659
8
This chapter should be cited as:
Myhre, G., D. Shindell, F.-M. Bréon, W. Collins, J. Fuglestvedt, J. Huang, D. Koch, J.-F. Lamarque, D. Lee, B. Mendoza,
T. Nakajima, A. Robock, G. Stephens, T. Takemura and H. Zhang, 2013: Anthropogenic and Natural Radiative Forc-
ing. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen,
J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA.
Coordinating Lead Authors:
Gunnar Myhre (Norway), Drew Shindell (USA)
Lead Authors:
François-Marie Bréon (France), William Collins (UK), Jan Fuglestvedt (Norway), Jianping Huang
(China), Dorothy Koch (USA), Jean-François Lamarque (USA), David Lee (UK), Blanca Mendoza
(Mexico), Teruyuki Nakajima (Japan), Alan Robock (USA), Graeme Stephens (USA), Toshihiko
Takemura (Japan), Hua Zhang (China)
Contributing Authors:
Borgar Aamaas (Norway), Olivier Boucher (France), Stig B. Dalsøren (Norway), John S. Daniel
(USA), Piers Forster (UK), Claire Granier (France), Joanna Haigh (UK), Øivind Hodnebrog
(Norway), Jed O. Kaplan (Switzerland/Belgium/USA), George Marston (UK), Claus J. Nielsen
(Norway), Brian C. O’Neill (USA), Glen P. Peters (Norway), Julia Pongratz (Germany), Michael
Prather (USA), Venkatachalam Ramaswamy (USA), Raphael Roth (Switzerland), Leon Rotstayn
(Australia), Steven J. Smith (USA), David Stevenson (UK), Jean-Paul Vernier (USA), Oliver Wild
(UK), Paul Young (UK)
Review Editors:
Daniel Jacob (USA), A.R. Ravishankara (USA), Keith Shine (UK)
Anthropogenic and Natural
Radiative Forcing
8
660
Table of Contents
Executive Summary ..................................................................... 661
8.1 Radiative Forcing ............................................................ 664
8.1.1 The Radiative Forcing Concept .................................. 664
Box 8.1: Definition of Radiative Forcing and Effective
Radiative Forcing ....................................................................... 665
Box 8.2: Grouping Forcing Compounds by Common
Properties .................................................................................. 668
8.1.2 Calculation of Radiative Forcing due to
Concentration or Emission Changes.......................... 668
8.2 Atmospheric Chemistry ................................................. 669
8.2.1 Introduction .............................................................. 669
8.2.2 Global Chemistry Modelling in Coupled Model
Intercomparison Project Phase 5 ............................... 670
8.2.3 Chemical Processes and Trace Gas Budgets .............. 670
8.3 Present-Day Anthropogenic Radiative Forcing ...... 675
8.3.1 Updated Understanding of the Spectral Properties of
Greenhouse Gases and Radiative Transfer Codes ...... 675
8.3.2 Well-mixed Greenhouse Gases ................................. 676
8.3.3 Ozone and Stratospheric Water Vapour ..................... 679
8.3.4 Aerosols and Cloud Effects ....................................... 682
8.3.5 Land Surface Changes ............................................... 686
8.4 Natural Radiative Forcing Changes: Solar
and Volcanic ...................................................................... 688
8.4.1 Solar Irradiance ......................................................... 688
8.4.2 Volcanic Radiative Forcing ........................................ 691
Box 8.3: Volcanic Eruptions as Analogues .............................. 693
8.5 Synthesis of Global Mean Radiative Forcing,
Past and Future ................................................................ 693
8.5.1 Summary of Radiative Forcing by Species and
Uncertainties ............................................................. 694
8.5.2 Time Evolution of Historical Forcing .......................... 698
8.5.3 Future Radiative Forcing ........................................... 700
8.6 Geographic Distribution of Radiative Forcing ....... 702
8.6.1 Spatial Distribution of Current Radiative Forcing ...... 702
8.6.2 Spatial Evolution of Radiative Forcing and Response
over the Industrial Era ............................................... 705
8.6.3 Spatial Evolution of Radiative Forcing and Response
for the Future ............................................................ 708
8.7 Emission Metrics ............................................................. 710
8.7.1 Metric Concepts ........................................................ 710
Box 8.4: Choices Required When Using Emission Metrics .... 711
8.7.2 Application of Metrics ............................................... 716
References .................................................................................. 721
Appendix 8.A: Lifetimes, Radiative Efficiencies and
Metric Values ................................................................................. 731
Frequently Asked Questions
FAQ 8.1 How Important Is Water Vapour to
Climate Change? .................................................... 666
FAQ 8.2 Do Improvements in Air Quality Have an
Effect on Climate Change? ................................... 684
Supplementary Material
Supplementary Material is available in online versions of the report.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
661
Executive Summary
It is unequivocal that anthropogenic increases in the well-mixed
greenhouse gases (WMGHGs) have substantially enhanced
the greenhouse effect, and the resulting forcing continues to
increase. Aerosols partially offset the forcing of the WMGHGs and
dominate the uncertainty associated with the total anthropogenic
driving of climate change.
As in previous IPCC assessments, AR5 uses the radiative forcing
1
(RF) concept, but it also introduces effective radiative forcing
2
(ERF). The RF concept has been used for many years and in previous
IPCC assessments for evaluating and comparing the strength of the
various mechanisms affecting the Earth’s radiation balance and thus
causing climate change. Whereas in the RF concept all surface and
tropospheric conditions are kept fixed, the ERF calculations presented
here allow all physical variables to respond to perturbations except
for those concerning the ocean and sea ice. The inclusion of these
adjustments makes ERF a better indicator of the eventual temperature
response. ERF and RF values are significantly different for anthropo-
genic aerosols owing to their influence on clouds and on snow cover.
These changes to clouds are rapid adjustments and occur on a time
scale much faster than responses of the ocean (even the upper layer) to
forcing. RF and ERF are estimated over the Industrial Era from 1750 to
2011 if other periods are not explicitly stated. {8.1, Box 8.1, Figure 8.1}
Industrial-Era Anthropogenic Forcing
The total anthropogenic ERF over the Industrial Era is 2.3 (1.1 to
3.3) Wm
–2
.
3
It is certain that the total anthropogenic ERF is positive.
Total anthropogenic ERF has increased more rapidly since 1970 than
during prior decades. The total anthropogenic ERF estimate for 2011 is
43% higher compared to the AR4 RF estimate for the year 2005 owing
to reductions in estimated forcing due to aerosols but also to contin-
ued growth in greenhouse gas RF. {8.5.1, Figures 8.15, 8.16}
Due to increased concentrations, RF from WMGHGs has
increased by 0.20 (0.18 to 0.22) Wm
–2
(8%) since the AR4 esti-
mate for the year 2005. The RF of WMGHG is 2.83 (2.54 to 3.12)
Wm
–2
. The majority of this change since AR4 is due to increases in the
carbon dioxide (CO
2
) RF of nearly 10%. The Industrial Era RF for CO
2
alone is 1.82 (1.63 to 2.01) Wm
–2
, and CO
2
is the component with the
largest global mean RF.
Over the last decade RF of CO
2
has an average
growth rate of 0.27 (0.24 to 0.30) Wm
–2
per decade. Emissions of CO
2
have made the largest contribution to the increased anthropogenic
forcing in every decade since the 1960s. The best estimate for ERF of
WMGHG is the same as the RF but with a larger uncertainty (±20%).
{8.3.2, 8.5.2, Figures 8.6, 8.18}
The net forcing by WMGHGs other than CO
2
shows a small
increase since the AR4 estimate for the year 2005. A small growth
in the CH
4
concentration has increased its RF by 2% to an AR5 value
of 0.48 (0.43 to 0.53) Wm
–2
. RF of nitrous oxide (N
2
O) has increased
by 6% since AR4 and is now 0.17 (0.14 to 0.20) Wm
–2
. N
2
O concen-
trations continue to rise while those of dichlorodifluoromethane (CFC-
12), the third largest WMGHG contributor to RF for several decades, is
falling due to its phase-out under the Montreal Protocol and amend-
ments. Since 2011 N
2
O has become the third largest WMGHG contrib-
utor to RF. The RF from all halocarbons (0.36 Wm
–2
) is very similar to
the value in AR4, with a reduced RF from chlorofluorocarbons (CFCs)
but increases from many of their substitutes. Four of the halocarbons
(trichlorofluoromethane (CFC-11), CFC-12, trichlorotrifluoroethane
(CFC-113) and chlorodifluoromethane (HCFC-22)) account for around
85% of the total halocarbon RF. The first three of these compounds
have declining RF over the last 5 years but their combined decrease
is compensated for by the increased RF from HCFC-22. Since AR4, the
RF from all HFCs has nearly doubled but still only amounts to 0.02
Wm
–2
. There is high confidence
4
that the overall growth rate in RF
from all WMGHG is smaller over the last decade than in the 1970s and
1980s owing to a reduced rate of increase in the combined non-CO
2
RF. {8.3.2; Figure 8.6}
Ozone and stratospheric water vapour contribute substantially
to RF. The total RF estimated from modelled ozone changes is 0.35
(0.15 to 0.55) Wm
–2
, with RF due to tropospheric ozone changes of
0.40 (0.20 to 0.60) Wm
–2
and due to stratospheric ozone changes of
–0.05 (–0.15 to +0.05) Wm
–2
. Ozone is not emitted directly into the
atmosphere but is formed by photochemical reactions. Tropospheric
ozone RF is largely attributed to anthropogenic emissions of methane
(CH
4
), nitrogen oxides (NO
x
), carbon monoxide (CO) and non-methane
volatile organic compounds (NMVOCs), while stratospheric ozone RF
results primarily from ozone depletion by halocarbons. Estimates are
also provided attributing RF to emitted compounds. Ozone-depleting
substances (ODS) cause ozone RF of –0.15 (–0.30 to 0.0) Wm
–2
, some
of which is in the troposphere. Tropospheric ozone precursors cause
ozone RF of 0.50 (0.30 to 0.70) Wm
–2
, some of which is in the strato-
sphere; this value is larger than that in AR4. There is robust evidence
that tropospheric ozone also has a detrimental impact on vegetation
physiology, and therefore on its CO
2
uptake, but there is a low confi-
dence on quantitative estimates of the RF owing to this indirect effect.
RF for stratospheric water vapour produced by CH
4
oxidation is 0.07
(0.02 to 0.12) Wm
–2
. The RF best estimates for ozone and stratospheric
1
Change in net downward radiative flux at the tropopause after allowing for stratospheric temperatures to readjust to radiative equilibrium, while holding surface and tropo-
spheric temperatures and state variables fixed at the unperturbed values.
2
Change in net downward radiative flux at the top of the atmosphere (TOA) after allowing for atmospheric temperatures, water vapour, clouds and land albedo to adjust, but
with global mean surface temperature or ocean and sea ice conditions unchanged (calculations presented in this chapter use the fixed ocean conditions method).
3
Uncertainties are given associated with best estimates of forcing. The uncertainty values represent the 5–95% (90%) confidence range.
4
In this Report, the following summary terms are used to describe the available evidence: limited, medium, or robust; and for the degree of agreement: low, medium, or high.
A level of confidence is expressed using five qualifiers: very low, low, medium, high, and very high, and typeset in italics, e.g., medium confidence. For a given evidence and
agreement statement, different confidence levels can be assigned, but increasing levels of evidence and degrees of agreement are correlated with increasing confidence (see
Section 1.4 and Box TS.1 for more details).
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
662
water vapour are either identical or consistent with the range in AR4.
{8.2, 8.3.3, Figure 8.7}
The magnitude of the aerosol forcing is reduced relative to AR4.
The RF due to aerosol–radiation interactions, sometimes referred to as
direct aerosol effect, is given a best estimate of –0.35 (–0.85 to +0.15)
Wm
–2
, and black carbon (BC) on snow and ice is 0.04 (0.02 to 0.09)
Wm
–2
. The ERF due to aerosol–radiation interactions is –0.45 (–0.95 to
+0.05) Wm
–2
. A total aerosol–cloud interaction
5
is quantified in terms
of the ERF concept with an estimate of –0.45 (–1.2 to 0.0) Wm
–2
. The
total aerosol effect (excluding BC on snow and ice) is estimated as ERF
of –0.9 (–1.9 to –0.1) Wm
–2
. The large uncertainty in aerosol ERF is the
dominant contributor to overall net Industrial Era forcing uncertain-
ty. Since AR4, more aerosol processes have been included in models,
and differences between models and observations persist, resulting in
similar uncertainty in the aerosol forcing as in AR4. Despite the large
uncertainty range, there is a high confidence that aerosols have offset
a substantial portion of WMGHG global mean forcing. {8.3.4, 8.5.1,
Figures 8.15, 8.16}
There is robust evidence that anthropogenic land use change
has increased the land surface albedo, which leads to an RF of
–0.15 ± 0.10 Wm
–2
. There is still a large spread of estimates owing to
different assumptions for the albedo of natural and managed surfaces
and the fraction of land use changes before 1750. Land use change
causes additional modifications that are not radiative, but impact the
surface temperature, in particular through the hydrologic cycle. These
are more uncertain and they are difficult to quantify, but tend to offset
the impact of albedo changes. As a consequence, there is low agree-
ment on the sign of the net change in global mean temperature as a
result of land use change. {8.3.5}
Attributing forcing to emissions provides a more direct link
from human activities to forcing. The RF attributed to methane
emissions is very likely
6
to be much larger (~1.0 W m
–2
) than that
attributed to methane concentration increases (~0.5 Wm
–2
) as concen-
tration changes result from the partially offsetting impact of emissions
of multiple species and subsequent chemical reactions. In addition,
emissions of CO are virtually certain to have had a positive RF, while
emissions of NO
X
are likely to have had a net negative RF at the global
scale. Emissions of ozone-depleting halocarbons are very likely to have
caused a net positive RF as their own positive RF has outweighed the
negative RF from the stratospheric ozone depletion that they have
induced. {8.3.3, 8.5.1, Figure 8.17, FAQ 8.2}
Forcing agents such as aerosols, ozone and land albedo changes
are highly heterogeneous spatially and temporally. These pat-
terns generally track economic development; strong negative aerosol
forcing appeared in eastern North America and Europe during the early
5
The aerosol–cloud interaction represents the portion of rapid adjustments to aerosols initiated by aerosol-cloud interactions, and is defined here as the total aerosol ERF minus
the ERF due to aerosol-radiation-interactions (the latter includes cloud responses to the aerosol–radiation interaction RF)
6
In this Report, the following terms have been used to indicate the assessed likelihood of an outcome or a result: Virtually certain 99–100% probability, Very likely 90–100%,
Likely 66–100%, About as likely as not 33–66%, Unlikely 0–33%, Very unlikely 0–10%, Exceptionally unlikely 0–1%. Additional terms (Extremely likely: 95–100%, More likely
than not >50–100%, and Extremely unlikely 0–5%) may also be used when appropriate. Assessed likelihood is typeset in italics, e.g., very likely (see Section 1.4 and Box TS.1
for more details).
7
Chapter 1 describes the Representative Concentration Pathways (RCPs) that are the primary scenarios discussed in this report.
20th century, extending to Asia, South America and central Africa by
1980. Emission controls have since reduced aerosol pollution in North
America and Europe, but not in much of Asia. Ozone forcing increased
throughout the 20th century, with peak positive amplitudes around
15°N to 30°N due to tropospheric pollution but negative values over
Antarctica due to stratospheric loss late in the century. The pattern
and spatial gradients of forcing affect global and regional temperature
responses as well as other aspects of climate response such as the
hydrologic cycle. {8.6.2, Figure 8.25}
Natural Forcing
Satellite observations of total solar irradiance (TSI) changes
from 1978 to 2011 show that the most recent solar cycle min-
imum was lower than the prior two. This very likely led to a small
negative RF of –0.04 (–0.08 to 0.00) Wm
–2
between 1986 and 2008.
The best estimate of RF due to TSI changes representative for the 1750
to 2011 period is 0.05 (to 0.10) Wm
–2
. This is substantially smaller
than the AR4 estimate due to the addition of the latest solar cycle
and inconsistencies in how solar RF has been estimated in earlier IPCC
assessments. There is very low confidence concerning future solar forc-
ing estimates, but there is high confidence that the TSI RF variations
will be much smaller than the projected increased forcing due to GHG
during the forthcoming decades. {8.4.1, Figures 8.10, 8.11}
The RF of volcanic aerosols is well understood and is greatest
for a short period (~2 years) following volcanic eruptions. There
have been no major volcanic eruptions since Mt Pinatubo in 1991, but
several smaller eruptions have caused a RF for the years 2008–2011 of
–0.11 (–0.15 to –0.08) Wm
–2
as compared to 1750 and –0.06 (–0.08
to –0.04) Wm
–2
as compared to 1999–2002. Emissions of CO
2
from
volcanic eruptions since 1750 have been at least 100 times smaller
than anthropogenic emissions. {8.4.2, 8.5.2, Figures 8.12, 8.13, 8.18}
There is very high confidence that industrial-era natural forcing
is a small fraction of the anthropogenic forcing except for brief
periods following large volcanic eruptions. In particular, robust
evidence from satellite observations of the solar irradiance and volcan-
ic aerosols demonstrates a near-zero (–0.1 to +0.1 Wm
–2
) change in
the natural forcing compared to the anthropogenic ERF increase of 1.0
(0.7 to 1.3) Wm
–2
from 1980 to 2011. The natural forcing over the last
15 years has likely offset a substantial fraction (at least 30%) of the
anthropogenic forcing. {8.5.2; Figures 8.18, 8.19, 8.20}
Future Anthropogenic Forcing and Emission Metrics
Differences in RF between the emission scenarios considered
here
7
are relatively small for year 2030 but become very large by
2100 and are dominated by CO
2
. The scenarios show a substantial
8
Anthropogenic and Natural Radiative Forcing Chapter 8
663
weakening of the negative total aerosol ERF. Nitrate aerosols are an
exception to this reduction, with a substantial increase, which is a
robust feature among the few available models for these scenarios.
The scenarios emphasized in this assessment do not span the range of
future emissions in the literature, however, particularly for near-term
climate forcers. {8.2.2, 8.5.3, Figures 8.2, 8.21, 8.22}
Emission metrics such as Global Warming Potential (GWP) and
Global Temperature change Potential (GTP) can be used to
quantify and communicate the relative and absolute contribu-
tions to climate change of emissions of different substances,
and of emissions from regions/countries or sources/sectors. The
metric that has been used in policies is the GWP, which integrates the
RF of a substance over a chosen time horizon, relative to that of CO
2
.
The GTP is the ratio of change in global mean surface temperature at
a chosen point in time from the substance of interest relative to that
from CO
2
. There are significant uncertainties related to both GWP and
GTP, and the relative uncertainties are larger for GTP. There are also
limitations and inconsistencies related to their treatment of indirect
effects and feedbacks. The values are very dependent on metric type
and time horizon. The choice of metric and time horizon depends on
the particular application and which aspects of climate change are
considered relevant in a given context. Metrics do not define policies
or goals but facilitate evaluation and implementation of multi-com-
ponent policies to meet particular goals. All choices of metric contain
implicit value-related judgements such as type of effect considered
and weighting of effects over time. This assessment provides updated
values of both GWP and GTP for many compounds. {8.7.1, 8.7.2, Table
8.7, Table 8.A.1, Supplementary Material Table 8.SM.16}
Forcing and temperature response can also be attributed to sec-
tors. From this perspective and with the GTP metric, a single year’s
worth of current global emissions from the energy and industrial sec-
tors have the largest contributions to global mean warming over the
next approximately 50 to 100 years. Household fossil fuel and biofuel,
biomass burning and on-road transportation are also relatively large
contributors to warming over these time scales, while current emis-
sions from sectors that emit large amounts of CH
4
(animal husbandry,
waste/landfills and agriculture) are also important over shorter time
horizons (up to 20 years). {8.7.2, Figure 8.34}
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
664
8.1 Radiative Forcing
There are a variety of ways to examine how various drivers contribute
to climate change. In principle, observations of the climate response
to a single factor could directly show the impact of that factor, or cli-
mate models could be used to study the impact of any single factor.
In practice, however, it is usually difficult to find measurements that
are influenced by only a single cause, and it is computationally pro-
hibitive to simulate the response to every individual factor of interest.
Hence various metrics intermediate between cause and effect are used
to provide estimates of the climate impact of individual factors, with
applications both in science and policy. Radiative forcing (RF) is one
of the most widely used metrics, with most other metrics based on RF.
In this chapter, we discuss RF from natural and anthropogenic compo-
nents during the industrial period, presenting values for 2011 relative
to 1750 unless otherwise stated, and projected values through 2100
(see also Annex II). In this section, we present the various definitions
of RF used in this chapter, and discuss the utility and limitations of
RF. These definitions are used in the subsequent sections quantifying
the RF due to specific anthropogenic (Section 8.3) and natural (Sec-
tion 8.4) causes and integrating RF due to all causes (Sections 8.5 and
8.6). Atmospheric chemistry relevant for RF is discussed in Section 8.2
and used throughout the chapter. Emission metrics using RF that are
designed to facilitate rapid evaluation and comparison of the climate
effects of emissions are discussed in Section 8.7.
8.1.1 The Radiative Forcing Concept
RF is the net change in the energy balance of the Earth system due to
some imposed perturbation. It is usually expressed in watts per square
meter averaged over a particular period of time and quantifies the
energy imbalance that occurs when the imposed change takes place.
Though usually difficult to observe, calculated RF provides a simple
quantitative basis for comparing some aspects of the potential climate
response to different imposed agents, especially global mean temper-
ature, and hence is widely used in the scientific community. Forcing is
often presented as the value due to changes between two particular
times, such as pre-industrial to present-day, while its time evolution
provides a more complete picture.
8.1.1.1 Defining Radiative Forcing
Alternative definitions of RF have been developed, each with its own
advantages and limitations. The instantaneous RF refers to an instan-
taneous change in net (down minus up) radiative flux (shortwave plus
longwave; in W m
–2
) due to an imposed change. This forcing is usually
defined in terms of flux changes at the top of the atmosphere (TOA)
or at the climatological tropopause, with the latter being a better indi-
cator of the global mean surface temperature response in cases when
they differ.
Climate change takes place when the system responds in order to
counteract the flux changes, and all such responses are explicitly
excluded from this definition of forcing. The assumed relation between
a sustained RF and the equilibrium global mean surface temperature
response (DT) is DT = lRF where l is the climate sensitivity parameter.
The relationship between RF and DT is an expression of the energy
balance of the climate system and a simple reminder that the steady-
state global mean climate response to a given forcing is determined
both by the forcing and the responses inherent in l.
Implicit in the concept of RF is the proposition that the change in net
irradiance in response to the imposed forcing alone can be separat-
ed from all subsequent responses to the forcing. These are not in fact
always clearly separable and thus some ambiguity exists in what may
be considered a forcing versus what is part of the climate response.
In both the Third Assessment Report (TAR) and AR4, the term radiative
forcing (RF, also called stratospherically adjusted RF, as distinct from
instantaneous RF) was defined as the change in net irradiance at the
tropopause after allowing for stratospheric temperatures to readjust to
radiative equilibrium, while holding surface and tropospheric tempera-
tures and state variables such as water vapour and cloud cover fixed at
the unperturbed values
8
. RF is generally more indicative of the surface
and tropospheric temperature responses than instantaneous RF, espe-
cially for agents such as carbon dioxide (CO
2
) or ozone (O
3
) change
that substantially alter stratospheric temperatures. To be consistent
with TAR and AR4, RF is hereafter taken to mean the stratospherically
adjusted RF.
8.1.1.2 Defining Effective Radiative Forcing
For many forcing agents the RF gives a very useful and appropriate way
to compare the relative importance of their potential climate effect.
Instantaneous RF or RF is not an accurate indicator of the temper-
ature response for all forcing agents, however. Rapid adjustments in
the troposphere can either enhance or reduce the flux perturbations,
leading to substantial differences in the forcing driving long-term cli-
mate change. In much the same way that allowing for the relatively
rapid adjustment of stratospheric temperatures provides a more useful
characterization of the forcing due to stratospheric constituent chang-
es, inclusion of rapid tropospheric adjustments has the potential to
provide more useful characterization for drivers in the troposphere (see
also Section 7.1.3).
Many of the rapid adjustments affect clouds and are not readily includ-
ed into the RF concept. For example, for aerosols, especially absorbing
ones, changes in the temperature distribution above the surface occur
due to a variety of effects, including cloud response to changing atmos-
pheric stability (Hansen et al., 2005; see Section 7.3.4.2) and cloud
absorption effects (Jacobson, 2012), which affect fluxes but are not
strictly part of RF. Similar adjustments take place for many forcings,
including CO
2
(see Section 7.2.5.6).
Aerosols also alter cloud properties via microphysical interactions
leading to indirect forcings (referred to as aerosol–cloud interactions;
8
Tropospheric variables were fixed except for the impact of aerosols on cloud albedo due to changes in droplet size with constant cloud liquid water which was considered an
RF in AR4 but is part of ERF in AR5.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
665
see Section 7.4). Although these adjustments are complex and not fully
quantified, they occur both on the microphysical scale of the cloud
particles as well as on a more macroscopic scale involving whole cloud
systems (e.g., Shine et al., 2003; Penner et al., 2006; Quaas et al., 2009).
A portion of these adjustments occurs over a short period, on cloud life
cycle time scales, and is not part of a feedback arising from the sur-
face temperature changes. Previously these type of adjustments were
sometimes termed ‘fast feedbacks’ (e.g., Gregory et al., 2004; Hansen
et al., 2005), whereas in AR5 they are denoted ‘rapid adjustments’ to
emphasize their distinction from feedbacks involving surface temper-
ature changes. Atmospheric chemistry responses have typically been
included under the RF framework, and hence could also be included in
a forcing encompassing rapid adjustments, which is important when
evaluating forcing attributable to emissions changes (Section 8.1.2)
and in the calculation of emission metrics (Section 8.7).
Studies have demonstrated the utility of including rapid adjustment in
comparison of forcing agents, especially in allowing quantification of
forcing due to aerosol-induced changes in clouds (e.g., effects previ-
ously denoted as cloud lifetime or semi-direct effects; see Figure 7.3)
that are not amenable to characterization by RF (e.g., Rotstayn and
Penner, 2001; Shine et al., 2003; Hansen et al., 2005; Lohmann et al.,
2010; Ban-Weiss et al., 2012). Several measures of forcing have been
introduced that include rapid adjustments. We term a forcing that
accounts for rapid adjustments the effective radiative forcing (ERF).
Conceptually, ERF represents the change in net TOA downward radi-
ative flux after allowing for atmospheric temperatures, water vapour
and clouds to adjust, but with global mean surface temperature or a
portion of surface conditions unchanged. The primary methods in use
for such calculations are (1) fixing sea surface temperatures (SSTs) and
sea ice cover at climatological values while allowing all other parts of
the system to respond until reaching steady state (e.g., Hansen et al.,
2005) or (2) analyzing the transient global mean surface temperature
response to an instantaneous perturbation and using the regression of
the response extrapolated back to the start of the simulation to derive
the initial ERF (Gregory et al., 2004; Gregory and Webb, 2008). The
ERF calculated using the regression technique has an uncertainty of
about 10% (for the 5 to 95% confidence interval) for a single 4 × CO
2
simulation (ERF ~7 W m
–2
) due to internal variability in the transient
climate (Andrews et al., 2012a), while given a similar length simulation
the uncertainty due to internal variability in ERF calculated using the
fixed-SST technique is much smaller and hence the latter may be more
suitable for very small forcings. Analysis of both techniques shows that
the fixed-SST method yields a smaller spread across models, even in
calculations neglecting the uncertainty in the regression fitting proce-
dure (Andrews et al., 2012a). As a portion of land area responses are
included in the fixed-SST technique, however, that ERF is slightly less
than it would be with surface temperature held fixed everywhere. It is
possible to adjust for this in the global mean forcing, though we do not
include such a correction here as we examine regional as well as global
ERF, but the land response will also introduce artificial gradients in
land–sea temperatures that could cause small local climate responses.
In contrast, there is no global mean temperature response included in
the regression method. Despite the low bias in fixed-SST ERF due to
land responses, results from a multi-model analysis of the forcing due
to CO
2
are 7% greater using this method than using the regression
technique (Andrews et al., 2012a) though this is within the uncertainty
range of the calculations. Although each technique has advantages,
forcing diagnosed using the fixed-SST method is available for many
more forcing agents in the current generation of climate models than
forcing diagnosed using the regression method. Hence for practical
purposes, ERF is hereafter used for results from the fixed-SST technique
unless otherwise stated (see also Box 8.1).
The conceptual relation between instantaneous RF, RF and ERF is illus-
trated in Figure 8.1. It implies the adjustments to the instantaneous RF
involve effects of processes that occur more rapidly than the time scale
of the response of the global mean surface temperature to the forcing.
However, there is no a priori time scale defined for adjustments to be
rapid with the fixed-SST method. The majority take place on time scales
Box 8.1 | Definition of Radiative Forcing and Effective Radiative Forcing
The two most commonly used measures of radiative forcing in this chapter are the radiative forcing (RF) and the effective radiative
forcing (ERF). RF is defined, as it was in AR4, as the change in net downward radiative flux at the tropopause after allowing for strato-
spheric temperatures to readjust to radiative equilibrium, while holding surface and tropospheric temperatures and state variables such
as water vapor and cloud cover fixed at the unperturbed values.
ERF is the change in net TOA downward radiative flux after allowing for atmospheric temperatures, water vapour and clouds to adjust,
but with surface temperature or a portion of surface conditions unchanged. Although there are multiple methods to calculate ERF, we
take ERF to mean the method in which sea surface temperatures and sea ice cover are fixed at climatological values unless otherwise
specified. Land surface properties (temperature, snow and ice cover and vegetation) are allowed to adjust in this method. Hence ERF
includes both the effects of the forcing agent itself and the rapid adjustments to that agent (as does RF, though stratospheric tem-
perature is the only adjustment for the latter). In the case of aerosols, the rapid adjustments of clouds encompass effects that have
been referred to as indirect or semi-direct forcings (see Figure 7.3 and Section 7.5), with some of these same cloud responses also
taking place for other forcing agents (see Section 7.2). Calculation of ERF requires longer simulations with more complex models than
calculation of RF, but the inclusion of the additional rapid adjustments makes ERF a better indicator of the eventual global mean tem-
perature response, especially for aerosols. When forcing is attributed to emissions or used for calculation of emission metrics, additional
responses including atmospheric chemistry and the carbon cycle are also included in both RF and ERF (see Section 8.1.2). The general
term forcing is used to refer to both RF and ERF.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
666
Frequently Asked Questions
FAQ 8.1 | How Important Is Water Vapour to Climate Change?
As the largest contributor to the natural greenhouse effect, water vapour plays an essential role in the Earth’s
climate. However, the amount of water vapour in the atmosphere is controlled mostly by air temperature, rather
than by emissions. For that reason, scientists consider it a feedback agent, rather than a forcing to climate change.
Anthropogenic emissions of water vapour through irrigation or power plant cooling have a negligible impact on
the global climate.
Water vapour is the primary greenhouse gas in the Earth’s atmosphere. The contribution of water vapour to the
natural greenhouse effect relative to that of carbon dioxide (CO
2
) depends on the accounting method, but can
be considered to be approximately two to three times greater. Additional water vapour is injected into the atmo-
sphere from anthropogenic activities, mostly through increased evaporation from irrigated crops, but also through
power plant cooling, and marginally through the combustion of fossil fuel. One may therefore question why there
is so much focus on CO
2
, and not on water vapour, as a forcing to climate change.
Water vapour behaves differently from CO
2
in one fundamental way: it can condense and precipitate. When air
with high humidity cools, some of the vapour condenses into water droplets or ice particles and precipitates. The
typical residence time of water vapour in the atmosphere is ten days. The flux of water vapour into the atmosphere
from anthropogenic sources is considerably less than from ‘natural’ evaporation. Therefore, it has a negligible
impact on overall concentrations, and does not contribute significantly to the long-term greenhouse effect. This is
the main reason why tropospheric water vapour (typically below 10 km altitude) is not considered to be an anthro-
pogenic gas contributing to radiative forcing.
Anthropogenic emissions do have a significant impact on water vapour in the stratosphere, which is the part of
the atmosphere above about 10 km. Increased concentrations of methane (CH
4
) due to human activities lead to
an additional source of water, through oxidation, which partly explains the observed changes in that atmospheric
layer. That stratospheric water change has a radiative impact, is considered a forcing, and can be evaluated. Strato-
spheric concentrations of water have varied significantly in past decades. The full extent of these variations is not
well understood and is probably less a forcingthan
a feedback process added to natural variability. The
contribution of stratospheric water vapour to warm-
ing, both forcing and feedback, is much smaller than
from CH
4
or CO
2
.
The maximum amount of water vapour in the air
is controlled by temperature. A typical column of
air extending from the surface to the stratosphere
in polar regions may contain only a few kilograms
of water vapour per square metre, while a simi-
lar column of air in the tropics may contain up to
70 kg. With every extra degree of air temperature,
the atmosphere can retain around 7% more water
vapour (see upper-left insert in the FAQ 8.1, Figure
1). This increase in concentration amplifies the green-
house effect, and therefore leads to more warming.
This process, referred to as the water vapour feed-
back, is well understood and quantified. It occurs in
all models used to estimate climate change, where
its strength is consistent with observations. Although
an increase in atmospheric water vapour has been
observed, this change is recognized as a climate feed-
back (from increased atmospheric temperature) and
should not be interpreted as a radiative forcing from
anthropogenic emissions. (continued on next page)
FAQ 8.1, Figure 1 | Illustration of the water cycle and its interaction with the
greenhouse effect. The upper-left insert indicates the relative increase of poten-
tial water vapour content in the air with an increase of temperature (roughly
7% per degree). The white curls illustrate evaporation, which is compensated by
precipitation to close the water budget. The red arrows illustrate the outgoing
infrared radiation that is partly absorbed by water vapour and other gases, a pro-
cess that is one component of the greenhouse effect. The stratospheric processes
are not included in this figure.
T
0
-4 T
0
-2 T
0
T
0
+2 T
0
+4
Temperature change
1.4
1.2
1.0
0.8
0.6
Water vapour
8
Anthropogenic and Natural Radiative Forcing Chapter 8
667
of seasons or less, but there is a spectrum of adjustment times. Chang-
es in land ice and snow cover, for instance, may take place over many
years. The ERF thus represents that part of the instantaneous RF that is
maintained over long time scales and more directly contributes to the
steady-state climate response. The RF can be considered a more limited
version of ERF. Because the atmospheric temperature has been allowed
to adjust, ERF would be nearly identical if calculated at the tropopause
instead of the TOA for tropospheric forcing agents, as would RF. Recent
work has noted likely advantages of the ERF framework for under-
standing model responses to CO
2
as well as to more complex forcing
agents (see Section 7.2.5.6).
The climate sensitivity parameter l derived with respect to RF can vary
substantially across different forcing agents (Forster et al., 2007). The
response to RF from a particular agent relative to the response to RF
from CO
2
has been termed the efficacy (Hansen et al., 2005). By includ-
ing many of the rapid adjustments that differ across forcing agents,
the ERF concept includes much of their relative efficacy and therefore
leads to more uniform climate sensitivity across agents. For example,
the influence of clouds on the interaction of aerosols with sunlight and
the effect of aerosol heating on cloud formation can lead to very large
differences in the response per unit RF from black carbon (BC) located
at different altitudes, but the response per unit ERF is nearly uniform
with altitude (Hansen et al., 2005; Ming et al., 2010; Ban-Weiss et al.,
2012). Hence as we use ERF in this chapter when it differs significantly
from RF, efficacy is not used hereinafter. For inhomogeneous forcings,
we note that the climate sensitivity parameter may also depend on the
horizontal forcing distribution, especially with latitude (Shindell and
Faluvegi, 2009; Section 8.6.2).
A combination of RF and ERF will be used in this chapter with RF pro-
vided to keep consistency with TAR and AR4, and ERF used to allow
quantification of more complex forcing agents and, in some cases, pro-
vide a more useful metric than RF.
8.1.1.3 Limitations of Radiative Forcing
Both the RF and ERF concepts have strengths and weaknesses in
addition to those discussed previously. Dedicated climate model sim-
ulations that are required to diagnose the ERF can be more compu-
tationally demanding than those for instantaneous RF or RF because
many years are required to reduce the influence of climate variability.
The presence of meteorological variability can also make it difficult to
isolate the ERF of small forcings that are easily isolated in the pair of
radiative transfer calculations performed for RF (Figure 8.1). For RF, on
the other hand, a definition of the tropopause is required, which can
be ambiguous.
In many cases, however, ERF and RF are nearly equal. Analysis of 11
models from the current Coupled Model Intercomparison Project Phase
5 (CMIP5) generation finds that the rapid adjustments to CO
2
cause
fixed-SST-based ERF to be 2% less than RF, with an intermodel stand-
ard deviation of 7% (Vial et al., 2013). This is consistent with an earlier
study of six GCMs that found a substantial inter-model variation in
the rapid tropospheric adjustment to CO
2
using regression analysis in
slab ocean models, though the ensemble mean adjustment was less
than 5% (Andrews and Forster, 2008). Part of the large uncertainty
range arises from the greater noise inherent in regression analyses of
single runs in comparison with fixed-SST experiments. Using fixed-SST
simulations, Hansen et al. (2005) found that ERF is virtually identical
to RF for increased CO
2
, tropospheric ozone and solar irradiance, and
within 6% for methane (CH
4
), nitrous oxide (N
2
O), stratospheric aer-
osols and for the aerosol–radiation interaction of reflective aerosols.
Shindell et al. (2013b) also found that RF and ERF are statistically equal
for tropospheric ozone. Lohmann et al. (2010) report a small increase
in the forcing from CO
2
using ERF instead of RF based on the fixed-SST
technique, while finding no substantial difference for CH
4
, RF due to
aerosol–radiation interactions or aerosol effects on cloud albedo. In
the fixed-SST simulations of Hansen et al. (2005), ERF was about 20%
less than RF for the atmospheric effects of BC aerosols (not including
microphysical aerosol–cloud interactions), and nearly 300% greater
for the forcing due to BC snow albedo forcing (Hansen et al., 2007).
ERF was slightly greater than RF for stratospheric ozone in Hansen
et al. (2005), but the opposite is true for more recent analyses (Shin-
dell et al., 2013b), and hence it seems most appropriate at present to
use RF for this small forcing. The various studies demonstrate that RF
provides a good estimate of ERF in most cases, as the differences are
very small, with the notable exceptions of BC-related forcings (Bond
et al., 2013). ERF provides better characterization of those effects, as
well as allowing quantification of a broader range of effects including
all aerosol–cloud interactions. Hence while RF and ERF are generally
quite similar for WMGHGs, ERF typically provides a more useful indica-
tion of climate response for near-term climate forcers (see Box 8.2). As
the rapid adjustments included in ERF differ in strength across climate
models, the uncertainty range for ERF estimates tends to be larger than
the range for RF estimates.
FAQ 8.1 (continued)
Currently, water vapour has the largest greenhouse effect in the Earth’s atmosphere. However, other greenhouse
gases, primarily CO
2
, are necessary to sustain the presence of water vapour in the atmosphere. Indeed, if these other
gases were removed from the atmosphere, its temperature would drop sufficiently to induce a decrease of water
vapour, leading to a runaway drop of the greenhouse effect that would plunge the Earth into a frozen state. So
greenhouse gases other than water vapour provide the temperature structure that sustains current levels of atmo-
spheric water vapour. Therefore, although CO
2
is the main anthropogenic control knob on climate, water vapour
is a strong and fast feedback that amplifies any initial forcing by a typical factor between two and three. Water
vapour is not a significant initial forcing, but is nevertheless a fundamental agent of climate change.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
668
Whereas the global mean ERF provides a useful indication of the even-
tual change in global mean surface temperature, it does not reflect
regional climate changes. This is true for all forcing agents, but is espe-
cially the case for the inhomogeneously distributed forcings because
they activate climate feedbacks based on their regional distribution.
For example, forcings over Northern Hemisphere (NH) middle and high
latitudes induce snow and ice albedo feedbacks more than forcings at
lower latitudes or in the Southern Hemisphere (SH) (e.g., Shindell and
Faluvegi, 2009).
In the case of agents that strongly absorb incoming solar radiation
(such as BC, and to a lesser extent organic carbon (OC) and ozone) the
TOA forcing provides little indication of the change in solar radiation
reaching the surface which can force local changes in evaporation and
alter regional and general circulation patterns (e.g., Ramanathan and
Carmichael, 2008; Wang et al., 2009). Hence the forcing at the surface,
or the atmospheric heating, defined as the difference between sur-
face and tropopause/TOA forcing, might also be useful metrics. Global
mean precipitation changes can be related separately to ERF within
the atmosphere and to a slower response to global mean temperature
changes (Andrews et al., 2010; Ming et al., 2010; Ban-Weiss et al.,
2012). Relationships between surface forcing and localized aspects of
climate response have not yet been clearly quantified, however.
In general, most widely used definitions of forcing and most forc-
ing-based metrics are intended to be proportional to the eventual
temperature response, and most analyses to date have explored the
global mean temperature response only. These metrics do not explic-
itly include impacts such as changes in precipitation, surface sunlight
available for photosynthesis, extreme events, and so forth, or regional
Box 8.2 | Grouping Forcing Compounds by Common Properties
As many compounds cause RF when their atmospheric concentration is changed, it can be useful to refer to groups of compounds with
similar properties. Here we discuss two primary groupings: well-mixed greenhouse gases (WMGHGs) and near-term climate forcers
(NTCFs).
We define as ‘well-mixed’ those greenhouse gases that are sufficiently mixed throughout the troposphere that concentration measure-
ments from a few remote surface sites can characterize the climate-relevant atmospheric burden; although these gases may still have
local variation near sources and sinks and even small hemispheric gradients. Global forcing per unit emission and emission metrics
for these gases thus do not depend on the geographic location of the emission, and forcing calculations can assume even horizontal
distributions. These gases, or a subset of them, have sometimes been referred to as ‘long-lived greenhouse gases’ as they are well
mixed because their atmospheric lifetimes are much greater than the time scale of a few years for atmospheric mixing, but the physical
property that causes the aforementioned common characteristics is more directly associated with their mixing within the atmosphere.
WMGHGs include CO
2
, N
2
O, CH
4
, SF
6
, and many halogenated species. Conversely, ozone is not a WMGHG.
We define ‘near-term climate forcers’ (NTCFs) as those compounds whose impact on climate occurs primarily within the first decade
after their emission. This set of compounds is composed primarily of those with short lifetimes in the atmosphere compared to WMGHGs,
and has been sometimes referred to as short-lived climate forcers or short-lived climate pollutants. However, the common property
that is of greatest interest to a climate assessment is the time scale over which their impact on climate is felt. This set of compounds
includes methane, which is also a WMGHG, as well as ozone and aerosols, or their precursors, and some halogenated species that
are not WMGHGs. These compounds do not accumulate in the atmosphere at decadal to centennial time scales, and so their effect on
climate is predominantly in the near term following their emission.
temperatures, which can differ greatly from the global mean. Hence
although they are quite useful for understanding the factors driving
global mean temperature change, they provide only an imperfect and
limited perspective on the factors driving broader climate change. In
addition, a metric based solely on radiative perturbations does not
allow comparison of non-RFs, such as effects of land cover change
on evapotranspiration or physiological impacts of CO
2
and O
3
except
where these cause further impacts on radiation such as through cloud
cover changes (e.g., Andrews et al., 2012b).
8.1.2 Calculation of Radiative Forcing due to
Concentration or Emission Changes
Analysis of forcing due to observed or modelled concentration changes
between pre-industrial, defined here as 1750, and a chosen later year
provides an indication of the importance of different forcing agents to
climate change during that period. Such analyses have been a main-
stay of climate assessments. This perspective has the advantage that
observational data are available to accurately quantify the concentra-
tion changes for several of the largest forcing components. Atmospher-
ic concentration changes, however, are the net result of variations in
emissions of multiple compounds and any climate changes that have
influenced processes such as wet removal, atmospheric chemistry or
the carbon cycle. Characterizing forcing according to concentration
changes thus mixes multiple root causes along with climate feedbacks.
Policy decisions are better informed by analysis of forcing attributable
to emissions, which the IPCC first presented in AR4. These analyses can
be applied to historical emissions changes in a ‘backward-looking’ per-
spective, as done for example, for major WMGHGs (den Elzen et al.,
2005; Hohne et al., 2011) and NTCFs (Shindell et al., 2009), or to current
8
Anthropogenic and Natural Radiative Forcing Chapter 8
669
IRF: Net
Flux
change
Temperature
fixed
everywhere
Radiative
forcing tries
to modify
original
temperature
Stratospheric
T adjust
RF: Net Flux
change at
tropopause
Tropospheric
temperature
fixed
Net Flux
change at
TOA
Ground
temperature
fixed
ERF: Net Flux
change at
TOA
Net Flux = 0
Temperature
adjusts
everywhere
T
0
T
s
a b c d
e
Online or offline pair of
radiative transfer
calculations within one
simulation
Calculation Methodology
Difference between
two offline radiative
transfer calculations
with prescribed surface
and tropospheric
conditions allowing
stratospheric
temperature to adjust
Difference between
two full atmospheric
model simulations
with prescribed
surface conditions
everywhere or
estimate based on
regression of
response in full
coupled atmosphere-
ocean simulation
Difference between
two full atmospheric
model simulations
with prescribed ocean
conditions (SSTs and
sea ice)
Difference between
two full coupled
atmosphere-ocean
model simulations
Climatological
Tropopause
Ocean
fixed
Figure 8.1 | Cartoon comparing (a) instantaneous RF, (b) RF, which allows stratospheric temperature to adjust, (c) flux change when the surface temperature is fixed over the whole
Earth (a method of calculating ERF), (d) the ERF calculated allowing atmospheric and land temperature to adjust while ocean conditions are fixed and (e) the equilibrium response
to the climate forcing agent. The methodology for calculation of each type of forcing is also outlined. DT
o
represents the land temperature response, while DT
s
is the full surface
temperature response. (Updated from Hansen et al., 2005.)
or projected future emissions in a ‘forward-looking’ view (see Section
8.7). Emissions estimates through time typically come from the scientific
community, often making use of national reporting for recent decades.
With the greater use of emission-driven models, for example, in CMIP5,
it is becoming more natural to estimate ERF resulting from emissions
of a particular species rather than concentration-based forcing. Such
calculations typically necessitate model simulations with chemical
transport models or chemistry–climate models, however, and require
careful consideration of which processes are included, especially when
comparing results to concentration-based forcings. In particular, simu-
lation of concentration responses to emissions changes requires incor-
porating models of the carbon cycle and atmospheric chemistry (gas
and aerosol phases). The requisite expansion of the modelling realm for
emissions-based forcing or emission metrics should in principle be con-
sistent for all drivers. For example, as the response to aerosol or ozone
precursor emissions includes atmospheric chemistry, the response to
CO
2
emissions should as well. In addition, if the CO
2
concentration
responses to CO
2
emissions include the impact of CO
2
-induced climate
changes on carbon uptake, then the effect of climate changes caused
by any other emission on carbon uptake should also be included. Simi-
larly, if the effects of atmospheric CO
2
concentration change on carbon
uptake are included, the effects of other atmospheric composition or
deposition changes on carbon uptake should be included as well (see
also Section 6.4.1). Comparable issues are present for other forcing
agents. In practice, the modelling realm used in studies of forcing
attributable to emissions has not always been consistent. Furthermore,
climate feedbacks have sometimes been included in the calculation
of forcing due to ozone or aerosol changes, as when concentrations
from a historical transient climate simulation are imposed for an ERF
calculation. In this chapter, we endeavour to clarify which processes
have been included in the various estimates of forcing attributed to
emissions (Sections 8.3 and 8.7).
RF or ERF estimates based on either historical emissions or concen-
trations provide valuable insight into the relative and absolute con-
tribution of various drivers to historical climate change. Scenarios of
changing future emissions and land use are also developed based on
various assumptions about socioeconomic trends and societal choices.
The forcing resulting from such scenarios is used to understand the
drivers of potential future climate changes (Sections 8.5.3 and 8.6).
As with historical forcings, the actual impact on climate depends on
both the temporal and spatial structure of the forcings and the rate of
response of various portions of the climate system.
8.2 Atmospheric Chemistry
8.2.1 Introduction
Most radiatively active compounds in the Earth’s atmosphere are
chemically active, meaning that atmospheric chemistry plays a large
role in determining their burden and residence time. In the atmosphere,
a gaseous chemically active compound can be affected by (1) interac-
tion with other species (including aerosols and water) in its immediate
vicinity and (2) interaction with solar radiation (photolysis). Physical
processes (wet removal and dry deposition) act on some chemical
compounds (gas or aerosols) to further define their residence time
in the atmosphere. Atmospheric chemistry is characterized by many
interactions and patterns of temporal or spatial variability, leading to
significant nonlinearities (Kleinman et al., 2001) and a wide range of
time scales of importance (Isaksen et al., 2009).
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
670
This section assesses updates in understanding of processes, modelling
and observations since AR4 (see Section 2.3) on key reactive species
contributing to RF. Note that aerosols, including processes responsible
for the formation of aerosols, are extensively described in Section 7.3.
8.2.2 Global Chemistry Modelling in Coupled Model
Intercomparison Project Phase 5
Because the distribution of NTCFs cannot be estimated from obser-
vations alone, coupled chemistry-climate simulations are required to
define their evolution and associated RF. While several CMIP5 mode-
ling groups performed simulations with interactive chemistry (i.e., com-
puted simultaneously within the climate model), many models used
as input pre-computed distributions of radiatively active gases and/
or aerosols. To assess the distributions of chemical species and their
respective RF, many research groups participated in the Atmospheric
Chemistry and Climate Model Intercomparison Project (ACCMIP).
The ACCMIP simulations (Lamarque et al., 2013) were defined to pro-
vide information on the long-term changes in atmospheric composi-
tion with a few, well-defined atmospheric simulations. Because of the
nature of the simulations (pre-industrial, present-day and future cli-
mates), only a limited number of chemistry-transport models (models
which require a full definition of the meteorological fields needed to
simulate physical processes and transport) participated in the ACCMIP
project, which instead drew primarily from the same General Circu-
lation Models (GCMs) as CMIP5 (see Lamarque et al., 2013 for a list
of the participating models and their configurations), with extensive
model evaluation against observations (Bowman et al., 2013; Lee et
al., 2013; Shindell et al., 2013c; Voulgarakis et al., 2013; Young et al.,
2013).
In all CMIP5/ACCMIP chemistry simulations, anthropogenic and bio-
mass burning emissions are specified. More specifically, a single set of
historical anthropogenic and biomass burning emissions (Lamarque et
al., 2010) and one set of emissions for each of the RCPs (van Vuuren
et al., 2011) was defined (Figure 8.2). This was designed to increase
the comparability of simulations. However, these uniform emission
specifications mask the existing uncertainty (e.g., Bond et al., 2007; Lu
et al., 2011), so that there is in fact a considerable range in the esti-
mates and time evolution of recent anthropogenic emissions (Granier
et al., 2011). Historical reconstructions of biomass burning (wildfires
and deforestation) also exhibit quite large uncertainties (Kasischke and
Penner, 2004; Ito and Penner, 2005; Schultz et al., 2008; van der Werf
et al., 2010). In addition, the RCP biomass burning projections do not
include the feedback between climate change and fires discussed in
Bowman et al. (2009), Pechony and Shindell (2010) and Thonicke et al.
(2010). Finally, the RCP anthropogenic precursor emissions of NTCFs
tend to span a smaller range than available from existing scenarios
(van Vuuren et al., 2011). The ACCMIP simulations therefore provide an
estimate of the uncertainty due to range of representation of physical
and chemical processes in models, but do not incorporate uncertainty
in emissions.
8.2.3 Chemical Processes and Trace Gas Budgets
8.2.3.1 Tropospheric Ozone
The RF from tropospheric ozone is strongly height- and latitude-de-
pendent through coupling of ozone change with temperature, water
vapour and clouds (Lacis et al., 1990; Berntsen et al., 1997; Worden
et al., 2008, 2011; Bowman et al., 2013). Consequently, it is necessary
to accurately estimate the change in the ozone spatio-temporal struc-
ture using global models and observations. It is also well established
that surface ozone detrimentally affects plant productivity (Ashmore,
2005; Fishman et al., 2010), albeit estimating this impact on climate,
although possibly significant, is still limited to a few studies (Sitch et
al., 2007; UNEP, 2011).
Tropospheric ozone is a by-product of the oxidation of carbon monox-
ide (CO), CH
4
, and non-CH
4
hydrocarbons in the presence of nitrogen
oxides (NO
x
). As emissions of these precursors have increased (Figure
8.2), tropospheric ozone has increased since pre-industrial times (Volz
and Kley, 1988; Marenco et al., 1994) and over the last decades (Parrish
et al., 2009; Cooper et al., 2010; Logan et al., 2012), but with important
regional variations (Section 2.2). Ozone production is usually limited
by the supply of HO
x
(OH + HO
2
) and NO
X
(NO + NO
2
) (Levy, 1971;
Logan et al., 1981). Ozone’s major chemical loss pathways in the trop-
osphere are through (1) photolysis (to O(
1
D), followed by reaction with
water vapour) and (2) reaction with HO
2
(Seinfeld and Pandis, 2006).
The former pathway leads to couplings between stratospheric ozone
(photolysis rate being a function of the overhead ozone column) and
climate change (through water vapour). Observed surface ozone abun-
dances typically range from less than 10 ppb over the tropical Pacific
Ocean to more than 100 ppb downwind of highly emitting regions. The
lifetime of ozone in the troposphere varies strongly with season and
location: it may be as little as a few days in the tropical boundary layer,
or as much as 1 year in the upper troposphere. Two recent studies give
similar global mean lifetime of ozone: 22.3 ± 2 days (Stevenson et al.,
2006) and 23.4 ± 2.2 days (Young et al., 2013).
For present (about 2000) conditions, the various components of the
budget of global mean tropospheric ozone are estimated from the
ACCMIP simulations and other model simulations since AR4 (Table
8.1). In particular, most recent models define a globally and annually
averaged tropospheric ozone burden of (337 ± 23 Tg, 1-σ). Differences
in the definition of the tropopause lead to inter-model variations of
approximately 10% (Wild, 2007). This multi-model mean estimate of
global annual tropospheric ozone burden has not significantly changed
since the Stevenson et al. (2006) estimates (344 ± 39 Tg, 1-σ), and
is consistent with the most recent satellite-based Ozone Monitoring
Instrument–Microwave Limb Sounder (OMI-MLS; Ziemke et al., 2011)
and Tropospheric Emission Spectrometer (TES; Osterman et al., 2008)
climatologies.
Estimates of the ozone chemical sources and sinks (uncertainty esti-
mates are quoted here as 1-σ) are less robust, with a net chemical
production (production minus loss) of 618 ± 275 Tg yr
–1
(Table 8.1),
larger than the Atmospheric Composition Change: a European Net-
work (ACCENT) results (442 ± 309 Tg yr
–1
; Stevenson et al., 2006). Esti-
mates of ozone deposition (1094 ± 264 Tg yr
–1
) are slightly increased
8
Anthropogenic and Natural Radiative Forcing Chapter 8
671
Black Carbon CH
4
CO
NH
3
SO
2
Organic Carbon
NMVOC NO
x
(TgC year
-1
)
(TgCH
4
year
-1
)
(TgCO year
-1
)
(TgNH
3
year
-1
)(TgNMVOC year
-1
)
(TgNO
2
year
-1
)
(TgC year
-1
)
(TgSO
2
year
-1
)
Historical
RCP2.6
RCP4.5
RCP6.0
RCP8.5
SRES B1
IS92a
GAINS-CLE
GAINS-MFR
SRES A2
policy (upper bound)
reference (upper bound)
policy (lower bound)
reference (lower bound)
Post-SRES scenarios
CH
4
: Rogelj et al. (2011)
Others: van Vuuren et al. (2008)
Asia Modeling Exercise (Calvin
et al., 2012)
2.6 W m
-2
(mean)
reference (mean)
Figure 8.2 | Time evolution of global anthropogenic and biomass burning emissions 1850–2100 used in CMIP5/ACCMIP following each RCP. Historical (1850–2000) values are
from Lamarque et al. (2010). RCP values are from van Vuuren et al. (2011). Emissions estimates from Special Report on Emission Scenarios (SRES) are discussed in Annex II; note that
black carbon and organic carbon estimates were not part of the SRES and are shown here only for completeness. The Maximum Feasible Reduction (MFR) and Current Legislation
(CLE) are discussed in Cofala et al. (2007); as biomass burning emissions are not included in that publication, a fixed amount, equivalent to the value in 2000 from the RCP esti-
mates, is added (see Annex II for more details; Dentener et al., 2006). The post-SRES scenarios are discussed in Van Vuuren et al. (2008) and Rogelj et al. (2011). For those, only the
range (minimum to maximum) is shown. Global emissions from the Asian Modelling Exercise are discussed in Calvin et al. (2012). Regional estimates are shown in Supplementary
Material Figure 8.SM.1 and Figure 8.SM.2 for the historical and RCPs.
since ACCENT (1003 ± 200 Tg yr
–1
) while estimates of the net influx of
ozone from the stratosphere to the troposphere (477 ± 96 Tg yr
–1
) have
slightly decreased since ACCENT (552 ± 168 Tg yr
–1
). Additional model
estimates of this influx (Hegglin and Shepherd, 2009; Hsu and Prather,
2009) fall within both ranges, as do estimates based on observations
(Murphy and Fahey, 1994; Gettelman et al., 1997; Olsen et al., 2002),
all estimates being sensitive to their choice of tropopause definition
and interannual variability.
Model simulations for present-day conditions or the recent past are
evaluated (Figure 8.3) against frequent ozonesonde measurements
(Logan, 1999; Tilmes et al., 2012) and additional surface, aircraft and
satellite measurements. The ACCMIP model simulations (Figure 8.3)
indicate 10 to 20% negative bias at 250 hPa in the SH tropical region,
and a slight underestimate in NH tropical region. Comparison with
satellite-based estimates of tropospheric ozone column (Ziemke et al.,
2011) indicates an annual mean bias of –4.3 ± 29 Tg (with a spatial
correlation of 0.87 ± 0.07, 1-σ) for the ACCMIP simulations (Young et
al., 2013). Overall, our ability to simulate tropospheric ozone burden
for present (about 2000) has not substantially changed since AR4.
Evaluation (using a subset of two ACCMIP models) of simulated trends
(1960s to present or shorter) in surface ozone against observations at
remote surface sites (see Section 2.2) indicates an underestimation,
especially in the NH (Lamarque et al., 2010). Although this limits the
ability to represent recent ozone changes, it is unclear how this trans-
lates into an uncertainty on changes since pre-industrial times.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
672
Table 8.1 | Summary of tropospheric ozone global budget model and observation estimates for present (about 2000) conditions. Focus is on modelling studies published since
AR4. STE stands for stratosphere–troposphere exchange. All uncertainties quoted as 1 standard deviation (68% confidence interval).
Burden Production Loss Deposition STE
Reference
Tg Tg yr
–1
Tg yr
–1
Tg yr
–1
Tg yr
–1
Modelling Studies
337 ± 23 4877 ± 853 4260 ± 645 1094 ± 264 477 ± 96 Young et al. (2013); ACCMIP
323 N/A N/A N/A N/A Archibald et al. (2011)
330 4876 4520 916 560 Kawase et al. (2011)
312 4289 3881 829 421 Huijnen et al. (2010)
334 3826 3373 1286 662 Zeng et al. (2010)
324 4870 4570 801 502 Wild and Palmer (2008)
314 N/A N/A 1035 452 Zeng et al. (2008)
319 4487 3999 N/A 500 Wu et al. (2007)
372 5042 4507 884 345 Horowitz (2006)
349 4384 3972 808 401 Liao et al. (2006)
344 ± 39 5110 ± 606 4668 ± 727 1003 ± 200 552 ± 168 Stevenson et al. (2006); ACCENT
314 ± 33 4465 ± 514 4114 ± 409 949 ± 222 529 ± 105 Wild (2007) (post-2000 studies)
N/A N/A N/A N/A 515 Hsu and Prather (2009)
N/A N/A N/A N/A 655 Hegglin and Shepherd (2009)
N/A N/A N/A N/A 383–451 Clark et al. (2007)
Observational Studies
333 N/A N/A N/A N/A Fortuin and Kelder (1998)
327 N/A N/A N/A N/A Logan (1999)
325 N/A N/A N/A N/A Ziemke et al. (2011); 60S–60N
319–351 N/A N/A N/A N/A Osterman et al. (2008); 60S–60N
N/A N/A N/A N/A 449 (192–872) Murphy and Fahey (1994)
N/A N/A N/A N/A 510 (450–590) Gettelman et al. (1997)
N/A N/A N/A N/A 500 ± 140 Olsen et al. (2001)
In most studies ‘pre-industrial’ does not identify a specific year but is
usually assumed to correspond to 1850s levels; no observational infor-
mation on ozone is available for that time period. Using the Lamarque
et al. (2010) emissions, the ACCMIP models (Young et al., 2013) are
unable to reproduce the low levels of ozone observed at Montsouris
1876–1886 (Volz and Kley, 1988). The other early ozone measurements
using the Schönbein paper are controversial (Marenco et al., 1994)
and assessed to be of qualitative use only. The main uncertainty in
estimating the pre-industrial to present-day change in ozone there-
fore remains the lack of constraint on emission trends because of the
very incomplete knowledge of pre-industrial ozone concentrations, of
which no new information is available. The uncertainty on pre-indus-
trial conditions is not confined to ozone but applies to aerosols as well
(e.g., Schmidt et al., 2012), although ice and lake core records provide
some constraint on pre-industrial aerosol concentrations.
The ACCMIP results provide an estimated tropospheric ozone increase
(Figure 8.4) from 1850 to 2000 of 98 ± 17 Tg (model range), similar
to AR4 estimates. Skeie et al. (2011a) found an additional 5% increase
in the anthropogenic contribution to the ozone burden between 2000
and 2010, which translates into an approximately 1.5% increase in
tropospheric ozone burden. A best estimate of the change in ozone
since 1850 is assessed at 100 ± 25 Tg (1-σ). Attribution simulations
(Stevenson et al., 2013) indicate unequivocally that anthropogenic
changes in ozone precursor emissions are responsible for the increase
between 1850 and present or into the future.
8.2.3.2 Stratospheric Ozone and Water Vapour
Stratospheric ozone has experienced significant depletion since the
1960s due to bromine and chlorine-containing compounds (Solomon,
1999), leading to an estimated global decrease of stratospheric ozone
of 5% between the 1970s and the mid-1990s, the decrease being
largest over Antarctica (Fioletov et al., 2002). Most of the ozone loss
is associated with the long-lived bromine and chlorine-containing
compounds (chlorofluorocarbons and substitutes) released by human
activities, in addition to N
2
O. This is in addition to a background level
of natural emissions of short-lived halogens from oceanic and volcanic
sources.
With the advent of the Montreal Protocol and its amendments, emis-
sions of chlorofluorocarbons (CFCs) and replacements have strongly
declined (Montzka et al., 2011), and signs of ozone stabilization and
even possibly recovery have already occurred (Mader et al., 2010; Salby
et al., 2012). A further consequence is that N
2
O emissions (Section
8.2.3.4) likely dominate all other emissions in terms of ozone-depleting
8
Anthropogenic and Natural Radiative Forcing Chapter 8
673
J F M A M J J A S O N D J
20
40
60
80
J F M A M J J A S O N D J
20
40
60
80
100
J F M A M J J A S O N D J
20
40
60
80
J F M A M J J A S O N D J
20
40
60
80
100
J F M A M J J A S O N D J
20
40
60
80
100
J F M A M J J A S O N D J
20
40
60
80
100
JFMAMJ JASONDJ
20
40
60
80
JFMAMJ JASONDJ
20
40
60
80
100
JFMAMJ JASONDJ
20
40
60
80
100
JFMAMJ JASONDJ
20
40
60
80
100
ACCMIP mean
ACCMIP models
ACCENT mean
r = 0.90, 0.96
mnbe = −12.2%, 3.0%
r = 0.61, 0.45
mnbe = −1.1%, 19.3%
r = 0.95, 0.97
mnbe = −2.2%, 10.9%
r = 0.94, 0.91
mnbe = −10.0%, −1.5%
r = 0.71, 0.59
mnbe = 5.0%, 15.0%
r = 0.99, 0.84
mnbe = 12.8%, 9.9%
r = 0.97, 0.98
mnbe = −2.6%, −8.5%
r = 0.97, 0.89
mnbe = −8.6%, −3.1%
r = 0.89, 0.81
mnbe = 7.8%, 10.8%
r = 0.95, 0.84
mnbe = 13.2%, 3.0%
250 hPa
500 hPa
750 hPa
90°S - 30°S 30°S - EQ EQ - 30°N 30°N - 90°N
Ozonesondes
Ozone volume mixing ratio (ppb)
Figure 8.3 | Comparisons between observations and simulations for the monthly mean ozone for ACCMIP results (Young et al., 2013). ACCENT refers to the model results in
Stevenson et al. (2006). For each box, the correlation of the seasonal cycle is indicated by the r value, while the mean normalized bias estimated is indicated by mnbe value.
Historical RCP2.6 RCP4.5 RCP6.0 RCP8.5
1850 1930 1980 2000 2000 2030 210020302100 2030 2100 2030 2100
0
100
200
300
400
500
600
Burden (Tg)
Modeled Past Modeled ProjectionsObservations
ACCENT
Osterman et al.
Ziemke et al.
Logan
Fortuin and Kelder
2000
mean
Figure 8.4 | Time evolution of global tropospheric ozone burden (in Tg(O
3
)) from 1850 to 2100 from ACCMIP results, ACCENT results (2000 only), and observational estimates
(see Table 8.1). The box, whiskers, line and dot show the interquartile range, full range, median and mean burdens and differences, respectively. The dashed line indicates the 2000
ACCMIP mean. (Adapted from Young et al., 2013.)
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
674
potential (Ravishankara et al., 2009). Chemistry-climate models with
resolved stratospheric chemistry and dynamics recently predicted an
estimated global mean total ozone column recovery to 1980 levels to
occur in 2032 (multi-model mean value, with a range of 2024 to 2042)
under the A1B scenario (Eyring et al., 2010a). Increases in the strato-
spheric burden and acceleration of the stratospheric circulation leads
to an increase in the stratosphere–troposphere flux of ozone (Shindell
et al., 2006c; Grewe, 2007; Hegglin and Shepherd, 2009; Zeng et al.,
2010). This is also seen in recent RCP8.5 simulations, with the impact
of increasing tropospheric burden (Kawase et al., 2011; Lamarque et
al., 2011). However, observationally based estimates of recent trends
in age of air (Engel et al., 2009; Stiller et al., 2012) do not appear
to be consistent with the acceleration of the stratospheric circulation
found in model simulations, possibly owing to inherent difficulties with
extracting trends from SF
6
observations (Garcia et al., 2011).
Oxidation of CH
4
in the stratosphere (see Section 8.2.3.3) is a signifi-
cant source of water vapour and hence the long-term increase in CH
4
leads to an anthropogenic forcing (see Section 8.3) in the stratosphere.
Stratospheric water vapour abundance increased by an average of 1.0
± 0.2 (1-σ) ppm during 1980–2010, with CH
4
oxidation explaining
approximately 25% of this increase (Hurst et al., 2011). Other factors
contributing to the long-term change in water vapour include changes
in tropical tropopause temperatures (see Section 2.2.2.1).
8.2.3.3 Methane
The surface mixing ratio of CH
4
has increased by 150% since pre-indus-
trial times (Sections 2.2.1.1.2 and 8.3.2.2), with some projections indi-
cating a further doubling by 2100 (Figure 8.5). Bottom-up estimates of
present CH
4
emissions range from 542 to 852 TgCH
4
yr
–1
(see Table 6.8),
while a recent top-down estimate with uncertainty analysis is 554 ±
56 TgCH
4
yr
–1
(Prather et al., 2012). All quoted uncertainties in Section
8.2.3.3 are defined as 1-σ.
The main sink of CH
4
is through its reaction with the hydroxyl radical
(OH) in the troposphere (Ehhalt and Heidt, 1973). A primary source
of tropospheric OH is initiated by the photodissociation of ozone, fol-
lowed by reaction with water vapour (creating sensitivity to humid-
ity, cloud cover and solar radiation) (Levy, 1971; Crutzen, 1973). The
Historical
RCP2.6
RCP4.5
RCP6.0
RCP8.5
SRES B1
IS92a
SRES A2
CH
4
(ppm) CO
2
(ppm) N
2
O (ppm)
Figure 8.5 | Time evolution of global-averaged mixing ratio of long-lived species1850–2100 following each RCP; blue (RCP2.6), light blue (RCP4.5), orange (RCP6.0) and red
(RCP8.5). (Based on Meinshausen et al., 2011b.)
other main source of OH is through secondary reactions (Lelieveld et
al., 2008), although some of those reactions are still poorly understood
(Paulot et al., 2009; Peeters et al., 2009; Taraborrelli et al., 2012). A
recent estimate of the CH
4
tropospheric chemical lifetime with respect
to OH constrained by methyl chloroform surface observations is 11.2
± 1.3 years (Prather et al., 2012). In addition, bacterial uptake in soils
provides an additional small, less constrained loss (Fung et al., 1991);
estimated lifetime = 120 ± 24 years (Prather et al., 2012), with another
small loss in the stratosphere (Ehhalt and Heidt, 1973); estimated life-
time = 150 ± 50 years (Prather et al., 2012). Halogen chemistry in the
troposphere also contributes to some tropospheric CH
4
loss (Allan et
al., 2007), estimated lifetime = 200 ± 100 years (Prather et al., 2012).
The ACCMIP estimate for present CH
4
lifetime with respect to trop-
ospheric OH varies quite widely (9.8 ± 1.6 years (Voulgarakis et al.,
2013)), slightly shorter than the 10.2 ± 1.7 years in (Fiore et al. (2009),
but much shorter than the methyl chloroform-based estimate of 11.2
± 1.3 years (Prather et al., 2012). A partial explanation for the range in
CH
4
lifetime changes can be found in the degree of representation of
chemistry in chemistry–climate models. Indeed, Archibald et al. (2010)
showed that the response of OH to increasing nitrogen oxides strongly
depends on the treatment of hydrocarbon chemistry in a model. The
impact on CH
4
distribution in the ACCMIP simulations is, however,
rather limited because most models prescribed CH
4
as a time-varying
lower-boundary mixing ratio (Lamarque et al., 2013).
The chemical coupling between OH and CH
4
leads to a significant
amplification of an emission impact; that is, increasing CH
4
emissions
decreases tropospheric OH which in turn increases the CH
4
lifetime
and therefore its burden. The OH-lifetime sensitivity for CH
4
, s_OH =
–δln(OH)/δln(CH4), was estimated in Chapter 4 of TAR to be 0.32,
implying a 0.32% decrease in tropospheric mean OH (as weighted by
CH
4
loss) for a 1% increase in CH
4
. The Fiore et al. (2009) multi-mod-
el (12 models) study provides a slightly smaller value (0.28 ± 0.03).
Holmes et al. (2013) gives a range 0.31 ± 0.04 by combining Fiore et al.
(2009), Holmes et al. (2011) and three new model results (0.36, 0.31,
0.27). Only two ACCMIP models reported values (0.19 and 0.26; Voul-
garakis et al., 2013). The projections of future CH
4
in Chapter 11 use
the Holmes et al. (2013) range and uncertainty, which at the 2-σ level
covers all but one model result. The feedback factor f, the ratio of the
8
Anthropogenic and Natural Radiative Forcing Chapter 8
675
lifetime of a CH
4
perturbation to the lifetime of the total CH
4
burden, is
calculated as f = 1/(1-s). Other CH
4
losses, which are relatively insensi-
tive to CH
4
burden, must be included so that f = 1.34 ± 0.06, (slightly
larger but within the range of the Stevenson et al. (2006) estimate of
1.29 ± 0.04, based on six models), leading to an overall perturbation
lifetime of 12.4 ± 1.4 years, which is used in calculations of metrics
in Section 8.7. Additional details are provided in the Supplementary
Material Section 8.SM.2.
8.2.3.4 Nitrous Oxide
Nitrous oxide (N
2
O) in 2011 has a surface concentration 19% above
its 1750 level (Sections 2.2.1.1.3 and 8.3.2.3). Increases in N
2
O lead to
depletion of mid- to upper-stratospheric ozone and increase in mid-lat-
itude lower stratospheric ozone (as a result of increased photolysis
rate from decreased ozone above). This impacts tropospheric chemistry
through increase in stratosphere–troposphere exchange of ozone and
odd nitrogen species and increase in tropospheric photolysis rates and
OH formation (Prather and Hsu, 2010). Anthropogenic emissions repre-
sent around 30 to 45% of the present-day global total, and are mostly
from agricultural and soil sources (Fowler et al., 2009) and fossil-fuel
activities. Natural emissions come mostly from microbial activity in the
soil. The main sink for N
2
O is through photolysis and oxidation reac-
tions in the stratosphere, leading to an estimated lifetime of 131 ±
10 years (Prather et al., 2012), slightly larger than previous estimates
(Prather and Hsu, 2010; Montzka et al., 2011). The addition of N
2
O
to the atmosphere changes its own lifetime through feedbacks that
couple N
2
O to stratospheric NO
y
and ozone depletion (Prather, 1998;
Ravishankara et al., 2009; Prather and Hsu, 2010), so that the lifetime
of a perturbation is less than that of the total burden, 121 ± 10 years
(1-σ; Prather et al., 2012) and is used in calculations of metrics (Sec-
tion 8.7).
8.2.3.5 Halogenated Species
Halogenated species can be powerful greenhouse gases (GHGs). Those
containing chlorine and bromine also deplete stratospheric ozone
and are referred to as ozone-depleting substances (ODSs). Most of
those compounds do not have natural emissions and, because of the
implementation of the Montreal Protocol and its amendments, total
emissions of ODSs have sharply decreased since the 1990s (Montzka
et al., 2011). For CFCs, perfluorocarbons (PFCs) and SF
6
the main loss
is through photolysis in the stratosphere. The CFC substitutes (hydro-
chlorofluorocarbons (HCFCs) and hydrofluorocarbons (HFCs)) are
destroyed by OH oxidation in the troposphere. Their global concen-
tration has steadily risen over the recent past (see Section 2.2.1.1.4).
8.2.3.6 Aerosols
Aerosol particles are present in the atmosphere with size ranges
from a few nanometres to tens of micrometres. They are the results
of direct emission (primary aerosols: BC, OC, sea salt, dust) into the
atmosphere or as products of chemical reactions (secondary inorganic
aerosols: sulphate, nitrate, ammonium; and secondary organic aero-
sols (SOAs)) occurring in the atmosphere. Secondary inorganic aero-
sols are the products of reactions involving sulphur dioxide, ammonia
and nitric oxide emissions. SOAs are the result of chemical reactions
of non-methane hydrocarbons (and their products) with the hydroxyl
radical (OH), ozone, nitrate (NO
3
) or photolysis (Hallquist et al., 2009).
Thus although many hydrocarbons in the atmosphere are of biogenic
origin, anthropogenic pollutants can have impacts on their conversion
to SOAs. There is tremendous complexity and still much uncertainty in
the processes involved in the formation of SOAs (Hallquist et al., 2009;
Carslaw et al., 2010). Additional information can be found in Section
7.3.2.
Once generated, the size and composition of aerosol particles can be
modified by additional chemical reactions, condensation or evapora-
tion of gaseous species and coagulation (Seinfeld and Pandis, 2006).
It is this set of processes that defines their physical, chemical and opti-
cal properties, and hence their impact on radiation and clouds, with
large regional and global differences (see Section 7.3.3). Furthermore,
their distribution is affected by transport and deposition, defining a
residence time in the troposphere of usually a few days (Textor et al.,
2006).
8.3 Present-Day Anthropogenic Radiative
Forcing
Human activity has caused a variety of changes in different forcing
agents in the atmosphere or land surface. A large number of GHGs
have had a substantial increase over the Industrial Era and some of
these gases are entirely of anthropogenic origin. Atmospheric aerosols
have diverse and complex influences on the climate. Human activity
has modified the land cover and changed the surface albedo. Some of
the gases and aerosols are directly emitted to the atmosphere where-
as others are secondary products from chemical reactions of emitted
species. The lifetimes of these different forcing agents vary substan-
tially. This section discusses all known anthropogenic forcing agents
of non-negligible importance and their quantification in terms of RF
or ERF based on changes in abundance over the 1750–2011 period.
In this section we determine the RFs for WMGHGs and heterogene-
ously distributed species in fundamentally different ways. As described
in Box 8.2, the concentrations of WMGHGs can be determined from
observations at a few surface sites. For the pre-industrial concentra-
tions these are typically from trapped air in polar ice or firn (see Sec-
tion 2.2.1). Thus the RFs from WMGHGs are determined entirely from
observations (Section 8.3.2). In contrast, we do not have sufficient
pre-industrial or present-day observations of heterogeneously distrib-
uted forcing agents (e.g., ozone and aerosols) to be able to character-
ize their RF; therefore we instead have to rely on chemistry–climate
models (Sections 8.3.3 and 8.3.4).
8.3.1 Updated Understanding of the Spectral Properties
of Greenhouse Gases and Radiative Transfer Codes
RF estimates are performed with a combination of radiative transfer
codes typical for GCMs as well as more detailed radiative transfer
codes. Physical properties are needed in the radiative transfer codes
such as spectral properties for gases. The HITRAN (HIgh Resolution
TRANsmission molecular absorption) database (Rothman, 2010) is
widely used in radiative transfer models. Some researchers studied
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
676
the difference among different editions of HITRAN databases for
diverse uses (Feng et al., 2007; Kratz, 2008; Feng and Zhao, 2009;
Fomin and Falaleeva, 2009; Lu et al., 2012). Model calculations have
shown that modifications of the spectroscopic characteristics tend to
have a modest effect on the determination of RF estimates of order
2 to 3% of the calculated RF attributed to the combined doubling of
CO
2
, N
2
O and CH
4
. These results showed that even the largest overall
RF induced by differences among the HITRAN databases is consider-
ably smaller than the range reported for the modelled RF estimates;
thus the line parameter updates to the HITRAN database are not a
significant source for discrepancies in the RF calculations appearing
in the IPCC reports. However, the more recent HITRAN data set is
still recommended, as the HITRAN process offers internal verification
and tends to progress closer to the best laboratory measurements.
It is found that the differences among the water vapour continuum
absorption formulations tend to be comparable to the differences
among the various HITRAN databases (Paynter and Ramaswamy,
2011); but use of the older Robert continuum formula produces signif-
icantly larger flux differences, thus, replacement of the older continu-
um is warranted (Kratz, 2008) and there are still numerous unresolved
issues left in the continuum expression, especially related to short-
wave radiative transfer (Shine et al., 2012). Differences in absorption
data from various HITRAN versions are very likely a small contributor
to the uncertainty in RF of GHGs.
Line-by-line (LBL) models using the HITRAN data set as an input are
the benchmark of radiative transfer models for GHGs. Some research-
ers compared different LBL models (Zhang et al., 2005; Collins et al.,
2006) and line-wing cutoff, line-shape function and gas continuum
absorption treatment effects on LBL calculations (Zhang et al., 2008;
Fomin and Falaleeva, 2009). The agreement between LBL codes has
been investigated in many studies and found to generally be within
a few percent (e.g., Collins et al., 2006; Iacono et al., 2008; Forster et
al., 2011a) and to compare well to observed radiative fluxes under
controlled situations (Oreopoulos et al., 2012). Forster et al. (2011a)
evaluated global mean radiatively important properties of chemistry
climate models (CCMs) and found that the combined WMGHG global
annual mean instantaneous RF at the tropopause is within 30% of LBL
models for all CCM radiation codes tested. The accuracies of the LW RF
due to CO
2
and tropospheric ozone increase are generally very good
and within 10% for most of the participation models, but problems
remained in simulating RF for stratospheric water vapour and ozone
changes with errors between 3% and 200% compared to LBL models.
Whereas the differences in the results from CCM radiation codes were
large, the agreement among the LW LBL codes was within 5%, except
for stratospheric water vapour changes.
Most intercomparison studies of the RF of GHGs are for clear-sky and
aerosol-free conditions; the introduction of clouds would greatly com-
plicate the targets of research and are usually omitted in the intercom-
parison exercises of GCM radiation codes and LBL codes (e.g., Collins
et al., 2006; Iacono et al., 2008). It is shown that clouds can reduce
the magnitude of RF due to GHGs by about 25% (Forster et al., 2005;
Worden et al., 2011; Zhang et al., 2011), but the influence of clouds
on the diversity in RF is found to be within 5% in four detailed radi-
ative transfer schemes with realistic cloud distributions (Forster et al.,
2005). Estimates of GHG RF are based on the LBL codes or the radiative
transfer codes compared and validated against LBL models, and the
uncertainty range from AR4 in the RF of GHG of 10% is retained. We
underscore that uncertainty in RF calculations in many GCMs is sub-
stantially higher owing both to radiative transfer codes and meteoro-
logical data such as clouds adopted in the simulations.
8.3.2 Well-mixed Greenhouse Gases
AR4 assessed the RF from 1750 to 2005 of the WMGHGs to be 2.63
W m
–2
. The four most important gases were CO
2
, CH
4
, dichlorodifluo-
romethane (CFC-12) and N
2
O in that order. Halocarbons, comprising
CFCs, HCFCs, HFCs, PFCs and SF
6
, contributed 0.337 W m
–2
to the total.
Uncertainties (90% confidence ranges) were assessed to be approxi-
mately 10% for the WMGHGs. The major changes to the science since
AR4 are the updating of the atmospheric concentrations, the inclusion
of new species (NF
3
and SO
2
F
2
) and discussion of ERF for CO
2
. Since
AR4 N
2
O has overtaken CFC-12 as the third largest contributor to RF.
The total WMGHG RF is now 2.83 (2.54 to 3.12) W m
–2
.
The RFs in this section are derived from the observed differences in
concentrations of the WMGHGs between 1750 and 2011. The con-
centrations of CO
2
, CH
4
and N
2
O vary throughout the pre-industrial
era, mostly due to varying climate, with a possible small contribution
from anthropogenic emissions (MacFarling Meure et al., 2006). These
variations do not contribute to uncertainty in the RF as strictly defined
here, but do affect the RF attribution to anthropogenic emissions. On
centennial time scales, variations in late Holocene concentrations of
CO
2
are around 10 ppm (see note to Table 2.1), much larger than
the uncertainty in the 1750 concentration. This would equate to a
variation in the RF of 10%. For CH
4
and N
2
O the centennial variations
are comparable to the uncertainties in the 1750 concentrations and
so do not significantly affect the estimate of the 1750 value used in
calculating RF.
8.3.2.1 Carbon Dioxide
The tropospheric mixing ratio of CO
2
has increased globally from 278
(276–280) ppm in 1750 to 390.5 (390.3 to 390.7) ppm in 2011 (see
Section 2.2.1.1.1). Here we assess the RF due to changes in atmos-
pheric concentration rather than attributing it to anthropogenic emis-
sions. Section 6.3.2.6 describes how only a fraction of the historical
CO
2
emissions have remained in the atmosphere. The impact of land
use change on CO
2
from 1850 to 2000 was assessed in AR4 to be 12 to
35 ppm (0.17 to 0.51 W m
–2
).
Using the formula from Table 3 of Myhre et al. (1998), and see Supple-
mentary Material Table 8.SM.1, the CO
2
RF (as defined in Section 8.1)
from 1750 to 2011 is 1.82 (1.63 to 2.01) W m
–2
. The uncertainty is dom-
inated by the radiative transfer modelling which is assessed to be 10%
(Section 8.3.1). The uncertainty in the knowledge of 1750 concentra-
tions contributes only 2% (see Supplementary Material Table 8.SM.2)
Table 8.2 shows the concentrations and RF in AR4 (2005) and 2011 for
the most important WMGHGs. Figure 8.6 shows the time evolution of
RF and its rate of change. Since AR4, the RF of CO
2
has increased by
0.16 W m
–2
and continues the rate noted in AR4 of almost 0.3 W m
–2
per decade. As shown in Figure 8.6(d) the rate of increase in the RF
8
Anthropogenic and Natural Radiative Forcing Chapter 8
677
from the WMGHGs over the last 15 years has been dominated by CO
2
.
Since AR4, CO
2
has accounted for more than 80% of the WMGHG RF
increase. The interannual variability in the rate of increase in the CO
2
RF is due largely to variation in the natural land uptake whereas the
trend is driven by increasing anthropogenic emissions (see Figure 6.8
in Section 6.3.1).
As described in Section 8.1.1.3, CO
2
can also affect climate through
physical effects on lapse rates and clouds, leading to an ERF that will
be different from the RF. Analysis of CMIP5 models (Vial et al., 2013)
found a large negative contribution to the ERF (20%) from the increase
in land surface temperatures which was compensated for by positive
contributions from the combined effects on water vapour, lapse rate,
albedo and clouds. It is therefore not possible to conclude with the
current information whether the ERF for CO
2
is higher or lower than
the RF. Therefore we assess the ratio ERF/RF to be 1.0 and assess our
uncertainty in the CO
2
ERF to be (–20% to 20%). We have medium
confidence in this based on our understanding that the physical pro-
cesses responsible for the differences between ERF and RF are small
enough to be covered within the 20% uncertainty.
There are additional effects mediated through plant physiology, reduc-
ing the conductance of the plant stomata and hence the transpiration
of water. Andrews et al. (2012b) find a physiological enhancement of
the adjusted forcing by 3.5% due mainly to reductions in low cloud.
This is smaller than a study with an earlier model by Doutriaux-Bouch-
er et al. (2009) which found a 10% effect. Longer-term impacts of
CO
2
on vegetation distributions also affect climate (O’ishi et al., 2009;
Andrews et al., 2012b) but because of the longer time scale we choose
to class these as feedbacks rather than rapid adjustments.
8.3.2.2 Methane
Globally averaged surface CH
4
concentrations have risen from 722 ±
25 ppb in 1750 to 1803 ± 2 ppb by 2011 (see Section 2.2.1.1.2). Over
that time scale the rise has been due predominantly to changes in
anthropogenic-related CH
4
. Anthropogenic emissions of other com-
pounds have also affected CH
4
concentrations by changing its remov-
al rate (Section 8.2.3.3). Using the formula from Myhre et al. (1998)
(see Supplementary Material Table 8.SM.1) the RF for CH
4
from 1750
to 2011 is 0.48 ± 0.05 W m
–2
, with an uncertainty dominated by the
radiative transfer calculation. This increase of 0.01 W m
–2
since AR4 is
due to the 29 ppb increase in the CH
4
mixing ratio. This is much larger
than the 11 ppb increase between TAR and AR4, and has been driven
by increases in net natural and anthropogenic emissions, but the rel-
ative contributions are not well quantified. Recent trends in CH
4
and
their causes are discussed in Sections 2.2.1.1.2 and 6.3.3.1. CH
4
con-
centrations do vary with latitude and decrease above the tropopause;
however, this variation contributes only 2% to the uncertainty in RF
(Freckleton et al., 1998).
In this section only the direct forcing from changing CH
4
concentrations
is addressed. CH
4
emissions can also have indirect effects on climate
through impacts on CO
2
, stratospheric water vapour, ozone, sulphate
aerosol and lifetimes of HFCs and HCFCs (Boucher et al., 2009; Shindell
et al., 2009; Collins et al., 2010). Some of these are discussed further in
Sections 8.3.3, 8.5.1 and 8.7.2.
8.3.2.3 Nitrous Oxide
Concentrations of nitrous oxide have risen from 270 ± 7 ppb in 1750 to
324.2 ± 0.1 ppb in 2011, an increase of 5 ppb since 2005 (see Section
2.2.1.1.3). N
2
O now has the third largest forcing of the anthropogenic
gases, at 0.17 ± 0.03 W m
–2
an increase of 6% since 2005 (see Table
Figure 8.6 | (a) Radiative forcing (RF) from the major well-mixed greenhouse gases
(WMGHGs) and groups of halocarbons from 1850 to 2011 (data from Tables A.II.1.1
and A.II.4.16), (b) as (a) but with a logarithmic scale, (c) RF from the minor WMGHGs
from 1850 to 2011 (logarithmic scale). (d) Rate of change in forcing from the major
WMGHGs and groups of halocarbons from 1850 to 2011.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
678
8.2) where the uncertainty is due approximately equally to the pre-in-
dustrial concentration and radiative transfer. Only the direct RF from
changing nitrous oxide concentrations is included. Indirect effects of
N
2
O emissions on stratospheric ozone are not taken into account here
but are discussed briefly in Section 8.7.2.
8.3.2.4 Other Well-mixed Greenhouse Gases
RFs of the other WMGHG are shown in Figure 8.6 (b and c) and Table
8.2. The contribution of groups of halocarbons to the rate of change
of WMGHG RF is shown in Figure 8.6 (d). Between 1970 and 1990
halocarbons made a significant contribution to the rate of change of
RF. The rate of change in the total WMGHG RF was higher in 1970 to
1990 with high confidence compared to the present owing to higher
contribution from non-CO
2
gases especially the halocarbons. Since the
Montreal Protocol and its amendments, the rate of change of RF from
halocarbons and related compounds has been much less, but still just
positive (total RF of 0.360 W m
–2
in 2011 compared to 0.351 W m
–2
in
2005) as the growth of HCFCs, HFCs, PFCs and other halogens (SF
6
,
SO
2
F
2
, NF
3
) RFs (total 0.022 W m
–2
since 2005) more than compensates
Notes:
a
Pre-industrial values are zero except for CO
2
(278 ppm), CH
4
(722 ppb), N
2
O (270 ppb) and CF
4
(35 ppt).
b
Total includes 0.007 W m
–2
to account for CFC-114, Halon-1211 and Halon-1301.
c
Total includes 0.009 W m
–2
forcing (as in AR4) to account for CFC-13, CFC-114, CFC-115, Halon-1211 and Halon-1301.
d
Defined here as CFCs + HCFCs + CH
3
CCl
3
+ CCl
4
.
e
The value for the 1750 methane concentrations has been updated from AR4 in this report, thus the 2005 methane RF is slightly lower than reported in AR4.
f
Estimates for halocarbons given in the table may have changed from estimates reported in AR4 owing to updates in radiative efficiencies and concentrations.
Table 8.2 | Present-day mole fractions (in ppt(pmol mol
–1
) except where specified) and RF (in W m
–2
) for the WMGHGs. Concentration data are averages of National Oceanic and
Atmospheric Administration (NOAA) and Advanced Global Atmospheric Gases Experiment (AGAGE) observations where available. CO
2
concentrations are the average of NOAA
and SIO. See Table 2.1 for more details of the data sources. The data for 2005 (the time of the AR4 estimates) are also shown. Some of the concentrations vary slightly from those
reported in AR4 owing to averaging different data sources. Radiative efficiencies for the minor gases are given in Table 8.A.1. Uncertainties in the RF for all gases are dominated
by the uncertainties in the radiative efficiencies. We assume the uncertainties in the radiative efficiencies to be perfectly correlated between the gases, and the uncertainties in the
present day and 1750 concentrations to be uncorrelated.
Concentrations (ppt) Radiative forcing
a
(W m
–2
)
Species 2011 2005 2011 2005
CO
2
(ppm) 391 ± 0.2 379 1.82 ± 0.19 1.66
CH
4
(ppb) 1803 ± 2 1774 0.48 ± 0.05 0.47
e
N
2
O (ppb) 324 ± 0.1 319 0.17 ± 0.03 0.16
CFC-11 238 ± 0.8 251 0.062 0.065
CFC-12 528 ± 1 542 0.17 0.17
CFC-13 2.7 0.0007
CFC-113 74.3 ± 0.1 78.6 0.022 0.024
CFC-115 8.37 8.36 0.0017 0.0017
HCFC-22 213 ± 0.1 169 0.0447 0.0355
HCFC-141b 21.4 ± 0.1 17.7 0.0034 0.0028
HCFC-142b 21.2 ± 0.2 15.5 0.0040 0.0029
HFC-23 24.0 ± 0.3 18.8 0.0043 0.0034
HFC-32 4.92 1.15 0.0005 0.0001
HFC-125 9.58 ± 0.04 3.69 0.0022 0.0008
HFC-134a 62.7 ± 0.3 34.3 0.0100 0.0055
HFC-143a 12.0 ± 0.1 5.6 0.0019 0.0009
HFC-152a 6.4 ± 0.1 3.4 0.0006 0.0003
SF
6
7.28 ± 0.03 5.64 0.0041 0.0032
SO
2
F
2
1.71 1.35 0.0003 0.0003
NF
3
0.9 0.4 0.0002 0.0001
CF
4
79.0 ± 0.1 75.0 0.0040 0.0036
C
2
F
6
4.16 ± 0.02 3.66 0.0010 0.0009
CH
3
CCl
3
6.32 ± 0.07 18.32 0.0004 0.0013
CCl
4
85.8 ± 0.8 93.1 0.0146 0.0158
CFCs 0.263 ± 0.026
b
0.273
c
HCFCs 0.052 ± 0.005 0.041
Montreal gases
d
0.330 ± 0.033 0.331
Total halogens 0.360 ± 0.036 0.351
f
Total 2.83 ± 0.029 2.64
8
Anthropogenic and Natural Radiative Forcing Chapter 8
679
for the decline in the CFCs, CH
3
CCl
3
and CCl
4
RFs (–0.013 W m
–2
since
2005). The total halocarbon RF is dominated by four gases, namely
CFC-12, trichlorofluoromethane (CFC-11), chlorodifluoromethane
(HCFC-22) and trichlorofluoroeethane (CFC-113) in that order, which
account for about 85% of the total halocarbon RF (see Table 8.2) . The
indirect RF from the impacts of ODSs is discussed in Section 8.3.3.2.
8.3.2.4.1 Chlorofluorocarbons and hydrochlorofluorocarbons
The CFCs and HCFCs contribute approximately 11% of the WMGHG
RF. Although emissions have been drastically reduced for CFCs, their
long lifetimes mean that reductions take substantial time to affect their
concentrations. The RF from CFCs has declined since 2005 (mainly due
to a reduction in the concentrations of CFC-11 and CFC-12), whereas
the RF from HCFCs is still rising (mainly due to HCFC-22).
8.3.2.4.2 Hydrofluorocarbons
The RF of HFCs is 0.02 W m
–2
and has close to doubled since AR4 (2005
concentrations). HFC-134a is the dominant contributor to RF of the
HFCs, with an RF of 0.01 W m
–2
.
8.3.2.4.3 Perfluorocarbons and sulphur hexafluoride
These gases have lifetimes of thousands to tens of thousands of years
(Table 8.A.1); therefore emissions essentially accumulate in the atmos-
phere on the time scales considered here. CF
4
has a natural source and
a 1750 concentration of 35 ppt (see Section 2.2.1.1.4). These gases
currently contribute 0.01 W m
–2
of the total WMGHG RF.
8.3.2.4.4 New species
Nitrogen trifluoride (NF
3
) is used in the electronics industry and sulfuryl
fluoride (SO
2
F
2
) is used as a fumigant. Both have rapidly increasing
emissions and high GWPs, but currently contribute only around 0.0002
W m
–2
and 0.0003 W m
–2
to anthropogenic RF, respectively (Weiss et
al., 2008; Andersen et al., 2009; Muhle et al., 2009; Arnold et al., 2013).
8.3.3 Ozone and Stratospheric Water Vapour
Unlike for the WMGHGs, the estimate of the tropospheric and strato-
spheric ozone concentration changes are almost entirely model based
for the full pre-industrial to present-day interval (though, especially for
the stratosphere, more robust observational evidence on changes is
available for recent decades; see Section 2.2).
AR4 assessed the RF (for 1750–2005) from tropospheric ozone to be
0.35 W m
–2
from multi-model studies with a high 95th percentile of
0.65 W m
–2
to allow for the possibility of model overestimates of the
pre-industrial tropospheric ozone levels. The stratospheric ozone RF
was assessed from observational trends from 1979 to 1998 to be –0.05
± 0.1 W m
–2
, with the 90% confidence range increased to reflect uncer-
tainty in the trend prior to 1979 and since 1998. In AR4 the RF from
stratospheric water vapour generated by CH
4
oxidation was assessed
to be +0.07 ± 0.05 W m
–2
based on Hansen et al. (2005).
Since AR4, there have been a few individual studies of tropospheric
or stratospheric ozone forcing (Shindell et al., 2006a, 2006c, 2013a;
Skeie et al., 2011a; Søvde et al., 2011), a multi-model study of strat-
ospheric ozone RF in the 2010 WMO stratospheric ozone assessment
(Forster et al., 2011b), and the ACCMIP multi-model study of tropo-
spheric and tropospheric + stratospheric chemistry models (Conley et
al., 2013; Stevenson et al., 2013). There is now greater understanding
of how tropospheric ozone precursors can affect stratospheric ozone,
and how ODSs can affect tropospheric ozone (Shindell et al., 2013a).
We assess the total ozone RF to be +0.35 (0.15 to 0.55) W m
–2
. This
can be split according to altitude or by emitted species (Shindell et
al., 2013a). We assess these contributions to be 0.40 (0.20 to 0.60) W
m
–2
for ozone in the troposphere and –0.05 ± 0.10 W m
–2
for ozone in
the stratosphere based on the studies presented in Table 8.3. Alterna-
tively, the contributions to the total ozone forcing can be attributed
as 0.50 (0.30 to 0.70) W m
–2
from ozone precursors and –0.15 (–0.3
to 0.0) W m
–2
from the effect of ODSs. The value attributed to ODSs is
assessed to be slightly smaller in magnitude than in the two studies
quoted in Table 8.3 (Søvde et al., 2011; Shindell et al., 2013a) because
the models used for these had stratospheric ozone RFs with higher
magnitudes than the ACCMIP mean (Conley et al., 2013). Differences
between the ERFs and RFs for tropospheric and stratospheric ozone
are likely to be small compared to the uncertainties in the RFs (Shin-
dell et al., 2013b), so the assessed values for the ERFs are the same
as those for the RFs.
The influence of climate change is typically included in ozone RF esti-
mates as those are based on modelled concentration changes, but the
available literature provides insufficient evidence for the sign and mag-
nitude of the impact and we therefore refrain from giving an estimate
except to assess that it is very likely to be smaller than the overall
uncertainty in the total RF. Unlike the WMGHGs, there are significant
latitudinal variations in the RFs from changes in tropospheric and strat-
ospheric ozone. The implications of inhomogeneous RFs are explored
in more detail in Section 8.6.
There has been one study since AR4 (Myhre et al., 2007) on the RF from
water vapour formed from the stratospheric oxidation of CH
4
(Section
8.3.3.3). This is consistent with the AR4 value and so has not led to
any change in the recommended value of 0.07 (0.02 to 0.12) W m
–2
since AR4.
8.3.3.1 Tropospheric Ozone
Ozone is formed in the troposphere by photochemical reactions of nat-
ural and anthropogenic precursor species (Section 8.2.3.1). Changes in
ozone above the tropopause due to emissions of stratospheric ODSs
can also affect ozone in the troposphere either by transport across
the tropopause or modification of photolysis rates. Changes in climate
have also affected tropospheric ozone concentrations (medium evi-
dence, low agreement) through changes in chemistry, natural emis-
sions and transport from the stratosphere (Isaksen et al., 2009).
The most recent estimates of tropospheric ozone RF come from
multi-model studies under ACCMIP (Conley et al., 2013; Lamarque et
al., 2013; Stevenson et al., 2013). The model ensemble reported only
1850–2000 RFs (0.34 W m
–2
) so the single-model results from Skeie et
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
680
al. (2011a) were used to expand the timespan to 1750–2010, adding
0.04 W m
–2
, and 0.02 W m
–2
to account for the periods 1750–1850 and
2000–2010 respectively. The best estimate of tropospheric ozone RF
taking into account the ACCMIP models and the Søvde et al. (2011)
results (the Skeie et al. (2011a) and Shindell et al. (2013a) models are
included in ACCMIP) is 0.40 (0.20 to 0.60) W m
–2
. The quantifiable
uncertainties come from the inter-model spread (–0.11 to 0.11 W m
–2
)
and the differences between radiative transfer models (–0.07 to 0.07
W m
–2
); all 5 to 95% confidence interval. Additional uncertainties arise
from the lack of knowledge of pre-industrial emissions and the rep-
resentation of chemical and physical processes beyond those included
in the current models. The tropospheric ozone RF is sensitive to the
assumed ‘pre-industrial’ levels. As described in Section 8.2.3.1, very
limited late 19th and early 20th
century observations of surface ozone
concentrations are lower than the ACCMIP models for the same period;
however, we assess that those observations are very uncertain. Skeie
et al. (2011a) and Stevenson et al. (2013) increase their uncertainty
ranges to 30% for 1 standard deviation which is equivalent to (–50%
to +50%) for the 5 to 95% confidence range and we adopt this for
AR5. The overall confidence in the tropospheric ozone RF is assessed
as high.
Because we have low confidence in the pre-industrial ozone observa-
tions, and these were extremely limited in spatial coverage, it is not
possible to calculate a purely observationally based ozone RF. However,
modern observations can be used to assess the performance of the
chemistry models. Bowman et al. (2013) used satellite retrievals from
the TES instrument to constrain the RF from the ACCMIP models. This
reduced the inter-model uncertainty by 30%; however, we still main-
tain overall the (–50% to +50%) 5 to 95% confidence range for AR5.
The time evolution of the tropospheric ozone forcing is shown in Figure
8.7. There is a noticeable acceleration in the forcing after 1950 and a
deceleration in the 1990s reflecting the time evolution of anthropo-
genic precursor emissions. Observational evidence for trends in ozone
concentrations is discussed in Section 2.2.2.3.
It can be useful to calculate a normalized radiative forcing (NRF) which
is an RF per change in ozone column in W m
–2
DU
–1
or W mol
–1
. This is
only an approximation as the NRF is sensitive to the vertical profile
of the ozone change and to the latitudinal profile to a smaller extent.
From Table 8.3 we assess the NRF to be 0.042 (0.037 to 0.047) W m
–2
DU
–1
(94 (83 to 105) W mol
–1
) similar to the value of 0.042 W m
–2
DU
–1
(94 W mol
–1
) in TAR (Ramaswamy et al., 2001).
A small number of studies have looked at attributing the ozone chang-
es among the anthropogenically emitted species. Søvde et al. (2011)
report a tropospheric ozone RF of 0.38 W m
–2
, 0.44 W m
–2
from ozone
precursors and –0.06 W m
–2
from the impact of stratospheric ozone
depletion on the troposphere. Shindell et al. (2013a) also calculate
that ODSs are responsible for about –0.06 W m
–2
of the tropospher-
ic ozone RF, and ozone precursors for about 0.41 W m
–2
. Six of the
models in Stevenson et al. (2013) and Shindell et al. (2009) performed
experiments to attribute the ozone RF to the individual precursor emis-
sions. An average of these seven model results leads to attributions of
0.24 ± 0.13 W m
–2
due to CH
4
emissions, 0.14 ± 0.09 W m
–2
from NO
X
emissions, 0.07 ± 0.03 W m
–2
from CO, and 0.04 ± 0.03W m
–2
from
1750 1800 1850 1900 1950 2000
-0.2
0.0
0.2
0.4
0.6
Ozone Radiative Forcing W m
-2
Tropospheric
Total
Stratospheric
( )
non-methane volatile organic compounds (NMVOCs). These results
were calculated by reducing the precursor emissions individually from
2000 to pre-industrial levels. The results were scaled by the total ozone
RFs attributed to ozone precursors (0.50 W m
–2
) to give the contri-
butions to the full 1750–2010 RF. Because of the nonlinearity of the
chemistry an alternative method of starting from pre-industrial con-
ditions and increasing precursor emissions singly may give a different
result. Note that as well as inducing an ozone RF, these ozone pre-
cursor species can also strongly affect the concentrations of CH
4
and
aerosols, adding extra terms (both positive and negative) to their total
indirect forcings. The contributions to the 1750–2010 CH
4
RF are again
based on Stevenson et al. (2013) and Shindell et al. (2009). The Steven-
son et al. (2013) values are for 1850–2000 rather than 1750 to 2011
so for these we distribute the CH
4
RF for 1750–1850 and 2000–2011
(0.06 W m
–2
) by scaling the CH
4
and CO contributions (assuming these
were the most significant contributors over those time periods). This
gives contributions of 0.58 ± 0.08, –0.29 ± 0.18, 0.07 ± 0.02 and 0.02
± 0.02 W m
–2
for changes from historical to present day emissions of
CH
4
(inferred emissions), NO
X
, CO and VOCs respectively (uncertainties
are 5 to 95% confidence intervals). The difference between the total
CH
4
RF attributed to ozone precursors here (0.38 W m
–2
) and the value
calculated from CH
4
concentration changes in Table 8.2 (0.48 W m
–2
)
is due to nonlinearities in the CH
4
chemistry because large single-step
changes were used. To allow an easier comparison between the con-
centration-based and emission-based approaches in Section 8.5.1 the
nonlinear term (+0.1 W m
–2
) is distributed between the four emitted
species according to their absolute magnitude so that they total 0.48
W m
–2
. The scaled results still lie within the uncertainty bounds of the
values quoted above. The impact of climate change over the historical
period on CH
4
oxidation is not accounted for in these calculations.
Tropospheric ozone can also affect the natural uptake of CO
2
by
decreasing plant productivity (see Sections 6.4.8.2 and 8.2.3.1) and
it is found that this indirect effect could have contributed to the total
CO
2
RF (Section 8.3.2.1; Sitch et al., 2007), roughly doubling the over-
all RF attributed to ozone precursors. Although we assess there to be
Figure 8.7 | Time evolution of the radiative forcing from tropospheric and stratospheric
ozone from 1750 to 2010. Tropospheric ozone data are from Stevenson et al. (2013)
scaled to give 0.40 W m
–2
at 2010. The stratospheric ozone RF follow the functional
shape of the Effective Equivalent Stratospheric Chlorine assuming a 3-year age of air
(Daniel et al., 2010) scaled to give –0.05 W m
–2
at 2010.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
681
Notes:
a
From multi-model results.
b
From Randel and Wu (2007) observation-based data set.
c
Using the REF chemistry, see Søvde et al. (2011).
d
Using the R2 chemistry.
e
The Atmospheric Chemistry and Climate Model Intercomparison Project (ACCMIP) tropospheric ozone RFs are from Stevenson et al. (2013). The stratospheric ozone values are from
Conley et al. (2013) calculations for 1850–2005 disregarding the Modèle de Chimie Atmosphérique a Grande Echelle (MOCAGE) model which showed excessive ozone depletion.
f
Only the Goddard Institute for Space Studies (GISS)-E2-R results (including bias correction) from the Shindell et al. (2013a) study are shown here rather than the multi-model result
presented in that paper.
Table 8.3 | Contributions of tropospheric and stratospheric ozone changes to radiative forcing (W m
–2
) from 1750 to 2011.
Troposphere Stratosphere
Longwave Shortwave Total
Normalized
Radiative Forcing
m W m
–2
DU
–1
Longwave Shortwave Total
AR4 (Forster
et al. (2007)
0.35
(0.25 to 0.65)
–0.05
(–0.15 to 0.05)
Shindell et al.
(2013a)
f
0.33
(0.31 to 0.35)
–0.08
(–0.10 to –0.06)
WMO (Forster
et al., 2011b)
–0.03
a
(–0.23 to +0.17)
+0.03
b
Søvde et al. (2011)
0.45
c
40 –0.12
0.38
d
39 –0.12
Skeie et al. (2011a)
0.41
(0.21 to 0.61)
38
ACCMIP
e
0.33
(0.24 to 0.42)
0.08
(0.06 to 0.10)
0.41
(0.21 to 0.61)
42
(37 to 47)
–0.13
(–0.26 to 0)
0.11
(0.03 to 0.19)
–0.02
(–0.09 to 0.05)
AR5
0.40
(0.20 to 0.60)
42
(37 to 47)
–0.05
(-0.15 to 0.05)
robust evidence of an effect, we make no assessment of the magnitude
because of lack of further corroborating studies.
8.3.3.2 Stratospheric Ozone
The decreases in stratospheric ozone due to anthropogenic emissions
of ODSs have a positive RF in the shortwave (increasing the flux into
the troposphere) and a negative RF in the longwave. This leaves a
residual forcing that is the difference of two larger terms. In the lower
stratosphere the longwave effect tends to be larger, whereas in the
upper stratosphere the shortwave dominates. Thus whether strat-
ospheric ozone depletion has contributed an overall net positive or
negative forcing depends on the vertical profile of the change (Forster
and Shine, 1997). WMO (2011) assessed the RF from 1979 to 2005
from observed ozone changes (Randel and Wu, 2007) and results from
16 models for the 1970s average to 2004. The observed and modelled
mean ozone changes gave RF values of different signs (see Table 8.3).
Negative net RFs arise from models with ozone decline in the lower-
most stratosphere, particularly at or near the tropopause.
The ACCMIP study also included some models with stratospheric
chemistry (Conley et al., 2013). One model in that study stood out as
having excessive ozone depletion. Removing that model leaves a strat-
ospheric ozone RF of –0.02 (–0.09 to 0.05) W m
–2
. These results are in
good agreement with the model studies from WMO (2011). Forster et
al. (2007) in AR4 calculated a forcing of –0.05 W m
–2
from observa-
tions over the period 1979–1998 and increased the uncertainty to 0.10
W m
–2
to encompass changes between the pre-industrial period and
2005. The RF from stratospheric ozone due to changes in emissions
of ozone precursors and ODSs is here assessed to be –0.05 (–0.15 to
0.05) taking into account all the studies listed in Table 8.3. This is in
agreement with AR4, although derived from different data. The time-
line of stratospheric ozone forcing is shown in Figure 8.7, making the
assumption that it follows the trajectory of the changes in EESC. It
reaches a minimum in the late 1990s and starts to recover after that.
The net global RF from ODSs taking into account the compensating
effects on ozone and their direct effects as WMGHGs is 0.18 (0.03 to
0.33) W m
–2
. The patterns of RF for these two effects are different so
the small net global RF comprises areas of positive and negative RF.
8.3.3.3 Stratospheric Water Vapour
Stratospheric water vapour is dependent on the amount entering from
the tropical troposphere and from direct injection by volcanic plumes
(Joshi and Jones, 2009) and aircraft, and the in situ chemical pro-
duction from the oxidation of CH
4
and hydrogen. This contrasts with
tropospheric water vapour which is almost entirely controlled by the
balance between evaporation and precipitation (see FAQ 8.1). We con-
sider trends in the transport (for instance, due to the Brewer–Dobson
circulation or tropopause temperature changes) to be climate feedback
rather than a forcing so the anthropogenic RFs come from oxidation of
CH
4
and hydrogen, and emissions from stratospheric aircraft.
Myhre et al. (2007) used observations of the vertical profile of CH
4
to
deduce a contribution from oxidation of anthropogenic CH
4
of 0.083
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
682
W m
–2
which compares with the value of 0.07 W m
–2
from calcula-
tions in a 2D model in Hansen et al. (2005). Both of these values are
consistent with AR4 which obtained the stratospheric water vapour
forcing by scaling the CH
4
direct forcing by 15%. Thus the time evolu-
tion of this forcing is also obtained by scaling the CH
4
forcing by 15%.
The best estimate and uncertainty range from AR4 of 0.07 (0.02 to
0.12) W m
–2
remain unchanged and the large uncertainty range is due
to large differences found in the intercomparison studies of radiative
transfer modelling for changes in stratospheric water vapour (see Sec-
tion 8.3.1).
RF from the current aircraft fleet through stratospheric water vapour
emissions is very small. Wilcox et al. (2012) estimate a contribution
from civilian aircraft in 2005 of 0.0009 (0.0003 to 0.0013) W m
–2
with high confidence in the upper limit. Water vapour emissions from
aircraft in the troposphere also contribute to contrails which are dis-
cussed in Section 8.3.4.5.
8.3.4 Aerosols and Cloud Effects
8.3.4.1 Introduction and Summary of AR4
In AR4 (Forster et al., 2007), RF estimates were provided for three aer-
osol effects. These were the RF of aerosol–radiation interaction (previ-
ously denoted as direct aerosol effect), RF of the aerosol–cloud inter-
action (previously denoted as the cloud albedo effect), and the impact
of BC on snow and ice surface albedo. See Chapter 7 and Figure 7.3 for
an explanation of the change in terminology between AR4 and AR5.
The RF due to aerosol–radiation interaction is scattering and absorp-
tion of shortwave and longwave radiation by atmospheric aerosols.
Several different aerosol types from various sources are present in the
atmosphere (see Section 8.2). Most of the aerosols primarily scatter
solar radiation, but some components absorb solar radiation to various
extents with BC as the most absorbing component. RF of aerosols in
the troposphere is often calculated at the TOA because it is similar to
tropopause values (Forster et al., 2007). A best estimate RF of –0.5 ±
0.4 W m
–2
was given in AR4 for the change in the net aerosol–radia-
tion interaction between 1750 and 2005 and a medium to low level of
scientific understanding (LOSU).
An increase in the hygroscopic aerosol abundance may enhance the
concentration of cloud condensation nuclei (CCN). This may increase
the cloud albedo and under the assumption of fixed cloud water con-
tent this effect was given a best estimate of –0.7 W m
–2
(range from
–1.8 to –0.3) in AR4 and a low LOSU.
BC in the snow or ice can lead to a decrease of the surface albedo.
This leads to a positive RF. In AR4 this mechanism was given a best RF
estimate of 0.1 ± 0.1 W m
–2
and a low LOSU.
Impacts on clouds from the ERF of aerosol–cloud interaction (includ-
ing both effects previously denoted as cloud lifetime and cloud albedo
effect) and the ERF of aerosol–radiation interaction (including both
effects previously denoted as direct aerosol effect and semi-direct
effect) were not strictly in accordance with the RF concept, because
they involve tropospheric changes in variables other than the forcing
agent at least in the available model estimates, so no best RF estimates
were provided in AR4 (see Section 8.1). However, the ERF of aerosol–
cloud and aerosol–radiation interactions were included in the discus-
sion of total aerosol effect in Chapter 7 in AR4 (Denman et al., 2007).
The mechanisms influenced by anthropogenic aerosol including the
aerosol cloud interactions are discussed in detail in this assessment in
Section 7.5 and summarized in the subsections that follow.
8.3.4.2 Radiation Forcing of the Aerosol–Radiation
Interaction by Component
Based on a combination of global aerosol models and observa-
tion-based methods, the best RF estimate of the aerosol–radiation
interaction in AR5 is –0.35 (–0.85 to +0.15) W m
–2
(see Section 7.5).
This estimate is thus smaller in magnitude than in AR4, however; with
larger uncertainty range. Overall, the estimate compared to AR4 is
more robust because the agreement between estimates from models
and observation-based methods is much greater (see Section 7.5). The
larger range arises primarily from analysis by observation-based meth-
ods (see Section 7.5).
The main source of the model estimate is based on updated simula-
tions in AeroCom (Myhre et al., 2013), which is an intercomparison
exercise of a large set of global aerosol models that includes extensive
evaluation against measurements. The assessment in Chapter 7 relies
to a large extent on this study for the separation in the various aerosol
components, except for BC where the assessment in Chapter 7 relies in
addition on Bond et al. (2013). The RF of aerosol–radiation interaction
is separated into seven components in this report; namely sulphate,
BC from fossil fuel and biofuel, OA from fossil fuel and biofuel, BC and
OA combined from biomass burning (BB), nitrate, SOA and mineral
dust. BC and OA from biomass burning are combined due to the joint
sources, whereas treated separately for fossil fuel and biofuel because
there is larger variability in the ratio of BC to OA in the fossil fuel
and biofuel emissions. This approach is consistent with TAR and AR4.
Table 8.4 compares the best estimates of RF due to aerosol–radiation
interaction for various components in this report with values in SAR,
TAR and AR4. In magnitude the sulphate and BC from use of fossil fuel
and biofuel dominate. It is important to note that the BB RF is small in
magnitude but consists of larger, offsetting terms in magnitude from
OA and BC (see Section 7.5.2). Changes in the estimates of RF due to
aerosol–radiation interaction of the various components have been
rather modest compared to AR4, except for BC from fossil fuel and
biofuel (see Section 7.5). SOA is a new component compared to AR4.
Anthropogenic SOA precursors contribute only modestly to the anthro-
pogenic change in SOA. The increase in SOA is mostly from biogenic
precursors and enhanced partitioning of SOA into existing particles
from anthropogenic sources and changes in the atmospheric oxidation
(Carlton et al., 2010). This change in SOA is therefore of anthropogenic
origin, but natural emission of SOA precursors is important (Hoyle et
al., 2011).
Note that the best estimate and the uncertainty for the total is not
equal to the sum of the aerosol components because the total is
estimated based on a combination of methods (models and observa-
tion-based methods), whereas the estimates for the components rely
mostly on model estimates.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
683
Global Mean Radiative Forcing (W m
–2
)
SAR TAR AR4 AR5
Sulphate aerosol –0.40 (–0.80 to –0.20) –0.40 (–0.80 to –0.20) –0.40 (–0.60 to –0.20) –0.40 (–0.60 to –0.20)
Black carbon aerosol from
fossil fuel and biofuel
+0.10 (+0.03 to +0.30) +0.20 (+0.10 to +0.40) +0.20 (+0.05 to +0.35) +0.40 (+0.05 to +0.80)
Primary organic aerosol
from fossil fuel and biofuel
Not estimated –0.10 (–0.30 to –0.03) –0.05 (0.00 to –0.10) –0.09 (–0.16 to –0.03)
Biomass burning –0.20 (–0.60 to –0.07) –0.20 (–0.60 to –0.07) +0.03(–0.09 to +0.15) –0.0 (–0.20 to +0.20)
Secondary organic aerosol Not estimated Not estimated Not estimated –0.03 (–0.27 to +0.20)
Nitrate Not estimated Not estimated –0.10 (–0.20 to 0.00) –0.11 (–0.30 to –0.03)
Dust Not estimated –0.60 to +0.40 –0.10 (–0.30 to +0.10) –0.10 (–0.30 to +0.10)
Total Not estimated Not estimated –0.50 (–0.90 to –0.10) –0.35 (–0.85 to +0.15)
The RF due to aerosol–radiation interaction during some time periods is
more uncertain than the current RF. Improvements in the observations
of aerosols have been substantial with availability of remote sensing
from the ground-based optical observational network AErosol RObotic
NETwork (AERONET) and the launch of the Moderate Resolution Imag-
ing Spectrometer (MODIS) and Multi-angle Imaging Spectro-Radiom-
eter (MISR) instruments (starting in 2000) as well as other satellite
data. This has contributed to constraining the current RF using aerosol
observations. The aerosol observations are very limited backward in
time, although there is growing constraint coming from new ice and
Table 8.4 | Global and annual mean RF (W m
–2
) due to aerosol–radiation interaction between 1750 and 2011 of seven aerosol components for AR5. Values and uncertainties
from SAR, TAR, AR4 and AR5 are provided when available. Note that for SAR, TAR and AR4 the end year is somewhat different than for AR5 with 1993, 1998 and 2005, respectively.
1850 1900 1950 2000
-0.4
-0.2
0.0
0.2
0.4
Radiative Forcing W m
-2
Sulfate
BC
OC
BioBurn
Nitrate
SOA
BC on snow
Total Aerosols
-0.60
0.80
-0.81
( )
lake core records, and uncertainties in the historical emission of aero-
sols and their precursors used in the global aerosol modeling are larger
than for current conditions. Emissions of carbonaceous aerosols are
particularly uncertain in the 1800s due to a significant biofuel source in
this period, in contrast to the SO
2
emissions which were very small until
the end of the 1800s. The uncertainty in the biomass burning emissions
also increases backward in time. Note that, for 1850, the biomass burn-
ing emissions from Lamarque et al. (2010) are quite different from the
previous estimates, but RF due to aerosol–radiation interaction is close
to zero for this component. Figure 8.8 shows an example of the time
evolution of the RF due to aerosol–radiation interaction as a total and
separated into six aerosol components. From 1950 to 1990 there was
a strengthening of the total RF due to aerosol–radiation interaction,
mainly due to a strong enhancement of the sulphate RF. After 1990 the
change has been small with even a weakening of the RF due to aero-
sol–radiation interaction, mainly due to a stronger BC RF as a result of
increased emissions in East and Southeast Asia.
8.3.4.3 Aerosol–Cloud Interactions
The RF by aerosol effects on cloud albedo was previously referred to
as the Twomey or cloud albedo effect (see Section 7.1). Although this
RF can be calculated, no estimate of this forcing is given because it
has heuristic value only and does not simply translate to the ERF due
to aerosol–cloud interaction. The total aerosol ERF, namely ERF due
to aerosol–radiation and aerosol–cloud interactions (excluding BC on
snow and ice) provided in Chapter 7 is estimated with a 5 to 95%
uncertainty between –1.9 and –0.1 W m
–2
with a best estimate value
of –0.9 W m
–2
(medium confidence). The likely range of this forcing
is between –1.5 and –0.4 W m
–2
. The estimate of ERF due to aero-
sol–radiation and aerosol-cloud interaction is lower (i.e., less negative)
than the corresponding AR4 RF estimate of –1.2 W m
–2
because the
latter was based mainly on GCM studies that did not take secondary
processes (such as aerosol effects on mixed-phase and/or convective
clouds and effects on longwave radiation) into account. This new best
estimate of ERF due to aerosol–radiation and aerosol–cloud interac-
tion is also consistent with the studies allowing cloud-scale process-
es and related responses and with the lower estimates of this forcing
inferred from satellite observations.
Figure 8.8 | Time evolution of RF due to aerosol–radiation interaction and BC on
snow and ice. Multi-model results for 1850, 1930, 1980 and 2000 from ACCMIP for
aerosol–radiation interaction (Shindell et al., 2013c) and BC on snow and ice (Lee et al.,
2013) are combined with higher temporal-resolution results from the Goddard Institute
for Space Studies (GISS)-E2 and Oslo-Chemical Transport Model 2 (OsloCTM2) models
(aerosol–radiation interaction) and OsloCTM2 (BC on snow and ice). Uncertainty ranges
(5 to 95%) for year 2010 are shown with vertical lines. Values next to the uncertainty
lines are for cases where uncertainties go beyond the scale. The total includes the RF
due to aerosol–radiation interaction for six aerosol components and RF due to BC on
snow and ice. All values have been scaled to the best estimates for 2011 given in Table
8.4. Note that time evolution for mineral dust is not included and the total RF due to
aerosol–radiation interaction is estimated based on simulations of the six other aerosol
components.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
684
Frequently Asked Questions
FAQ 8.2 | Do Improvements in Air Quality Have an Effect on Climate Change?
Yes they do, but depending on which pollutant(s) they limit, they can either cool or warm the climate. For example,
whereas a reduction in sulphur dioxide (SO
2
) emissions leads to more warming, nitrogen oxide (NO
x
) emission
control has both a cooling (through reducing of tropospheric ozone) and a warming effect (due to its impact on
methane lifetime and aerosol production). Air pollution can also affect precipitation patterns.
Air quality is nominally a measure of airborne surface pollutants, such as ozone, carbon monoxide, NO
x
and aerosols
(solid or liquid particulate matter). Exposure to such pollutants exacerbates respiratory and cardiovascular diseases,
harms plants and damages buildings. For these reasons, most major urban centres try to control discharges of air-
borne pollutants.
Unlike carbon dioxide (CO
2
) and other well-mixed greenhouse gases, tropospheric ozone and aerosols may last in
the atmosphere only for a few days to a few weeks, though indirect couplings within the Earth system can prolong
their impact. These pollutants are usually most potent near their area of emission or formation, where they can
force local or regional perturbations to climate, even if their globally averaged effect is small.
Air pollutants affect climate differently according to their physical and chemical characteristics. Pollution-generated
greenhouse gases will impact climate primarily through shortwave and longwave radiation, while aerosols can in
addition affect climate through cloud–aerosol interactions.
Controls on anthropogenic emissions of methane (FAQ 8.2, Figure 1) to lower surface ozone have been identified as
‘win–win’ situations. Consequences of controlling other ozone precursors are not always as clear. NO
x
emission con-
trols, for instance, might be expected to have a cooling effect as they reduce tropospheric ozone, but their impact
on CH
4
lifetime and aerosol formation is more likely instead to cause overall warming.
Satellite observations have identified increasing atmospheric concentrations of SO
2
(the primary precursor to scat-
tering sulphate aerosols) from coal-burning power plants over eastern Asia during the last few decades. The most
recent power plants use scrubbers to reduce such emissions (albeit not the concurrent CO
2
emissions and associated
long-term climate warming). This improves air quality, but also reduces the cooling effect of sulphate aerosols and
therefore exacerbates warming. Aerosol cooling occurs through aerosol–radiation and aerosol–cloud interactions
and is estimated at –0.9 W m
–2
(all aerosols combined, Section 8.3.4.3) since pre-industrial, having grown especially
during the second half of the 20th century when anthropogenic emissions rose sharply. (continued on next page)
FAQ 8.2, Figure 1 | Schematic diagram of the impact of pollution controls on specific emissions and climate impact. Solid black line indicates known impact; dashed
line indicates uncertain impact.
Methane
Ammonia
Black
carbon
Organic
carbon
Sulfur
dioxide
Carbon
monoxide
Nitrogen
oxides
VOCs
Cooling
Warming
Ozone pollution
controls
Particulate
matter controls
8
Anthropogenic and Natural Radiative Forcing Chapter 8
685
One reason an expert judgment estimate of ERF due to aerosol–radi-
ation and aerosol–cloud interaction is provided rather than ERF due
to aerosol–cloud interaction specifically is that the individual contri-
butions are very difficult to disentangle. These contributions are the
response of processes that are the outputs from a system that is con-
stantly readjusting to multiple nonlinear forcings. Assumptions of inde-
pendence and linearity are required to deduce ERF due to aerosol–radi-
ation interaction and ERF due to aerosol–cloud interaction (although
there is no a priori reason why the individual ERFs should be simply
additive). Under these assumptions, ERF due to aerosol–cloud interac-
tion is deduced as the difference between ERF due to aerosol–radia-
tion and aerosol–cloud interaction and ERF due to aerosol–radiation
interaction alone. This yields an ERF due to aerosol–cloud interaction
estimate of –0.45 W m
–2
which is much smaller in magnitude than the
–1.4 W m
–2
median forcing value of the models summarized in Figure
7.19 and is also smaller in magnitude than the AR4 estimates of –0.7
W m
–2
for RF due to aerosol–cloud interaction.
8.3.4.4 Black Carbon Deposition in Snow and Ice
Because absorption by ice is very weak at visible and ultraviolet (UV)
wavelengths, BC in snow makes the snow darker and increases absorp-
tion. This is not enough darkening to be seen by eye, but it is enough
to be important for climate (Warren and Wiscombe, 1980; Clarke and
Noone, 1985). Several studies since AR4 have re-examined this issue
and find that the RF may be weaker than the estimates of Hansen
and Nazarenko (2004) in AR4 (Flanner et al., 2007; Koch et al., 2009a;
Rypdal et al., 2009; Lee et al., 2013). The anthropogenic BC on snow/
ice is assessed to have a positive global and annual mean RF of +0.04
W m
−2
, with a 0.02–0.09 W m
−2
5 to 95% uncertainty range (see fur-
ther description in Section 7.5.2.3). This RF has a two to four times
larger global mean surface temperature change per unit forcing than
a change in CO
2
.
In Figure 8.8, the time evolution of global mean RF due to BC on snow
and ice is shown based on multi-model simulations in ACCMIP (Lee et
al., 2013) for 1850, 1930, 1980 and 2000. The results show a maximum
in the RF in 1980 with a small increase since 1850 and a 20% lower
RF in 2000 compared to 1980. Those results are supported by obser-
vations. The BC concentration in the Arctic atmosphere is observed to
be declining since 1990, at least in the Western Hemisphere portion
(Sharma et al., 2004), which should lead to less deposition of BC on
the snow surface. Surveys across Arctic during 1998 and 2005 to 2009
showed that the BC content of Arctic snow appears to be lower than
in 1984 (Doherty et al., 2010) and found BC concentrations in Canada,
Alaska and the Arctic Ocean (e.g., Hegg et al., 2009), about a factor of
2 lower than measured in the 1980s (e.g., Clarke and Noone, 1985).
Large-area field campaigns (Huang et al., 2011; Ye et al., 2012) found
that the BC content of snow in northeast China is comparable to values
found in Europe. The steep drop off in BC content of snow with latitude
in northeast China may indicate that there is not much BC in the Arctic
coming from China (Huang et al., 2011; Ye et al., 2012; Wang et al.,
2013). The change in the spatial pattern of emission of BC is a main
cause for the difference in the temporal development of RF due to BC
on snow and ice compared to the BC from RF due to aerosol–radiation
interaction over the last decades.
FAQ 8.2 (continued)
Black carbon or soot, on the other hand, absorbs heat in the atmosphere (leading to a 0.4 W m
–2
radiative forcing
from anthropogenic fossil and biofuel emissions) and, when deposited on snow, reduces its albedo, or ability to
reflect sunlight. Reductions of black carbon emissions can therefore have a cooling effect, but the additional inter-
action of black carbon with clouds is uncertain and could lead to some counteracting warming.
Air quality controls might also target a specific anthropogenic activity sector, such as transportation or energy pro-
duction. In that case, co-emitted species within the targeted sector lead to a complex mix of chemistry and climate
perturbations. For example, smoke from biofuel combustion contains a mixture of both absorbing and scattering
particles as well as ozone precursors, for which the combined climate impact can be difficult to ascertain.
Thus, surface air quality controls will have some consequences on climate. Some couplings between the targeted
emissions and climate are still poorly understood or identified, including the effects of air pollutants on precipi-
tation patterns, making it difficult to fully quantify these consequences. There is an important twist, too, in the
potential effect of climate change on air quality. In particular, an observed correlation between surface ozone
and temperature in polluted regions indicates that higher temperatures from climate change alone could worsen
summertime pollution, suggesting a ‘climate penalty’. This penalty implies stricter surface ozone controls will be
required to achieve a specific target. In addition, projected changes in the frequency and duration of stagnation
events could impact air quality conditions. These features will be regionally variable and difficult to assess, but
better understanding, quantification and modelling of these processes will clarify the overall interaction between
air pollutants and climate.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
686
8.3.4.5 Contrails and Contrail-Induced Cirrus
AR4 assessed the RF of contrails (persistent linear contrails) as +0.01
(–0.007 to +0.02) W m
–2
and provided no estimate for contrail induced
cirrus. In AR5, Chapter 7 gives a best estimate of RF due to contrails of
+0.01 (+0.005 to +0.03) W m
–2
and an ERF estimate of the combined
contrails and contrail-induced cirrus of +0.05 (+0.02 to +0.15) W
m
–2
. Since AR4, the evidence for contrail-induced cirrus has increased
because of observational studies (for further details see Section 7.2.7).
8.3.5 Land Surface Changes
8.3.5.1 Introduction
Anthropogenic land cover change has a direct impact on the Earth radi-
ation budget through a change in the surface albedo. It also impacts
the climate through modifications in the surface roughness, latent heat
flux and river runoff. In addition, human activity may change the water
cycle through irrigation and power plant cooling, and also generate
direct input of heat to the atmosphere by consuming energy. Land use
change, and in particular deforestation, also has significant impacts on
WMGHG concentration, which are discussed in Section 6.3.2.2. Poten-
tial geo-engineering techniques that aim at increasing the surface
albedo are discussed in Section 7.7.2.3.
AR4 referenced a large number of RF estimates resulting from a change
in land cover albedo. It discussed the uncertainties due to the recon-
struction of historical vegetation, the characterization of present-day
vegetation and the surface radiation processes. On this basis, AR4 gave
a best estimate of RF relative to 1750 due to land use related surface
albedo at –0.2 ± 0.2 Wm
–2
with a level of scientific understanding at
medium-low.
8.3.5.2 Land Cover Changes
Hurtt et al. (2006) estimates that 42 to 68% of the global land sur-
face was impacted by land use activities (crop, pasture, wood harvest)
during the 1700–2000 period. Until the mid-20th century most land
use change took place over the temperate regions of the NH. Since
then, reforestation is observed in Western Europe, North America and
China as a result of land abandonment and afforestation efforts, while
deforestation is concentrated in the tropics. After a rapid increase of
the rate of deforestation during the 1980s and 1990s, satellite data
indicate a slowdown in the past decade (FAO, 2012).
Since AR4, Pongratz et al. (2008) and Kaplan et al. (2011) extended
existing reconstructions on land use back in time to the past millenni-
um, accounting for the progress of agriculture technique and historical
events such as the black death or war invasions. As agriculture was
already widespread over Europe and South Asia by 1750, the RF, which
is defined with respect to this date, is weaker than the radiative flux
change from the state of natural vegetation cover (see Figure 8.9).
Deforestation in Europe and Asia during the last millennium led to a
significant regional negative forcing. Betts et al. (2007) and Goosse
et al. (2006) argue that it probably contributed to the ‘Little Ice Age’,
together with natural solar and volcanic activity components, before
the increase in GHG concentration led to temperatures similar to those
experienced in the early part of the second millennium. There is still
significant uncertainty in the anthropogenic land cover change, and in
particular its time evolution (Gaillard et al., 2010).
8.3.5.3 Surface Albedo and Radiative Forcing
Surface albedo is the ratio between reflected and incident solar flux
at the surface. It varies with the surface cover. Most forests are darker
(i.e., lower albedo) than grasses and croplands, which are darker
than barren land and desert. As a consequence, deforestation tends
to increase the Earth albedo (negative RF) while cultivation of some
bright surfaces may have the opposite effect. Deforestation also leads
to a large increase in surface albedo in case of snow cover as low vege-
tation accumulates continuous snow cover more readily in early winter
allowing it to persist longer in spring. This causes average winter
albedo in deforested areas to be generally much higher than that of a
tree-covered landscape (Bernier et al., 2011).
The pre-industrial impact of the Earth albedo increase due to land use
change, including the reduced snow masking by tall vegetation, is esti-
mated to be on the order of –0.05 W m
–2
(Pongratz et al., 2009). Since
then, the increase in world population and agriculture development
led to additional forcing. Based on reconstruction of land use since the
beginning of the Industrial Era, Betts et al. (2007) and Pongratz et al.
(2009) computed spatially and temporally distributed estimates of the
land use RF. They estimate that the shortwave flux change induced by
the albedo variation, from fully natural vegetation state to 1992, is on
the order of –0.2 W m
–2
(range –0.24 to –0.21W m
–2
). The RF, defined
with respect to 1750, is in the range –0.17 to –0.18 W m
–2
. A slightly
stronger value (–0.22 W m
–2
) was found by Davin et al. (2007) for the
period 1860–1992.
In recent years, the availability of global scale MODIS data (Schaaf et
al., 2002) has improved surface albedo estimates (Rechid et al., 2009).
These data have been used by Myhre et al (2005a) and Kvalevag et al.
(2010). They argue that the observed albedo difference between nat-
ural vegetation and croplands is less than usually assumed in climate
simulations, so that the RF due to land use change is weaker than in
estimates that do not use the satellite data. On the other hand, Nair et
al. (2007) show observational evidence of an underestimate of the sur-
face albedo change in land use analysis in southwest Australia. Overall,
there is still a significant range of RF estimates for the albedo com-
ponent of land use forcing. This is mostly due to the range of albedo
change as a result of land use change, as shown in an inter-comparison
of seven atmosphere–land models (de Noblet-Ducoudre et al., 2012).
Deforestation has a direct impact on the atmospheric CO
2
concen-
tration and therefore contributes to the WMGHG RF as quantified in
Section 8.3.2. Conversely, afforestation is a climate mitigation strate-
gy to limit the CO
2
concentration increase. Several authors have com-
pared the radiative impact of deforestation/afforestation that results
from the albedo change with the greenhouse effect of CO
2
released/
sequestered. Pongratz et al. (2010) shows that the historic land use
change has had a warming impact (i.e., greenhouse effect dominates)
at the global scale and over most regions with the exception of Europe
and India. Bala et al. (2007) results show latitudinal contrast where
the greenhouse effect dominates for low-latitude deforestation while
8
Anthropogenic and Natural Radiative Forcing Chapter 8
687
1992
1900
1750
-4.5 -3 -1 1 3 4.50000
W m
-2
1400 1600 1800 2000
-0.2
-0.1
0.0
SW flux change
(W m
-2
)
-0.1
0.0
Rad. forc. (W m
-2
)
1750
( )
Figure 8.9 | Change in top of the atmosphere (TOA) shortwave (SW) flux (W m
–2
)
following the change in albedo as a result of anthropogenic Land Use Change for three
periods (1750, 1900 and 1992 from top to bottom). By definition, the RF is with respect
to 1750, but some anthropogenic changes had already occurred in 1750. The lower
right inset shows the globally averaged impact of the surface albedo change to the TOA
SW flux (left scale) as well as the corresponding RF (right scale) after normalization to
the 1750 value. Based on simulations by Pongratz et al. (2009).
the combined effect of albedo and evapotranspiration impact does at
high latitude. These results are also supported by Bathiany et al. (2010).
Similarly, Lohila et al. (2010) shows that the afforestation of boreal
peatlands results in a balanced RF between the albedo and green-
house effect. Overall, because of the opposite impacts, the potential of
afforestation to mitigate climate change is limited (Arora and Monte-
negro, 2011) while it may have undesired impacts on the atmospheric
circulation, shifting precipitation patterns (Swann et al., 2012).
8.3.5.4 Other Impacts of Land Cover Change on the
Earth’s Albedo
Burn scars resulting from agriculture practices, uncontrolled fires or
deforestation (Bowman et al., 2009) have a lower albedo than unper-
turbed vegetation (Jin and Roy, 2005). On the other hand, at high lat-
itude, burnt areas are more easily covered by snow, which may result
in an overall increase of the surface albedo. Surface blackening of nat-
ural vegetation due to fire is relatively short lived and typically disap-
pears within one to a few years (Jin et al., 2012). Myhre et al. (2005b)
estimates a global albedo-related radiative effect due to African fires
of 0.015 W m
–2
.
Over semi-arid areas, the development of agriculture favours the gen-
eration of dust. Mulitza et al. (2010) demonstrates a very large increase
of dust emission and deposition in the Sahel concomitant with the
development of agriculture in this area. This, together with the anal-
ysis of dust sources (Ginoux et al., 2010), suggests that a significant
fraction of the dust that is transported over the Atlantic has an anthro-
pogenic origin and impacts the Earth albedo. There is no full estimate
of the resulting RF, however. The dust RF estimate in Section 8.3.4.2
includes both land use contributions and change in wind-driven emis-
sions. Both dust and biomass burning aerosol may impact the Earth
surface albedo as these particles can be deposed on snow, which has
a large impact on its absorption, in particular for soot. This is discussed
in Section 8.3.4.4.
Urban areas have an albedo that is 0.01 to 0.02 smaller than adjacent
croplands (Jin et al., 2005). There is the potential for a strong increase
through white roof coating with the objective of mitigating the heat
island effect (Oleson et al., 2010). Although the global scale impact is
small, local effects can be very large, as shown by Campra et al. (2008)
that reports a regional (260 km
2
) 0.09 increase in albedo and –20 W
m
–2
RF as a consequence of greenhouse horticulture development.
8.3.5.5 Impacts of Surface Change on Climate
Davin et al. (2007) argues that the climate sensitivity to land use forc-
ing is lower than that for other forcings, due to its spatial distribution
but also the role of non-radiative processes. Indeed, in addition to the
impact on the surface albedo, land use change also modifies the evap-
oration and surface roughness, with counterbalancing consequences
on the lower atmosphere temperature. There is increasing evidence
that the impact of land use on evapotranspiration—a non-RF on cli-
mate—is comparable to, but of opposite sign than, the albedo effect,
so that RF is not as useful a metric as it is for gases and aerosols. For
instance, Findell et al. (2007) climate simulations show a negligible
impact of land use change on the global mean temperature, although
there are some significant regional changes.
Numerical climate experiments demonstrate that the impact of land use
on climate is much more complex than just the RF. This is due in part
to the very heterogeneous nature of land use change (Barnes and Roy,
2008), but mostly due to the impact on the hydrological cycle through
evapotranspiration, root depth and cloudiness (van der Molen et al.,
2011). As a consequence, the forcing on climate is not purely radiative
and the net impact on the surface temperature may be either positive
or negative depending on the latitude (Bala et al., 2007). Davin and de
Noblet-Ducoudre (2010) analyses the impact on climate of large-scale
deforestation; the albedo cooling effect dominates for high latitude
whereas reduced evapotranspiration dominates in the tropics. This lat-
itudinal trend is confirmed by observations of the temperature differ-
ence between open land and nearby forested land (Lee et al., 2011).
Irrigated areas have continuously increased during the 20th century
although a slowdown has been observed in recent decades (Bonfils
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
688
and Lobell, 2007). There is clear evidence that irrigation leads to local
cooling of several degrees (Kueppers et al., 2007). Irrigation also
affects cloudiness and precipitation (Puma and Cook, 2010). In the
United States, DeAngelis et al. (2010) found that irrigation in the Great
Plains in the summer produced enhanced precipitation in the Midwest
1000 km to the northeast.
8.3.5.6 Conclusions
There is still a rather wide range of estimates of the albedo change
due to anthropogenic land use change, and its RF. Although most
published studies provide an estimate close to –0.2 Wm
–2
, there is
convincing evidence that it may be somewhat weaker as the albedo
difference between natural and anthropogenic land cover may have
been overestimated. In addition, non-radiative impact of land use have
a similar magnitude, and may be of opposite sign, as the albedo effect
(though these are not part of RF). A comparison of the impact of land
use change according to seven climate models showed a wide range of
results (Pitman et al., 2009), partly due to difference in the implemen-
tation of land cover change, but mostly due to different assumptions
on ecosystem albedo, plant phenology and evapotranspiration. There
is no agreement on the sign of the temperature change induced by
anthropogenic land use change. It is very likely that land use change
led to an increase of the Earth albedo with a RF of –0.15 ± 0.10 W m
–2
,
but a net cooling of the surface—accounting for processes that are not
limited to the albedo—is about as likely as not.
8.4 Natural Radiative Forcing Changes:
Solar and Volcanic
Several natural drivers of climate change operate on multiple time
scales. Solar variability takes place at many time scales that include
centennial and millennial scales (Helama et al., 2010), as the radiant
energy output of the Sun changes. Also, variations in the astronomical
alignment of the Sun and the Earth (Milankovitch cycles) induce cycli-
cal changes in RF, but this is substantial only at millennial and longer
time scales (see Section 5.2.1.1). Volcanic forcing is highly episodic,
but can have dramatic, rapid impacts on climate. No major asteroid
impacts occurred during the reference period (1750–2012) and thus
this effect is not considered here. This section discusses solar and
volcanic forcings, the two dominant natural contributors of climate
change since the pre-industrial time.
8.4.1 Solar Irradiance
In earlier IPCC reports the forcing was estimated as the instantaneous
RF at TOA. However, due to wavelength-albedo dependence, solar radi-
ation-wavelength dependence and absorption within the stratosphere
and the resulting stratospheric adjustment, the RF is reduced to about
78% of the TOA instantaneous RF (Gray et al., 2009). There is low con-
fidence in the exact value of this number, which can be model and time
scale dependent (Gregory et al., 2004; Hansen et al., 2005). AR4 gives
an 11-year running mean instantaneous TOA RF between 1750 and
the present of 0.12 W m
–2
with a range of estimates of 0.06 to 0.30 W
m
–2
, equivalent to a RF of 0.09 W m
–2
with a range of 0.05 to 0.23 W
m
–2
. For a consistent treatment of all forcing agents, hereafter we use
RF while numbers quoted from AR4 will be provided both as RF and
instantaneous RF at TOA.
8.4.1.1 Satellite Measurements of Total Solar Irradiance
Total solar irradiance (TSI) measured by the Total Irradiance Monitor
(TIM) on the spaceborne Solar Radiation and Climate Experiment
(SORCE) is 1360.8 ± 0.5 W m
–2
during 2008 (Kopp and Lean, 2011)
which is ~4.5 W m
–2
lower than the Physikalisch-Meteorologisches
Observatorium Davos (PMOD) TSI composite during 2008 (Frohlich,
2009).The difference is probably due to instrumental biases in meas-
urements prior to TIM. Measurements with the PREcision MOnitor
Sensor (PREMOS) instrument support the TIM absolute values (Kopp
and Lean, 2011). The TIM calibration is also better linked to national
standards which provides further support that it is the most accurate
(see Supplementary Material Section 8.SM.6). Given the lower TIM TSI
values relative to currently used standards, most general circulation
models are calibrated to incorrectly high values. However, the few
tenths of a percent bias in the absolute TSI value has minimal con-
sequences for climate simulations because the larger uncertainties in
cloud properties have a greater effect on the radiative balance. As the
maximum-to-minimum TSI relative change is well-constrained from
observations, and historical variations are calculated as changes rela-
tive to modern values, a revision of the absolute value of TSI affects RF
by the same fraction as it affects TSI. The downward revision of TIM TSI
with respect to PMOD, being 0.3%, thus has a negligible impact on RF,
which is given with a relative uncertainty of several tenths of a percent.
Since 1978, several independent space-based instruments have direct-
ly measured the TSI. Three main composite series were constructed,
referred to as the Active Cavity Radiometer Irradiance Monitor (ACRIM)
(Willson and Mordvinov, 2003), the Royal Meteorological Institute of
Belgium (RMIB) (Dewitte et al., 2004) and the PMOD (Frohlich, 2006)
series. There are two major differences between ACRIM and PMOD.
The first is the rapid drift in calibration between PMOD and ACRIM
before 1981. This arises because both composites employ the Hickey–
Frieden (HF) radiometer data for this interval, while a re-evaluation of
the early HF degradation has been implemented by PMOD but not by
ACRIM. The second one, involving also RMIB, is the bridging of the gap
between the end of ACRIM I (mid-1989) and the beginning of ACRIM
II (late 1991) observations, as it is possible that a change in HF data
occurred during this gap. This possibility is neglected in ACRIM and
thus its TSI increases by more than 0.5 W m
–2
during solar cycle (SC)
22. These differences lead to different long-term TSI trends in the three
composites (see Figure 8.10): ACRIM rises until 1996 and subsequently
declines, RMIB has an upward trend through 2008 and PMOD shows a
decline since 1986 which unlike the other two composites, follows the
solar-cycle-averaged sunspot number (Lockwood, 2010). Moreover, the
ACRIM trend implies that the TSI on time scales longer than the SC is
positively correlated with the cosmic ray variation indicating a decline
in TSI throughout most of the 20th century (the opposite to most TSI
reconstructions produced to date; see Section 8.4.1.2). Furthermore,
extrapolating the ACRIM TSI long-term drift would imply a brighter
Sun in the Maunder minimum (MM) than now, again opposite to most
TSI reconstructions (Lockwood and Frohlich, 2008). Finally, analysis
of instrument degradation and pointing issues (Lee et al., 1995) and
independent modeling based on solar magnetograms (Wenzler et al.,
8
Anthropogenic and Natural Radiative Forcing Chapter 8
689
2009; Ball et al., 2012), confirm the need for correction of HF data, and
we conclude that PMOD is more accurate than the other composites.
TSI variations of approximately 0.1% were observed between the
maximum and minimum of the 11-year SC in the three composites
mentioned above (Kopp and Lean, 2011). This variation is mainly due
to an interplay between relatively dark sunspots, bright faculae and
bright network elements (Foukal and Lean, 1988; see Section 5.2.1.2).
A declining trend since 1986 in PMOD solar minima is evidenced in
Figure 8.10. Considering the PMOD solar minima values of 1986 and
2008, the RF is –0.04 W m
–2
. Our assessment of the uncertainty range
of changes in TSI between 1986 and 2008 is –0.08 to 0.0 W m
–2
and
thus very likely negative, and includes the uncertainty in the PMOD
data (Frohlich, 2009; see Supplementary Material Section 8.SM.6) but
is extended to also take into account the uncertainty of combining the
satellite data.
For incorporation of TIM data with the previous and overlapping data,
in Figure 8.10 we have standardized the composite time series to the
TIM series (over 2003–2012, the procedure is explained in Supplemen-
tary Material Section 8.SM.6. Moreover as we consider annual averag-
es, ACRIM and PMOD start at 1979 because for 1978 both composites
have only two months of data.
8.4.1.2 Total Solar Irradiance Variations Since Preindustrial Time
The year 1750, which is used as the preindustrial reference for estimat-
ing RF, corresponds to a maximum of the 11-year SC. Trend analysis are
usually performed over the minima of the solar cycles that are more
stable. For such trend estimates, it is then better to use the closest
SC minimum, which is in 1745. To avoid trends caused by compar-
ing different portions of the solar cycle, we analyze TSI changes using
multi-year running means. For the best estimate we use a recent TSI
reconstruction by Krivova et al. (2010) between 1745 and 1973 and
from 1974 to 2012 by Ball et al. (2012). The reconstruction is based
( )
Figure 8.10 | Annual average composites of measured total solar irradiance: The
Active Cavity Radiometer Irradiance Monitor (ACRIM) (Willson and Mordvinov, 2003),
the Physikalisch-Meteorologisches Observatorium Davos (PMOD) (Frohlich, 2006) and
the Royal Meteorological Institute of Belgium (RMIB) (Dewitte et al., 2004).These com-
posites are standardized to the annual average (2003–2012) Total Irradiance Monitor
(TIM) (Kopp and Lean, 2011) measurements that are also shown.
on physical modeling of the evolution of solar surface magnetic flux,
and its relationship with sunspot group number (before 1974) and
sunspot umbra and penumbra and faculae afterwards. This provides
a more detailed reconstruction than other models (see the time series
in Supplementary Material Table 8.SM.3). The best estimate from our
assessment of the most reliable TSI reconstruction gives a 7-year run-
ning mean RF between the minima of 1745 and 2008 of 0.05 W m
–2
.
Our assessment of the range of RF from TSI changes is 0.0 to 0.10
W m
–2
which covers several updated reconstructions using the same
7-year running mean past-to-present minima years (Wang et al., 2005;
Steinhilber et al., 2009; Delaygue and Bard, 2011), see Supplementa-
ry Material Table 8.SM.4. All reconstructions rely on indirect proxies
that inherently do not give consistent results. There are relatively large
discrepancies among the models (see Figure 8.11).With these consid-
erations, we adopt this value and range for AR5. This RF is almost half
of that in AR4, in part because the AR4 estimate was based on the
previous solar cycle minimum while the AR5 estimate includes the drop
of TSI in 2008 compared to the previous two SC minima (see 8.4.1).
Concerning the uncertainty range, in AR4 the upper limit corresponded
to the reconstruction of Lean (2000), based on the reduced brightness
of non-cycling Sun-like stars assumed typical of a Maunder minimum
(MM) state. The use of such stellar analogues was based on the work
of Baliunas and Jastrow (1990), but more recent surveys have not
reproduced their results and suggest that the selection of the original
set was flawed (Hall and Lockwood, 2004; Wright, 2004); the lower
limit from 1750 to present in AR4 was due to the assumed increase
in the amplitude of the 11-year cycle only. Thus the RF and uncertain-
ty range have been obtained in a different way in AR5 compared to
AR4. Maxima to maxima RF give a higher estimate than minima to
minima RF, but the latter is more relevant for changes in solar activity.
Given the medium agreement and medium evidence, this RF value has
a medium confidence level (although confidence is higher for the last
three decades). Figure 8.11 shows several TSI reconstructions modelled
using sunspot group numbers (Wang et al., 2005; Krivova et al., 2010;
( )
Figure 8.11 | Reconstructions of total solar irradiance since1745; annual resolution
series from Wang et al. (2005) with and without an independent change in the back-
ground level of irradiance, Krivova et al. (2010) combined with Ball et al. (2012) and
5-year time resolution series from Steinhilber et al. (2009) and Delaygue and Bard
(2011). The series are standardized to the Physikalisch-Meteorologisches Observato-
rium Davos (PMOD) measurements of solar cycle 23 (1996–2008) (PMOD is already
standardized to Total Irradiance Monitor).
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
690
Ball et al., 2012) and sunspot umbra and penumbra and faculae (Ball
et al., 2012), or cosmogenic isotopes (Steinhilber et al., 2009; Delaygue
and Bard, 2011). These reconstructions are standardized to PMOD SC
23 (1996–2008) (see also Supplementary Material Section 8.SM.6).
For the MM-to-present AR4 gives a TOA instantaneous RF range of
0.1 to 0.28 W m
–2
, equivalent to 0.08 to 0.22 Wm
–2
with the RF defi-
nition used here. The reconstructions in Schmidt et al. (2011) indicate
a MM-to-present RF range of 0.08 to 0.18 W m
–2
, which is within
the AR4 range although narrower. As discussed above, the estimates
based on irradiance changes in Sun-like stars are not included in this
range because the methodology has been shown to be flawed. A more
detailed explanation of this is found in Supplementary Material Section
8.SM.6. For details about TSI reconstructions on millennia time scales
see Section 5.2.1.2.
8.4.1.3 Attempts to Estimate Future Centennial Trends of
Total Solar Irradiance
Cosmogenic isotope and sunspot data (Rigozo et al., 2001; Solanki and
Krivova, 2004; Abreu et al., 2008) reveal that currently the Sun is in a
grand activity maximum that began about 1920 (20th century grand
maximum). However, SC 23 showed an activity decline not previous-
ly seen in the satellite era (McComas et al., 2008; Smith and Balogh,
2008; Russell et al., 2010). Most current estimations suggest that the
forthcoming solar cycles will have lower TSI than those for the past 30
years (Abreu et al., 2008; Lockwood et al., 2009; Rigozo et al., 2010;
Russell et al., 2010). Also there are indications that the mean magnetic
field in sunspots may be diminishing on decadal level. A linear expan-
sion of the current trend may indicate that of the order of half the
sunspot activity may disappear by about 2015 (Penn and Livingston,
2006). These studies only suggest that the Sun may have left the 20th
century grand maximum and not that it is entering another grand min-
imum. But other works propose a grand minimum during the 21st cen-
tury, estimating an RF within a range of -0.16 to 0.12 Wm
–2
between
this future minimum and the present-day TSI (Jones et al., 2012). How-
ever, much more evidence is needed and at present there is very low
confidence concerning future solar forcing estimates.
Nevertheless, even if there is such decrease in the solar activity, there
is a high confidence that the TSI RF variations will be much smaller
in magnitude than the projected increased forcing due to GHG (see
Section 12.3.1).
8.4.1.4 Variations in Spectral Irradiance
8.4.1.4.1 Impacts of ultraviolet variations on the stratosphere
Ozone is the main gas involved in stratospheric radiative heating.
Ozone production rate variations are largely due to solar UV irradi-
ance changes (HAIGH, 1994), with observations showing statistical-
ly significant variations in the upper stratosphere of 2 to 4% along
the SC (Soukharev and Hood, 2006). UV variations may also produce
transport-induced ozone changes due to indirect effects on circulation
(Shindell et al., 2006b). In addition, statistically significant evidence for
an 11-year variation in stratospheric temperature and zonal winds is
attributed to UV radiation (Frame and Gray, 2010). The direct UV heat-
ing of the background ozone is dominant and over twice as large as
the ozone heating in the upper stratosphere and above, while indirect
solar and terrestrial radiation through the SC-induced ozone change
is dominant below about 5 hPa (Shibata and Kodera, 2005). The RF
due to solar-induced ozone changes is a small fraction of the solar RF
discussed in Section 8.4.1.1 (Gray et al., 2009).
8.4.1.4.2 Measurements of spectral irradiance
Solar spectral irradiance (SSI) variations in the far (120 to 200 nm)
and middle (200 to 300 nm) ultraviolet (UV) are the primary driver for
heating, composition, and dynamic changes of the stratosphere, and
although these wavelengths compose a small portion of the incoming
radiation they show large relative variations between the maximum
and minimum of the SC compared to the corresponding TSI chang-
es. As UV heating of the stratosphere over a SC has the potential to
influence the troposphere indirectly, through dynamic coupling, and
therefore climate (Haigh, 1996; Gray et al., 2010), the UV may have a
more significant impact on climate than changes in TSI alone would
suggest. Although this indicates that metrics based only on TSI are not
appropriate, UV measurements present several controversial issues
and modelling is not yet robust.
Multiple space-based measurements made in the past 30 years indi-
cated that UV variations account for about 30% of the SC TSI varia-
tions, while about 70% were produced within the visible and infrared
(Rottman, 2006). However, current models and data provide the range
of 30 to 90% for the contribution of the UV variability below 400 nm to
TSI changes (Ermolli et al., 2013), with a more probable value of ~60%
(Morrill et al., 2011; Ermolli et al., 2013). The Spectral Irradiance Mon-
itor (SIM) on board SORCE (Harder et al., 2009) shows, over the SC 23
declining phase, measurements that are rather inconsistent with prior
understanding, indicating that additional validation and uncertainty
estimates are needed (DeLand and Cebula, 2012; Lean and Deland,
2012). A wider exposition can be found in Supplementary Material
Section 8.SM.6.
8.4.1.4.3 Reconstructions of preindustrial ultraviolet variations
The Krivova et al. (2010) reconstruction is based on what is known
about spectral contrasts of different surface magnetic features and the
relationship between TSI and magnetic fields. The authors interpolated
backwards to the year 1610 based on sunspot group numbers and
magnetic information. The Lean (2000) model is based on historical
sunspot number and area and is scaled in the UV using measurements
from the Solar Stellar Irradiance Comparison Experiment (SOLSTICE)
on board the Upper Atmosphere Research Satellite (UARS). The results
show smoothed 11-year UV SSI changes between 1750 and the pres-
ent of about 25% at about 120 nm, about 8% at 130 to 175 nm, ~4%
at 175 to 200 nm, and about 0.5% at 200 to 350 nm. Thus, the UV SSI
appears to have generally increased over the past four centuries, with
larger trends at shorter wavelengths. As few reconstructions are avail-
able, and recent measurements suggest a poor understanding of UV
variations and their relationship with solar activity, there is very low
confidence in these values.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
691
8.4.1.5 The Effects of Cosmic Rays on Clouds
Changing cloud amount or properties modify the Earth’s albedo and
therefore affect climate. It has been hypothesized that cosmic ray flux
create atmospheric ions which facilitates aerosol nucleation and new
particle formation with a further impact on cloud formation (Dickinson,
1975; Kirkby, 2007). High solar activity means a stronger heliospheric
magnetic field and thus a more efficient screen against cosmic rays.
Under the hypothesis underlined above, the reduced cosmic ray flux
would promote fewer clouds amplifying the warming effect expected
from high solar activity. There is evidence from laboratory, field and
modelling studies that ionization from cosmic ray flux may enhance
aerosol nucleation in the free troposphere (Merikanto et al., 2009;
Mirme et al., 2010; Kirkby et al., 2011). However, there is high con-
fidence (medium evidence and high agreement) that the cosmic ray–
ionization mechanism is too weak to influence global concentrations
of cloud condensation nuclei or their change over the last century or
during a SC in a climatically significant way (Harrison and Ambaum,
2010; Erlykin and Wolfendale, 2011; Snow-Kropla et al., 2011). A
detailed exposition is found in Section 7.4.6.
8.4.2 Volcanic Radiative Forcing
8.4.2.1 Introduction
Volcanic eruptions that inject substantial amounts of SO
2
gas into
the stratosphere are the dominant natural cause of externally forced
climate change on the annual and multi-decadal time scales, both
because of the multi-decadal variability of eruptions and the time
scale of the climate system response, and can explain much of the
pre-industrial climate change of the last millennium (Schneider et
al., 2009; Brovkin et al., 2010; Legras et al., 2010; Miller et al., 2012).
Although volcanic eruptions inject both mineral particles (called ash
or tephra) and sulphate aerosol precursor gases (predominantly SO
2
)
into the atmosphere, it is the sulphate aerosols, which because of their
small size are effective scatterers of sunlight and have long lifetimes,
that are responsible for RF important for climate. Global annually aver-
aged emissions of CO
2
from volcanic eruptions since 1750 have been
at least 100 times smaller than anthropogenic emissions and incon-
sequential for climate on millennial and shorter time scales (Gerlach,
2011). To be important for climate change, sulphur must be injected
into the stratosphere, as the lifetime of aerosols in the troposphere is
only about one week, whereas sulphate aerosols in the stratosphere
from tropical eruptions have a lifetime of about one year, and those
from high-latitude eruptions last several months. Most stratospheric
aerosols are from explosive eruptions that directly put sulphur into the
stratosphere, but Bourassa et al. (2012, 2013) showed that sulphur
injected into the upper troposphere can then be lifted into the strato-
sphere over the next month or two by deep convection and large scale
Asian summer monsoon circulation, although Vernier et al. (2013) and
Fromm et al. (2013) suggested that direct injection was also important.
Robock (2000), AR4 (Forster et al., 2007) and Timmreck (2012) provide
summaries of this relatively well understood forcing agent.
There have been no major volcanic eruptions since Mt Pinatubo in
1991 (Figure 8.12), but several smaller eruptions have caused a RF for
the years 2008–2011 of –0.11 (–0.15 to –0.08) W m
–2
, approximately
twice the magnitude of the 1999–2002 RF of –0.06 (–0.08 to –0.04)
W m
–2
, consistent with the trends noted in Solomon et al. (2011).
However, the CMIP5 simulations discussed elsewhere in this report
did not include the recent small volcanic forcing in their calculations.
New work has also produced a better understanding of high latitude
eruptions, the hydrological response to volcanic eruptions (Trenberth
and Dai, 2007; Anchukaitis et al., 2010), better long-term records of
past volcanism and better understanding of the effects of very large
eruptions.
There are several ways to measure both the SO
2
precursor and sul-
phate aerosols in the stratosphere, using balloons, airplanes, and both
ground- and satellite-based remote sensing. Both the infrared and
ultraviolet signals sensed by satellite instruments can measure SO
2
,
and stratospheric aerosol measurements by space-based sensors have
been made on a continuous basis since 1978 by a number of instru-
ments employing solar and stellar occultation, limb scattering, limb
emission, and lidar strategies (Thomason and Peter, 2006; Kravitz et al.,
2011; Solomon et al., 2011).
Forster et al. (2007) described four mechanisms by which volcanic
forcing influences climate: RF due to aerosol–radiation interaction;
differential (vertical or horizontal) heating, producing gradients and
changes in circulation; interactions with other modes of circulation,
such as El Niño-Southern Oscillation (ENSO); and ozone depletion with
its effects on stratospheric heating, which depends on anthropogenic
chlorine (stratospheric ozone would increase with a volcanic eruption
under low-chlorine conditions). In addition, the enhanced diffuse light
from volcanic aerosol clouds impacts vegetation and hence the carbon
cycle (Mercado et al., 2009) and aerosol–cloud interaction of sulphate
aerosols on clouds in the troposphere can also be important (Schmidt
et al., 2010), though Frolicher et al. (2011) showed that the impacts of
the 1991 Mt Pinatubo eruption on the carbon cycle were small.
8.4.2.2 Recent Eruptions
The background stratospheric aerosol concentration was affected by
several small eruptions in the first decade of the 21st century (Nagai et
al., 2010; Vernier et al., 2011; Neely et al., 2013; see also Figure 8.13),
with a very small contribution from tropospheric pollution (Siddaway
and Petelina, 2011; Vernier et al., 2011), and had a small impact on RF
(Solomon et al., 2011). Two recent high-latitude eruptions, of Kasa-
tochi Volcano (52.1°N, 175.3°W) on August 8, 2008 and of Sarychev
Volcano (48.1°N, 153.2°E) on June 12–16, 2009, each injected ~1.5
Tg(SO
2
) into the stratosphere, but did not produce detectable climate
response. Their eruptions, however, led to better understanding of the
dependence of the amount of material and time of year of high-lat-
itude injections to produce climate impacts (Haywood et al., 2010;
Kravitz et al., 2010, 2011). The RF from high-latitude eruptions is a
function of seasonal distribution of insolation and the 3- to 4-month
lifetime of high-latitude volcanic aerosols. Kravitz and Robock (2011)
showed that high-latitude eruptions must inject at least 5 Tg(SO
2
) into
the lower stratosphere in the spring or summer, and much more in fall
or winter, to have a detectible climatic response.
On April 14, 2010 the Eyjafjallajökull volcano in Iceland (63.6°N,
19.6°W) began an explosive eruption phase that shut down air traffic
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
692
in Europe for 6 days and continued to disrupt it for another month. The
climatic impact of Eyjafjallajökull was about 10,000 times less than
that of Mt Pinatubo; however, because it emitted less than 50 ktonnes
SO
2
and its lifetime in the troposphere was 50 times less than if it had
been injected into the stratosphere, and was therefore undetectable
amidst the chaotic weather noise in the atmosphere (Robock, 2010).
2011 saw the continuation of a number of small eruptions with signif-
icant tropospheric SO
2
and ash injections, including Puyehue-Cordón
Caulle in Chile, Nabro in Eritrea, and Grimsvötn in Iceland. None have
been shown to have produced an important RF, but the June 13, 2011
Nabro eruption resulted in the largest stratospheric aerosol cloud since
the 1991 Mt Pinatubo eruption (Bourassa et al., 2012), more than 1.5
Tg(SO
2
).
Figure 8.12 shows reconstructions of volcanic aerosol optical depth
since 1750. Figure 8.13 shows details of the vertical distribution of
stratospheric aerosols in the tropics since 1985. The numerous small
eruptions in the past decade are evident, but some of them were at
higher latitudes and their full extent is not captured in this plot.
Figure 8.13 | (Top) Monthly mean extinction ratio (525 nm) profile evolution in the tropics [20°N to 20°S] from January 1985 through December 2012 derived from Stratospheric
Aerosol and Gas Experiment (SAGE) II extinction in 1985–2005 and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) scattering ratio in 2006–2012,
after removing clouds below 18 km based on their wavelength dependence (SAGE II) and depolarization properties (CALIPSO) compared to aerosols. Black contours represent the
extinction ratio in log-scale from 0.1 to 100. The position of each volcanic eruption occurring during the period is displayed with its first two letters on the horizontal axis, where
tropical eruptions are noted in red. The eruptions were Nevado del Ruiz (Ne), Augustine (Au), Chikurachki (Ch), Kliuchevskoi (Kl), Kelut (Ke), Pinatubo (Pi), Cerro Hudson (Ce), Spur
(Sp), Lascar (La), Rabaul (Ra), Ulawun (Ul), Shiveluch (Sh), Ruang (Ru), Reventador (Re), Manam (Ma), Soufrière Hills (So), Tavurvur (Ta), Okmok (Ok), Kasatochi (Ka), Victoria (Vi*—
forest fires with stratospheric aerosol injection), Sarychev (Sa), Merapi (Me), Nabro (Na). (Updated from Figure 1 from Vernier et al., 2011.) (Bottom) Mean stratospheric aerosol
optical depth (AOD) in the tropics [20°N to 20°S] between the tropopause and 40 km since 1985 from the SAGE II (black line), the Global Ozone Monitoring by Occultation of Stars
(GOMOS) (red line), and CALIPSO (blue line). (Updated from Figure 5 from Vernier et al., 2011.)
Figure 8.12 | Volcanic reconstructions of global mean aerosol optical depth (at 550
nm). Gao et al. (2008) and Crowley and Unterman (2013) are from ice core data, and
end in 2000 for Gao et al. (2008) and 1996 for Crowley and Unterman (2013). Sato et
al. (1993) includes data from surface and satellite observations, and has been updated
through 2011. (Updated from Schmidt et al., 2011.)
8
Anthropogenic and Natural Radiative Forcing Chapter 8
693
8.4.2.3 Records of Past Volcanism and Effects of Very
Large Eruptions
Although the effects of volcanic eruptions on climate are largest in
the 2 years following a large stratospheric injection, and the winter
warming effect in the NH has been supported by long-term records
(Fischer et al., 2007), there is new work indicating extended volcanic
impacts via long-term memory in the ocean heat content and sea level
(Stenchikov et al., 2009; Gregory, 2010; Otterä et al., 2010). Zanchet-
tin et al. (2012) found changes in the North Atlantic Ocean circulation
that imply strengthened northward oceanic heat transport a decade
after major eruptions, which contributes to the emergence of extensive
winter warming over the continental NH along with persistent cooling
over Arctic regions on decadal time scales, in agreement with Zhong et
al. (2011) and Miller et al. (2012).
New work on the mechanisms by which a supereruption (Self and
Blake, 2008) could force climate has focused on the 74,000 BP eruption
of the Toba volcano (2.5°N, 99.0°E). Robock et al. (2009) used simu-
lations of up to 900 times the 1991 Mt Pinatubo sulphate injection to
show that the forcing is weaker than that predicted based on a linear
relationship with the sulphate aerosol injection. The results agreed
with a previous simulation by Jones et al. (2005). They also showed
that chemical interactions with ozone had small impacts on the forcing
and that the idea of Bekki et al. (1996) that water vapour would limit
and prolong the growth of aerosols was not supported. Timmreck et al.
(2010) however, incorporating the idea of Pinto et al. (1989) that aer-
osols would grow and therefore both have less RF per unit mass and
fall out of the atmosphere more quickly, found much less of a radiative
impact from such a large stratospheric input.
8.4.2.4 Future Effects
We expect large eruptions over the next century but cannot predict
when. Ammann and Naveau (2003) and Stothers (2007) suggested an
80-year periodicity in past eruptions, but the data record is quite short
and imperfect, and there is no mechanism proposed that would cause
this. While the period 1912–1963 was unusual for the past 500 years in
having no large volcanic eruptions, and the period 1250–1300 had the
most globally climatically significant eruptions in the past 1500 years
(Gao et al., 2008), current knowledge only allows us to predict such
periods on a statistical basis, assuming that the recent past distribu-
tions are stationary. Ammann and Naveau (2003), Gusev (2008), and
Deligne et al. (2010) studied these statistical properties and Ammann
and Naveau (2010) showed how they could be used to produce a sta-
tistical distribution for future simulations. Although the future forcing
from volcanic eruptions will depend only on the stratospheric aerosol
loading for most forcing mechanisms, the future effects on reducing
ozone will diminish as ozone depleting substances diminish in the
future (Eyring et al., 2010b).
8.5 Synthesis of Global Mean Radiative
Forcing, Past and Future
The RF can be used to quantify the various agents that drive climate
change over the Industrial Era or the various contributions to future
climate change. There are multiple ways in which RF can be attribut-
ed to underlying causes, each providing various perspectives on the
importance of the different factors driving climate change. This section
evaluates the RF with respect to emitted component and with respect
to the ultimate atmospheric concentrations. The uncertainties in the RF
Box 8.3 | Volcanic Eruptions as Analogues
Volcanic eruptions provide a natural experiment of a stratospheric aerosol cloud that can serve to inform us of the impacts of the pro-
posed production of such a cloud as a means to control the climate, which is one method of geoengineering (Rasch et al., 2008); see
Section 7.7. For example, Trenberth and Dai (2007) showed that the Asian and African summer monsoon, as well as the global hydro-
logical cycle, was weaker for the year following the 1991 Mt Pinatubo eruption, which is consistent with climate model simulations
(Robock et al., 2008). MacMynowski et al. (2011) showed that because the climate system response of the hydrological cycle is rapid,
forcing from volcanic eruptions, which typically last about a year, can serve as good analogues for longer-lived forcing. The formation
of sulphate aerosols, their transport and removal, their impacts on ozone chemistry, their RF, and the impacts on whitening skies all
also serve as good analogues for geoengineering proposals. Volcanic impacts on the carbon cycle because of more diffuse radiation
(Mercado et al., 2009) and on remote sensing can also be useful analogues, and the impacts of contrail-generated sub-visual cirrus
(Long et al., 2009) can be used to test the long-term impacts of a permanent stratospheric cloud.
Smoke from fires generated by nuclear explosions on cities and industrial areas, which could be lofted into the stratosphere, would
cause surface cooling and a reduction of stratospheric ozone (Mills et al., 2008). Volcanic eruptions that produce substantial strato-
spheric aerosol clouds also serve as an analogue that supports climate model simulations of the transport and removal of stratospheric
aerosols, their impacts on ozone chemistry, their RF, and the climate response. The use of the current global nuclear arsenal still has
the potential to produce nuclear winter, with continental temperatures below freezing in summer (Robock et al., 2007a; Toon et al.,
2008), and the use of only 100 nuclear weapons could produce climate change unprecedented in recorded human history (Robock et
al., 2007b), with significant impacts on global agriculture (Özdoğan et al., 2013; Xia and Robock, 2013).
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
694
agents vary and the confidence levels for these are presented in this
section. Finally, this section shows historical and scenarios of future
time evolution of RF.
8.5.1 Summary of Radiative Forcing by Species and
Uncertainties
Table 8.5 has an overview of the RF agents considered here and each of
them is given a confidence level for the change in RF over the Industrial
Era to the present day. The confidence level is based on the evidence
(robust, medium, and limited) and the agreement (high, medium, and
low; see further description in Chapter 1). The confidence level of the
forcing agents goes beyond the numerical values available in estimates
and is an assessment for a particular forcing agent to have a real
Evidence Agreement
Confidence
Level
Basis for Uncertainty Estimates
(more certain / less certain)
Change in Under-
standing Since AR4
Well-mixed
greenhouse gases
Robust High Very high
Measured trends from different observed data sets and differences
between radiative transfer models
No major change
Tropospheric
ozone
Robust Medium High
Observed trends of ozone in the troposphere and model results for
the industrial era/Differences between model estimates of RF
No major change
Stratospheric
ozone
Robust Medium High
Observed trends in stratospheric and total ozone and model-
ling of ozone depletion/Differences between estimates of RF
No major change
Stratospheric
water vapour
from CH
4
Robust Low Medium
Similarities in results of independent methods to estimate the
RF/Known uncertainty in RF calculations
Elevated owing to more studies
Aerosol–radiation
interactions
Robust Medium High
A large set of observations and converging independent estimates of
RF/Differences between model estimates of RF
Elevated owing to more robust esti-
mates from independent methods
Aerosol–cloud
interactions
Medium Low Low
Variety of different observational evidence and modelling activities/
Spread in model estimates of ERF and differences between
observations and model results
ERF in AR5 has a similar
confidence level to RF in AR4
Rapid adjustment
aerosol–radiation
interactions
Medium Low Low
Observational evidence combined with results from different types of
models/Large spread in model estimates
Elevated owing to
increased evidence
Total aerosol
effect
Medium Medium Medium
A large set of observations and model results, independent methods
to derive ERF estimates/Aerosol–cloud interaction processes
and anthropogenic fraction of CCN still fairly uncertain
Not provided previously
Surface albedo
(land use)
Robust Medium High
Estimates of deforestation for agricultural purposes and well known
physical processes/Spread in model estimates of RF
Elevated owing to the availability
of high-quality satellite data
Surface albedo
(BC aerosol on
snow and ice)
Medium Low Low
Observations of snow samples and the link between BC content
in snow and albedo/Large spread in model estimates of RF
No major change
Contrails Robust Low Medium
Contrails observations , large number of model estimates/Spread in
model estimates of RF and uncertainties in contrail optical properties
Elevated owing to more studies
Contrail- induced
cirrus
Medium Low Low
Observations of a few events of contrail induced cirrus/Extent of
events uncertain and large spread in estimates of ERF
Elevated owing to additional
studies increasing the evidence
Solar irradiance Medium Medium Medium
Satellite information over recent decades and small uncertainty
in radiative transfer calculations/Large relative spread in
reconstructions based on proxy data
Elevated owing to better
agreement of a weak RF
Volcanic aerosol Robust Medium High
Observations of recent volcanic eruptions/Reconstructions of
past eruptions
Elevated owing to improved
understanding
Table 8.5 | Confidence level for the forcing estimate associated with each forcing agent for the 1750–2011 period. The confidence level is based on the evidence and the agree-
ment as given in the table. The basis for the confidence level and change since AR4 is provided. See Figure 1.11 for further description of the evidence, agreement and confidence
level. The colours are adopted based on the evidence and agreement shown in Figure 1.11. Dark green is “High agreement and Robust evidence”, light green is either “High
agreement and Medium evidence” or “Medium agreement and Robust evidence”, yellow is either “High agreement and limited evidence” or “Medium agreement and Medium
evidence” or “Low agreement and Robust evidence”, orange is either “Medium agreement and Limited evidence” or “Low agreement and Medium evidence” and finally red is
“Low agreement and Limited evidence”. Note, that the confidence levels given in Table 8.5 are for 2011 relative to 1750 and for some of the agents the confidence level may be
different for certain portions of the Industrial Era.
value within the estimated range. Some of the RF agents have robust
evidence such as WMGHG with well documented increases based on
high precision measurements as well as contrails as additional clouds
which can be seen by direct observations. However, for some forcing
agents the evidence is more limited regarding their existence such as
aerosol influence on cloud cover. The consistency in the findings for
a particular forcing agent determines the evaluation of the evidence.
A combination of different methods, for example, observations and
modeling, and thus the understanding of the processes causing the
forcing is important for this evaluation. The agreement is a qualitative
judgment of the difference between the various estimates for a par-
ticular RF agent. Figure 1.11 shows how the combined evidence and
agreement results in five levels for the confidence level.
Notes:
The confidence level for aerosol–cloud interactions includes rapid adjustments (which include what was previously denoted as cloud lifetime effect or second indirect aerosol effect). The separate
confidence level for the rapid adjustment for aerosol–cloud interactions is very low. For aerosol–radiation interaction the table provides separate confidence levels for RF due to aerosol–radiation
interaction and rapid adjustment associated with aerosol–radiation interaction.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
695
Evidence is robust for several of the RF agents because of long term
observations of trends over the industrial era and well defined links
between atmospheric or land surfaced changes and their radiative
effects. Evidence is medium for a few agents where the anthropogenic
changes or the link between the forcing agent and its radiative effect
are less certain. Medium evidence can be assigned in cases where
observations or modelling provide a diversity of information and thus
not a consistent picture for a given forcing agent. We assess the evi-
dence to be limited only for rapid adjusment associated with aerosol–
cloud interaction where model studies in some cases indicate changes
but direct observations of cloud alterations are scarce. High agreement
is given only for the WMGHG where the relative uncertainties in the RF
estimates are much smaller than for the other RF agents. Low agree-
ment can either be due to large diversity in estimates of the magnitude
of the forcing or from the fact that the method to estimate the forcing
has a large uncertainty. Stratospheric water vapour is an example of
the latter with modest difference in the few available estimates but
a known large uncertainty in the radiative transfer calculations (see
further description in Section 8.3.1).
Figure 8.14 shows the development of the confidence level over the
last four IPCC assessments for the various RF mechanisms. In the pre-
vious IPCC reports level of scientific understanding (LOSU) has been
used instead of confidence level. For comparison with previous IPCC
assessments the LOSU is converted approximately to confidence level.
Note that LOSU and confidence level use different terms for their rank-
ings. The figure shows generally increasing confidence levels but also
that more RF mechanisms have been included over time. The confi-
dence levels for the RF due to aerosol–radiation interactions, surface
albedo due to land use and volcanic aerosols have been raised and
are now at the same ranking as those for change in stratospheric and
tropospheric ozone. This is due to an increased understanding of key
parameteres and their uncertainties for the elevated RF agents. For
tropospheric and stratospheric ozone changes, research has shown fur-
ther complexities with changes primarily influencing the troposphere
or the stratosphere being linked to some extent (see Section 8.3.3). The
rapid adjustment associated with aerosol–cloud interactions is given
the confidence level very low and had a similar level in AR4. For rapid
adjustment associated with aerosol–radiation interactions (previously
denoted as semi-direct effect) the confidence level is low and is raised
compared to AR4, as the evidence is improved and is now medium (see
Section 7.5.2).
Table 8.6 shows the best estimate of the RF and ERF (for AR5 only)
for the various RF agents from the various IPCC assessments. The RF
due to WMGHG has increased by 16% and 8% since TAR and AR4,
Figure 8.14 | Confidence level of the forcing mechanisms in the 4 last IPCC assessments. In the previous IPCC assessments the level of scientific understanding (LOSU) has been
adopted instead of confidence level, but for comparison with previous IPCC assessments the LOSU is converted approximately to confidence level. The thickness of the bars repre-
sents the relative magnitude of the current forcing (with a minimum value for clarity of presentation). LOSU for the RF mechanisms was not available in the first IPCC Assessment
(Houghton et al., 1990). Rapid adjustments associated with aerosol–cloud interactions (shown as RA aero. –cloud interac.) which include what was previously referred to as the
second indirect aerosol effect or cloud lifetime effect whereas rapid adjustments associated with aerosol–radiation interactions (shown as RA aero.-rad. interac.) were previously
referred to as the semi-direct effect (see Figure 7.3). In AR4 the confidence level for aerosol–cloud interaction was given both for RF due to aerosol–cloud interaction and rapid
adjustment associated with aerosol–cloud interaction. Generally the aerosol–cloud interaction is not separated into various components in AR5, hence the confidence levels for
ERF due to aerosol–cloud interaction in AR5 and for RF due to aerosol–cloud interaction from previous IPCC reports are compared. The confidence level for the rapid adjustment
associated with aerosol–cloud interaction is comparable for AR4 and AR5. The colours are adopted based on the evidence and agreement shown in Figure 1.11. Dark green is “High
agreement and Robust evidence”, light green is either “High agreement and Medium evidence” or “Medium agreement and Robust evidence”, yellow is either “High agreement
and limited evidence” or “Medium agreement and Medium evidence” or “Low agreement and Robust evidence”, orange is either “Medium agreement and Limited evidence” or
“Low agreement and Medium evidence” and finally red is “Low agreement and Limited evidence”.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
696
respectively. This is due mainly to increased concentrations (see Sec-
tion 8.3.2), whereas the other changes for the anthropogenic RF
agents compared to AR4 are due to re-evaluations and in some cases
from improved understanding. An increased number of studies, addi-
tional observational data and better agreement between models and
observations can be the causes for such re-evaluations. The best esti-
mates for RF due to aerosol–radiation interactions, BC on snow and
solar irradiance are all substantially decreased in magnitude compared
to AR4; otherwise the modifications to the best estimates are rather
small. For the RF due to aerosol–radiation interaction and BC on snow
the changes in the estimates are based on additional new studies since
AR4 (see Section 8.3.4 and Section 7.5). For the change in the estimate
of the solar irradiance it is a combination on how the RF is calculated,
new evidence showing some larger earlier estimates were incorrect,
and a downward trend observed during recent years in the solar activ-
ity that has been taken into account (see Section 8.4.1). The estimate
for ERF due to to aerosol–cloud interaction includes rapid adjustment
but still this ERF is smaller in magnitude than the AR4 RF estimate due
to aerosol–cloud interactions without rapid adjustments (a theoreti-
cal construct not quantified in AR5). The uncertainties for ERF due to
CO
2
increase when compared to RF (see Section 8.3.2). We assume the
relative ERF uncertainties for CO
2
apply to all WMGHG. For the short-
lived GHG we do not have sufficient information to include separate
ERF uncertainty to each of these forcing agents (see Section 8.1.1.3).
However, for these forcing mechanisms the RF uncertainties are larger
than for the WMGHG and thus it is unlikely that rapid adjustments
change the uncertainties substantially.
Figure 8.15 shows the RF for agents listed in Table 8.6 over the
1750–2011 period. The methods for calculation of forcing estimates
are described in Section 8.3 and 8.4. For some of the components the
forcing estimates are based on observed abundance whereas some
are estimated from a combination of model simulations and observa-
tions and for others are purely model based. Solid bars are given for
ERF, whereas RF values are given as (additional) hatched bars. Similarly
the uncertainties are given for ERF in solid lines and dotted lines for
RF. An important assumption is that different forcing mechanisms can
be treated additively to calculate the total forcing (see Boucher and
Haywood, 2001; Forster et al., 2007; Haywood and Schulz, 2007). Total
ERF over the Industrial Era calculated from Monte Carlo simulations
are shown in Figure 8.16, with a best estimate of 2.29 W m
–2
. For each
of the forcing agents a probability density function (PDF) is generated
based on uncertainties provided in Table 8.6. The combination of the
individual RF agents to derive total forcing follows the same approach
as in AR4 (Forster et al., 2007) which is based on the method in Bouch-
er and Haywood (2001). The PDF of the GHGs (sum of WMGHG, ozone
and stratospheric water vapour) has a more narrow shape than the
PDF for the aerosols owing to the much lower relative uncertainty.
Global Mean Radiative Forcing (W m
–2
) ERF (W m
–2
)
SAR
(1750–1993)
TAR
(1750–1998)
AR4
(1750–2005)
AR5
(1750–2011)
Comment AR5
Well-mixed
greenhouse gases
(CO
2
, CH
4
, N
2
O, and
halocarbons)
2.45 (2.08 to 2.82) 2.43 (2.19 to 2.67) 2.63 (2.37 to 2.89) 2.83 (2.54 to 3.12) Change due to increase
in concentrations
2.83 (2.26 to 3.40)
Tropospheric ozone +0.40 (0.20 to 0.60) +0.35 (0.20 to 0.50) +0.35 (0.25 to 0.65) +0.40 (0.20 to 0.60) Slightly modified estimate
Stratospheric ozone –0.1 (–0.2 to –0.05) –0.15 (–0.25 to –0.05) –0.05 (–0.15 to +0.05) –0.05 (–0.15 to +0.05) Estimate unchanged
Stratospheric water
vapour from CH
4
Not estimated +0.01 to +0.03 +0.07 (+0.02, +0.12) +0.07 (+0.02 to +0.12) Estimate unchanged
Aerosol–radia-
tion interactions
Not estimated Not estimated –0.50 (–0.90 to –0.10) –0.35 (–0.85 to +0.15) Re-evaluated to be
smaller in magnitude
–0.45 (–0.95 to +0.05)
Aerosol–cloud
interactions
0 to –1.5
(sulphate only)
0 to –2.0
(all aerosols)
–0.70 (–1.80 to –0.30)
(all aerosols)
Not estimated Replaced by ERF and re-evaluated
to be smaller in magnitude
–0.45 (–1.2 to 0.0)
Surface albedo
(land use)
Not estimated –0.20 (–0.40 to 0.0) –0.20 (–0.40 to 0.0) –0.15 (–0.25 to –0.05) Re-evaluated to be slightly
smaller in magnitude
Surface albedo
(black carbon aero-
sol on snow and ice)
Not estimated Not estimated +0.10 (0.0 to +0.20) +0.04 (+0.02 to +0.09) Re-evaluated to be weaker
Contrails
Not estimated +0.02 (+0.006
to +0.07)
+0.01 (+0.003
to +0.03)
+0.01 (+0.005
to +0.03)
No major change
Combined contrails
and contrail-
induced cirrus
Not estimated 0 to +0.04 Not estimated Not estimated 0.05 (0.02 to 0.15)
Total anthropogenic
Not estimated Not estimated 1.6 (0.6 to 2.4) Not estimated Stronger positive due to changes
in various forcing agents
2.3 (1.1 to 3.3)
Solar irradiance
+0.30 (+0.10
to +0.50)
+0.30 (+0.10
to +0.50)
+0.12 (+0.06
to +0.30)
+0.05 (0.0 to +0.10) Re-evaluated to be weaker
Table 8.6 | Summary table of RF estimates for AR5 and comparison with the three previous IPCC assessment reports. ERF values for AR5 are included. For AR5 the values are given
for the period 1750–2011, whereas earlier final years have been adopted in the previous IPCC assessment reports.
Notes:
Volcanic RF is not added to the table due to the periodic nature of volcanic eruptions, which makes it difficult to compare to the other forcing mechanisms.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
697
NaturalAnthropogenic
Well Mixed
Greenhouse Gases
Forcing agent
Radiative forcing of climate between 1750 and 2011
-1 0 1 2 3
Radiative Forcing (W m
-2
)
CO
2
Other WMGHG CH
4
N
2
O
Halocarbons
TroposphericStratospheric Ozone
Stratospheric water
vapour from CH
4
Land Use
Black carbon
on snow
Surface Albedo
Contrail induced cirrus
Contrails
Aerosol-Radiation Interac.
Aerosol-Cloud Interac.
Total anthropogenic
Solar irradiance
Figure 8.15 | Bar chart for RF (hatched) and ERF (solid) for the period 1750–2011, where the total ERF is derived from Figure 8.16. Uncertainties (5 to 95% confidence range)
are given for RF (dotted lines) and ERF (solid lines).
Figure 8.16 | Probability density function (PDF) of ERF due to total GHG, aerosol
forcing and total anthropogenic forcing. The GHG consists of WMGHG, ozone and
stratospheric water vapour. The PDFs are generated based on uncertainties provided in
Table 8.6. The combination of the individual RF agents to derive total forcing over the
Industrial Era are done by Monte Carlo simulations and based on the method in Boucher
and Haywood (2001). PDF of the ERF from surface albedo changes and combined con-
trails and contrail-induced cirrus are included in the total anthropogenic forcing, but
not shown as a separate PDF. We currently do not have ERF estimates for some forcing
mechanisms: ozone, land use, solar, etc. For these forcings we assume that the RF is
representative of the ERF and for the ERF uncertainty an additional uncertainty of 17%
has been included in quadrature to the RF uncertainty. See Supplementary Material Sec-
tion 8.SM.7 and Table 8.SM.4 for further description on method and values used in the
calculations. Lines at the top of the figure compare the best estimates and uncertainty
ranges (5 to 95% confidence range) with RF estimates from AR4.
Therefore, the large uncertainty in the aerosol forcing is the main
cause of the large uncertainty in the total anthropogenic ERF. The total
anthropogenic forcing is virtually certain to be positive with the prob-
ability for a negative value less than 0.1%. Compared to AR4 the total
anthropogenic ERF is more strongly positive with an increase of 43%.
This is caused by a combination of growth in GHG concentration, and
thus strengthening in forcing of WMGHG, and weaker ERF estimates of
aerosols (aerosol–radiation and aerosol–cloud interactions) as a result
of new assessments of these effects.
Figure 8.17 shows the forcing over the Industrial Era by emitted com-
pounds (see Supplementary Material Tables 8.SM.6 and 8.SM.7 for
actual numbers and references). It is more complex to view the RF
by emitted species than by change in atmospheric abundance (Figure
8.15) since the number of emitted compounds and changes leading to
RF is larger than the number of compounds causing RF directly (see
Section 8.3.3). The main reason for this is the indirect effect of sever-
al compounds and in particular components involved in atmospheric
chemistry (see Section 8.2). To estimate the RF by the emitted com-
pounds in some cases the emission over the entire Industrial Era is
needed (e.g., for CO
2
) whereas for other compounds (such as ozone
and CH
4
) quite complex simulations are required (see Section 8.3.3).
CO
2
is the dominant positive forcing both by abundance and by emit-
ted compound. Emissions of CH
4
, CO, and NMVOC all lead to excess
CO
2
as one end product if the carbon is of fossil origin and is the reason
why the RF of direct CO
2
emissions is slightly lower than the RF of
abundance change of CO
2
. For CH
4
the contribution from emission is
estimated to be almost twice as large as that from the CH
4
concen-
Greenhouse
gases
Total
anthropogenic
Aerosols
AR4 RF
-2 0 2 4
Effective radiative forcing (Wm
-2
)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Probability density function
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
698
tration change, 0.97 (0.80 to 1.14) W m
–2
versus 0.48 (0.43 to 0.53)
W m
–2
, respectively. This is because emission of CH
4
leads to ozone
production, stratospheric water vapour, CO
2
(as mentioned above), and
importantly affects its own lifetime (Section 8.2). Actually, emissions of
CH
4
would lead to a stronger RF via the direct CH
4
greenhouse effect
(0.64 W m
–2
) than the RF from abundance change of CH
4
(0.48 W m
–2
).
This is because other compounds have influenced the lifetime of CH
4
and reduced the abundance of CH
4
, most notably NO
x
. Emissions of CO
(0.23 (0.18 to 0.29) W m
–2
) and NMVOC (0.10 (0.06 to 0.14) W m
–2
)
have only indirect effects on RF through ozone production, CH
4
and
CO
2
and thus contribute an overall positive RF. Emissions of NO
X
, on
the other hand, have indirect effects that lead to positive RF through
ozone production and also effects that lead to negative RF through
reduction of CH
4
lifetime and thus its concentration, and through con-
tributions to nitrate aerosol formation. The best estimate of the overall
effect of anthropogenic emissions of NO
X
is a negative RF (-0.15 (-0.34
to +0.02) W m
–2
). Emissions of ammonia also contribute to nitrate aer-
osol formation, with a small offset due to compensating changes in
sulphate aerosols. Additionally indirect effects from sulphate on atmos-
pheric compounds are not included here as models typically simulate
a small effect, but there are large relative differences in the response
between models. Impacts of emissions other than CO
2
on the carbon
cycle via changes in atmospheric composition (ozone or aerosols) are
also not shown owing to the limited amount of available information.
For the WMGHG, the ERF best estimate is the same as the RF. The
uncertainty range is slightly larger, however. The total emission-based
ERF of WMGHG is 3.00 (2.22 to 3.78) W m
–2
. That of CO
2
is 1.68 (1.33
to 2.03) W m
–2
; that of CH
4
is 0.97 (0.74 to 1.20) W m
–2
; that of strat-
ospheric ozone-depleting halocarbons is 0.18 (0.01 to 0.35) W m
–2
.
Emissions of BC have a positive RF through aerosol–radiation interac-
tions and BC on snow (0.64 W m
–2
, see Section 8.3.4 and Section 7.5).
The emissions from the various compounds are co-emitted; this is in
particular the case for BC and OC from biomass burning aerosols. The
net RF of biomass burning emissions for aerosol–radiation interactions
is close to zero, but with rather strong positive RF from BC and negative
RF from OC (see Sections 8.3.4 and 7.5). The ERF due to aerosol–cloud
interactions is caused by primary anthropogenic emissions of BC, OC
and dust as well as secondary aerosol from anthropogenic emissions
of SO
2
, NO
X
and NH
3
. However, quantification of the contribution from
the various components to the ERF due to aerosol–cloud interactions
has not been attempted in this assessment.
8.5.2 Time Evolution of Historical Forcing
The time evolution of global mean forcing is shown in Figure 8.18 for
the Industrial Era. Over all time periods during the Industrial Era CO
2
and other WMGHG have been the dominant term, except for short-
er periods with strong volcanic eruptions. The time evolution shows
an almost continuous increase in the magnitude of anthropogenic
ERF. This is the case both for CO
2
and other WMGHGs as well as sev-
eral individual aerosol components. The forcing from CO
2
and other
WMGHGs has increased somewhat faster since the 1960s. Emissions
of CO
2
have made the largest contribution to the increased anthropo-
genic forcing in every decade since the 1960s. The total aerosol ERF
(aerosol–radiation interaction and aerosol–cloud interaction) has the
strongest negative forcing (except for brief periods with large volcanic
forcing), with a strengthening in the magnitude similar to many of the
other anthropogenic forcing mechanisms with time. The global mean
forcing of aerosol–radiation interactions was rather weak until 1950
but strengthened in the latter half of the last century and in particular
in the period between 1950 and 1980. The RF due to aerosol–radiation
interaction by aerosol component is shown in Section 8.3.4 (Figure 8.8).
Although there is high confidence for a substantial enhancement in the
negative aerosol forcing in the period 1950–1980, there is much more
uncertainty in the relative change in global mean aerosol forcing over
the last two decades (1990–2010). Over the last two decades there
has been a strong geographic shift in aerosol and aerosol precursor
Figure 8.17 | RF bar chart for the period 1750–2011 based on emitted compounds
(gases, aerosols or aerosol precursors) or other changes. Numerical values and their
uncertainties are shown in Supplementary Material Tables 8.SM.6 and 8.SM.7. Note
that a certain part of CH
4
attribution is not straightforward and discussed further in
Section 8.3.3. Red (positive RF) and blue (negative forcing) are used for emitted com-
ponents which affect few forcing agents, whereas for emitted components affecting
many compounds several colours are used as indicated in the inset at the upper part
the figure. The vertical bars indicate the relative uncertainty of the RF induced by each
component. Their length is proportional to the thickness of the bar, that is, the full length
is equal to the bar thickness for a ±50% uncertainty. The net impact of the individual
contributions is shown by a diamond symbol and its uncertainty (5 to 95% confidence
range) is given by the horizontal error bar. ERFaci is ERF due to aerosol–cloud interac-
tion. BC and OC are co-emitted, especially for biomass burning emissions (given as
Biomass Burning in the figure) and to a large extent also for fossil and biofuel emissions
(given as Fossil and Biofuel in the figure where biofuel refers to solid biomass fuels).
SOA have not been included because the formation depends on a variety of factors not
currently sufficiently quantified.
( )
8
Anthropogenic and Natural Radiative Forcing Chapter 8
699
emissions (see Section 2.2.3), and there are some uncertainties in these
emissions (Granier et al., 2011). In addition to the regional changes in
the aerosol forcing there is also likely a competition between various
aerosol effects. Emission data indicate a small increase in the BC emis-
sions (Granier et al., 2011) but model studies also indicate a weak
enhancement of other aerosol types. Therefore, the net aerosol forc-
ing depends on the balance between absorbing and scattering aero-
sols for aerosol–radiation interaction as well as balance between the
changes in aerosol–radiation and aerosol–cloud interactions. In the
ACCMIP models, for example, the RF due to aerosol–radiation inter-
action becomes less negative during 1980 to 2000, but total aerosol
ERF becomes more negative (Shindell et al., 2013c). There is a very low
confidence for the trend in the total aerosol forcing during the past two
to three decades, even the sign; however, there is high confidence that
the offset from aerosol forcing to WMGHG forcing during this period
was much smaller than over the 1950–1980 period.
The volcanic RF has a very irregular temporal pattern and for certain
years has a strongly negative RF. There has not been a major volcanic
eruption in the past decade, but some weaker eruptions give a current
RF that is slightly negative relative to 1750 and slightly stronger in
magnitude compared to 1999–2002 (see Section 8.4.2).
Figure 8.19 shows linear trends in forcing (anthropogenic, natural and
total) over four different time periods. Three of the periods are the
same as chosen in Box 9.2 (1984–1998, 1998–2011 and 1951–2011)
and the period 1970–2011 is shown in Box 13.1. Monte Carlo sim-
ulations are performed to derive uncertainties in the forcing based
on ranges given in Table 8.6 and the derived linear trends. Further,
these uncertainties are combined with uncertainties derived from
shifting time periods ±2 years and the full 90% confidence range is
shown in Figure 8.19 (in Box 9.2 only the total forcing is shown with
uncertainties derived from the forcing uncertainty without sensitivity
to time period). For the anthropogenic forcing sensitivity to the selec-
tion of time periods is very small with a maximum contribution to the
uncertainties shown in Figure 8.19 of 2%. However, for the natural
forcing the sensitivity to time periods is the dominant contributor to
the overall uncertainty (see Supplementary Material Figure 8.SM.3) for
the relatively short periods 1998–2011 and 1984–1998, whereas this
is not the case for the longer periods. For the 1998–2011 period the
natural forcing is very likely negative and has offset 2 to 89% of the
anthropogenic forcing. It is likely that the natural forcing change has
offset at least 30% of the anthropogenic forcing increase and very
likely that it has offset at least 10% of the anthropogenic increase. For
the 1998–2011 period both the volcanic and solar forcings contribute
to this negative natural forcing, with the latter dominating. For the
other periods shown in Figure 8.19 the best estimate of the natural is
much smaller in magnitude than the anthropogenic forcing, but note
that the natural forcing is very dependent on the selection of time
period near the 1984–1998 interval. Over the period 1951–2011 the
trend in anthropogenic forcing is almost 0.3 W m
–2
per decade and
thus anthropogenic forcing over this period is more than 1.5 W m
–2
. The
anthropogenic forcing for 1998–2011 is 30% higher and with smaller
uncertainty than for the 1951–2011 period. Note that due to large
WMGHG forcing (Section 8.3.2) the anthropogenic forcing was similar
in the late 1970s and early 1980s to the 1998–2011 period. The reason
for the reduced uncertainty in the 1998–2011 anthropogenic forcing
is the larger domination of WMGHG forcing and smaller contribution
from aerosol forcing compared to previous periods. Similar to the
results for 1970–2011 in Figure 8.19, Box 13.1 shows that the global
energy budget is dominated by anthropogenic forcing compared to
the natural forcing, except for the two major volcanic eruption in this
period as can be easily seen in Figure 8.18.
Figure 8.20 shows the forcing between 1980 and 2011. Compared
to the whole Industrial Era the dominance of the CO
2
is larger for
this recent period both with respect to other WMGHG and the total
anthropogenic RF. The forcing due to aerosols is rather weak leading
( )
( )
Figure 8.18 | Time evolution of forcing for anthropogenic and natural forcing mecha-
nisms. Bars with the forcing and uncertainty ranges (5 to 95% confidence range) at
present are given in the right part of the figure. For aerosol the ERF due to aerosol–
radiation interaction and total aerosol ERF are shown. The uncertainty ranges are for
present (2011 versus 1750) and are given in Table 8.6. For aerosols, only the uncertainty
in the total aerosol ERF is given. For several of the forcing agents the relative uncertainty
may be larger for certain time periods compared to present. See Supplementary Material
Table 8.SM.8 for further information on the forcing time evolutions. Forcing numbers
provided in Annex II. The total antropogenic forcing was 0.57 (0.29 to 0.85) W m
–2
in 1950, 1.25 (0.64 to 1.86) W m
–2
in 1980 and 2.29 (1.13 to 3.33) W m
–2
in 2011.
Figure 8.19 | Linear trend in anthropogenic, natural and total forcing for the indicated
time periods. The uncertainty ranges (5 to 95% confidence range) are combined from
uncertainties in the forcing values (from Table 8.6) and the uncertainties in selection
of time period. Monte Carlo simulations were performed to derive uncertainties in the
forcing based on ranges given in Table 8.6 and linear trends in forcing. The sensitivity to
time periods has been derived from changing the time periods by ±2 years.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
700
Figure 8.20 | Bar chart for RF (hatched) and ERF (solid) for the period 1980–2011,
where the total anthropogenic ERF are derived from Monte-Carlo simulations similar to
Figure 8.16. Uncertainties (5 to 95% confidence range) are given for RF (dotted lines)
and ERF (solid lines).
to a very strong net positive ERF for the 1980–2011 period. More than
40% of the total anthropogenic ERF has occurred over the 1980–2011
period with a value close to 1.0 (0.7 to 1.3) W m
–2
. The major contri-
bution to the uncertainties in the time evolution of the anthropogenic
forcing is associated with the aerosols (see Section 8.5.1). Despite this,
anthropogenic ERF is very likely considerably more positive than the
natural RF over the decadal time periods since 1950. This is in par-
ticular the case after 1980, where satellite data are available that pro-
vide important measurements to constrain the natural RF mechanisms
(e.g., the volcanic RF change between 2007–2011 and 1978–1982 is
0.06 W m
–2
and the representative change in solar irradiance over the
1980–2011 period is –0.06 W m
–2
) with total natural RF of 0.0 (-0.1 to
+0.1) W m
–2
.
8.5.3 Future Radiative Forcing
Projections of global mean RF are assessed based on results from mul-
tiple sources examining the RF due to RCP emissions: the ACCMIP ini-
tiative (see Section 8.2) provides analysis of the RF or ERF due to aer-
osols and ozone (Shindell et al., 2013c), while WMGHG, land use and
stratospheric water RFs are taken from the results of calculations with
the reduced-complexity Model for the Assessment of Greenhouse-gas
Induced Climate Change 6 (MAGICC6) driven by the RCP emissions
and land use (Meinshausen et al., 2011a). While MAGICC6 also esti-
mated ozone and aerosol RF, those values differ substantially from the
ACCMIP values and are considered less realistic. Additional discussion
of biases in the MAGICC6 results due to the simplified representations
Natural Anthropogenic
Well Mixed
Greenhouse Gases
Forcing agent
-0.5 0.0 0.5 1.0
Radiative Forcing (W m
-2
)
Radiative forcing of climate between 1980 and 2011
CO
2
Other WMGHG
CH
4
N
2
O
Halocarbons
TroposphericStratospheric
Ozone
Stratospheric water
vapour from CH
4
BC on snow
+ Land Use
Surface Albedo
Contrail induced cirrus
Contrails
Aero.-Rad. Interac.
Aero.-Cloud Interac.
Total anthropogenic
Solar irradiance
of atmospheric chemistry and the carbon cycle, along with further dis-
cussion on the representativeness of the RCP projections in context
with the broader set of scenarios in the literature, is presented in Sec-
tion 11.3.5 and Section 12.3 (also see Section 8.2). As the ACCMIP
project provided projected forcings primarily at 2030 and 2100, we
hereafter highlight those times. Although understanding the relative
contributions of various processes to the overall effect of aerosols on
forcing is useful, we emphasize the total aerosol ERF, which includes
all aerosol–radiation and aerosol–cloud interactions, as this is the
most indicative of the aerosol forcing driving climate change. We also
present traditional RF due to aerosol–radiation interaction (previously
called direct aerosol effect) but do not examine further the various
components of aerosol ERF. Aerosol forcing estimates, both mean and
uncertainty ranges, are derived from the 10 ACCMIP models, 8 of which
are also CMIP5 models. We analyze forcing during the 21st century
(relative to 2000), and hence the WMGHG forcing changes are in addi-
tion to persistent forcing from historical WMGHG increases.
Analysis of forcing at 2030 relative to 2000 shows that under RCP2.6,
total ozone (tropospheric and stratospheric) forcing is near zero, RF
due to aerosol–radiation interaction is positive but small, and hence
WMGHG forcing dominates changes over this time period (Figure 8.21).
WMGHG forcing is dominated by increasing CO
2
, as declining CH
4
and
increasing N
2
O have nearly offsetting small contributions to forcing.
Aerosol ERF was not evaluated for this RCP under ACCMIP, and values
cannot be readily inferred from RF due to aerosol–radiation interaction
as these are not directly proportional. Under RCP8.5, RF due to aerosol–
radiation interaction in 2030 is weakly negative, aerosol ERF is positive
with a fairly small value and large uncertainty range, total ozone forc-
ing is positive but small (~0.1 W m
–2
), and thus WMGHG forcing again
dominates with a value exceeding 1 W m
–2
. As with RCP2.6, WMGHG
forcing is dominated by CO
2
, but under this scenario the other WMGHGs
all contribute additional positive forcing. Going to 2100, ozone forcing
diverges in sign between the two scenarios, consistent with changes in
the tropospheric ozone burden (Figure 8.4) which are largely attribut-
able to projected CH
4
emissions, but is small in either case. Ozone RF
is the net impact of a positive forcing from stratospheric ozone recov-
ery owing to reductions in anthropogenic ozone-depleting halocarbon
emissions in both scenarios and a larger impact from changes in tropo-
spheric precursors (Shindell et al., 2013c) which have a negative forcing
in RCP2.6 and a positive forcing in RCP8.5.
The two scenarios are fairly consistent in their trends in RF due to aero-
sol–radiation interaction by component (Figure 8.21). There is positive
RF due to aerosol–radiation interaction due to reductions in sulfate
aerosol. This is largely offset by negative RF due to aerosol–radiation
interaction by primary carbonaceous aerosols and especially by nitrate
(though nearly all CMIP5 models did not include nitrate), leaving net
aerosol RF due to aerosol–radiation interaction values that are very
small, 0.1 W m
–2
or less in magnitude, in either scenario at 2030 and
2100. Nitrate aerosols continue to increase through 2100 as ammo-
nia emissions rise steadily due to increased use of agricultural ferti-
lizer even as all other aerosol precursor emissions decrease (Figure
8.2), including sulphur dioxide which drives the reduction in sulphate
aerosol that also contributes to additional formation of nitrate aerosols
in the future (Bauer et al., 2007; Bellouin et al., 2011). Aerosol ERF is
likely similar at this time in all scenarios given that they all have greatly
8
Anthropogenic and Natural Radiative Forcing Chapter 8
701
reduced emissions of all aerosols and aerosol precursors other than
ammonia. Aerosol ERF shows a large positive value at 2100 relative
to 2000, nearly returning to its 1850 levels (the 2100 versus 1850 ERF
represents a decrease in ERF of 91% relative to the 2000 versus 1850
value), as is expected given the RCP emissions. Thus although some
models project large increases in nitrate RF in the future, the reduc-
tion in overall aerosol loading appears to lead to such a strong reduc-
tion in aerosol ERF that the impact of aerosols becomes very small
under these RCPs. Of course the projections of drastic reductions in
primary aerosol as well as aerosol and ozone precursor emissions may
be overly optimistic as they assume virtually all nations in the world
become wealthy and that emissions reductions are directly dependent
on wealth. The RCPs also contain substantially lower projected growth
in HFC emissions than in some studies (e.g., Velders et al., 2009).
Although aerosol ERF becomes less negative by nearly 1 W m
–2
from
2000 to 2100, this change is still small compared with the increased
WMGHG forcing under RCP8.5, which is roughly 6 W m
–2
during this
time (Figure 8.21). Roughly 5 W m
–2
of this WMGHG forcing comes
from CO
2
, with substantial additional forcing from increases in both
CH
4
and nitrous oxide and only a very small negative forcing from
reductions in halocarbons. Under RCP2.6, the WMGHG forcing is
only about 0.5 W m
–2
during this time, as relatively strong decreases
in CH
4
and halocarbon forcing offset roughly 40% of the increased
CO
2
forcing, which is itself far less than under RCP8.5. Hence under
this scenario, the projected future forcing due to aerosol reductions is
actually stronger than the WMGHG forcing. Viewing the timeseries of
the various forcings, however, indicates that aerosol ERF is returning
to its pre-industrial levels, so that net forcing becomes increasingly
dominated by WMGHGs regardless of scenario during the 21st cen-
tury (Figure 8.22). As the forcing is so heavily dominated by WMGHGs
at 2100, and the WMGHG concentrations (CO
2
) or emissions (others)
were chosen to match forcing targets, all the scenarios show net forc-
ing values at that time that are fairly close to the scenarios’ target
values. The reduced aerosol forcing, with its large uncertainty, leads
to a pronounced decrease in the uncertainty of the total net forcing
by 2100. Based on the spread across ACCMIP models (using ERF for
aerosols and converting to ERF for GHGs), the 90% confidence interval
(CI) is about 20% for the 2100 net forcing, versus 26% for 2030 under
RCP8.5 and 45–61% for 1980 and 2000 (Shindell et al., 2013c). The
total ERF due to all causes has been independently estimated based
on the transient response in the CMIP5 models and a linear forc-
ing-response relationship derived through regression of the modelled
response to an instantaneous increase in CO
2
(Forster et al., 2013).
Uncertainties based on model spread behave similarly, with the 90%
CI for net total ERF decreasing from 53% for 2003 to only 24 to 34%
for 2100. Forcing relative to 2000 due to land use (via albedo only) and
stratospheric water vapor changes are not shown separately as their
projected values under the four RCPs are quite small: –0.09 to 0.00 and
–0.03 to 0.10 W m
–2
, respectively.
The CMIP5 forcing estimates (Forster et al., 2013) for the total project-
ed 2030 and 2100 ERF are slightly smaller than the results obtained
from the ACCMIP models (or the RCP targets; see Section 12.3.3).
Examining the subset of models included in both this regression anal-
ysis and in ACCMIP shows that the ACCMIP subset show forcings on
the low side of the mean value obtained from the full set of CMIP5
analyzed, indicating that the discrepancy between the methods is not
related to analysis of a different set of models. Instead, it may reflect
nonlinearities in the response to forcing that are not represented by
the regression analysis of the response to abrupt CO
2
increase experi-
ments (Long and Collins, 2013) or differences in the response to other
forcing agents relative to the response to CO
2
used in deriving the
CMIP5 estimates (see also 12.3.3).
Natural forcings will also change in the future. The magnitudes cannot
be reliably projected, but are likely to be small at multi-decadal scales
(see Section 8.4). Brief episodic volcanic forcing could be large, however.
0
2
4
6
8
Effective radiative forcing (W m
-2
)
CO
2
Other WMGHG
O
3
+Strat H
2
O
Total RFari
Aerosol ERF
Total
-0.4
-0.2
0.0
0.2
0.4
0.6
Sulphate Carbonaceous Nitrate SOA
RCP2.6
RCP8.5
RCP2.6
RCP8.5
}
}
2030 vs 2000
2100 vs 2000
1850 1900 1950 2000 2050 2100
0
2
4
6
8
Effective radiative forcing (W m
-2
)
Ozone
Aerosol
WMGHG
Net
RCP8.5
RCP2.6
RCP4.5
RCP6.0
Figure 8.21 | Radiative forcing relative to 2000 due to anthropogenic composition
changes based on ACCMIP models for aerosols (with aerosol ERF scaled to match
the best estimate of present-day forcing) and total ozone and RCP WMGHG forcings.
Ranges are one standard deviation in the ACCMIP models and assessed relative uncer-
tainty for WMGHGs and stratospheric water vapor. Carbonaceous aerosols refer to pri-
mary carbonaceous, while SOA are secondary organic aerosols. Note that 2030 ERF for
RCP2.6 was not available, and hence the total shown for that scenario is not perfectly
comparable to the other total values. RFari is RF due to aerosol–radiation interaction.
Figure 8.22 | Global mean anthropogenic forcing with symbols indicating the times at
which ACCMIP simulations were performed (solid lines with circles are net; long dashes
with squares are ozone; short dashes with diamonds are aerosol; dash-dot are WMGHG;
colours indicate the RCPs with red for RCP8.5, orange RCP6.0, light blue RCP4.5, and
dark blue RCP2.6). RCPs 2.6, 4.5 and 6.0 net forcings at 2100 are approximate values
using aerosol ERF projected for RCP8.5 (modified from Shindell et al., 2013c). Some
individual components are omitted for some RCPs for visual clarity.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
702
8.6 Geographic Distribution of Radiative
Forcing
The forcing spatial pattern of the various RF mechanisms varies sub-
stantially in space and in time, especially for the NTCFs. The spatial pat-
tern is of interest to the extent that it may influence climate response
(Section 8.6.2.2) as is being particularly investigated in the ACCMIP
simulations.
8.6.1 Spatial Distribution of Current Radiative Forcing
The WMGHGs such as CO
2
have the largest forcing in the subtropics,
decreasing toward the poles, with the largest forcing in warm and dry
regions and smaller values in moist regions and in high-altitude regions
(Taylor et al., 2011). For the NTCFs (Box 8.2) their concentration spatial
pattern and therefore their RF pattern are highly inhomogeneous, and
again meteorological factors such as temperature, humidity, clouds,
and surface albedo influence how concentration translates to RF.
Figure 8.23 shows the RF spatial distribution of the major NTCFs togeth-
er with standard deviation among the ACCMIP models (Shindell et al.,
2013c) the net anthropogenic composition (WMGHG+ozone+aerosol)
forcing is also shown (lower left panel). These models used unified
anthropogenic emissions of aerosol and ozone precursors (Supplemen-
tary Material Figure 8.SM.2), so that the model diversity in RF is due
only to differences in model chemical and climate features and natu-
ral emissions, and would be larger if uncertainty in the anthropogenic
emissions were also included. In general, the confidence in geograph-
ical distribution is lower than for global mean, due to uncertainties in
chemistry, transport and removal of species.
The negative RF due to aerosol–radiation interaction (first row; defined
in Figure 7.3) is greatest in the NH and near populated and biomass
burning regions. The standard deviation for the net RF due to aerosol–
radiation interaction is typically largest over regions where vegetation
changes are largest (e.g., South Asia and central Africa), due to uncer-
tainties in biomass burning aerosol optical properties and in treatment
of secondary organic aerosols. Carbonaceous aerosol forcing (second
row) is greatest in South and East Asia and can be negative in biomass
burning regions due to large weakly absorbing organic components.
Absorbing aerosols also have enhanced positive forcing when they
overlie high albedo surfaces such as cryosphere, desert or clouds, with
as much as 50% of BC RF resulting from BC above clouds (Zarzycki
and Bond, 2010).
Figure 8.24 compares the aerosol RFs for ACCMIP (Shindell et al.,
2013c), which are representative of the CMIP5 experiments, with those
from the AeroCom model intercomparison (Myhre et al., 2013) which
includes sixteen models that used unified meteorology and are more
extensively compared to measurements (e.g., Koch et al., 2009b; Koffi
et al., 2012). The forcing results are very similar, establishing the repre-
sentativeness and validity of the ACCMIP aerosol simulations.
The net aerosol ERF (Figure 8.23; third row), includes both aerosol–
radiation and aerosol–cloud interactions. The spatial pattern corre-
lates with the RF (first row), except with stronger effect in the out-
flow regions over oceans. The flux change is larger in the NH than the
SH (e.g., by nearly a factor of 3; Ming et al., 2007). Rapid adjustment
associated with aerosol–radiation and aerosol–cloud interactions may
enhance or reduce cloud cover depending on the region, cloud dynam-
ics and aerosol loading (e.g., Randles and Ramaswamy, 2008; Koch
and Del Genio, 2010; Persad et al., 2012). In general, the ocean-land
forcing pattern differs from that reported in AR4, where the forcing due
to aerosol–cloud interaction were larger over land than ocean (Forster
et al., 2007), and this continues to be a source of uncertainty. Since
AR4, Quaas et al. (2009) showed using satellite retrievals that the cor-
relation between AOD changes and droplet number changes is strong-
er over oceans than over land and that models tend to overestimate
the strength of the relation over land. Penner et al. (2011) showed
that satellite retrievals, due to their dependence on present-day condi-
tions, may underestimate the forcing due to aerosol–cloud interaction,
especially over land, although this model analysis may overestimate
the cloud condensation nucleus to AOD relation (Quaas et al., 2011).
Wang and Penner (2009) also showed that if models include boundary
layer nucleation and increase the fraction of sulphur emitted as a pri-
mary particle, the effect over land is increased relative to over ocean
(see also Section 7.5.3). The aerosol ERF standard deviation is large
in biomass burning regions, as for the RF, and in regions where cloud
effects differ among models (e.g., northern North America, northeast
Asia, Amazonia). The spread in aerosol ERF is much larger than for the
RF alone, although the relative standard deviation is no larger (Shindell
et al., 2013c).
For components that primarily scatter radiation, the radiative effect at
the surface is similar to the RF (according to the definition in Section
8.1.1). However for components that absorb radiation in the atmos-
phere the radiation reaching the surface is reduced (Forster et al., 2007;
Ramanathan and Carmichael, 2008; Andrews et al., 2010). This absorp-
tion of incoming solar radiation alters the vertical temperature profile
in the atmospheric column and can thus change atmospheric circula-
tion and cloud formation. The aerosol atmospheric absorption (Figure
8.23, bottom right), or the difference between ERF and the analogous
radiative flux reaching the surface including rapid adjustments, has a
spatial pattern that to lowest order tracks the carbonaceous aerosol
forcing, but is also affected by cloud changes, where e.g., cloud loss
could enhance atmospheric absorption. Atmospheric aerosol absorp-
tion patterns thus mirror the ERF due to aerosol–cloud interaction pat-
tern, with larger forcing over continents.
Ozone RF is calculated using the methodology described in Shindell et
al. (2013c), but applied to the larger set of models in ACCMIP (Steven-
son et al., 2013). The net ozone RF (Figure 8.23; fourth row) is largest
in subtropical latitudes, and is more positive in the NH than the SH.
Pollution in the NH accounts for positive tropospheric forcing; strato-
spheric ozone loss has caused negative SH polar forcing. Model stand-
ard deviation is largest in the polar regions where lower stratosphere/
upper troposphere changes differ in the models (Young et al., 2013).
Overall, the confidence in aerosol and ozone RF spatial patterns is
medium and lower than that for the global mean due to the large
regional standard deviations (Figure 8.23), and is exacerbated in aero-
sol ERF patterns due to uncertainty in cloud responses.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
703
All aerosol RF
0.29 W m
-2
Ozone RF Carbonaceous aerosol RF
0.21 W m
-2
0.08 (0.18) W m
-2
0.16 (0.19) W m
-2
-.38-.62-.88 -.12 .62.38.12 .88
-.38-.62-.88 -.12 .62.38.12 .88
-.38-.62-.88 -.12 .62.38.12 .88
.70.350 1.05 2.101.751.40 2.45
.70.350 1.05 2.101.751.40 2.45
.12.060 .18 .36.30.24 .42
-0.26 W m
-2
0.14 (0.30) W m
-2
All aerosol ERF
-1.5-2.5-3.5 -.5 1.25.75.25 1.75
.70.350 1.05 2.101.751.40 2.45
0.29 (1.27) W m
-2
-1.17 W m
-2
All aerosol eective
atmospheric forcing
1.42 W m
-2
-1.5-2.5-3.5 -.5 2.51.5.5 3.5
1.46 W m
-2
-1.0-2.0-3.0 -0.5 2.01.00.5 3.0
Total anthropogenic
composition forcing
Preindustrial to Present-Day Forcing
Multi-model mean Standard deviation
Multi-model mean Multi-model mean
Figure 8.23 | Spatial pattern of ACCMIP models 1850 to 2000 forcings, mean values (left) and standard deviation (right) for aerosols and ozone (top four rows). Values above are
the average of the area-weighted global means, with the area weighted mean of the standard deviation of models at each point provided in parenthesis. Shown are net aerosol RF
due to aerosol–radiation interaction (top, 10 models), carbonaceous aerosol RF due to aerosol-radiation interaction (2nd row, 7 models), aerosol ERF (3rd row, 8 models), ozone
(4th row, 11 models), total anthropogenic composition forcing (WMGHG+ozone+aerosols; bottom left), aerosol atmospheric absorption including rapid adjustment (bottom right,
6 models). Note that RF and ERF means are shown with different colour scales, and standard deviation colour scales vary among rows.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
704
All aerosol RF
Preindustrial to present-day forcing
0.25 W m
-2
-.38-.62-.88 -.12 .62.38.12 .88
.70.350 1.05 2.101.751.40 2.45
ACCMIPACCMIP AeroComAeroCom
-0.27 W m
-2
0.15 (0.27) W m
-2
0.18 W m
-2
0.07 (0.12) W m
-2
Fossil & biofuel Black Carbon RF
0.10 (0.12) W m
-2
-0.26 W m
-2
0.14 (0.30) W m
-2
Multi-model mean Standard deviation
Figure 8.24 | Spatial pattern of ACCMIP and 16 AeroCom models 1850 to 2000 RF due to aerosol–radiation interaction, mean values (left) and standard deviation (right). Note
that different carbonaceous aerosol diagnostics are used here compared to Figure 8.23, due to available AeroCom fields. Values above are the average of the area-weighted global
means, with the area weighted mean of the standard deviation of models at each point provided in parentheses.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
705
8.6.2 Spatial Evolution of Radiative Forcing and
Response over the Industrial Era
8.6.2.1 Regional Forcing Changes During the Industrial Era
The spatial distribution of the WMGHG RF has shifted only slightly over
the industrial period; however the RF spatial distributions for NTCFs
has shifted with emissions, due to the timing of regional development
and implementation of pollution standards (Supplementary Material
Figures 8.SM.1 and 8.SM.2 show regional trends and emissions maps;
Lamarque et al., 2013). Figure 8.25 shows how the distributions of
aerosol and ozone forcings are modelled to have changed up to 1930,
1980 and 2000. Substantial industrial coal-burning in the early part of
the 20th century occurred in the northeastern United States and West-
ern Europe, leading to stronger sulphate and BC forcing near those
2000 vs 1850
All aerosol RF
1980 vs 18501930 vs 1850
Carbonaceous aerosol RF
-.38-.62-.88 -.12 .62.38.12 .88
0.21 W m
-2
0.17 W m
-2
0.07 W m
-2
-1.5-2.5-3.5 -.5 1.25.75.25 1.75
-1.03 W m
-2
-0.27 W m
-2
0.29 W m
-2
Ozone RF
0.27 W m
-2
0.09 W m
-2
-0.29 W m
-2
-0.06 W m
-2
-0.26 W m
-2
All aerosol ERF
-1.17 W m
-2
Figure 8.25 | Multi-model mean RF due to aerosol–radiation interaction of all aerosols, carbonaceous aerosols, ozone, and aerosol ERF (W m
–2
) for the indicated times based on
the ACCMIP simulations. Global area-weighted means are given in the upper right.
regions (Figure 8.25, left). Between 1950 and 1970, coal burning for
power generation increased while coal burning for other purposes was
replaced by oil and natural gas and motor vehicle usage grew rapidly in
these regions, leading to more sulphate and less BC. Peak aerosol forc-
ing in North America and Europe occurred around 1970–1980 (Figure
8.25, second column), while Asian development led to increased bio-
fuel and fossil fuel sources of aerosols and ozone precursors toward
the end of the century. During the final decades of the century, des-
ulphurization controls reduced sulphur emissions from North America
and Europe, resulting in reduced negative forcing in these regions and
positive Arctic aerosol forcing. The SH ozone hole developed during the
final three decades, with negative forcing over high latitudes. Biomass
burning generated ozone and carbonaceous aerosols in NH high-lati-
tudes early in the century, with increased tropical burning from mid to
late century.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
706
Aerosol ERF grew rapidly from 1930 to 1980, as did RF due to aero-
sol–radiation interaction, with a spatial structure reflecting both the
influence of aerosol–radiation and aerosol–cloud interactions that are
especially strong over pollution outflow regions and over areas with
high surface albedo. From 1980 to 2000, aerosol ERF continued to
become more negative even as negative RF due to aerosol–radiation
interaction grew weaker, with the spatial pattern showing strengthen-
ing of aerosol ERF over Asia and weakening of aerosol ERF over North
America and Europe.
Soil dust has changed since the pre-industrial due to land disturbance
and resulting desertification (a forcing) and to changes in climate (a
feedback). Mahowald et al. (2010) showed approximate doubling in
dust loading over the 20th century (–0.1 W m
–2
; consistent with the
best estimate in Section 7.5.2; Section 8.3.4.2), primarily from the
Saharan and Middle Eastern Deserts, with largest increase from the
1950s to the 1980s (–0.3 W m
–2
), followed by a leveling. The increased
dustiness reduces model precipitation within the Saharan source
region, improving agreement with observed precipitation.
Aerosol loading changes during the past century have impacted radi-
ation at the surface (Section 2.3.3), with peak radiation reductions
in North America and Europe in the 1980s, and ongoing reduction in
South and East Asia (Wild, 2009). The AR4 and CMIP5 models simu-
lated these trends but underestimated their magnitude, the decadal
temperature variations and the diurnal temperature range over land
(Wild, 2009; see Chapter 9).
Changes in spatial patterns of species and their forcing over the cen-
tury are difficult to validate due to sparse observations of short-lived
species. Some constraint comes from limited historical observations
in ice core records and from shorter trends beginning in late century
from satellite and surface-based site measurements. The emissions
estimates for historical species are very uncertain, especially for car-
bonaceous aerosols and dust. Therefore, the confidence in the histor-
ical forcing pattern changes is low for RF due to aerosol–radiation
interaction and ozone, and very low for ERF, carbonaceous aerosols
and dust.
8.6.2.2 Relationship Between Regional Forcing Patterns and
Climate Response During the Industrial Era
An increasing body of research considers how spatial variations in RF
affect climate response. Detection and attribution methods have had
limited success in discerning statistically significant regional climate
signals from regional forcing, due to large internal climate variability
at regional scales, uncertainty in model processes and sparse region-
al observational records (Chapter 10). Meanwhile, research including
model sensitivity studies for NTCFs, which vary strongly in space in
time, explores climate response patterns.
In AR4 (Forster et al., 2007; Knutti et al., 2008) it was argued that the
spatial pattern of forcing is not indicative of the pattern of climate
response. Rather, the response is linked more closely to TOA flux result-
ing from the climate feedback spatial patterns (Boer and Yu, 2003;
Taylor et al., 2011; Ming and Ramaswamy, 2012), with the lapse rate,
surface albedo and cloud feedbacks explaining most of the temperature
response. Yet Crook and Forster (2011) showed that both the spatial
distribution of climate feedbacks and of heterogeneous forcing played
important roles in the patterns of 20th century temperature changes.
Other studies since AR4 have probed relationships between forcing
patterns and climate responses.
Broad links between forcing and climate response have been identi-
fied. Shindell et al. (2010) used multiple models to show that surface
temperature changes are much more sensitive to latitudinal than longi-
tudinal variations in forcing. Shindell and Faluvegi (2009) used a model
inverse approach to infer that NH aerosol reduction was associated
with more than 70% of Arctic warming from the 1970s to the 2000s,
and that Arctic and much of the SH surface temperature changes are
strongly affected by remote forcing changes (alsoSection 10.3.1.1.4).
Voulgarakis and Shindell (2010) defined a regional transient tempera-
ture sensitivity parameter, or temperature response per unit forcing for
each 4-degree latitude band. Using observed surface air temperature
changes they showed that the parameter is best constrained from 50°S
to 25°N, where the value is 0.35°C (W m
–2
)
–1
, smaller than at northern
higher latitudes, and 35% smaller than in AR4 models.
Some aerosol model studies have demonstrated highly localized cli-
mate response to regional forcing. Significant regional cooling and
hydrological shifts in the eastern USA and in Eastern Asia during the
last half of the 20th century were modelled and attributed to local
aerosols (Leibensperger et al., 2008, 2012a, 2012b; Chang et al., 2009)
and localized warming projected for aerosol reductions (Mickley et al.,
2012). Observations have also linked historical trends in aerosols and
temperature (Ruckstuhl et al., 2008; Philipona et al., 2009).
Since AR4, there has been new research on aerosol influences on the
hydrologic cycle (also Sections 7.4, 7.6.4, 10.3.3.1 and 11.3.2.4.3).
Increased aerosol loading, with greater surface energy flux reduction
in the NH, has been implicated in the observed southward shift of the
Intertropical Convergence Zone (ITCZ) towards the hemisphere with
smaller surface energy reduction: southward up to the 1980s with a
reversal since (e.g., Denman et al., 2007; Zhang et al., 2007). Several
studies have modelled an associated reduction in NH precipitation
and associated shifts in the Hadley circulation (e.g., Rotstayn et al.,
2000; Williams et al., 2001; Ming et al., 2011). The ITCZ shift may in
turn be responsible for broad regional precipitation changes, includ-
ing drying of the Sahel (e.g., Rotstayn and Lohmann, 2002; Biasutti
and Giannini, 2006; Kawase et al., 2010; Ackerley et al., 2011) and
northwestern Brazil (Cox et al., 2008), both of which peaked in the
1980s. These hemispheric asymmetric ITCZ effects are overlaid on
thermodynamic aerosol effects which moisten subtropical regions,
countering GHG-induced drying of these regions (Ming et al., 2011).
Studies indicate that aerosols are more effective than an equivalent
WMGHG forcing for shifting precipitation, and that historical trends
in several areas cannot be explained without including aerosol forcing
(Bollasina et al., 2011; Booth et al., 2012; Shindell et al., 2012a; Shin-
dell et al., 2012b). However, confidence in attribution of any human
influence on zonal shifts in precipitation distribution is only medium
(Section 10.3.2.2).
There is increasing evidence but limited agreement that absorbing
aerosols influence cloud distributions (Section 7.3.4.2). Absorbing
8
Anthropogenic and Natural Radiative Forcing Chapter 8
707
aerosols apparently have complex influences on precipitation in mon-
soon regions. Model studies of Stephens et al. (2004) and Miller et
al. (2004) showed that dust absorption over Africa enhances low-lev-
el convergence, vertical velocities and therefore local monsoon cir-
culation and precipitation. On the other hand, Kawase et al. (2010)
showed that biomass burning BC may cause the decreasing precipita-
tion trend seen in tropical Africa during austral summer, due to reduc-
tion in evaporation and enhanced subsidence. The aerosol effects on
the Indian monsoon are similarly complex, and have been the sub-
ject of numerous studies (e.g., Ramanathan et al., 2005; Chung and
Ramanathan, 2006; Lau et al., 2006; Wang et al., 2009; Bollasina et
al., (2011), but a clear picture of how the regional aerosol forcing
correlates with responses has not yet fully emerged. Attribution of
changes in monsoon to human influence generally has low confidence
(Section 10.3).
Stratospheric ozone loss modelling has demonstrated an effect on the
SH stratosphere similar to increased GHGs, cooling stratospheric tem-
peratures, strengthening the polar vortex and shifting the westerly jet
poleward; however causing cooler Antarctic surface temperatures, with
larger influence on austral summer conditions (Son et al., 2009; McLan-
dress et al., 2011; Thompson et al., 2011; see also Sections 10.3.3 and
11.3.2.4.3.) In the troposphere, models indicate that increased tropo-
spheric ozone has caused warming, proportionally more in the NH and
notably to the Arctic during winter, mainly during the second half of the
20th century (Shindell et al., 2006a).
Albedo changes due to land use and land cover changes exert a heter-
ogeneous climate forcing (Figure 8.9). The surface albedo brightened
on the one hand due to a shift from forest to brighter croplands, caus-
ing local cooling (e.g., Eliseev and Mokhov, 2011; Lee et al., 2011), but
also darkened due to the re-expansion of forests to higher latitudes
(Esper and Schweingruber, 2004) and increased vegetation height in
snowy regions (Bonfils et al., 2012; also Section 8.3.5). Model studies
have shown cooling from land use and land cover changes, especial-
ly over NH continents, although without demonstrating a detectable
signal in observations (Matthews et al., 2004).
In addition to land use and climate-induced vegetation changes, CO
2
affects vegetation forcing indirectly, reducing transpiration from plants
as stomata open less with increasing CO
2
, resulting in localized atmos-
pheric drying and warming (Section 11.3.2.3.1; Joshi and Gregory,
2008). These are not included in the standard RF (Section 8.1) and may
be considered feedbacks (Section 8.3.2). This is modelled to be largest
over the Amazon, the central African forest, and to some extent over
boreal and temperate forests (Andrews et al., 2011). In the coupled
climate modelling study of Lawrence and Chase (2010), the vegetation
changes caused significant reduction in evapotranspiration, drying and
warming in tropical and subtropical regions, with insignificant cool-
ing at higher latitudes. Overall, vegetation changes may have caused
modest cooling at high latitudes and warming at low latitudes, but the
uncertainties are large and confidence is very low.
Deposition of BC on snow and ice, and loss of snow and ice darken
the surface, reduces albedo, and enhances climate warming. Substan-
tial snow-cover reduction of North America leads to warmer North
American summertime temperature in models having a strong snow
albedo feedback. These forcings can also have non-local impacts that
result from enhanced land-ocean temperature contrast, increasing sur-
face convergence over land and divergence over oceans. A poleward
intensification of the high pressure patterns and subtropical jet may
also result (Fletcher et al., 2009). BC contributions to snow darken-
ing reduces snow cover, however the magnitude of the effect is very
uncertain (see Sections 7.5.2.3 and 8.3.4.4). A model study calculated
BC-albedo reduction to cause about 20% Arctic snow/ice cover reduc-
tion and 20% of Arctic warming over the previous century (Koch et al.,
2011). However, reductions in Arctic soot during the past two decades
(e.g., Hegg et al., 2009) have likely reversed that trend (e.g., Koch et al.,
2011; Skeie et al., 2011b; Lee et al., 2013). Cryospheric feedbacks and
atmospheric dynamical responses in models have an associated pole-
ward shift in the temperature response to aerosol–cloud interactions
(Kristjansson et al., 2005; Koch et al., 2009a; Chen et al., 2010).
Solar spectral (UV) irradiance variations along the solar cycle induce
ozone responses by modifying the ozone production rate through pho-
tolysis of molecular oxygen (Section 8.4.1.4.1), and the resulting dif-
ferential heating can drive circulation anomalies that lead to regional
temperature and precipitation changes (Haigh, 1999; Shindell et al.,
2006b; Frame and Gray, 2010; Gray et al., 2010). Such solar forcing
may influence natural modes of circulation such as the Northern Annu-
lar Mode (e.g., Shindell et al., 2001; de la Torre et al., 2006; Ineson et al.,
2011), the South Asian Summer Monsoon (Fan et al., 2009), the South-
ern Annular Mode (Kuroda and Kodera, 2005; Roscoe and Haigh, 2007)
or the ENSO (Mann et al., 2005). The pattern of temperature response
is less uniform than the forcing, for example, warming in the NH, but
little response in the SH due to temperature moderation by wind speed
enhancement effects on ocean circulation (Swingedouw et al., 2011).
Regional responses to solar forcing are mediated by the stratosphere,
so that reproducing such change requires spectrally varying solar forc-
ing rather than TSI forcing (Lee et al., 2009; Section 8.4.1.4).
Stratospheric aerosol clouds (also Section 8.4.2.2) from tropical erup-
tions spread poleward and can cover an entire hemisphere or the
globe, depending on the initial latitudinal spread. The aerosol eruption
cloud from the 1963 Agung was confined mainly to the SH; the 1982 El
Chichón mainly to the NH; and the 1991 Pinatubo covered the globe,
all with an e-folding lifetime of about 1 year (e.g., Antuña et al., 2003).
High-latitude eruptions typically stay confined to the high-latitude
regions with shorter lifetimes of 2 to 4 months (Kravitz and Robock,
2011). Volcanic aerosols primarily scatter solar radiation back to space,
but also absorb longwave radiation with the former larger by an order
of magnitude. Stratospheric aerosol absorption heats the layer where
they reside and produces distinct vertical and horizontal distributions
of the heating rate. The temperature and chemical effects of the aer-
osols also enhance ozone destruction, which somewhat counteracts
the radiative heating (Stenchikov et al., 2002). For tropical eruptions,
this may affect atmospheric dynamics, with a stronger polar vortex, a
positive mode of the Arctic Oscillation, and winter warming over NH
continents (Robock, 2000). Climate responses to solar and volcanic
forcings are further discussed in the context of detection and attribu-
tion of millennial climate change (see Section 10.7).
The study of how climate responds to regionally varying patterns of
forcing is critical for understanding how local activities impact regional
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
708
climate; however, the studies are exploratory and generally evoke very
low confidence. However there is medium to high confidence in some
qualitative but robust features, such as the damped warming of the
NH and shifting of the ITCZ from aerosols, and positive feedbacks from
high-latitude snow and ice albedo changes.
8.6.3 Spatial Evolution of Radiative Forcing and
Response for the Future
Most components of aerosols and ozone precursors are estimated to
decrease toward the end of this century in the RCPs except CH
4
in
RCP8.5 (Figure 8.2) and nitrate aerosols, though some species reach
the maximum amounts of emissions around the mid-21st century
(Figure 8.2). The RCPs therefore contrast with the emission scenarios
for TAR and AR4, which were based on Special Report on Emissions
Scenarios (SRES) and have future projections of larger increase in the
near-term climate forcers (NTCFs). It has been questioned whether
such low emission of NTCFs is possible in the future given the current
policies (Pozzer et al., 2012). This section surveys spatial differences in
the RF of aerosols and ozone for the future based on the RCPs.
Figure 8.26 shows the global distributions of changes in aerosol and
ozone forcings in 2030 and 2100 relative to 2000 for RCP2.6 and 8.5
(Shindell et al., 2013c). Both scenarios indicate reduced aerosol load-
ing, and thus positive forcing over Europe, North America and Asia by
2100 where RF is above +0.5 W m
–2
because of substantial reduction
of scattering aerosols. The global mean RF due to aerosol–radiation
interaction is estimated to be +0.12 and +0.08 W m
–2
for RCP2.6 and
8.5, respectively, in 2100. Though the RF by total anthropogenic aer-
osols is positive, reduced BC contributes substantial negative forcing
especially over the similar regions. The global mean carbonaceous RF
including both the effects of BC and OC is estimated to be –0.20 and
–0.11 W m
–2
for RCP2.6 and 8.5, respectively, in 2100. Early in the
century, on the other hand, both scenarios indicate increased negative
aerosol forcing over South Asia, with reversal between 2030 and 2100.
Emissions of BC, OC and SO
2
will reach their maximums early and
middle in the century for RCP2.6 and 8.5, respectively in India. In RCP6,
high emission levels of SO
2
in China persist until the mid-21st century
(Supplementary Material Figure 8.SM.1), and then it is predicted to
keep a high negative RF due to aerosol–radiation interaction over East
Asia. The RF due to aerosol–radiation interaction for carbonaceous aer-
osol is positive over East and South Asia in 2030 relative to 2000 for
RCP8.5 because BC emission is also larger in 2030. Over central and
southern Africa, a change in the future RF due to aerosol–radiation
interaction based on RCPs is not clear mainly because of uncertainties
in the wildfires emissions (see Section 7.3.5.3). The global mean total
RF due to aerosol–radiation interaction in the future is rather small due
to offsetting effects, with reductions in BC, increases in nitrate aero-
sols, and reductions in scattering aerosols each causing substantially
more forcing than the net.
Emissions and atmospheric loadings of natural aerosols are affected
by climate change. There is, however, no consensus among studies on
future trends of their changes for major natural aerosols, mineral dust
and sea salt, as indicated in Section 7.3.5.1. The spatial pattern of the
aerosol forcing may be influenced by natural aerosols due to reduction
in sea ice cover leading to increased emission of high-latitude sea
salt (Struthers et al., 2011; Takemura, 2012) and SOA from vegetation
changes (Tsigaridis and Kanakidou, 2007).
The simulations applying the RCPs indicate that the latitude of max-
imum emission of NTCFs, and therefore of maximum RF, is projected
to shift somewhat southward for the next few decades (in 2030 of
Figure 8.26). The shift of peak aerosol loading southward is expected to
cause the ITCZ to continue to shift northward. This, in combination with
warming and drying over tropical land, has been modelled to lead to
greatly enhanced drought conditions in the Amazon (Cox et al., 2008).
On the other hand, if the low-latitude aerosol is sufficiently absorbing,
broadening of the ITCZ convergence region and enhanced cloud cover
could result, as modelled for dust (Perlwitz and Miller, 2010).
Reductions in high-latitude BC are expected to contribute to reducing
Arctic forcing (e.g., Koch et al., 2011), due to reduction in BC deposi-
tion on snow as well as in absorption of sunlight over bright surface.
On the other hand, reduction in mid-high-latitude scattering aerosols
may offset all or part of the impact of the local Arctic forcing change
(Shindell et al., 2010; Koch et al., 2011).
Figure 8.26 also shows the ozone RF in 2030 and 2100 relative to 2000,
which includes changes both in tropospheric and stratospheric ozone.
Recovery of ozone in the stratosphere in the 21st century will result in
positive forcing in the SH high latitudes in comparison with the year
2000 for both the pathways. This is because of the reduced emissions
of ozone-depleting substances controlled under the Montreal Protocol,
with a small additional effect from a feedback of changes in temper-
ature and in the vertical circulation due to changes in stratospheric
compositions (Kawase et al., 2011; Lamarque et al., 2011). In the trop-
osphere, on the other hand, a large difference in the CH
4
emissions
between RCP8.5 and the other pathways shown in Figure 8.2 leads
to a different RF trend outside the SH high latitudes. Ozone recovery
in the stratosphere and ozone increase in the troposphere leads to a
positive RF all over the globe in RCP8.5 with a mean of +0.26 W m
–2
in 2100. The cancellation between positive RF due to ozone increase in
the stratosphere and negative RF due to ozone decrease in the tropo-
sphere results in a global mean RF of –0.12 W m
–2
in RCP2.6.
Figure 8.26 also shows the global distributions of changes in ERF due
to both aerosol–radiation and aerosol–cloud interactions in 2030 and
2100 relative to 2000 for RCP8.5. Although the ERF includes rapid
adjustments and therefore its magnitude is much larger than that of
RF due to aerosol–radiation interaction, the spatial pattern is generally
similar to RF. The ERF in 2100 shows positive values relative to 2000
in North America, Europe and Asia even with RCP8.5, which indicates
the aerosol forcing is projected to approach to the pre-industrial level.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
709
All aerosol RF
RCP 2.6
All aerosol RF
RCP 8.5
0.09 W m
-2
0.12 W m
-2
0.01 W m
-2
0.08 W m
-2
Carbonaceous aerosol RF
RCP 2.6
Carbonaceous aerosol RF
RCP 8.5
Ozone RF
RCP 2.6
Ozone RF
RCP 8.5
-0.12 W m
-2
0.26 W m
-2
-.38-.62-.88 -.12 .62.38.12 .88
2030 2100
-0.02 W m
-2
-0.11 W m
-2
-0.03 W m
-2
-0.20 W m
-2
0.00 W m
-2
0.11 W m
-2
-.75-1.25-1.75 -.25 2.51.5.5 3.5
0.27 W m
-2
0.90 W m
-2
All aerosol ERF
RCP 8.5
Figure 8.26 | Multi-model mean RF (W m
–2
) due to aerosol–radiation interaction of all anthropogenic aerosols (first and second rows) and anthropogenic carbonaceous (BC+OC)
aerosols (third and fourth rows), and total ozone (fifth and sixth rows) in 2030 (left) and 2100 (right) relative to 2000 for RCP2.6 (top each) and RCP8.5 (bottom each) based on
the ACCMIP simulations. The seventh row shows multi-model mean ERF (W m
–2
) by all anthropogenic aerosols in 2030 (left) and 2100 (right) relative to 2000 for RCP8.5. Global
area-weighted means are given in the upper right of each panel.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
710
8.7 Emission Metrics
8.7.1 Metric Concepts
8.7.1.1 Introduction
To quantify and compare the climate impacts of various emissions,
it is necessary to choose a climate parameter by which to measure
the effects; that is, RF, temperature response, and so forth. Thus, var-
ious choices are needed for the steps down the cause–effect chain
from emissions to climate change and impacts (Figure 8.27 and Box
8.4). Each step in the cause effect chain requires a modelling frame-
work. For assessments and evaluation one may—as an alternative to
models that explicitly include physical processes resulting in forcing
and responses—apply simpler measures or metrics that are based on
results from complex models. Metrics are used to quantify the contri-
butions to climate change of emissions of different substances and can
thus act as ‘exchange rates’ in multi-component policies or compar-
isons of emissions from regions/countries or sources/sectors. Metrics
are also used in areas such as Life Cycle Assessments and Integrated
Assessment Modelling (e.g., by IPCC WGIII).
Metrics can be given in absolute terms (e.g., K kg
–1
) or in relative terms
by normalizing to a reference gas — usually CO
2
. To transform the
effects of different emissions to a common scale — often called ‘CO
2
equivalent emissions’—the emission (E
i
) of component i can be mul-
tiplied with the adopted normalized metric (M
i
): M
i
× E
i
= CO
2
-eq
i
.
Ideally, the climate effects of the calculated CO
2
equivalent emissions
should be the same regardless of the mix of components emitted.
However, different components have different physical properties, and
a metric that establishes equivalence with regard to one effect cannot
guarantee equivalence with regard to other effects and over extended
time periods, for example, Lauder et al. (2013), O’Neill (2000), Smith
and Wigley (2000), Fuglestvedt et al. (2003).
Figure 8.27 | The cause–effect chain from emissions to climate change and impacts showing how metrics can be defined to estimate responses to emissions (left) and for develop-
ment of multi-component mitigation (right). The relevance of the various effects increases downwards but at the same time the uncertainty also increases. The dotted line on the
left indicates that effects and impacts can be estimated directly from emissions, while the arrows on the right side indicate how these estimates can be used in development of
strategies for reducing emissions. (Adapted from Fuglestvedt et al., 2003, and Plattner et al., 2009.)
Metrics do not define goals and policy—they are tools that enable
evaluation and implementation of multi-component policies (i.e.,
which emissions to abate). The most appropriate metric will depend
on which aspects of climate change are most important to a particu-
lar application, and different climate policy goals may lead to differ-
ent conclusions about what is the most suitable metric with which
to implement that policy, for example, Plattner et al. (2009); Tol et al.
(2012). Metrics that have been proposed include physical metrics as
well as more comprehensive metrics that account for both physical and
economic dimensions (see 8.7.1.5 and WGIII, Chapter 3).
This section provides an assessment that focuses on the scientific
aspects and utility of emission metrics. Extending such an assessment
to include more policy-oriented aspects of their performance and
usage such as simplicity, transparency, continuity, economic implica-
tions of usage of one metric over another, and so forth, is not given
here as this is beyond the scope of WGI. However, consideration of
such aspects is vital for user-assessments. In the following, the focus is
on the more well-known Global Warming Potential (GWP) and Global
Temperature change Potential (GTP), though other concepts are also
briefly discussed.
8.7.1.2 The Global Warming Potential Concept
The Global Warming Potential (GWP) is defined as the time-integrat-
ed RF due to a pulse emission of a given component, relative to a
pulse emission of an equal mass of CO
2
(Figure 8.28a and formula).
The GWP was presented in the First IPCC Assessment (Houghton et al.,
1990), stating ‘It must be stressed that there is no universally accepted
methodology for combining all the relevant factors into a single global
warming potential for greenhouse gas emissions. A simple approach
has been adopted here to illustrate the difficulties inherent in the
concept, ...’. Further, the First IPCC Assessment gave no clear physical
interpretation of the GWP.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
711
A direct interpretation is that the GWP is an index of the total energy
added to the climate system by a component in question relative to
that added by CO
2
. However, the GWP does not lead to equivalence
with temperature or other climate variables (Fuglestvedt et al., 2000,
2003; O’Neill, 2000; Daniel et al., 2012; Smith and Wigley, 2000;
Tanaka et al., 2009). Thus, the name ‘Global Warming Potential’ may be
somewhat misleading, and ‘relative cumulative forcing index’ would
be more appropriate. It can be shown that the GWP is approximately
equal to the ratio (normalizing by the similar expression for CO
2
) of the
equilibrium temperature response due to a sustained emission of the
species or to the integrated temperature response for a pulse emission
(assuming efficacies are equal for the gases that are compared; O’Neill,
2000; Prather, 2002; Shine et al., 2005a; Peters et al., 2011a; Azar and
Johansson, 2012).
The GWP has become the default metric for transferring emissions of
different gases to a common scale; often called ‘CO
2
equivalent emis-
sions’ (e.g., Shine, 2009). It has usually been integrated over 20, 100
or 500 years consistent with Houghton et al. (1990). Note, however
that Houghton et al. presented these time horizons as ‘candidates for
discussion [that] should not be considered as having any special sig-
nificance’. The GWP for a time horizon of 100 years was later adopted
as a metric to implement the multi-gas approach embedded in the
United Nations Framework Convention on Climate Change (UNFCCC)
and made operational in the 1997 Kyoto Protocol. The choice of time
horizon has a strong effect on the GWP values — and thus also on the
calculated contributions of CO
2
equivalent emissions by component,
sector or nation. There is no scientific argument for selecting 100 years
compared with other choices (Fuglestvedt et al., 2003; Shine, 2009).
The choice of time horizon is a value judgement because it depends
Box 8.4 | Choices Required When Using Emission Metrics
Time frames: One can apply a backward-looking (i.e., historical) or a forward-looking perspective on the responses to emissions. In
the forward-looking case one may use pulses of emissions, sustained emissions or emission scenarios. All choices of emission perturba-
tions are somewhat artificial and idealized, and different choices serve different purposes. One may use the level (e.g., degrees Celsius)
or rate of change (e.g., degrees Celsius per decade). Furthermore, the effects of emissions may be estimated at a particular time or be
integrated over time up to a chosen time horizon. Alternatively, discounting of future effects may be introduced (i.e., a weighting of
effects over time).
Type of effect or end-point: Radiative forcing, temperature change or sea level change, for example, could be examined (Figure
8.27). Metrics may also include eco/biological or socioeconomic damages. The choice of climate impact parameters is related to which
aspects of climate change are considered relevant for interpretation of ‘dangerous anthropogenic interference with the climate system’
(UNFCCC Article 2).
Spatial dimension for emission and response: Equal-mass emissions of NTCFs from different regions can induce varying global
mean climate responses, and the climate response also has a regional component irrespective of the regional variation in emissions.
Thus, metrics may be given for region of emission as well as region of response.
Some of the choices involved in metrics are scientific (e.g., type of model, and how processes are included or parameterized in the
models). Choices of time frames and climate impact are policy-related and cannot be based on science alone, but scientific studies can
be used to analyse different approaches and policy choices.
Figure 8.28 | (a) The Absolute Global Warming Potential (AGWP) is calculated by
integrating the RF due to emission pulses over a chosen time horizon; for example, 20
and 100 years (vertical lines). The GWP is the ratio of AGWP for component i over AGWP
for the reference gas CO
2
. The blue hatched field represents the integrated RF from a
pulse of CO
2
, while the green and red fields represent example gases with 1.5 and 13
years lifetimes, respectively. (b) The Global Temperature change Potential (GTP) is based
on the temperature response at a selected year after pulse emission of the same gases;
e.g., 20 or 100 years (vertical lines). See Supplementary Material Section 8.SM.11 for
equations for calculations of GWP and GTP.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
712
( )
on the relative weight assigned to effects at different times. Other
important choices include the background atmosphere on which the
GWP calculations are superimposed, and the way indirect effects and
feedbacks are included (see Section 8.7.1.4).
For some gases the variation in GWP with time horizon mainly reflects
properties of the reference gas, not the gas for which the GWP is cal-
culated. The GWP for NTCFs decreases with increasing time horizon, as
GWP is defined with the integrated RF of CO
2
in the denominator. As
shown in Figure 8.29, after about five decades the development in the
GWP for CH
4
is almost entirely determined by CO
2
. However, for long-
lived gases (e.g., SF
6
) the development in GWP is controlled by both the
increasing integrals of RF from the long-lived gas and CO
2
.
8.7.1.3 The Global Temperature change Potential Concept
Compared to the GWP, the Global Temperature change Potential (GTP;
Shine et al., 2005a) goes one step further down the cause–effect
chain (Figure 8.27) and is defined as the change in global mean sur-
face temperature at a chosen point in time in response to an emission
pulse—relative to that of CO
2
. Whereas GWP is integrated in time
(Figure 8.28a), GTP is an end-point metric that is based on tempera-
ture change for a selected year, t, (see Figure 8.28b with formula). Like
for the GWP, the impact from CO
2
is normally used as reference, hence,
for a component i, GTP(t)
i
= AGTP(t)
i
/ AGTP(t)
CO2
= ∆T((t)
i
/∆T(t)
CO2
,
where AGTP is the absolute GTP giving temperature change per unit
emission (see Supplementary Material Section 8.SM.11 for equations
and parameter values). Shine et al. (2005a) presented the GTP for both
pulse and sustained emission changes based on an energy balance
model as well as analytical equations. A modification was later intro-
duced (Shine et al., 2007) in which the time horizon is determined by
the proximity to a target year as calculated by using scenarios and
climate models (see Section 8.7.1.5).
Like GWP, the GTP values can be used for weighting the emissions
to obtain ‘CO
2
equivalents’ (see Section 8.7.1.1). This gives the
Figure 8.29 | Development of AGWP-CO
2
, AGWP-CH
4
and GWP-CH
4
with time hori-
zon. The yellow and blue curves show how the AGWPs changes with increasing time
horizon. Because of the integrative nature the AGWP for CH
4
(yellow curve) reaches a
constant level after about five decades. The AGWP for CO
2
continues to increase for cen-
turies. Thus the ratio which is the GWP (black curve) falls with increasing time horizon.
temperature effects of emissions relative to that of CO
2
for the chosen
time horizon. As for GWP, the choice of time horizon has a strong effect
on the metric values and the calculated contributions to warming.
In addition, the AGTP can be used to calculate the global mean temper-
ature change due to any given emission scenario (assuming linearity)
using a convolution of the emission scenarios and AGTP
i
:
(8.1)
where i is component, t is time, and s is time of emission (Berntsen and
Fuglestvedt, 2008; Peters et al., 2011b; Shindell et al., 2011).
By accounting for the climate sensitivity and the exchange of heat
between the atmosphere and the ocean, the GTP includes physical pro-
cesses that the GWP does not. The GTP accounts for the slow response
of the (deep) ocean, thereby prolonging the response to emissions
beyond what is controlled by the decay time of the atmospheric con-
centration. Thus the GTP includes both the atmospheric adjustment
time scale of the component considered and the response time scale
of the climate system.
The GWP and GTP are fundamentally different by construction and dif-
ferent numerical values can be expected. In particular, the GWPs for
NTCFs, over the same time frames, are higher than GTPs due to the
integrative nature of the metric. The GTP values can be significantly
affected by assumptions about the climate sensitivity and heat uptake
by the ocean. Thus, the relative uncertainty ranges are wider for the
GTP compared to GWP (see Section 8.7.1.4). The additional uncertainty
is a typical trade-off when moving along the cause–effect chain to an
effect of greater societal relevance (Figure 8.27). The formulation of the
ocean response in the GTP has a substantial effect on the values; thus
its characterization also represents a trade-off between simplicity and
accuracy. As for GWP, the GTP is also influenced by the background
atmosphere, and the way indirect effects and feedbacks are included
(see Section 8.7.1.4).
8.7.1.4 Uncertainties and Limitations related to Global Warming
Potential and Global Temperature change Potential
The uncertainty in the numerator of GWP; that is, the AGWP
i
(see for-
mula in Figure 8.28a) is determined by uncertainties in lifetimes (or
perturbation lifetimes) and radiative efficiency. Inclusion of indirect
effects increases uncertainties (see below). For the reference gas CO
2
,
the uncertainty is dominated by uncertainties in the impulse response
function (IRF) that describes the development in atmospheric concen-
tration that follows from an emission pulse (Joos et al., 2013); see Box
6.2 and Supplementary Material Section 8.SM.12. The IRF is sensitive
to model representation of the carbon cycle, pulse size and background
CO
2
concentrations and climate.
Based on a multi-model study, Joos et al. (2013) estimate uncertain-
ty ranges for the time-integrated IRF for CO
2
to be ±15% and ±25%
(5 to 95% uncertainty range) for 20- and 100-year time horizons,
respectively. Assuming quadratic error propagation, and ±10% uncer-
tainty in radiative efficiency, the uncertainty ranges in AGWP for CO
2
were estimated to be ±18% and ±26% for 20 and 100 years. These
∆ =
!

!
− 
!
!
!
8
Anthropogenic and Natural Radiative Forcing Chapter 8
713
uncertainties affect all metrics that use CO
2
as reference. Reisinger et
al. (2010) and Joos et al. (2013) show that these uncertainties increase
with time horizon.
The same factors contribute to uncertainties in the GTP, with an addi-
tional contribution from the parameters describing the ocean heat
uptake and climate sensitivity. In the first presentation of the GTP,
Shine et al. (2005a) used one time constant for the climate response in
their analytical expression. Improved approaches were used by Bouch-
er and Reddy (2008), Collins et al. (2010) and Berntsen and Fuglestvedt
(2008) that include more explicit representations of the deep ocean
that increased the long-term response to a pulse forcing. Over the
range of climate sensitivities from AR4, GTP
50
for BC was found to vary
by a factor of 2, the CH
4
GTP
50
varied by about 50%, while for N
2
O
essentially no dependence was found (Fuglestvedt et al., 2010). AGTPs
for CO
2
were also calculated in the multi-model study by Joos et al.
(2013). They found uncertainty ranges in AGTP that are much larger
than for AGWP; ±45% and ±90% for 20 and 100 years (5 to 95%
uncertainty range). These uncertainty ranges also reflect the signal-to-
noise ratio, and not only uncertainty in the physical mechanisms.
There are studies combining uncertainties in various input parameters.
Reisinger et al. (2011) estimated the uncertainty in the GWP for CH
4
and found an uncertainty of –30 to +40% for the GWP
100
and –50 to
+75% for GTP
100
of CH
4
(for 5 to 95% of the range). Boucher (2012)
performed a Monte Carlo analysis with uncertainties in perturbation
lifetime and radiative efficiency, and for GWP
100
for CH
4
(assuming a
constant background atmosphere) he found ±20%, and –40 to +65 for
GTP
100
(for 5 to 95% uncertainty range).
Here we estimate uncertainties in GWP values based on the uncer-
tainties given for radiative efficiencies (Section 8.3.1), perturbation
lifetimes, indirect effects and in the AGWP for the reference gas CO
2
(see Supplementary Material Section 8.SM.12). For CH
4
GWP we esti-
mate an uncertainty of ±30% and ±40% for 20- and 100-year time
horizons, respectively (for 5 to 95% uncertainty range). The uncertainty
is dominated by AGWP for CO
2
and indirect effects. For gases with life-
times of a century or more the uncertainties are of the order of ±20%
and ±30% for 20- and 100-year horizons. The uncertainty in GWPs for
gases with lifetimes of a few decades is estimated to be of the order
of ±25% and ±35% for 20 and 100 years. For shorter-lived gases, the
uncertainties in GWPs will be larger (see Supplementary Material Sec-
tion 8.SM.12 for a discussion of contributions to the total uncertainty.)
For GTP, few uncertainty estimates are available in the literature. Based
on the results from Joos et al. (2013), Reisinger et al. (2010) and Bou-
cher (2012) we assess the uncertainty to be of the order of ±75% for
the CH
4
GTP
100
.
The metric values are also strongly dependent on which processes
are included in the definition of a metric. Ideally all indirect effects
(Sections 8.2 and 8.3) should be taken into account in the calculation
of metrics. The indirect effects of CH
4
on its own lifetime, tropospher-
ic ozone and stratospheric water have been traditionally included in
its GWP. Boucher et al. (2009) have quantified an indirect effect on
CO
2
when fossil fuel CH
4
is oxidized in the atmosphere. Shindell et
al. (2009) estimated the impact of reactive species emissions on both
gaseous and aerosol forcing species and found that ozone precursors,
including CH
4
, had an additional substantial climate effect because
they increased or decreased the rate of oxidation of SO
2
to sulphate
aerosol. Studies with different sulphur cycle formulations have found
lower sensitivity (Collins et al., 2010; Fry et al., 2012). Collins et al.
(2010) postulated an additional component to their GWPs and GTPs
for ozone precursors due to the decreased productivity of plants under
higher levels of surface ozone. This was estimated to have the same
magnitude as the ozone and CH
4
effects. This effect, however, has
so far only been examined with one model. In a complex and inter-
connected system, feedbacks can become increasingly complex, and
uncertainty of the magnitude and even direction of feedback increases
the further one departs from the primary perturbation, resulting in a
trade-off between completeness and robustness, and hence utility for
decision-making.
Gillett and Matthews (2010) included climate–carbon feedbacks in
calculations of GWP for CH
4
and N
2
O and found that this increased
the values by about 20% for 100 years. For GTP of CH
4
they found
an increase of ~80%. They used numerical models for their studies
and suggest that climate–carbon feedbacks should be considered and
parameterized when used in simple models to derive metrics. Col-
lins et al. (2013) parameterize the climate-carbon feedback based on
Friedlingstein et al. (2006) and Arora et al. (2013) and find that this
more than doubles the GTP
100
for CH
4
. Enhancement of the GTP for
CH
4
due to carbon–climate feedbacks may also explain the higher GTP
values found by Reisinger et al. (2010).
The inclusion of indirect effects and feedbacks in metric values has
been inconsistent in the IPCC reports. In SAR and TAR, a carbon model
without a coupling to a climate model was used for calculation of IRF
for CO
2
(Joos et al., 1996), while in AR4 climate-carbon feedbacks were
included for the CO
2
IRF (Plattner et al., 2008). For the time horizons
20 and 100 years, the AGWP
CO2
calculated with the Bern3D-LPJ model
is, depending on the pulse size, 4 to 5% and 13 to 15% lower, respec-
tively, when carbon cycle–climate feedbacks are not included (Joos
et al., 2013). While the AGWP for the reference gas CO
2
included cli-
mate–carbon feedbacks, this is not the case for the non-CO
2
gas in the
numerator of GWP, as recognized by Gillett and Matthews (2010), Joos
et al. (2013), Collins et al. (2013) and Sarofim (2012). This means that
the GWPs presented in AR4 may underestimate the relative impacts
of non-CO
2
gases. The different inclusions of feedbacks partially repre-
sent the current state of knowledge, but also reflect inconsistent and
ambiguous definitions. In calculations of AGWP for CO
2
in AR5 we use
the IRF for CO
2
from Joos et al. (2013) which includes climate–carbon
feedbacks. Metric values in AR5 are presented both with and without
including climate–carbon feedbacks for non-CO
2
gases. This feedback
is based on the carbon-cycle response in a similar set of models (Arora
et al., 2013) as used for the reference gas (Collins et al., 2013).
The effect of including this feedback for the non-reference gas increas-
es with time horizon due to the long-lived nature of the initiated CO
2
perturbation (Table 8.7). The relative importance also increases with
decreasing lifetime of the component, and is larger for GTP than GWP
due to the integrative nature of GWP. We calculate an increase in the
CH
4
GWP
100
of 20%. For GTP
100
, however, the changes are much larger;
of the order of 160%. For the shorter time horizons (e.g., 20 years)
the effect of including this feedback is small (<5%) for both GWP
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
714
Table 8.7 | GWP and GTP with and without inclusion of climate–carbon feedbacks (cc fb) in response to emissions of the indicated non-CO
2
gases (climate-carbon feedbacks in
response to the reference gas CO
2
are always included).
Lifetime (years) GWP
20
GWP
100
GTP
20
GTP
100
CH
4
b
12.4
a
No cc fb 84 28 67 4
With cc fb 86 34 70 11
HFC-134a
13.4 No cc fb 3710 1300 3050 201
With cc fb 3790 1550 3170 530
CFC-11
45.0 No cc fb 6900 4660 6890 2340
With cc fb 7020 5350 7080 3490
N
2
O
121.0
a
No cc fb 264 265 277 234
With cc fb 268 298 284 297
CF
4
50,000.0 No cc fb 4880 6630 5270 8040
With cc fb 4950 7350 5400 9560
and GTP. For the more long-lived gases the GWP
100
values increase
by 10 to 12%, while for GTP
100
the increase is 20 to 30%. Table 8.A.1
gives metric values including the climate–carbon feedback for CO
2
only, while Supplementary Material Table 8.SM.16 gives values for all
halocarbons that include the climate–carbon feedback. Though uncer-
tainties in the carbon cycle are substantial, it is likely that including
the climate–carbon feedback for non-CO
2
gases as well as for CO
2
provides a better estimate of the metric value than including it only
for CO
2
.
Emission metrics can be estimated based on a constant or variable
background climate and this influences both the adjustment times and
the concentration–forcing–temperature relationships. Thus, all metric
values will need updating due to changing atmospheric conditions
as well as improved input data. In AR5 we define the metric values
with respect to a constant present-day condition of concentrations and
climate. However, under non-constant background, Joos et al. (2013)
found decreasing CO
2
AGWP
100
for increasing background levels (up to
23% for RCP8.5). This means that GWP for all non-CO
2
gases (except
CH
4
and N
2
O) would increase by roughly the same magnitude. Reising-
er et al. (2011) found a reduction in AGWP for CO
2
of 36% for RCP8.5
from 2000 to 2100 and that the CH
4
radiative efficiency and AGWP
also decrease with increasing CH
4
concentration. Accounting for both
effects, the GWP
100
for CH
4
would increase by 10 to 20% under low
and mid-range RCPs by 2100, but would decrease by up to 10% by
mid-century under the highest RCP. While these studies have focused
on the background levels of GHGs, the same issues apply for tempera-
ture. Olivié et al. (2012) find different temperature IRFs depending on
the background climate (and experimental set up).
User related choices (see Box 8.4) such as the time horizon can greatly
affect the numerical values obtained for CO
2
equivalents. For a change
in time horizon from 20 to 100 years, the GWP for CH
4
decreases by
a factor of approximately 3 and its GTP by more than a factor of 10.
Short-lived species are most sensitive to this choice. Some approaches
have removed the time horizon from the metrics (e.g., Boucher, 2012),
but discounting is usually introduced which means that a discount rate
r (for the weighting function e
–rt
) must be chosen instead. The choice of
discount rate is also value based (see WGIII, Chapter 3).
For NTCFs the metric values also depend on the location and timing
of emission and whether regional or global metrics are used for these
gases is also a choice for the users. Metrics are usually calculated for
pulses, but some studies also give metric values that assume constant
emissions over the full time horizon (e.g., Shine et al., 2005a; Jacobson,
2010). It is important to be aware of the idealized assumption about
constant future emissions (or change in emissions) of the compound
being considered if metrics for sustained emissions are used.
8.7.1.5 New Metric Concepts
New metric concepts have been developed both to modify physical
metrics to address shortcomings as well as to replace them with met-
rics that account for economic dimensions of problems to which met-
rics are applied. Modifications to physical metrics have been proposed
to better represent CO
2
emissions from bioenergy, regional patterns of
response, and for peak temperature limits.
Emissions of CO
2
from the combustion of biomass for energy in nation-
al emission inventories are currently assumed to have no net RF, based
on the assumption that these emissions are compensated by biomass
regrowth (IPCC, 1996). However, there is a time lag between combus-
tion and regrowth, and while the CO
2
is resident in the atmosphere
it leads to an additional RF. Modifications of the GWP and GTP for
bioenergy (GWP
bio
, GTP
bio
) have been developed (Cherubini et al., 2011;
Cherubini et al., 2012). The GWP
bio
give values generally between zero
(current default for bioenergy) and one (current for fossil fuel emissions)
(Cherubini et al., 2011), and negative values are possible for GTP
bio
due to the fast time scale of atmospheric–ocean CO
2
exchange relative
to the growth cycle of biomass (Cherubini et al., 2012). GWP
bio
and
GTP
bio
have been used in only a few applications, and more research is
needed to assess their robustness and applicability. Metrics for bioge-
ophysical effects, such as albedo changes, have been proposed (Betts,
2000; Rotenberg and Yakir, 2010) , but as for NTCFs regional variations
Notes:
Uncertainties related to the climate–carbon feedback are large, comparable in magnitude to the strength of the feedback for a single gas.
a
Perturbation lifetime is used in the calculation of metrics.
b
These values do not include CO
2
from methane oxidation. Values for fossil methane are higher by 1 and 2 for the 20 and 100 year metrics, respectively (Table 8.A.1).
8
Anthropogenic and Natural Radiative Forcing Chapter 8
715
are important (Claussen et al., 2001) and the RF concept may not be
adequate (Davin et al., 2007).
New concepts have also been developed to capture information
about regional patterns of responses and cancelling effects that are
lost when global mean metrics are used. The use of nonlinear damage
functions to capture information on the spatial pattern of responses
has been explored (Shine et al., 2005b; Lund et al., 2012). In addi-
tion, the Absolute Regional Temperature Potential (ARTP) (Shindell,
2012; Collins et al., 2013) has been developed to provide estimates
of impacts at a sub-global scale. ARTP gives the time-dependent tem-
perature response in four latitude bands as a function of the regional
forcing imposed in all bands. These metrics, as well as new regional
precipitation metrics (Shindell et al., 2012b), require additional studies
to determine their robustness.
Alternatives to the single basket approach adopted by the Kyoto Pro-
tocol are a component-by-component approach or a multi-basket
approach (Rypdal et al., 2005; Daniel et al., 2012; Sarofim, 2012; Jack-
son, 2009). Smith et al. (2012) show how peak temperature change is
constrained by cumulative emissions (see 12.5.4) for gases with long
lifetimes and emissions rates for shorter-lived gases (including CH
4
).
Thus, they divide gases into two baskets and present two metrics that
can be used for estimating peak temperature for various emission sce-
narios. This division of gases into the two baskets is sensitive to the
time of peak temperature in the different scenarios. The approach uses
time invariant metrics that do not account for the timing of emissions
relative to the target year. The choice of time horizon is implicit in the
scenario assumed and this approach works only for a peak scenario.
A number of new metrics have been developed to add economic
dimensions to purely physically based metrics such as the GWP and
GTP. The use of physical metrics in policy contexts has been criticized
by economists (Reilly and Richards, 1993; Schmalensee, 1993; Hammitt
et al., 1996; Reilly et al., 1999; Bradford, 2001; De Cara et al., 2008). A
prominent use of metrics is to set relative prices of gases when imple-
menting a multi-gas policy. Once a particular policy has been agreed
on, economic metrics can address policy goals more directly than phys-
ical metrics by accounting not only for physical dimensions but also
for economic dimensions such as mitigation costs, damage costs and
discount rates (see WGIII, Chapter 3; Deuber et al., 2013).
For example, if mitigation policy is set within a cost-effectiveness
framework with the aim of making the least cost mix of emissions
reductions across components to meet a global temperature target,
the ‘price ratio’ (Manne and Richels, 2001), also called the Global Cost
Potential (GCP) (Tol et al., 2012), most directly addresses the goal. The
choice of target is a policy decision; metric values can then be calcu-
lated based on an agreed upon target. Similarly, if policy is set within
a cost–benefit framework, the metric that directly addresses the policy
goal is the ratio of the marginal damages from the emission of a gas
(i.e., the damage costs to society resulting from an incremental increase
in emissions) relative to the marginal damages of an emission of CO
2
,
known as the Global Damage Potential (GDP) (Kandlikar, 1995). Both
types of metrics are typically determined within an integrated climate–
economy model, since they are affected both by the response of the
climate system as well as by economic factors.
If other indexes, such as the GWP, are used instead of an economic
cost-minimizing index, costs to society will increase. Cost implications
at the project or country level could be substantial under some cir-
cumstances (Godal and Fuglestvedt, 2002; Shine, 2009; Reisinger et
al., 2013). However, under idealized conditions of full participation in
mitigation policy, the increase is relatively small at the global level,
particularly when compared to the cost savings resulting from a multi-
(as opposed to single-) gas mitigation strategy even when based on
an imperfect metric (O’Neill, 2003; Aaheim et al., 2006; Johansson et
al., 2006; Johansson, 2012; Reisinger et al., 2013; Smith et al., 2013).
Purely physical metrics continue to be used in many contexts due at
least in part to the added uncertainties in mitigation and damage
costs, and therefore in the values of economic metrics (Boucher, 2012).
Efforts have been made to view purely physical metrics such as GWPs
and GTPs as approximations of economic indexes. GTPs, for example,
can be interpreted as an approximation of a Global Cost Potential
designed for use in a cost-effectiveness setting (Shine et al., 2007; Tol
et al., 2012). Quantitative values for time-dependent GTPs reproduce
in broad terms several features of the Global Cost Potential such as the
rising value of metrics for short-lived gases as a climate policy target is
approached (Tanaka et al., 2013). Figure 8.30 shows how contributions
of N
2
O, CH
4
and BC to warming in the target year changes over time.
The contributions are given relative to CO
2
and show the effects of
emission occurring at various times. Similarly, GWPs can be interpret-
ed as approximations of the Global Damage Potential designed for a
cost–benefit framework (Tol et al., 2012). These interpretations of the
GTP and GWP imply that using even a purely physical metric in an eco-
nomic policy context involves an implicit economic valuation.
In both cases, a number of simplifying assumptions must be made
for these approximations to hold (Tol et al., 2012). For example, in
the case of the GWP, the influence of emissions on RF, and therefore
implicitly on costs to society, beyond the time horizon is not taken
into account, and there are substantial numerical differences between
GWP and GDP values (Marten and Newbold, 2012). In the case of the
GTP, the influence of emissions on temperature change (and costs) is
Figure 8.30 | Global Temperature change Potential (GTP(t)) for CH
4
, nitrous oxide
and BC for each year from year of emission to the time at which the temperature
change target is reached. The (time-invariant) GWP
100
is also shown for N
2
O and CH
4
for comparison.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
716
included only at the time the target is reached, but not before nor
after. Other metrics have been developed to more closely approximate
GCPs or GDPs. The Cost-Effective Temperature Potential (CETP) repro-
duces values of the GCP more closely than does the GTP (Johansson,
2012). It is similar to the GTP but accounts for post-target temperature
effects based on an assumption about how to value costs beyond the
time the target is reached. Metrics have also been proposed that take
into account forcing or temperature effects that result from emissions
trajectories over broad time spans, and that behave similarly to GCP
and GTP (Tanaka et al., 2009; Manning and Reisinger, 2011) or to GWP
(e.g., O’Neill, 2000; Peters et al., 2011a; Gillett and Matthews, 2010;
Azar and Johansson, 2012).
8.7.1.6 Synthesis
In the application and evaluation of metrics, it is important to distin-
guish between two main sources of variation in metric values. While
scientific choices of input data have to be made, there are also choic-
es involving value judgements. For some metrics such choices are not
always explicit and transparent. The choice of metric type and time
horizon will for many components have a much larger effect than
improved estimates of input parameters and can have strong effects
on perceived impacts of emissions and abatement strategies.
In addition to progress in understanding of GWP, new concepts have
been introduced or further explored since AR4. Time variant metrics
introduce more dynamical views of the temporal contributions that
accounts for the proximity to a prescribed target (in contrast to the tra-
ditional static GWP). Time variant metrics can be presented in a format
that makes changing metric values over time predictable.
As metrics use parameters further down the cause effect chain the met-
rics become in general more policy relevant, but at the same time the
uncertainties increase. Furthermore, metrics that account for regional
variations in sensitivity to emissions or regional variation in response
could give a very different emphasis to various emissions. Many spe-
cies, especially NTCFs, produce distinctly regionally heterogeneous RF
and climate response patterns. These aspects are not accounted for in
the commonly used global scale metrics.
The GWPs and GTPs have had inconsistent treatment of indirect effects
and feedbacks. The GWPs reported in AR4 include climate–carbon
feedbacks for the reference gas CO
2
but not for the non-CO
2
gases.
Such feedbacks may have significant impacts on metrics and should be
treated consistently. More studies are needed to assess the importance
of consistent treatment of indirect effects/feedbacks in metrics.
The weighting of effects over time—choice of time horizon in the
case of GWP and GTP—is value based. Discounting is an alternative,
which also includes value judgements and is equally controversial. The
weighting used in the GWP is a weight equal to one up to the time hori-
zon and zero thereafter, which is not in line with common approaches
for evaluation of future effects in economics (e.g., as in WGIII, Chapter
3). Adoption of a fixed horizon of e.g., 20, 100 or 500 years will inev-
itably put no weight on the long-term effect of CO
2
beyond the time
horizon (Figure 8.28 and Box 6.1). While GWP integrates the effects up
to a chosen time horizon the GTP gives the temperature just for one
chosen year with no weight on years before or after. The most appro-
priate metric depends on the particular application and which aspect
of climate change is considered relevant in a given context. The GWP
is not directly related to a temperature limit such as the 2°C target
(Manne and Richels, 2001; Shine et al., 2007; Manning and Reisinger,
2011; Smith et al., 2012; Tol et al., 2012; Tanaka et al., 2013), whereas
some economic metrics and physical end-point metrics like the GTP
may be more suitable for this purpose.
To provide metrics that can be useful to the users and policymakers
a more effective dialog and discussion on three topics is needed: (1)
which applications particular metrics are meant to serve; (2) how com-
prehensive metrics need to be in terms of indirect effects and feed-
backs, and economic dimensions; and—related to this (3) how impor-
tant it is to have simple and transparent metrics (given by analytical
formulations) versus more complex model-based and thus model-de-
pendent metrics. These issues are also important to consider in a wider
disciplinary context (e.g., across the IPCC Working Groups). Finally, it
is important to be aware that all metric choices, even ‘traditional’ or
‘widely used’ metrics, contain implicit value judgements as well as
large uncertainties.
8.7.2 Application of Metrics
8.7.2.1 Metrics for Carbon Dioxide, Methane, Nitrous Oxide,
Halocarbons and Related Compounds
Updated (A)GWP and (A)GTP values for CO
2
, CH
4
, N
2
O, CFCs, HCFCs,
bromofluorocarbons, halons, HFCs, PFCs, SF
6
, NF
3
, and related halogen-
containing compounds are given for some illustrative and tentative
time horizons in Tables 8.7, 8.A.1 and Supplementary Material Table
8.SM.16. The input data and methods for calculations of GWPs and
GTPs are documented in the Supplementary Material Section 8.SM.13.
Indirect GWPs that account for the RF caused by depletion of strat-
ospheric ozone (consistent with Section 8.3.3) are given for selected
gases in Table 8.A.2.
The confidence in the ability to provide useful metrics at time scales of
several centuries is very low due to nonlinear effects, large uncertain-
ties for multi-century processes and strong assumptions of constant
background conditions. Thus, we do not give metric values for longer
time scales than 100 years (see discussion in Supplementary Material
Section 8.SM.11). However, these time scales are important to consider
for gases such as CO
2
, SF
6
and PFCs. For CO
2
, as much as 20 to 40% of
the initial increase in concentration remains after 500 years. For PFC-
14, 99% of an emission is still in the atmosphere after 500 years. The
effects of emissions on these time scales are discussed in Chapter 12.
The GWP values have changed from previous assessments due to
new estimates of lifetimes, impulse response functions and radiative
efficiencies. These are updated due to improved knowledge and/or
changed background levels. Because CO
2
is used as reference, any
changes for this gas will affect all metric values via AGWP changes.
Figure 8.31 shows how the values of radiative efficiency (RE), integrat-
ed impulse response function (IRF) and consequentially AGWP for CO
2
have changed from earlier assessments relative to AR5 values. The net
effect of change in RE and IRF is an increase of approximately 1% and
8
Anthropogenic and Natural Radiative Forcing Chapter 8
717
6% from AR4 to AR5 in AGWP for CO
2
for 20 and 100 years, respective-
ly (see Supplementary Material Section 8.SM.12). These increases in
the AGWP of the reference gas lead to corresponding decreases in the
GWPs for all non-CO
2
gases. Continued increases in the atmospheric
levels of CO
2
will lead to further changes in GWPs (and GTPs) in the
future.
To understand the factors contributing to changes relative to AR4,
comparisons are made here using the AR5 values that include climate–
carbon feedbacks for CO
2
only. Relative to AR4 the CH
4
AGWP has
changed due to changes in perturbation lifetime, a minor change in RE
due to an increase in background concentration, and changes in the
estimates of indirect effects. The indirect effects on O
3
and stratospheric
H
2
O are accounted for by increasing the effect of CH
4
by 50% and 15%,
respectively (see Supplementary Material Table 8.SM.12). The ozone
effect has doubled since AR4 taking into account more recent studies
as detailed in Sections 8.3.3 and 8.5.1. Together with the changes in
AGWP for CO
2
the net effect is increased GWP values of CH
4
.
The GWPs for N
2
O are lower here compared to AR4. A longer perturba-
tion lifetime is used in AR5, while the radiative efficiency is lower due
to increased abundances of CH
4
and N
2
O. In addition, the reduction in
CH
4
via stratospheric O
3
, UV fluxes and OH levels due to increased N
2
O
abundance is included in GWPs and GTP. Owing to large uncertainties
related to altitude of changes, we do not include the RF from strato-
spheric ozone changes as an indirect effect of N
2
O.
Lifetimes for most of the halocarbons are taken from WMO (2011) and
many of these have changed from AR4. The lifetimes of CFC-114, CFC-
115 and HCF-161 are reduced by approximately 40%, while HFC-152
Figure 8.31 | Changes in the radiative efficiency (RE), integrated impulse response
function (IRF) and Absolute Global Warming Potential (AGWP) for CO
2
for 100 years
from earlier IPCC Assessment Reports normalized relative to the values given in AR5.
The ‘original’ values are calculated based on the methods explained or value reported
in each IPCC Assessment Report. The ‘updated’ values are calculated based on the
methods used in AR5, but the input values from each Assessment Report. The differ-
ence is primarily in the formula for the RE, which was updated in TAR. The different
integrated IRF in TAR relates to a different parameterisation of the same IRF (WMO,
1999). Changes represent both changes in scientific understanding and a changing
background atmospheric CO
2
concentration (note that y-axis starts from 0.8). The lines
connecting individual points are meant as a visual guide and not to represent the values
between different Assessment Reports.
is reduced by one third. Among the hydrofluoroethers (HFEs) there are
also several large changes in lifetimes. In addition, substantial updates
of radiative efficiencies are made for several important gases; CFC-
11, CFC-115, HCFC-124, HCFC-225cb, HFC-143a, HFC-245fa, CCl
4
,
CHCl
3
, and SF
6
. The radiative efficiency for carbon tetrachloride (CCl
4
)
is higher now and the GWP
100
has increased by almost 25% from
AR4. Uncertainties in metric values are given in Section 8.7.1.4. See
also Supplementary Material Section 8.SM.12 and footnote to Table
8.A.1. As can be seen from Table 8.A.2, some ODS have strong indi-
rect effects through stratospheric ozone forcing, which for some of the
gases reduce their net GWP
100
values substantially (and for the halons,
to large negative values). Note that, consistent with Section 8.3.3, the
uncertainties are large; ±100% for this indirect effect.
When climate-carbon feedbacks are included for both the non-CO
2
and
reference gases, all metric values increase relative to the methodolo-
gy used in AR4, sometimes greatly (Table 8.7, Supplementary Material
Table 8.SM.16). Though the uncertainties range for these metric values
is greater, as uncertainties in climate-carbon feedbacks are substantial,
these calculations provide a more consistent methodology.
8.7.2.2 Metrics for Near-Term Climate Forcers
The GWP concept was initially used for the WMGHGs, but later for
NTCFs as well. There are, however, substantial challenges related to
calculations of GWP (and GTP) values for these components, which
is reflected in the large ranges of values in the literature. Below we
present and assess the current status of knowledge and quantification
of metrics for various NTCFs.
8.7.2.2.1 Nitrogen oxides
Metric values for NO
X
usually include the short-lived ozone effect,
CH
4
changes and the CH
4
-controlled O
3
response. NO
X
also causes RF
through nitrate formation, and via CH
4
it affects stratospheric H
2
O and
through ozone it influences CO
2
. In addition, NO
x
affects CO
2
through
nitrogen deposition (fertilization effect). Due to high reactivity and
the many nonlinear chemical interactions operating on different time
scales, as well as heterogeneous emission patterns, calculation of net
climate effects of NO
X
is difficult. The net effect is a balance of large
opposing effects with very different temporal behaviours. There is also
a large spread in values among the regions due to variations in chem-
ical and physical characteristics of the atmosphere.
As shown in Table 8.A.3 the GTP and GWP values are very different.
This is due to the fundamentally different nature of these two metrics
(see Figure 8.28) and the way they capture the temporal behaviour of
responses to NO
x
emissions. Time variation of GTP for NO
X
is complex,
which is not directly seen by the somewhat arbitrary choices of time
horizon, and the net GTP is a fine balance between the contributing
terms. The general pattern for NO
X
is that the short-lived ozone forc-
ing is always positive, while the CH
4
-induced ozone forcing and CH
4
forcing are always negative (see Section 8.5.1). Nitrate aerosols from
NO
x
emission are not included in Table 8.A.3. For the GTP, all estimates
for NO
X
from surface sources give a negative net effect. As discussed
in Section 8.7.1.4 Collins et al. (2010) and Shindell et al. (2009) imple-
mented further indirect effects, but these are not included in Table
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
718
8.A.3 due to large uncertainties. The metric estimates for NO
X
reflect
the level of knowledge, but they also depend on experimental design,
treatment of transport processes, and modelling of background levels.
The multi-model study by Fry et al. (2012) shows the gaseous chemistry
response to NO
X
is relatively robust for European emissions, but that
the uncertainty is so large that for some regions of emissions it is not
possible to conclude whether NO
X
causes cooling or warming.
8.7.2.2.2 Carbon monoxide and volatile organic compounds
Emissions of carbon monoxide (CO) and volatile organic compounds
(VOCs) lead to production of ozone on short time scales. By affecting
OH and thereby the levels of CH
4
they also initiate a positive long-term
ozone effect. With its lifetime of 2 to 3 months, the effect of CO emis-
sions is less dependent on location than is the case for NO
X
(see Table
8.A.4). There is also less variation across models. However, Collins et
al. (2010) found that inclusion of vegetation effects of O
3
increased the
GTP values for CO by 20 to 50%. By including aerosol responses Shin-
dell et al. (2009) found an increase in GWP
100
by a factor of ~2.5. CO of
fossil origin will also have a forcing effect by contributing to CO
2
levels.
This effect adds 1.4 to 1.6 to the GWP
100
for CO (Daniel and Solomon,
1998; Derwent et al., 2001). (The vegetation and aerosol effects are not
included in the numbers in Table 8.A.4.)
VOC is not a well-defined group of hydrocarbons. This group of gases
with different lifetimes is treated differently across models by lump-
ing or using representative key species. However, the spread in metric
values in Table 8.A.5 is moderate across regions, with highest values
for emissions in South Asia (of the four regions studied). The effects
via ozone and CH
4
cause warming, and the additional effects via inter-
actions with aerosols and via the O
3
–CO
2
link increase the warming
effect further. Thus, the net effects of CO and VOC are less uncertain
than for NO
X
for which the net is a residual between larger terms of
opposite sign. However, the formation of SOAs is usually not included
in metric calculations for VOC, which introduces a cooling effect and
increased uncertainty.
8.7.2.2.3 Black carbon and organic carbon
Most of the metric values for BC in the literature include the aero-
sol–radiation interaction and the snow/ice albedo effect of BC, though
whether external or internal mixing is used varies between the studies.
Bond et al. (2011) calculate GWPs and find that when the albedo effect
is included the values increase by 5 to 15%. Studies have shown, how-
ever, that the climate response per unit forcing to this mechanism is
stronger than for WMGHG (see Section 7.5).
Bond et al. (2013) assessed the current understanding of BC effects
and calculated GWP and GTP for BC that includes aerosol–radiation
interaction, aerosol–cloud interactions and albedo. As shown in Table
8.A.6 the uncertainties are wide for both metrics (for 90% uncertain-
ty range) reflecting the current challenges related to understanding
and quantifying the various effects (see Sections 7.5, 8.3.4 and 8.5.1).
Their aerosol–radiation interaction effect is about 65% of the total
effect while the albedo effect is approximately 20% of the aerosol–
radiation interaction effect. Based on two studies (Rypdal et al., 2009;
Bond et al., 2011), the GWP and GTP metrics were found to vary with
the region where BC is emitted by about ±30% . For larger regions
of emissions, Collins et al. (2013) calculated GWPs and GTPs for the
direct effect of BC and found somewhat lower variations among the
regions.
Several studies have focused on the effects of emissions of BC and
OC from different regions (Bauer et al., 2007; Koch et al., 2007; Naik
et al., 2007; Reddy and Boucher, 2007; Rypdal et al., 2009). However,
examination of results from these models (Fuglestvedt et al., 2010)
reveals that there is not a robust relationship between the region of
emission and the metric value — hence, regions that yield the highest
metric value in one study, do not, in general, do so in the other studies.
The metric values for OC are quite consistent across studies, but fewer
studies are available (see Table 8.A.6). A brief overview of metric
values for other components is given in the Supplementary Material
Section 8.SM.14.
8.7.2.2.4 Summary of status of metrics for near-term climate forcers
The metrics provide a format for comparing the magnitudes of the
various emissions as well as for comparing effects of emissions from
different regions. They can also be used for comparing results from
different studies. Much of the spread in results is due to differences in
experimental design and how the models treat physical and chemical
processes. Unlike most of the WMGHGs, many of the NTCFs are tightly
coupled to the hydrologic cycle and atmospheric chemistry, leading to
a much larger spread in results as these are highly complex processes
that are difficult to validate on the requisite small spatial and short
temporal scales. The confidence level is lower for many of the NTCF
compared to WMGHG and much lower where aerosol–cloud interac-
tions are important (see Section 8.5.1). There are particular difficulties
for NO
X
, because the net impact is a small residual of opposing effects
with quite different spatial distributions and temporal behaviour.
Although climate–carbon feedbacks for non-CO
2
emissions have not
been included in the NTCF metrics (other than CH
4
) presented here,
they can greatly increase those values (Collins et al., 2013) and likely
provide more realistic results.
8.7.2.3 Impact by Emitted Component
We now use the metrics evaluated here to estimate climate impacts
of various components (in a forward looking perspective). Figure 8.32
shows global anthropogenic emissions of some selected components
weighted by the GWP and GTP. The time horizons are chosen as exam-
ples and illustrate how the perceived impacts of components—relative
to the impact of the reference gas—vary strongly as function of impact
parameter (integrated RF in GWP or end-point temperature in GTP)
and with time horizon.
We may also calculate the temporal development of the temperature
responses to pulse or sustained emissions using the AGTP metric.
Figure 8.33 shows that for a one-year pulse the impacts of NTCF decay
quickly owing to their atmospheric adjustment times even if effects are
prolonged due to climate response time (in the case of constant emis-
sions the effects reach approximately constant levels since the emis-
sions are replenished each year, except for CO
2
, which has a fraction
8
Anthropogenic and Natural Radiative Forcing Chapter 8
719
remaining in the atmosphere on time scales of centuries). Figure 8.33
also shows how some components have strong short-lived effects of
both signs while CO
2
has a weaker initial effect but one that persists
to create a long-lived warming effect. Note that there are large uncer-
tainties related to the metric values (as discussed in Section 8.7.1.4);
especially for the NTCFs.
These examples show that the outcome of comparisons of effects of
emissions depends strongly on choice of time horizon and metric type.
Such end-user choices will have a strong influence on the calculat-
ed contributions from NTCFs versus WMGHGs or non-CO
2
versus CO
2
emissions. Thus, each specific analysis should use a design chosen in
light of the context and questions being asked.
8.7.2.4 Metrics and Impacts by Sector
While the emissions of WMGHGs vary strongly between sectors, the cli-
mate impacts of these gases are independent of sector. The latter is not
the case for chemically active and short-lived components, due to the
dependence of their impact on the emission location. Since most sectors
have multiple co-emissions, and for NTCFs some of these are warm-
ing while others are cooling, the net impact of a given sector requires
explicit calculations. Since AR4, there has been significant progress in
the understanding and quantification of climate impacts of NTCFs from
sectors such as transportation, power production and biomass burning
(Berntsen and Fuglestvedt, 2008; Skeie et al., 2009; Stevenson and Der-
went, 2009; Lee et al., 2010; Unger et al., 2010; Dahlmann et al., 2011).
Supplementary Material Table 8.SM.18 gives an overview of recent pub-
lished metric values for various components by sector.
The impact from sectors depends on choice of metric, time horizon,
pulse versus sustained emissions and forward versus backward looking
perspective (see Section 8.7.1 and Box 8.4). Unger et al. (2010) calcu-
lated RF for a set of components emitted from each sector. RF at chosen
points in time (20 and 100 years) for sustained emissions was used by
Unger et al. (2010) as the metric for comparison. This is comparable
10 yrs 20 yrs 100 yrs
10 yrs
20 yrs 100 yrs
20
0
-20
5
10
0
-5
CO
2
equivalent emissions (Pg CO
2
-eq)
CO
2
equivalent emissions (PgC-eq)
GWP GTP
NO
X
CO
BC
OC
SO
2
CO
2
CH
4
N
2
O
Figure 8.32 | Global anthropogenic emissions weighted by GWP and GTP for chosen time horizons (aerosol–cloud interactions are not included). Emission data for 2008 are
taken from the EDGAR database. For BC and OC emissions for 2005 are from Shindell et al. (2012a). The units are ‘CO
2
equivalents’ which reflects equivalence only in the impact
parameter of the chosen metric (integrated RF over the chosen time horizon for GWP; temperature change at the chosen point in time for GTP), given as Pg(CO
2
)
eq
(left axis) and
given as PgC
eq
(right axis). There are large uncertainties related to the metric values and consequentially also to the calculated CO
2
equivalents (see text).
Time Horizon (yr)
Temperature impact (10
-3
K)
20 40 60 80
-20
-30
-10
0
0
10
20
Figure 8.33 | Temperature response by component for total anthropogenic emissions
for a 1-year pulse. Emission data for 2008 are taken from the EDGAR database and for
BC and OC for 2005 from Shindell et al. (2012a). There are large uncertainties related
to the AGTP values and consequentially also to the calculated temperature responses
(see text).
to using integrated RF up to the chosen times for pulse emissions (as
in GWPs). Such studies are relevant for policymaking that focuses on
regulating the total activity of a sector or for understanding the con-
tribution from a sector to climate change. On the other hand, the fixed
mix of emissions makes it less general and relevant for emission sce-
narios. Alternatively, one may adopt a component-by-component view
which is relevant for policies directed towards specific components (or
sets of components, as controlling an individual pollutant in isolation
is usually not practical). But this view will not capture interactions and
non-linearities within the suite of components emitted by most sectors.
The effects of specific emission control technologies or policies or pro-
jected societal changes on the mix of emissions is probably the most
relevant type of analysis, but there are an enormous number of possi-
ble actions and regional details that could be investigated. Henze et al.
(2012) demonstrate a method for providing highly spatially resolved
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
720
estimates of forcing per component, and caution that RF aggregated
over regions or sectors may not represent the impacts of emissions
changes on finer scales.
Metrics for individual land-based sectors are often similar to the global
mean metric values (Shindell et al., 2008). In contrast, metrics for emis-
sions from aviation and shipping usually show large differences from
global mean metric values (Table 8.A.3 versus Table 8.SM.18). Though
there can sometimes be substantial variation in the impact of land-
based sectors across regions, and for a particular region even from one
sector to another, variability between different land-based sources is
generally smaller than between land, sea and air emissions.
NO
x
from aviation is one example where the metric type is especial-
ly important. GWP
20
values are positive due to the strong response
of short-lived ozone. Reported GWP
100
and GTP
100
values are of either
sign, however, due to the differences in balance between the individ-
ual effects modelled. Even if the models agree on the net effect of
NO
X
, the individual contributions can differ significantly, with large
uncertainties stemming from the relative magnitudes of the CH
4
and
O
3
responses (Myhre et al., 2011) and the background tropospheric
concentrations of NO
X
(Holmes et al., 2011; Stevenson and Derwent,
2009). Köhler et al. (2013), find strong regional sensitivity of ozone
and CH
4
to NO
X
particularly at cruise altitude. Generally, they find the
strongest effects at low latitudes. For the aviation sector contrails and
contrail induced cirrus are also important. Based on detailed studies
in the literature, Fuglestvedt et al. (2010) produced GWP and GTP for
contrails, water vapor and contrail-induced cirrus.
The GWP and GTPs for NO
X
from shipping are strongly negative for
all time horizons. The strong positive effect via O
3
due to the low-NO
X
environment into which ships generally emit NO
X
is outweighed by the
stronger effect on CH
4
destruction due to the relatively lower latitudes
of these emissions compared to land-based sources.
In addition to having large emissions of NO
X
the shipping sector has
large emission of SO
2
. The direct GWP
100
for shipping ranges from –11
to –43 (see Supplementary Material Table 8.SM.18). Lauer et al. (2007)
reported detailed calculations of the indirect forcing specifically for this
sector and found a wide spread of values depending on the emission
inventory. Righi et al. (2011) and Peters et al. (2012) calculate indirect
effects that are 30 to 50% lower than the indirect forcing reported by
Lauer et al. (2007). The values from Shindell and Faluvegi (2010) for
SO
2
from power generation are similar to those for shipping.
Although the various land transport sectors often are treated as one
aggregate (e.g., road transport) there are important subdivisions. For
instance, Bond et al. (2013) points out that among the BC-rich sec-
tors they examined, diesel vehicles have the most clearly positive net
impact on forcing. Studies delving even further have shown substantial
differences between trucks and cars, gasoline and diesel vehicles, and
low-sulphur versus high-sulphur fuels. Similarly, for power production
there are important differences depending on fuel type (coal, oil, gas;
e.g., Shindell and Faluvegi, 2010).
In the assessment of climate impacts of current emissions by sectors
we give examples and apply a forward-looking perspective on effects
in terms of temperature change. The AGTP concept can be used to
study the effects of the various components for chosen time horizons.
A single year’s worth of current global emissions from the energy and
industrial sectors have the largest contributions to warming after 100
years (see Figure 8.34a). Household fossil fuel and biofuel, biomass
burning and on-road transportation are also relatively large contribu-
tors to warming over 100-year time scales. Those same sectors, along
with sectors that emit large amounts of CH
4
(animal husbandry, waste/
landfills and agriculture), are most important over shorter time hori-
zons (about 20 years; see Figure 8.34b).
Analysing climate change impacts by using the net effect of particular
activities or sectors may—compared to other perspectives—provide
more insight into how societal actions influence climate. Owing to
large variations in mix of short- and long-lived components, as well
as cooling and warming effects, the results will also in these cases
depend strongly on choice of time horizon and climate impact param-
eter. Improved understanding of aerosol–cloud interactions, and how
those are attributed to individual components is clearly necessary to
refine estimates of sectoral or emitted component impacts.
( )
( )
Figure 8.34 | Net global mean temperature change by source sector after (a) 100
and (b) 20 years (for 1-year pulse emissions). Emission data for 2008 are taken from
the EDGAR database. For BC and OC anthropogenic emissions are from Shindell et al.
(2012a) and biomass burning emissions are from Lamarque et al. (2010), see Supple-
mentary Material Section 8.SM.17. There are large uncertainties related to the AGTP
values and consequentially also to the calculated temperature responses (see text).
(a)
(b)
8
Anthropogenic and Natural Radiative Forcing Chapter 8
721
References
Aaheim, A., J. Fuglestvedt, and O. Godal, 2006: Costs savings of a flexible multi-gas
climate policy. Energy J. (Special Issue No. 3), 485–501.
Abreu, J., J. Beer, F. Steinhilber, S. Tobias, and N. Weiss, 2008: For how long will the
current grand maximum of solar activity persist? Geophys. Res. Lett., 35, L20109.
Ackerley, D., B. B. B. Booth, S. H. E. Knight, E. J. Highwood, D. J. Frame, M. R. Allen,
and D. P. Rowell, 2011: Sensitivity of twentieth-century Sahel rainfall to sulfate
aerosol and CO
2
forcing. J. Clim., 24, 4999–5014.
Allan, W., H. Struthers, and D. C. Lowe, 2007: Methane carbon isotope effects
caused by atomic chlorine in the marine boundary layer: Global model results
compared with Southern Hemisphere measurements. J. Geophys. Res. Atmos.,
112, D04306.
Ammann, C. M., and P. Naveau, 2003: Statistical analysis of tropical explosive
volcanism occurrences over the last 6 centuries. Geophys. Res. Lett., 30, 1210.
Ammann, C. M., and P. Naveau, 2010: A statistical volcanic forcing scenario
generator for climate simulations. J. Geophys. Res. Atmos., 115, D05107.
Anchukaitis, K. J., B. M. Buckley, E. R. Cook, B. I. Cook, R. D. D’Arrigo, and C. M.
Ammann, 2010: Influence of volcanic eruptions on the climate of the Asian
monsoon region. Geophys. Res. Lett., 37, L22703.
Andersen, M., D. Blake, F. Rowland, M. Hurley, and T. Wallington, 2009: Atmospheric
chemistry of sulfuryl fluoride: Reaction with OH radicals, CI atoms and O
3
,
atmospheric lifetime, IR spectrum, and global warming potential. Environ. Sci.
Technol., 43, 1067–1070.
Andersen, M., V. Andersen, O. Nielsen, S. Sander, and T. Wallington, 2010: Atmospheric
chemistry of HCF
2
O(CF
2
CF
2
O)(x)CF
2
H (x=2–4): Kinetics and mechanisms of the
chlorine-atom-initiated oxidation. Chemphyschem, 11, 4035–4041.
Andrews, T., and P. M. Forster, 2008: CO
2
forcing induces semi-direct effects with
consequences for climate feedback interpretations. Geophys. Res. Lett., 35,
L04802.
Andrews, T., M. Doutriaux-Boucher, O. Boucher, and P. M. Forster, 2011: A regional
and global analysis of carbon dioxide physiological forcing and its impact on
climate. Clim. Dyn., 36, 783–792.
Andrews, T., J. Gregory, M. Webb, and K. Taylor, 2012a: Forcing, feedbacks and climate
sensitivity in CMIP5 coupled atmosphere-ocean climate models. Geophys. Res.
Lett., 39, L09712.
Andrews, T., P. Forster, O. Boucher, N. Bellouin, and A. Jones, 2010: Precipitation,
radiative forcing and global temperature change. Geophys. Res. Lett., 37,
doi:10.1029/2010GL043991, L14701.
Andrews, T., M. Ringer, M. Doutriaux-Boucher, M. Webb, and W. Collins, 2012b:
Sensitivity of an Earth system climate model to idealized radiative forcing.
Geophys. Res. Lett., 39, L10702.
Antuña, J. C., A. Robock, G. Stenchikov, J. Zhou, C. David, J. Barnes, and L. Thomason,
2003: Spatial and temporal variability of the stratospheric aerosol cloud
produced by the 1991 Mount Pinatubo eruption. J. Geophys. Res. Atmos., 108,
4624.
Archibald, A. T., M. E. Jenkin, and D. E. Shallcross, 2010: An isoprene mechanism
intercomparison. Atmos. Environ., 44, 5356–5364.
Archibald, A. T., et al., 2011: Impacts of HO(x) regeneration and recycling in the
oxidation of isoprene: Consequences for the composition of past, present and
future atmospheres. Geophys. Res. Lett., 38, L05804.
Arnold, T., et al., 2013: Nitrogen trifluoride global emissions estimated from updated
atmospheric measurements, Proc. Natl. Acad. Sci. U.S.A, 110, 2029-2034.
Arora, V. K., and A. Montenegro, 2011: Small temperature benefits provided by
realistic afforestation efforts. Nature Geosci., 4, 514–518.
Arora, V. K., et al., 2013: Carbon-concentration and carbon-climate feedbacks in
CMIP5 Earth system models. J. Clim., 26, 5289-5314.
Ashmore, M. R., 2005: Assessing the future global impacts of ozone on vegetation.
Plant Cell Environ., 28, 949–964.
Azar, C., and D. J. A. Johansson, 2012: On the relationship between metrics to
compare greenhouse gases—the case of IGTP, GWP and SGTP. Earth Syst.
Dynam., 3, 139–147.
Baasandorj, M., A. R. Ravishankara, and J. B. Burkholder, 2011: Atmospheric
chemistry of (Z)-CF
3
CH=CHCF
3
: OH radical reaction rate coefficient and global
warming potential. J. Phys. Chem. A, 115, 10539–10549.
Baasandorj, M., G. Knight, V. Papadimitriou, R. Talukdar, A. Ravishankara, and J.
Burkholder, 2010: Rate coefficients for the gas-phase reaction of the hydroxyl
radical with CH
2
= CHF and CH
2
= CF
2
. J. Phys. Chem. A, 114, 4619–4633.
Bala, G., K. Caldeira, M. Wickett, T. J. Phillips, D. B. Lobell, C. Delire, and A. Mirin, 2007:
Combined climate and carbon-cycle effects of large-scale deforestation. Proc.
Natl. Acad. Sci. U.S.A., 104, 6550–6555.
Baliunas, S., and R. Jastrow, 1990: Evidence for long-term brightness changes of
solar-type stars. Nature, 348, 520–523.
Ball, W., Y. Unruh, N. Krivova, S. Solanki, T. Wenzler, D. Mortlock, and A. Jaffe, 2012:
Reconstruction of total solar irradiance 1974–2009. Astron. Astrophys., 541,
A27.
Ban-Weiss, G., L. Cao, G. Bala, and K. Caldeira, 2012: Dependence of climate forcing
and response on the altitude of black carbon aerosols. Clim. Dyn., 38, 897–911.
Barnes, C. A., and D. P. Roy, 2008: Radiative forcing over the conterminous United
States due to contemporary land cover land use albedo change. Geophys. Res.
Lett., 35, L09706.
Bathiany, S., M. Claussen, V. Brovkin, T. Raddatz, and V. Gayler, 2010: Combined
biogeophysical and biogeochemical effects of large-scale forest cover changes
in the MPI earth system model. Biogeosciences, 7, 1383–1399.
Bauer, S., D. Koch, N. Unger, S. Metzger, D. Shindell, and D. Streets, 2007: Nitrate
aerosols today and in 2030: A global simulation including aerosols and
tropospheric ozone. Atmos. Chem. Phys., 7, 5043–5059.
Bekki, S., J. A. Pyle, W. Zhong, R. Toumi, J. D. Haigh, and D. M. Pyle, 1996: The role of
microphysical and chemical processes in prolonging the climate forcing of the
Toba eruption. Geophys. Res. Lett., 23, 2669–2672.
Bellouin, N., J. Rae, A. Jones, C. Johnson, J. Haywood, and O. Boucher, 2011: Aerosol
forcing in the Climate Model Intercomparison Project (CMIP5) simulations by
HadGEM2–ES and the role of ammonium nitrate. J. Geophys. Res. Atmos., 116,
D20206.
Bernier, P. Y., R. L. Desjardins, Y. Karimi-Zindashty, D. Worth, A. Beaudoin, Y. Luo, and S.
Wang, 2011: Boreal lichen woodlands: A possible negative feedback to climate
change in eastern North America. Agr. Forest Meteorol., 151, 521–528.
Berntsen, T., and J. Fuglestvedt, 2008: Global temperature responses to current
emissions from the transport sectors. Proc. Natl. Acad. Sci. U.S.A., 105, 19154–
19159.
Berntsen, T. K., et al., 1997: Effects of anthropogenic emissions on tropospheric
ozone and its radiative forcing. J. Geophys. Res. Atmos., 102, 28101–28126.
Betts, R., 2000: Offset of the potential carbon sink from boreal forestation by
decreases in surface albedo. Nature, 408, 187–190.
Betts, R. A., P. D. Falloon, K. K. Goldewijk, and N. Ramankutty, 2007: Biogeophysical
effects of land use on climate: Model simulations of radiative forcing and large-
scale temperature change. Agr. Forest Meteorol., 142, 216–233.
Biasutti, M., and A. Giannini, 2006: Robust Sahel drying in response to late 20th
century forcings. Geophys. Res. Lett., 33, L11706.
Blowers, P., K. F. Tetrault, and Y. Trujillo-Morehead, 2008: Global warming potential
predictions for hydrofluoroethers with two carbon atoms. Theor. Chem. Acc.,
119, 369–381.
Blowers, P., D. Moline, K. Tetrault, R. Wheeler, and S. Tuchawena, 2007: Prediction of
radiative forcing values for hydrofluoroethers using density functional theory
methods. J. Geophys. Res. Atmos., 112, D15108.
Boer, G. J., and B. Yu, 2003: Climate sensitivity and response. Clim. Dyn., 20, 415–429.
Bollasina, M. A., Y. Ming, and V. Ramaswamy, 2011: Anthropogenic aerosols and the
weakening of the South Asian summer monsoon. Science, 334, 502–505.
Bond, T., C. Zarzycki, M. Flanner, and D. Koch, 2011: Quantifying immediate radiative
forcing by black carbon and organic matter with the Specific Forcing Pulse.
Atmos. Chem. Phys., 11, 1505–1525.
Bond, T. C., et al., 2007: Historical emissions of black and organic carbon aerosol
from energy-related combustion, 1850–2000. Global Biogeochem. Cycles, 21,
Gb2018.
Bond, T. C., et al., 2013: Bounding the role of black carbon in the climate system:
A scientific assessment. J. Geophys. Res. Atmos., 118, doi:10.1002/jgrd.50171,
5380-5552.
Bonfils, C., and D. Lobell, 2007: Empirical evidence for a recent slowdown in
irrigation-induced cooling. Proc. Natl. Acad. Sci. U.S.A., 104, 13582–13587.
Bonfils, C. J. W., T. J. Phillips, D. M. Lawrence, P. Cameron-Smith, W. J. Riley, and Z. M.
Subin, 2012: On the influence of shrub height and expansion on northern high
latitude climate. Environ. Res. Lett., 7, 015503.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
722
Booth, B., N. Dunstone, P. Halloran, T. Andrews, and N. Bellouin, 2012: Aerosols
implicated as a prime driver of twentieth-century North Atlantic climate
variability. Nature, 485, 534–534.
Boucher, O., 2012: Comparison of physically- and economically-based CO
2
–
equivalences for methane. Earth Syst. Dyn., 3, 49–61.
Boucher, O., and J. Haywood, 2001: On summing the components of radiative forcing
of climate change. Clim. Dyn., 18, 297–302.
Boucher, O., and M. Reddy, 2008: Climate trade-off between black carbon and
carbon dioxide emissions. Energy Policy, 36, 193–200.
Boucher, O., P. Friedlingstein, B. Collins, and K. P. Shine, 2009: The indirect global
warming potential and global temperature change potential due to methane
oxidation. Environ. Res. Lett., 4, 044007.
Bourassa, A. E., et al., 2012: Large volcanic aerosol load in the stratosphere linked to
Asian monsoon transport. Science, 337, 78–81.
Bourassa, A. E., et al., 2013: Response to Comments on “Large Volcanic Aerosol Load
in the Stratosphere Linked to Asian Monsoon Transport”. Science, 339, 6120.
Bowman, D., et al., 2009: Fire in the Earth System. Science, 324, 481–484.
Bowman, K. W., et al., 2013: Evaluation of ACCMIP outgoing longwave radiation
from tropospheric ozone using TES satellite observations. Atmos. Chem. Phys.,
13, 4057–4072.
Bradford, D., 2001: Global change – Time, money and tradeoffs. Nature, 410, 649–
650.
Bravo, I., et al., 2010: Infrared absorption spectra, radiative efficiencies, and global
warming potentials of perfluorocarbons: Comparison between experiment and
theory. J. Geophys. Res. Atmos., 115, D24317.
Brovkin, V., et al., 2010: Sensitivity of a coupled climate-carbon cycle model to large
volcanic eruptions during the last millennium. Tellus B, 62, 674–681.
Calvin, K., et al., 2012: The role of Asia in mitigating climate change: Results from the
Asia modeling exercise. Energ. Econ., 34, S251–S260.
Campra, P., M. Garcia, Y. Canton, and A. Palacios-Orueta, 2008: Surface temperature
cooling trends and negative radiative forcing due to land use change toward
greenhouse farming in southeastern Spain. J. Geophys. Res. Atmos., 113,
D18109.
Carlton, A. G., R. W. Pinder, P. V. Bhave, and G. A. Pouliot, 2010: To what extent can
biogenic SOA be controlled? Environ. Sci. Technol., 44, 3376–3380.
Carslaw, K. S., O. Boucher, D. V. Spracklen, G. W. Mann, J. G. L. Rae, S. Woodward, and
M. Kulmala, 2010: A review of natural aerosol interactions and feedbacks within
the Earth system. Atmos. Chem. Phys., 10, 1701–1737.
Chang, W. Y., H. Liao, and H. J. Wang, 2009: Climate responses to direct radiative
forcing of anthropogenic aerosols, tropospheric ozone, and long-lived
greenhouse gases in Eastern China over 1951–2000. Adv. Atmos. Sci., 26, 748–
762.
Chen, W. T., A. Nenes, H. Liao, P. J. Adams, J. L. F. Li, and J. H. Seinfeld, 2010: Global
climate response to anthropogenic aerosol indirect effects: Present day and year
2100. J. Geophys. Res. Atmos., 115, D12207.
Cherubini, F., G. Guest, and A. Strømman, 2012: Application of probablity distributions
to the modelleing of biogenic CO
2
fluxes in life cycle assessment. Global Change
Biol. , 4, doi:10.1111/j.1757–1707.2011.01156.x, 784-798.
Cherubini, F., G. Peters, T. Berntsen, A. Stromman, and E. Hertwich, 2011: CO
2
emissions from biomass combustion for bioenergy: Atmospheric decay and
contribution to global warming. Global Change Biol. Bioenerg., 3, 413–426.
Chung, C. E., and V. Ramanathan, 2006: Weakening of North Indian SST gradients
and the monsoon rainfall in India and the Sahel. J. Clim., 19, 2036–2045.
Clark, H. L., M. L. Cathala, H. Teyssedre, J. P. Cammas, and V. H. Peuch, 2007: Cross-
tropopause fluxes of ozone using assimilation of MOZAIC observations in a
global CTM. Tellus B, 59, 39–49.
Clarke, A. D., and K. J. Noone, 1985: Soot in the Arctic Snowpack—A cause for
peturbations in radiative-transfer. Atmos. Environ., 19, 2045–2053.
Claussen, M., V. Brovkin, and A. Ganopolski, 2001: Biogeophysical versus
biogeochemical feedbacks of large-scale land cover change. Geophys. Res. Lett.,
28, 1011–1014.
Cofala, J., M. Amann, Z. Klimont, K. Kupiainen, and L. Hoglund-Isaksson, 2007:
Scenarios of global anthropogenic emissions of air pollutants and methane until
2030. Atmos. Environ., 41, 8486–8499.
Collins, W., R. Derwent, C. Johnson, and D. Stevenson, 2002: The oxidation of organic
compounds in the troposphere and their global warming potentials. Clim.
Change, 52, 453–479.
Collins, W. D., et al., 2006: Radiative forcing by well-mixed greenhouse gases:
Estimates from climate models in the Intergovernmental Panel on Climate
Change (IPCC) Fourth Assessment Report (AR4). J. Geophys. Res. Atmos., 111,
D14317.
Collins, W. J., S. Sitch, and O. Boucher, 2010: How vegetation impacts affect climate
metrics for ozone precursors. J. Geophys. Res. Atmos., 115, D23308.
Collins, W. J., M. M. Fry, H. Yu, J. S. Fuglestvedt, D. T. Shindell, and J. J. West, 2013:
Global and regional temperature-change potentials for near-term climate
forcers. Atmos. Chem. Phys., 13, 2471–2485.
Conley, A. J., J. F. Lamarque, F. Vitt, W. D. Collins, and J. Kiehl, 2013: PORT, a CESM tool
for the diagnosis of radiative forcing. Geosci. Model Dev., 6, 469–476.
Cooper, O. R., et al., 2010: Increasing springtime ozone mixing ratios in the free
troposphere over western North America. Nature, 463, 344–348.
Cox, P. M., et al., 2008: Increasing risk of Amazonian drought due to decreasing
aerosol pollution. Nature, 453, 212–215.
Crook, J., and P. Forster, 2011: A balance between radiative forcing and climate
feedback in the modeled 20th century temperature response. J. Geophys. Res.
Atmos., 116, D17108.
Crowley, T. J., and M. B. Unterman, 2013: Technical details concerning development
of a 1200 yr proxy index for global volcanism. Earth Syst. Sci. Data, 5, 187-197.
Crutzen, P., 1973: Discussion of chemistry of some minor constituents in stratosphere
and troposhere. Pure Appl. Geophys., 106, 1385–1399.
Dahlmann, K., V. Grewe, M. Ponater, and S. Matthes, 2011: Quantifying the
contributions of individual NO
x
sources to the trend in ozone radiative forcing.
Atmos. Environ., 45, 2860–2868.
Daniel, J., and S. Solomon, 1998: On the climate forcing of carbon monoxide. J.
Geophys. Res. Atmos., 103, 13249–13260.
Daniel, J., S. Solomon, and D. Abritton, 1995: On the evaluation of halocarbon
radiative forcing and global warming potentials. J. Geophys. Res. Atmos., 100,
1271–1285.
Daniel, J., E. Fleming, R. Portmann, G. Velders, C. Jackman, and A. Ravishankara,
2010: Options to accelerate ozone recovery: Ozone and climate benefits. Atmos.
Chem. Phys., 10, 7697–7707.
Daniel, J., S. Solomon, T. Sanford, M. McFarland, J. Fuglestvedt, and P. Friedlingstein,
2012: Limitations of single-basket trading: Lessons from the Montreal Protocol
for climate policy. Clim. Change, 111, 241–248.
Davin, E., N. de Noblet-Ducoudre, and P. Friedlingstein, 2007: Impact of land cover
change on surface climate: Relevance of the radiative forcing concept. Geophys.
Res. Lett., 34, L13702.
Davin, E. L., and N. de Noblet-Ducoudre, 2010: Climatic impact of global-scale
deforestation: Radiative versus nonradiative orocesses. J. Clim., 23, 97–112.
De Cara, S., E. Galko, and P. Jayet, 2008: The global warming potential paradox:
Implications for the design of climate policy. In: Design of Climate Policy [R.
Guesnerie and H. Tulkens (eds.)]. The MIT Press, Cambridge, MA, USA, pp. 359–
384.
de la Torre, L., et al., 2006: Solar influence on Northern Annular Mode spatial
structure and QBO modulation. Part. Accel. Space Plasma Phys. Sol. Radiat.
Earth. Atmos. Clim., 37, 1635–1639.
de Noblet-Ducoudre, N., et al., 2012: Determining robust impacts of land-use-
induced land cover changes on surface climate over North America and Eurasia:
Results from the first set of LUCID experiments. J. Clim., 25, 3261–3281.
DeAngelis, A., F. Dominguez, Y. Fan, A. Robock, M. D. Kustu, and D. Robinson, 2010:
Evidence of enhanced precipitation due to irrigation over the Great Plains of the
United States. J. Geophys. Res. Atmos., 115, D15115.
DeLand, M., and R. Cebula, 2012: Solar UV variations during the decline of Cycle 23.
J. Atmos. Sol. Terres. Phys., 77, 225–234.
Delaygue, G., and E. Bard, 2011: An Antarctic view of Beryllium-10 and solar activity
for the past millennium. Clim. Dyn., 36, 2201–2218.
Deligne, N. I., S. G. Coles, and R. S. J. Sparks, 2010: Recurrence rates of large explosive
volcanic eruptions. J. Geophys. Res. Sol. Earth, 115, B06203.
den Elzen, M., et al., 2005: Analysing countries’ contribution to climate change:
Scientific and policy-related choices. Environ. Sci. Policy, 8, 614–636.
Denman, K. L., et al., 2007: Couplings between changes in the climate system
and biogeochemistry. In: Climate Change 2007: The Physical Science Basis.
Contribution of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning,
Z. Chen, M. Marquis, K. B. Averyt, M. Tignor and H. L. Miller (eds.)] Cambridge
University Press, Cambridge, United Kingdom and New York, NY, USA, 499-587.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
723
Dentener, F., et al., 2006: The global atmospheric environment for the next
generation. Environ. Sci. Technol., 40, 3586–3594.
Derwent, R., W. Collins, C. Johnson, and D. Stevenson, 2001: Transient behaviour of
tropospheric ozone precursors in a global 3-D CTM and their indirect greenhouse
effects. Clim. Change, 49, 463–487.
Deuber, O., G. Luderer, and O. Edenhofer, 2013: Physico-economic evaluation of
climate metrics: A conceptual framework. Environ. Sci. Policy, 29, 37–45.
Dewitte, S., D. Crommelynck, S. Mekaoui, and A. Joukoff, 2004: Measurement and
uncertainty of the long-term total solar irradiance trend. Solar Phys., 224, 209–
216.
Dickinson, R., 1975: Solar variability and lower atmosphere. Bull. Am. Meteorol. Soc.,
56, 1240–1248.
Doherty, S. J., S. G. Warren, T. C. Grenfell, A. D. Clarke, and R. E. Brandt, 2010: Light-
absorbing impurities in Arctic snow. Atmos. Chem. Phys., 10, 11647–11680.
Doutriaux-Boucher, M., M. Webb, J. Gregory, and O. Boucher, 2009: Carbon dioxide
induced stomatal closure increases radiative forcing via a rapid reduction in low
cloud. Geophys. Res. Lett., 36, doi:10.1029/2008GL036273, L02703.
Ehhalt, D. H., and L. E. Heidt, 1973: Vertical profiles of CH
4
in troposphere and
stratosphere. J. Geophys. Res., 78, 5265–5271.
Eliseev, A. V., and I.I. Mokhov, 2011: Effect of including land-use driven radiative
forcing of the surface albedo of land on climate response in the 16th-21st
centuries. Izvestiya Atmos. Ocean. Phys., 47, 15–30.
Engel, A., et al., 2009: Age of stratospheric air unchanged within uncertainties over
the past 30 years. Nature Geosci., 2, 28–31.
Erlykin, A., and A. Wolfendale, 2011: Cosmic ray effects on cloud cover and their
relevance to climate change. J. Atmos. Sol. Terres. Phys., 73, 1681–1686.
Ermolli, I., K. Matthes, T. Dudok de Wit, N. A. Krivova, K. Tourpali, M. Weber, Y. C.
Unruh, L. Gray, U. Langematz, P. Pilewskie, E. Rozanov, W. Schmutz, A. Shapiro,
S. K. Solanki, and T. N. Woods, 2013: Recent variability of the solar spectral
irradiance and its impact on climate modelling, Atmospheric Chemistry and
Physics, 13, 3945-3977.
Esper, J., and F. H. Schweingruber, 2004: Large-scale treeline changes recorded in
Siberia. Geophys. Res. Lett., 31, L06202.
Eyring, V., et al., 2010a: Sensitivity of 21st century stratospheric ozone to greenhouse
gas scenarios. Geophys. Res. Lett., 37, L16807.
Eyring, V., et al., 2010b: Multi-model assessment of stratospheric ozone return dates
and ozone recovery in CCMVal-2 models. Atmos. Chem. Phys., 10, 9451–9472.
Fan, F. X., M. E. Mann, and C. M. Ammann, 2009: Understanding changes in the
Asian summer monsoon over the past millennium: Insights from a long-term
coupled model simulation. J. Clim., 22, 1736–1748.
FAO, 2012: State of the world’s forests. Food and Agriculture Organization of the
United Nations, Rome, Italy, 60 pp.
Feng, X., and F. Zhao, 2009: Effect of changes of the HITRAN database on
transmittance calculations in the near-infrared region. J. Quant. Spectrosc.
Radiat. Transfer, 110, 247–255.
Feng, X., F. Zhao, and W. Gao, 2007: Effect of the improvement of the HITIRAN
database on the radiative transfer calculation. J. Quant. Spectrosc. Radiat.
Transfer, 108, 308–318.
Findell, K. L., E. Shevliakova, P. C. D. Milly, and R. J. Stouffer, 2007: Modeled impact of
anthropogenic land cover change on climate. J. Clim., 20, 3621–3634.
Fioletov, V. E., G. E. Bodeker, A. J. Miller, R. D. McPeters, and R. Stolarski, 2002: Global
and zonal total ozone variations estimated from ground-based and satellite
measurements: 1964–2000. J. Geophys. Res. Atmos., 107, 4647.
Fiore, A. M., et al., 2009: Multimodel estimates of intercontinental source-receptor
relationships for ozone pollution. J. Geophys. Res. Atmos., 114, D04301.
Fischer, E. M., J. Luterbacher, E. Zorita, S. F. B. Tett, C. Casty, and H. Wanner, 2007:
European climate response to tropical volcanic eruptions over the last half
millennium. Geophys. Res. Lett., 34, L05707.
Fishman, J., et al., 2010: An investigation of widespread ozone damage to the
soybean crop in the upper Midwest determined from ground-based and satellite
measurements. Atmos. Environ., 44, 2248–2256.
Flanner, M. G., C. S. Zender, J. T. Randerson, and P. J. Rasch, 2007: Present-day climate
forcing and response from black carbon in snow. J. Geophys. Res. Atmos., 112,
D11202.
Fletcher, C. G., P. J. Kushner, A. Hall, and X. Qu, 2009: Circulation responses to snow
albedo feedback in climate change. Geophys. Res. Lett., 36, L09702.
Fomin, B. A., and V. A. Falaleeva, 2009: Recent progress in spectroscopy and its effect
on line-by-line calculations for the validation of radiation codes for climate
models. Atmos. Oceanic Opt., 22, 626–629.
Forster, P., and K. Shine, 1997: Radiative forcing and temperature trends from
stratospheric ozone changes. J. Geophys. Res. Atmos., 102, 10841–10855.
Forster, P., et al., 2005: Resolution of the uncertainties in the radiative forcing of
HFC-134a. J. Quant. Spectrosc. Radiat. Transfer, 93, 447–460.
Forster, P., et al., 2007: Changes in Atmospheric Constituents and in Radiative Forcing.
In: Climate Change 2007: The Physical Science Basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B.
Averyt, M. Tignor and H. L. Miller (eds.)] Cambridge University Press, Cambridge,
United Kingdom and New York, NY, USA, 129-234.
Forster, P., et al., 2011a: Evaluation of radiation scheme performance within
chemistry climate models. J. Geophys. Res. Atmos., 116, D10302.
Forster, P. M., T. Andrews, P. Good, J. M. Gregory, L. S. Jackson, and M. Zelinka, 2013:
Evaluating adjusted forcing and model spread for historical and future scenarios
in the CMIP5 generation of climate models. J. Geophys. Res. Atmos., 118, 1139–
1150.
Forster, P. M., et al., 2011b: Stratospheric changes and climate. In: Scientific
Assessment of Ozone Depletion: 2010. Global Ozone Research and
MonitoringProject–Report No. 52, World Meteorological Organization, Geneva,
Switzerland, 516 pp.
Fortuin, J. P. F., and H. Kelder, 1998: An ozone climatology based on ozonesonde and
satellite measurements. J. Geophys. Res. Atmos., 103, 31709–31734.
Foukal, P., and J. Lean, 1988: Magnetic modulation of solar luminosity by phtospheric
activity. Astrophys. J., 328, 347–357.
Fowler, D., et al., 2009: Atmospheric composition change: Ecosystems-atmosphere
interactions. Atmos. Environ., 43, 5193–5267.
Frame, T., and L. Gray, 2010: The 11-yr solar cycle in ERA-40 data: An update to 2008.
J. Clim., 23, 2213–2222.
Freckleton, R., E. Highwood, K. Shine, O. Wild, K. Law, and M. Sanderson, 1998:
Greenhouse gas radiative forcing: Effects of averaging and inhomogeneities in
trace gas distribution. Q. J. R. Meteorol. Soc., 124, 2099–2127.
Friedlingstein, P., et al., 2006: Climate-carbon cycle feedback analysis: Results from
the C(4)MIP model intercomparison. J. Clim., 19, 3337–3353.
Frohlich, C., 2006: Solar irradiance variability since 1978—Revision of the PMOD
composite during solar cycle 21. Space Sci. Rev., 125, 53–65.
Frohlich, C., 2009: Evidence of a long-term trend in total solar irradiance. Astron.
Astrophys., 501, L27–L30.
Frolicher, T. L., F. Joos, and C. C. Raible, 2011: Sensitivity of atmospheric CO
2
and
climate to explosive volcanic eruptions. Biogeosciences, 8, 2317–2339.
Fromm, M., G. Nedoluha, and Z. Charvat, 2013: Comment on “Large Volcanic
Aerosol Load in the Stratosphere Linked to Asian Monsoon Transport”. Science,
339, 647–c.
Fry, M., et al., 2012: The influence of ozone precursor emissions from four world
regions on tropospheric composition and radiative climate forcing. J. Geophys.
Res. Atmos., 117, D07306.
Fuglestvedt, J., T. Berntsen, O. Godal, and T. Skodvin, 2000: Climate implications of
GWP-based reductions in greenhouse gas emissions. Geophys. Res. Lett., 27,
409–412.
Fuglestvedt, J., T. Berntsen, O. Godal, R. Sausen, K. Shine, and T. Skodvin, 2003:
Metrics of climate change: Assessing radiative forcing and emission indices.
Clim. Change, 58, 267–331.
Fuglestvedt, J. S., et al., 2010: Transport impacts on atmosphere and climate: Metrics.
Atmos. Environ., 44, 4648–4677.
Fung, I., J. John, J. Lerner, E. Matthews, M. Prather, L. P. Steele, and P. J. Fraser, 1991:
3-Dimensional model synthesis of the global methane cycle. J. Geophys. Res.
Atmos., 96, 13033–13065.
Gaillard, M. J., et al., 2010: Holocene land-cover reconstructions for studies on land
cover-climate feedbacks. Clim. Past, 6, 483–499.
Gao, C. C., A. Robock, and C. Ammann, 2008: Volcanic forcing of climate over
the past 1500 years: An improved ice core-based index for climate models. J.
Geophys. Res. Atmos., 113, D23111.
Garcia, R. R., W. J. Randel, and D. E. Kinnison, 2011: On the determination of age of
air trends from atmospheric trace species. J. Atmos. Sci., 68, 139–154.
Gerlach, T., 2011: Volcanic versus anthropogenic carbon dioxide. Eos, 92, 201–202.
Gettelman, A., J. Holton, and K. Rosenlof, 1997: Mass fluxes of O
3
, CH
4
, N
2
O and
CF
2
Cl
2
in the lower stratosphere calculated from observational data. J. Geophys.
Res. Atmos., 102, 19149–19159.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
724
Gillett, N., and H. Matthews, 2010: Accounting for carbon cycle feedbacks in a
comparison of the global warming effects of greenhouse gases. Environ. Res.
Lett., 5, 034011.
Ginoux, P., D. Garbuzov, and N. C. Hsu, 2010: Identification of anthropogenic and
natural dust sources using Moderate Resolution Imaging Spectroradiometer
(MODIS) Deep Blue level 2 data. J. Geophys. Res. Atmos., 115, D05204.
Godal, O., and J. Fuglestvedt, 2002: Testing 100-year global warming potentials:
Impacts on compliance costs and abatement profile. Clim. Change, 52, 93–127.
Goosse, H., et al., 2006: The origin of the European “Medieval Warm Period”. Clim.
Past, 2, 99–113.
Granier, C., et al., 2011: Evolution of anthropogenic and biomass burning emissions
of air pollutants at global and regional scales during the 1980–2010 period.
Clim. Change, 109, 163–190.
Gray, L., S. Rumbold, and K. Shine, 2009: Stratospheric temperature and radiative
forcing response to 11-year solar cycle changes in irradiance and ozone. J.
Atmos. Sci., 66, 2402–2417.
Gray, L., et al., 2010: Solar influences on climate. Rev. Geophys., 48, RG4001.
Gregory, J., and M. Webb, 2008: Tropospheric adjustment induces a cloud component
in CO
2
forcing. J. Clim., 21, 58–71.
Gregory, J., et al., 2004: A new method for diagnosing radiative forcing and climate
sensitivity. Geophys. Res. Lett., 31, L03205.
Gregory, J. M., 2010: Long-term effect of volcanic forcing on ocean heat content.
Geophys. Res. Lett., 37, L22701.
Grewe, V., 2007: Impact of climate variability on tropospheric ozone. Sci. Tot.
Environ., 374, 167–181.
Gusev, A. A., 2008: Temporal structure of the global sequence of volcanic eruptions:
Order clustering and intermittent discharge rate. Phys. Earth Planet. Inter., 166,
203–218.
Haigh, J., 1994: The role of stratospheric ozone in modulating the solar radiative
forcing of climate. Nature, 370, 544–546.
Haigh, J., 1999: A GCM study of climate change in response to the 11-year solar
cycle. Q. J. R. Meteorol. Soc., 125, 871–892.
Haigh, J. D., 1996: The impact of solar variability on climate. Science, 272, 981–984.
Hall, J., and G. Lockwood, 2004: The chromospheric activity and variability of cycling
and flat activity solar-analog stars. Astrophys. J., 614, 942–946.
Hallquist, M., et al., 2009: The formation, properties and impact of secondary organic
aerosol: Current and emerging issues. Atmos. Chem. Phys., 9, 5155–5236.
Hammitt, J., A. Jain, J. Adams, and D. Wuebbles, 1996: A welfare-based index for
assessing environmental effects of greenhouse-gas emissions. Nature, 381,
301–303.
Hansen, J., and L. Nazarenko, 2004: Soot climate forcing via snow and ice albedos.
Proc. Natl. Acad. Sci. U.S.A., 101, 423–428.
Hansen, J., et al., 2005: Efficacy of climate forcings. J. Geophys. Res. Atmos., 110,
D18104.
Hansen, J., et al., 2007: Climate simulations for 1880–2003 with GISS modelE. Clim.
Dyn., 29, 661–696.
Harder, J., J. Fontenla, P. Pilewskie, E. Richard, and T. Woods, 2009: Trends in solar
spectral irradiance variability in the visible and infrared. Geophys. Res. Lett.,
36, L07801.
Harrison, R., and M. Ambaum, 2010: Observing Forbush decreases in cloud at
Shetland. J. Atmos. Sol. Terres. Phys., 72, 1408–1414.
Haywood, J., and M. Schulz, 2007: Causes of the reduction in uncertainty in the
anthropogenic radiative forcing of climate between IPCC (2001) and IPCC
(2007). Geophys. Res. Lett., 34, L20701.
Haywood, J. M., et al., 2010: Observations of the eruption of the Sarychev volcano
and simulations using the HadGEM2 climate model. J. Geophys. Res. Atmos.,
115, D21212.
Heathfield, A., C. Anastasi, A. McCulloch, and F. Nicolaisen, 1998: Integrated infrared
absorption coefficients of several partially fluorinated ether compounds:
CF
3
OCF
2
H, CF
2
HOCF
2
H, CH
3
OCF
2
CF
2
H, CH
3
OCF
2
CFClH, CH
3
CH
2
OCF
2
CF
2
H,
CF
3
CH
2
OCF
2
CF
2
H AND CH
2
=CHCH
2
OCF
2
CF
2
H. Atmos. Environ., 32, 2825–2833.
Hegg, D. A., S. G. Warren, T. C. Grenfell, S. J. Doherty, T. V. Larson, and A. D. Clarke,
2009: Source attribution of black carbon in Arctic snow. Environ. Sci. Technol.,
43, 4016–4021.
Hegglin, M. I., and T. G. Shepherd, 2009: Large climate-induced changes in ultraviolet
index and stratosphere-to-troposphere ozone flux. Nature Geosci., 2, 687–691.
Helama, S., M. Fauria, K. Mielikainen, M. Timonen, and M. Eronen, 2010: Sub-
Milankovitch solar forcing of past climates: Mid and late Holocene perspectives.
Geol. Soc. Am. Bull., 122, 1981–1988.
Henze, D. K., et al., 2012: Spatially refined aerosol direct radiative forcing efficiencies.
Environ. Sci. Technol., 46, 9511–9518.
Hohne, N., et al., 2011: Contributions of individual countries’ emissions to climate
change and their uncertainty. Clim. Change, 106, 359–391.
Holmes, C., Q. Tang, and M. Prather, 2011: Uncertainties in climate assessment for
the case of aviation NO. Proc. Natl. Acad. Sci. U.S.A., 108, 10997–11002.
Holmes, C. D., M. J. Prather, O. A. Sovde, and G. Myhre, 2013: Future methane,
hydroxyl, and their uncertainties: Key climate and emission parameters for future
predictions. Atmos. Chem. Phys., 13, 285–302.
Horowitz, L. W., 2006: Past, present, and future concentrations of tropospheric ozone
and aerosols: Methodology, ozone evaluation, and sensitivity to aerosol wet
removal. J. Geophys. Res. Atmos., 111, D22211.
Houghton, J. T., G. J. Jenkins, and J. J. Ephraums (eds.), 1990: Climate Change. The
IPCC Scientific Assessment. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA, 364 pp.
Hoyle, C., et al., 2011: A review of the anthropogenic influence on biogenic secondary
organic aerosol. Atmos. Chem. Phys., 11, 321–343.
Hsu, J., and M. J. Prather, 2009: Stratospheric variability and tropospheric ozone. J.
Geophys. Res. Atmos., 114, D06102.
Huang, J., Q. Fu, W. Zhang, X. Wang, R. Zhang, H. Ye, and S. Warren, 2011: Dust and
black carbon in seasonal snow across northern China. Bull. Am. Meteorol. Soc.,
92, 175–181.
Huijnen, V., et al., 2010: The global chemistry transport model TM5: Description and
evaluation of the tropospheric chemistry version 3.0. Geosci. Model Dev., 3,
445–473.
Hurst, D. F., et al., 2011: Stratospheric water vapor trends over Boulder, Colorado:
Analysis of the 30 year Boulder record. J. Geophys. Res. Atmos., 116, D02306.
Hurtt, G. C., et al., 2006: The underpinnings of land-use history: Three centuries
of global gridded land-use transitions, wood-harvest activity, and resulting
secondary lands. Global Change Biol., 12, 1208–1229.
Iacono, M. J., J. S. Delamere, E. J. Mlawer, M. W. Shephard, S. A. Clough, and W. D.
Collins, 2008: Radiative forcing by long-lived greenhouse gases: Calculations
with the AER radiative transfer models. J. Geophys. Res. Atmos., 113, D13103.
Ineson, S., A. A. Scaife, J. R. Knight, J. C. Manners, N. J. Dunstone, L. J. Gray, and
J. D. Haigh, 2011: Solar forcing of winter climate variability in the Northern
Hemisphere. Nature Geosci., 4, 753–757.
IPCC, 1996: Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories.
Intergovernmental Panel of Climate Change.
Isaksen, I., et al., 2009: Atmospheric composition change: Climate-chemistry
interactions. Atmos. Environ., 43, 5138–5192.
Ito, A., and J. E. Penner, 2005: Historical emissions of carbonaceous aerosols from
biomass and fossil fuel burning for the period 1870–2000. Global Biogeochem.
Cycles, 19, Gb2028.
Jackson, S., 2009: Parallel pursuit of near-term and long-term climate mitigation.
Science, 326, 526–527.
Jacobson, M., 2010: Short-term effects of controlling fossil-fuel soot, biofuel soot
and gases, and methane on climate, Arctic ice, and air pollution health. J.
Geophys. Res. Atmos., 115, D14209.
Jacobson, M., 2012: Investigating cloud absorption effects: Global absorption
properties of black carbon, tar balls, and soil dust in clouds and aerosols. J.
Geophys. Res. Atmos., 117, D06205.
Javadi, M., O. Nielsen, T. Wallington, M. Hurley, and J. Owens, 2007: Atmospheric
chemistry of 2-ethoxy-3,3,4,4,5–pentafluorotetra-hydro-2,5-bis[1,2,2,2-
tetrafluoro-1-(trifluoromethyl)ethyl]-furan: Kinetics, mechanisms, and products
of CL atom and OH radical initiated oxidation. Environ. Sci. Technol., 41, 7389–
7395.
Jin, M. L., R. E. Dickinson, and D. L. Zhang, 2005: The footprint of urban areas on
global climate as characterized by MODIS. J. Clim., 18, 1551–1565.
Jin, Y., and D. P. Roy, 2005: Fire-induced albedo change and its radiative forcing at
the surface in northern Australia. Geophys. Res. Lett., 32, L13401.
Jin, Y. F., J. T. Randerson, M. L. Goulden, and S. J. Goetz, 2012: Post-fire changes in
net shortwave radiation along a latitudinal gradient in boreal North America.
Geophys. Res. Lett., 39, L13403.
Johansson, D., 2012: Economics- and physical-based metrics for comparing
greenhouse gases. Clim. Change, 110, 123–141.
Johansson, D., U. Persson, and C. Azar, 2006: The cost of using global warming
potentials: Analysing the trade off between CO
2
, CH
4
and N
2
O. Clim. Change, 77,
doi:10.1007/s10584-006-9054-1, 291–309.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
725
Jones, G., M. Lockwood, and P. Stott, 2012: What influence will future solar
activity changes over the 21st century have on projected global near-surface
temperature changes? J. Geophys. Res. Atmos., 117, D05103.
Jones, G. S., J. M. Gregory, P. A. Stott, S. F. B. Tett, and R. B. Thorpe, 2005: An AOGCM
simulation of the climate response to a volcanic super-eruption. Clim. Dyn., 25,
725–738.
Joos, F., M. Bruno, R. Fink, U. Siegenthaler, T. Stocker, and C. LeQuere, 1996: An
efficient and accurate representation of complex oceanic and biospheric models
of anthropogenic carbon uptake. Tellus B, 48, 397–417.
Joos, F., et al., 2013: Carbon dioxide and climate impulse response functions for the
computation of greenhouse gas metrics: A multi-model analysis. Atmos. Chem.
Phys., 13, 2793–2825.
Joshi, M., and J. Gregory, 2008: Dependence of the land-sea contrast in surface
climate response on the nature of the forcing. Geophys. Res. Lett., 35, L24802.
Joshi, M. M., and G. S. Jones, 2009: The climatic effects of the direct injection of
water vapour into the stratosphere by large volcanic eruptions. Atmos. Chem.
Phys., 9, 6109–6118.
Kandlikar, M., 1995: The relative role of trace gas emissions in greenhouse abatement
policies. Energ. Policy, 23, 879–883.
Kaplan, J. O., K. M. Krumhardt, E. C. Ellis, W. F. Ruddiman, C. Lemmen, and K. K.
Goldewijk, 2011: Holocene carbon emissions as a result of anthropogenic land
cover change. Holocene, 21, 775–791.
Kasischke, E. S., and J. E. Penner, 2004: Improving global estimates of atmospheric
emissions from biomass burning. J. Geophys. Res. Atmos., 109, D14S01.
Kawase, H., T. Nagashima, K. Sudo, and T. Nozawa, 2011: Future changes in
tropospheric ozone under Representative Concentration Pathways (RCPs).
Geophys. Res. Lett., 38, L05801.
Kawase, H., M. Abe, Y. Yamada, T. Takemura, T. Yokohata, and T. Nozawa, 2010:
Physical mechanism of long-term drying trend over tropical North Africa.
Geophys. Res. Lett., 37, L09706.
Kirkby, J., 2007: Cosmic rays and climate. Surv. Geophys., 28, 333–375.
Kirkby, J., et al., 2011: Role of sulphuric acid, ammonia and galactic cosmic rays in
atmospheric aerosol nucleation. Nature, 476, 429–433.
Kleinman, L. I., P. H. Daum, Y. N. Lee, L. J. Nunnermacker, S. R. Springston, J. Weinstein-
Lloyd, and J. Rudolph, 2001: Sensitivity of ozone production rate to ozone
precursors. Geophys. Res. Lett., 28, 2903–2906.
Knutti, R., et al., 2008: A review of uncertainties in global temperature projections
over the twenty-first century. J. Clim., 21, 2651–2663.
Koch, D., and A. D. Del Genio, 2010: Black carbon semi-direct effects on cloud cover:
Review and synthesis. Atmos. Chem. Phys., 10, 7685–7696.
Koch, D., T. Bond, D. Streets, N. Unger, and G. van der Werf, 2007: Global impacts
of aerosols from particular source regions and sectors. J. Geophys. Res. Atmos.,
112, D02205.
Koch, D., S. Menon, A. Del Genio, R. Ruedy, I. Alienov, and G. A. Schmidt, 2009a:
Distinguishing aerosol impacts on climate over the past century. J. Clim., 22,
2659–2677.
Koch, D., et al., 2011: Coupled aerosol-chemistry-climate twentieth-century transient
model investigation: Trends in short-lived species and climate responses. J. Clim.,
24, 2693–2714.
Koch, D., et al., 2009b: Evaluation of black carbon estimations in global aerosol
models. Atmos. Chem. Phys., 9, 9001–9026.
Koehler, M. O., G. Raedel, K. P. Shine, H. L. Rogers, and J. A. Pyle, 2013: Latitudinal
variation of the effect of aviation NO
x
emissions on atmospheric ozone and
methane and related climate metrics. Atmos. Environ., 64, 1–9.
Koffi, B., et al., 2012: Application of the CALIOP layer product to evaluate the vertical
distribution of aerosols estimated by global models: AeroCom phase I results. J.
Geophys. Res. Atmos., 117, D10201.
Kopp, G., and J. Lean, 2011: A new, lower value of total solar irradiance: Evidence
and climate significance. Geophys. Res. Lett., 38, L01706.
Kratz, D., 2008: The sensitivity of radiative transfer calculations to the changes in
the HITRAN database from 1982 to 2004. J. Quant. Spectrosc. Radiat. Transfer,
109, 1060–1080.
Kravitz, B., and A. Robock, 2011: Climate effects of high-latitude volcanic eruptions:
Role of the time of year. J. Geophys. Res. Atmos., 116, D01105.
Kravitz, B., A. Robock, and A. Bourassa, 2010: Negligible climatic effects from the
2008 Okmok and Kasatochi volcanic eruptions. J. Geophys. Res. Atmos., 115,
D00L05.
Kravitz, B., et al., 2011: Simulation and observations of stratospheric aerosols from
the 2009 Sarychev volcanic eruption. J. Geophys. Res. Atmos., 116, D18211.
Kristjansson, J. E., T. Iversen, A. Kirkevag, O. Seland, and J. Debernard, 2005:
Response of the climate system to aerosol direct and indirect forcing: Role of
cloud feedbacks. J. Geophys. Res. Atmos., 110, D24206.
Krivova, N., L. Vieira, and S. Solanki, 2010: Reconstruction of solar spectral irradiance
since the Maunder minimum. J. Geophys. Res. Space Phys., 115, A12112.
Kueppers, L. M., M. A. Snyder, and L. C. Sloan, 2007: Irrigation cooling effect:
Regional climate forcing by land-use change. Geophys. Res. Lett., 34, L03703.
Kuroda, Y., and K. Kodera, 2005: Solar cycle modulation of the southern annular
mode. Geophys. Res. Lett., 32, L13802.
Kvalevag, M. M., G. Myhre, G. Bonan, and S. Levis, 2010: Anthropogenic land cover
changes in a GCM with surface albedo changes based on MODIS data. Int. J.
Climatol., 30, 2105–2117.
Lacis, A. A., D. J. Wuebbles, J. A. Logan, 1990: Radiative forcing of climate by changes
in the vertical-distribution of ozone, J. Geophys. Res., 95, 9971-9981.
Lamarque, J., et al., 2011: Global and regional evolution of short-lived radiatively-
active gases and aerosols in the Representative Concentration Pathways. Clim.
Change, 109, 191–212.
Lamarque, J., et al., 2010: Historical (1850–2000) gridded anthropogenic and
biomass burning emissions of reactive gases and aerosols: Methodology and
application. Atmos. Chem. Phys., 10, 7017–7039.
Lamarque, J. F., et al., 2013: The Atmospheric Chemistry and Climate Model
Intercomparison Project (ACCMIP): Overview and description of models,
simulations and climate diagnostics. Geosci. Model Dev., 6, 179–206.
Lau, K. M., M. K. Kim, and K. M. Kim, 2006: Asian summer monsoon anomalies
induced by aerosol direct forcing: The role of the Tibetan Plateau. Clim. Dyn.,
26, 855–864.
Lauder, A. R., I. G. Enting, J. O. Carter, N. Clisby, A. L. Cowie, B. K. Henry, and M.
R. Raupach, 2013: Offsetting methane emissions—An alternative to emission
equivalence metrics. Int. J. Greenh. Gas Control, 12, 419–429.
Lauer, A., V. Eyring, J. Hendricks, P. Jockel, and U. Lohmann, 2007: Global model
simulations of the impact of ocean-going ships on aerosols, clouds, and the
radiation budget. Atmos. Chem. Phys., 7, 5061–5079.
Lawrence, P. J., and T. N. Chase, 2010: Investigating the climate impacts of global
land cover change in the community climate system model. Int. J. Climatol., 30,
2066–2087.
Lean, J., 2000: Evolution of the sun’s spectral irradiance since the Maunder Minimum.
Geophys. Res. Lett., 27, 2425–2428.
Lean, J., and M. Deland, 2012: How does the sun’s spectrum vary? J. Clim., 25,
2555–2560.
Lee, D. S., et al., 2010: Transport impacts on atmosphere and climate: Aviation.
Atmos. Environ., 44, 4678–4734.
Lee, J., D. Shindell, and S. Hameed, 2009: The influence of solar forcing on tropical
circulation. J. Clim., 22, 5870–5885.
Lee, R., M. Gibson, R. Wilson, and S. Thomas, 1995: Long-term total solar irradiance
variability during sunspot cycle-22. J. Geophys. Res. Space Phys., 100, 1667–
1675.
Lee, X., et al., 2011: Observed increase in local cooling effect of deforestation at
higher latitudes. Nature, 479, 384–387.
Lee, Y. H., et al., 2013: Evaluation of preindustrial to present-day black carbon and its
albedo forcing from Atmospheric Chemistry and Climate Model Intercomparison
Project (ACCMIP). Atmos. Chem. Phys., 13, 2607–2634.
Legras, B., O. Mestre, E. Bard, and P. Yiou, 2010: A critical look at solar-climate
relationships from long temperature series. Clim. Past, 6, 745–758.
Leibensperger, E. M., L. J. Mickley, and D. J. Jacob, 2008: Sensitivity of US air quality
to mid-latitude cyclone frequency and implications of 1980–2006 climate
change. Atmos. Chem. Phys., 8, 7075–7086.
Leibensperger, E. M., et al., 2012a: Climatic effects of 1950–2050 changes in US
anthropogenic aerosols—Part 1: Aerosol trends and radiative forcing. Atmos.
Chem. Phys., 12, 3333–3348.
Leibensperger, E. M., et al., 2012b: Climatic effects of 1950–2050 changes in US
anthropogenic aerosols—Part 2: Climate response. Atmos. Chem. Phys., 12,
3349–3362.
Lelieveld, J., et al., 2008: Atmospheric oxidation capacity sustained by a tropical
forest. Nature, 452, 737–740.
Levy, H., 1971: Normal atmosphere—Large radical and formaldehyde concentrations
predicted. Science, 173, 141–143.
Liao, H., W. T. Chen, and J. H. Seinfeld, 2006: Role of climate change in global
predictions of future tropospheric ozone and aerosols. J. Geophys. Res. Atmos.,
111, D12304.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
726
Lockwood, M., 2010: Solar change and climate: An update in the light of the current
exceptional solar minimum. Proc. R. Soc. London A, 466, 303–329.
Lockwood, M., and C. Frohlich, 2008: Recent oppositely directed trends in solar
climate forcings and the global mean surface air temperature. II. Different
reconstructions of the total solar irradiance variation and dependence on
response time scale. Proc. R. Soc. London A, 464, 1367–1385.
Lockwood, M., A. Rouillard, and I. Finch, 2009: The rise and fall of open solar flux
during the current grand solar maximum. Astrophys. J., 700, 937–944.
Logan, J. A., 1999: An analysis of ozonesonde data for the troposphere:
Recommendations for testing 3-D models and development of a gridded
climatology for tropospheric ozone. J. Geophys. Res. Atmos., 104, 16115–16149.
Logan, J. A., M. J. Prather, S. C. Wofsy, and M. B. McElroy, 1981: Tropospheric
chemistry—A global perspective. J. Geophys. Res. Oceans Atmos., 86, 7210–
7254.
Logan, J. A., et al., 2012: Changes in ozone over Europe: Analysis of ozone
measurements from sondes, regular aircraft (MOZAIC) and alpine surface sites.
J. Geophys. Res. Atmos., 117, D09301.
Lohila, A., et al., 2010: Forestation of boreal peatlands: Impacts of changing albedo
and greenhouse gas fluxes on radiative forcing. J. Geophys. Res. Biogeosci., 115,
G04011.
Lohmann, U., et al., 2010: Total aerosol effect: Radiative forcing or radiative flux
perturbation? Atmos. Chem. Phys., 10, 3235–3246.
Long, C. N., E. G. Dutton, J. A. Augustine, W. Wiscombe, M. Wild, S. A. McFarlane, and
C. J. Flynn, 2009: Significant decadal brightening of downwelling shortwave in
the continental United States. J. Geophys. Res. Atmos., 114, D00D06.
Long, D., and M. Collins, 2013: Quantifying global climate feedbacks, responses and
forcing under abrupt and gradual CO
2
forcing. Clim. Dyn., 41, 2471-2479.
Lu, P., H. Zhang, and X. Jing, 2012: The effects of different HITRAN versions on
calculated long-wave radiation and uncertainty evaluation. Acta Meteorol. Sin.,
26, 389–398.
Lu, Z., Q. Zhang, and D. G. Streets, 2011: Sulfur dioxide and primary carbonaceous
aerosol emissions in China and India, 1996–2010. Atmos. Chem. Phys., 11,
9839–9864.
Lund, M., T. Berntsen, J. Fuglestvedt, M. Ponater, and K. Shine, 2012: How much
information is lost by using global-mean climate metrics? an example using the
transport sector. Clim. Change, 113, 949–963.
MacFarling Meure, C., et al., 2006: Law Dome CO
2
, CH
4
and N
2
O ice core records
extended to 2000 years BP. Geophys. Res. Lett., 33, L14810.
MacMynowski, D., H. Shin, and K. Caldeira, 2011: The frequency response of
temperature and precipitation in a climate model. Geophys. Res. Lett., 38,
L16711.
Mader, J. A., J. Staehelin, T. Peter, D. Brunner, H. E. Rieder, and W. A. Stahel, 2010:
Evidence for the effectiveness of the Montreal Protocol to protect the ozone
layer. Atmos. Chem. Phys., 10, 12161–12171.
Mahowald, N. M., et al., 2010: Observed 20th century desert dust variability: Impact
on climate and biogeochemistry. Atmos. Chem. Phys., 10, 10875–10893.
Mann, M., M. Cane, S. Zebiak, and A. Clement, 2005: Volcanic and solar forcing of
the tropical Pacific over the past 1000 years. J. Clim., 18, 447–456.
Manne, A., and R. Richels, 2001: An alternative approach to establishing trade-offs
among greenhouse gases. Nature, 410, 675–677.
Manning, M., and A. Reisinger, 2011: Broader perspectives for comparing different
greenhouse gases. Philos. Trans. R. Soc. London A, 369, 1891–1905.
Marenco, A., H. Gouget, P. Nedelec, J. P. Pages, and F. Karcher, 1994: Evidence of
a long-term increase in troposheric ozone from PIC Du Midi Data Series—
Consequences—Positive radiative forcing. J. Geophys. Res. Atmos., 99, 16617–
16632.
Marten, A. L., and S. C. Newbold, 2012: Estimating the social cost of non-CO2 GHG
emissions: Methane and nitrous oxide. Energ. Policy, 51, 957–972.
Matthews, H. D., A. J. Weaver, K. J. Meissner, N. P. Gillett, and M. Eby, 2004: Natural
and anthropogenic climate change: Incorporating historical land cover change,
vegetation dynamics and the global carbon cycle. Clim. Dyn., 22, 461–479.
McComas, D., R. Ebert, H. Elliott, B. Goldstein, J. Gosling, N. Schwadron, and R.
Skoug, 2008: Weaker solar wind from the polar coronal holes and the whole
Sun. Geophys. Res. Lett., 35, L18103.
McLandress, C., T. G. Shepherd, J. F. Scinocca, D. A. Plummer, M. Sigmond, A. I.
Jonsson, and M. C. Reader, 2011: Separating the dynamical effects of climate
change and ozone depletion. Part II. Southern Hemisphere troposphere. J. Clim.,
24, 1850–1868.
Meinshausen, M., T. Wigley, and S. Raper, 2011a: Emulating atmosphere-ocean
and carbon cycle models with a simpler model, MAGICC6–Part 2: Applications.
Atmos. Chem. Phys., 11, 1457–1471.
Meinshausen, M., et al., 2011b: The RCP greenhouse gas concentrations and their
extensions from 1765 to 2300. Clim. Change, 109, 213–241.
Mercado, L. M., N. Bellouin, S. Sitch, O. Boucher, C. Huntingford, M. Wild, and P. M.
Cox, 2009: Impact of changes in diffuse radiation on the global land carbon sink.
Nature, 458, 1014–1017.
Merikanto, J., D. Spracklen, G. Mann, S. Pickering, and K. Carslaw, 2009: Impact of
nucleation on global CCN. Atmos. Chem. Phys., 9, 8601–8616.
Mickley, L. J., E. M. Leibensperger, D. J. Jacob, and D. Rind, 2012: Regional warming
from aerosol removal over the United States: Results from a transient 2010–
2050 climate simulation. Atmos. Environ., 46, 545–553.
Miller, G. H., et al., 2012: Abrupt onset of the Little Ice Age triggered by volcanism
and sustained by sea-ice/ocean feedbacks. Geophys. Res. Lett., 39, L02708.
Miller, R. L., I. Tegen, and J. Perlwitz, 2004: Surface radiative forcing by soil dust
aerosols and the hydrologic cycle. J. Geophys. Res. Atmos., 109, D04203.
Mills, M. J., O. B. Toon, R. P. Turco, D. E. Kinnison, and R. R. Garcia, 2008: Massive
global ozone loss predicted following regional nuclear conflict. Proc. Natl. Acad.
Sci. U.S.A., 105, 5307–5312.
Ming, Y., and V. Ramaswamy, 2012: Nonlocal component of radiative flux
perturbation. Geophys. Res. Lett., 39, L22706.
Ming, Y., V. Ramaswamy, and G. Persad, 2010: Two opposing effects of absorbing
aerosols on global-mean precipitation. Geophys. Res. Lett., 37, L13701.
Ming, Y., V. Ramaswamy, and G. Chen, 2011: A model investigation of aerosol-
induced changes in boreal winter extratropical circulation. J. Clim., 24, 6077–
6091.
Ming, Y., V. Ramaswamy, L. J. Donner, V. T. J. Phillips, S. A. Klein, P. A. Ginoux, and L. W.
Horowitz, 2007: Modeling the interactions between aerosols and liquid water
clouds with a self-consistent cloud scheme in a general circulation model. J.
Atmos. Sci., 64, 1189–1209.
Mirme, S., A. Mirme, A. Minikin, A. Petzold, U. Horrak, V.-M. Kerminen, and M.
Kulmala, 2010: Atmospheric sub-3 nm particles at high altitudes. Atm. Chem.
Phys., 10, 437–451.
Montzka, S. A., E. J. Dlugokencky, and J. H. Butler, 2011: Non-CO
2
greenhouse gases
and climate change. Nature, 476, 43–50.
Morrill, J., L. Floyd, and D. McMullin, 2011: The solar ultraviolet spectrum estimated
using the Mg II Index and Ca II K disk activity. Solar Physics, 269, 253–267.
Muhle, J., et al., 2009: Sulfuryl fluoride in the global atmosphere. J. Geophys. Res.
Atmos., 114, D05306.
Mulitza, S., et al., 2010: Increase in African dust flux at the onset of commercial
agriculture in the Sahel region. Nature, 466, 226–228.
Murphy, D., and D. Fahey, 1994: An estimate of the flux of stratospheric reactive
nitrogen and ozone into the troposphere. J. Geophys. Res. Atmos., 99, 5325–
5332.
Myhre, G., M. M. Kvalevag, and C. B. Schaaf, 2005a: Radiative forcing due to
anthropogenic vegetation change based on MODIS surface albedo data.
Geophys. Res. Lett., 32, L21410.
Myhre, G., E. J. Highwood, K. P. Shine, and F. Stordal, 1998: New estimates of
radiative forcing due to well mixed greenhouse gases. Geophys. Res. Lett., 25,
2715–2718.
Myhre, G., Y. Govaerts, J. M. Haywood, T. K. Berntsen, and A. Lattanzio, 2005b:
Radiative effect of surface albedo change from biomass burning. Geophys. Res.
Lett., 32, L20812.
Myhre, G., J. Nilsen, L. Gulstad, K. Shine, B. Rognerud, and I. Isaksen, 2007: Radiative
forcing due to stratospheric water vapour from CH
4
oxidation. Geophys. Res.
Lett., 34, L01807.
Myhre, G., et al., 2011: Radiative forcing due to changes in ozone and methane
caused by the transport sector. Atmos. Environ., 45, 387–394.
Myhre, G., et al., 2013: Radiative forcing of the direct aerosol effect from AeroCom
Phase II simulations. Atmos. Chem. Phys., 13, 1853–1877.
Nagai, T., B. Liley, T. Sakai, T. Shibata, and O. Uchino, 2010: Post-Pinatubo evolution
and subsequent trend of the stratospheric aerosol layer observed by mid-
latitude lidars in both hemispheres. Sola, 6, 69–72.
Naik, V., D. L. Mauzerall, L. W. Horowitz, M. D. Schwarzkopf, V. Ramaswamy, and M.
Oppenheimer, 2007: On the sensitivity of radiative forcing from biomass burning
aerosols and ozone to emission location. Geophys. Res. Lett., 34, L03818.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
727
Nair, U. S., D. K. Ray, J. Wang, S. A. Christopher, T. J. Lyons, R. M. Welch, and R. A. Pielke,
2007: Observational estimates of radiative forcing due to land use change in
southwest Australia. J. Geophys. Res. Atmos., 112, D09117.
Neely, R. R., et al., 2013: Recent anthropogenic increases in SO
2
from Asia have
minimal impact on stratospheric aerosol. Geophys. Res. Lett., 40, 999-1004.
O’ishi, R., A. Abe-Ouchi, I. Prentice, and S. Sitch, 2009: Vegetation dynamics and plant
CO
2
responses as positive feedbacks in a greenhouse world. Geophys. Res. Lett.,
36, L11706.
O’Neill, B., 2000: The jury is still out on global warming potentials. Clim. Change,
44, 427–443.
O’Neill, B., 2003: Economics, natural science, and the costs of global warming
potentials—An editorial comment. Clim. Change, 58, 251–260.
Oleson, K. W., G. B. Bonan, and J. Feddema, 2010: Effects of white roofs on urban
temperature in a global climate model. Geophys. Res. Lett., 37, L03701.
Olivié, D. J. L., G. Peters, and D. Saint-Martin, 2012: Atmosphere response time scales
estimated from AOGCM experiments. J. Climate, 25, 7956–7972.
Olsen, S. C., C. A. McLinden, and M. J. Prather, 2001: Stratospheric N
2
O–NOy system:
Testing uncertainties in a three-dimensional framework, J. Geophys. Res., 106,
28771.
Oreopoulos, L., et al., 2012: The Continual Intercomparison of Radiation Codes:
Results from Phase I. J. Geophys. Res. Atmos., 117, D06118.
Osterman, G. B., et al., 2008: Validation of Tropospheric Emission Spectrometer (TES)
measurements of the total, stratospheric, and tropospheric column abundance
of ozone. J. Geophys. Res. Atmos., 113, D15S16.
Otterä, O. H., M. Bentsen, H. Drange, and L. L. Suo, 2010: External forcing as a
metronome for Atlantic multidecadal variability. Nature Geosci., 3, 688–694.
Özdoğan, M., A. Robock, and C. J. Kucharik, 2013: Impacts of a nuclear war in South
Asia on soybean and maize production in the Midwest United States. Clim.
Change, 116, 373–387.
Parrish, D. D., D. B. Millet, and A. H. Goldstein, 2009: Increasing ozone in marine
boundary layer inflow at the west coasts of North America and Europe. Atmos.
Chem. Phys., 9, 1303–1323.
Paulot, F., J. D. Crounse, H. G. Kjaergaard, A. Kurten, J. M. St Clair, J. H. Seinfeld,
and P. O. Wennberg, 2009: Unexpected epoxide formation in the gas-phase
photooxidation of isoprene. Science, 325, 730–733.
Paynter, D., and V. Ramaswamy, 2011: An assessment of recent water vapor
continuum measurements upon longwave and shortwave radiative transfer. J.
Geophys. Res. Atmos., 116, D20302.
Pechony, O., and D. Shindell, 2010: Driving forces of global wildfires over the past
millennium and the forthcoming century. Proc. Natl. Acad. Sci. U.S.A., 107,
19167–19170.
Peeters, J., T. L. Nguyen, and L. Vereecken, 2009: HO(x) radical regeneration in the
oxidation of isoprene. Phys. Chem. Chem. Phys., 11, 5935–5939.
Penn, M., and W. Livingston, 2006: Temporal changes in sunspot umbral magnetic
fields and temperatures. Astrophys. J., 649, L45–L48.
Penner, J., L. Xu, and M. Wang, 2011: Satellite methods underestimate indirect
climate forcing by aerosols. Proc. Natl. Acad. Sci. U.S.A., 108, 13404–13408.
Penner, J., et al., 2006: Model intercomparison of indirect aerosol effects. Atmos.
Chem. Phys., 6, 3391–3405.
Perlwitz, J., and R. L. Miller, 2010: Cloud cover increase with increasing aerosol
absorptivity: A counterexample to the conventional semidirect aerosol effect. J.
Geophys. Res. Atmos., 115, D08203.
Persad, G. G., Y. Ming, and V. Ramaswamy, 2012: Tropical tropospheric-only responses
to absorbing aerosols. J. Clim., 25, 2471–2480.
Peters, G., B. Aamaas, T. Berntsen, and J. Fuglestvedt, 2011a: The integrated global
temperature change potential (iGTP) and relationships between emission
metrics. Environ. Res. Lett., 6, 044021.
Peters, G. P., B. Aamaas, M. T. Lund, C. Solli, and J. S. Fuglestvedt, 2011b: Alternative
“Global Warming” metrics in life cycle assessment: A case study with existing
transportation data. Environ. Sci. Technol., 45, 8633–8641.
Peters, K., P. Stier, J. Quaas, and H. Grassl, 2012: Aerosol indirect effects from shipping
emissions: Sensitivity studies with the global aerosol-climate model ECHAM-
HAM. Atmos. Chem. Phys., 12, 5985–6007.
Philipona, R., K. Behrens, and C. Ruckstuhl, 2009: How declining aerosols and rising
greenhouse gases forced rapid warming in Europe since the 1980s. Geophys.
Res. Lett., 36, L02806.
Pinto, J. P., R. P. Turco, and O. B. Toon, 1989: Self-limiting physical and chemical
effects in volcanic-eruption clouds. J. Geophys. Res. Atmos., 94, 11165–11174.
Pitman, A. J., et al., 2009: Uncertainties in climate responses to past land cover
change: First results from the LUCID intercomparison study. Geophys. Res. Lett.,
36, L14814.
Plattner, G.-K., T. Stocker, P. Midgley, and M. Tignor, 2009: IPCC Expert Meeting
on the Science of Alternative Metrics: Meeting Report. IPCC Working Group I,
Technical Support Unit.
Plattner, G. K., et al., 2008: Long-term climate commitments projected with climate-
carbon cycle models. J. Clim., 21, 2721–2751.
Pongratz, J., C. Reick, T. Raddatz, and M. Claussen, 2008: A reconstruction of global
agricultural areas and land cover for the last millennium. Global Biogeochem.
Cycles, 22, Gb3018.
Pongratz, J., C. H. Reick, T. Raddatz, and M. Claussen, 2010: Biogeophysical versus
biogeochemical climate response to historical anthropogenic land cover change.
Geophys. Res. Lett., 37, L08702.
Pongratz, J., T. Raddatz, C. H. Reick, M. Esch, and M. Claussen, 2009: Radiative
forcing from anthropogenic land cover change since AD 800. Geophys. Res. Lett.,
36, L02709.
Pozzer, A., et al., 2012: Effects of business-as-usual anthropogenic emissions on air
quality. Atmos. Chem. Phys., 12, 6915–6937.
Prather, M., 2002: Lifetimes of atmospheric species: Integrating environmental
impacts. Geophys. Res. Lett., 29, 2063.
Prather, M., and J. Hsu, 2010: Coupling of nitrous oxide and methane by global
atmospheric chemistry. Science, 330, 952–954.
Prather, M. J., 1998: Time scales in atmospheric chemistry: Coupled perturbations to
N
2
O, NOy, and O
3
. Science, 279, 1339–1341.
Prather, M. J., C. D. Holmes, and J. Hsu, 2012: Reactive greenhouse gas scenarios:
Systematic exploration of uncertainties and the role of atmospheric chemistry.
Geophys. Res. Lett., 39, L09803.
Puma, M. J., and B. I. Cook, 2010: Effects of irrigation on global climate during the
20th century. J. Geophys. Res. Atmos., 115, D16120.
Quaas, J., O. Boucher, N. Bellouin, and S. Kinne, 2011: Which of satellite- or model-
based estimates is closer to reality for aerosol indirect forcing? Proc. Natl. Acad.
Sci. U.S.A., 108, E1099.
Quaas, J., et al., 2009: Aerosol indirect effects—general circulation model
intercomparison and evaluation with satellite data. Atmos. Chem. Phys., 9,
8697–8717.
Rajakumar, B., R. Portmann, J. Burkholder, and A. Ravishankara, 2006: Rate
coefficients for the reactions of OH with CF
3
CH
2
CH
3
(HFC-263fb), CF
3
CHFCH
2
F
(HFC-245eb), and CHF
2
CHFCHF
2
(HFC-245ea) between 238 and 375 K. J. Phys.
Chem. A, 110, 6724–6731.
Ramanathan, V., and G. Carmichael, 2008: Global and regional climate changes due
to black carbon. Nature Geosci., 1, 221–227.
Ramanathan, V., et al., 2005: Atmospheric brown clouds: Impacts on South Asian
climate and hydrological cycle. Proc. Natl. Acad. Sci. U.S.A., 102, 5326–5333.
Ramaswamy, V., et al., 2001: Radiative forcing of climate change. In: Climate
Change 2001: The Scientific Basis. Contribution of Working Group I to the
Third Assessment Report of the Intergovernmntal Panel on Climate Change [J.
T. Houghton, Y. Ding, D. J. Griggs, M. Noquer, P. J. van der Linden, X. Dai, K.
Maskell and C. A. Johnson (eds.)]. Cambride University Press, Cambridge, United
Kingdom and New York, NY, USA, 349-416.
Randel, W., and F. Wu, 2007: A stratospheric ozone profile data set for 1979–2005:
Variability, trends, and comparisons with column ozone data. J. Geophys. Res.
Atmos., 112, D06313.
Randles, C. A., and V. Ramaswamy, 2008: Absorbing aerosols over Asia: A Geophysical
Fluid Dynamics Laboratory general circulation model sensitivity study of model
response to aerosol optical depth and aerosol absorption. J. Geophys. Res.
Atmos., 113, D21203.
Rasch, P. J., et al., 2008: An overview of geoengineering of climate using stratospheric
sulphate aerosols. Philos. Trans. R. Soc. A, 366, 4007–4037.
Ravishankara, A. R., J. S. Daniel, and R. W. Portmann, 2009: Nitrous oxide (N
2
O): The
dominant ozone-depleting substance emitted in the 21st century. Science, 326,
123–125.
Rechid, D., T. Raddatz, and D. Jacob, 2009: Parameterization of snow-free land
surface albedo as a function of vegetation phenology based on MODIS data and
applied in climate modelling. Theor. Appl. Climatol., 95, 245–255.
Reddy, M., and O. Boucher, 2007: Climate impact of black carbon emitted from
energy consumption in the world’s regions. Geophys. Res. Lett., 34, L11802.
Reilly, J., et al., 1999: Multi-gas assessment of the Kyoto Protocol. Nature, 401,
549–555.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
728
Reilly, J. M., and K. R. Richards, 1993: Climate-change damage and the trace-gas-
index issue. Environ. Resour. Econ., 3, 41–61.
Reisinger, A., M. Meinshausen, and M. Manning, 2011: Future changes in global
warming potentials under representative concentration pathways. Environ. Res.
Lett., 6, 024020.
Reisinger, A., M. Meinshausen, M. Manning, and G. Bodeker, 2010: Uncertainties of
global warming metrics: CO
2
and CH
4
. Geophys. Res. Lett., 37, L14707.
Reisinger, A., P. Havlik, K. Riahi, O. Vliet, M. Obersteiner, and M. Herrero, 2013:
Implications of alternative metrics for global mitigation costs and greenhouse
gas emissions from agriculture. Clim. Change, 117, 677-690.
Righi, M., C. Klinger, V. Eyring, J. Hendricks, A. Lauer, and A. Petzold, 2011: Climate
impact of biofuels in shipping: global model studies of the aerosol indirect
effect. Environ. Sci. Technol., 45, 3519–3525.
Rigozo, N., E. Echer, L. Vieira, and D. Nordemann, 2001: Reconstruction of Wolf
sunspot numbers on the basis of spectral characteristics and estimates of
associated radio flux and solar wind parameters for the last millennium. Sol.
Phys., 203, 179–191.
Rigozo, N., D. Nordemann, E. Echer, M. Echer, and H. Silva, 2010: Prediction of solar
minimum and maximum epochs on the basis of spectral characteristics for the
next millennium. Planet. Space Sci., 58, 1971–1976.
Robock, A., 2000: Volcanic eruptions and climate. Rev. Geophys., 38, 191–219.
Robock, A., 2010: New START, Eyjafjallajökull, and Nuclear Winter. Eos, 91, 444–445.
Robock, A., L. Oman, and G. L. Stenchikov, 2007a: Nuclear winter revisited with
a modern climate model and current nuclear arsenals: Still catastrophic
consequences. J. Geophys. Res. Atmos., 112, D13107.
Robock, A., L. Oman, and G. L. Stenchikov, 2008: Regional climate responses to
geoengineering with tropical and Arctic SO
2
injections. J. Geophys. Res. Atmos.,
113, D16101.
Robock, A., L. Oman, G. L. Stenchikov, O. B. Toon, C. Bardeen, and R. P. Turco, 2007b:
Climatic consequences of regional nuclear conflicts. Atmos. Chem. Phys., 7,
2003–2012.
Robock, A., C. M. Ammann, L. Oman, D. Shindell, S. Levis, and G. Stenchikov, 2009:
Did the Toba volcanic eruption of similar to 74 ka BP produce widespread
glaciation? J. Geophys. Res. Atmos., 114, D10107.
Rogelj, J., et al., 2011: Emission pathways consistent with a 2 degrees C global
temperature limit. Nature Clim. Change, 1, 413–418.
Roscoe, H. K., and J. D. Haigh, 2007: Influences of ozone depletion, the solar cycle
and the QBO on the Southern Annular Mode. Q. J. R. Meteorol. Soc., 133, 1855–
1864.
Rotenberg, E., and D. Yakir, 2010: Contribution of semi-arid forests to the climate
system. Science, 327, 451–454.
Rothman, L., 2010: The evolution and impact of the HITRAN molecular spectroscopic
database. J. Quant. Spectrosc. Radiat. Transfer, 111, 1565–1567.
Rotstayn, L., and J. Penner, 2001: Indirect aerosol forcing, quasi forcing, and climate
response. J. Clim., 14, 2960–2975.
Rotstayn, L. D., and U. Lohmann, 2002: Tropical rainfall trends and the indirect
aerosol effect. J. Clim., 15, 2103–2116.
Rotstayn, L. D., B. F. Ryan, and J. E. Penner, 2000: Precipitation changes in a GCM
resulting from the indirect effects of anthropogenic aerosols. Geophys. Res. Lett.,
27, 3045–3048.
Rottman, G., 2006: Measurement of total and spectral solar irradiance. Space Sci.
Rev., 125, 39–51.
Ruckstuhl, C., et al., 2008: Aerosol and cloud effects on solar brightening and the
recent rapid warming. Geophys. Res. Lett., 35, L12708.
Russell, C., J. Luhmann, and L. Jian, 2010: How unprecedented a solar minimum?
Rev. Geophys., 48, RG2004.
Rypdal, K., N. Rive, T. Berntsen, Z. Klimont, T. Mideksa, G. Myhre, and R. Skeie, 2009:
Costs and global impacts of black carbon abatement strategies. Tellus B., 61,
625–641.
Rypdal, K., et al., 2005: Tropospheric ozone and aerosols in climate agreements:
Scientific and political challenges. Environ. Sci. Policy, 8, 29–43.
Salby, M. L., E. A. Titova, and L. Deschamps, 2012: Changes of the Antarctic ozone
hole: Controlling mechanisms, seasonal predictability, and evolution. J. Geophys.
Res. Atmos., 117, D10111.
Sarofim, M., 2012: The GTP of methane: Modeling analysis of temperature impacts of
methane and carbon dioxide reductions. Environ. Model. Assess., 17, 231–239.
Sato, M., J. E. Hansen, M. P. McCormick, and J. B. Pollack, 1993: Stratospheric aerosol
optical depths, 1850–1990. J. Geophys. Res. Atmos., 98, 22987–22994.
Schaaf, C., et al., 2002: First operational BRDF, albedo nadir reflectance products
from MODIS. Remote Sens. Environ., 83, 135–148.
Schmalensee, R., 1993: Comparing greenhouse gases for policy purposes. Energy J.,
14, 245–256.
Schmidt, A., K. S. Carslaw, G. W. Mann, M. Wilson, T. J. Breider, S. J. Pickering, and
T. Thordarson, 2010: The impact of the 1783–1784 AD Laki eruption on global
aerosol formation processes and cloud condensation nuclei. Atmos. Chem. Phys.,
10, 6025–6041.
Schmidt, A., et al., 2012: Importance of tropospheric volcanic aerosol for indirect
radiative forcing of climate. Atmos. Chem. Phys., 12, 7321–7339.
Schmidt, G., et al., 2011: Climate forcing reconstructions for use in PMIP simulations
of the last millennium (v1.0). Geosci. Model Dev., 4, 33–45.
Schneider, D. P., C. M. Ammann, B. L. Otto-Bliesner, and D. S. Kaufman, 2009: Climate
response to large, high-latitude and low-latitude volcanic eruptions in the
Community Climate System Model. J. Geophys. Res. Atmos., 114, D15101.
Schultz, M. G., et al., 2008: Global wildland fire emissions from 1960 to 2000. Global
Biogeochem. Cycles, 22, Gb2002.
Seinfeld, J. H., and S. N. Pandis, 2006: Atmospheric Chemistry and Physics: From Air
Pollution to Climate Change. John Wiley & Sons, Hoboken, NJ, USA.
Self, S., and S. Blake, 2008: Consequences of explosive supereruptions. Elements,
4, 41–46.
Sharma, S., D. Lavoue, H. Cachier, L. Barrie, and S. Gong, 2004: Long-term trends of
the black carbon concentrations in the Canadian Arctic. J. Geophys. Res. Atmos.,
109, D15203.
Shibata, K., and K. Kodera, 2005: Simulation of radiative and dynamical responses
of the middle atmosphere to the 11-year solar cycle. J. Atmos. Sol. Terres. Phys.,
67, 125–143.
Shindell, D., and G. Faluvegi, 2009: Climate response to regional radiative forcing
during the twentieth century. Nature Geosci., 2, 294–300.
Shindell, D., and G. Faluvegi, 2010: The net climate impact of coal-fired power plant
emissions. Atmos. Chem. Phys., 10, 3247–3260.
Shindell, D., G. Schmidt, M. Mann, D. Rind, and A. Waple, 2001: Solar forcing of
regional climate change during the maunder minimum. Science, 294, 2149–
2152.
Shindell, D., G. Faluvegi, A. Lacis, J. Hansen, R. Ruedy, and E. Aguilar, 2006a: Role of
tropospheric ozone increases in 20th-century climate change. J. Geophys. Res.
Atmos., 111, D08302.
Shindell, D., G. Faluvegi, R. Miller, G. Schmidt, J. Hansen, and S. Sun, 2006b: Solar and
anthropogenic forcing of tropical hydrology. Geophys. Res. Lett., 33, L24706.
Shindell, D., M. Schulz, Y. Ming, T. Takemura, G. Faluvegi, and V. Ramaswamy, 2010:
Spatial scales of climate response to inhomogeneous radiative forcing. J.
Geophys. Res. Atmos., 115, D19110.
Shindell, D., et al., 2008: Climate forcing and air quality change due to regional
emissions reductions by economic sector. Atmos. Chem. Phys., 8, 7101–7113.
Shindell, D., et al., 2011: Climate, health, agricultural and economic impacts of
tighter vehicle-emission standards. Nature Clim. Change, 1, 59–66.
Shindell, D., et al., 2013a: Attribution of historical ozone forcing to anthropogenic
emissions. Nature Clim. Change, 3, 567-570.
Shindell, D., et al., 2012a: Simultaneously mitigating near-term climate change and
improving human health and food Security. Science, 335, 183–189.
Shindell, D. T., 2012: Evaluation of the absolute regional temperature potential.
Atmos. Chem. Phys., 12, 7955–7960.
Shindell, D. T., A. Voulgarakis, G. Faluvegi, and G. Milly, 2012b: Precipitation response
to regional radiative forcing. Atmos. Chem. Phys., 12, 6969–6982.
Shindell, D. T., G. Faluvegi, D. M. Koch, G. A. Schmidt, N. Unger, and S. E. Bauer, 2009:
Improved attribution of climate forcing to emissions. Science, 326, 716–718.
Shindell, D. T., et al., 2006c: Simulations of preindustrial, present-day, and 2100
conditions in the NASA GISS composition and climate model G-PUCCINI. Atmos.
Chem. Phys., 6, 4427–4459.
Shindell, D. T., et al., 2013b: Interactive ozone and methane chemistry in GISS-E2
historical and future climate simulations. Atmos. Chem. Phys., 13, 2653–2689.
Shindell, D. T., et al., 2013c: Radiative forcing in the ACCMIP historical and future
climate simulations. Atmos. Chem. Phys., 13, 2939–2974.
Shine, K., 2009: The global warming potential-the need for an interdisciplinary
retrial. Clim. Change, 96, 467–472.
Shine, K., J. Cook, E. Highwood, and M. Joshi, 2003: An alternative to radiative
forcing for estimating the relative importance of climate change mechanisms.
Geophys. Res. Lett., 30, 2047.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
729
Shine, K., J. Fuglestvedt, K. Hailemariam, and N. Stuber, 2005a: Alternatives to
the global warming potential for comparing climate impacts of emissions of
greenhouse gases. Clim. Change, 68, 281–302.
Shine, K., T. Berntsen, J. Fuglestvedt, and R. Sausen, 2005b: Scientific issues in the
design of metrics for inclusion of oxides of nitrogen in global climate agreements.
Proc. Natl. Acad. Sci. U.S.A., 102, 15768–15773.
Shine, K., T. Berntsen, J. Fuglestvedt, R. Skeie, and N. Stuber, 2007: Comparing the
climate effect of emissions of short- and long-lived climate agents. Philos. Trans.
R. Soc. A, 365, 1903–1914.
Shine, K. P., I. V. Ptashnik, and G. Raedel, 2012: The water vapour continuum: Brief
history and recent developments. Surv. Geophys., 33, 535–555.
Siddaway, J. M., and S. V. Petelina, 2011: Transport and evolution of the 2009
Australian Black Saturday bushfire smoke in the lower stratosphere observed by
OSIRIS on Odin. J. Geophys. Res. Atmos., 116, D06203.
Sitch, S., P. M. Cox, W. J. Collins, and C. Huntingford, 2007: Indirect radiative forcing
of climate change through ozone effects on the land-carbon sink. Nature, 448,
791–794.
Skeie, R., T. Berntsen, G. Myhre, K. Tanaka, M. Kvalevag, and C. Hoyle, 2011a:
Anthropogenic radiative forcing time series from pre-industrial times until 2010.
Atmos. Chem. Phys., 11, 11827–11857.
Skeie, R., T. Berntsen, G. Myhre, C. Pedersen, J. Strom, S. Gerland, and J. Ogren,
2011b: Black carbon in the atmosphere and snow, from pre-industrial times
until present. Atmos. Chem. Phys., 11, 6809–6836.
Skeie, R. B., J. Fuglestvedt, T. Berntsen, M. T. Lund, G. Myhre, and K. Rypdal, 2009:
Global temperature change from the transport sectors: Historical development
and future scenarios. Atmos. Environ., 43, 6260–6270.
Smith, E., and A. Balogh, 2008: Decrease in heliospheric magnetic flux in this solar
minimum: Recent Ulysses magnetic field observations. Geophys. Res. Lett., 35,
L22103.
Smith, S., and M. Wigley, 2000: Global warming potentials: 1. Climatic implications
of emissions reductions. Clim. Change, 44, 445–457.
Smith, S., J. Karas, J. Edmonds, J. Eom, and A. Mizrahi, 2013: Sensitivity of multi-gas
climate policy to emission metrics. Clim. Change, 117, 663–675.
Smith, S. M., J. A. Lowe, N. H. A. Bowerman, L. K. Gohar, C. Huntingford, and M.
R. Allen, 2012: Equivalence of greenhouse-gas emissions for peak temperature
limits. Nature Clim. Change, 2, 535–538.
Snow-Kropla, E., J. Pierce, D. Westervelt, and W. Trivitayanurak, 2011: Cosmic
rays, aerosol formation and cloud-condensation nuclei: Sensitivities to model
uncertainties. Atmos. Chem. Phys., 11, 4001–4013.
Solanki, S., and N. Krivova, 2004: Solar irradiance variations: From current
measurements to long-term estimates. Sol. Phys., 224, 197–208.
Solomon, S., 1999: Stratospheric ozone depletion: A review of concepts and history.
Rev. Geophys., 37, 275–316.
Solomon, S., J. S. Daniel, R. R. Neely, J. P. Vernier, E. G. Dutton, and L. W. Thomason,
2011: The persistently variable “background” stratospheric aerosol layer and
global climate change. Science, 333, 866–870.
Son, S. W., N. F. Tandon, L. M. Polvani, and D. W. Waugh, 2009: Ozone hole and
Southern Hemisphere climate change. Geophys. Res. Lett., 36, L15705.
Soukharev, B., and L. Hood, 2006: Solar cycle variation of stratospheric ozone:
Multiple regression analysis of long-term satellite data sets and comparisons
with models. J. Geophys. Res. Atmos., 111, D20314.
Søvde, O., C. Hoyle, G. Myhre, and I. Isaksen, 2011: The HNO
3
forming branch of the
HO
2
+ NO reaction: Pre-industrial-to-present trends in atmospheric species and
radiative forcings. Atmos. Chem. Phys., 11, 8929–8943.
Steinhilber, F., J. Beer, and C. Frohlich, 2009: Total solar irradiance during the
Holocene. Geophys. Res. Lett., 36, L19704.
Stenchikov, G., A. Robock, V. Ramaswamy, M. D. Schwarzkopf, K. Hamilton, and S.
Ramachandran, 2002: Arctic Oscillation response to the 1991 Mount Pinatubo
eruption: Effects of volcanic aerosols and ozone depletion. J. Geophys. Res.
Atmos., 107, 4803.
Stenchikov, G., T. L. Delworth, V. Ramaswamy, R. J. Stouffer, A. Wittenberg, and F. R.
Zeng, 2009: Volcanic signals in oceans. J. Geophys. Res. Atmos., 114, D16104.
Stephens, G. L., N. B. Wood, and L. A. Pakula, 2004: On the radiative effects of dust
on tropical convection. Geophys. Res. Lett., 31, L23112.
Stevenson, D., and R. Derwent, 2009: Does the location of aircraft nitrogen oxide
emissions affect their climate impact? Geophys. Res. Lett., 36, L17810.
Stevenson, D. S., et al., 2013: Tropospheric ozone changes, radiative forcing and
attribution to emissions in the Atmospheric Chemistry and Climate Model
Intercomparison Project (ACCMIP). Atmos. Chem. Phys., 13, 3063–3085.
Stevenson, D. S., et al., 2006: Multimodel ensemble simulations of present-day and
near-future tropospheric ozone. J. Geophys. Res. Atmos., 111, D08301.
Stiller, G. P., et al., 2012: Observed temporal evolution of global mean age of
stratospheric air for the 2002 to 2010 period. Atmos. Chem. Phys., 12, 3311–
3331.
Stothers, R. B., 2007: Three centuries of observation of stratospheric transparency.
Clim. Change, 83, 515–521.
Struthers, H., et al., 2011: The effect of sea ice loss on sea salt aerosol concentrations
and the radiative balance in the Arctic. Atmos. Chem. Phys., 11, 3459–3477.
Swann, A. L. S., I. Y. Fung, and J. C. H. Chiang, 2012: Mid-latitude afforestation shifts
general circulation and tropical precipitation. Proc. Natl. Acad. Sci. U.S.A., 109,
712–716.
Swingedouw, D., L. Terray, C. Cassou, A. Voldoire, D. Salas-Melia, and J. Servonnat,
2011: Natural forcing of climate during the last millennium: Fingerprint of solar
variability. Clim. Dyn., 36, 1349–1364.
Takemura, T., 2012: Distributions and climate effects of atmospheric aerosols from
the preindustrial era to 2100 along Representative Concentration Pathways
(RCPs) simulated using the global aerosol model SPRINTARS. Atmos. Chem.
Phys., 12, 11555–11572.
Tanaka, K., D. Johansson, B. O’Neill, and J. Fuglestvedt, 2013: Emission metrics under
the 2°C climate stabilization. Clim. Change Lett. , 117, 933-941.
Tanaka, K., B. O’Neill, D. Rokityanskiy, M. Obersteiner, and R. Tol, 2009: Evaluating
Global Warming Potentials with historical temperature. Clim. Change, 96, 443–
466.
Taraborrelli, D., et al., 2012: Hydroxyl radical buffered by isoprene oxidation over
tropical forests. Nature Geosci., 5, 190–193.
Taylor, P. C., R. G. Ellingson, and M. Cai, 2011: Seasonal variations of climate
feedbacks in the NCAR CCSM3. J. Clim., 24, 3433–3444.
Textor, C., et al., 2006: Analysis and quantification of the diversities of aerosol life
cycles within AeroCom. Atmos. Chem. Phys., 6, 1777–1813.
Thomason, L., and T. Peter, 2006: Assessment of Stratospheric Aerosol Properties
(ASAP). SPARC ReportsWCRP-124, WMO/TD- No. 1295, SPARC Report No. 4.
Thompson, D. W. J., S. Solomon, P. J. Kushner, M. H. England, K. M. Grise, and D. J.
Karoly, 2011: Signatures of the Antarctic ozone hole in Southern Hemisphere
surface climate change. Nature Geosci., 4, 741–749.
Thonicke, K., A. Spessa, I. C. Prentice, S. P. Harrison, L. Dong, and C. Carmona-
Moreno, 2010: The influence of vegetation, fire spread and fire behaviour on
biomass burning and trace gas emissions: Results from a process-based model.
Biogeosciences, 7, 1991–2011.
Tilmes, S., et al., 2012: Technical Note: Ozonesonde climatology between 1995 and
2011: Description, evaluation and applications. Atmos. Chem. Phys., 12, 7475–
7497.
Timmreck, C., 2012: Modeling the climatic effects of large explosive volcanic
eruptions. Climate Change, 3, 545–564.
Timmreck, C., et al., 2010: Aerosol size confines climate response to volcanic super-
eruptions. Geophys. Res. Lett., 37, L24705.
Tol, R., T. Berntsen, B. O’Neill, J. Fuglestvedt, and K. Shine, 2012: A unifying framework
for metrics for aggregating the climate effect of different emissions. Environ.
Res. Lett., 7, 044006.
Toon, O. B., A. Robock, and R. P. Turco, 2008: Environmental consequences of nuclear
war. Physics Today, 61, 37–42.
Trenberth, K. E., and A. Dai, 2007: Effects of Mount Pinatubo volcanic eruption on
the hydrological cycle as an analog of geoengineering. Geophys. Res. Lett., 34,
L15702.
Tsigaridis, K., and M. Kanakidou, 2007: Secondary organic aerosol importance in the
future atmosphere. Atmos. Environ., 41, 4682–4692.
UNEP, 2011: Near-term Climate Protection and Clean Air Benefits: Actions
for Controlling Short-Lived Climate Forcers. United Nations Environment
Programme (UNEP), 78 pp.
Unger, N., T. C. Bond, J. S. Wang, D. M. Koch, S. Menon, D. T. Shindell, and S. Bauer,
2010: Attribution of climate forcing to economic sectors. Proc. Natl. Acad. Sci.
U.S.A., 107, 3382–3387.
van der Molen, M. K., B. J. J. M. van den Hurk, and W. Hazeleger, 2011: A dampened
land use change climate response towards the tropics. Clim. Dyn., 37, 2035–
2043.
van der Werf, G. R., et al., 2010: Global fire emissions and the contribution of
deforestation, savanna, forest, agricultural, and peat fires (1997–2009). Atmos.
Chem. Phys., 10, 11707–11735.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
730
van Vuuren, D., J. Edmonds, M. Kainuma, K. Riahi, and J. Weyant, 2011: A special
issue on the RCPs. Clim. Change, 109, 1–4.
Van Vuuren, D. P., et al., 2008: Temperature increase of 21st century mitigation
scenarios. Proc. Natl. Acad. Sci. U.S.A., 105, 15258–15262.
Vasekova, E., E. Drage, K. Smith, and N. Mason, 2006: FTIR spectroscopy and
radiative forcing of octafluorocyclobutane and octofluorocyclopentene. J. Quant.
Spectrosc. Radiat. Transfer, 102, 418–424.
Velders, G. J. M., D. W. Fahey, J. S. Daniel, M. McFarland, and S. O. Andersen, 2009:
The large contribution of projected HFC emissions to future climate forcing. Proc.
Natl. Acad. Sci. U.S.A., 106, 10949–10954.
Vernier, J. P., L. W. Thomason, T. D. Fairlie, P. Minnis, R. Palikonda, and K. M. Bedka,
2013: Comment on “Large Volcanic Aerosol Load in the Stratosphere Linked to
Asian Monsoon Transport”. Science, 339, 647-d.
Vernier, J. P., et al., 2011: Major influence of tropical volcanic eruptions on the
stratospheric aerosol layer during the last decade. Geophys. Res. Lett., 38,
L12807.
Vial, J., J.-L. Dufresne, and S. Bony, 2013: On the interpretation of inter-model spread
in CMIP5 climate sensitivity estimates. Clim. Dyn., doi:10.1007/s00382-013-
1725-9, in press.
Volz, A., and D. Kley, 1988: Evaluation of the Montsouris Series of ozone
measurements made in the 19
th
century. Nature, 332, 240–242.
Voulgarakis, A., and D. T. Shindell, 2010: Constraining the sensitivity of regional
climate with the use of historical observations. J. Clim., 23, 6068–6073.
Voulgarakis, A., et al., 2013: Analysis of present day and future OH and methane
lifetime in the ACCMIP simulations. Atmos. Chem. Phys., 13, 2563–2587.
Wang, C., D. Kim, A. Ekman, M. Barth, and P. Rasch, 2009: Impact of anthropogenic
aerosols on Indian summer monsoon. Geophys. Res. Lett., 36, L21704.
Wang, M., and J. E. Penner, 2009: Aerosol indirect forcing in a global model with
particle nucleation. Atmos. Chem. Phys., 9, 239–260.
Wang, X., S. J. Doherty, and J. Huang, 2013: Black carbon and other light-absorbing
impurities in snow across Northern China. J. Geophys. Res. Atmos., 118, 1471–
1492.
Wang, Y., J. Lean, and N. Sheeley, 2005: Modeling the sun’s magnetic field and
irradiance since 1713. Astrophys. J., 625, 522–538.
Warren, S., and W. Wiscombe, 1980: A model for the spectral albedo of snow. 2.
Snow containing atmospheric aerosols. J. Atmos. Sci., 37, 2734–2745.
Weiss, R., J. Muhle, P. Salameh, and C. Harth, 2008: Nitrogen trifluoride in the global
atmosphere. Geophys. Res. Lett., 35, L20821.
Wenzler, T., S. Solanki, and N. Krivova, 2009: Reconstructed and measured total solar
irradiance: Is there a secular trend between 1978 and 2003? Geophys. Res. Lett.,
36, L11102.
Wilcox, L., K. Shine, and B. Hoskins, 2012: Radiative forcing due to aviation water
vapour emissions. Atmos. Environ., 63, 1–13.
Wild, M., 2009: How well do IPCC-AR4/CMIP3 climate models simulate global
dimming/brightening and twentieth-century daytime and nighttime warming?
J. Geophys. Res. Atmos., 114, D00D11.
Wild, O., 2007: Modelling the global tropospheric ozone budget: Exploring the
variability in current models. Atmos. Chem. Phys., 7, 2643–2660.
Wild, O., and P. I. Palmer, 2008: How sensitive is tropospheric oxidation to
anthropogenic emissions? Geophys. Res. Lett., 35, L22802.
Wild, O., M. Prather, and H. Akimoto, 2001: Indirect long-term global radiative
cooling from NOx emissions. Geophys. Res. Lett., 28, 1719–1722.
Williams, K. D., A. Jones, D. L. Roberts, C. A. Senior, and M. J. Woodage, 2001: The
response of the climate system to the indirect effects of anthropogenic sulfate
aerosol. Clim. Dyn., 17, 845–856.
Willson, R., and A. Mordvinov, 2003: Secular total solar irradiance trend during solar
cycles 21–23. Geophys. Res. Lett., 30, 1199.
WMO, 1999: Scientific Assessment of Ozone Depletion: 1998. Global Ozone Research
and Monitoring Project. Report No. 44.World Meteorological Organization,
Geneva, Switzerland.
WMO, 2011: Scientific Assessment of Ozone Depletion: 2010. Global Ozone
Research and Monitoring Project-Report. World Meteorological Organisation,
Geneva, Switzerland, 516 pp.
Worden, H., K. Bowman, S. Kulawik, and A. Aghedo, 2011: Sensitivity of outgoing
longwave radiative flux to the global vertical distribution of ozone characterized
by instantaneous radiative kernels from Aura-TES. J. Geophys. Res. Atmos., 116,
D14115.
Worden, H., K. Bowman, J. Worden, A. Eldering, and R. Beer, 2008: Satellite
measurements of the clear-sky greenhouse effect from tropospheric ozone.
Nature Geosci., 1, 305–308.
Wright, J., 2004: Do we know of any Maunder minimum stars? Astron. J., 128,
1273–1278.
Wu, S. L., L. J. Mickley, D. J. Jacob, J. A. Logan, R. M. Yantosca, and D. Rind, 2007: Why
are there large differences between models in global budgets of tropospheric
ozone? J. Geophys. Res. Atmos., 112, D05302.
Xia, L. L., and A. Robock, 2013: Impacts of a nuclear war in South Asia on rice
production in Mainland China. Clim. Change, 116, 357–372.
Ye, H., R. Zhang, J. Shi, J. Huang, S. G. Warren, and Q. Fu, 2012: Black carbon in
seasonal snow across northern Xinjiang in northwestern China. Environ. Res.
Lett., 7, 044002.
Young, P. J., et al., 2013: Pre-industrial to end 21st century projections of tropospheric
ozone from the Atmospheric Chemistry and Climate Model Intercomparison
Project (ACCMIP). Atmos. Chem. Phys., 13, 2063–2090.
Zanchettin, D., et al., 2012: Bi-decadal variability excited in the coupled ocean-
atmosphere system by strong tropical volcanic eruptions. Clim. Dyn., 39, 419–
444.
Zarzycki, C. M., and T. C. Bond, 2010: How much can the vertical distribution of black
carbon affect its global direct radiative forcing? Geophys. Res. Lett., 37, L20807.
Zeng, G., J. A. Pyle, and P. J. Young, 2008: Impact of climate change on tropospheric
ozone and its global budgets. Atmos. Chem. Phys., 8, 369–387.
Zeng, G., O. Morgenstern, P. Braesicke, and J. A. Pyle, 2010: Impact of stratospheric
ozone recovery on tropospheric ozone and its budget. Geophys. Res. Lett., 37,
L09805.
Zhang, H., G. Y. Shi, and Y. Liu, 2005: A comparison between the two line-by-line
integration algorithms. Chin. J. Atmos. Sci., 29, 581–593.
Zhang, H., G. Shi, and Y. Liu, 2008: The effects of line-wing cutoff in LBL integration
on radiation calculations. Acta Meteorol. Sin., 22, 248–255.
Zhang, H., J. Wu, and P. Luc, 2011: A study of the radiative forcing and global
warming potentials of hydrofluorocarbons. J. Quant. Spectrosc. Radiat. Transfer,
112, 220–229.
Zhang, X. B., et al., 2007: Detection of human influence on twentieth-century
precipitation trends. Nature, 448, 461–465.
Zhong, Y., G. H. Miller, B. L. Otto-Bliesner, M. M. Holland, D. A. Bailey, D. P. Schneider,
and A. Geirsdottir, 2011: Centennial-scale climate change from decadally-paced
explosive volcanism: A coupled sea ice-ocean mechanism. Clim. Dyn., 37, 2373–
2387.
Ziemke, J. R., S. Chandra, G. J. Labow, P. K. Bhartia, L. Froidevaux, and J. C. Witte,
2011: A global climatology of tropospheric and stratospheric ozone derived from
Aura OMI and MLS measurements. Atmos. Chem. Phys., 11, 9237–9251.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
731
Appendix 8.A: Lifetimes, Radiative Efficiencies and Metric Values
Table 8.A.1 | Radiative efficiencies (REs), lifetimes/adjustment times, AGWP and GWP values for 20 and 100 years, and AGTP and GTP values for 20, 50 and 100 years. Climate–carbon feedbacks are included for CO
2
while no climate
feedbacks are included for the other components (see discussion in Sections 8.7.1.4 and 8.7.2.1, Supplementary Material and notes below the table; Supplementary Material Table 8.SM.16 gives analogous values including climate–carbon
feedbacks for non-CO
2
emissions). For a complete list of chemical names and CAS numbers, and for accurate replications of metric values, see Supplementary Material Section 8.SM.13 and references therein.
Acronym, Common
Name or Chemi-
cal Name
Chemical
Formula
Lifetime
(Years)
Radiative
Efficiency
(W m
–2
ppb
–1
)
AGWP
20-year
(W m
–2
yr kg
–1
)
GWP
20-year
AGWP
100-year
(W m
–2
yr kg
–1
)
GWP
100-year
AGTP
20-year
(K kg
–1
)
GTP
20-year
AGTP
50-year
(K kg
–1
)
GTP
50-year
AGTP
100-year
(K kg
–1
)
GTP
100-year
Carbon dioxide CO
2
see* 1.37e-5 2.49e-14 1 9.17e-14 1 6.84e-16 1 6.17e-16 1 5.47e-16 1
Methane CH
4
12.4
†
3.63e-4 2.09e-12 84 2.61e-12 28 4.62e-14 67 8.69e-15 14 2.34e-15 4
Fossil methane‡ CH
4
12.4
†
3.63e-4 2.11e-12 85 2.73e-12 30 4.68e-14 68 9.55e-15 15 3.11e-15 6
Nitrous Oxide N
2
O 121
†
3.00e-3 6.58e-12 264 2.43e-11 265 1.89e-13 277 1.74e-13 282 1.28e-13 234
Chlorofluorocarbons
CFC-11 CCl
3
F 45.0 0.26 1.72e-10 6900 4.28e-10 4660 4.71e-12 6890 3.01e-12 4890 1.28e-12 2340
CFC-12 CCl
2
F
2
100.0 0.32 2.69e-10 10,800 9.39e-10 10,200 7.71e-12 11,300 6.75e-12 11,000 4.62e-12 8450
CFC-13 CClF
3
640.0 0.25 2.71e-10 10,900 1.27e-09 13,900 7.99e-12 11,700 8.77e-12 14,200 8.71e-12 15,900
CFC-113 CCl
2
FCClF
2
85.0 0.30 1.62e-10 6490 5.34e-10 5820 4.60e-12 6730 3.85e-12 6250 2.45e-12 4470
CFC-114 CClF
2
CClF
2
190.0 0.31 1.92e-10 7710 7.88e-10 8590 5.60e-12 8190 5.56e-12 9020 4.68e-12 8550
CFC-115 CClF
2
CF
3
1,020.0 0.20 1.46e-10 5860 7.03e-10 7670 4.32e-12 6310 4.81e-12 7810 4.91e-12 8980
Hydrochlorofluorocarbons
HCFC-21 CHCl
2
F 1.7 0.15 1.35e-11 543 1.35e-11 148 1.31e-13 192 1.59e-14 26 1.12e-14 20
HCFC-22 CHClF
2
11.9 0.21 1.32e-10 5280 1.62e-10 1760 2.87e-12 4200 5.13e-13 832 1.43e-13 262
HCFC-122 CHCl
2
CF
2
Cl 1.0 0.17 5.43e-12 218 5.43e-12 59 4.81e-14 70 6.25e-15 10 4.47e-15 8
HCFC-122a CHFClCFCl
2
3.4 0.21 2.36e-11 945 2.37e-11 258 2.91e-13 426 2.99e-14 48 1.96e-14 36
HCFC-123 CHCl
2
CF
3
1.3 0.15 7.28e-12 292 7.28e-12 79 6.71e-14 98 8.45e-15 14 6.00e-15 11
HCFC-123a CHClFCF
2
Cl 4.0 0.23 3.37e-11 1350 3.39e-11 370 4.51e-13 659 4.44e-14 72 2.81e-14 51
HCFC-124 CHClFCF
3
5.9 0.20 4.67e-11 1870 4.83e-11 527 7.63e-13 1120 7.46e-14 121 4.03e-14 74
HCFC-132c CH
2
FCFCl
2
4.3 0.17 3.07e-11 1230 3.10e-11 338 4.27e-13 624 4.14e-14 67 2.58e-14 47
HCFC-141b CH
3
CCl
2
F 9.2 0.16 6.36e-11 2550 7.17e-11 782 1.27e-12 1850 1.67e-13 271 6.09e-14 111
HCFC-142b CH
3
CClF
2
17.2 0.19 1.25e-10 5020 1.82e-10 1980 3.01e-12 4390 8.46e-13 1370 1.95e-13 356
HCFC-225ca CHCl
2
CF
2
CF
3
1.9 0.22 1.17e-11 469 1.17e-11 127 1.17e-13 170 1.38e-14 22 9.65e-15 18
HCFC-225cb CHClFCF
2
CClF
2
5.9 0.29 4.65e-11 1860 4.81e-11 525 7.61e-13 1110 7.43e-14 120 4.01e-14 73
(E)-1-Chloro-3,3,3-
trifluoroprop-1-ene
trans-CF
3
CH=CHCl 26.0 days 0.04 1.37e-13 5 1.37e-13 1 1.09e-15 2 1.54e-16 <1 1.12e-16 <1
(continued on next page)
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
732
(continued on next page)
Table 8.A.1 (continued)
Acronym, Common Name
or Chemical Name
Chemical
Formula
Lifetime
(Years)
Radiative
Efficiency
(W m
–2
ppb
–1
)
AGWP
20-year
(W m
–2
yr kg
–1
)
GWP
20-year
AGWP
100-year
(W m
–2
yr kg
–1
)
GWP
100-year
AGTP
20-year
(K kg
–1
)
GTP
20-year
AGTP
50-year
(K kg
–1
)
GTP
50-year
AGTP
100-year
(K kg
–1
)
GTP
100-year
Hydrofluorocarbons
HFC-23 CHF
3
222.0 0.18 2.70e-10 10,800 1.14e-09 12,400 7.88e-12 11,500 7.99e-12 13,000 6.95e-12 12,700
HFC-32 CH
2
F
2
5.2 0.11 6.07e-11 2430 6.21e-11 677 9.32e-13 1360 8.93e-14 145 5.17e-14 94
HFC-41 CH
3
F 2.8 0.02 1.07e-11 427 1.07e-11 116 1.21e-13 177 1.31e-14 21 8.82e-15 16
HFC-125 CHF
2
CF
3
28.2 0.23 1.52e-10 6090 2.91e-10 3170 3.97e-12 5800 1.84e-12 2980 5.29e-13 967
HFC-134 CHF
2
CHF
2
9.7 0.19 8.93e-11 3580 1.02e-10 1120 1.82e-12 2660 2.54e-13 412 8.73e-14 160
HFC-134a CH
2
FCF
3
13.4 0.16 9.26e-11 3710 1.19e-10 1300 2.09e-12 3050 4.33e-13 703 1.10e-13 201
HFC-143 CH
2
FCHF
2
3.5 0.13 3.00e-11 1200 3.01e-11 328 3.76e-13 549 3.82e-14 62 2.49e-14 46
HFC-143a CH
3
CF
3
47.1 0.16 1.73e-10 6940 4.41e-10 4800 4.76e-12 6960 3.12e-12 5060 1.37e-12 2500
HFC-152 CH
2
FCH
2
F 0.4 0.04 1.51e-12 60 1.51e-12 16 1.25e-14 18 1.71e-15 3 1.24e-15 2
HFC-152a CH
3
CHF
2
1.5 0.10 1.26e-11 506 1.26e-11 138 1.19e-13 174 1.47e-14 24 1.04e-14 19
HFC-161 CH
3
CH
2
F 66.0 days 0.02 3.33e-13 13 3.33e-13 4 2.70e-15 4 3.76e-16 <1 2.74e-16 <1
HFC-227ca CF
3
CF
2
CHF
2
28.2 0.27 1.27e-10 5080 2.42e-10 2640 3.31e-12 4830 1.53e-12 2480 4.41e-13 806
HFC-227ea CF
3
CHFCF
3
38.9 0.26 1.34e-10 5360 3.07e-10 3350 3.61e-12 5280 2.12e-12 3440 7.98e-13 1460
HFC-236cb CH
2
FCF
2
CF
3
13.1 0.23 8.67e-11 3480 1.11e-10 1210 1.94e-12 2840 3.92e-13 636 1.01e-13 185
HFC-236ea CHF
2
CHFCF
3
11.0 0.30
a
1.03e-10 4110 1.22e-10 1330 2.18e-12 3190 3.53e-13 573 1.06e-13 195
HFC-236fa CF
3
CH
2
CF
3
242.0 0.24 1.73e-10 6940 7.39e-10 8060 5.06e-12 7400 5.18e-12 8400 4.58e-12 8380
HFC-245ca CH
2
FCF
2
CHF
2
6.5 0.24
b
6.26e-11 2510 6.56e-11 716 1.07e-12 1570 1.09e-13 176 5.49e-14 100
HFC-245cb CF
3
CF
2
CH
3
47.1 0.24 1.67e-10 6680 4.24e-10 4620 4.58e-12 6690 3.00e-12 4870 1.32e-12 2410
HFC-245ea CHF
2
CHFCHF
2
3.2 0.16
c
2.15e-11 863 2.16e-11 235 2.59e-13 378 2.70e-14 44 1.79e-14 33
HFC-245eb CH
2
FCHFCF
3
3.1 0.20
c
2.66e-11 1070 2.66e-11 290 3.15e-13 460 3.31e-14 54 2.20e-14 40
HFC-245fa CHF
2
CH
2
CF
3
7.7 0.24 7.29e-11 2920 7.87e-11 858 1.35e-12 1970 1.51e-13 245 6.62e-14 121
HFC-263fb CH
3
CH
2
CF
3
1.2 0.10
c
6.93e-12 278 6.93e-12 76 6.31e-14 92 8.02e-15 13 5.70e-15 10
HFC-272ca CH
3
CF
2
CH
3
2.6 0.07 1.32e-11 530 1.32e-11 144 1.46e-13 213 1.61e-14 26 1.09e-14 20
HFC-329p CHF
2
CF
2
CF
2
CF
3
28.4 0.31 1.13e-10 4510 2.16e-10 2360 2.94e-12 4290 1.37e-12 2220 3.96e-13 725
HFC-365mfc CH
3
CF
2
CH
2
CF
3
8.7 0.22 6.64e-11 2660 7.38e-11 804 1.30e-12 1890 1.62e-13 262 6.24e-14 114
HFC-43-10mee CF
3
CHFCHFCF
2
CF
3
16.1 0.42
b
1.08e-10 4310 1.51e-10 1650 2.54e-12 3720 6.62e-13 1070 1.54e-13 281
HFC-1132a CH
2
=CF
2
4.0 days 0.004
d
3.87e-15 <1 3.87e-15 <1 3.08e-17 <1 4.35e-18 <1 3.18e-18 <1
HFC-1141 CH
2
=CHF 2.1 days 0.002
d
1.54e-15 <1 1.54e-15 <1 1.23e-17 <1 1.73e-18 <1 1.27e-18 <1
(Z)-HFC-1225ye CF
3
CF=CHF(Z) 8.5 days 0.02 2.14e-14 <1 2.14e-14 <1 1.70e-16 <1 2.40e-17 <1 1.76e-17 <1
(E)-HFC-1225ye CF
3
CF=CHF(E) 4.9 days 0.01 7.25e-15 <1 7.25e-15 <1 5.77e-17 <1 8.14e-18 <1 5.95e-18 <1
(Z)-HFC-1234ze CF
3
CH=CHF(Z) 10.0 days 0.02 2.61e-14 1 2.61e-14 <1 2.08e-16 <1 2.93e-17 <1 2.14e-17 <1
HFC-1234yf CF
3
CF=CH
2
10.5 days 0.02 3.22e-14 1 3.22e-14 <1 2.57e-16 <1 3.62e-17 <1 2.65e-17 <1
(E)-HFC-1234ze trans-CF
3
CH=CHF 16.4 days 0.04 8.74e-14 4 8.74e-14 <1 6.98e-16 <1 9.82e-17 <1 7.18e-17 <1
(Z)-HFC-1336 CF
3
CH=CHCF
3
(Z) 22.0 days 0.07
d
1.54e-13 6 1.54e-13 2 1.23e-15 2 1.73e-16 <1 1.26e-16 <1
8
Anthropogenic and Natural Radiative Forcing Chapter 8
733
(continued on next page)
Table 8.A.1 (continued)
Acronym, Common Name or Chemical Name Chemical Formula
Lifetime
(Years)
Radia-
tive
Effi-
ciency
(W m
–2
ppb
–1
)
AGWP
20-year
(W m
–2
yr kg
–1
)
GWP
20-year
AGWP
100-year
(W m
–2
yr kg
–1
)
GWP
100-year
AGTP
20-year
(K kg
–1
)
GTP
20-year
AGTP
50-year
(K kg
–1
)
GTP
50-year
AGTP
100-year
(K kg
–1
)
GTP
100-year
HFC-1243zf CF
3
CH=CH
2
7.0 days 0.01 1.37e-14 1 1.37e-14 <1 1.09e-16 <1 1.53e-17 <1 1.12e-17 <1
HFC-1345zfc C
2
F
5
CH=CH
2
7.6 days 0.01 1.15e-14 <1 1.15e-14 <1 9.19e-17 <1 1.30e-17 <1 9.48e-18 <1
3,3,4,4,5,5,6,6,6-Nonafluorohex-1-ene C
4
F
9
CH=CH
2
7.6 days 0.03 1.25e-14 <1 1.25e-14 <1 9.92e-17 <1 1.40e-17 <1 1.02e-17 <1
3,3,4,4,5,5,6,6,7,7,8,8,8-Tridecafluorooct-1-ene C
6
F
13
CH=CH
2
7.6 days 0.03 9.89e-15 <1 9.89e-15 <1 7.87e-17 <1 1.11e-17 <1 8.12e-18 <1
3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,10-Hep-
tadecafluorodec-1-ene
C
8
F
17
CH=CH
2
7.6 days 0.03 8.52e-15 <1 8.52e-15 <1 6.79e-17 <1 9.57e-18 <1 7.00e-18 <1
Chlorocarbons and Hydrochlorocarbons
Methyl chloroform CH
3
CCl
3
5.0 0.07 1.44e-11 578 1.47e-11 160 2.17e-13 317 2.07e-14 34 1.22e-14 22
Carbon tetrachloride CCl
4
26.0 0.17 8.69e-11 3480 1.59e-10 1730 2.24e-12 3280 9.68e-13 1570 2.62e-13 479
Methyl chloride CH
3
Cl 1.0 0.01
a
1.12e-12 45 1.12e-12 12 9.93e-15 15 1.29e-15 2 9.20e-16 2
Methylene chloride CH
2
Cl
2
0.4 0.03
b
8.18e-13 33 8.18e-13 9 6.78e-15 10 9.26e-16 2 6.72e-16 1
Chloroform CHCl
3
0.4 0.08 1.50e-12 60 1.50e-12 16 1.25e-14 18 1.70e-15 3 1.24e-15 2
1,2-Dichloroethane CH
2
ClCH
2
Cl 65.0 days 0.01 8.24e-14 3 8.24e-14 <1 6.67e-16 <1 9.29e-17 <1 6.77e-17 <1
Bromocarbons, Hydrobromocarbons and Halons
Methyl bromide CH
3
Br 0.8 0.004 2.16e-13 9 2.16e-13 2 1.87e-15 3 2.47e-16 <1 1.78e-16 <1
Methylene bromide CH
2
Br
2
0.3 0.01 9.31e-14 4 9.31e-14 1 7.66e-16 1 1.05e-16 <1 7.65e-17 <1
Halon-1201 CHBrF
2
5.2 0.15 3.37e-11 1350 3.45e-11 376 5.17e-13 756 4.96e-14 80 2.87e-14 52
Halon-1202 CBr
2
F
2
2.9 0.27 2.12e-11 848 2.12e-11 231 2.43e-13 356 2.61e-14 42 1.75e-14 32
Halon-1211 CBrClF
2
16.0 0.29 1.15e-10 4590 1.60e-10 1750 2.70e-12 3950 6.98e-13 1130 1.62e-13 297
Halon-1301 CBrF
3
65.0 0.30 1.95e-10 7800 5.77e-10 6290 5.46e-12 7990 4.16e-12 6750 2.28e-12 4170
Halon-2301 CH
2
BrCF
3
3.4 0.14 1.59e-11 635 1.59e-11 173 1.96e-13 286 2.01e-14 33 1.32e-14 24
Halon-2311 / Halothane CHBrClCF
3
1.0 0.13 3.77e-12 151 3.77e-12 41 3.35e-14 49 4.34e-15 7 3.10e-15 6
Halon-2401 CHFBrCF
3
2.9 0.19 1.68e-11 674 1.68e-11 184 1.94e-13 283 2.07e-14 34 1.39e-14 25
Halon-2402 CBrF
2
CBrF
2
20.0 0.31 8.59e-11 3440 1.35e-10 1470 2.12e-12 3100 7.08e-13 1150 1.66e-13 304
Fully Fluorinated Species
Nitrogen trifluoride NF
3
500.0 0.20 3.19e-10 12,800 1.47e-09 16,100 9.39e-12 13,700 1.02e-11 16,500 9.91e-12 18,100
Sulphur hexafluoride SF
6
3,200.0 0.57 4.37e-10 17,500 2.16e-09 23,500 1.29e-11 18,900 1.47e-11 23,800 1.54e-11 28,200
(Trifluoromethyl) sulphur pentafluoride SF
5
CF
3
800.0 0.59 3.36e-10 13,500 1.60e-09 17,400 9.93e-12 14,500 1.10e-11 17,800 1.11e-11 20,200
Sulphuryl fluoride SO
2
F
2
36.0 0.20 1.71e-10 6840 3.76e-10 4090 4.58e-12 6690 2.55e-12 4140 9.01e-13 1650
PFC-14 CF
4
50,000.0 0.09 1.22e-10 4880 6.08e-10 6630 3.61e-12 5270 4.12e-12 6690 4.40e-12 8040
PFC-116 C
2
F
6
10,000.0 0.25 2.05e-10 8210 1.02e-09 11,100 6.07e-12 8880 6.93e-12 11,200 7.36e-12 13,500
PFC-c216 c-C
3
F
6
3,000.0 0.23
e
1.71e-10 6850 8.44e-10 9200 5.06e-12 7400 5.74e-12 9310 6.03e-12 11,000
PFC-218 C
3
F
8
2,600.0 0.28 1.66e-10 6640 8.16e-10 8900 4.91e-12 7180 5.56e-12 9010 5.83e-12 10,700
PFC-318 c-C
4
F
8
3,200.0 0.32 1.77e-10 7110 8.75e-10 9540 5.25e-12 7680 5.96e-12 9660 6.27e-12 11,500
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
734
Acronym, Common Name or Chemical Name Chemical Formula
Lifetime
(Years)
Radia-
tive
Effi-
ciency
(W m
–2
ppb
–1
)
AGWP
20-year
(W m
–2
yr kg
–1
)
GWP
20-year
AGWP
100-year
(W m
–2
yr kg
–1
)
GWP
100-year
AGTP
20-year
(K kg
–1
)
GTP
20-year
AGTP
50-year
(K kg
–1
)
GTP
50-year
AGTP
100-year
(K kg
–1
)
GTP
100-year
PFC-31-10 C
4
F
10
2,600.0 0.36 1.71e-10 6870 8.44e-10 9200 5.08e-12 7420 5.75e-12 9320 6.02e-12 11,000
Perfluorocyclopentene c-C
5
F
8
31.0 days 0.08
f
1.71e-13 7 1.71e-13 2 1.37e-15 2 1.92e-16 <1 1.40e-16 <1
PFC-41-12 n-C
5
F
12
4,100.0 0.41 1.58e-10 6350 7.84e-10 8550 4.69e-12 6860 5.33e-12 8650 5.62e-12 10,300
PFC-51-14 n-C
6
F
14
3,100.0 0.44 1.47e-10 5890 7.26e-10 7910 4.35e-12 6370 4.94e-12 8010 5.19e-12 9490
PFC-61-16 n-C
7
F
16
3,000.0 0.50 1.45e-10 5830 7.17e-10 7820 4.31e-12 6290 4.88e-12 7920 5.13e-12 9380
PFC-71-18 C
8
F
18
3,000.0 0.55 1.42e-10 5680 6.99e-10 7620 4.20e-12 6130 4.76e-12 7710 5.00e-12 9140
PFC-91-18 C
10
F
18
2,000.0 0.55 1.34e-10 5390 6.59e-10 7190 3.98e-12 5820 4.49e-12 7290 4.68e-12 8570
Perfluorodecalin (cis) Z-C
10
F
18
2,000.0 0.56 1.35e-10 5430 6.64e-10 7240 4.01e-12 5860 4.52e-12 7340 4.72e-12 8630
Perfluorodecalin (trans) E-C
10
F
18
2,000.0 0.48 1.18e-10 4720 5.77e-10 6290 3.48e-12 5090 3.93e-12 6380 4.10e-12 7500
PFC-1114 CF
2
=CF
2
1.1 days 0.002 2.68e-16 <1 2.68e-16 <1 2.13e-18 <1 3.00e-19 <1 2.20e-19 <1
PFC-1216 CF
3
CF=CF
2
4.9 days 0.01 6.42e-15 <1 6.42e-15 <1 5.11e-17 <1 7.21e-18 <1 5.27e-18 <1
Perfluorobuta-1,3-diene CF
2
=CFCF=CF
2
1.1 days 0.003 3.29e-16 <1 3.29e-16 <1 2.61e-18 <1 3.69e-19 <1 2.70e-19 <1
Perfluorobut-1-ene CF
3
CF
2
CF=CF
2
6.0 days 0.02 8.38e-15 <1 8.38e-15 <1 6.67e-17 <1 9.41e-18 <1 6.88e-18 <1
Perfluorobut-2-ene CF
3
CF=CFCF
3
31.0 days 0.07 1.62e-13 6 1.62e-13 2 1.30e-15 2 1.82e-16 <1 1.33e-16 <1
Halogenated Alcohols and Ethers
HFE-125 CHF
2
OCF
3
119.0 0.41 3.10e-10 12,400 1.14e-09 12,400 8.91e-12 13,000 8.14e-12 13,200 5.97e-12 10,900
HFE-134 (HG-00) CHF
2
OCHF
2
24.4 0.44 2.90e-10 11,600 5.10e-10 5560 7.42e-12 10,800 3.02e-12 4900 7.83e-13 1430
HFE-143a CH
3
OCF
3
4.8 0.18 4.72e-11 1890 4.80e-11 523 6.95e-13 1020 6.66e-14 108 3.99e-14 73
HFE-227ea CF
3
CHFOCF
3
51.6 0.44 2.22e-10 8900 5.92e-10 6450 6.15e-12 8980 4.22e-12 6850 1.98e-12 3630
HCFE-235ca2 (enflurane) CHF
2
OCF
2
CHFCl 4.3 0.41 5.30e-11 2120 5.35e-11 583 7.36e-13 1080 7.14e-14 116 4.44e-14 81
HCFE-235da2 (isoflurane) CHF
2
OCHClCF
3
3.5 0.42 4.49e-11 1800 4.50e-11 491 5.62e-13 822 5.72e-14 93 3.73e-14 68
HFE-236ca CHF
2
OCF
2
CHF
2
20.8 0.56
g
2.42e-10 9710 3.89e-10 4240 6.03e-12 8820 2.10e-12 3400 4.98e-13 912
HFE-236ea2 (desflurane) CHF
2
OCHFCF
3
10.8 0.45 1.39e-10 5550 1.64e-10 1790 2.93e-12 4280 4.64e-13 753 1.42e-13 260
HFE-236fa CF
3
CH
2
OCF
3
7.5 0.36 8.35e-11 3350 8.98e-11 979 1.53e-12 2240 1.68e-13 273 7.54e-14 138
HFE-245cb2 CF
3
CF
2
OCH
3
4.9 0.33 5.90e-11 2360 6.00e-11 654 8.77e-13 1280 8.40e-14 136 4.99e-14 91
HFE-245fa1 CHF
2
CH
2
OCF
3
6.6 0.31 7.22e-11 2900 7.59e-11 828 1.25e-12 1820 1.27e-13 206 6.35e-14 116
HFE-245fa2 CHF
2
OCH
2
CF
3
5.5 0.36 7.25e-11 2910 7.45e-11 812 1.15e-12 1670 1.10e-13 179 6.21e-14 114
2,2,3,3,3-Pentafluoropropan-1-ol CF
3
CF
2
CH
2
OH 0.3 0.14 1.72e-12 69 1.72e-12 19 1.42e-14 21 1.95e-15 3 1.42e-15 3
HFE-254cb1 CH
3
OCF
2
CHF
2
2.5 0.26 2.76e-11 1110 2.76e-11 301 2.99e-13 438 3.34e-14 54 2.28e-14 42
HFE-263fb2 CF
3
CH
2
OCH
3
23.0 days 0.04 1.22e-13 5 1.22e-13 1 9.72e-16 1 1.37e-16 <1 9.98e-17 <1
HFE-263m1 CF
3
OCH
2
CH
3
0.4 0.13 2.70e-12 108 2.70e-12 29 2.25e-14 33 3.06e-15 5 2.22e-15 4
3,3,3-Trifluoropropan-1-ol CF
3
CH
2
CH
2
OH 12.0 days 0.02 3.57e-14 1 3.57e-14 <1 2.85e-16 <1 4.01e-17 <1 2.93e-17 <1
HFE-329mcc2 CHF
2
CF
2
OCF
2
CF
3
22.5 0.53 1.68e-10 6720 2.81e-10 3070 4.23e-12 6180 1.59e-12 2580 3.93e-13 718
HFE-338mmz1 (CF
3
)
2
CHOCHF
2
21.2 0.44 1.48e-10 5940 2.40e-10 2620 3.70e-12 5410 1.31e-12 2130 3.14e-13 575
(continued on next page)
Table 8.A.1 (continued)
8
Anthropogenic and Natural Radiative Forcing Chapter 8
735
Acronym, Common Name or Chemical Name Chemical Formula
Lifetime
(Years)
Radia-
tive
Effi-
ciency
(W m
–2
ppb
–1
)
AGWP
20-year
(W m
–2
yr kg
–1
)
GWP
20-year
AGWP
100-year
(W m
–2
yr kg
–1
)
GWP
100-year
AGTP
20-year
(K kg
–1
)
GTP
20-year
AGTP
50-year
(K kg
–1
)
GTP
50-year
AGTP
100-year
(K kg
–1
)
GTP
100-year
HFE-338mcf2 CF
3
CH
2
OCF
2
CF
3
7.5 0.44 7.93e-11 3180 8.52e-11 929 1.45e-12 2120 1.60e-13 259 7.16e-14 131
Sevoflurane (HFE-347mmz1) (CF
3
)
2
CHOCH
2
F 2.2 0.32 1.98e-11 795 1.98e-11 216 2.06e-13 302 2.37e-14 38 1.64e-14 30
HFE-347mcc3 (HFE-7000) CH
3
OCF
2
CF
2
CF
3
5.0 0.35 4.78e-11 1910 4.86e-11 530 7.18e-13 1050 6.87e-14 111 4.05e-14 74
HFE-347mcf2 CHF
2
CH
2
OCF
2
CF
3
6.6 0.42 7.45e-11 2990 7.83e-11 854 1.29e-12 1880 1.31e-13 212 6.55e-14 120
HFE-347pcf2 CHF
2
CF
2
OCH
2
CF
3
6.0 0.48
h
7.86e-11 3150 8.15e-11 889 1.30e-12 1900 1.27e-13 206 6.81e-14 124
HFE-347mmy1 (CF
3
)
2
CFOCH
3
3.7 0.32 3.32e-11 1330 3.33e-11 363 4.27e-13 624 4.28e-14 69 2.76e-14 51
HFE-356mec3 CH
3
OCF
2
CHFCF
3
3.8 0.30 3.53e-11 1410 3.55e-11 387 4.60e-13 673 4.58e-14 74 2.94e-14 54
HFE-356mff2 CF
3
CH
2
OCH
2
CF
3
105.0
days
0.17 1.54e-12 62 1.54e-12 17 1.26e-14 18 1.74e-15 3 1.26e-15 2
HFE-356pcf2 CHF
2
CH
2
OCF
2
CHF
2
5.7 0.37 6.40e-11 2560 6.59e-11 719 1.03e-12 1500 9.97e-14 162 5.50e-14 101
HFE-356pcf3 CHF
2
OCH
2
CF
2
CHF
2
3.5 0.38 4.08e-11 1640 4.09e-11 446 5.11e-13 747 5.20e-14 84 3.39e-14 62
HFE-356pcc3 CH
3
OCF
2
CF
2
CHF
2
3.8 0.32 3.77e-11 1510 3.79e-11 413 4.91e-13 718 4.89e-14 79 3.14e-14 57
HFE-356mmz1 (CF
3
)
2
CHOCH
3
97.1 days 0.15 1.25e-12 50 1.25e-12 14 1.02e-14 15 1.41e-15 2 1.02e-15 2
HFE-365mcf3 CF
3
CF
2
CH
2
OCH
3
19.3 days 0.05 8.51e-14 3 8.51e-14 <1 6.80e-16 <1 9.56e-17 <1 6.99e-17 <1
HFE-365mcf2 CF
3
CF
2
OCH
2
CH
3
0.6 0.26
i
5.35e-12 215 5.35e-12 58 4.53e-14 66 6.10e-15 10 4.40e-15 8
HFE-374pc2 CHF
2
CF
2
OCH
2
CH
3
5.0 0.30 5.65e-11 2260 5.75e-11 627 8.48e-13 1240 8.12e-14 132 4.79e-14 88
4,4,4-Trifluorobutan-1-ol CF
3
(CH
2
)
2
CH
2
OH 4.0 days 0.01 1.73e-15 <1 1.73e-15 <1 1.38e-17 <1 1.94e-18 <1 1.42e-18 <1
2,2,3,3,4,4,5,5-Octafluorocyclopentanol -(CF
2
)
4
CH(OH)- 0.3 0.16 1.18e-12 47 1.18e-12 13 9.67e-15 14 1.33e-15 2 9.69e-16 2
HFE-43-10pccc124 (H-Galden 1040x, HG-11) CHF
2
OCF
2
OC
2
F
4
OCHF
2
13.5 1.02 2.00e-10 8010 2.58e-10 2820 4.52e-12 6600 9.46e-13 1530 2.38e-13 436
HFE-449s1 (HFE-7100) C
4
F
9
OCH
3
4.7 0.36 3.80e-11 1530 3.86e-11 421 5.54e-13 809 5.32e-14 86 3.21e-14 59
n-HFE-7100 n-C
4
F
9
OCH
3
4.7 0.42 4.39e-11 1760 4.45e-11 486 6.39e-13 934 6.14e-14 99 3.70e-14 68
i-HFE-7100 i-C
4
F
9
OCH
3
4.7 0.35 3.68e-11 1480 3.73e-11 407 5.35e-13 783 5.14e-14 83 3.10e-14 57
HFE-569sf2 (HFE-7200) C
4
F
9
OC
2
H
5
0.8 0.30 5.21e-12 209 5.21e-12 57 4.52e-14 66 5.97e-15 10 4.29e-15 8
n-HFE-7200 n-C
4
F
9
OC
2
H
5
0.8 0.35
i
5.92e-12 237 5.92e-12 65 5.14e-14 75 6.78e-15 11 4.87e-15 9
i-HFE-7200 i-C
4
F
9
OC
2
H
5
0.8 0.24 4.06e-12 163 4.06e-12 44 3.52e-14 52 4.65e-15 8 3.34e-15 6
HFE-236ca12 (HG-10) CHF
2
OCF
2
OCHF
2
25.0 0.65 2.75e-10 11,000 4.91e-10 5350 7.06e-12 10,300 2.94e-12 4770 7.75e-13 1420
HFE-338pcc13 (HG-01) CHF
2
OCF
2
CF
2
OCHF
2
12.9 0.86 2.10e-10 8430 2.67e-10 2910 4.69e-12 6860 9.28e-13 1500 2.42e-13 442
1,1,1,3,3,3-Hexafluoropropan-2-ol (CF
3
)
2
CHOH 1.9 0.26 1.67e-11 668 1.67e-11 182 1.66e-13 243 1.97e-14 32 1.38e-14 25
HG-02
HF
2
C–(OCF
2
CF
2
)
2
–
OCF
2
H
12.9 1.24
i
1.97e-10 7900 2.50e-10 2730 4.40e-12 6430 8.70e-13 1410 2.27e-13 415
HG-03
HF
2
C–(OCF
2
CF
2
)
3
–
OCF
2
H
12.9 1.76
i
2.06e-10 8270 2.62e-10 2850 4.60e-12 6730 9.10e-13 1480 2.37e-13 434
HG-20 HF
2
C–(OCF
2
)
2
–OCF
2
H 25.0 0.92
i
2.73e-10 10,900 4.86e-10 5300 7.00e-12 10,200 2.91e-12 4730 7.68e-13 1400
HG-21
HF
2
C–OCF
2
CF
2
OC-
F
2
OCF
2
O–CF
2
H
13.5 1.71
i
2.76e-10 11,100 3.57e-10 3890 6.23e-12 9110 1.31e-12 2120 3.29e-13 602
(continued on next page)
Table 8.A.1 (continued)
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
736
Acronym, Common Name or Chemical Name Chemical Formula
Lifetime
(Years)
Radia-
tive
Effi-
ciency
(W m
–2
ppb
–1
)
AGWP
20-year
(W m
–2
yr kg
–1
)
GWP
20-year
AGWP
100-year
(W m
–2
yr kg
–1
)
GWP
100-year
AGTP
20-year
(K kg
–1
)
GTP
20-year
AGTP
50-year
(K kg
–1
)
GTP
50-year
AGTP
100-year
(K kg
–1
)
GTP
100-year
HG-30 HF
2
C–(OCF
2
)
3
–OCF
2
H 25.0 1.65
i
3.77e-10 15,100 6.73e-10 7330 9.68e-12 14,100 4.03e-12 6530 1.06e-12 1940
1-Ethoxy-1,1,2,2,3,3,3-heptafluoropropane CF
3
CF
2
CF
2
OCH
2
CH
3
0.8 0.28
i
5.56e-12 223 5.56e-12 61 4.80e-14 70 6.36e-15 10 4.57e-15 8
Fluoroxene CF
3
CH
2
OCH=CH
2
3.6 days 0.01
i
4.97e-15 <1 4.97e-15 <1 3.95e-17 <1 5.58e-18 <1 4.08e-18 <1
1,1,2,2-Tetrafluoro-1-(fluoromethoxy)ethane CH
2
FOCF
2
CF
2
H 6.2 0.34
i
7.68e-11 3080 7.99e-11 871 1.29e-12 1880 1.28e-13 207 6.68e-14 122
2-Ethoxy-3,3,4,4,5-pentafluorotetrahydro-2,5-bis[1,2,2,2-
tetrafluoro-1-(trifluoromethyl)ethyl]-furan
C
12
H
5
F
19
O
2
1.0 0.49
j
5.09e-12 204 5.09e-12 56 4.53e-14 66 5.86e-15 10 4.19e-15 8
Fluoro(methoxy)methane CH
3
OCH
2
F 73.0 days 0.07
g
1.15e-12 46 1.15e-12 13 9.34e-15 14 1.30e-15 2 9.46e-16 2
Difluoro(methoxy)methane CH
3
OCHF
2
1.1 0.17
g
1.32e-11 528 1.32e-11 144 1.18e-13 173 1.52e-14 25 1.08e-14 20
Fluoro(fluoromethoxy)methane CH
2
FOCH
2
F 0.9 0.19
g
1.20e-11 479 1.20e-11 130 1.05e-13 153 1.37e-14 22 9.84e-15 18
Difluoro(fluoromethoxy)methane CH
2
FOCHF
2
3.3 0.30
g
5.65e-11 2260 5.66e-11 617 6.88e-13 1010 7.11e-14 115 4.69e-14 86
Trifluoro(fluoromethoxy)methane CH
2
FOCF
3
4.4 0.33
g
6.82e-11 2730 6.89e-11 751 9.59e-13 1400 9.27e-14 150 5.72e-14 105
HG’-01 CH
3
OCF
2
CF
2
OCH
3
2.0 0.29 2.03e-11 815 2.03e-11 222 2.06e-13 301 2.42e-14 39 1.68e-14 31
HG’-02 CH
3
O(CF
2
CF
2
O)
2
CH
3
2.0 0.56 2.16e-11 868 2.16e-11 236 2.19e-13 320 2.57e-14 42 1.79e-14 33
HG’-03 CH
3
O(CF
2
CF
2
O)
3
CH
3
2.0 0.76 2.03e-11 812 2.03e-11 221 2.05e-13 299 2.41e-14 39 1.67e-14 31
HFE-329me3 CF
3
CFHCF
2
OCF
3
40.0 0.48 1.79e-10 7170 4.17e-10 4550 4.85e-12 7090 2.89e-12 4690 1.12e-12 2040
3,3,4,4,5,5,6,6,7,7,7-Undecafluoroheptan-1-ol CF
3
(CF
2
)
4
CH
2
CH
2
OH 20.0 days 0.06 3.91e-14 2 3.91e-14 <1 3.12e-16 <1 4.39e-17 <1 3.21e-17 <1
3,3,4,4,5,5,6,6,7,7,8,8,9,9,9-Pentadecafluorononan-1-ol CF
3
(CF
2
)
6
CH
2
CH
2
OH 20.0 days 0.07 3.00e-14 1 3.00e-14 <1 2.40e-16 <1 3.37e-17 <1 2.46e-17 <1
3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-Non-
adecafluoroundecan-1-ol
CF
3
(CF
2
)
8
CH
2
CH
2
OH 20.0 days 0.05 1.72e-14 <1 1.72e-14 <1 1.37e-16 <1 1.93e-17 <1 1.41e-17 <1
2-Chloro-1,1,2-trifluoro-1-methoxyethane CH
3
OCF
2
CHFCl 1.4 0.21 1.12e-11 449 1.12e-11 122 1.05e-13 153 1.31e-14 21 9.24e-15 17
PFPMIE (perfluoropolymethylisopropyl ether)
CF
3
OCF(CF
3
)
CF
2
OCF
2
OCF
3
800.0 0.65 1.87e-10 7500 8.90e-10 9710 5.52e-12 8070 6.11e-12 9910 6.15e-12 11,300
HFE-216 CF
3
OCF=CF
2
8.4 days 0.02 1.92e-14 <1 1.92e-14 <1 1.53e-16 <1 2.15e-17 <1 1.58e-17 <1
Trifluoromethyl formate HCOOCF
3
3.5 0.31
i
5.37e-11 2150 5.39e-11 588 6.73e-13 984 6.85e-14 111 4.47e-14 82
Perfluoroethyl formate HCOOCF
2
CF
3
3.5 0.44
i
5.30e-11 2130 5.32e-11 580 6.64e-13 971 6.76e-14 110 4.41e-14 81
Perfluoropropyl formate HCOOCF
2
CF
2
CF
3
2.6 0.50
i
3.45e-11 1380 3.45e-11 376 3.80e-13 555 4.19e-14 68 2.85e-14 52
Perfluorobutyl formate HCOOCF
2
CF
2
CF
2
CF
3
3.0 0.56
i
3.59e-11 1440 3.59e-11 392 4.19e-13 613 4.45e-14 72 2.97e-14 54
2,2,2-Trifluoroethyl formate HCOOCH
2
CF
3
0.4 0.16
i
3.07e-12 123 3.07e-12 33 2.55e-14 37 3.48e-15 6 2.52e-15 5
3,3,3-Trifluoropropyl formate HCOOCH
2
CH
2
CF
3
0.3 0.13
i
1.60e-12 64 1.60e-12 17 1.31e-14 19 1.80e-15 3 1.31e-15 2
1,2,2,2-Tetrafluoroethyl formate HCOOCHFCF
3
3.2 0.35
i
4.30e-11 1720 4.31e-11 470 5.17e-13 755 5.39e-14 87 3.57e-14 65
1,1,1,3,3,3-Hexafluoropropan-2-yl formate HCOOCH(CF
3
)
2
3.2 0.33
i
3.05e-11 1220 3.05e-11 333 3.66e-13 535 3.81e-14 62 2.53e-14 46
Perfluorobutyl acetate CH
3
COOCF
2
CF
2
CF
2
CF
3
21.9 days 0.12
i
1.52e-13 6 1.52e-13 2 1.21e-15 2 1.71e-16 <1 1.25e-16 <1
Perfluoropropyl acetate CH
3
COOCF
2
CF
2
CF
3
21.9 days 0.11
i
1.59e-13 6 1.59e-13 2 1.27e-15 2 1.78e-16 <1 1.30e-16 <1
Perfluoroethyl acetate CH
3
COOCF
2
CF
3
21.9 days 0.10
i
1.89e-13 8 1.89e-13 2 1.51e-15 2 2.12e-16 <1 1.55e-16 <1
Trifluoromethyl acetate CH
3
COOCF
3
21.9 days 0.07
i
1.90e-13 8 1.90e-13 2 1.52e-15 2 2.14e-16 <1 1.56e-16 <1
(continued on next page)
Table 8.A.1 (continued)
8
Anthropogenic and Natural Radiative Forcing Chapter 8
737
Acronym, Common Name or Chemical Name Chemical Formula
Lifetime
(Years)
Radia-
tive
Effi-
ciency
(W m
–2
ppb
–1
)
AGWP
20-year
(W m
–2
yr kg
–1
)
GWP
20-year
AGWP
100-year
(W m
–2
yr kg
–1
)
GWP
100-year
AGTP
20-year
(K kg
–1
)
GTP
20-year
AGTP
50-year
(K kg
–1
)
GTP
50-year
AGTP
100-year
(K kg
–1
)
GTP
100-year
Methyl carbonofluoridate FCOOCH
3
1.8 0.07
i
8.74e-12 350 8.74e-12 95 8.60e-14 126 1.03e-14 17 7.21e-15 13
1,1-Difluoroethyl carbonofluoridate FCOOCF
2
CH
3
0.3 0.17
i
2.46e-12 99 2.46e-12 27 2.02e-14 30 2.78e-15 5 2.02e-15 4
1,1-Difluoroethyl 2,2,2-trifluoroacetate CF
3
COOCF
2
CH
3
0.3 0.27
i
2.83e-12 113 2.83e-12 31 2.33e-14 34 3.20e-15 5 2.32e-15 4
Ethyl 2,2,2-trifluoroacetate CF
3
COOCH
2
CH
3
21.9 days 0.05
i
1.26e-13 5 1.26e-13 1 1.00e-15 1 1.41e-16 <1 1.03e-16 <1
2,2,2-Trifluoroethyl 2,2,2-trifluoroacetate CF
3
COOCH
2
CF
3
54.8 days 0.15
i
6.27e-13 25 6.27e-13 7 5.06e-15 7 7.07e-16 1 5.15e-16 <1
Methyl 2,2,2-trifluoroacetate CF
3
COOCH
3
0.6 0.18
i
4.80e-12 192 4.80e-12 52 4.08e-14 60 5.47e-15 9 3.95e-15 7
Methyl 2,2-difluoroacetate HCF
2
COOCH
3
40.1 days 0.05
i
3.00e-13 12 3.00e-13 3 2.41e-15 4 3.38e-16 <1 2.47e-16 <1
Difluoromethyl 2,2,2-trifluoroacetate CF
3
COOCHF
2
0.3 0.24
i
2.48e-12 99 2.48e-12 27 2.04e-14 30 2.81e-15 5 2.04e-15 4
2,2,3,3,4,4,4-Heptafluorobutan-1-ol C
3
F
7
CH
2
OH 0.6 0.20 3.10e-12 124 3.10e-12 34 2.61e-14 38 3.52e-15 6 2.55e-15 5
1,1,2-Trifluoro-2-(trifluoromethoxy)-ethane CHF
2
CHFOCF
3
9.8 0.35 9.91e-11 3970 1.14e-10 1240 2.03e-12 2960 2.88e-13 467 9.74e-14 178
1-Ethoxy-1,1,2,3,3,3-hexafluoropropane CF
3
CHFCF
2
OCH
2
CH
3
0.4 0.19 2.14e-12 86 2.14e-12 23 1.77e-14 26 2.43e-15 4 1.76e-15 3
1,1,1,2,2,3,3-Heptafluoro-3-(1,2,2,2-
tetrafluoroethoxy)-propane
CF
3
CF
2
CF
2
OCHFCF
3
67.0 0.58 1.98e-10 7940 5.95e-10 6490 5.57e-12 8140 4.29e-12 6960 2.39e-12 4380
2,2,3,3-Tetrafluoro-1-propanol CHF
2
CF
2
CH
2
OH 91.3 days 0.11 1.19e-12 48 1.19e-12 13 9.72e-15 14 1.35e-15 2 9.79e-16 2
2,2,3,4,4,4-Hexafluoro-1-butanol CF
3
CHFCF
2
CH
2
OH 94.9 days 0.19 1.56e-12 63 1.56e-12 17 1.27e-14 19 1.76e-15 3 1.28e-15 2
2,2,3,3,4,4,4-Heptafluoro-1-butanol CF
3
CF
2
CF
2
CH
2
OH 0.3 0.16 1.49e-12 60 1.49e-12 16 1.23e-14 18 1.69e-15 3 1.23e-15 2
1,1,2,2-Tetrafluoro-3-methoxy-propane CHF
2
CF
2
CH
2
OCH
3
14.2 days 0.03 4.82e-14 2 4.82e-14 <1 3.84e-16 <1 5.41e-17 <1 3.96e-17 <1
perfluoro-2-methyl-3-pentanone CF
3
CF
2
C(O)CF(CF
3
)
2
7.0 days 0.03 9.14e-15 <1 9.14e-15 <1 7.27e-17 <1 1.03e-17 <1 7.51e-18 <1
3,3,3-Trifluoro-propanal CF
3
CH
2
CHO 2.0 days 0.004 9.86e-16 <1 9.86e-16 <1 7.84e-18 <1 1.11e-18 <1 8.10e-19 <1
2-Fluoroethanol CH
2
FCH
2
OH 20.4 days 0.02 8.07e-14 3 8.07e-14 <1 6.45e-16 <1 9.07e-17 <1 6.63e-17 <1
2,2-Difluoroethanol CHF
2
CH
2
OH 40.0 days 0.04 2.78e-13 11 2.78e-13 3 2.23e-15 3 3.12e-16 <1 2.28e-16 <1
2,2,2-Trifluoroethanol CF
3
CH
2
OH 0.3 0.10 1.83e-12 73 1.83e-12 20 1.50e-14 22 2.07e-15 3 1.50e-15 3
1,1’-Oxybis[2-(difluoromethoxy)-1,1,2,2-tetrafluoroethane HCF
2
O(CF
2
CF
2
O)
2
CF
2
H 26.0 1.15
k
2.47e-10 9910 4.51e-10 4920 6.38e-12 9320 2.75e-12 4460 7.45e-13 1360
1,1,3,3,4,4,6,6,7,7,9,9,10,10,12,12-hexa-
decafluoro-2,5,8,11-Tetraoxadodecane
HCF
2
O(CF
2
CF
2
O)
3
CF
2
H 26.0 1.43
k
2.26e-10 9050 4.12e-10 4490 5.83e-12 8520 2.51e-12 4080 6.81e-13 1250
1,1,3,3,4,4,6,6,7,7,9,9,10,10,12,12,13,13,15,15-eico-
safluoro-2,5,8,11,14-Pentaoxapentadecane
HCF
2
O(CF
2
CF
2
O)
4
CF
2
H 26.0 1.46
k
1.83e-10 7320 3.33e-10 3630 4.71e-12 6880 2.03e-12 3300 5.50e-13 1010
Table 8.A.1 (continued)
Notes:
For CH
4
we estimate an uncertainty of ±30% and ±40% for 20- and 100-year time horizon, respectively (for 90% uncertainty range). The uncertainty is dominated by AGWP for CO
2
and indirect effects. The uncertainty in GWP for N
2
O is estimated to ±20% and
±30% for 20- and 100-year time horizon, with the largest contributions from CO
2
. The uncertainty in GWP for HFC-134a is estimated to ±25% and ±35% for 20- and 100-year time horizons while for CFC-11 the GWP the corresponding numbers are approximately
±20% and ±35% (not accounting for the indirect effects). For CFC-12 the corresponding numbers are ±20 and ±30. The uncertainties estimated for HFC-134a and CFC-11 are assessed as representative for most other gases with similar or longer lifetimes. For
shorter-lived gases, the uncertainties will be larger. For GTP, few estimates are available in the literature. The uncertainty is assessed to be of the order of ±75% for the methane GTP
100
.
* No single lifetime can be given. The impulse response function for CO
2
from Joos et al. (2013) has been used. See also Supplementary Material Section 8.SM.11.
† Perturbation lifetime is used in calculation of metrics, not the lifetime of the atmospheric burden.
(continued on next page)
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
738
Table 8.A.1 Notes (continued)
‡ Metric values for CH
4
of fossil origin include the oxidation to CO
2
(based on Boucher et al., 2009). In applications of these values, inclusion of the CO
2
effect of fossil methane must be done with caution to avoid any double-counting because CO
2
emissions
numbers are often based on total carbon content. Methane values without the CO
2
effect from fossil methane are thus appropriate for fossil methane sources for which the carbon has been accounted for elsewhere, or for biospheric methane sources for which
there is abalance between CO
2
taken up by the biosphere and CO
2
produced from CH
4
oxidization. The addition effect on GWP and GTP represents lower limits from Boucher et al. (2009) and assume 50% of the carbon is deposited as formaldehyde to the
surface and is then lost. The upper limit in Boucher et al. (2009) made the assumption that this deposited formaldehyde was subsequently further oxidized to CO
2
.
a
RE is unchanged since AR4.
b
RE is unchanged since AR4 except the absolute forcing is increased by a factor of 1.04 to account for the change in the recommended RE of CFC-11.
c
Based on Rajakumar et al. (2006) (lifetime correction factor has been applied to account for non-homogeneous horizontal and vertical mixing).
d
Based on instantaneous RE from Baasandorj et al. (2010); Baasandorj et al. (2011) (correction factors have been applied to account for stratospheric temperature adjustment and non-homogeneous horizontal and vertical mixing).
e
Based on instantaneous RE from ab initio study of Bravo et al. (2010) (a factor 1.10 has been applied to account for stratospheric temperature adjustment).
f
Based on average instantaneous RE reported in literature (Vasekova et al., 2006; Bravo et al., 2010) (correction factors have been applied to account for stratospheric temperature adjustment and non-homogeneous horizontal and vertical mixing).
g
Based on instantaneous RE from ab initio studies of Blowers et al. (2007, 2008)(correction factors have been applied to account for stratospheric temperature adjustment and non-homogeneous horizontal and vertical mixing).
h
Based on instantaneous RE from Heathfield et al. (1998) (correction factors have been applied to account for stratospheric temperature adjustment and non-homogeneous horizontal and vertical mixing).
i
Note that calculation of RE is based on calculated (ab initio) absorption cross-section and uncertainties are therefore larger than for calculations using experimental absorption cross section.
j
Based on instantaneous RE from Javadi et al. (2007) (correction factors have been applied to account for stratospheric temperature adjustment and non-homogeneous horizontal and vertical mixing).
k
Based on instantaneous RE from Andersen et al. (2010) (correction factors have been applied to account for stratospheric temperature adjustment and non-homogeneous horizontal and vertical mixing).
The GTP values are calculated with a temperature impulse response function taken from Boucher and Reddy (2008). See also Supplementary Material Section 8.SM.11.
8
Anthropogenic and Natural Radiative Forcing Chapter 8
739
Table 8.A.2 | Halocarbon indirect GWPs from ozone depletion using the EESC-based
method described in WMO (2011), adapted from Daniel et al. (1995). A radiative forcing
in year 2011 of –0.15 (–0.30 to 0.0) W m
–2
relative to preindustrial times is used (see
Section 8.3.3). Uncertainty on the indirect AGWPs due to the ozone forcing uncertainty
is ±100%.
Gas GWP
100
CFC-11 –2640
CFC-12 –2100
CFC-113 –2150
CFC-114 –914
CFC-115 –223
HCFC-22 –98
HCFC-123 –37
HCFC-124 –46
HCFC-141b –261
HCFC-142b –152
CH
3
CCl
3
–319
CCl
4
–2110
CH
3
Br –1250
Halon-1211 –19,000
Halon-1301 –44,500
Halon-2402 –32,000
HCFC-225ca –40
HCFC-225cb –60
GWP GTP
H = 20 H = 100 H = 20 H = 100
NO
X
East Asia
a
6.4 (±38.1) –5.3 (±11.5) –55.6 (±23.8) –1.3 (±2.1)
NO
X
EU + North Africa
a
–39.4 (±17.5) –15.6 (±5.8) –48.0 (±14.9) –2.5 (±1.3)
NO
X
North America
a
–2.4 (±30.3) –8.2 (±10.3) –61.9 (±27.8) –1.7 (±2.1)
NO
X
South Asia
a
–40.7 (±88.3) –25.3 (±29.0) –124.6 (±67.4) –4.6 (±5.1)
NO
X
four above regions
a
–15.9 (±32.7) –11.6 (±10.7) –62.1 (±26.2) –2.2 (±2.1)
Mid-latitude NOx
c
–43 to +23 –18 to +1.6 –55 to –37 –2.9 to –0.02
Tropical NO
x
c
43 to 130 –28 to –10 –260 to –220 –6.6 to –5.4
NO
X
global
b
19 –11 –87 –2.9
NO
X
global
d
–108 ± 35
–335 ± 110
–560 ± 279
–31 ± 10
–95 ± 31
–159 ± 79
Table 8.A.3 | GWP and GTP for NO
X
from surface sources for time horizons of 20 and 100 years from the literature. All values are on a per kilogram of nitrogen basis. Uncertainty
for numbers from Fry et al. (2012) and Collins et al. (2013) refer to 1-σ. For the reference gas CO
2
, RE and IRF from AR4 are used in the calculations. The GWP
100
and GTP
100
values
can be scaled by 0.94 and 0.92, respectively, to account for updated values for the reference gas CO
2
. For 20 years the changes are negligible.
Notes:
a
Fry et al. (2012) (updated by including stratospheric H
2
O) and Collins et al. (2013).
b
Fuglestvedt et al. (2010); based on Wild et al. (2001).
c
Fuglestvedt et al. (2010).
d
Shindell et al. (2009). Three values are given: First, without aerosols, second, direct aerosol effect included (sulfate and nitrate), third, direct and indirect aerosol effects included. Uncertainty
ranges from Shindell et al. (2009) are given for 95% confidence levels.
8
Chapter 8 Anthropogenic and Natural Radiative Forcing
740
GWP GTP
H = 20 H = 100 H = 20 H = 100
CO East Asia
a
5.4 (±1.7) 1.8 (±0.6) 3.5 (±1.3) 0.26 (±0.12)
CO EU + North Africa
a
4.9 (±1.5) 1.6 (±0.5) 3.2 (±1.2) 0.24 (±0.11)
CO North America
a
5.6 (±1.8) 1.8 (±0.6) 3.7 (±1.3) 0.27 (±0.12)
CO South Asia
a
5.7 (±1.3) 1.8 (±0.4) 3.4 (±1.0) 0.27 (±0.10)
CO four regions above
a
5.4 (±1.6) 1.8 (±0.5) 3.5 (±1.2) 0.26 (±0.11)
CO global
b
6 to 9.3 2 to 3.3 3.7 to 6.1 0.29 to 0.55
CO global
c
7.8 ± 2.0
11.4 ± 2.9
18.6 ± 8.3
2.2 ± 0.6
3.3 ± 0.8
5.3 ± 2.3
GWP GTP
H = 20 H = 100 H = 20 H = 100
BC total, global
c
3200 (270 to 6200) 900 (100 to 1700) 920 (95 to 2400) 130 (5 to 340)
BC (four regions)
d
1200 ± 720 345 ± 207 420 ± 190 56 ± 25
BC global
a
1600 460 470 64
BC aerosol–radiation interaction +albedo, global
b
2900 ± 1500 830 ± 440
OC global
a
–240 –69 –71 –10
OC global
b
–160 (–60 to –320) –46 (–18 to –19)
OC (4 regions)
d
–160 ± 68 –46 ± 20 –55 ± 16 –7.3±2.1
GWP GTP
H = 20 H = 100 H = 20 H = 100
VOC East Asia
a
16.3 (±6.4) 5.0 (±2.1) 8.4 (±4.6) 0.7 (±0.4)
VOC EU + North Africa
a
18.0 (±8.5) 5.6 (±2.8) 9.5 (±6.5) 0.8 (±0.5)
VOC North America
a
16.2 (±9.2) 5.0 (±3.0) 8.6 (±6.4) 0.7 (±0.5)
VOC South Asia
a
27.8 (±5.6) 8.8 (±1.9) 15.7 (±5.0) 1.3 (±0.5)
VOC four regions above 18.7 (±7.5) 5.8 (±2.5) 10.0 (±5.7) 0.9 (±0.5)
VOC global
b
14 4.5 7.5 0.66
Table 8.A.4 | GWP and GTP for CO for time horizons of 20 and 100 years from the literature. Uncertainty for numbers from Fry et al. (2012) and Collins et al. (2013) refer to 1-σ.
For the reference gas CO
2
, RE and IRF from AR4 are used in the calculations. The GWP
100
and GTP
100
values can be scaled by 0.94 and 0.92, respectively, to account for updated
values for the reference gas CO
2
. For 20 years the changes are negligible.
Notes:
a
Fry et al. (2012) (updated by including stratospheric H
2
O) and Collins et al. (2013).
b
Fuglestvedt et al. (2010).
c
Shindell et al. (2009). Three values are given: First, without aerosols, second, direct aerosol effect included, third, direct and indirect aerosol effects included. Uncertainty ranges from Shindell et
al. (2009) are given for 95% confidence levels.
Table 8.A.5 | GWP and GTP for VOCs for time horizons of 20 and 100 years from the literature. Uncertainty for numbers from Fry et al. (2012) and Collins et al. (2013) refer to
1-σ. For the reference gas CO
2
, RE and IRF from AR4 are used in the calculations. The GWP
100
and GTP
100
values can be scaled by 0.94 and 0.92, respectively, to account for updated
values for the reference gas CO
2
. For 20 years the changes are negligible.
Notes:
a
Fry et al. (2012) (updated by including stratospheric H
2
O) and Collins et al. (2013).
b
Fuglestvedt et al. (2010) based on Collins et al. (2002).
The values are given on a per kilogram of C basis.
Table 8.A.6 | GWP and GTP from the literature for BC and OC for time horizons of 20 and 100 years. For the reference gas CO
2
, RE and IRF from AR4 are used in the calculations.
The GWP
100
and GTP
100
values can be scaled by 0.94 and 0.92, respectively, to account for updated values for the reference gas CO
2
. For 20 years the changes are negligible.
Notes:
a
Fuglestvedt et al. (2010).
b
Bond et al. (2011). Uncertainties for OC are asymmetric and are presented as ranges.
c
Bond et al. (2013). Metric values are given for total effect.
d
Collins et al. (2013). The four regions are East Asia, EU + North Africa, North America and South Asia (as also given in Fry et al., 2012). Only aerosol-radiation interaction is included.