effects of scientific and 
technical developments and use the information in the legislative 
assessment of matters pending before the Congress” (U.S. House of 
Representatives 2007, see H.R. Rep. No. 110-198, at 30 (2007)). 
GAO established a permanent operational technology assessment group 
within its Applied Research and Methods team: the Center for Science, 
Technology, and Engineering. GAO defines technology assessment as 
the thorough and balanced analysis of significant primary, secondary, 
indirect, and delayed interactions of a technological innovation with 
society, the environment, and the economy and the present and foreseen 
consequences and effects of those interactions. 

[16] Energy is also transferred mechanically (not by radiation) from 
Earth’s surface to the atmosphere and clouds by evaporation and 
convection. 

[17] The Stefan-Boltzmann law, named after Jožef Stefan and Ludwig 
Boltzmann, states that the total power radiated per unit of surface 
area of a black body per unit of time is directly proportional to the 
fourth power of the black body’s thermodynamic temperature T. The 
Stefan–Boltzmann constant?is equal to 5.6704 x 10–8 watts per square 
meter per absolute temperature measured in Kelvin to the fourth power 
(W/m2/K4). 

[18] We used the AFRL Technology Readiness Level Calculator to assess 
maturity (see section 8.1). For a rating of TRL 2 or higher, the basic 
requirement is a system concept on a global scale; for a rating of TRL 
3 or higher, analytical and experimental demonstration of proof of 
concept is required, and for a rating of TRL 4 or higher, system 
demonstration with a breadboard unit is required. These requirements 
apply regardless of a technology’s scientific basis or the extent to 
which the techniques it incorporates are well established. 

[19] In 2010, the atmospheric concentration of CO2 was about 390 ppm; 
around the year 1750, it was about 280 ppm. 

[20] Sorbent refers to a solution or solid that selectively absorbs a 
specific gas. 

[21] Geologic sequestration of CO2 is a relatively new idea. 

[22] Psi indicates pounds per square inch. A megapascal is 1 million 
pascals; a pascal is a measure of force per unit area, defined as 1 
newton per square meter. A newton is the force that produces an 
acceleration of 1 meter per second per second when exerted on a mass 
of 1 kilogram. Atmospheric pressure at sea level is 14.7 psi, or 
roughly 0.1 MPa. 

[23] Thermodynamic efficiency refers to the ratio of the thermodynamic 
minimum energy requirement to the actual amount of energy used in the 
process (Zeman 2007). 

[24] Conservative estimates of the potential to store CO2 emissions 
geologically in North America range from 3,300 to 12,600 gigatons-—
that is, enough to store the CO2 output of several coal-fired power 
plants for many decades. 

[25] Well logging is the process of measuring and recording the rock 
and fluid properties of geologic formations through drilled boreholes. 
It is common in the oil and gas industry for helping to find potential 
reservoirs, as well for gathering data to support geotechnical 
studies. Resistivity is a characteristic electrical property of 
materials defined as the electrical resistance of a conductor of unit 
cross-sectional area and unit length. 

[26] The Department of Energy and NETL lead the federal agencies in 
supporting carbon capture and sequestration (CCS) research and field 
demonstrations. The coal-fired Mountaineer Power Plant, run by 
American Electric Power in West Virginia, has conducted a one-of-a-
kind small-scale CCS demonstration that integrated CO2 capture from 
the flue stack, injecting the CO2 into an underground formation at the 
plant site. Also, the Sleipner project, run by Statoil of Norway, 
sequesters approximately 1 megaton of CO2 per year in a deep saline 
aquifer. 

[27] Direct air capture of CO2 is expected to cost more than CO2 
capture from the flue stack of a coal-fired power plant where CO2 
concentration is substantially higher (Ranjan and Herzog 2010). 
Engineers from American Electric Power indicated that the present cost 
of capturing CO2 from a flue stack is estimated at about $50 per ton 
in contrast to the likely high cost of direct air capture. 

[28] These estimates apply only to the energy costs of the process. 
Adding capital and operations costs would increase them significantly 
(Ranjan 2010). 

[29] In geology, a fault is a planar fracture or discontinuity in a 
volume of rock, across which displacement has been significant. Large 
faults within Earth’s crust result from the action of tectonic forces. 

[30] A variant of direct air capture, CCS captures CO2 from a fixed 
location such as the effluent stream of a coal-fired power plant. The 
large technical and scientific literature on CCS has brought it to the 
attention of government agencies, electric power generation 
corporations, and the enhanced oil recovery community (GAO 2010c; GAO 
2008a). We excluded CCS from our analysis because it is not generally 
considered to involve deliberate modification of Earth’s climate 
system and was therefore beyond our scope. As a forerunner of direct 
air capture, CCS is a key part of the bioenergy with CO2 sequestration 
(BECS) method, which, at large scale, is considered to be climate 
engineering. 

[31] Pyrolysis refers to the thermochemical decomposition of organic 
material at elevated temperatures in the absence of oxygen or where 
its supply is limited. 

[32] The term CO2–C equivalent describes the extent of global warming 
caused by a given type and amount of greenhouse gas, using the 
functionally equivalent amount or concentration of CO2 as the 
reference. 

[33] China has recently accomplished afforestation on a large scale 
for reasons unrelated to global climate change mitigation. 

[34] Emissions pricing can provide financial incentives for carbon 
sequestration. This range of estimates of the global economic 
potential of land-use management assumes a price of $100 per ton of 
CO2 sequestered. 

[35] Nabuurs and colleagues described trade-offs that could affect net 
sequestration from land-use management. For example, a moratorium on 
timber harvesting could increase the carbon sequestered in forests but 
could also result in the substitution of energy-intensive building 
materials, such as cement or concrete, for wood in the construction of 
buildings (Nabuurs et al. 2007). 

[36] One expert noted that natural disturbances might not 
significantly challenge carbon sequestration through land-use 
management in the long term. 

[37] Enhanced weathering of silicate and carbonate rocks can be 
represented by CaSiO3 + 2CO2 + H2O yields Ca2++ 2HCO3 + SiO2 and CaCO3 
+ CO2 + H2O yields Ca2++ 2HCO3. 

[38] One proposal would spread crushed olivine, a type of silicate 
rock, on agricultural and forested lands to sequester CO2 and improve 
soil quality (Schuiling and Krijgsman 2006). Another proposal would 
cause the CO2 emissions from a power plant to react with crushed 
limestone (mainly calcium carbonate) in the presence of seawater to 
spontaneously produce calcium bicarbonate ions (Rau et al. 2007). 

[39] This represents a substantially large storage capacity compared 
to the total cumulative anthropogenic carbon additions to oceans of 
about 100 gigatons since preindustrial times. 

[40] One ton of carbon corresponds to 3.67 tons of CO2. 

[41] Despite the fact that oceans exchange large quantities of CO2 
with the atmosphere in a natural process, comparatively little is 
known about sequestering CO2 by ocean fertilization. 

[42] Anoxia means the absence of oxygen. Algal blooms in the ocean can 
deplete available oxygen in the water, leading to dead or anoxic zones. 

[43] While the published research has focused on sulfate aerosols 
(Royal Society 2009), other aerosols such as alumina (Teller et al. 
1997) and self-levitated nanoparticles (Keith 2010) have also been 
considered. 

[44] The total rough cost estimate for the cloud brightening system 
would be $2.4 billion to $4.8 billion. This cost estimate is made up 
of Salter, Sortino, and Latham’s estimates of $3.1 million for the 
first 2 years of engineering; $39 million for the next 3 years for 
final design, including construction of a prototype; $47 million 
for production tooling; and production costs of $2.3 billion to 
$4.7 billion ($1.56 million to $3.13 million each) for 1,500 45-meter, 
300-ton wind-driven spray vessels (Salter et al. 2008). 

[45] In 1992, reflecting 1 percent of solar radiation was thought to 
counteract the global warming from doubling the concentration of CO2 
in the atmosphere (NAS 1992). Up-to-date modeling studies indicate 
that reflection of about 1.8 percent is required (Govindasamy and 
Caldeira 2000; Govindasamy et al. 2002; Caldeira and Wood 2008). 

[46] In this context, plant productivity is net primary productivity, 
which is net carbon uptake by vegetation. 

[47] Gaps or deficiencies in observational networks could also 
interfere with the ability to monitor the effect of deployed climate 
engineering technologies. Monitoring would allow scientists to verify 
the effectiveness of technologies and help ensure their safety. 

[48] One example of a measure of climate sensitivity would be “the 
response of global mean temperature to a doubling of [the atmospheric 
concentration of] carbon dioxide” (Bader et al. 2008, 2). 

[49] A discrepancy in carbon output and uptake by Earth’s systems 
remains unresolved. To preserve mass balance in today’s best estimates 
of the global carbon budget requires including an unknown terrestrial 
carbon sink of about 1.8 billion tons of carbon per year (R. A. 
Houghton et al. 1998). 

[50] The eruption of Mount Pinatubo in 1991 released a large quantity 
of sulfate aerosols into the atmosphere, causing average global 
temperatures to fall. Scientists attending the 2010 Asilomar 
conference said that current observational networks are inadequate to 
collect data following such an eruption that could help improve 
scientific knowledge about atmospheric mechanisms related to aerosol-
based SRM technologies. 

[51] According to NIST scientists, both ground- and space-based 
measurements exhibit these types of variation. The space-based 
variations are largely attributable to calibration inaccuracies that 
can largely be corrected by adjustments using measurements taken 
during satellite overlap. The ground-based variations are both 
geographic and temporal and are likely to include contributions from 
global dimming, urban aerosols, and sensor calibration inaccuracies. 

[52] The President’s fiscal year 2012 budget request for NASA cut 
much of the funding for the CLARREO mission and called for an extended 
preformulation period for the mission and science team to identify 
implementation options for obtaining climate change measurements 
without using CLARREO satellites. 

[53] The network is expected to be fully operational in 2016. 

[54] Grids for atmospheric circulation models have been refined from 
resolving areas the size of Colorado in 1990 to the size of South 
Carolina in 1995, and to about the size of Rhode Island (4,000 km2) in 
2007. Meanwhile, weather models have been run with a resolution of 
less than 1,000 km2 for over 20 years. 

[55] More computer resources can be used for finer numerical grids, 
greater number of runs for statistical estimation, or more climate 
processes; consensus on the optimal resource allocation does not exist. 

[56] Existing models were designed to distinguish long-term climate 
trends (Fraser et al. 2008). 

[57] The advent of GPUs follows on three revolutions in computing 
operations: (1) integrated circuits, (2) vector computing, and (3) 
parallel computing (which made current forecasting possible). GPUs 
complete intense, split-second calculations efficiently to render 
virtual representations of the real world. 

[58] Among the experts we consulted, primary areas of expertise 
spanned two broad categories: (1) physical science or technical 
research related to climate engineering or climate change and (2) 
social science, law, ethics, or other related fields with applications 
in climate engineering or climate change. 

[59] Although we attempted to consult diverse experts representing the 
full range of views on climate engineering, the relative numbers who 
expressed a particular view to us may not reflect the entire community 
of those with similar kinds of expertise. However, for transparency, 
we provide the specific numbers of experts who told us that they 
advocated certain views. We note that not all experts expressed an 
opinion on all issues. 

[60] Two-thirds of the experts we consulted about the future (31 of 
45) advocated starting significant research now or in the very near 
future. 

[61] Four of the 45 experts we consulted about the future stated that 
they opposed research on climate engineering. (One of these 4 made 
exceptions for certain kinds of research, such as computer modeling.) 

[62] Twenty-nine of 45 experts envisioned a federal research effort, 
and as detailed below, 26 of these mentioned one or more of these 
three features. 

[63] Thirty-one of 45 experts said they advocated research now or in 
the near future, and 4 opposed research. The remainder did not clearly 
state whether they advocated starting research now. 

[64] Specifically, we reported that 13 federal agencies had identified 
at least 52 research activities, totaling about $100.9 million, as 
relevant to climate engineering in fiscal years 2009 and 2010 
(GAO 2010a)—$1.9 million for activities to investigate specific 
climate engineering approaches and $99 million related to conventional 
mitigation strategies or basic science that could be applied to 
improving understanding of climate engineering. 

[65] We note, however, that national leaders might not base decisions 
on information about research results, even if such information were 
available. 

[66] This insurance view reflects the hedging strategy described in 
foresight literature whereby, faced with uncertainty, decision-makers 
choose a strategy that they anticipate will work reasonably well 
across all alternatives to avoid potentially disastrous low-
probability outcomes (Popper et al. 2005). 

[67] Overall, of the 31 experts advocating research now, 27 recognized 
risks associated with it, including risks from conducting it (11 
experts) and from using its results (26 experts). One advocate who 
believes that it is urgent to start research now also said that 
guidelines are needed to decide when research “has become too 
dangerous to continue.” 

[68] We reported earlier (GAO 2010b, 13) that in 2008 the parties to 
the London Convention and London Protocol issued a decision stating 
that ocean fertilization that is not legitimate scientific research is 
contrary to the aims of the agreements and should not be allowed. The 
treaties’ scientific bodies are developing an assessment framework for 
countries to use and evaluate whether research proposals are 
legitimate scientific research (GAO 2010a, 33). In 2010, the parties 
to the Convention on Biological Diversity invited countries to 
consider the following guidance: (1) ensure that ocean fertilization 
activities are consistent with the London Convention and Protocol and 
decisions issued by the conference of the parties to those treaties 
and (2) ensure that except for certain small-scale scientific research 
studies, no climate-related geoengineering activity that may affect 
biodiversity take place until there is an adequate scientific basis on 
which to justify it and appropriate consideration of the associated 
risks and impacts. 

[69] Additionally, of the 52 research activities federal agencies 
identified as relevant to geoengineering in our 2010 report, only one 
project’s activity description specifically mentioned risk (GAO 2010a). 

[70] Four of the experts we consulted opposed conducting research. 
Additionally, because of questions some reviewers of a draft of this 
report raised, we note that three of the four opponents of research 
had primary expertise in fields such as social science, law, ethics or 
other related fields (rather than physical science); these three 
provided the direct statements of research opponents that we quote in 
this section. 

[71] These risks (international conflict, drought, and famine) that an 
expert cited as potentially deriving from climate engineering research 
and deployment are similar to those associated with climate change. 
Research opponents also pointed to other possible risks; for example, 
some potential SRM technologies have been associated with the 
depletion of ozone and interference with the use of solar-energy 
technology. 

[72] Of the 31 experts who advocated starting research now, 29 also 
advocated or envisioned a federal research effort; 26 of these 
envisioned one or more of the three specific features discussed in 
this section. Some experts also anticipated the development of 
technologies by the private sector. 

[73] One expert told us that some nations’ research may be hidden 
because it is not specifically labeled as climate engineering or 
because it is covert. 

[74] One approach to international research cooperation is illustrated 
by the International Space Station, with its five main partners: 
Canada, Japan, Russia, the United States, and the European Space 
Agency (which includes a number of countries) (GAO 2009c). 

[75] Twenty-four experts (of 31 who advocated climate engineering 
research now) specifically envisioned an international approach for 
federal research. 

[76] One of our scenarios describes the lessened possibility of 
conflict because nations, having cooperated on research, have a basis 
for cooperating in a sudden crisis. Our 2010 report indicated that 
“several of the experts we interviewed as well as the NRC study 
emphasized the potential for international tension, distrust, or even 
conflict over geoengineering deployment” and discussed international 
agreements and governance challenges (GAO 2010a, 17 and 26–37). 

[77] Twenty-three experts (of 31 who advocated starting research now) 
favored engaging the public or national leaders or both in a federal 
effort. 

[78] Eighteen experts (of 31 who advocated starting research now) 
specifically envisioned an anticipatory, foresight approach for 
federal research. 

[79] A horizon scan is a systematic examination of ongoing trends, 
emerging developments, persistent problems that may have changed, and 
novel and unexpected issues. Horizon scans are sometimes structured 
strategically to consider potential threats and opportunities 
separately. 

[80] For example, Long (2010) has said that an adaptive approach is 
appropriate for climate engineering because climate is a complex, 
nonlinear system. Such an approach might include monitoring the 
results of an intervention, comparing observations to predictions, 
deciding whether the research is proceeding in the right direction, 
and making a new set of decisions about what to do. 

[81] Additionally, 10 experts told us that either analytical 
information on the cost of a potential climate-engineering research 
program is lacking or they did not know of such information. 

[82] An additional $99 million supported these other activities. 

[83] Carbon constraint policies aim to limit or reduce carbon 
emissions. Greenhouse gas emissions pricing is one type of carbon 
constraint that would encourage people to reduce emissions by making 
them more expensive. Despite ongoing debate over climate change 
legislation, the U.S. Congress did not enact legislation in 2010, and 
its prospects are uncertain. 

[84] Knowledge Networks Inc. fielded the survey of a statistically 
representative sample of 1,006 respondents July 19 to August 5, 2010, 
using its online research panel. We used the term “geoengineering” in 
our survey questions and other information we provided about climate 
engineering because we had used the term in earlier work. All 
estimates from the survey are subject to sampling error. In terms of 
the margin of error at the 95 percent confidence level, the sampling 
error for estimates based on the total sample is plus or minus 4 
percentage points and, for estimates based on subgroups of the sample, 
is plus or minus 9 percentage points, unless otherwise noted. Because 
the overall response rate was low and sources of nonsampling error may 
have contributed to total survey error, we rounded survey results to 
the nearest 5 percentage points. We describe our methodology in more 
detail in section 8.1.3. 

[85] The sampling errors for the following demographic subgroup 
estimates, in terms of the margin of error at the 95 percent 
confidence level, are plus or minus 12 percentage points for the 
percentage of those with less than a high school education who believe 
research should be done on geoengineering, plus or minus 13 percentage 
points for the percentage of blacks who believe research should be 
done on geoengineering, and plus or minus 14 percentage points for the 
percentage of Hispanics who believe research should be done on 
geoengineering. The margin of error for the remaining subgroup 
estimates is plus or minus 9 percentage points. 

[86] Because our focus for this report was on public perceptions of 
climate engineering, our survey was not designed to assess public 
views of climate change more broadly. It did, however, ask several 
questions about climate change and energy policy similar to those in 
prior surveys of the U.S. adult population. While comparisons between 
our survey and others’ surveys are not conclusive because of 
historical, methodological, and measurement differences, we found a 
general similarity between the distribution of our results and those 
from other sample surveys. 

[87] We did not vary effectiveness in the split-ballot design. In each 
ballot group, respondents learned about one technology that the Royal 
Society’s 2009 report had identified as highly effective (either 
capturing CO2 from the air or injecting stratospheric aerosols) and 
one that it had identified as relatively less effective (either 
increasing reflection from Earth’s surface or fertilizing the oceans). 
The design did not allow us to determine how experts’ assessments of 
the different technologies’ effectiveness might affect public 
reactions to geoengineering. 

[88] The preindustrial CO2 concentration is reported to have been 280 
ppm. In 2010, the atmospheric concentration of CO2 was estimated in 
the literature as 390 ppm. In the year 2100, the concentration 
projected for a mitigation scenario is 500 ppm. 

[89] The high effectiveness rating the Royal Society gave for these 
two technologies could not be confirmed and validated by reports in 
the literature. We did not assign an overall qualitative rating to 
these technologies because of conflicting indications in the 
literature about their effectiveness. 

[90] The word “equilibrium” indicates a steady state response to 
specify climatic conditions, such as the concentration of CO2 and 
variables related to climate engineering. 

[91] Additionally, in preparing for these activities we interviewed 
other experts who provided background information or recommended some 
of the experts listed in sections 8.4 to 8.6. 

[92] We also conducted two focus groups with science and engineering 
graduate students participating in Arizona State University’s 
Consortium for Science, Policy & Outcomes (CSPO), one before and one 
after the public focus group. We did not make any changes to the focus 
group protocol as a result of the CSPO focus group conducted before 
the public focus group. 

[93] Before we pretested our survey, students in the Science, 
Technology, and Public Policy Program at the Gerald R. Ford School of 
Public Policy, University of Michigan, provided input on issues 
related to governance and surveying public opinion in the area of 
climate engineering. 

[94] The six external experts who participated in building the 
scenarios are listed in section 8.4. Additionally, GAO’s Chief 
Scientist, Timothy Persons, served as host and ex officio member of 
the group. 

[95] The only exception consists of minor corrections to a footnote. 

[96] MITRE (2011) illustrates how qualitative judgments of probability 
and severity may be combined according to risk management literature. 

[97] The twin goals of these efforts are to accelerate mitigation 
efforts (that is, reduce carbon dioxide, or CO2, emissions) and raise 
incentives for private-sector research, including research on carbon 
dioxide removal (CDR) and sequestration. 

[98] Similarly, without adequate information on fast-acting 
technologies, it would be difficult for leaders to decide how to 
respond to a surprise deployment by a single nation, terrorist group, 
or some other “rogue” geoengineering effort. 

[99] The moderate reductions are not at the scale required to 
transform energy or energy-intensive industrial sectors. 

[100] One option, among others, for housing a dedicated research 
program, would be the U.S. Global Climate Change Research Program. 

[101] Also, relative to Scenario I, the increased knowledge might 
better prepare decision-makers for responding to a “rogue” deployment. 

[102] Note, however, that in this scenario, decision-makers who reject 
fast-acting technologies would lack alternative, more gradual 
approaches for dealing with the problem. For example, because CDR 
technologies were “left on the drawing board” rather than being 
further developed and deployed, decision-makers would not have the 
option of ramping up existing direct air capture efforts. Decades 
would be likely to be needed to prepare for such an effort. 

[End of section] 

GAO’s mission: 

The Government Accountability Office, the audit, evaluation, and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the 
performance and accountability of the federal government for the 
American people. GAO examines the use of public funds; evaluates 
federal programs and policies; and provides analyses, recommendations, 
and other assistance to help Congress make informed oversight, policy, 
and funding decisions. GAO’s commitment to good government is 
reflected in its core values of accountability, integrity, and 
reliability. 

To obtain copies of GAO reports and testimony:" 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO’s website [hyperlink, http://www.gao.gov]. Each 
weekday afternoon, GAO posts on its website newly released reports, 
testimony, and correspondence. To have GAO e-mail you a list of newly 
posted products, go to [hyperlink, http://www.gao.gov] and select “E-
mail Updates.” 

To order by phone: 

The price of each GAO publication reflects GAO’s actual cost of 
production and distribution and depends on the number of pages in the 
publication and whether the publication is printed in color or black 
and white. Pricing and ordering information is posted on GAO’s 
website, [hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or 
TDD (202) 512-2537. 

Orders may be paid for with American Express, Discover Card, 
MasterCard, Visa, check, or money order. Call for additional 
information. 

Congressional Relations contact: 

Ralph Dawn, Managing Director, at (202) 512-4400 or dawnr@gao.govU.S. 
Government Accountability Office, 441 G Street NW, Room 7125, 
Washington, DC 20548. 

Public Affairs contact: 

Chuck Young, Managing Director, at (202) 512-4800 or youngc1@gao.gov 
U.S. Government Accountability Office, 441 G Street NW, Room 7149, 
Washington, DC 20548. 

To report fraud, waste, and abuse in federal programs: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470. 

[End of publication]