867
This chapter should be cited as:
Bindoff, N.L., P.A. Stott, K.M. AchutaRao, M.R. Allen, N. Gillett, D. Gutzler, K. Hansingo, G. Hegerl, Y. Hu, S. Jain, I.I.
Mokhov, J. Overland, J. Perlwitz, R. Sebbari and X. Zhang, 2013: Detection and Attribution of Climate Change:
from Global to Regional. 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:
Nathaniel L. Bindoff (Australia), Peter A. Stott (UK)
Lead Authors:
Krishna Mirle AchutaRao (India), Myles R. Allen (UK), Nathan Gillett (Canada), David Gutzler
(USA), Kabumbwe Hansingo (Zambia), Gabriele Hegerl (UK/Germany), Yongyun Hu (China),
Suman Jain (Zambia), Igor I. Mokhov (Russian Federation), James Overland (USA), Judith
Perlwitz (USA), Rachid Sebbari (Morocco), Xuebin Zhang (Canada)
Contributing Authors:
Magne Aldrin (Norway), Beena Balan Sarojini (UK/India), Jürg Beer (Switzerland), Olivier
Boucher (France), Pascale Braconnot (France), Oliver Browne (UK), Ping Chang (USA), Nikolaos
Christidis (UK), Tim DelSole (USA), Catia M. Domingues (Australia/Brazil), Paul J. Durack (USA/
Australia), Alexey Eliseev (Russian Federation), Kerry Emanuel (USA), Graham Feingold (USA),
Chris Forest (USA), Jesus Fidel González Rouco (Spain), Hugues Goosse (Belgium), Lesley Gray
(UK), Jonathan Gregory (UK), Isaac Held (USA), Greg Holland (USA), Jara Imbers Quintana
(UK), William Ingram (UK), Johann Jungclaus (Germany), Georg Kaser (Austria), Veli-Matti
Kerminen (Finland), Thomas Knutson (USA), Reto Knutti (Switzerland), James Kossin (USA),
Mike Lockwood (UK), Ulrike Lohmann (Switzerland), Fraser Lott (UK), Jian Lu (USA/Canada),
Irina Mahlstein (Switzerland), Valérie Masson-Delmotte (France), Damon Matthews (Canada),
Gerald Meehl (USA), Blanca Mendoza (Mexico), Viviane Vasconcellos de Menezes (Australia/
Brazil), Seung-Ki Min (Republic of Korea), Daniel Mitchell (UK), Thomas Mölg (Germany/
Austria), Simone Morak (UK), Timothy Osborn (UK), Alexander Otto (UK), Friederike Otto (UK),
David Pierce (USA), Debbie Polson (UK), Aurélien Ribes (France), Joeri Rogelj (Switzerland/
Belgium), Andrew Schurer (UK), Vladimir Semenov (Russian Federation), Drew Shindell (USA),
Dmitry Smirnov (Russian Federation), Peter W. Thorne (USA/Norway/UK), Muyin Wang (USA),
Martin Wild (Switzerland), Rong Zhang (USA)
Review Editors:
Judit Bartholy (Hungary), Robert Vautard (France), Tetsuzo Yasunari (Japan)
Detection and Attribution
of Climate Change:
from Global to Regional
10
868
10
Table of Contents
Executive Summary ..................................................................... 869
10.1 Introduction ...................................................................... 872
10.2 Evaluation of Detection and Attribution
Methodologies ................................................................. 872
10.2.1 The Context of Detection and Attribution ................. 872
10.2.2 Time Series Methods, Causality and Separating
Signal from Noise ...................................................... 874
Box 10.1: How Attribution Studies Work ................................ 875
10.2.3 Methods Based on General Circulation Models
and Optimal Fingerprinting ....................................... 877
10.2.4 Single-Step and Multi-Step Attribution and the
Role of the Null Hypothesis ....................................... 878
10.3 Atmosphere and Surface .............................................. 878
10.3.1 Temperature .............................................................. 878
Box 10.2: The Sun’s Influence on the Earth’s Climate ........... 885
10.3.2 Water Cycle ............................................................... 895
10.3.3 Atmospheric Circulation and Patterns of
Variability .................................................................. 899
10.4 Changes in Ocean Properties....................................... 901
10.4.1 Ocean Temperature and Heat Content ...................... 901
10.4.2 Ocean Salinity and Freshwater Fluxes ....................... 903
10.4.3 Sea Level ................................................................... 905
10.4.4 Oxygen and Ocean Acidity ........................................ 905
10.5 Cryosphere ........................................................................ 906
10.5.1 Sea Ice ...................................................................... 906
10.5.2 Ice Sheets, Ice Shelves and Glaciers .......................... 909
10.5.3 Snow Cover ............................................................... 910
10.6 Extremes ............................................................................ 910
10.6.1 Attribution of Changes in Frequency/Occurrence
and Intensity of Extremes.......................................... 910
10.6.2 Attribution of Weather and Climate Events ............... 914
10.7 Multi-century to Millennia Perspective .................... 917
10.7.1 Causes of Change in Large-Scale Temperature over
the Past Millennium .................................................. 917
10.7.2 Changes of Past Regional Temperature ..................... 919
10.7.3 Summary: Lessons from the Past ............................... 919
10.8 Implications for Climate System Properties
and Projections ................................................................ 920
10.8.1 Transient Climate Response ...................................... 920
10.8.2 Constraints on Long-Term Climate Change and the
Equilibrium Climate Sensitivity .................................. 921
10.8.3 Consequences for Aerosol Forcing and Ocean
Heat Uptake .............................................................. 926
10.8.4 Earth System Properties ............................................ 926
10.9 Synthesis ............................................................................ 927
10.9.1 Multi-variable Approaches ........................................ 927
10.9.2 Whole Climate System .............................................. 927
References .................................................................................. 940
Frequently Asked Questions
FAQ 10.1 Climate Is Always Changing. How Do We
Determine the Causes of Observed
Changes? ................................................................. 894
FAQ 10.2 When Will Human Influences on Climate
Become Obvious on Local Scales? ....................... 928
Supplementary Material
Supplementary Material is available in online versions of the report.
869
10
Detection and Attribution of Climate Change: from Global to Regional Chapter 10
1
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).
2
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).
Executive Summary
Atmospheric Temperatures
More than half of the observed increase in global mean surface
temperature (GMST) from 1951 to 2010 is very likely
1
due to the
observed anthropogenic increase in greenhouse gas (GHG) con-
centrations. The consistency of observed and modeled changes across
the climate system, including warming of the atmosphere and ocean,
sea level rise, ocean acidification and changes in the water cycle, the
cryosphere and climate extremes points to a large-scale warming
resulting primarily from anthropogenic increases in GHG concentra-
tions. Solar forcing is the only known natural forcing acting to warm
the climate over this period but it has increased much less than GHG
forcing, and the observed pattern of long-term tropospheric warming
and stratospheric cooling is not consistent with the expected response
to solar irradiance variations. The Atlantic Multi-decadal Oscillation
(AMO) could be a confounding influence but studies that find a signif-
icant role for the AMO show that this does not project strongly onto
1951–2010 temperature trends. {10.3.1, Table 10.1}
It is extremely likely that human activities caused more than
half of the observed increase in GMST from 1951 to 2010. This
assessment is supported by robust evidence from multiple studies
using different methods. Observational uncertainty has been explored
much more thoroughly than previously and the assessment now con-
siders observations from the first decade of the 21st century and sim-
ulations from a new generation of climate models whose ability to
simulate historical climate has improved in many respects relative to
the previous generation of models considered in AR4. Uncertainties in
forcings and in climate models’ temperature responses to individual
forcings and difficulty in distinguishing the patterns of temperature
response due to GHGs and other anthropogenic forcings prevent a
more precise quantification of the temperature changes attributable to
GHGs. {9.4.1, 9.5.3, 10.3.1, Figure 10.5, Table 10.1}
GHGs contributed a global mean surface warming likely to be
between 0.5°C and 1.3°C over the period 1951–2010, with the
contributions from other anthropogenic forcings likely to be
between –0.6°C and 0.1°C, from natural forcings likely to be
between –0.1°C and 0.1°C, and from internal variability likely
to be between –0.1°C and 0.1°C. Together these assessed contri-
butions are consistent with the observed warming of approximately
0.6°C over this period. {10.3.1, Figure 10.5}
It is virtually certain that internal variability alone cannot
account for the observed global warming since 1951. The
observed global-scale warming since 1951 is large compared to cli-
mate model estimates of internal variability on 60-year time scales. The
Northern Hemisphere (NH) warming over the same period is far out-
side the range of any similar length trends in residuals from reconstruc-
tions of the past millennium. The spatial pattern of observed warming
differs from those associated with internal variability. The model-based
simulations of internal variability are assessed to be adequate to make
this assessment. {9.5.3, 10.3.1, 10.7.5, Table 10.1}
It is likely that anthropogenic forcings, dominated by GHGs,
have contributed to the warming of the troposphere since 1961
and very likely that anthropogenic forcings, dominated by the
depletion of the ozone layer due to ozone-depleting substanc-
es, have contributed to the cooling of the lower stratosphere
since 1979. Observational uncertainties in estimates of tropospheric
temperatures have now been assessed more thoroughly than at the
time of AR4. The structure of stratospheric temperature trends and
multi-year to decadal variations are well represented by models and
physical understanding is consistent with the observed and modelled
evolution of stratospheric temperatures. Uncertainties in radiosonde
and satellite records make assessment of causes of observed trends in
the upper troposphere less confident than an assessment of the overall
atmospheric temperature changes. {2.4.4, 9.4.1, 10.3.1, Table 10.1}
Further evidence has accumulated of the detection and attri-
bution of anthropogenic influence on temperature change in
different parts of the world. Over every continental region, except
Antarctica, it is likely that anthropogenic influence has made a sub-
stantial contribution to surface temperature increases since the mid-
20th century. The robust detection of human influence on continental
scales is consistent with the global attribution of widespread warming
over land to human influence. It is likely that there has been an anthro-
pogenic contribution to the very substantial Arctic warming over the
past 50 years. For Antarctica large observational uncertainties result
in low confidence
2
that anthropogenic influence has contributed to
the observed warming averaged over available stations. Anthropo-
genic influence has likely contributed to temperature change in many
sub-continental regions. {2.4.1, 10.3.1, Table 10.1}
Robustness of detection and attribution of global-scale warm-
ing is subject to models correctly simulating internal variabili-
ty. Although estimates of multi-decadal internal variability of GMST
need to be obtained indirectly from the observational record because
the observed record contains the effects of external forcings (meaning
the combination of natural and anthropogenic forcings), the standard
deviation of internal variability would have to be underestimated in
climate models by a factor of at least three to account for the observed
warming in the absence of anthropogenic influence. Comparison with
observations provides no indication of such a large difference between
climate models and observations. {9.5.3, Figures 9.33, 10.2, 10.3.1,
Table 10.1}
870
Chapter 10 Detection and Attribution of Climate Change: from Global to Regional
10
The observed recent warming hiatus, defined as the reduction
in GMST trend during 1998–2012 as compared to the trend
during 1951–2012, is attributable in roughly equal measure to
a cooling contribution from internal variability and a reduced
trend in external forcing (expert judgement, medium confi-
dence). The forcing trend reduction is primarily due to a negative forc-
ing trend from both volcanic eruptions and the downward phase of the
solar cycle. However, there is low confidence in quantifying the role of
forcing trend in causing the hiatus because of uncertainty in the mag-
nitude of the volcanic forcing trends and low confidence in the aerosol
forcing trend. Many factors, in addition to GHGs, including changes
in tropospheric and stratospheric aerosols, stratospheric water vapour,
and solar output, as well as internal modes of variability, contribute to
the year-to-year and decade- to-decade variability of GMST. {Box 9.2,
10.3.1, Figure 10.6}
Ocean Temperatures and Sea Level Rise
It is very likely that anthropogenic forcings have made a sub-
stantial contribution to upper ocean warming (above 700 m)
observed since the 1970s. This anthropogenic ocean warming has
contributed to global sea level rise over this period through thermal
expansion. New understanding since AR4 of measurement errors and
their correction in the temperature data sets have increased the agree-
ment in estimates of ocean warming. Observations of ocean warming
are consistent with climate model simulations that include anthropo-
genic and volcanic forcings but are inconsistent with simulations that
exclude anthropogenic forcings. Simulations that include both anthro-
pogenic and natural forcings have decadal variability that is consistent
with observations. These results are a major advance on AR4. {3.2.3,
10.4.1, Table 10.1}
It is very likely that there is a substantial contribution from
anthropogenic forcings to the global mean sea level rise since
the 1970s. It is likely that sea level rise has an anthropogenic con-
tribution from Greenland melt since 1990 and from glacier mass loss
since 1960s. Observations since 1971 indicate with high confidence
that thermal expansion and glaciers (excluding the glaciers in Antarc-
tica) explain 75% of the observed rise. {10.4.1, 10.4.3, 10.5.2, Table
10.1, 13.3.6}
Ocean Acidification and Oxygen Change
It is very likely that oceanic uptake of anthropogenic carbon
dioxide has resulted in acidification of surface waters which
is observed to be between –0.0014 and –0.0024 pH units per
year. There is medium confidence that the observed global pattern
of decrease in oxygen dissolved in the oceans from the 1960s to the
1990s can be attributed in part to human influences. {3.8.2, Box 3.2,
10.4.4, Table 10.1}
The Water Cycle
New evidence is emerging for an anthropogenic influence on
global land precipitation changes, on precipitation increases
in high northern latitudes, and on increases in atmospheric
humidity. There is medium confidence that there is an anthropogenic
contribution to observed increases in atmospheric specific humidi-
ty since 1973 and to global scale changes in precipitation patterns
over land since 1950, including increases in NH mid to high latitudes.
Remaining observational and modelling uncertainties, and the large
internal variability in precipitation, preclude a more confident assess-
ment at this stage. {2.5.1, 2.5.4, 10.3.2, Table 10.1}
It is very likely that anthropogenic forcings have made a dis-
cernible contribution to surface and subsurface oceanic salini-
ty changes since the 1960s. More than 40 studies of regional and
global surface and subsurface salinity show patterns consistent with
understanding of anthropogenic changes in the water cycle and ocean
circulation. The expected pattern of anthropogenic amplification of cli-
matological salinity patterns derived from climate models is detected
in the observations although there remains incomplete understanding
of the observed internal variability of the surface and sub-surface salin-
ity fields. {3.3.2, 10.4.2, Table 10.1}
It is likely that human influence has affected the global water
cycle since 1960. This assessment is based on the combined evidence
from the atmosphere and oceans of observed systematic changes that
are attributed to human influence in terrestrial precipitation, atmos-
pheric humidity and oceanic surface salinity through its connection
to precipitation and evaporation. This is a major advance since AR4.
{3.3.2, 10.3.2, 10.4.2, Table 10.1}
Cryosphere
Anthropogenic forcings are very likely to have contributed to
Arctic sea ice loss since 1979. There is a robust set of results from
simulations that show the observed decline in sea ice extent is simu-
lated only when models include anthropogenic forcings. There is low
confidence in the scientific understanding of the observed increase in
Antarctic sea ice extent since 1979 owing to the incomplete and com-
peting scientific explanations for the causes of change and low confi-
dence in estimates of internal variability. {10.5.1, Table 10.1}
Ice sheets and glaciers are melting, and anthropogenic influ-
ences are likely to have contributed to the surface melting of
Greenland since 1993 and to the retreat of glaciers since the
1960s. Since 2007, internal variability is likely to have further enhanced
the melt over Greenland. For glaciers there is a high level of scientific
understanding from robust estimates of observed mass loss, internal
variability and glacier response to climatic drivers. Owing to a low level
of scientific understanding there is low confidence in attributing the
causes of the observed loss of mass from the Antarctic ice sheet since
1993. {4.3.3, 10.5.2, Table 10.1}
It is likely that there has been an anthropogenic component to
observed reductions in NH snow cover since 1970. There is high
agreement across observations studies and attribution studies find a
human influence at both continental and regional scales. {10.5.3, Table
10.1}
871
10
Detection and Attribution of Climate Change: from Global to Regional Chapter 10
Climate Extremes
There has been a strengthening of the evidence for human influ-
ence on temperature extremes since the AR4 and IPCC Special
Report on Managing the Risks of Extreme Events and Disasters
to Advance Climate Change Adaptation (SREX) reports. It is very
likely that anthropogenic forcing has contributed to the observed
changes in the frequency and intensity of daily temperature extremes
on the global scale since the mid-20th century. Attribution of changes
in temperature extremes to anthropogenic influence is robustly seen in
independent analyses using different methods and different data sets.
It is likely that human influence has substantially increased the prob-
ability of occurrence of heatwaves in some locations. {10.6.1, 10.6.2,
Table 10.1}
In land regions where observational coverage is sufficient for
assessment, there is medium confidence that anthropogen-
ic forcing has contributed to a global-scale intensification of
heavy precipitation over the second half of the 20th century.
There is low confidence in attributing changes in drought over global
land areas since the mid-20th century to human influence owing to
observational uncertainties and difficulties in distinguishing decad-
al-scale variability in drought from long-term trends. {10.6.1, Table
10.1}
There is low confidence in attribution of changes in tropical
cyclone activity to human influence owing to insufficient obser-
vational evidence, lack of physical understanding of the links
between anthropogenic drivers of climate and tropical cyclone
activity and the low level of agreement between studies as to
the relative importance of internal variability, and anthropo-
genic and natural forcings. This assessment is consistent with that
of SREX. {10.6.1, Table 10.1}
Atmospheric Circulation
It is likely that human influence has altered sea level pressure
patterns globally. Detectable anthropogenic influence on changes
in sea level pressure patterns is found in several studies. Changes in
atmospheric circulation are important for local climate change since
they could lead to greater or smaller changes in climate in a particular
region than elsewhere. There is medium confidence that stratospheric
ozone depletion has contributed to the observed poleward shift of the
southern Hadley Cell border during austral summer. There are large
uncertainties in the magnitude of this poleward shift. It is likely that
stratospheric ozone depletion has contributed to the positive trend
in the Southern Annular Mode seen in austral summer since the mid-
20th century which corresponds to sea level pressure reductions over
the high latitudes and an increase in the subtropics. There is medium
confidence that GHGs have also played a role in these trends of the
southern Hadley Cell border and the Southern Annular Mode in Austral
summer. {10.3.3, Table 10.1}
A Millennia to Multi-Century Perspective
Taking a longer term perspective shows the substantial role
played by anthropogenic and natural forcings in driving climate
variability on hemispheric scales prior to the twentieth century.
It is very unlikely that NH temperature variations from 1400 to 1850
can be explained by internal variability alone. There is medium confi-
dence that external forcing contributed to NH temperature variability
from 850 to 1400 and that external forcing contributed to European
temperature variations over the last five centuries. {10.7.2, 10.7.5,
Table 10.1}
Climate System Properties
The extended record of observed climate change has allowed
a better characterization of the basic properties of the climate
system that have implications for future warming. New evidence
from 21st century observations and stronger evidence from a wider
range of studies have strengthened the constraint on the transient
climate response (TCR) which is estimated with high confidence to
be likely between 1°C and 2.5°C and extremely unlikely to be greater
than 3°C. The Transient Climate Response to Cumulative CO
2
Emissions
(TCRE) is estimated with high confidence to be likely between 0.8°C
and 2.5°C per 1000 PgC for cumulative CO
2
emissions less than about
2000 PgC until the time at which temperatures peak. Estimates of the
Equilibrium Climate Sensitivity (ECS) based on multiple and partly
independent lines of evidence from observed climate change indicate
that there is high confidence that ECS is extremely unlikely to be less
than 1°C and medium confidence that the ECS is likely to be between
1.5°C and 4.5°C and very unlikely greater than 6°C. These assessments
are consistent with the overall assessment in Chapter 12, where the
inclusion of additional lines of evidence increases confidence in the
assessed likely range for ECS. {10.8.1, 10.8.2, 10.8.4, Box 12.2}
Combination of Evidence
Human influence has been detected in the major assessed com-
ponents of the climate system. Taken together, the combined
evidence increases the level of confidence in the attribution of
observed climate change, and reduces the uncertainties associ-
ated with assessment based on a single climate variable. From
this combined evidence it is virtually certain that human influ-
ence has warmed the global climate system. Anthropogenic influ-
ence has been identified in changes in temperature near the surface
of the Earth, in the atmosphere and in the oceans, as well as changes
in the cryosphere, the water cycle and some extremes. There is strong
evidence that excludes solar forcing, volcanoes and internal variability
as the strongest drivers of warming since 1950. {10.9.2, Table 10.1}
872
Chapter 10 Detection and Attribution of Climate Change: from Global to Regional
10
10.1 Introduction
This chapter assesses the causes of observed changes assessed in
Chapters 2 to 5 and uses understanding of physical processes, climate
models and statistical approaches. The chapter adopts the terminolo-
gy for detection and attribution proposed by the IPCC good practice
guidance paper on detection and attribution (Hegerl et al., 2010) and
for uncertainty Mastrandrea et al. (2011). Detection and attribution of
impacts of climate changes are assessed by Working Group II, where
Chapter 18 assesses the extent to which atmospheric and oceanic
changes influence ecosystems, infrastructure, human health and activ-
ities in economic sectors.
Evidence of a human influence on climate has grown stronger over
the period of the four previous assessment reports of the IPCC. There
was little observational evidence for a detectable human influence on
climate at the time of the First IPCC Assessment Report. By the time
of the second report there was sufficient additional evidence for it to
conclude that ‘the balance of evidence suggests a discernible human
influence on global climate’. The Third Assessment Report found that
a distinct greenhouse gas (GHG) signal was robustly detected in the
observed temperature record and that ‘most of the observed warming
over the last fifty years is likely to have been due to the increase in
greenhouse gas concentrations.
With the additional evidence available by the time of the Fourth Assess-
ment Report, the conclusions were further strengthened. This evidence
included a wider range of observational data, a greater variety of more
sophisticated climate models including improved representations of
forcings and processes and a wider variety of analysis techniques.
This enabled the AR4 report to conclude that ‘most of the observed
increase in global average temperatures since the mid-20th century is
very likely due to the observed increase in anthropogenic greenhouse
gas concentrations’. The AR4 also concluded that ‘discernible human
influences now extend to other aspects of climate, including ocean
warming, continental-average temperatures, temperature extremes
and wind patterns.
A number of uncertainties remained at the time of AR4. For example,
the observed variability of ocean temperatures appeared inconsist-
ent with climate models, thereby reducing the confidence with which
observed ocean warming could be attributed to human influence. Also,
although observed changes in global rainfall patterns and increases
in heavy precipitation were assessed to be qualitatively consistent
with expectations of the response to anthropogenic forcings, detec-
tion and attribution studies had not been carried out. Since the AR4,
improvements have been made to observational data sets, taking more
complete account of systematic biases and inhomogeneities in obser-
vational systems, further developing uncertainty estimates, and cor-
recting detected data problems (Chapters 2 and 3). A new set of sim-
ulations from a greater number of AOGCMs have been performed as
part of the Coupled Model Intercomparison Project Phase 5 (CMIP5).
These new simulations have several advantages over the CMIP3 sim-
ulations assessed in the AR4 (Hegerl et al., 2007b). They incorporate
some moderate increases in resolution, improved parameterizations,
and better representation of aerosols (Chapter 9). Importantly for attri-
bution, in which it is necessary to partition the response of the climate
system to different forcings, most CMIP5 models include simulations of
the response to natural forcings only, and the response to increases in
well mixed GHGs only (Taylor et al., 2012).
The advances enabled by this greater wealth of observational and
model data are assessed in this chapter. In this assessment, there is
increased focus on the extent to which the climate system as a whole
is responding in a coherent way across a suite of climate variables
such as surface mean temperature, temperature extremes, ocean heat
content, ocean salinity and precipitation change. There is also a global
to regional perspective, assessing the extent to which not just global
mean changes but also spatial patterns of change across the globe can
be attributed to anthropogenic and natural forcings.
10.2 Evaluation of Detection and Attribution
Methodologies
Detection and attribution methods have been discussed in previous
assessment reports (Hegerl et al., 2007b) and the IPCC Good Practice
Guidance Paper (Hegerl et al., 2010), to which we refer. This section
reiterates key points and discusses new developments and challenges.
10.2.1 The Context of Detection and Attribution
In IPCC Assessments, detection and attribution involve quantifying the
evidence for a causal link between external drivers of climate change
and observed changes in climatic variables. It provides the central,
although not the only (see Section 1.2.3) line of evidence that has
supported statements such as ‘the balance of evidence suggests a dis-
cernible human influence on global climate’ or ‘most of the observed
increase in global average temperatures since the mid-20th century is
very likely due to the observed increase in anthropogenic greenhouse
gas concentrations.
The definition of detection and attribution used here follows the ter-
minology in the IPCC guidance paper (Hegerl et al., 2010). ‘Detection
of change is defined as the process of demonstrating that climate or
a system affected by climate has changed in some defined statistical
sense without providing a reason for that change. An identified change
is detected in observations if its likelihood of occurrence by chance
due to internal variability alone is determined to be small’ (Hegerl
et al., 2010). Attribution is defined as ‘the process of evaluating the
relative contributions of multiple causal factors to a change or event
with an assignment of statistical confidence’. As this wording implies,
attribution is more complex than detection, combining statistical anal-
ysis with physical understanding (Allen et al., 2006; Hegerl and Zwiers,
2011). In general, a component of an observed change is attributed to
a specific causal factor if the observations can be shown to be consist-
ent with results from a process-based model that includes the causal
factor in question, and inconsistent with an alternate, otherwise iden-
tical, model that excludes this factor. The evaluation of this consistency
in both of these cases tak