Docstoc

uno

Document Sample
uno Powered By Docstoc
					Interannual and Seasonal Variations of CMAQ-simulated
tropospheric NO2 in Asia and comparison with GOME
satellite data
- Combination of bottom-up and top-down analysis -




 Itsushi UNO*, Youjiang HE,
 Research Institute for Applied Mechanics, Kyushu University, Kasuga, Fukuoka,
 JAPAN

 Toshimasa OHARA, Jun-ichi KUROKAWA, Hiroshi TANIMOTO
 National Institute for Environmental Studies, Tsukuba, Ibaraki, JAPAN

 Kazuyo YAMAJI
 Frontier Research Center for Global Change, Yokohama, Kanagawa, JAPAN
How can we understand this year-by-year trend ?
Can we simulate these recent increase of NO 2 by CMAQ?
Can we reproduce satellite observed horizontal distribution
of NO2 ?
                NO2 from GOME measurements
Tropospheric NO2:
• Sources:
    – anthropogenic: transport, energy, biomass burning
    – natural: soil emissions, fires, lightning
• short lifetime, emission dominated

Data:
• 8 years of global measurements (1996.1 - 2003.6)
• Retrieval V2.0 data using
    – monthly AMFs based on MOZART 1997 profiles
    – surface reflectivity climatology
    – Stratosphere contribution by SLIMSCAT model
    – Provide Tropospheric NO2 column densities
• aerosol a priori assumptions
 a priori information is used, but no trend in a priori
• only daytime measurements (10:30LT ;40x320km;every 4 days)
Model simulation (Full calendar year calculation)
Past: 1980,1985,1990,1995 Recent: 1996-2006.3
                       Analysis Method
GOME-NO2 swath data (40x320km)           GOME-NO2 interpolate into 0.5x0.5˚
Observation 10:30LT
                                                REAS 1.1 Emission Inventory
                                                0.5 x 0.5 ˚ Lon-Lat Mesh




 CMAQ NO2 PS system (80km grid)
 3 hr interval




                                  Interpolate PS to Lon-Lat system. Use 3UTC data.
                                  Tropospheric NO2 column below 10km is integrated
                                  to get NO2 VCDS.
Sensitivity Experiments



EyyMyy   Year-by-year emission and      Control run
         meteorology
E00Myy   Fixed emission for year 2000   Fixed emission run



                                        Examination Domain

                                        CEC(1000km x 1000km)

                                        Japan

                                        (Korea)
Comparison of year 2000 annual average




    GOME-CMAQ
                            Model results under-
                            estimate NO2 VCDs over
                            the large source region
                            (especially Beijing
                            Region), overestimate
                            Taiwan and Korea.
GOME_NO 2 = -5.55E14 + 2.41 × CMAQ_NO 2 (molecule·cm2 ) (R=0.919).
                             Seasonal variation of NO2 VCDs
        China CEC region (7 year average)                                                   Japan




Maximum values of the NO2 columns occur in December even though the wind speed is higher. This indicates that the effect
of the longer chemical lifetime of NO2 is more important than that of strong wind. While the minimum value is observed in
July and August because of the strong vertical mixing, the short lifetime of NO2 and the inflow of relatively clean air from
the Pacific Ocean side. For CEC, CMAQ VCDs corresponds to 64% of the value of GOME VCDs in July.
CMAQ/REAS results under-estimate GOME.
But almost same under-estimates were reported
by many global CTMs inter-comparison paper by
Noije et al. (2006; ACP)




       CMAQ
                                                CMAQ



    Intercomparion results for year 2000.
    Green lines are satellite retrieval from 3 different groups.
Scatter plot of monthly averaged value of GOME NO2 and
CMAQ NO2 over China CEC.
                              This figure shows the scatter of
                              monthly averaged NO2 VCDs for
                              GOME and CMAQ EyyMyy. Red
                              numbers represent data from CEC
                              (last digit of the year). Blue symbols
                              are data from Japan.

                              This plot indicates that GOME NO2
                              is more enhanced when the CMAQ
                              NO2 concentration becomes higher
                              (i.e., emission becomes higher);
                              most of these conditions occur after
                              the year 2000.

                              The exact reason why the
                              relationship between CMAQ NO2
                              and GOME NO2 becomes nonlinear
                              remains unclear
Emission Trend Analysis by GOME and CMAQ/REAS
                                       An increasing trend of 1996–1998 and
Trend of GOME NO2, CMAQ NO2 and REAS   2000–2002 for GOME and CMAQ/REAS
NOx emission normalized to 2000.       shows a good agreement (GOME is
                                       approximately 10–11%·yr-1, whereas
                                       CMAQ/REAS is 8–9%·yr-1). The greatest
                                       difference also can be found between
                                       1998 and 2000. The CMAQ/REAS result
                                       shows only a few percentage points of
                                       increase, but GOME gives more than
                                       8%·yr-1 of increase.


                                       The most likely explanation is that the
                                       REAS emission trend (based on Chinese
                                       data) underestimates the rapid growth of
                                       emissions. This result highlights that
                                       combinations of CTM based on bottom-
                                       up inventories with satellite top-down
                                       estimates can play an important role in
                                       improving emission inventory estimates
                                       and provide very useful information that
                                       advances the development of a reliable
                                       CTM simulation.
O3 Fields




Tanimoto et al. (2006)
  Concluding Remarks

Systematic analyses of interannual and seasonal variations of tropospheric NO2
vertical column densities (VCDs) based on GOME satellite data and the CMAQ
were presented

The horizontal distribution of annual averaged GOME NO2 VCDs for 2000
generally agrees with CMAQ/REAS results. However, CMAQ results
underestimate GOME retrievals by factors of 2–4 over polluted industrial regions
such as Central East China

Evolution of the tropospheric columns of NO2 above Japan and CEC between
1996 and 2003 was examined.

Recent trends of annual emission increases in CEC were examined.

This study shows that the combinations of CMAQ based on bottom-up
inventories with satellite top-down estimates can play an important role
for air quality study.


             More detailed can be found in Uno et al. (2006, Atmos Chem. Phys. Submitted)
                                      OMI NO2 (from TEMIS web page)

Future Directions

1) High resolution CMAQ and
Satellite (Aura/OMI NO2)

2) Emission Inversion by adjoint


                                      Log10 [CMAQ NO2] (20km grid2)




                    CMAQ Grid2
                    RAMS Grid 2



                      CMAQ Grid1

                        RAMS Grid 1

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:5
posted:6/25/2011
language:English
pages:16