Evaluation and improvement of Air Pollutant emission inventory for Asian region by using Satellite column densities data.

Gakuji KURATA*, Pichnaree Lalitaporn, Yuzuru Matsuoka, Kyoto University, Japan *E-mail: gkurata@athehost.env.kyoto-u.ac.jp

2012 ACCENT-IGAC-GEIA Conference, Emission to Address Science and Policy Needs, 11-13 June, 2012, Toulouse, France

AcknowledgmentsThis research was partially supported by the Ministry of Education, Science, Sports and Culture, Japan, Grant-in-Aid for Science Research (B) , 21360254 , 2012. and the Global Environment Research Fund (S-6) by the Ministry of the Environment of Japan.

Satellite observations of tropospheric NO2 vertical column densities (VCDs) over SoutheastAsia including China and Japan are analyzed based on measurements from four satellitesensors; GOME, SCIAMACHY, OMI, and GOME-2 during the time period from 1996 to2011. The inter-annual variations and the consistency between the different satellite datasetsare investigated and compared with several emission inventory for Asian region. Thetropospheric NO2 VCDs over the study area have been simulated with Community Multi-scale Air Quality (CMAQ) model and then comparably analyzed with those retrieved fromsatellite observations in order to validate the accuracy of the emission inventories. The fifteenyears tropospheric NO2 VCDs data (1996-2011) from GOME, SCIAMACHY, OMI, andGOME-2 shows high increasing trends in China, especially in Beijing and Shanghai. Most ofthe results from the model simulations of horizontal tropospheric NO2 VCDs distributiongenerally agree well with satellite measurements. Overall, the discrepancies among theCMAQ model and satellite retrievals are mainly due to inaccurate emission inventories fedinto the model and the uncertainties in the satellite retrievals. However, as a result of theconsistency between satellite-retrieved and model simulated tropospheric NO2 VCDs, itsuggests that integration of satellite data with air quality model can be used to evaluate andimprove the accuracy of emission inventories.

Abstract

1. Satellite retrievals: Satellite-based tropospheric NO2 columns are retrieved from level-2 products of GOME, SCIAMACHY, OMI and GOME-2 published in the TEMIS website (http://www.temis.nl).

2. Emission inventory: REAS emission: Regional Emission inventory in Asia. MACCity emission: Global emission inventory. Kyoto Univ. emission(AIM): Regional Emission in Asia.

3. Model description: WRF 3.3 80km mesh (Jan –Dec , 2005)

NCEP-CFSR (0.5degree) Noah land-surface model WSM 6-class graupel scheme

CMAQ 4.7 Chemistry: CB-05- AERO5 Boundary condition : MOZART4

Methodology and Data

Outline of the study

GCMOutput

LanduseTerrain

WRF

EmissionMesh data

Meteo.Field

CalculatedConcentration

HealthImpact

BoundaryCondition

Chemical TransportModel

CMAQ

Co-benefitAnalysis

DeathDisease

Impact AssessmentExposure

Outdoor

MicroEnvironment

Indoor●Indoor Emission

(Cooking, Heating,Hot water, Lighting)(Oil, Coal, Wood, Charcoal, etc)

●Time use data(Each Cohort)

● Room / House / Building parameter

● Ventilation condition

Meteorological Model

Target AreaLocal Administrative level

N = 6,695

Large Point Source N = 16,956

SectorsPower PlantIron and SteelCementPetrochemicalPaper and Pulpother IndustryPassenger transportationFreight transportationCommercialResidential

Target Year : 2005 (2010) (2020) (2030)

Emission Inventory of Asian Countries

Application to Asian Countries

Emission Mesh

Estimation of Emission

Collection and Organization of Information of Large Point

Source and Area activity

ArcGIS

Monthly average of CMAQ NO2 VCD at Satellite over-pass time(10:30 LST): There are clear annual variation in northern part of China. It seems that there is no influence of a long-range transport.

FEB MAR APR

MAY JUN JUL AUG

SEP OCT NOV DEC

(molecs/cm2)

Model output for NO2 VCDsJAN

Comparison between CMAQ output and Satellite NO2 VCD

The comparison between NO2 VCDs from CMAQ & SCIAMACHY at Satellite over-pass (10:30 LST)

The Ratio of NO2 VCDs of Model vs. SCIAMACHY at satellite over-pass (10:30 LST) for every 3 month average.

Summary• Regarding the qualitative relationship between the satellite NO2 VCDs data and

emission inventory around the megacities, it became clear that it is well inagreement especially in Beijing and Shanghai.

• It was clearly shown that systematic errors exists in our original emissioninventory used in the CMAQ simulation by the comparison between modelsimulation and satellite observation for Year 2005.

• In particular, the systematic underestimate exists in the area along the shore ofChina and the Indochinese Peninsula.

• On the other hand, overestimation was seen around several area and cities, suchas northern India and Singapore.

• The tendency of an underestimate may be strong in the winter of the NorthernHemisphere at high latitude. Our assumption of a seasonal variation may not beright.

• It can be expected that this kind of analyses can provide compensation ofemission source data with useful information.

Seasonal Variability of NO2

0

50

100

Jan-

96Ju

l-96

Jan-

97Ju

l-97

Jan-

98Ju

l-98

Jan-

99Ju

l-99

Jan-

00Ju

l-00

Jan-

01Ju

l-01

Jan-

02Ju

l-02

Jan-

03Ju

l-03

Jan-

04Ju

l-04

Jan-

05Ju

l-05

Jan-

06Ju

l-06

Jan-

07Ju

l-07

Jan-

08Ju

l-08

Jan-

09Ju

l-09

Jan-

10Ju

l-10

Jan-

11Ju

l-11

NO

2(1

015m

olec

. cm

-2) Tropospheric NO2 columns over Beijing

GOME GOME-2 SCIAMACHY OMI

0

10

20

Jan-

96Ju

l-96

Jan-

97Ju

l-97

Jan-

98Ju

l-98

Jan-

99Ju

l-99

Jan-

00Ju

l-00

Jan-

01Ju

l-01

Jan-

02Ju

l-02

Jan-

03Ju

l-03

Jan-

04Ju

l-04

Jan-

05Ju

l-05

Jan-

06Ju

l-06

Jan-

07Ju

l-07

Jan-

08Ju

l-08

Jan-

09Ju

l-09

Jan-

10Ju

l-10

Jan-

11Ju

l-11

NO

2(1

015m

olec

. cm

-2) Tropospheric NO2 columns over Bangkok

GOME GOME-2 SCIAMACHY OMI

0

10

20

Jan-

96Ju

l-96

Jan-

97Ju

l-97

Jan-

98Ju

l-98

Jan-

99Ju

l-99

Jan-

00Ju

l-00

Jan-

01Ju

l-01

Jan-

02Ju

l-02

Jan-

03Ju

l-03

Jan-

04Ju

l-04

Jan-

05Ju

l-05

Jan-

06Ju

l-06

Jan-

07Ju

l-07

Jan-

08Ju

l-08

Jan-

09Ju

l-09

Jan-

10Ju

l-10

Jan-

11Ju

l-11

NO

2(1

015m

olec

. cm

-2) Tropospheric NO2 columns over Jakarta

GOME GOME-2 SCIAMACHY OMI

Time series of monthly tropospheric NO2 columns from GOME, SCIAMACHY, OMI & GOME-2 satellites for the Megacities in SEA including China & Japan from 1996-2011 were compared.

Shanghai has the highest increasing trend of 21.5% per year followed by Beijing with 14.1% per year (Ref. year 1996).

Mid/Low-latitude zone: maximum of tropospheric NO2 columns can be seen during wintertime (November-February) & minimum during summertime (June-August).

Equator-latitude zone: maximum of tropospheric NO2 columns can be seen during dry season (June-August) & minimum during rainy season (December-February).

Comparison of Satellite data and Emission inventories (REAS)The comparison of REAS NOx emissions & annual average of tropospheric NO2 columns from GOME, SCIAMACHY & GOME-2 satellites during 1996-2009.

The cities that located in mainland (Shanghai, Beijing, Bangkok, Hanoi and Phnom Penh): present relatively good relationships between REAS NOxemissions and tropospheric NO2 columns (R > 0.7).

The cities that located near coastal area (Naypyidaw, Dili, Singapore): the relationships between REAS NOxemissions and tropospheric NO2 columns didn’t show a good agreement. We need to identify the reason. (emission? or meteorology? )

NO2 columns: 27.80 % yr-1

REAS NOx: 5.00 % yr-1

R = 0.900

50000

100000

150000

0

10

20

30

40

RE

AS

NO

x(1

015m

olec

. cm

-2yr

-1)

Sate

llite

NO

2(1

015m

olec

. cm

-2)

NOx emissions & NO2 columns: Shanghai

GOME GOME-2 SCIAMACHY avg allNOx 線形 (avg all) 線形 (NOx )

NO2 columns: 14.56 % yr-1

REAS NOx: -0.75 % yr-1

R = -0.51 60000

65000

70000

75000

80000

0

5

10

15

RE

AS

NO

x(1

015m

olec

. cm

-2yr

-1)

Sate

llite

NO

2(1

015m

olec

. cm

-2)

NOx emissions & NO2 columns: Singapore

GOME GOME-2 SCIAMACHY avg allNOx 線形 (avg all) 線形 (NOx )linear(avg all) linear(NOx)

linear(NOx)linear(avg all)

Comparison of Satellite data & Emission inventory(MACCity) The long-term trend including seasonal variation were compared between MACCityNOx emissions & tropospheric NO2 columns from satellites during 1996-2010.

The cities that located in mainland (Shanghai, Beijing and Hanoi): the seasonal cycle of NOx emissions and tropospheric NO2 columns are in good agreement (R > 0.65).

The cities that located near coastal area: the correlations of MACCity NOxemissions and tropospheric NO2 columns are low and the seasonal variation of tropospheric NO2 columns from satellites were not clear.

NO2 columns: 16.39% yr-1

MACCity NOx: 7.02% yr-1

R = 0.68

0

10000

20000

30000

40000

50000

0

20

40

60

80

100

Jan-

96Ju

l-96

Jan-

97Ju

l-97

Jan-

98Ju

l-98

Jan-

99Ju

l-99

Jan-

00Ju

l-00

Jan-

01Ju

l-01

Jan-

02Ju

l-02

Jan-

03Ju

l-03

Jan-

04Ju

l-04

Jan-

05Ju

l-05

Jan-

06Ju

l-06

Jan-

07Ju

l-07

Jan-

08Ju

l-08

Jan-

09Ju

l-09

Jan-

10Ju

l-10

MA

CC

ityN

Ox

(1015

mol

ec. c

m-2

yr-1

)

Sate

llite

NO

2(1

015m

olec

. cm

-2)

NOx emissions & NO2 columns: Beijing

GOME GOME-2 SCIA OMIavg all NOx 線形 (avg all) 線形 (NOx )

NO2 columns: 13.81% yr-1

MACCity NOx: -0.80% yr-1

R = -0.36

0

10000

20000

30000

40000

50000

0

5

10

15

20

Jan-

96Ju

l-96

Jan-

97Ju

l-97

Jan-

98Ju

l-98

Jan-

99Ju

l-99

Jan-

00Ju

l-00

Jan-

01Ju

l-01

Jan-

02Ju

l-02

Jan-

03Ju

l-03

Jan-

04Ju

l-04

Jan-

05Ju

l-05

Jan-

06Ju

l-06

Jan-

07Ju

l-07

Jan-

08Ju

l-08

Jan-

09Ju

l-09

Jan-

10Ju

l-10

MA

CC

ityN

Ox

(1015

mol

ec. c

m-2

yr-1

)

Sate

llite

NO

2(1

015m

olec

. cm

-2)

NOx emissions & NO2 columns: Singapore

GOME GOME-2 SCIA OMIavg all NOx 線形 (avg all) 線形 (NOx )linear(avg all) linear(NOx)

linear(NOx)linear(avg all)