1 1 Model studies of some atmospheric aerosols and comparisons with measurements K. G e o r g i e v I P P – B A S, S o f i a, B u l g a r i a.

1 1

Model studies of some atmospheric Model studies of some atmospheric aerosols and comparisons with aerosols and comparisons with

measurementsmeasurements

K. G e o r g i e vK. G e o r g i e vI P P – B A S,I P P – B A S,

S o f i a, B u l g a r i aS o f i a, B u l g a r i a

2 2

IntroductionIntroduction

DG Joint Research Centre of European Commission

Institute for Environment and Sustainability

(IES – JRC, Ispra, Italy)

M. Krool, F. Dentener, E. Vignati

3 3

AerosolsAerosols

big and increasing importance for the assessment of air quality and climate forcing;

play a crucial role for global temperature modifications;

research of aerosols in the surface layer is mainly motivated of their impact on human health and possible ecological effects.

4 4

AerosolsAerosols

Aerosols are defined as relatively stable suspensions of solid or liquid particles in a gas.

The aerosols can be classified as primary (those that are emitted in particulate form directly from sources) and secondary (particles produced in the atmosphere).

5 5

AerosolsAerosols

The formation of the secondary aerosols can be done mainly by :

reaction of gases to form low–vapor–pressure products;

reaction of gases on the surfaces of existing particles to form condensed phase products;

chemical reactions within the aerosol itself (for example, SO2, oxidation to sulfate)

6 6

TM5 ModelTM5 Model

TM5TM5 – 3D global chemistry Transport ModelAllows two-way nested zooming which leads to possibility to run the model on relatively very fine space grid (11) (longitude x latitude) over selected regions (Europe is most often used but North America, Africa, Asia and South America can be treated separately or in combinations)The coarsest space resolution is (64) and between these two is a grid (32)

7 7

Zooming EuropeZooming Europe

8 8

Zooming AfricaZooming Africa

9 9

Zooming AsiaZooming Asia

10 10

Zooming North AmericaZooming North America

11 11

Vertical resolutionVertical resolution

Dashed lines – the 60 hybrid sigma-pressure (terrain following) levels of the operational ECMWF model;

Solid lines – the subset employed by the 25 layer European zoom model;

five layers – boundary layer

ten layers – free troposphere

ten layers – stratosphere

12 12

The modelThe model

The TM5 model is designed for

Global studies of atmospheric chemistrysuch as

intercontinental and interhemispheric exchange;

effects of grid refinement on the budgets of chemically active compounds.

13 13

The modelThe model

The TM5 model is an offline model, using

preprocessed meteorological fields from

ECMWF.

14 14

The modelThe modelSplittingSplitting

The basic model operations (advection,

convection, sources, chemistry) are solved

by symmetrical operator splitting

15 15

The modelThe modelSplittingSplitting

Symmetrical splitting can not always be presented in an zooming algorithmNote: X and Y – horizontal advection

Z – vertical advection V – vertical diffusion and

convection C – chemistry (incl. emissions

and depositions)

16 16

The modelThe modelSplittingSplitting

Let

– the parents write the boundary conditions to their children

– the parents are updated with the information by their children

17 17

The model – splittingThe model – splittingThree-region European-focused TM5 versionThree-region European-focused TM5 version

t ………………………………………………………………..t + ΔT/2

Region 1 XYZ……………………………………………………………...VC

Region 2 ……..XYZ…………………..VC CVZYX…………………

Region 3 ……………...XYZVC CVZYX……………….CVZYX XYZVC

t + ΔT/2 …………………………………………………………….. t + ΔT

Region 1 CVZYX…………………………………………………………………..

Region 2 ………… CVZYX…………………… XYZ……………………VC

Region 3 ……………………… CVZYX XYZVC............. XYZVC CVZYX……

18 18

The modelThe modelChemistryChemistry

Gas phase chemistry is calculated using the

CBM-IV chemical mechanism solved by

means of the EBI method.

Photochemistry and aerosols are coupled in

the IPCC version of the TM5 model.

19 19

The modelThe modelAerosolAerosol ChemistryChemistry

Aerosols are assumed internally mixed and in an

accumulation mode size distribution for the

calculations of both scavenging and depositions.

They can contain sulphate, ammonium and nitrate

and are described using bulk approach.

20 20

The modelThe modelAerosolAerosol ChemistryChemistry

The water attached to the particles is determined

from the ambient relatively humidity. Sulphate is

reduced to its aerosol phase. It is obtained by the

oxidation of sulphur dioxide in the gas phase by

OH radical. In the aqueous phase this oxidation is

done by H2O2 and ozone.

21 21

Aerosol studiedAerosol studied

sulfate (SO4,);

ammonium (NH4);

nitrate (NO3)

sulfur dioxide (SO2) , ammonia (NH3) , nitric acid (HNO3)

(due to some chemical transformations which play an

essential role in creation and the live cycle of SO4, NH4, NO3)

22 22

Aerosol measurementsAerosol measurements

There are relatively enough EMEP stations (57) with

measurements for sulfate and sulfur dioxide. The stations

with available measurements for ammonium and ammonia

are 23 while the stations presenting measurements for nitrate

and nitric acid are only 14 and not all of them have

measurements for both.

23 23

Aerosol measurementsAerosol measurements

The measurements reported by the EMEP network

stations are surface measurements. The model

output predictions which are discussed and used

for the comparisons are taken in the first model

layer.

24 24

RResults vs measurements for esults vs measurements for sulphatesulphate

Year 2000: yearly average output of SO4

Yearly average Yearly average Yearly average Station

Correlation Remarks

Station

Correlation Remarks

Station

Correlation Remarks

AT02 0.56 + GB02 0.62 - PL02 0.24 -

CH01 0.61 + GB04 0.50 - PL03 0.16 - bad

CH02 0.52 + GB06 0.75 - PL04 0.45 - bad

CH05 0.55 + GB07 0.67 - PL05 0.32 - bad

DK03 0.60 - GB13 0.77 - OK RU16 0.37 + bad

ES04 0.66 - GB14 0.80 - OK RU18 0.13 + bad

25 25

RResults vs measurements for esults vs measurements for sulphatesulphate

The correlation between measurements end model results is relatively good.

The model results underestimate the measurements in the most of the stations (65 %).

26 26

RResults vs measurements for esults vs measurements for sulphatesulphate and and sulphur dioxidesulphur dioxide

60

8

39

21

14

37 7

21

0 1110 14

111

3

114

10 1

0

10

20

30

40

50

60

Nu

mb

er o

f st

atio

ns

Total North part Central part South part GB

Location of the station

Correlation coefficient analysis on the yearly base for SO4

Number of records (stations) With correlation coefficient ≥ 0.75With correlation coefficient ≥ 0.50 With correlation coefficient < 0.50

27 27

RResults vs measurements for esults vs measurements for sulphatesulphate

In total:

65 % corr. coeff. > 0.5

>13 % > 0.75

“GB” : > 90% corr. coeff. > 0.5

36% corr. coeff. > 0.75

“North” + “Central”: 50% corr. coeff. > 0.5

28 28

RResults vs measurements for esults vs measurements for sulphatesulphate and and sulphur dioxidesulphur dioxide

TM5 model

overpredicts sulphur dioxide

and

underpredicts sulphate

29 29

RResults vs measurements for esults vs measurements for ammoniumammonium and and ammoniaammonia

TM5 model

overestimate ammonium

and

underestimate ammonia

30 30

RResults vs measurements for esults vs measurements for nitratenitrate and and nitric acidnitric acid

TM5 model

overpredicts nitrate

and

underpredicts nitric acid

31 31

ImplementationImplementation

TM5 program has been coded in Fortran 90

Implemented and tested on: IBM p690+ SGI Origin 3800 MAC OSX SUN Cluster