1
Black carbon vertical profiles strongly affect its radiative forcing uncertainty
2
B. H. Samset, G. Myhre, M. Schulz, Y. Balkanski, S. Bauer, N. Bellouin, T. K. Berntsen, M. Chin, T.
3
Diehl, R. E. Easter, S. J. Ghan, T. Iversen, A. Kirkevåg, J.-F. Lamarque, G. Lin, J. Penner, Ø. Seland,
4
R. B. Skeie, P. Stier, T. Takemura, K. Tsigaridis, K. Zhang
5
6
Unlike most atmospheric aerosols, black carbon (BC) absorbs solar radiation. This
7
warming effect of BC has led to suggestions, both in the scientific community1-4 and
8
among policy makers, for reduction of BC emissions to mitigate global warming. The
9
uncertainties in the radiative forcing (RF) of the direct aerosol effect of BC are however
10
large5-7, hampering mitigation studies8. Among the causes of these model uncertainties
11
are assumptions about the vertical concentration profiles of BC9. There are also
12
significant discrepancies between models and observations10. Here eleven global aerosol
13
models are used to show that the diversity in the simulated vertical profile of BC mass
14
contributes between 20% and 50% of the uncertainty in modeled BC RF. A significant
15
fraction of this variability comes from high altitudes as, globally, more than 40% of the
16
total BC RF is found above 5km. BC emission regions and areas with transported BC
17
are found to have differing characteristics. These insights into the importance of the
18
vertical profile of BC suggest that observational studies are needed to better
19
characterize the global distribution of BC, including in the upper troposphere.
20
Aerosol radiative forcing is a measure of the effect a given change in aerosol concentrations
21
has on the atmospheric energy balance. The efficiency with which BC can induce RF is
22
however dependent on external factors such as surface albedo, water vapor, background
23
aerosol distributions and, most notably, the presence of clouds9,11. The BC forcing is therefore
24
highly sensitive to the full 3D distribution of BC concentration12. Anthropogenic direct BC
25
forcing estimated by the current major global aerosol models varies between 0.05 W/m2 and
1
13
. This range is similar to earlier estimates6,7, and difficulties in reducing the
26
0.38 W/m2
27
range highlights the urgent need to understand the different causes of this model variability in
28
BC forcing.
29
Several studies have previously indicated that the vertical transport is one area where the
30
models still differ significantly14-17. This study uses input from 11 global aerosol models
31
participating in the AeroCom model intercomparison proect to compare and study the impacts
32
of modeled BC vertical forcing profiles. By combining the models’ own concentration
33
profiles with a common 4D (spatial and temporal) efficiency profile (EP) of RF per gram of
34
BC (see Methods), we are able to recalculate the induced RF of the BC direct aerosol effect at
35
various altitudes and spatial regions. By comparing calculations using a 1D profile with the
36
full 4D distribution in clear and cloudy skies, we can also isolate the contributions from the
37
cloud field and variations due to regional differences.
38
We first quantify the fraction of the total modeled RF variability attributable to vertical
39
profiles alone. Figure 1a shows the global mean anthropogenic BC burden from 11 current
40
models participating in AeroCom Phase 2, in addition to three models from AeroCom Phase 1
41
included to indicate the magnitude of variability due to model developments. See Methods for
42
details. We find a multi-model mean of 0.19 mg/m2, but with a relative standard deviation
43
(RSD) of 32% and a model spread from 0.09 to 0.37 mg/m2. Supplementary Table 1 lists the
44
numbers for individual models. Under an assumption of equal forcing per gram, this alone
45
will cause a large diversity in the total BC forcing predicted by the models. AeroCom Phases
46
1 and 2 found a similar burden range, as shown by the whisker boxes in Figure 1a. Note that
47
the AeroCom P2 results are for BC from fossil and biofuel burning (BCFF) only. Figure 1b
48
shows the RF for each model, recalculated from the concentration profiles using a common
49
EP12 as described in Methods. The recalculated RF values are consequently highly correlated
50
(Pearson corr. coeff. =95%) with the burden values. Howeve, dividing the recalculated RF
2
51
by the global mean burdens gives global estimates of BC forcing efficiency (radiative forcing
52
exerted per gram of BC aerosol), that are independent of the burden value simulated by the
53
host model. This is shown in Figure 1c. If the modeled spatial and temporal aerosol
54
distributions, notably the vertical profiles, were also identical this spread would vanish. The
55
remaining diversity therefore carries information on the impact of the vertical profiles of BC
56
concentration on the model RF spread, separate from the variability due to burden differences.
57
To quantify this impact, we analyze the multi-model RSDs and correlations, as fully discussed
58
in Methods. We isolate the contribution by the vertical distributions alone by performing
59
parallel analyses using global monthly BC mass profiles and EPs, removing horizontal
60
variability, and global, annual mean profiles, removing temporal variability. We then observe
61
that in AeroCom P1 and P2 there is an anticorrelation between burdens and forcing
62
efficiencies, while the impact of the vertical distribution variability (quantified by the fraction
63
of mass above 5km, discussed below) is positively correlated with the recalculated RF. This
64
allows us to calculate a range for the impact of BC mass vertical profiles alone. We find that
65
of the total model variability of BC RF, no more than 50% but likely more than 20% is
66
attributable to differences in vertical profiles.
67
Next we compare global and regional BC vertical profiles and EPs. As also shown in previous
68
model comparisons15-17, there are large differences in the modeled BC concentrations. Figure
69
2a-c shows the annual mean vertical profiles of BC concentration from all models, globally
70
and for two selected regions (Arctic and China; see Methods for definitions). Globally, the
71
concentrations quickly decrease with altitude while the forcing efficiency increases. The
72
Arctic is a region with negligible indigenous BC emissions, but where the effects of
73
transported aerosols are relevant18. While all models show a significant BC contribution at
74
high altitudes the model spread in the Arctic is quite large, ranging from virtually constant
75
below 200hPa to double-peaked structures with maxima at 900 hPa and 200hPa. This
3
76
highlights the differences in transport and wet removal schemes between the climate models.
77
AeroCom P1 and P2 models can be seen to perform equally well where we are able to
78
compare. China has large surface emissions of BC, and also sees contributions from transport
79
at high altitudes, as is evident from Figure 2c.
80
Figure 2d-f shows the corresponding recalculated forcing efficiency as a function of altitude,
81
divided by model layer height to remove differences due to vertical resolution. The resulting
82
vertical profiles illustrate at what altitudes RF is generated. Globally, RF can be seen to be
83
mostly exerted in a range around 800hPa for most models, with a secondary peak between
84
400hPa and 200hPa for some models. The China region follows the global pattern, but with
85
the low altitude peak closer to the ground, reflecting the emissions there. The Arctic region,
86
however, has a strong peak at high altitudes for most models, evidencing the relative
87
importance of high altitude BC RF there. These features are explained by the differences in
88
regional concentration and forcing efficiency profiles shown in Figure 2a-c.
89
Figure 2g-i illustrates the model spread in accumulated forcing, by integrating the absolute
90
forcing per model layer from the surface and upwards. Globally, the majority of models
91
contain 50% of their forcing below 500hPa (above approximately 5km); however there is a
92
large spread around this value. Regionally, this picture is quite different. In the Arctic all
93
models have a significant forcing component above 5km, due both to a high fraction of
94
aerosol at high altitudes and a strong forcing efficiency caused by high surface albedo For the
95
industrial regions of China the opposite is true, with the bulk of the aerosol located close to
96
the ground where the forcing efficiency is weak.
97
We find, in general, that a significant fraction of modeled BC RF is exerted at high altitudes,
98
with a notable regional pattern. Figure 3 maps the model mean fraction of aerosol mass (3a)
99
and induced forcing (3b) from model layers of altitude above 500hPa, or approximately 5km.
4
100
Firstly, the distinction between BC emission regions and regions with mostly transported BC
101
is evident. The former have small mass and RF fractions at high altitudes, typically 10-20%,
102
while transport regions show RF fractions up to 80%. Figure 3c compares the global and
103
regional (Arctic, China and Europe) mean values of the mass and RF fractions. Note that the
104
fraction of RF above 5km is systematically higher than the mass fraction, due to the strongly
105
increasing shape of the RF efficiency profiles. Globally, more than 40% of the model
106
simulated RF from BC comes from model layers above 5km, while only 24% of the mass is
107
found in this region. This illustrates the importance of validating model vertical profiles and
108
transport codes not only in industrial regions, but also in transport regions with low
109
indigenous BC emissions.
110
Supplemental figure 1 shows the RF and mass fractions in four altitude bands, globally and
111
for the three regions. Such information is useful for comparing models with observational
112
data, which can aid future constructions of best estimates for global BC forcing.
113
We have shown that, even on global mean, a significant fraction of the variability in the BC
114
forcing per gram is due to differences in vertical profiles. It is instructive to further divide this
115
variability into the components that make up the efficiency profiles. In Figure 4a we show the
116
zonal mean forcing efficiency for all models, and investigate the relative importance of the
117
cloud field and of regional differences in albedo for the resulting spread. In Figure 4c we have
118
run the analysis using a global, annual mean efficiency profile instead of the full time
119
dependent 3D profile, and in addition used clear sky conditions. The forcing efficiency has a
120
weak but non-vanishing model spread, and is only weakly dependent on latitude since the
121
effects of varying surface albedo, which normally strongly increases the forcing efficiency at
122
the poles, are averaged out in the global profile. Not all vertical sensitivity of BC forcing is
123
due to the aerosols being above or below clouds12. The remainder, caused predominantly by
124
Rayleigh scattering, water vapor and the competing effects of other aerosols, is the cause of
5
125
the variability here. RSD on global mean values is 10% (see Figure 4b). Figure 4d shows the
126
same analysis using the full 3D efficiency profile. The RSD increases to 13%, and we now
127
see the latitudinal effects of the high polar planetary albedo. Figure 4e shows the analysis
128
using the global mean efficiency profile again, but under all-sky conditions. Again the RSD
129
increases to 13%, due to the cloud field. The model variability in forcing efficiency that is due
130
to vertical profile differences can therefore be decomposed into three factors: Equal and
131
significant contributions from the cloud field and from regional differences, and a major
132
contribution from the underlying sensitivity of BC forcing to altitude even in the absence of
133
clouds and albedo differences. Harmonizing model treatment of clouds and albedo is therefore
134
not sufficient to remove uncertainties in BC forcing due to vertical profiles. Combining the
135
effects of clouds and regional variations with the intrinsic vertical variability of the forcing
136
efficiency yields the total variability picture seen in Figure 4a.
137
Different models will likely have different BC efficiency profiles. To quantify the sensitivity
138
of the present analysis to the shape of the profile used, we reran the analysis with an EP that
139
was changed by 20% at the top of the atmosphere and unchanged at the surface, with a linear
140
interpolation in between, resulting in a weaker EP. This changed the global fraction of RF
141
above 5km by less than 5%, indicating that the results are relatively stable within reasonable
142
variations of the EP.
143
Results in the present study are given for total BC aerosol emissions only. However, due to
144
different source regions, the vertical profiles of BCFF could be different from BC. Six models
145
also provided concentration profiles for BCFF. Using the same efficiency profiles we
146
performed the analysis also for BCFF, and found results consistent with what we have
147
presented for BC (not shown). While the absolute forcing numbers differ due to lower total
148
burdens for BCFF, the variability in vertical profiles is very similar to that for total BC. Our
149
conclusions here are therefore also applicable to model comparison results on BCFF only.
6
150
Results presented here have all used emissions from year 2000. One model (CAM4-Oslo) also
151
provided simulations for year 2006. While the BC burden was 60% higher for 2006 emissions,
152
the forcing per burden and vertical profiles were invariant. This gives confidence that the
153
variability due to vertical profiles can indeed be regarded as independent of that due to the
154
present day emissions dataset.
155
Results are also similar for the P1 and P2 submissions of the IMPACT model, where no major
156
aerosol microphysical changes have been performed. For OsloCTM2, however, where both an
157
ageing scheme and modifications to the washout of BC were added, forcing per burden
158
changes between P1 and P2 by as much as half of the full range observed. Hence, the
159
transport scheme and model treatment of BC are crucial factors in determining the modeled
160
value of forcing per burden, and both of these factors are closely linked to the vertical
161
distribution.
162
In conclusion, we have shown that the previously documented large spread in BC aerosol
163
concentration profiles is enhanced for vertical BC RF profiles. Using 11 global aerosol
164
models, we show that most models globally simulate 40% of their BC forcing above 5km, and
165
that regionally the fraction can be above 70%. The spread between models is however quite
166
large, and we computed that between 25% and 50% of the differences in modeled BC RF can
167
be attributed to differences in vertical profiles. Harmonizing the treatment of clouds and
168
albedo between models is found to not be sufficient to remove this variability. To propose
169
efficient mitigation measures for BC, its radiative forcing needs to be well understood both in
170
emission and transport regions. Further model improvements and comparisons with data are
171
needed, and should focus on both of these types of region. It is however clear that while the
172
BC vertical profiles are important, they are not sufficient to explain the remaining differences
173
between global aerosol models.
7
174
Methods
175
Model simulations
176
The AeroCom initiative asks models to simulate the direct aerosol radiative effect under as
177
similar conditions as possible. All models run single year simulations using the same base
178
years for aerosol emissions (2000 and 1850), with unchanged meteorology to exclude indirect
179
effects. Both circulation models and transport models have participated. Two sets of model
180
intercomparisons have been performed, here labeled AeroCom Phase 1
181
Phase 2 13. See the references or the main AeroCom website (aerocom.met.no) for details.
182
For the present study, monthly 3D BC concentration fields from 11 global aerosol models
183
participating in AeroCom Phase 2 are used. For six of the models, concentration fields of BC
184
from fossil and biofuel burning only (BCFF) are also used. Three of the same modeling
185
groups also provided BC concentration profiles to AeroCom Phase 1, and these fields are also
186
included here. The IMPACT model has not undergone major changes between the AeroCom
187
phases, while the OsloCTM2 model has been heavily revised with the addition of both an
188
ageing scheme for BC and improved treatment of BC washout. Changes in treatment of BC
189
from CAM4-Oslo between P1 and P2 are small. The concentration fields and the 3D model
190
resolutions are the only inputs taken from each separate model. The 1850 concentrations are
191
subtracted from the 2000 ones, leaving the contribution due to anthropogenic BC or BCFF
192
emissions.
193
From AeroCom P2, participating models are NCAR-CAM3.520, CAM4-Oslo21, CAM5.122,
194
GISS-modelE23, GOCART-v424, HadGEM225, IMPACT26, INCA27, ECHAM5-HAM28,
195
OsloCTM229 and SPRINTARS30. From AeroCom P1, participating models are UiO_GCM31
196
(earlier version of CAM4-Oslo), IMPACT32 and OsloCTM233.
197
RF recalculation
8
6,16,19
and AeroCom
198
Samset and Myhre 2011 presented a set of vertical profiles of the direct radiative forcing per
199
gram of aerosol for BCFF, i.e. the amount of top-of-atmosphere shortwave radiative forcing
200
seen by a particular model (OsloCTM2) per gram of aerosols at a given altitude. Here we term
201
these efficiency profiles (EP). In the specialized literature, the term normalized radiative
202
forcing is more commonly used. We here employ the full time dependent 3D efficiency
203
profiles, either for all sky or clear sky conditions as described above. For each model grid
204
point and time step, we multiply the modeled BC concentration by the EP to get the
205
contribution to the total shortwave, top-of-atmosphere BC radiative forcing. This gives us
206
intercomparable 3D RF fields, something that is not immediately available from each model.
207
We label this the recalculated RF, to emphasize that it is heavily correlated with the burden
208
and should not be taken directly as an estimate of BC forcing of each model. Since the model
209
that was used to produce the profiles of RF per gram has among the strongest global mean
210
forcing efficiencies13, most recalculated RFs can be expected to be stronger than their host
211
model would predict. However, by dividing the recalculated RF by the total aerosol burden of
212
the host model, we extract the variability in forcing per gram from each model that is due only
213
to vertical profiles, and can subsequently use the vertical RF profiles to study it.
214
As our method removes contributions to the spread in BC RF due to differences in model
215
specific treatment of clouds, water uptake and microphysics, we are left solely with variations
216
due to the concentration profiles and the total aerosol burden. Dividing by the burden, we can
217
study the contribution to the variability due to regional differences (high or low surface
218
albedo, indigenous emissions, transport region or both), and due to the presence of clouds. To
219
study the latter effects, we ran parallel analyses using only a global mean profile of RF per
220
gram (labeled GP above), a distinct profile of RF per gram for clear sky conditions (labeled
221
CS), or both (labeled CSGP).
222
Region definitions
9
223
“Europe” is defined as the box covering longitudes -10 to 30, latitudes 38 to 60. “China”
224
covers longitudes 105 to 135, latitudes 15 to 45. “Arctic” covers latitudes 70 to 90, all
225
longitudes. Europe and China represent regions with high industrial BC emissions and
226
significant fractions of the global BCFF emissions, making their total BC forcing sensitive to
227
pure vertical transport and wet scavenging. The Arctic is a region with low indigenous
228
emissions, but important BC contributions at high altitudes transported from other regions. Its
229
total BC forcing is therefore sensitive to model differences in vertical and long range transport.
230
Estimating the variability caused by vertical distributions
231
We show that there is a residual variability in the forcing efficiency that is due to the
232
variations in the 4D (spatial and temporal) aerosol mass distributions, with a relative standard
233
deviation (RSD) of 16%. This number is dominated by the vertical distributions, but will also
234
have components from horizontal and temporal variability between models. To isolate the
235
contribution from vertical distributions, we performed the analysis using global mean BC and
236
efficiency profiles per month, averaging out any horizontal differences, and then using global,
237
annual mean profiles, to also remove temporal variability. The resulting RSD from vertical
238
profiles alone is 13%. This is approximately 30% of the variability on forcing per burden in
239
AeroCom P1 and P2. Three models in P2 have mass extinction coefficients that deviate
240
significantly from recommendations given in the literature34. Removing contributions from
241
these three models reduces the RSD on the forcing per burden to 32%, subsequently
242
increasing the estimate of the variability due to vertical profiles to 40%.
243
The burden and forcing efficiencies in P1 and P2 have approximately equal RSDs, so if they
244
were uncorrelated the vertical distribution could be said to contribute 20% of the variability
245
on RF. This is however not the case. We first note that in both AeroCom P1 and P2, the
246
burden and forcing per burden are weakly anticorrelated. This is apparent from the fact that
247
the RSD on the RF (shown in Figure 1) is lower what it would be if the errors on burden and
10
248
forcing per burden were uncorrelated. Quantifying this effect using numbers from AeroCom
249
P2, we find a weak Pearson correlation coefficient of =-0.46 (or -0.40 if we again remove
250
the three outlier models).
251
In the present analysis, we can estimate the impact of the vertical distributions by the fraction
252
of mass simulated above 5km (M5k), shown in Supplementary table 1. We observe that M5k
253
is strongly correlated with the recalculated RF (=0.70) and forcing per burden (=0.97), as
254
expected due to the efficiency profile used for the present analysis. However M5k is also
255
weakly positively correlated with the total burden (=0.48). The vertical distribution
256
variability therefore does not contribute to the observed anticorrelation between burden and
257
forcing per burden in AeroComP1 and P2.
258
We can assume that forcing per burden is determined by a combination of the vertical profile
259
(positive correlation with burden) and a set of uncorrelated global variables such as BC
260
optical properties (which must then have a combined negative correlation with burden). The
261
variability on BC RF would be higher if the models had not compensated for high burden
262
with a low forcing efficiency and vice versa. Since the vertical variability rather leads to a
263
high efficiency for a high burden, its impact on the RF variability is likely stronger than the
264
estimate of 20% above. The present analysis does not allow for rigorous quantification, but it
265
is unlikely to be larger than the 50% of the variability on total BC RF caused by forcing per
266
burden. Hence we have a range of 20% to 50% of the variability of modeled BC RF caused by
267
differences in vertical distributions.
268
11
269
Tables
270
271
Supplementary Table 1: Modeled BC burden, RF calculated by use of full 3D efficiency profiles (RF) or by a global mean
272
profile (RFGP), and forcing per gram (NRF). All numbers shown for global mean and for three selected regions.
273
RF_fraction shows the fraction of the total BC forcing simulated within the stated region. M>5km and RF>5km show the
274
fractions of aerosol mass and RF, respectively, simulated above an altitude of 5km (500hPa).
Global
Model
Arctic
Burden
RF
NRF
RFGP
RF fraction
Mass>5km
RF>5km
[mg/m2]
[W/m2]
[W/g]
[W/g]
[%]
[%]
[%]
NCAR-CAM3.5
0.14
0.25
1801
0.27
100
20.3
39.3
CAM4-Oslo
0.22
0.54
2445
0.57
100
46.0
CAM5.1
0.09
0.16
1655
0.18
100
18.1
GISS-modelE
0.21
0.40
1905
0.44
100
GOCART-v4
0.19
0.38
1982
0.41
HadGEM2
0.37
0.81
2185
IMPACT
0.17
0.24
INCA
0.23
ECHAM5-HAM
Burden
RF
NRF
RFGP
RF fraction
Mass>5km
[mg/m2]
[W/m2]
[W/g]
[W/g]
[%]
[%]
[%]
NCAR-CAM3.5
0.05
0.20
3907
0.16
2.47
63.2
76.1
64.7
CAM4-Oslo
0.20
0.79
3877
0.60
4.41
60.9
71.0
36.4
CAM5.1
0.02
0.07
4187
0.06
1.41
81.6
88.5
30.1
56.4
GISS-modelE
0.16
0.64
3889
0.49
4.74
65.7
77.2
100
27.1
46.3
GOCART-v4
0.14
0.52
3746
0.38
4.15
53.2
64.1
0.86
100
33.6
50.6
HadGEM2
0.34
1.19
3465
0.86
4.40
40.3
52.1
1450
0.28
100
5.8
13.1
IMPACT
0.05
0.16
3214
0.11
1.95
35.9
48.1
0.41
1833
0.45
100
28.9
53.8
INCA
0.07
0.29
4016
0.24
2.12
78.8
89.2
0.18
0.26
1492
0.31
100
10.8
25.7
ECHAM5-HAM
0.03
0.11
4078
0.08
1.21
75.5
89.2
OsloCTM2
0.20
0.40
2003
0.44
100
30.1
48.3
OsloCTM2
0.07
0.27
3879
0.21
2.03
71.0
82.4
SPRINTARS
0.19
0.37
1959
0.41
100
30.3
54.2
SPRINTARS
0.08
0.34
4413
0.26
2.77
83.1
89.9
CAM-Oslo-P1
0.17
0.34
1955
0.36
100
26.4
46.9
CAM-Oslo-P1
0.12
0.46
3716
0.34
4.10
48.9
61.1
IMPACT-P1
0.19
0.30
1546
0.34
100
14.2
30.7
IMPACT-P1
0.11
0.35
3266
0.26
3.48
39.4
52.9
OsloCTM2-P1
0.19
0.27
1477
0.33
100
11.7
24.2
OsloCTM2-P1
0.02
0.07
3637
0.05
0.74
63.0
80.4
Mean
0.19
0.37
1835
0.40
100
23.8
42.2
Mean
0.10
0.39
3806
0.29
2.85
61.5
73.0
Std.dev
0.06
0.16
290
0.16
0
10.8
14.5
Std.dev
0.09
0.31
335
0.23
1.34
15.9
15.0
RF>5km
Europe
Model
Model
RF>5km
China
Burden
RF
NRF
RFGP
RF fraction
Mass>5km
RF>5km
[mg/m2]
[W/m2]
[W/g]
[W/g]
[%]
[%]
[%]
NCAR-CAM3.5
0.20
0.27
1386
0.32
1.15
16.2
40.8
CAM4-Oslo
0.36
0.73
2029
0.80
1.41
36.2
CAM5.1
0.21
0.25
1230
0.30
1.70
9.1
GISS-modelE
0.62
0.79
1276
0.91
2.03
GOCART-v4
0.30
0.48
1626
0.54
HadGEM2
0.58
1.09
1864
1.20
IMPACT
0.36
0.37
1036
INCA
0.45
0.54
ECHAM5-HAM
0.39
OsloCTM2
0.30
SPRINTARS
CAM-Oslo-P1
Model
Burden
RF
NRF
RFGP
RF fraction
Mass>5km
[mg/m2]
[W/m2]
[W/g]
[W/g]
[%]
[%]
[%]
NCAR-CAM3.5
0.89
1.17
1312
1.42
8.88
9.7
25.0
60.1
CAM4-Oslo
0.84
1.35
1612
1.55
4.66
22.7
48.3
25.0
CAM5.1
0.65
0.74
1139
0.93
8.87
8.1
24.4
16.5
43.6
GISS-modelE
1.60
1.89
1185
2.43
8.78
10.4
30.5
1.34
21.6
44.0
GOCART-v4
1.12
1.50
1340
1.82
7.46
11.3
28.4
1.40
27.3
47.4
HadGEM2
2.08
3.29
1581
3.82
7.61
15.9
34.2
0.45
1.61
4.3
13.0
IMPACT
0.83
0.94
1138
1.21
7.32
3.3
9.2
1206
0.63
1.37
17.4
49.6
INCA
1.06
1.27
1198
1.53
5.70
13.7
40.8
0.41
1052
0.47
1.62
5.8
19.4
ECHAM5-HAM
1.31
1.39
1061
1.80
9.85
2.8
9.1
0.47
1557
0.54
1.23
19.5
41.0
OsloCTM2
1.27
1.95
1544
2.26
9.17
15.4
33.0
0.23
0.38
1658
0.43
1.08
27.9
58.7
SPRINTARS
1.33
1.66
1248
2.04
8.30
10.2
28.3
0.36
0.52
1425
0.59
1.62
16.5
38.5
CAM-Oslo-P1
0.67
0.89
1336
1.06
4.96
10.5
27.4
IMPACT-P1
0.50
0.54
1081
0.63
1.87
8.5
24.9
IMPACT-P1
0.96
0.87
904
1.20
5.41
5.6
19.8
OsloCTM2-P1
0.38
0.45
1189
0.53
1.71
5.9
16.0
OsloCTM2-P1
0.84
1.00
1189
1.27
6.82
6.6
18.1
Mean
0.37
0.52
1401
0.60
1.51
16.6
37.3
Mean
1.10
1.42
1270
1.74
7.41
10.4
26.9
Std.dev
0.13
0.22
308
0.24
0.27
9.5
15.2
Std.dev
0.39
0.66
202
0.75
1.69
5.3
10.9
275
276
12
277
Figures
278
279
Figure 1: Modeled BC global mean (a) burden, (b) RF and (c) forcing efficiency (RF per gram of BC). Yellow boxes with
280
whiskers indicate mean, one standard deviation and max/min values. Mean values and spreads for AeroCom P1 and P2
281
(hatched whisker boxes) are taken from Schulz et al 2006 and Myhre et al 2012 respectively.
13
282
283
Figure 2: Comparison of modeled concentration and RF profiles. (a-c) BC concentration vertical profiles, global mean and
284
for two selected regions. Overlain is the annual mean forcing efficiency profile for the selected region (grey dashed line).
285
Solid lines show AeroCom P2 submissions, dashed lines show P1. (d-f) BC RF per height, divided by the modeled global
286
mean BC burden, globally and for three selected regions. (g-i) Vertical profile of integrated absolute BC RF. Lines indicate
287
the 50% mark and 500hPa altitude.
14
288
289
290
291
Figure 3: Black carbon mass and induced forcing at high altitudes. (a) Fraction of modeled BC mass above 5km. (b)
292
Fraction of modeled BC RF originating above 5km. (c) Mean fraction of mass (orange) and RF (yellow) globally and for
293
three selected regions. Boxes indicate one standard deviation on the model spread, whiskers show maximum and
294
minimum values.
15
295
296
297
298
Figure 4: Breakdown of spread in BC forcing efficiency. (a) Zonal mean of BC RF per gram using 3D efficiency profiles and
299
all sky conditions, for all models. (c-e) Model mean (solid line) and maximum/minimum (dashed lines) when instead
300
using (c) a global, annual mean efficiency profile and clear sky conditions, (d) 3D profile and clear sky conditions, and (e)
301
a global profile and all sky conditions. (b) shows the global mean values for the four cases. Boxes show one standard
302
deviation on the model spread, whiskers show maximum/minimum values.
16
303
304
305
Supplementary figure 1: Mass and forcing fractions in four altitude bands, globally and for three regions. Yellow boxes
306
show mass fractions, orange boxes show RF fractions. (a) Global mean, (b) Arctic, (c) Europe, (d) China.
307
308
1
Time for early action. Nature 460, 12-12, doi:Doi 10.1038/460012a (2009).
309
2
Grieshop, A. P., Reynolds, C. C. O., Kandlikar, M. & Dowlatabadi, H. A black-carbon mitigation wedge. Nat Geosci 2,
310
311
533-534, doi:Doi 10.1038/Ngeo595 (2009).
3
312
313
314
Shindell, D. et al. Simultaneously Mitigating Near-Term Climate Change and Improving Human Health and Food
Security. Science 335, 183-189, doi:DOI 10.1126/science.1210026 (2012).
4
Hansen, J., Sato, M., Ruedy, R., Lacis, A. & Oinas, V. Global warming in the twenty-first century: An alternative
scenario. P Natl Acad Sci USA 97, 9875-9880 (2000).
17
315
5
316
317
221-227, doi:10.1038/ngeo156 (2008).
6
318
319
7
Feichter, J. & Stier, P. Assessment of black carbon radiative effects in climate models. doi:DOI: 10.1002/wcc.180
(2012).
8
322
323
Schulz, M. et al. Radiative forcing by aerosols as derived from the AeroCom present-day and pre-industrial
simulations. Atmospheric Chemistry and Physics 6, 5225-5246 (2006).
320
321
Ramanathan, V. & Carmichael, G. Global and regional climate changes due to black carbon. Nature Geoscience 1,
Koch, D. et al. Soot microphysical effects on liquid clouds, a multi-model investigation. Atmos Chem Phys 11,
1051-1064, doi:DOI 10.5194/acp-11-1051-2011 (2011).
9
324
Zarzycki, C. M. & Bond, T. C. How much can the vertical distribution of black carbon affect its global direct
radiative forcing? Geophys Res Lett 37, L20807, doi:Artn L20807
325
Doi 10.1029/2010gl044555 (2010).
326
10
327
328
(2009).
11
329
330
Koch, D. et al. Evaluation of black carbon estimations in global aerosol models. Atmos Chem Phys 9, 9001-9026
Haywood, J. M. & Shine, K. P. Multi-spectral calculations of the direct radiative forcing of tropospheric sulphate
and soot aerosols using a column model. Q J Roy Meteor Soc 123, 1907-1930 (1997).
12
331
Samset, B. H. & Myhre, G. Vertical dependence of black carbon, sulphate and biomass burning aerosol radiative
forcing. Geophys Res Lett 38, doi:Artn L24802
332
Doi 10.1029/2011gl049697 (2011).
333
13
Myhre, G. Radiative forcing of the direct aerosol effect from AeroCom Phase II simulations. (2012).
334
14
Koffi, B. et al. Application of the CALIOP layer product to evaluate the vertical distribution of aerosols estimated
335
by global models: AeroCom phase I results. J Geophys Res-Atmos 117, doi:Artn D10201
336
Doi 10.1029/2011jd016858 (2012).
337
15
338
339
models. Geophys Res Lett 37 (2010).
16
340
341
344
Textor, C. et al. Analysis and quantification of the diversities of aerosol life cycles within AeroCom. Atmos Chem
Phys 6, 1777-1813 (2006).
17
342
343
Schwarz, J. P. et al. Global scale black carbon profiles observed in the remote atmosphere and compared to
Textor, C. et al. The effect of harmonized emissions on aerosol properties in global models - an AeroCom
experiment. Atmos Chem Phys 7, 4489-4501 (2007).
18
Shindell, D. T. et al. A multi-model assessment of pollution transport to the Arctic. Atmos Chem Phys 8, 5353-5372
(2008).
18
345
19
346
347
for AeroCom. Atmos Chem Phys 6, 4321-4344 (2006).
20
348
349
21
22
23
Koch, D. et al. Coupled Aerosol-Chemistry-Climate Twentieth-Century Transient Model Investigation: Trends in
Short-Lived Species and Climate Responses. J Climate 24, 2693-2714, doi:Doi 10.1175/2011jcli3582.1 (2011).
24
356
357
Liu, X. et al. Toward a minimal representation of aerosols in climate models: description and evaluation in the
Community Atmosphere Model CAM5. Geosci Model Dev 5, 709-739, doi:10.5194/gmd-5-709-2012 (2012).
354
355
Kirkevåg, A. et al. Kirkevåg, A. T. Iversen, Ø. Seland, C. Hoose, J. E. Kristjáson, H. Struthers, A. Ekman, S. Ghan, J.
Griesfeller, D. Nilsson, and M. Schulz. In prep. (2012).
352
353
Lamarque, J. F. et al. CAM-chem: description and evaluation of interactive atmospheric chemistry in the
Community Earth System Model. Geosci Model Dev 5, 369-411, doi:DOI 10.5194/gmd-5-369-2012 (2012).
350
351
Kinne, S. et al. Emissions of primary aerosol and precursor gases in the years 2000 and 1750 prescribed data-sets
Chin, M. et al. Light absorption by pollution, dust, and biomass burning aerosols: a global model study and
evaluation with AERONET measurements. Ann Geophys-Germany 27, 3439-3464 (2009).
25
358
Bellouin, N. et al. Aerosol forcing in the Climate Model Intercomparison Project (CMIP5) simulations by HadGEM2ES and the role of ammonium nitrate. J Geophys Res-Atmos 116, doi:Artn D20206
359
Doi 10.1029/2011jd016074 (2011).
360
26
361
362
epoxides, organic nitrates and peroxides. Atmos Chem Phys 12, 4743-4774, doi:10.5194/acp-12-4743-2012 (2012).
27
363
364
28
Zhang, K. et al. The global aerosol-climate model ECHAM-HAM, version 2: sensitivity to improvements in process
representations. Atmos. Chem. Phys. Discuss. 12, 7545-7615, doi:10.5194/acpd-12-7545-2012 (2012).
29
367
368
Szopa, S. et al. Aerosol and ozone changes as forcing for climate evolution between 1850 and 2100. Climate
Dynamics Submitted (2012).
365
366
Lin, G., Penner, J., Sillmann, S., Taraborrelli, D. & Lelieveld, J. Global modeling of SOA formation from dicarbonyls,
Skeie, R. B. et al. Black carbon in the atmosphere and snow, from pre-industrial times until present. Atmos Chem
Phys 11, 6809-6836, doi:10.5194/acp-11-6809-2011 (2011).
30
369
Takemura, T., Nozawa, T., Emori, S., Nakajima, T. Y. & Nakajima, T. Simulation of climate response to aerosol
direct and indirect effects with aerosol transport-radiation model. J Geophys Res-Atmos 110, doi:Artn D02202
370
Doi 10.1029/2004jd005029 (2005).
371
31
372
Kirkevag, A. & Iversen, T. Global direct radiative forcing by process-parameterized aerosol optical properties. J
Geophys Res-Atmos 107, doi:Artn 4433
373
Doi 10.1029/2001jd000886 (2002).
374
32
375
Liu, X. H., Penner, J. E. & Herzog, M. Global modeling of aerosol dynamics: Model description, evaluation, and
interactions between sulfate and nonsulfate aerosols. J Geophys Res-Atmos 110, doi:Artn D18206
19
376
Doi 10.1029/2004jd005674 (2005).
377
33
378
Myhre, G. et al. Modeling the solar radiative impact of aerosols from biomass burning during the Southern African
Regional Science Initiative (SAFARI-2000) experiment. J Geophys Res-Atmos 108, doi:Artn 8501
379
Doi 10.1029/2002jd002313 (2003).
380
34
381
382
Bond, T. C., Habib, G. & Bergstrom, R. W. Limitations in the enhancement of visible light absorption due to mixing
state. J Geophys Res-Atmos 111, doi:Artn D20211
Doi 10.1029/2006jd007315 (2006).
383
384
20