Q. J . R. Mereorol. SOC. (1994). 120, pp. 1683-1688 551.590.3:551.521.3
Solar heating rates after a volcanic eruption: The importance of SO2 absorption
By D. J. LARY*, M. BALLUCH and S. BEKKI
University of Cambridge, UK
(Received 1 February 1994; revised 20 April 1994)
SUMMARY
Studies of the impact of volcanic eruptions on the climate have mainly focused on the radiative effects of
enhanced aerosol loadings. However, initially, volcanoes release not aerosols, but large amounts of sulphur
dioxide (SO,), which subsequently leads to the production of sulphate aerosols. SO2 absorbs strongly in the
ultraviolet and the visible region of the spectrum. This paper shows that accounting for this absorption is
important if an accurate determination of change in heating rates is to be made in the period shortly after a
volcanic eruption.
1. INTRODUCTION
The volcanic eruption of Mount Pinatubo in June 1991 injected approximately 20 Mt of sulphur
dioxide (SOz) into the upper troposphere and lower stratosphere (Bluth et al. 1992; Read et af.
1993). SO2 concentrations exceeded one part per million by volume (p.p.m.v.) in the centre of the
volcanic SOz cloud shortly after the eruption. However, the SO2 levels decreased rapidly because
of dispersion and oxidization by O H (Bekki and Pyle 1994). The SO2 released from the volcano
is subsequently converted to sulphate aerosols. When radiative effects of volcanic eruptions are
considered, they are usually focused on the effects of sulphate aerosols (for example, Vogelmann
et al. 1992). Recently Bekki et al. (1993) have shown the importance of the absorption of ultraviolet
radiation by volcanic SOz for the photolysis rate of molecular oxygen after the volcanic eruption
of Mount Pinatubo. In like manner, this note uses a time series of SO2 profiles over the period
just after an eruption to quantify the effects which the absorption of solar radiation by SO2 alone
can have on atmospheric heating rates. The magnitude and duration of these effects are also
discussed.
2. CALCULATIONS
Instantaneous and diurnal average solar heating rates for clear-sky conditions were calculated
using a detailed spherical radiative-transfer model which contains a description of all orders of
multiple scattering described by Lary and Balluch (1993). For the calculations presented in this
note the scattering by air molecules was assumed to be isotropic and no infrared cooling was
included. A set of different profiles (described below) was used for the calculations.
The calculations presented here include the effect of the absorption of solar radiation by SO2
(Bekki and Pyle 1992). SOz absorbs strongly between 180 nm and 235 nm, weakly between 260 nm
and 340 nm and very weakly between 340 nm and 390 nm. The values of SO2 absorption cross-
sections adopted in this work are taken from Yung (1982) for the 180-235 nm region and from
Okabe (1978) for the 260-390 nm region.
By analogy with the volcanic eruption of Mount Pinatubo, the calculations are performed for
the summer solstice with the centre of the volcanic SO2 cloud at 25 km over the equator (Read et
al. 1993). A mid-latitude summer ozone profile is specified. The conclusions drawn are not
significantly affected by the ozone profile used. If a tropical ozone profile (which has a lower total
ozone column) is used it results in slightly more heating due to SO2 absorption within the SOz
cloud. Using Total Ozone Mapping Spectrometer (TOMS) data, Bluth et al. (1992) analysed the
decay rate of the volcanic SOz cloud produced by the Mount Pinatubo eruption. They inferred the
change in the global average SOz column as a function of time: 3.6 x 10l8 on 17 June (a day
after the eruption), 8.5 x lo” on 23 June and 2.1 x 10’’ cm-’ on 30 June. These values of SOz
column correspond to layers 5 km thick, centred at 25 km, containing 7.8,1.85, and 0.45 p.p.m.v. of
SOz respectively. Four different values for the SO2 column are considered in this study. They are
aimed at representing different periods of the eruption. The four profiles used are shown in Fig.
1. and the total columns listed in Table 1.
* Corresponding author: Cambridge Centre for Atmospheric Science, Dept. Chemistry, University of
Cambridge, Lensfield Road, Cambridge CB2 lEW, UK.
1683

1684 NOTES AND CORRESPONDENCE
SO2 (volume mixing ratio)
Figure 1. The four SOz profiles used in this study (corresponding to runs 1 to 4).
TABLE 1. THE TOTAL SO2 COLUMNS FOR THE FOUR SO2 PROFILES
CONSIDERED
Run SO2 column Approximate time after eruption
1 1.61 x 1019cm-2
2 6.44 x 10'" cm-' a few hours
3 1.85 x lo1* cm-' a few days
4 3.86 x 1017 cm-2 about 10 days
To highlight the importance of the absorption at wavelengths greater than 235 nm, two sets
of heating-rate calculations were performed. In the first set of calculations the full SOz cross-section
was used. In the second set the SOz absorption cross-section for wavelengths longer than 235 nm
were set to zero.
The results from the first set of calculations, which included the SOz absorption at all
wavelengths, are shown in Fig. 2. The heating rates are considerably enhanced by the SO2
absorption. This enhancement decreases with time as the SO2 is oxidized and dispersed.
To see more clearly the effect of including the SO2 absorption it is useful to look at absolute
and percentage-difference plots. The absolute and percentage differences in the diurnal average
heating rate between a control calculation (with no SO2 absorption) and four calculations using
the four different SO2 profiles shown in Fig. 1 are shown in Figs. 3 and 4 respectively. In the case
of run 1, corresponding to the highest SO2 concentration, there is an approximate doubling in the
diurnal average heating rate within the SO2 layer, i.e. where most of the solar energy is deposited
owing to the enhanced absorption. For the instantaneous heating rate the enhancement in the
heating rate reaches 135% for an overhead sun within the SO2 layer. Below the SO2 layer there is
a reduction in the diurnal average heating rate of up to 20% due to the reduced penetration of
solar radiation to these altitudes. There is also a reduction in the diurnal average heating rate
above the SO, layer, as less ground-reflected radiation is able to pass back up through the layer
and, of course, less radiation reaches the ground to be reflected in the first place.

NOTES AND CORRESPONDENCE 1685
0.03 0.06 0.1 0.3 0.6 1.0 3.0 0.03 0.m 0.1 0.3 0.6 1.0 3.0 6.0
I 4 I 1 1 1 1 1 I , 4 1 I I I I I I I I 1 I , 1 1 1 1 1 I , , I 40' 4a I I 1 ' 1 1 1
h
E s
Run3 -
10 1 I 1 , 1 , , I , I 8 I ' , ( I 1 " I
-35
-30
-25
-20
-15
I I 1 1 40
-35
-30
-25
-20
Run 4 -I5
1 1 1 3 1 I 8 ~ 1 ~ 1 1 1 I , ! 10
0.03 0.m 0.1 0.3 0.6 1.0 3.0 0.03 0.w 0.1 0.3 0.6 1.0 3.0 6.0
Diurnal Average Heating Rate at the Equator (WDay)
Figure 2. The calculated diurnal average heating rates for the summer solstice made using four SO2 profiles
(dashed lines) shown in Fig. 1 . In each case the solid line is the control run which does not include any heating
due to SO2 absorption.
A few days after the eruption (run 3) the enhancement in the diurnal average heating rate is
already much less, peaking at approximately 30%. The vertical extent of the enhancement is also
reduced, it is now confined to the SO2 layer itself. In like manner the reduction in the diurnal
average heating rate above and below the SO, layer is also smaller. Ten days after the eruption
(run 4) the peak enhancement in the diurnal average heating rate is just over 5% at 25 km, the
centre of the SO2 layer. This is much smaller than it was a few days earlier, but it is still a significant
enhancement of the heating rate.
The results from the second set of calculations, where only SO2 absorption at wavelengths
less than 235 nm were included, showed almost no change in the heating rate compared with the
control run which included no SO2 absorption. For the profile containing the most SO2 there is a
slight increase at the top of the SO2 layer of a fraction of a per cent and below this a slight reduction.
This is because little solar radiation of wavelengths less than 310nm gets through to the lower
stratosphere under normal conditions. So there is virtually no change in the heating rate, as even
though the strongest SO2 absorption occurs at wavelengths less than 235 nm there are virtually no
photons for the SO, to absorb. For this reason, if a volcano injected a large amount of SO2 into
an atmosphere with a very low ozone content in the lower stratosphere, the enhancement in the
heating rate due to SO2 absorption would be higher than in the calculations presented here. This
is because more light of wavelengths less than 310 nm wou!d reach the lower stratosphere where
it could be absorbed by SO2. One of the most obvious results of these calculations is that, as would
be expected, the region of SOz absorption which is most important for atmospheric heating rates
is the absorption at wavelengths greater than 235 nm.
It is important to note that the radiative effects due to SO2 absorption occur long before
significant amounts of sulphate aerosol have formed. The peak heating due to SO, absorption
occurs during the first few days after the volcanic eruption. During this period the increase in the

1686
4
40.
35
30
25
20
15
40
35
30
25
20
I5
10
r:ll
NOTES AND CORRESPONDENCE
-0.25 0.0 0.25 0.5 0.75
1.25 1 -
-------
/
Run 1
Run 3
-rrTrTm
1.25 1.5
i 0 0 025 0.5 075 1 0 125
I I I I I 1 , I I 1 I I I I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 L
Run 2
I I
1.75

5 0 0 0.25 0 5 075 1 0 125 15 1
40
35
30
25
20
15
40
35
30
25
20 (5
10
5
Absolute Difference in Diurnal Average Heating Rate (WDay)
Figure 3. The absolute difference in the calculated summer solstice diurnal average heating rates between a
control run (with no SO2 absorption) and calculations performed with SO2 absorption using the four SO2
profiles shown in Fig. 1.
heating rate can be up to 1 K day-', which completely outweighs any radiative effects due to
aerosols at this time (see, for example, Tie et al. (1994)). By the time significant amounts of
sulphate aerosol have formed the heating due to SO2 absorption has dropped to a negligible level.
One implication of this is that the large additional heating which occurs for a short time owing to
SO2 absorption immediately after the eruption will cause more upwelling, transporting the volcanic
plume to higher altitudes and so causing the aerosol to last longer in the stratosphere.
3. CONCLUSIONS
A large increase in atmospheric SO2 in the lower stratosphere, such as occurs immediately
after a volcanic eruption, will lead to a large, but short lived, enhancement of the diurnal average
heating rate which can locally exceed 1 K day-' (depending on the loading of SOz). Above and
below the layer of enhanced SO2 there is a reduction of the diurnal average heating rate of as
much as 30%. The enhancement in the heating rate is almost entirely due to the weak SO2
absorption bands at wavelengths greater than 235 nm. The enhancement in the heating rate will
be greater if the lower stratosphere has low levels of ozone. This study has not included the effect
of aerosols to highlight the importance of the SO2 absorption alone.
In conclusion, if an accurate assessment of the climatic effects of volcanoes is required then
the ultraviolet/visible heating of SOz should be included 0s well as the radiative effects of aerosols
for the short period just after the volcanic eruption. The additional heating due to SO2 absorption
immediately after the eruption may contribute to the upwelling of the volcanic plume to higher
altitudes, so causing the aerosol to spend longer in the atmosphere.

10
3s
30
2s
20
-
E I s
Y
a $ 4 0
9 3 5
30
25
20
15
10
Run 1
NOTES AND CORRESPONDENCE
-25 0 2s 50 7s 100 -25 0 2 5 5 0 7 5 1 0 0
I
Run 2 -
1687
40
35
30
25
20
15
40
35
30
25
20
15
10
-25 0 25 50 75 100 -25 0 2 5 5 0 7 5 1 0 0
Percentage Difference in the Diurnal Average Heating Rate
(Relative to the reference profile)
Figure 4. As Fig. 3, but for the percentage difference.
ACKNOWLEDGEMENTS
This work is part of the UK Universities Global Atmospheric Modelling Programme funded
by the Natural Environment Research Council. It was partly supported by the grant STEP0016
from DGXII and STEP-(391-0139 of the Commission of the European Communities. The authors
would like to thank J. A. Pyle and M. McIntyre for their support and useful conversations.
REFERENCES
Bekki, S. and Pyle, J. A.
Bekki, S., Toumi, R. and Pyle, J. A.
Bluth, G. J. S., Doiron. S. D.,
Schnetzler, C. C.,
Krueger, A. J. and
Walter, L. S.
Lary, D. J. and Balluch, M.
Okabe, H.
1992 2-D assessment of the impact of aircraft sulphur emissions on
the stratospheric sulphate aerosol layer. Geophys. Res.
Lett., 20,723-726
A two-dimensional modeling study of the volcanic eruption of
Mount Pinatubo. 1. Geophys. Res., in press
Tropical ozone production and destruction by gaseous sulphur
photochemistry following the eruption of Mount
Pinatubo. Nature, 362, 331-333
Global tracking of the SO2 clouds from the June 1991 Mount
Pinatubo eruptions. Geophys. Res. Len., 19, 151-154
1994
1993
1992
1993
1978
Solar heating rates: The importance of spherical geometry. 1.
Photochemishy of small molecules. Wdey-Interscience, New
Amos. Sci., 50, 3983-3993
York

1688 NOTES AND CORRESPONDENCE
Read, W. G., Froidevaux, L. and 1993 Microwave Limb Sounder measurements of stratospheric SO2
from the Mt. Pinatubo volcano. Geophys. Res. Leu., M,
1299-1302
Two-dimensional simulation of Pinatubo aerosol and its effect
on stratospheric ozone. J. Geophys. Res., in press
Enhancements in biologically effective ultraviolet radiation
following volcanic eruptions. Narure, 359, 47-49
Photochemistry of the stratosphere of Venus-Implications for
atmospheric evolution. Icarus, 51, 199-247
Waters, J. W.
Tie, X . er al. 1994
Vogelmann, A. M., Ackerman, T. P. 1992
Yung, Y. L. and DeMore, W. B. 1982
and Turco, R. P.