SIMULATION OF SULFATE AEROSOL IN EAST ASIA
USING MODELS-3/CMAQ WITH RAMS METEOROLOGICAL DATA
             Seiji Sugata1, Daewon W. Byun2, and ItsusM Uno3

             'National Institute for Environmental Studies
              Tsukuba, Ibaraki, Japan 305-0053
             National Exposure Research Lab, EPA
              Research Triangle Park, NC 27711
             3Kyushu University
              Kasuga, Fukuoka, Japan 816-8580
INTRODUCTION

    Interest has grown to understand the transport and chemistry of pollutants originating
from eastern Asia,  which is  undergoing rapid industrialization.   Acid deposition has
become one of key environmental issues due to the increased use of high-sulfur fossil fuels
in the area.  During the last decade, comprehensive air quality models have been used to
assess the severity of acid deposition problems and to develop effective emissions control
strategies for North America and Europe. Recently, U.S. EPA has developed and publicly
released the Models-3 Community Multiscale Air Quality (CMAQ)  modeling system
(Byun  and Ching, 1999).   It is a comprehensive modeling system that consists of a
meteorological model, an emissions processing and projection system, and several interface
processors such as Meteorology-Chemistry Interface  Processor (MCIP), as well as  the
CMAQ Chemical Transport Model (CCTM).  One of key design objectives of Models-3
CMAQ was to achieve flexibility that enables linkage of different science processors and
modules  to build appropriate air quality models to meet user's needs.   It  adapts a
generalized coordinate system  with governing  equations for the fully  compressible
atmosphere to allow linkage of different description of atmospheric dynamics for multi-
scale applications. It uses efficient modular structure with minimal data dependency and a
set of a generalized chemistry solver module and chemical mechanism reader to handle
multi-pollutant problems. The Models-3 CMAQ modeling system utilizes the Mesoscale
Model Generation 5  (MM5) as the default meteorological driver. Until now,  it  has been
tested only with a few applications in USA.
    The present study attempts to address a few challenges in utilizing the flexibility of the
system.   We apply  the CMAQ system with the meteorological data provided  by  the
Regional  Atmospheric Modeling System (RAMS) and to a different geographical area -
East Asia covering the eastern half of China, Korean peninsula, and the Islands  of Japan.

-------
To demonstrate the model performance, we compare the results with observed time series
of non-sea salt sulfate that are available at several sites in the southern part of Japan during
January 1997,

ADAPTATION OF MODELS-3/CMAQ TO RAMS

    There are some significant differences between RAMS and MM5, such as the vertical
coordinate, horizontal grid system, and available meteorological parameters in the output.
Although the CMAQ CTM uses a generalized coordinate system, the interface processors
released were  designed to work with MM5  output exclusively.  Some of the interface
programs b*oth in the RAMS and CMAQ systems were modified to allow this linkage.
They  are a  postprocessor routine of RAMS to output  data in I/O API  (input/output
application programming interface) format as required by the Models-3/CMAQ system and
a reader subroutine in the MCIP of CMAQ.
    Except for a few minor changes in the CTM code to output wet-deposition data at the
interval defined by the user and to incorporate daily  photolytic rates  for long-term
simulations,  we used the default CMAQ CTM version as released in 1999.  The default
options are:
• Advection with piece-wise parabolic method (PPM)
• Vertical diffusion with K-theory parameterization
• Deposition flux as bottom boundary condition for the vertical diffusion
• Mass conservation adjustment scheme
• Horizontal diffusion with scale dependent diffusivity
• Carbon Bond 4 (CB-4) chemistry mechanism with isoprene chemistry
« QSSA gas-phase reaction solver
• Emissions injected in the vertical diffusion module
• Aqueous-phase reactions and convective cloud mixing
* Modal approach aerosol size distribution and dynamics
For the detailed description of the science algorithms, refer to Byun and Ching (1999).
APPLICATION OF RAMS/CMAQ FOR EAST ASIA

     To provide meteorological data for CMAQ simulations, RAMS was ran from 27
December 1996 to 31 January 1997. The modeling domain is 4,000 x 4,000 km2 on the
rotated polar-streographic map projection centered at the (35 N, 130 E) with 80 km mesh.
The model extends vertically up to  18 km, which is represented with the 23 layers in the
sigma-z coordinates.  For example, lowest four vertical layers are placed at elevations 47.7,
157.3, 288.7,  and 446.5 m.   RAMS  options used for  the run  were; non-hydrostatic
dynamics, simplified  Kuo for cloud parameterization, Mellor-Yamada 2.5 for vertical
diffusion, and Louis (1979) surface flux parameterizations.  The RAMS runs used large-
scale meteorological data provided by the ECMWF analysis. Then, the RAMS output is
fed into MCIP to generate all the necessary meteorological parameters for the CMAQ CTM
simulations.
     There is  no detailed emissions inventory yet for the East Asia region.   For the
simulation, we modified the emissions database from the Global Emissions Inventory
Activity 
-------
observations in the region. Ammonia emissions are obtained from Murano et al. (1995) and
Zhao and Wang (1994). Descriptions of emissions processing methods can be found in
Carmichael et al. (1998) and Uno et al. (1997).

        1997/01/11.04:00:00     1997/01/12.04:00:00
        1997/01/13.04:00
1997/01/14.04:00:00
      i  i i.. i  . i  i ,  , . i
Figure 1. Sulfate aerosol (ASO4, Aitken mode and accumulation mode combined) concentration in
(fig I m3) and horizontal wind at the surface from 11-14 January 1997. Time is in GMT,
     Figure 1 illustrates one of the typical evolution patterns of sulfate aerosol (ASO4)
distributions during winter in the East Asia. The center of high concentration area was near
Qingdao on 11 January. Then the area started to deform in an elongated shape with major
axis from southwest to northeast direction on 13th. Although the peaks became smaller on
14th, the edge of high concentration area reached southern Japan.  The eastward movement
of a high-pressure system, with strong southwesterly winds behind it, caused the elongated
pattern.  After the system passed through on  13th,  wind became northerly around the
Korean peninsula and  southern Japan.  The  northerly  wind brought high  sulfate
concentrations to the southern Japan. This sulfate aerosol distribution is associated with a
typical winter pressure pattern often referred to as, "high in the west and low in the east"
illustrated in Figure 2.

-------
                     r/1:Qingdao,  2:Cheju,   3:Fukue;
                     Sty „.,!, K .,. nr.^m r y ^ ,.,* p. y m K,y u, r ^ i,,,, |.'1
Figure 2. Typical weather pattern for January. Monthly averaged sea level pressure and horizontal wind at
the surface calculated by RAMS in January 1997. Three observation sites (Qingdao, Cheju, and Fukue) are
identified.
    Figure 3 (a) presents observed monthly precipitation and Figure 3 (b) shows the same
simulated by RAMS.  RAMS successfully reproduced three large precipitation areas: area
near the center of Japan facing the Japan Sea; the sea east of Japan; and the area between
Taiwan and Kyushu island.  Figures 3 (c) and (d) show simulated monthly mean SO2 and
sulfate aerosol (ASO4) concentrations.  Regions of high sulfur concentrations are directly
related with the high anthropogenic emissions sources while sulfate aerosol concentrations
are distributed with much smoother gradient fanning out  from source regions toward
downwind areas.  Figures 3 (e) and (f) present dry and wet monthly deposition amounts of
ASO4, respectively.   The total amount of wet deposition is much larger than the dry
deposition, with the ratio of 20:1.
    Observed time series of sulfate aerosol concentrations are available at several sites in
China, Korea,  and Japan. These observations were obtained from the ion-chromatography
chemical analysis of Teflon filter measurements with 6- to 8-hour sampling periods. Figure
4 compares the model results with observations at Qingdao, Cheju, and Fukue. Model data
show  delay of peaks in the time series from upwind to downwind sites,  i.e., from Qingdao
to Cheju to Fukue.
COMPARISON OF SULFUR BUDGET WITH OTHER STUDIES

     Because there have been only a few comparable regional modeling studies on  the
sulfur budget in this region, we compare the present modeling results with previous studies
on the global sulfur budget (Chin et al. 1996, Langner and Rodhe 1991, Pham et al. 1995,
and Takemura 2000).  Using the process  analysis feature available with CMAQ, we have
estimated sulfur cycle budget for the eastern Asia simulation.  Figure 5 summarizes  the
monthly sulfur budget obtained from the CMAQ simulation for January 1997. The CMAQ
budget analysis indicates that one-third of the total sulfate aerosol in the model domain  has
resulted from gas-phase oxidation of SO2.1 note that this is approximately what we found
with RADM over  the eastern U.S. (McHenry and Dennis,  1994). Therefore,  while  our
regional analysis differs from global analyses, that is, the fraction is much larger than that

-------
in the global ones, it is in agreement with an earlier regional analysis using RADM. One of
possible reasons for the large fraction is that the regional model has much smaller area over
the oceans with little or no fresh sulfur emissions, which gives clouds less time to find SO2
and do the aqueous-phase conversion.  The other possible reasons  is that winter is a dry
season in East Asia, which gives winter sulfur more opportunity for gas-phase conversion
relative to the annual average. Therefore, we expect to predict a higher fraction  of sulfate
from gas-phase conversion than that for an annual global analyses for both reasons.
            (a) Observed Precip. (mm)
                          i.. fti....i
(b) RAMS Precip. (mm)
             (e)ASO4 Dry Dep. (kg/hectare)    (i) ASO4 Wet Dep.(kg/hectare)
             r ii'iiiii'iirr'ulii' il i nilti'nii ttlii'n I ii'nh	111 I'M I H'I i n i'i' I n'l 111 n 111 i'i i i i A >l i I'l 1111. i »\—
 Figure 3. (a) Observed monthly total precipitation, (b) RAMS simulated monthly precipitation; and CMAQ
 simulated (c) monthly mean SO2 concentration, (d) monthly mean sulfate aerosol (ASO4), (e) monthly total
 sulfate dry deposition amount, and (d) monthly total sulfate wet deposition amount for January, 1997.

-------
            90 -i
            60 -
            30 -
                                (a)  Qingddo
                 2468 10121416182022242628  31
                                 (b)   Cheju
            20 -
            10 -
                 2468 10121416182022242628  31
                                 (c )   Fukue
                 2468 10121416182022242628  31
                                  JANUARY

Figure 4. Observed and simulated time series of non-seasalt sulfate aerosol concentrations in(flg/ m ) in
January 1997. Dot (•) points represent measurements.
    We have found that 40% of the total sulfur is removed from the simulation domain by
horizontal advection and diffusion. This loss or gain (if there are more fluxes coming in
then going out) due to the horizontal transport is a natural feature of the regional models
(budget for this component is zero for the global models). From the wind field analysis, the
loss due to the transport is mostly at the western boundary of the domain. Vertical transport
and in-cloud mixing result in zero net budget as expected. Simulated  lifetimes (turnover
times)  of SO2 and  sulfate were 1.4 day and 3.1 day, respectively.  These numbers were
close to those with the previous studies.  Table 1 summarizes the ratio of gas- and aqueous-
phase conversion of sulfur species and time scales of SO2 and sulfate aerosol of the present
study and the previous global studies in the literature.

-------
         Diffusion
         (7)
Advection
(13)
                                 Diffusion
                                 (0)
Advection
(20)
              \       /
                  [1.4 day]
Emission
(100)
         gas
         (22)
                                 aqueous
                                 (40)
                      [3.1 day]
                            Dry
                            Deposition
                            (18)
                    Wet
                    Deposition
                    (40)
 Dry
 Deposition
 (2)
Figure 5. Process analysis illustration of sulfur circulation in the Bast Asia.

Table 1. Comparison of sulfur budget and life time scales of SO2 and sulfate aerosol among
global scale studies and the present study.

Present Study
Chin etal.( 1996)
Phametal.(1995)
Takemura (2000)
Langner and Rodhe
(1991)
Ratio between gas-
and aqueous-phase
conversion of SO,
1: 1.8
1:5.5
1:9.0
1:9.5
1:5.4 or 1: 1.8
Lifetime of SO2
1.4 day
1.2 day
0.6 day
1.2 day
1.2 or 1.8 day
Lifetime of sulfate
aerosol
3.1 day
3.9 day
4.7 day
2.6 day
3.2 or 6.1 day
CONCLUSIVE REMARKS

     We have presented simulation results of RAMS/CMAQ system applied to East Asia
covering eastern half of China, Korean peninsula, and Japan Islands for January 1997. The
simulation results were reasonable when compared with sulfate observations available at a
few sites in the region. The estimated sulfur budget is comparable with other studies and it
provides regional insights on the importance of different physical and chemical processes in
determining source-receptor relations of the acid deposition problem in East Asia.

ACKNOWLEDGMENTS AND DISCLAIMER

     The authors express their appreciation to Drs. Shiro Hatakeyama of NEBS and S.-G. Shim of
KAIST for providing the observation data. This paper has been reviewed in accordance with
the U.S. Environmental Protection Agency's peer and administrative review policies and
approved for presentation and publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.

-------
REFERENCE

Akimoto, EL, and H. Narita, 1994, Distribution of SO2, NO, and C02 emissions from fuel combustion and
        industrial activities in Asia with 1° x 1° resolution, Atmos. Environ,, 28,213-225,
Byun, D.W, and J.K.S. Ching, ed., 1999, Science Algorithms of the EPA Models-3 Community Multi-scale Air
        Quality (CMAQ) Modeling System, NERL, Research Triangle Park, NC. [Available from National
        Exposure Research Laboratory, U.S. Environmental Protection Agency, Research "Triangle park, NC
        27711]
Carmidhael, G.R., I. Uno, MJ, Phandis, Y. Zhang, and Y. Sunwoo, 1998, Tropospheric ozone prediction and
 1       transport in the springtime in East Asia. J. Geophys. Res., 103,10,649-10,671.
Chin, M,, D.J. Jacob, G.M, Gardner, M.S. Foreman Fowler, and PA. Spiro, 1996, A global three-dimensional
        mode,! of tropospheric sulfate. J. Geophys. Res., 101,18,776-18,690.
Langner J. and I. Rodhe, 1991, A global three-dimensional model of the tropospheric sulfur cycle. J. Atmos.
       'CAern., 13,225-263.
Louis, J.-F., 1979, A parameteric model of vertical eddy fluxes in the atmosphere. Boundary-Layer Meteor.,
        17,187-202.
McHenry, J. N. and Dennis, R. L., 1994, The relative importance of oxidation pathway and clouds to
        atmospheric ambient sulfate production as predicted by the Regional Acid Deposition Model, J.
        Applied Meteor., 33,890-905.
Murano, J., S. Hatakeyama, T. Mizuguchi, and N. Kuba, 1995, Gridded ammonia fluxes in Japan, Water Air
        and Soil Pollution, 85,1,915-1,920.
Pham, M., J. F. Muller, G. P. Btasseur, C. Granier, and G. Megie, 1995, A three-dimensional study of the
        tropospheric sulfur cycle. /. Geophys. Res., 100,26,061-26,092.
Piccot,S., S. D. Watson, J.W. Jones, 1992, A global inventory of volatile organic compound emissions from
        anthropogenic sources, J.Geophy.Res., 97,9,897-9,912.
Takemura, T., H. Okamoto, Y. Maruyama, A.  Numaguti, A. Higurashi, and T. Nakajima, 2000, Global three-
        dimensional simulation of aerosol optical thickness distribution of various origins, submitted to J.
        Geophys. Res.
Uno, I., T. Ohara and K. Murano, 1998, Simulated acidic aerosol long-range transport  and deposition over
        East Asia - Role of synoptic scale weather systems, Air Pollution Modeling and its Application Vol.,
        22, S.E. Gryning and N. Chaumerliac ed., 185-193, Plenum Pub. Co.
Zhao, D. and A. Wang, 1994,  Estimation of anthropogenic ammonia emissions in Asia, Atmos. Environ., 28,
        689-694.

-------
a NEIL RTP-AMn~nO-089 TECHNICAL REPORT DATA
1 . REPORT NO . 2 .
EPA/60Q/A-00/032

4. TITLE AND SUBTITLE
Simulation of sulfate aerosol in east Asia using Models-
3/CMAQ with RAMS meteorological data
7, AUTHOR (S)
Seiji Stigata* , D.W. Byun2, and Itsushi Uno3
9. PERFORMING ORGANIZATION NAME AND ADDRESS
"National Institute for Environmental Studies, Tsukuba,
Ibaraki, Japan 305-0053
2Same as Block 12
3Kyushu University, Kasuga, Fukuoka , Japan 816-08580
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Research Triangle Park, NG 27711
3 . RECIPIENT ' S ACCESSION NO.
5 . REPORT DATE
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Proceedings, FY-00
14. SPONSORING AGENCY CODE
EPA/600/9
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The present study attempts to address a few challenges in utilizing the flexibility of the Models-3 Community Multiscale Air
Quality (CMAQ) modeling system. We apply the CMAQ system with the meteorological data provided by the Regional
Atmospheric Modeling System (RAMS) and to a different geographical area, East Asia covering eastern half of China,
Korean perninsula, and Japan Islands. To demonstrate the model performance, we compare the results with two time series
of non-sea salt sulfate that are available at several sites in the southern part of Japan during January 1997.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS

18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
b. IDENTIFIERS/ OPEN
ENDED TERMS

19. SECURITY CLASS (This
Report)
UNCLASSIFIED
20. SECURITY CLASS (This
Page)
UNCLASSIFIED
c . COSATI

21. NO. OF
PAGES
8
22. PRICE

-------