Coupled Model Intercomparison Project 5 (CMIP5)
1
Simulations of Climate Following Volcanic Eruptions
2
Simon Driscoll,1 Alessio Bozzo,2,3 Lesley J. Gray1,3, Alan Robock4, Georgiy Stenchikov4,5
Simon Driscoll Alessio Bozzo Lesley J. Gray A. Robock G. L. Stenchikov
1Atmospheric, Oceanic and Planetary
Abstract. The ability of the climate models submitted to the Coupled
3
Model Intercomparison Project 5 (CMIP5) database to simulate the North-
4
ern Hemisphere winter climate following a large tropical volcanic eruption
5
is assessed. When sulfate aerosols are produced by volcanic emissions into
6
the tropical stratosphere and spread by the stratospheric circulation, it not
7
only causes globally averaged tropospheric cooling but also a localized heat-
8
ing in the lower stratosphere, which can cause major dynamical feedbacks.
9
Observations show a lower stratospheric and surface response during the fol-
10
lowing Northern Hemisphere (NH) winter, that resembles the positive phase
11
Physics, University of Oxford, Oxford, UK
2School of Geosciences, University of Edinburgh, UK
3National Centre of Atmospheric Sciences (NCAS), UK
4Department of Environmental Sciences, Rutgers University, New Brunswick, New Jersey, USA
5King Abdullah University of Science and Technology, Division of Physical Sciences and Engineering, Thuwal 23955-6900, Saudi Arabia
of the Northern Annular Mode for up to two years after the eruption. Sim-
12
ulations from 12 CMIP5 models that represent tropical eruptions in the 19th
13
and 20th century are examined, focusing on the regional impacts during the
14
NH winter season. The models generally fail to capture the NH dynamical
15
response following eruptions and tend to overestimate the cooling in the trop-
16
ical troposphere. The findings are confirmed by a superposed epoch analy-
17
sis of the North Atlantic Oscillation index for each model. The study con-
18
firms previous similar evaluations and raises concern for the ability of cur-
19
rent climate models to simulate the response of a major mode of global cir-
20
culation variability to external forcings. This is also of concern for the ac-
21
curacy of geoengineering modeling studies that seek to emulate the atmo-
22
spheric dynamical response to stratosphere-injected particles.
23
1. Introduction
For a volcano to have a significant long-term impact on the climate it must inject a
24
sufficient amount of sulfur containing gases into the stratosphere [Robock, 2000]. Once in
25
the stratosphere the sulfate gas undergoes a chemical reaction to produce sulfate aerosol.
26
The e-folding time of the sulfate gas to particle conversion is typically 35 days [Forster
27
et al., 2007]. Sulfate aerosol scatters back to space the incoming shortwave radiation
28
(SW) and also absorbs solar near infrared (NIR) radiation and upwelling long wave (LW)
29
radiation from the surface and atmosphere below [Stenchikov et al., 2006] (S06 hereafter).
30
For a given mass load, the scattering of SW radiation is modulated by the particle size
31
distribution and as the aerosol particle size increases, scattering of incoming SW radiation
32
decreases [Timmreck et al., 2009;Rasch et al., 2008]. The decrease in incoming shortwave
33
radiation results in a cooling of Earth’s surface [Robock and Mao, 1995]. The typical
34
e-folding lifetime for volcanic aerosols is about 12-14 months [Lambert et al., 1993; Baran
35
and Foot, 1994; Barnes and Hofmann, 1997]. However they can have a longer residence
36
time if they are injected into the tropics because of uplifting due to the Brewer-Dobson
37
circulation [Budyko, 1977; Stenchikov et al., 1998; Kravitz and Robock, 2011], causing
38
surface cooling for about two years following an eruption.
39
In contrast, localized equatorial heating of up to 3 K [Stenchikov et al., 2002] occurs
40
in the lower stratosphere due to the increase in near IR solar and LW absorption by the
41
sulfate aerosols. For a tropical volcanic eruption the heating in the tropical stratosphere
42
creates an anomalous temperature/density gradient between the equator and poles. By
43
the thermal wind relationship, this causes a strengthening of the zonal winds, which results
44
in a strengthened stratospheric polar vortex. In addition, reduced surface temperatures in
45
the tropical regions reduce the meridional surface temperature gradient, and this has been
46
associated with a reduction in the Eliassen Palm (EP) Flux - essentially, a measure of
47
planetary wave activity from the troposphere into the stratosphere [Andrews et al., 1987]
48
- and hence a stronger, less disturbed vortex. Further, chemical reactions which result in
49
ozone depletion serve to cool and strengthen the vortex, and the reduced temperatures
50
cause more NH ozone depletion, creating a positive feedback loop [Stenchikov et al., 2002].
51
A substantial body of research has indicated an influence of the stratospheric vortex
52
on high latitude circulations at Earth’s surface, with a strengthened vortex associated
53
with a positive North Atlantic Oscillation/Arctic Oscillation [Baldwin and Dunkerton,
54
2001]. The North Atlantic Oscillation (NAO) is an index corresponding to the difference
55
in mean sea level pressure (MSLP) between the Azores and Iceland [Rodwell et al., 1999;
56
Hurrell and Deser, 2009], and the Arctic Oscillation (AO) is defined as the first hemi-
57
spheric empirical orthogonal function (EOF) of sea level pressure variability [Thompson
58
and Wallace, 1998;Stenchikov et al., 2002]. Essentially the NAO can be thought of as the
59
AO over the Atlantic region [Christiansen, 2008]. A positive AO corresponds to anoma-
60
lously low pressure over the pole, and anomalously high pressure at the midlatitudes, and
61
vice versa for the AO. After large volcanic eruptions a positive phase of the AO has been
62
observed for the following 1 to 2 winters [Robock and Mao, 1992;Stenchikov et al., 2002].
63
The associated stronger westerly winds cause anomalous advection of warm oceanic air
64
overland, and this results in anomalously warm temperatures over major NH landmasses
65
(S06). Thus, as a result of the combined result of the surface cooling and lower strato-
66
spheric tropical heating, a dynamical feedback occurs during NH winter, which results in
67
surface warming over Northern America, Northern Europe and Russia [Robock and Mao,
68
1992]. Negative surface temperature anomalies in the Middle East are also a distinctive
69
feature of post-volcanic winters consistent with the positive phase of the AO (S06).
70
Climate model simulations of the historical period have, so far, been able to produce a
71
slightly strengthened stratospheric vortex, but much weaker than the observations, and
72
have failed to reproduce a positive AO and warming/cooling patterns over Eurasia and
73
the Middle East respectively for the two NH winters following volcanic eruptions. S06
74
analyzed seven models used for the Fourth Assessment Report of the Intergovernmental
75
Panel on Climate Change [IPCC, 2007]. They included all the models that specifically
76
represented volcanic eruptions by including a layer of aerosol, and excluded those that
77
either did not represent them, or represented them simply by a reduction in the solar
78
constant. They found that the temperature increase in the lower equatorial stratosphere,
79
caused by radiative heating from the aerosol, was reproduced by all the models. How-
80
ever, the models showed less agreement with the observed post eruption NH winter polar
81
lower stratospheric cooling. Further, the 50 hPa polar geopotential height (indicative of
82
the strength of the stratospheric polar vortex) in the models generally showed almost
83
no change whereas the observations show a large negative anomaly of about 200 m, re-
84
vealing a statistically significant stronger than average polar vortex at the 90% level.
85
Furthermore, the AO responses in the model simulations were significantly weaker than
86
in observations, indeed, Otter˚a [2008] notes that some model simulations show no AO
87
response. Correspondingly the strength and spatial pattern of the surface temperature
88
anomalies were not well reproduced.
89
Since the previous analysis of S06, who used simulations from the the World Climate Re-
90
search Programme’s (WCRP’s) Coupled Model Intercomparison Project phase 3 (CMIP3)
91
multi-model dataset [Meehl et al., 2007], climate models have undergone changes and im-
92
provements, spatial and vertical resolution have been increased, and many now include
93
the indirect aerosol effect, which is linked to cloud distribution which can be large but is
94
very uncertain. This is therefore likely to increase the spread of model responses in the
95
troposphere. In this study, we repeat the analysis of SO6 using CMIP5 model simulations
96
and focus our analysis on the impact of the largest volcanic eruptions on the NH winter
97
circulation. The models and experiments are described in section 2, results are presented
98
in section 3, and in section 4 we present our discussion and conclusions.
99
2. Models and Experiments
The model runs analyzed in this study come from the historic simulations of the climate
100
of the 20th century as standardized for the CMIP5. Models were forced with natural and
101
anthropogenic forcings from the late 19th century to the early 2000s. Although the ma-
102
jor external forcings (such as solar, greenhouse gases, land use) are standardized based
103
on the most recent observational databases, no specific recommendations were issued for
104
other forcings such as the stratospheric injection of sulfate aerosols from explosive volcanic
105
eruptions. As for the CMIP3, most modeling groups imposed the stratospheric emissions
106
for volcanic eruptions either from the reconstructions of Amman et al. [2003] (AM) or
107
from the updated version of Sato et al. [1993] (ST). The AM dataset provides monthly
108
latitudinal distributions of stratospheric optical depth for each volcanic event in 64 lati-
109
tude bands, computed with an explicit representation of the spread of the aerosol cloud,
110
taking into account the seasonal variations in stratospheric transport. A fixed particle
111
size distribution is assumed for all eruptions, with spherical droplets of sulfuric acid of
112
effective radius of 0.42 µm.The ST dataset provides latitudinal zonal mean stratospheric
113
optical depths for 4 layers between 15 km and 35 km together with variations of the par-
114
ticle’s effective radius based on the observations of the 1991 Mt. Pinatubo and 1982 El
115
Chich´on eruptions.
116
Unlike the other models, MRI-CGCM3 computes interactively the radiative effect of the
117
stratospheric volcanic aerosol. It includes the aerosol model MASINGAR mk-2 [Tanaka
118
et al., 2003], which calculates five species (sulfate, black carbon, organic carbon, mineral
119
dust, and sea-salt) of aerosols from emissions and other processes, including sulfate aerosol
120
of volcanic origin. The aerosol model is interactively coupled with the atmospheric compo-
121
nent that calculates radiation and cloud microphysics and utilizes the inventory of volcanic
122
SO2 emissions provided byStothers [1996],Bluth et al.[1997],Andres and Kasgnoc[1998],
123
andStothers[2001] and the optical properties of spherical sulfate aerosol droplets provided
124
by OPAC (Optical Properties of Aerosol and Clouds, [Hess et al., 1998]).
125
We restricted model analysis to those models that were both forced with volcanic aerosol
126
in the stratosphere and had at least 2 ensemble members, which yielded a total of 12
127
different climate models. The models with a brief description of the basic characteristics
128
are listed in Table 1. Not all variables were available for all models at the time of this
129
work: in Table 1 we list which variables are analyzed for each model. Only 2 models,
130
GISS-R and CCSM, in their updated version, are common to both our analysis and that
131
of S06.
132
Table 2 lists the nine major volcanic eruptions between 40◦S and 40◦N over 1883-present
133
day as well as the anomaly period, and the latitudes of eruption. Information for this
134
Table is taken from S06. As with S06 the eruptions listed in 2 are a subset of the volcanic
135
events analyzed by Robock and Mao [1992]. In the same approach as S06, high-latitude
136
eruptions from those studied by Robock and Mao are not included because they appear to
137
produce a qualitatively different effect on circulation than lower-latitude eruptions [Robock
138
and Mao, 1995; Oman et al., 2005; Kravitz and Robock, 2011]. The volcanoes listed in
139
Table 2 also correspond to the volcanoes south of 40◦N in Christiansen [2008] with the
140
caveat that we use different dates for the first winter after the eruptions of Santa Mar´ıa
141
and Fuego, shifting them forward one year with respect to Christiansen’s convention. The
142
implication of this choice is explored in section 3.4.
143
For comparison with observations the reanalysis of the 20th Century version 2 (20CRv2,
144
[Compo et al., 2011]) is employed. From this dataset we will use only near-surface tem-
145
perature and Mean Sea-Level Pressure (MSLP) fields for the period of 1871 to 2008.
146
More information about the database is provided at http://www.esrl.noaa.gov/psd/. The
147
ERA40 [Uppala et al., 2005] and NCEP/NCAR [Kistler et al., 2001] reanalysis fields are
148
also used to compare with middle atmosphere circulation changes during the winter season
149
for the largest eruptions after 1950.
150
To isolate the anomalies of the post-volcanic seasons and generate the average volcanic
151
composite, we adopt the same averaging procedure employed by S06, choosing a different
152
reference time for each eruption and averaging two winter seasons after each eruption.
153
Using a large number of eruptions and minimum of two ensemble members per model
154
should help to average out spurious effects, for example due to incorrect sampling of the
155
El Ni˜no Southern Oscillation (ENSO) cycles, which cannot be controlled in these coupled
156
ocean atmosphere simulations.
157
The statistical significance of anomalies from the mean climatology is evaluated with a
158
local two-tailed t-test. Christiansen [2008] showed through analysis of observations that
159
the largest volcanic eruptions of the 20th Century tend to be followed by a positive index
160
of the North Atlantic Oscillation (NAO). He noted that the NAO signal is strongest and
161
significant in the first year after the eruption and does not appear to be influenced by
162
ENSO events or by the specific volcanic eruption chosen for the composite.
163
The NAO index for all the model ensembles is computed to test whether the simu-
164
lated dynamical response to volcanic forcing projects onto the NAO index as observed by
165
Christiansen [2008] in the observations. The NAO index is computed for each ensemble of
166
each model, as in Christiansen [2008]: the EOFs are calculated from the monthly winter
167
MSLP anomalies north of 20◦N and between 110◦W and 70◦E for the period 1948-2000,
168
weighted by the square root of the grid area. The index is then computed for the total
169
period 1860-2000 by projecting the monthly anomalies of the mean sea level pressure onto
170
the first EOF. Model results are compared to the same index computed from the MSLP
171
fields from the 20CRv2 dataset using a superposed epoch analysis for the nine volcanic
172
eruptions listed in Table 2 for each model. The eruptions in each ensemble member are
173
considered to be independent events. The effective number of eruptions for each model
174
depends on the number of the ensemble members.
175
The statistical significance of the epoch analysis is estimated using 5,000 random vol-
176
canic surrogate composites drawn from the whole epoch matrix to preserve the structure
177
of the sample and comparing the distribution obtained for each time lag to the epoch
178
results. The normalization procedure adopted in [Adams et al., 2003] was used to avoid
179
possible biases due to single outliers in each volcanic window.
180
We also tested for the occurrence of positive NAO for both in the first and second
181
post-volcanic winter and its significance is tested using a bi-nomial distribution with
182
the probability of the single event (σ) estimated from the full timeseries. As noted in
183
Christiansen [2008],σ is in general different from 0.5 which is the probability distribution
184
of the NAO index not being normal.
185
The main conclusions are robust with respect to the definition of the winter season
186
(DJF or DJFM) and we will present here the results for the NAO index computed for the
187
DJF composite to allow comparison with previous results in the literature.
188
3. Results
3.1. Direct radiative effect of volcanic aerosol
Due to a lack of direct information on the radiative forcing of volcanic aerosol for each
189
model, we choose to use the time series of the anomalies in the reflected short wave
190
(SW) radiation at the top of atmosphere (TOA) (Fig. 1) as a rough proxy for the global
191
radiative effect of the stratospheric aerosol, as in S06 (their Fig. 1). All the models
192
perform consistently and show the increase in the reflected SW radiation corresponding
193
to the major explosive eruptions and do not show any appreciable differences compared
194
with the AR4 models shown in S06. The largest anomaly in the reflected SW radiation
195
is observed for the bcc-csm1-1 model whereas MRI-CGCM3 simulates the lowest signal
196
among the models. MRI-CGCM3 computes interactively the effect of the volcanic aerosol
197
from the stratospheric SO2 load and shows a lower scattering efficiency of incoming SW
198
radiation with respect to the other models.
199
As noted in S06, larger spread among the model response is observed for the early
200
eruptions and less uncertainty appears for the most recent El Chich´on and Pinatubo
201
events. Notably, the largest effect on the reflected SW radiation for the eruptions pre-
202
1900 is observed in the models that adopt the [Amman et al., 2003] reconstruction.
203
3.2. Surface temperature and mean sea level pressure
Fig. 2 and 3 shows the NH composites of surface temperature and MSLP for the post-
204
volcanic winter season (as given in the fourth column in Table 2) in the reanalysis and
205
models. Fig. 2 shows in the reanalysis the well known significant surface warming signal
206
over northern Europe and Asia. The anomalies are about 1-2 K over Scandinavia and
207
central-eastern Asia. Significant cooling is observed over NE Canada and also, though
208
not significant, over the Middle East. As noted in S06, a warming signal also appears on
209
the Eastern Pacific but this could be spurious due to a positive ENSO sampling bias. A
210
general cooling is observed in the Tropical region, although weak and barely significant.
211
The surface temperature in the Arctic region appears unusually warm, but the reliability of
212
the reconstructed lower tropospheric temperature at high latitudes reduces the significance
213
of the anomaly [Compo et al., 2011].
214
Large variability among the models is observed in the NH response: the observed warm-
215
ing in the northern Eurasia is simulated by a few models but is much weaker than in the
216
observations. For example, GISS-E2-H and GISS-E2-R simulate the northern European
217
warming reasonably well but the maximum amplitude is only 0.5K. The cooling over
218
NE Canada seems to be simulated more widely, independent of how well the northern
219
Eurasian warming is captured. Some models (CSIRO-Mk3.6, HadGEM2-ES, NorESM1)
220
simulate a general cooling in the Asian-European area, opposite to the observations, and
221
the majority show a significant cooling in the tropical lower latitudes, of order of 0.2K
222
and over all basins.
223
The observed surface temperature anomalies in the NH post-volcanic winters are closely
224
related to changes in the winter circulation as confirmed by the mean sea-level pressure
225
(MSLP) anomalies (Fig. 3). In agreement with previous studies (e.g., S06), in the reanal-
226
ysis a significant positive NAO-like pattern marks the north-Atlantic region, with negative
227
pressure anomalies in the Arctic region and positive over the north Atlantic. Notice that
228
the minimum and maximum of the positive anomaly are both displaced northward with
229
respect to the position of the canonical maximum of the leading variability mode in the
230
MSLP anomalies in the region [Hurrell and Deser, 2009].
231
Again, there are large differences in the MSLP patterns in the CMIP5 models as shown
232
in Fig. 3. Only CNRM-CM5 and CanESM2 reproduce a weak dipole over the north
233
Atlantic, whereas NorESM1 shows anomalies opposite to the observed. The other models
234
show only weak anomalies with weak statistically significance. The two GISS models,
235
although managing to simulate weak but reasonable surface temperature anomalies, do
236
not show any significant anomaly in the MSLP.
237
The analysis of the surface temperature and MSLP hence shows a poor agreement of
238
the historic CMIP5 runs with the observations in the first two NH winters after large
239
tropical eruptions and no improvement is seen with respect to the findings of S06 based
240
on seven models participating to the CMIP3.
241
3.3. Geopotential
The anomalies in the geopotential in the upper troposphere and mid stratosphere help
242
to define the vertical structure of the circulation changes in the post volcanic NH winter
243
seasons. Due to the high uncertainty in the 20CRv2 reconstructions of upper air fields
244
[Compo et al., 2011], we decide to analyse only the last four eruptions since 1950 using the
245
ERA40 dataset. In the upper troposphere (Figure 4), the observed 200 hPa geopotential
246
anomalies are linked to the MSLP anomalies, with a general decrease over the North Pole
247
surrounded by positive geopotential in the mid latitudes and a strong dipole over the
248
North Atlantic region. A general decrease in the geopotential dominates at low latitudes,
249
consistent with the generalized cooling tendency observed in the tropical troposphere.
250
The anomaly pattern in the troposphere is mirrored in the stratosphere by a cold and
251
deep polar night vortex, as observed in the 50 hPa geopotential anomalies (Figure 5)
252
showing a large statistically significant decrease in geopotential height over the pole of
253
around 200 m (Fig. 5). A weaker anomaly at 50 hPa is observed at low latitude, with
254
a geopotential increase of about 25 m which has been attributed to the direct heating
255
effect of the volcanic aerosol in the lower tropical stratosphere [Ramachandran et al.,
256
2000; Stenchikov et al., 1998]. The low 50 hPa geopotential at high latitude is associated
257
with a colder polar lower stratosphere, which suggests a stronger and persistent polar
258
vortex. Recent studies suggest that this might be a characteristic of the early stage of the
259
post-volcanic winter season. For example Graf et al. [2007] observed no clear weakening
260
of the wave activity during post-volcanic winter and Mitchell et al. [2011] show that the
261
polar vortex in the upper stratosphere is weaker than normal from the end of January
262
into February after the three major volcanic eruptions since 1960.
263
A few of the models capture a similar structure in the stratosphere (see Fig. 5) as in
264
the reanalysis, though much weaker. HadGEM2, MPI, CNRM-CM5 and MRI simulate
265
a decrease in the geopotential of order of 25 m, but other models show no significant
266
anomaly at high latitudes. Most of the models reproduce a statistical significant increase
267
in the geopotential at low latitude in agreement with the observations.
268
As observed for the MSLP, the modelled geopotential anomalies at 200 hPa are highly
269
variable (Fig. 4). The dipole between high and mid latitudes is weakly reproduced by
270
four models (HadGEM2-ES, CNRM-CM5, MPI and MRI). Three of them also simulate
271
a significant uniform decrease in the geopotential roughly south of 30N, stronger than in
272
the observations. Some of the other models (GISS-E2-H, Can-ESM2, MPI) simulate the
273
opposite anomaly, with a general increase in the geopotential at low latitudes, in contrast
274
to S06 where all of the analysed models showed a decrease in the geopotential in the
275
tropical and sub-tropical latitudes.
276
3.4. NAO index
As noted in section 3.2, the observed anomalies in the MSLP in the post-volcanic winters
277
are not well reproduced by the CMIP5 models. The observed MSLP anomalies in the
278
winters after the largest volcanic eruptions since 1880, project onto the leading variability
279
mode of the NH circulation, especially the NAO index, with a significant prevalence of
280
positive NAO in the first winter after the eruption [Fischer et al., 2007], both in terms of
281
amplitude and number of positive events [Christiansen, 2008].
282
In this section we test whether looking at the principal modes of variability can help
283
to better isolate the dynamic response in the model simulations. As mentioned in section
284
2, we use the same time convention adopted by Fischer et al. [2007] and S06 to identify
285
the 1st and 2nd winter after each eruption. The majority of the volcanoes erupted in
286
the spring-early summer but two erupted in autumn, the minor eruption of the Fuego in
287
October 1974 and the large eruption of Santa Mar´ıa at the end of October 1902. It can be
288
argued that their full effect is unlikely to be present in the first winter immediately after
289
the eruption and therefore the first winter should be considered to be a full year after the
290
eruption time, as listed in our Table 2. This differs from the time convention adopted
291
by Christiansen [2008] who considered the first winter immediately after the eruption for
292
all the volcanoes, hence changing the years of winters considered for the two eruptions
293
of Fuego and Santa Mar´ıa. In his paper he reported the robustness of his results when
294
those two eruptions are excluded from the analysis. However, we show here that with the
295
different dating convention the results are affected when these two eruptions are included.
296
When all nine eruptions south of 40◦N as listed in Table 2 are included, the 20CRv2
297
shows a clear prevalence of positive NAO index in the first year after the eruptions (Fig.
298
6, 20CRv2, lag 0). The amplitude is significant at the 5% level with seven volcanoes out
299
of nine with positive NAO in the first winter and the occurrence is significant at the 9%
300
level. No significant signals are observed for the second post-volcanic winter.
301
Only two post-volcanic winters show a negative NAO, after the eruptions of Agung and
302
Quizapu, which both erupted in the southern hemisphere. Quizapu has the weakest effect
303
on the stratospheric optical depth between 30◦S and 30◦N among all the analyzed volca-
304
noes (S06) and Agung’s aerosol was mostly concentrated south of the Equator [Robock,
305
2000]. This could affect the dynamics associated with the forcing of the NAO circulation.
306
It has also to be noted that the winter 1903-04 after the Santa Mar´ıa eruption has a NAO
307
signal close to zero (0.03, also consistent in the DJFM composite with -0.04 as confirmed
308
in Christiansen [2008], his Fig. 2). This brings the effective number of sizable positive
309
NAO events to six, which is not significant at the 10% level.
310
Among the 12 models analysed in this study, four models show a positive NAO signal
311
at lag 0 (Fig. 7, HadCM3 at the 5% significance level, GISS-E2-R at the 9% significance
312
level, CanESM2 at the 2% significance level and CNRM-CM5 at the 3% significance
313
level) but only CNRM-CM5 and CanESM2 show also a significant number of events at
314
lag 0 (52/90, p=0.1 for CNRM-CM5 and 28/45, p=0.07 for CanESM2). The analysis is
315
confirmed by the MSLP gridded anomalies shown in Fig. 3 where these two models also
316
show a weak NAO-like dipole even when averaged across 2 winter seasons. The MRI is
317
the only model that shows a positive NAO signal in the second winter after the eruptions
318
(5% level, p=0.08) but the model appears to have a positive NAO at all lag times, so it
319
is not clear whether this response is necessarily associated with the volcanic eruption.
320
The other models show no significant positive anomaly at lag 0, but many spurious
321
signals are detected at various lags for different models. CSIRO-Mk3.6 displays a negative
322
NAO at lag 0, while other models (NorESM1-M and CCSM4) show negative NAO at lag
323
-1. HadCM3 and CNRM-CM5 detect a positive NAO at lag -3: the signal could partially
324
be due to the degenerate contribution of the Krakatau eruptions that happens 3 years
325
before the 1886 eruption of Tarawera and shows a positive NAO in both of these models
326
(not shown). Finally, no strong signals are displayed by HadGEM2-ES at lag -1 and
327
NorESM1-M at lag +1: such signals could both be unphysical and occur by chance or
328
they could also depend on periodicity sampled in the epoch analysis at the same frequency
329
of the volcanic signal. We have not analyzed in detail the origin of the spurious result of
330
these two models.
331
As mentioned above, when a different convention is used to identify the closest winters
332
affected by the eruption of Santa Mar´ıa and Fuego, changes are observed in the epoch
333
analysis. Figure 7 examines the robustness of the analysis with respect to the choice of the
334
winters after Santa Mar´ıa and Fuego, using the convention adopted inChristiansen[2008].
335
Since the reanalysis are based on a limited sample, they prove to be highly sensitive to
336
changes in the epoch key date. The signal at lag 0 becomes now highly significant (1%
337
level) with an occurrence of 7 positive NAO out of 9 events (p=0.09). Most of the change
338
in the signal comes from the Santa Mar´ıa event, which shows a strong positive NAO in the
339
winter 1902-1903, immediately after the eruption and positively contributes to enhance
340
the epoch composite at lag 0.
341
The largest effect of the change of the year of the first winter after the eruptions of
342
Santa Mar´ıa and Fuego is observed for HadGEM2, which does not detect any significant
343
signal at any lag. CanESM2 and CNRM-CM5 both still detect a positive NAO at lag 0
344
(respectively at 5% and 2% level) but the number of events is now significant only for
345
CNRM-CM5 (54/90, p=0.04). HadCM3 shows a weaker positive NAO at lag 0 that is
346
not significant at the 10% level. Again, the analysis detects a positive NAO signal at lag
347
1 for MRI (10%) but with only 16 cases out of 27, it is not significant at the 10% level.
348
The main conclusions of this section are 1) the superposed epoch analysis of the 20CRv2
349
NAO index confirms previous findings of a positive NAO in the first winter following the
350
major tropical eruptions in the 19th and 20th century, but the strength of the signal is
351
sensitive to the choice of the key dates for each eruption. 2) as observed in the previous
352
sections, the models struggle to reproduce a detectable positive NAO signal in the first
353
post-eruption winter. Even with 10 ensemble members, the CNRM-CM5 and CSIRO-
354
Mk3.6 model results are sensitive to changes in the definition of the post-volcanic key
355
dates.
356
4. Discussion and Conclusions
All available models submitted to the CMIP5 archive as of December 2011 that had
357
a reasonably realistic representation of volcanic eruptions and number of samples have
358
been analyzed for their ability to simulate post-volcanic radiative and dynamic responses.
359
With substantially different dynamics between the models it was hoped to find at least one
360
model simulation that was dynamically consistent with observations, showing improve-
361
ment since S06. Disappointingly, we found that again, as with S06, despite relatively
362
consistent post volcanic radiative changes, none of the models manage to simulate a suf-
363
ficiently strong dynamical response. Although all the models reproduce reasonably well
364
the increase in geopotential height in the lower stratosphere at low latitudes, none of the
365
models simulate a sufficiently strong reduction in the geopotential height at high lati-
366
tudes and correspondingly the MSLP pressure fields and temperature fields show major
367
differences with respect to the observed anomalies. This is despite some models having
368
10 ensemble members, giving a strong signal to noise ratio.
369
It is unclear why models fails to simulate the dynamics following volcanic eruptions.
370
The dynamical mechanism proposed by Stenchikov et al. [2002] (their Fig. 13), involves
371
lower stratosphere tropical heating caused by the presence of volcanic aerosols which gives
372
rise to a stronger polar vortex due to the thermal wind relationship. A stronger vortex
373
also could be due to a decrease in planetary wave forcing from the troposphere, although
374
the evidence for this is unclear. The modelling results ofStenchikov et al.[2004] showed a
375
decreased EP flux into the stratosphere following the Pinatubo eruption but observations
376
suggest an increase in the EP flux following the Agung, Fuego, El Chich´on and Pinatubo
377
eruptions [Graf et al., 2007].
378
There are therefore still uncertainties in the dynamical mechanisms following volcanic
379
eruptions for example regarding the wave propagation through the polar stratosphere as
380
seen in EP flux diagnostics [Graf et al., 2007;Bozzo and Tett, 2011].
381
In addition, the degree of El Ni˜no influence and interaction following volcanic eruptions
382
is unknown. Based on the epoch analysis of post-volcanic winters stratified according
383
to the ENSO phase, [Christiansen, 2008] concluded that the ENSO does not change the
384
impact of volcanic eruptions on the Northern Hemisphere winter circulation, although
385
the low number of cases imposes caveats on the conclusions. A recent work [Graf and
386
Zanchettin, 2012] argues that ENSO has a different effect on the Northern Hemispheric
387
winter circulation when the differences between Central-Pacific (CP) and East-Pacific
388
(EP) El Ni˜no events are taken into account. In particular, CP El Ni˜no events appear to
389
have a significant effect on winter NH circulation, with a tendency towards a negative NAO
390
index. According to their definition, CP El Ni˜no occurred in 1963-1964 and 1991-1992 but
391
not in 1982-1983, which could explain the strong Eurasian warming signal observed after El
392
Chich´on, even though a strong El Ni˜no event was taking place, and the relatively disturbed
393
vortex in January 1992 [Graf et al., 2007]. Moreover, biases in model representations of
394
ENSO variability [Guilyardi, 2006] could in the same way affect their response to volcanic
395
forcing. The issue is also complicated by the intrinsic problems in defining the modes of
396
ENSO variability [Takahashi et al., 2011]. In our analysis the large number of ensemble
397
members should help to smooth out possible contaminations induced by the Pacific SST
398
variability.
399
Finally, Stenchikov et al. [2004] found that including the Quasi-Biennial Oscillation
400
(QBO) in the model made a substantial difference to the volcanic impact on the vortex.
401
They found in observations following the Pinatubo eruption that the vortex was strength-
402
ened more in the second winter than the first, despite more aerosol being present in the
403
stratosphere in the first winter. They proposed that this could be explained by the QBO
404
being in the East phase in the first winter, which tends to weaken the vortex, and was in
405
the West phase in the second winter, which tends to strengthen it. They concluded that a
406
model with a QBO in the correct phase could better represent the dynamical simulation
407
of the Pinatubo eruption. We note here that none of the models tested have a QBO in
408
them, which could affect the performance of the dynamical simulation.
409
Another factor which could account for the poor simulation of the dynamical response
410
following a volcanic eruption is related to how the aerosol is imposed in the model. We
411
note that it is typical for a model to employ a very crude representation of aerosol in four
412
latitude bands [Marshall et al., 2009], and the question of the suitability of this aerosol
413
representation has been raised before [Otter˚a, 2008;Marshall et al., 2009]. Another reason
414
for the “common failure” of models to simulate the dynamics following volcanic eruptions
415
may be their representation of the AO. Otter˚a [2008] notes that it may be that models
416
have a general basic inadequacy that does not allow a sufficiently strong AO response to
417
large-scale forcing. Others have pointed to ozone as being an important factor [Stenchikov
418
et al., 2002; Otter˚a, 2008], however, as noted by Marshall et al. [2009] the response to
419
the past major eruptions (before major ozone loss and larger amounts of ozone destroying
420
chlorine in the atmosphere) is similar to that of El Chich´on and Pinatubo combined,
421
which suggests that inclusion of ozone chemistry is unlikely to be a major factor in the
422
simulation of a volcanic eruption.
423
The impact of volcanic eruptions on surface climate is the closest natural analogue to
424
sulfate aerosol geoengineering, despite the differences in injection method and duration of
425
the perturbation. Unlike sulfate aerosol geoengineering, the ability of models to accurately
426
reproduce the response to volcanic eruptions can be tested against observations. Despite
427
it being likely that a more uniform profile of aerosol in the stratosphere would occur
428
from geoengineering than following volcanic eruptions, the results of GCM simulations
429
of stratospheric geoengineering need to be considered in the light of their limitations
430
when it comes to certain aspects of their responses to volcanic eruptions. This is of
431
concern not only for the temperature response, but also for the precipitation response, as
432
the dynamical effects following an eruption can often overwhelm the radiative response
433
[Anchukaitis et al., 2010]. Accordingly, research into the climate response to volcanic
434
eruptions and their simulations is an area of major importance, not only in its own right,
435
but for stratospheric aerosol geoengineering.
436
Acknowledgments. A.R. is supported by NSF grant ATM-0730452. Simon Driscoll
437
acknowledges financial support from the SPICE (Stratospheric Particle Injection for Cli-
438
mate Engineering) project, jointly funded by the UK EPSRC (Engineering and Physical
439
Sciences Research Council) and NERC (Natural Environment Research Council). Lesley
440
Gray is funded by the UK NERC National Centre for Atmospheric Research (NCAS)
441
Climate Directorate. Alessio Bozzo is jointly supported by NCAS and the NSF grant
442
ATM-0296007 and acknowledges the support of the SAGES Centre for Earth System
443
Dynamics at the University of Edinburgh.
444
References
Adams, B. J., M. E. Mann, and C. M. Ammann (2003), Proxy evidence for an El Ni˜no-like
445
response to volcanic forcing, Nature,426(6964), 274–278.
446
Amman, C. M., G. A. Meehl, W. M. Washington, and C. S. Zender (2003), A monthly and
447
latitudinally varying volcanic forcing dataset in simulations of 20th century climate.,
448
Geophys. Res. Lett., 30, 1657, doi:10.1029/2003GL016,875.
449
Anchukaitis, K. J., B. M. Buckley, E. R. Cook, B. I. Cook, R. D. D’Arrigo, and C. M.
450
Ammann (2010), Influence of volcanic eruptions on the climate of the asian monsoon
451
region, Geophys. Res. Lett.,37(22), doi:10.1029/2010GL044843.
452
Andres, R. J., and A. D. Kasgnoc (1998), A time-averaged inventory of subaerial volcanic
453
sulfur emissions, J. Geophys. Res., 103(D19), 25,251–25,261, doi:10.1029/98JD02091.
454
Andrews, D., J. Holton, and C. Leovy (1987), Middle Atmosphere Dynamics, Academic
455
Pr, Orlando, Fla.
456
Baldwin, M. P., and T. J. Dunkerton (2001), Stratospheric harbingers of anomalous
457
weather regimes, Science, 294(5542), 581–584, doi:10.1126/science.1063315.
458
Baran, A. J., and J. S. Foot (1994), New application of the operational sounder hirs
459
in determining a climatology of sulphuric acid aerosol from the pinatubo eruption, J.
460
Geophys. Res.,99(D12), 25,673–25,679, doi:10.1029/94JD02044.
461
Barnes, J. E., and D. J. Hofmann (1997), Lidar measurements of stratospheric
462
aerosol over mauna loa observatory, Geophys. Res. Lett., 24(15), 1923–1926, doi:
463
10.1029/97GL01943.
464
Bluth, G. J. S., W. I. Rose, I. E. Sprod, and A. J. Krueger (1997), Stratospheric loading
465
of sulfur from explosive volcanic eruptions, J. of Geology, 105(6), 671–684.
466
Bozzo, A., and S. Tett (2011), The northern annular mode and large volcanic eruptions:
467
the role of the early-winter conditions, submitted to J. Geophys. Res., -, –.
468
Budyko, M. I. (1977), Climatic Changes, vol. 261 pp., American Geophysical Union.
469
Christiansen, B. (2008), Volcanic Eruptions, Large-Scale Modes in the Northern Hemi-
470
sphere, and the El Ni˜no–Southern Oscillation, J. Clim., 21, 910–922.
471
Chylek, P., J. Li, M. K. Dubey, M. Wang, and G. Lesins (2011), Observed and model
472
simulated 20th Century Arctic temperature variability: Canadian Earth System Model
473
CanESM2, Atmospheric Chemistry and Physics Discussions,11(8), 22,893–22,907, doi:
474
10.5194/acpd-11-22893-2011.
475
Collins, M., S. F. B. Tett, and C. Cooper (2001), The internal climate variability of
476
HadCM3, a version of the Hadley Centre coupled model without flux adjustments,
477
Climate Dynamics,17, 61–81, 10.1007/s003820000094.
478
Collins, W. J., et al. (2011), Development and evaluation of an Earth-System model
479
- HadGEM2, Geoscientific Model Development, 4(4), 1051–1075, doi:10.5194/gmd-4-
480
1051-2011.
481
Compo, G. P., et al. (2011), The twentieth century reanalysis project, Q. J. R. Meteorol.
482
Soc.,137(654), 1–28, doi:10.1002/qj.776.
483
Fischer, E. M., J. Luterbacher, E. Zorita, S. F. B. Tett, C. Casty, and H. Wanner (2007),
484
European climate response to tropical volcanic eruptions over the last half millennium,
485
Geophys. Res. Lett., 34(5), doi:10.1029/2006GL027992.
486
Forster, P., et al. (2007), Changes in atmospheric constituents and in radiative forcing,
487
in Climate Change 2007: The Physical Science Basis. Contribution of Working Group
488
I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change,
489
edited by S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. Averyt, M. Tignor,
490
and H. Miller, ISBN 978-0-521-88009-1 (pb: 978-0-521-70596-7), Cambridge University
491
Press, Cambridge, United Kingdom and New York, NY, USA.
492
Gent, P. R., et al. (2011), The Community Climate System Model Version 4, Journal of
493
Climate, 24(19), 4973–4991, doi:10.1175/2011JCLI4083.1.
494
Graf, H.-F., and D. Zanchettin (2012), Central Pacific El Ni˜no, the “subtropical bridge”
495
and Eurasian climate, J. Geophys. Res.,117(D1), doi:10.1029/2011JD016493.
496
Graf, H.-F., Q. Li, and M. A. Giorgetta (2007), Volcanic effects on climate: revisiting the
497
mechanisms, Atmospheric Chemistry and Physics, 7(17), 4503–4511, doi:10.5194/acp-
498
7-4503-2007.
499
Guilyardi, E. (2006), El Ni˜no–mean state–seasonal cycle interactions in a multi-model
500
ensemble, Climate Dynamics, 26, 329–348, 10.1007/s00382-005-0084-6.
501
Hess, M., P. Koepke, and I. Schult (1998), Optical Properties of Aerosols and Clouds:
502
The software package OPAC, Bull. Am. Met. Soc., 79, 831–844.
503
Hurrell, J. W., and C. Deser (2009), North Atlantic climate variability: The role of
504
the North Atlantic Oscillation, Journal of Marine Systems, 78(1), 28 – 41, doi:
505
10.1016/j.jmarsys.2008.11.026.
506
IPCC (2007),Climate Change 2007: Synthesis Report. Contribution of Working Groups I,
507
II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate
508
Change, IPCC, Geneva, Switzerland, 104 pp.
509
Kirkev˚ag, A., T. Iversen, Ø. Seland, J. B. Debernard, T. Storelvmo, and J. E. Kristj`ansson
510
(2008), Aerosol-cloud-climate interactions in the climate model cam-oslo, Tellus A,
511
60(3), 492–512, doi:10.1111/j.1600-0870.2008.00313.x.
512
Kirkev˚ag, A., T. Iversen, Ø. Seland, H. Struthers, C. Hoose, and D. Nilsson (2011),
513
Aerosols and their radiative effects on climate in cam4-oslo/noresm. the importance of
514
natural aerosols for estimates of anthropogenic impacts, In prep.
515
Kistler, R., et al. (2001), The NCEP–NCAR 50–Year Reanalysis: Monthly Means CD–
516
ROM and Documentation,Bulletin of the American Meteorological Society,82(2), 247–
517
267, doi:10.1175/1520-0477(2001)082.
518
Kravitz, B., and A. Robock (2011), Climate effects of high-latitude volcanic eruptions:
519
Role of the time of year, J. Geophys. Res., 116(D1), doi:10.1029/2010JD014448.
520
Lambert, A., R. G. Grainger, J. J. Remedios, C. D. Rodgers, M. Corney, and F. W. Taylor
521
(1993), Measurements of the evolution of the Mt. Pinatubo aerosol cloud by ISAMS,
522
Geophys. Res. Lett., 20(12), 1287–1290, doi:10.1029/93GL00827.
523
Marshall, A. G., A. A. Scaife, and S. Ineson (2009), Enhanced seasonal prediction of eu-
524
ropean winter warming following volcanic eruptions, Journal of Climate,22(23), 6168–
525
6180, doi:10.1175/2009JCLI3145.1.
526
Marsland, S., H. Haak, J. Jungclaus, M. Latif, and F. Roeske (2003), The Max-Planck-
527
Institute global ocean/sea ice model with orthogonal curvilinear coordinates., Ocean
528
Modell, 5, 91–127.
529
Meehl, G., C. Covey, T. Delworth, M. Latif, B. McAvaney, J. Mitchell, R. Stouffer, and
530
K. Taylor (2007), The WCRP CMIP3 multi-model dataset: a new era in climate change
531
research, Bulletin of the American Meteorological Society, 88, 1383–1394.
532
Mitchell, D. M., L. J. Gray, and A. J. Charlton-Perez (2011), The structure and evolution
533
of the stratospheric vortex in response to natural forcings, J. Geophys. Res.,116(D15),
534
doi:10.1029/2011JD015788.
535
Mizuta, R., et al. (2012), Climate simulations using MRI-AGCM3.2 with 20-km grid, J.
536
Meteor. Soc. Japan,in press.
537
Oman, L., A. Robock, G. Stenchikov, G. A. Schmidt, and R. Ruedy (2005), Cli-
538
matic response to high-latitude volcanic eruptions, J. Geophys. Res., 110(D13), doi:
539
10.1029/2004JD005487.
540
Otter˚a, O. (2008), Simulating the effects of the 1991 Mount Pinatubo volcanic eruption
541
using the ARPEGE atmosphere general circulation model, Advances in Atmospheric
542
Sciences, 25, 213–226, 10.1007/s00376-008-0213-3.
543
Raddatz, T., C. Reick, W. Knorr, J. Kattge, E. Roeckner, R. Schnur, K.-G. Schnitzler,
544
P. Wetzel, and J. Jungclaus (2007), Will the tropical land biosphere dominate the
545
climate–carbon cycle feedback during the twenty-first century?, Climate Dynamics,29,
546
565–574, 10.1007/s00382-007-0247-8.
547
Ramachandran, S., V. Ramaswamy, G. L. Stenchikov, and A. Robock (2000), Radiative
548
impact of the Mount Pinatubo volcanic eruption: Lower stratospheric response, J.
549
Geophys. Res.,105(D19), 24,409–24,429.
550
Rasch, P. J., S. Tilmes, R. P. Turco, A. Robock, L. Oman, C.-C. J. Chen, G. L. Stenchikov,
551
and R. R. Garcia (2008), An overview of geoengineering of climate using stratospheric
552
sulphate aerosols, Philosophical Transactions of the Royal Society A: Mathematical,
553
Physical and Engineering Sciences, 366(1882), 4007–4037, doi:10.1098/rsta.2008.0131.
554
Robock, A. (2000), Volcanic eruptions and climate, Rev. Geophys., 38(2), 191–219.
555
Robock, A., and J. Mao (1992), Winter warming from large volcanic eruptions, Geophys.
556
Res. Lett., 19(24), 2405–2408, doi:10.1029/92GL02627.
557
Robock, A., and J. Mao (1995), The volcanic signal in surface temperature observations,
558
J. Clim., 8, 1086–1103.
559
Rodwell, M. J., D. P. Rowell, and C. K. Folland (1999), Oceanic forcing of the wintertime
560
North Atlantic Oscillation and European climate, Nature, 398(6725), 320–323.
561
Rotstayn, L. D., M. A. Collier, M. R. Dix, Y. Feng, H. B. Gordon, S. P. O’Farrell, I. N.
562
Smith, and J. Syktus (2010), Improved simulation of australian climate and enso-related
563
rainfall variability in a global climate model with an interactive aerosol treatment,
564
International Journal of Climatology,30(7), 1067–1088, doi:10.1002/joc.1952.
565
Sato, M., J. Hansen, M. McCormick, and J. Pollack (1993), Stratospheric aerosol optical
566
depths, J. Geophys. Res., 98, 22,987–22,994, doi:10.1029/93JD02,553.
567
Schmidt, G. A., et al. (2006), Present-Day Atmospheric Simulations Using GISS ModelE:
568
Comparison to In Situ, Satellite, and Reanalysis Data,Journal of Climate,19(2), 153–
569
192, doi:10.1175/JCLI3612.1.
570
Stenchikov, G., A. Robock, V. Ramaswamy, M. D. Schwarzkopf, K. Hamilton, and S. Ra-
571
machandran (2002), Arctic Oscillation response to the 1991 Mount Pinatubo erup-
572
tion: Effects of volcanic aerosols and ozone depletion, J. Geophys. Res., 107(D24),
573
doi:10.1029/2002JD002090.
574
Stenchikov, G., K. Hamilton, A. Robock, V. Ramaswamy, and M. D. Schwarzkopf (2004),
575
Arctic oscillation response to the 1991 Pinatubo eruption in the SKYHI general cir-
576
culation model with a realistic quasi-biennial oscillation, J. Geophys. Res., 109(D3),
577
doi:10.1029/2003JD003699.
578
Stenchikov, G., K. Hamilton, R. J. Stouffer, A. Robock, V. Ramaswamy, B. Santer, and
579
H.-F. Graf (2006), Arctic Oscillation response to volcanic eruptions in the IPCC AR4
580
climate models, J. Geophys. Res., 111(D7), doi:10.1029/2005JD006286.
581
Stenchikov, G. L., I. Kirchner, A. Robock, H.-F. Graf, J. C. Antu˜na, R. G. Grainger,
582
A. Lambert, and L. Thomason (1998), Radiative forcing from the 1991 mount pinatubo
583
volcanic eruption, J. Geophys. Res.,103(D12), 13,837–13,857, doi:10.1029/98JD00693.
584
Stothers, R. B. (1996), Major optical depth perturbations to the stratosphere from vol-
585
canic eruptions: Pyrheliometric period, 1881 1960, J. Geophys. Res., 101(D2), 3901–
586
3920, doi:10.1029/95JD03237.
587
Stothers, R. B. (2001), Major optical depth perturbations to the stratosphere from vol-
588
canic eruptions: Stellar extinction period, 1961 1978,J. Geophys. Res.,106(D3), 2993–
589
3003, doi:10.1029/2000JD900652.
590
Takahashi, K., A. Montecinos, K. Goubanova, and B. Dewitte (2011), ENSO regimes:
591
Reinterpreting the canonical and Modoki El Ni˜no, Geophys. Res. Lett., 38(10), doi:
592
10.1029/2011GL047364.
593
Tanaka, T. Y., K. Orito, T. T. Sekiyama, K. Shibata, M. Chiba, and H. Tanaka (2003),
594
MASINGAR, a global tropospheric aerosol chemical transport model coupled with
595
MRI/JMA98 GCM: Model description, Meteor. Geophys., 53(4), 119–138.
596
Thompson, D. W. J., and J. M. Wallace (1998), The Arctic oscillation signature in the
597
wintertime geopotential height and temperature fields,Geophys. Res. Lett.,25(9), 1297–
598
1300, doi:10.1029/98GL00950.
599
Timmreck, C., S. J. Lorenz, T. J. Crowley, S. Kinne, T. J. Raddatz, M. A. Thomas,
600
and J. H. Jungclaus (2009), Limited temperature response to the very large AD 1258
601
volcanic eruption, Geophys. Res. Lett., 36(21), doi:10.1029/2009GL040083.
602
Uppala, S. M., et al. (2005), The ERA-40 re-analysis, Quarterly Journal of the Royal
603
Meteorological Society, 131(612), 2961–3012, doi:10.1256/qj.04.176.
604
Voldoire, A., and coauthors (2011), The CNRM-CM5.1 global climate model:
605
Description and basic evaluation, technical doc, CNRM-CM, available at
606
http://www.cnrm.meteo.fr/cmip5/.
607
Wu, T., R. Yu, and F. Zhang (2008), A modified dynamic framework for the atmospheric
608
spectral model and its application, Journal of the Atmospheric Sciences, 65(7), 2235–
609
2253, doi:10.1175/2007JAS2514.1.
610
Wu, T., et al. (2012), The 20th century global carbon cycle from the Beijing Climate
611
Center Climate System Model (BCC CSM), J. Clim.,submitted.
612
Yukimoto, S., and coauthors (2011), Meteorological Research Institute-Earth System
613
Model v1 (MRI-ESM1) – Model Description, Tech. Rep. Meteor. Res. Inst. 64, 88pp.,
614
The Meteorological Research Institute.
615
Figure Captions.
616
Figure 1.
617
3-months running average of global averaged and de-seasonalised TOA outgoing shortwave
618
radiation anomalies for 11 models in the CMIP5 database over the period of 1860-2000 (only
619
GISS-E2-R is shown being not different from GISS-E2-H). For each model the ensemble mean
620
is shown. The red line at the bottom shows the volcanic aerosol optical depth (AOD) at 550nm
621
fromSato et al. [1993] (and updates). The grey bars at the top of plots indicate the occurrence
622
of the 9 volcanic eruptions listed in Table 2.
623
Figure 2.
624
Composite near-surface temperature anomalies (K) for the two following winters of the past
625
nine most recent large tropical volcanic eruptions (Table 2) in all models and the 20th century
626
reanalysis (20CRv2). Hatching displays areas at or over 95% Significance using a two tailed
627
t-test.
628
Figure 3.
629
Composite MSLP anomalies (hPa) for the two following winters of the past nine most re-
630
cent large tropical volcanic eruptions in all models and the 20th century reanalysis (20CRv2).
631
Hatching displays areas at or over 95% Significance using a two tailed t-test.
632
Figure 4.
633
Composite 200 hPa Geopotential Height anomalies (m) for the two following winters of the
634
past nine most recent large tropical volcanic eruptions in all models and ERA40 reanalysis. The
635
anomalies in the reanalysis are computed for the 4 eruptions after 1960. Hatching displays areas
636
at or over 95% Significance using a two tailed t-test.
637
Figure 5.
638
Composite 50 hPa Geopotential Height anomalies (hPa) for the two following winters of the
639
past nine most recent large tropical volcanic eruptions in all models and ERA40 reanalysis. The
640
anomalies in the reanalysis are computed for the 4 eruptions after 1960. Hatching displays areas
641
at or over 95% Significance using a two tailed t-test.
642
Figure 6.
643
Superposed epoch analysis for the winter (DJF) NAO index for the 12 CMIP5 models and
644
the 20th century reanalysis (20CRv2) for the 9 eruptions listed in Table 2. The average over 9
645
volcanic eruptions is shown at different lag time. Lag 0 indicates the first winter after a volcanic
646
eruption. The horizontal lines show, from bottom to top, the 1st, 5th, 95th and 99th percentiles
647
of the Monte Carlo distribution, which correspond to a one-tailed confidence interval of 1%
648
(continuous lines) and 5% (dashed lines). In the top right corner of each plot is indicated the
649
number of winters at lag 0 with positive NAO and the relative p-value.
650
Figure 7.
651
As Fig 6 but using the convention adopted inChristiansen [2008] for the first winter after the
652
eruptions of Santa Mar´ıa (1902-1903) and Fuego (1974-1975).
653
