Journal of Environmental Protection, 2011, 2, 778-795
doi:10.4236/jep.2011.26090 Published Online August 2011 (http://www.SciRP.org/journal/jep)
Effect of Asian Dust Storms on the Ambient SO2
Concentration over North-East India: A Case
Study
Timmy Francis
Physical Research Laboratory, Ahmedabad, India.
Received May 3rd, 2011; revised June 12th, 2011; accepted July 21st, 2011.
ABSTRACT
Ambient SO2 concen tration at a high rain fall site, Shillong (25.67˚N, 91.91˚E, 1064 m ASL), located in North-East In-
dia, was measured during March 2009 and January 2010 with the aim to understand the effect of long range transport
of pollutants from North-East Asia on the ambient SO2 levels at this relatively clean site. The concentrations recorded
during the former sampling period were very high (Max: 262.3 ppb)—which decayed down gradually towards the end
the sampling period—whereas those during the latter sampling period were well within the acceptable limits (Max:
29.7 ppb). This elevated SO2 concentrations during March 2009 is p roposed to have association with a major cold air
outbreak and an associated cyclone preceding one of the dust storm events reported in China, and a resultant sudden
change in wind trajectory leading to the long range transport of pollutants to the sampling site. The argument is for-
mulated on the basis of the back trajectory analysis performed using HYSPLIT for the month of March 2009—the plots
clearly showed a drastic change in wind trajectories between 8th and 15th of March 2009 wherein the winds traveled
over some of the highly polluted regions such as the Perm region of Russia—and on the results from model runs per-
formed using the global 3-D model of tropospheric chemistry, GEOS-Chem (v8-03-01)—it clearly showed the tropo-
spheric SO2 over Perm region in Russia peaking during Nov, Dec, Jan, Feb and Mar every year, possibly due to central
heating. The observation of long range transport of SO2 from the highly industrialized areas of Perm in Russia to
North-East India during dust storm events has importan t implications to the present und erstanding on its relative con-
tribution to the Asian pollutant outflow to the Pacific during spring as the GEOS-Chem model run s also showed regions
in and around Russia with relatively high concentrations of atmospheric NOx, Peroxyacetyl Nitrate, Lumped Per-
oxypropionyl Nitrate, HNO3, HNO4, C3H8, C2H6, SO4, NH4, Inorganic Sulphur Nitrates and Lumped Alkyl Nitrate.
Keywords: SO2, GEOS-Chem, HYSPLIT, Cold Air Outbreaks, Asian Dust Storms, Asian Pollutant Outflow
1. Introduction
Asian region is fast evolving as one of the major energy
consuming regions in the world for a rapidly growing
population and emerging economy [1], projected mega
cities in Asia as one of the large emission sources of SO 2.
As per IPCC 2007, there has been a regional shift in the
emissions of SO2 from US and Europe to South-East
Asia, with implication to shift in atmospheric radiative
forcing patterns for these regions [2-4]. In this scenario,
monitoring ambient SO2 concentration coupled with
wind trajectory analysis in the North-East region of India
—characterized by high rain fall—has special significance
to assessing the role of long range transport from North-
East Asia in controlling the pollutant characteristics over
this region. With this objective ambient SO2 concentra-
tions were measured during March 2009 and January
2010 at a high rainfall site in North-East India, Shillong
(25.67˚N, 91.91˚E, 1064 m Above Sea Level (ASL)).
2. Materials and Methods
2.1. Site Description: Shillong (25.67˚N, 91.91˚E,
1064 m ASL)
The sampling site (25.67˚N, 91.91˚E, 1064 m ASL) is
located about 18 km away from the Shillong city, located
on the Shillong Plateau which is an outlier of the plateau
of peninsular India and is composed primarily of ancient
rocks. The high est point is Sh illo ng peak , at 19 61 m ASL
located 5 km south of the city of Shillong. It is on

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study779
2
the Shillong Plateau, the only major pop-up structure in
the northern Indian shield. Due to its latitude and high
elevation Shillong has a sub-tropical climate with mild
summers and chilly to cold winters. The region is sub ject
to vagaries of the monsoon. The average annual rain fall
is about 2200 mm. The monsoon arrives in June and it
rains almost until the end of August. Perched at an alti-
tude of 1521 m ASL, the Sh illong city stretch es for about
6 km on an elevated tract. The city is situated on a pla-
teau bound on the north by the Umiam gorge, on the
northwest by the great mass of the Diengiei Hills that
rise up to a height of 1852 m ASL, and on the northeast
by the hills of the Assam valley. The proximity of this
high rainfall site to some of the highly polluted regions
in North-East Asia and hence the possibility of long
range transport makes this study location strategically
important.
2.2. Experiment: SO2 Monitoring System
The experimental set up comprised of a primary UV
fluorescence SO2 Monitor (Thermo—43i Trace Level
Enhanced (TLE)), an air compressor for producing zero-
level (zero) gas and a dynamic gas calibrator (Thermo—
146i) for preparing diluted SO2 standard gas. Using
pulsed fluorescence technology, the Model 43i-TLE
measures the amount of sulfur dioxide in the air with an
instrumental detection limit of 50 pptv (300 second av-
eraging time). The UV fluorescence SO2 monitor was
calibrated onsite every 10 days with standard SO2 gas (2
ppmv with N2 balance gas, Spectra, USA) and zero air.
Five minute averaged SO2 raw data was accumulated in
the SO2 analyzer which in turn was periodically down-
loaded and fed to a computer through an ethernet port.
Discussions on the performances of similar systems can
be seen elsewhere [5-7].
2.3. The HYSPLIT Trajectory Model
The HYSPLIT (HYbrid Single-Particle Lagrangian Inte-
grated Trajectory) model, developed by NOAA ARL
[8,9], was used for the back trajectory analysis for this
work. HYSPLIT uses a modeled vertical velocity scheme;
it can depict the vertical motion of the relevant air parcel.
The model computes the advection of a single pollutant
particle, or simply its trajectory.
2.4. GEOS-Chem Model Description
2.4.1. Gener al Description
The GEOS-Chem global 3-D model of tropospheric che-
mistry (v8-03-01; h ttp://acmg.seas.harvard.ed u/geos/)
driven by GEOS-5 assimilated meteorological observa-
tions from the NASA Global Modeling and Assimilation
Office (GMAO) is employed for this work. Meteoro-
logical fields in the GEOS-5 have 6-h temporal resolu-
tion (3-hour resolution for surface fields and mixing
depths) and a horizontal resolution of 0.5˚ latitude ×
0.667˚ longitude, with 72 levels in the vertical extending
from the surface to approximately 0.01 hPa. The model
is applied to a global simulation of Ozone-NOx-VOC-
aerosol chemistry. General descriptions of GEOS-Chem
are given by [10] and [11]. The detailed descriptions of
the GEOS-Chem aerosol emission inventories and simu-
lation evaluations are provided in [11] and [12] for sulfur
and ammonia, in [13] for dust, and in [14] for recent up-
dates on emission inventories over the China region. The
simulations were conducted for January-December 2009
at 4˚ × 5˚ resolutions. Th ey were initialized on 1st Janu -
ary 2009 using GEOS-Chem fields generated by a one
year spin-up simulation with 4˚ × 5˚ resolution.
2.4.2. Emission Inventories Included in the Model
Runs
In the Model runs, the following emission inventories
were used:
1) The EMEP—the Co-operative Programme for
Monitoring and Evaluation of the Long-range Transmis-
sion of Air Pollutants in Europe—inventory for 2000
[15].
2) The Big Bend Regional Aerosol and Visibility Ob-
servational (BRAVO) study inventory for 1999 [16].
3) The global sulfur emission based on the Emissions
Database for Global Atmospheric Research (EDGAR)
inventory [17].
4) The Streets inventory [18].
5) The Canada Criteria Air Contaminants (CAC) in-
ventory for 2 0 02
(http://www.ec.gc.ca/pdb/cac/cac_home_e.cfm) and
6) The Environmental Protection Agency (EPA/NEI05)
inventory.
The bio-fuel emissions and biogenic emissions also
were included in the model runs with the Mod el of Emi-
ssions of Gases and Aerosols from Nature (MEGAN)
v2.1 [19] inventory used for the latter.
2.5. GAMAP Visualization Toolkit
The Global Atmospheric Model Analysis Package (GA-
MAP) (Version 2.15; http://acmg.seas.harvard.edu/ga-
map/) is used for plotting the output from the GEOS-
Chem runs. GAMAP is a self-contained, consistent, and
user-friendly software package for reading and visualiz-
ing output from Chemical Tracer Models (CTM’s) and
consists of a suite of routines written in IDL (Interactive
Data Language). It can produce line plots, 2D plots, 2D
animations, or 3D iso-contour surface plots and can read
2D, 3D or 4D dat a bl ocks.
3. Results and Discussions
Ambient SO2 concentrations were measured during

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study 780 2
March 2009 and January 2010 at the high rainfall site,
Shillong (25.67˚N, 91.91˚E, 1064 m ASL) in North-East
India, using the pulsed UV fluorescence analyzer. During
the measurements the inlet of the SO2 analyzer was
maintained at an altitude of about 15 m above the ground,
while the analyzer was housed in an air conditioned
room. The integration time for the analyzer was set at 5
minutes. An Automated Weather Station (AWS) located
within a distance of 1/2 km from the sampling site was
employed to monitor the various atmospheric parameters
such as th e wind speed , wind direction , relative hu midity,
atmospheric pressure etc, throughout the sampling pe-
riod.
3.1. Major Features during March 2009
Figure 1 reveals the temporal changes of the SO2 con-
centrations at Shillong, during March 2009 . The concen-
tration levels recorded during the initial days of the sam-
pling were very high (Max: 262.3 ppb) which decayed
down gradually towards the end the sampling period.
This observed high SO2 concentrations is explained as
due to a transient long range pollutant transport from
North-East Asia associated with the cold air outbreak
preceding the dust storm events reported in China [20]
on 14th March 200 9 and in Nor th- East India [21] on 17 th
March 2009. The former [20], using MODIS 1B data of
relevant regions, showed evidences for the dust storm in
western Gansu and Neimenggu in China while the latter
[21], based on Terra/Aqua MODIS data, reported the
dust storm over Guwahati, North-East India. Based on
the mesoscale model (MM5) derived wind speed direc-
tions at 850 hPa overlaid on sea level pressure on 17th
March 2009, [21] found a persistent Nor th-Easterly flow
with high wind speed (~6 m/s) over the region resulting
in mobilization and lifting of dust particles in to the at-
mosphere. They analyzed NCEP temperature/relative
humidity (RH) anomalies variations and found ~0.6˚C
increase in surface air temperature and ~–4% reduction
in RH during March 2009, which they said has resulted
in dry conditions over the region. They attributed the
very high value of Terra MODIS AOD550 (~1.3) along
with lower value of Alpha (~0.78) to coarse mode dust
aerosol particles over the region due to dust event oc-
curred on 17th March 2009.
The above argument of long range transport as the
mechanism responsible for the observed high SO2 values
at the North-East Indian sampling site is based on the
results provided collectively by the back trajectory
analysis performed using HYSPLIT for the sampling
period and the model runs performed using the Chemical
Transport Model GEOS-Chem. The arguments are fur-
ther supported with the local meteorological parameters
measured using nearby Automated Weather Station
Days (March 2009)
789 10111213141516171819202122232425262728293031
SO
2
Concentration (ppbv)
0
50
100
150
200
250
Figure 1. The measured SO2 time series profile during
March 2009 at the sampling site, Shillong.
(AWS). The time series Mineral Dust Optical Depth
(MOPD) profiles created using GEOS-Chem (discussed
in the following sections) for the month of March 2009
also gave evidence for the dust storm movement reported
in China on 14th March 2009.
3.1.1. Interpreting the High SO2 Values
3.1.1.1. Sou rce Identification B ased on HYS P LIT
Trajectories
The 7 days back trajectories beginning from the sam-
pling site were generated (Figure 2) using HYSPLIT for
whole the sampling month, with one trajectory plot made
each on 1st, 8th, 15th , 2 2nd and 29th of March 2009. The
trajectories computed to identify the source regions were
for altitudes 500 m, 1000 m and 4000 m Above Ground
Level (AGL). The plots showed clearly that between 8th
and 15th of March 2009 there was a drastic change in
wind trajectories. During this period the winds which
were traveling in the eastward direction suddenly got
bend and started traveling towards North-East till it reach
~60˚N and from there it again got bend and started trav-
eling towards the south to reach the sampling site. It was
in line with this observed sudden change in the wind
trajectory (between 8th and 15th of March 2009), the
high SO2 concentrations were recorded. The back trajec-
tories plotted for th e fo llo wing week s ag ain show ed wind
patterns similar to the ones persisted prior 8th March
2009. The observed decay in concentration levels during
the days succeeding the transient long range pollutant
transport were in line with this normal trajectory pattern.
3.1.1.2. S o urce Identification Using GEOS-Chem Model
The GEOS-Chem model (v8-03-01) was run from 1st
January 2009 to 31st December 2009, after spinning up
the model fields for one year, and the output averaged for
the first half of each month was generated. The results
were then plotted using GAMAP to find out the possible

Effect of Asian Dust Storms on the Ambient SO2 Concentration over North-East India: A Case Study 781

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study 782 2
Figure 2. The 7-day back trajectories originating at the sampling site plotted using HYSPLIT for March 2009.
SO2 source regions along the back trajectories obtained
between 8th and 15th of March 2009. The plots clearly
showed that many of the regions over which the winds
traveled during this period are characterized by high
concentrations of SO2. Out of these the Perm region in
Russia has the highest SO2 concentration in the GEOS-
Chem generated plots.
3.1.1.2. 1. The Perm City , Rus si a
Perm, projected in this paper as the source region for the
high SO2 containing air parcel detected at the sampling
site, is a city in the eastern part of Russia, near the Urals
Mountains. It is a major center of heavy industries, che-
mical industries (varnishes, paints, fertilizers, sulfuric
acid, etc.), petrochemical and oil refining industries, de-
fense production, timber and wood processing industries,
food industry and so on. This regional area has substan-
tial deposits of natural resources (oil, gas, potassium,
magnesium salt, precious and jobbing stones, gold and
chromium ores and so on.) which is the base for both the
extractive industries and related economic sectors of the
region. The region of Perm has one of the richest mineral
resource bases in Russia and has many mines.
The Perm region is known for its long frosty, snowy
Ural winters. However, the region’s severe continental
climate causes abrupt seasonal changes, from a cold
winter followed by a mild, usually warm spring and then
by a hot summer. The cold winter necessitates central
heating especially during the months November, De-
cember, January, February, and March. The general pre-
cipitation trend over Perm region shows lowest precipi-
tation events during the months February and March
followed by slightly high er precipitation events in April.
3.1.1.3. Anomalies Detected in Local Meteorology
The Automated Weather Station (AWS)—located within
500m from the sampling location—provided meteoro-
logical parameters such as wind speed & wind direction
(Figure 3), relative humidity (Figure 4) and atmospheric
pressure (Figure 5) throughout the month of March 2009
except for a small non-operational period from 20 - 23rd
March 2009. Statistically the wind direction mostly re-
mained South-East with second directional preference
for South-West. The wind speed showed a transient in-
crease between 9th and 13th of March 2009 with the
maximum speed recorded of 3.2 m/s. The relative hu-
midity (RH) remained more or less on the higher side
throughout th e sampling period with a small dip reco rd ed
between 10th and 12th followed by two spikes each on
13th and 14th. The highest RH value observed for the
month was on 17th with the peak relative humidity value
recorded of 99%. The Atmospheric pressure also kept

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study783
2
Days (March 2009)
12345678910111213141516171819202122232425262728293031
Wind speed (m/s)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Wind direction (Deg)
0
100
200
300
400
Wind s peed
Wind direction
Figure 3. Wind speed and wind direction during the sam-
pling period, obtained from nearby AWS.
Days (March 2009)
1 23 45 67 8910111213141516171819202122232425262728293031
Relative Humidity (%)
0
20
40
60
80
100
120
Figure 4. Relative humidity during the sampling period,
obtained from nearby AWS.
Days (March 2009)
1 23 45 67 8910111213141516171819202122232425262728293031
Atmospheric Pressure (hPa)
890
892
894
896
898
900
902
904
Figure 5. Atmospheric pressure during the sampling period,
obtained from nearby AWS.
unusually fluctuating with a dip observed between 3rd
and 7th (lowest value 891 hPa) and a hump observed
between 12 th and 15th of Marc h 2009 (Max value 903.3
hPa).
3.1.1.4. Dust Storm Detection Using GEOS-Chem
To visualize and understand the source and the trajectory
of the dust storms reported [20] in China on 14th March
2009 and over North-East India [ 21] on 17 th March 2009,
the GEOS-Chem model was run with the inventories
discussed previously, and generated the time series Min-
eral Dust Optical Depth (MOPD) data for the month of
March 2009. The plots generated from these data for the
days 10th to 20th of March 2009 (Figure 6) clearly
showed this formation and the movement of the dust
storm. The plots also showed high MOPD values for the
Indo-Gangetic Plane also during 13th and 14th March
2009. A second phase of high MOPD values for North-
East India started on 17th March 2009 which remained
high in the following few days (18th, 19th and 20th
March 2009).
3.1.1.4.1. Cold Air Outbreaks and Dust Storms in Asia
It is known that Asian dust storms, especially those of
Chinese origin occur during spring [22], when the cli-
mate over East Asia reaches its maximum dryness of the
year, the surface start warming, and strong winds sweep
over the desert areas leading to highly unstable synoptic
systems. The percentages of dust storms generally occur-
ring in China during March, April, and May are 20%,
58%, and 22%, respectively [23].
Asian dust storms are always associated with cold air
outbreaks, resulting in the Mongolian cyclonic depres-
sion and frontal system. It is known that polar front
causes mid-latitude jet stream (by the thermal wind rela-
tionship) with troughs and ridges to lead to patterns of
divergence aloft. This upper air disturbance causes a sig-
nificant pressure decrease in the lower atmosphere as a
result of cyclogenesis, leading to the development of a
low pressure system with low central sea level pressures
which continue to deepen as time progress to cause cy-
clone and a resulting dust mobilization.
The reason the sampling was planned during this sea-
son at this North-Eastern Indian site was to understand
the possible effects of such cold air outbreaks and the
associated atmospheric disturbance preceding the Chi-
nese dust storms in altering the atmospheric pollutant
transport pattern to North-East India.
To visualize this air disturbance associated with the
cold air outbreak occurred during the above sampling
period, trajectory matrices were plotted on 8th, 11th,
15th, 22nd and 29th of March 2009 with so urce at multi-
ple locations (Figure 7). The plots clearly indicated a
major atmospheric disturbance between 8th and 15th of
March 2009.

Effect of Asian Dust Storms on the Ambient SO2 Concentration over North-East India: A Case Study 784

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study785
2
Figure 6. The Mineral Dust Optical Depth (MOPD) profile obtained during 10th to 20th of March 2009 from GEOS-Chem.

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study 786 2

Effect of Asian Dust Storms on the Ambient SO2 Concentration over North-East India: A Case Study 787
Figure 7. The Trajectory Matrix plotted using HYSPLIT for the cold air out break period.
3.2. Transport from Russian Region: Implication
to the Asian Pollutant Outflow
The observation that SO2 got transported from the highly
industrialized areas of Perm in Russia to North-East In-
dia during the dust storm event in March 2009 has im-
portant implications to its share in the Asian pollutant
outflow to the Pacific during spring, every year. To make
this picture clear, 7 day back trajectories starting at co-
ordinates (18˚N, 130˚E) and (22˚N, 110˚E), (Figure 8)
were plotted which showed the air parcel from Russian
region reaching the Pacific during the dust storm days.
Based on these observations, it is projected here that the
contribution from this region is non negligible while as-
sessing the impact of Asian pollutant outflow, as the
Perm region is characterized by a wide range of indus-
tries.
The GEOS-Chem generated SO2 plots for every month
(averaged over the first 15 days) plotted for the whole
year 2009 (Figure 9) clearly showed the tropospheric
SO2 over Perm region peaking during November, De-
cember, January, February and March every year, possi-
bly because of central heating. The lesser amount of pre-
cipitation in the Perm region during the months February
and March result in longer atmospheric retention time for
these pollutants during these months. The concentration
levels go decreasing when there is increase in precipita-
tion events starting from April along with the receding
winter to reduce the central heating requirements.
The model runs were also performed to find out the
concentration levels of other major pollutants over this
region during March 2009. It was found that regions in
and around Russia has relatively high concentrations of
atmospheric NOx, Peroxyacetyl Nitrate (PAN), Lumped
Peroxypropionyl Nitrate (PPN), HNO3, HNO4, C3H8,
C2H6, SO4, NH4, Inorganic Sulphur Nitrates (NIT), and
Lumped Alkyl Nitrate (R4N2) (Figure 10).
The model runs also showed (figures not included here)
that similar to SO 2, these pollutants also peak during No-
vember, December, January, February, and March every
year, again possibly has association with central heating
and climatology discussed above. But the fact that cold
air outbreaks preceding the dust storm events in Asia oc-
cur only during March, April and May makes th e relative
contribution from this region significant only during
March, when the atmospheric pollutant peak for the region
overlaps with the dust storm months (March, April, May).
So it suggest that during the dust storm events in March,
the contribution from regions in and around Russia may
not be neglected while considering the impact of Asian
pollutant outflow to the Pacific, which becomes less sig-
nificant during cold air outbreaks and the associated dust

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study 788 2
Figure 8. The 7-day back trajectories showing the air parcel from Russian region reaching the Pacific during the dust storm
events.

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study789
2
Figure 9. SO2 profile averaged for the first half of every month for 2009 from GEOS-Chem model runs showing the SO2
peaking over the Perm region in Russia, during the months November, December, January, February and March.

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study 790 2
Figure 10. NOx, Peroxyacetyl Nitrate (PAN), Lumped Peroxypropionyl Nitrate (PPN), HNO3, HNO4, C3H8, C2H6, SO4, NH4,
Inorganic Sulphur Nitrates (NIT) and Lumped Alkyl Nitrate (R4N2) profiles averaged for the first half of March 2009 from
GEOS-Chem model runs.

Effect of Asian Dust Storms on the Ambient SO2 Concentration over North-East India: A Case Study 791
SO2 levels were observed between 8th and 16th of March
2009 and in general the concentrations resembled more
or less to the background levels expected for this high
rainfall site in the absence of long range transports.
storms happenin g in April, and May.
The model runs also showed (Figure 11) that the pol-
lutants N2O5, NH3, hydrophilic black carbon, hydrophilic
organic carbon, hydrophobic black carbon and hydro-
phobic organic carbon are present only in smaller levels
in the Perm region compared to the other known polluted
regions in North East Asia, and so are not expected to
contribute significantly to the Asian pollutant outflow.
3.4. Major Features during January 2010
Unlike March 2009 the SO2 concentrations recorded dur-
ing January 2010 (Figure 13) were well within the ac-
ceptable limits, with maximum value going up to 29.7 ppb
only and a minimum value recorded of 1.94 ppb. The
wind back trajectory analysis ( Figure 14) showed that the
winds were coming from West Asian region and had no
component coming from North-East Asia ruling out the
possibility of any long range transport from Russia. So the
concentration levels recorded may be taken as the back
ground SO2 values for thi s North-Eastern Indi an site.
3.3. SO2 Time-Series Data (March 2009): Model
Versus Observations
Figure 12 shows the time series SO2 profile obtained for
the nearest co-ordinates (26˚N, 90˚E) from the GEOS-
Chem model runs for the month of March 2009. The
model was not able to pick up the magnitude of this ma-
jor transport event even though a slight increase in the
Figure 11. N2O5, NH3, hydrophilic black carbon (BCPI), hydrophilic organic carbon (OCPI), hydrophobic black carbon
(BCPO), hydrophobic organic carbon (OCPO) profiles averaged for the first half of March 2009 from GEOS-Chem model
runs.

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study 792 2
Days (March 2009)
78910 11 12 13 1415 16 17 1819 20 2122 23 24 2526 27 2829 30
SO
2
Concentration (ppbv)
0
1
2
3
4
GEOS-Chem generated SO
2
time series
Figure 12. SO2 time series profile generated using GEOS-Chem for the sampling period in March 2009 for the nearest
co-ordinates (26˚N, 90˚E).
Days (January 2010 )
891011 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
SO
2
Concentration (ppbv)
0
5
10
15
20
25
30
Figure 13. The measure d SO2 time series profile during January 2010 at the sampling site, Shillong.
3.5. SO2 Time-Series Data (January 2010):
Model Versus Observations
In the case of January 2010, the time series output from
the model (Figure 15) matched more or less with the
observations both magnitude wise and to certain extend
shape wise. The diurnal variation observed in the expe-
rimental data is reproduced in the model generated time
series output .
4. Conclusions
Ambient SO2 concentrations at Shillong (25.67˚N,
91.91˚E, 1064 m ASL), a high rain fall site located in
North East India, measured during March 2009 and
January 2010 showed significant differences in magni-
tude, with concentrations going to a maximum of 262.3
ppb during the former sampling period—which decayed
down gradually towards the end the sampling period -
against the maximum value recorded during the latter
period of 29.7 ppb. This elevated SO2 concentrations
during March 2009 is explained as due to a long range
transport associated with the cold air outbreak preceding
one of the dust storm events reported in China. This ar-
ument of long range transport for the observed high SO2 g

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study793
2
Figure 14. The 7-day back trajectories originating at the sampling site plotted using HYSPLIT for January 2010.

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study 794 2
Days (January 2010)
8910 11 12 13 14 1516 17 18 19 20 21 22 2324 25 26 27 28 2930
SO
2
Concentration (ppbv)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5 GEOS-Chem generated SO
2
time series
Figure 15. SO2 time series profile generated using GEOS- Chem for the sampling period in January 2010 for the nearest
co-ordinates (26˚N, 90˚E).
values during March 2009 is formulated on the basis of
the back trajectory analysis performed using HYSPLIT
for the sampling period—the plots clearly showed a dras-
tic change in wind trajectories between 8th and 15th of
March 2009 wherein the winds travelled over some of
the highly polluted regions such as the Perm region of
Russia—and on the basis of the results from simulations
performed using the global 3-D model of tropospheric
chemistry, GEOS-Chem (v8-03-01)—it clearly showed
the tropospheric SO2 over Perm region peaking during
November, December, Janu ary, February and March ev er y
year, possi bl y due to central heating.
The observation of long range transport of SO2 from
the highly industrialized areas of Perm in Russia to
North-East India during the dust storm event in March
2009 is projected to have important implications to its
share in the Asian pollutant outflow to the Pacific during
spring, every year. Separate runs of the model revealed
that the regions in and around Russia have high concen-
trations of other atmospheric pollutants such as NOx,
Peroxyacetyl Nitrate, HNO3, Lumped Peroxypropionyl
Nitrate, HNO3, HNO4, C3H8, C2H6, SO4, NH4, Inorganic
Sulphur Nitrates and Lumped Alkyl Nitrate. The model
runs also showed that some of the other atmospheric
pollutants such as N2O5, NH3, hydrophilic black carbon,
hydrophilic organic carbon, hydrophobic black carbon
and hydrophobic org anic carbon ar e present on ly in neg-
ligible levels in the Perm region compared to the other
polluted regions in North-East Asia and so will not con-
tribute significantly to the Asian pollutant outflow.
The comparison of the time series SO2 profile ob-
tained for the nearest co-ordinates (26˚N, 90˚E) from the
GEOS-Chem model with the experimental data, for the
month of March 2009, revealed that the model was not
able to pick up the magnitude of this transport whereas
there is a good correlation between the two for January
2010.
5. Acknowledgements
The ISRO-Geosphere Biosphere Program (Department
of Space, Government of India) is gratefully acknowl-
edged for the partial financial support for this study. The
author thanks Prof. M. M. Sarin, Physical Research
Laboratory, India, for valuable scien tific discussions and
support during the course of this study. The author also
like to thank Dr. P. P. Nageshwara Rao, Mr. S. S. Kundu,
and Mr. Arup Borgohain (North Eastern Space Applica-
tions centre, ISRO), for the logistics provided, during the
sampling at NESAC, Shillong.
REFERENCES
[1] S. K. Guttikunda, G. R. Carmichael, G. Calori, et al.,
“The Contribution of Megacities to Regional Sulfur Pol-
lution in Asia,” Atmospheric Environment, Vol. 37, No. 1,
2003, pp. 11-22. doi:10.1016/S1352-2310(02)00821-X
[2] O. Boucher and M. Pham, “History of Sulfate Aerosol
Radiative Forcings,” Geophysical Research Letters, Vol.
29, No. 9, 2002, p. 1308. doi:10.1029/2001GL014048
[3] M. Pham, O. Boucher and D. Hauglustaine, “Changes in
Atmospheric Sulfur Burdens and Concentrations and Re-
sulting Radiative Forcings under IPCC SRES Emission
Scenarios for 1990–2100,” Journal of Geophysi-
cal Research, Vol. 110, No. D6, 2005, p. D06112.
doi:10.1029/2004JD005125
[4] S. J. Smith, E. Conception, R. Andres, et al., “Historical

Effect of Asian Dust Storms on the Ambient SO Concentration over North-East India: A Case Study795
2
Sulphur Dioxide Emissions 1850-2000: Methods and
Results,” Research Report, No. PNNL-14537, Pacific
Northwest National Laborat ory, Richland, 2004.
[5] Y. Igarashi, Y. Sawa, K. Yoshioka, et al., “Monitoring
the SO2 Concentration at the Summit of Mt. Fuji and a
Comparison with Other Trace Gases during Winter,”
Journal of Geophysical Research, Vol. 109, No. D17,
2004, pp. 1-21. doi:10.1029/2003JD004428
[6] W. T. Luke, “Evaluation of a Commercial Pulsed Fluo-
rescence Detector for the Measurement of Low-Level
SO2 Concentrations during the Gas-Phase Sulfur Inter-
comparison Experiment,” Journal of Geophysical Re-
search, Vol. 102, No. D13, 1997, pp. 16255-16265.
doi:10.1029/96JD03347
[7] M. Luria, J. F. Boatman, J. Harris, et al., “Atmospheric
Sulfur Dioxide at Mauna Loa, Hawaii,” Journal of Geo-
physical Research, Vol. 97, No. D5, 1992, pp. 6011-
6022.
[8] R. R. Draxler and G. D. Rolph, “HYSPLIT (Hybrid Sin-
gle-Particle Lagrangian Integrated Trajectory) Model,”
NOAA, Air Resources Laboratory, Silver Spring, 2011.
http://www.arl.noaa.gov/ready/hysplit4.html
[9] G. D. Rolph, “Real-Time Environmental Applications
and Display System (READY),” NOAA Air Resource
Lab, Silver Spring, Md, 2003.
http://www.arl.noaa.gov/ready/hysplit4.html
[10] I. Bey, D. J. Jacob, R. M. Yantosca, et al., “Global Mod-
eling of Tropospheric Chemistry with Assimilated Mete-
orology: Model Description and Evaluation,” Journal of
Geophysical Research, Vol. 106, No. D19, 2001, pp.
23073-23095. doi:10.1029/2001JD000807
[11] R. J. Park, D. J. Jacob, B. D. Field, et al., “Natural and
Transboundary Pollution Influences on Sulfate-Nitrate-
Ammonium Aerosols in the United States: Implications
for Policy,” Journal of Geophysical Research, Vol. 109,
No. D15, 2004, CiteID: D15204.
[12] J. Wang, D. J. Jacob and S. T. Martin, “Sensitivity of
Sulfate Direct Climate Forcing to the Hysteresis of Parti-
cle Phase Transitions,” Journal of Geophysical Research,
Vol. 113, No. D11, 2008, pp. D11207.1-D11207.15.
[13] T. D. Fairlie, D. J. Jacob and R. J. Park, “The Impact of
Transpacific Transport of Mineral Dust in the United
States,” Atmospheric Environment, Vol. 41, No. 6, 2007,
pp. 1251-1266. doi:10.1016/j.atmosenv.2006.09.048
[14] D. Chen, Y. Wang, M. B. McElroy, et al., “Regional CO
Pollution and Export in China Simulated by the
High-Resolution Nested-Grid GEOS-Chem Model,” At-
mospheric Chemistry and Physics, Vol. 9, No. 11, 2009,
pp. 3825-3839.
[15] V. Vestreng and H. Klein, “Emission Data Reported to
UNECE/EMEP. Quality Assurance and Trend Analysis
and Presentation of WebDab,” MSC-W Status Report
2002, Norwegian Meteorological Institute, Oslo, 2002.
[16] H. Kuhns, M. Green and V. Etyemezian, “Big Bend Re-
gional Aerosol and Visibility Observational (BRAVO)
study emissions inventory,” Desert Research Institute,
Las Vegas, 2003.
[17] J. G. J. Olivier and J. J. M. Berdowski, “Global Emis-
sions Sources and Sinks, in the Climate Systemp,”
Balkema Publishers/Swets & Zeitlinger Publishers, Lisse,
2001, pp. 33-78.
[18] D. G. Streets, Q. Zhang, L. Wang, et al., “Revisiting
China’s CO Emissions after the Transport and Chemical
Evolution over the Pacific (TRACE-P) Mission: Synthe-
sis of Inventories, Atmospheric Modeling, and Observa-
tions,” Journal of Geophysical Research, Vol. 111, No.
D14, 2006, Cite ID: D14306.
[19] A. Guenther, T. Karl, P. Harley, et al., “Estimates of
Global Terrestrial Isoprene Emissions Using MEGAN
(Model of Emissions of Gases and Aerosols from Na-
ture),” Atmospheric Chemistry and Physics, Vol. 6, No.
11, 2006, pp. 3181-3210. doi:10.5194/acp-6-3181-2006
[20] X. Li, and W. Song, “Dust Storm Detection Based on
Modis Data,” International Archives of Photogrammetry,
Remote Sensing and Spatial Information Sciences,
Liaoning Technology University, Shengyang, pp. 169-
172.
[21] A. R. Sharma, S. K. Kharol and K. V. S. Badarinath, “An
Unusual Dust Event over North-Eastern India and Its
Association with Extreme Climatic Conditions—A Study
Using Satellite Data,” Aerosols & Clouds: Climate
Change Perspectives, ISATA Bulletin, 2000, pp. 214-
217.
[22] T. S. Liu, “Loess and the Environment,” China Ocean,
Beijing, 1985, pp. 251.
[23] J. Sun, M. Zhang and T. Liu, “Spatial and Temporal
Characteristics of Dust Storms in China and its Sur-
rounding Regions, 1960-1999: Relations to Source Area
and Climate,” Journal of Geophysical Research, Vol. 106,
No. D10, 2001, pp. 10325-10333.
doi:10.1029/2000JD900665