tinus pulles how to establish the uncertainties? particulate emission inventory for europe

16
Tinus Pulles How to establish the uncertainties? Particulate Emission Inventory for Europe

Upload: reginald-casey

Post on 13-Dec-2015

226 views

Category:

Documents


2 download

TRANSCRIPT

Tinus Pulles

How to establish the uncertainties?

Particulate Emission Inventory for Europe

Tinus Pulles

Outline

• Background• Methods

• Inventory methods• Uncertainty estimation

• The inventory• Uncertainties • Conclusions

Tinus Pulles

Background: Objective

• This presentation uses data from a project performed for DG Environment to support EU Member States in developing a PM2.5 inventory

• Quality requirements:• Transparency means that the assumptions and methodologies used

for the inventory should be clearly explained • Comparability means that estimates of emissions should be

comparable among Member States• Completeness means that an inventory covers all sources • Accuracy is a relative measure of the exactness of an emission or

removal estimate

• Partners in the project where: • TNO (NL, lead),• AEA-T (UK) • IVL (SE)

Tinus Pulles

Inventory approach

• Act_ID• Sector• Location• Time• Value

Activities

• Tech_IDTechnologies

• EF_ID• Tech_ID• Pollutant• Value

Emission factors

• S_T_ID• Act_ID• Tech_ID• fraction

Select_Tech

TFEIP

Guidebook PM2.5 update

TEAMTNO Emissions Assessment Model

Tinus Pulles

Data sourcesEmission Factors from CEPMEIP: power plants

Sum of EF PollutantCRF/NFRCode FuelName Technology EF_Unit PM2.5 PM10 TSP1.A.1.a Heavy Fuel Oil Autoproducer electricity, heat and CHP plants, and public heat

plants: Low S-fuel with optimized burner or ESPg/GJ 27.5 33 33

Autoproducer electricity, heat and CHP plants, and public heat plants: Low-medium S-fuel, conventional installation

g/GJ 108 180 240

Autoproducer electricity, heat and CHP plants, and public heat plants: Medium S-fuel, conventional installation

g/GJ 120 240 480

Lignite/Brown Coal Autoproducer electricity, heat and CHP plants, and public heat plants: Conventional ESP (removal efficiency around 98%)

g/GJ 180 720 1440

Autoproducer electricity, heat and CHP plants, and public heat plants: Efficient 3 or 4-field ESP only (removal efficiency about 99.5%)

g/GJ 70 150 200

Autoproducer electricity, heat and CHP plants, and public heat plants: FGD or fabric filter; <20 mg/Nm3 stack PM concentration; BAT

g/GJ 24 32 36

Natural Gas Autoproducer electricity, heat and CHP plants, and public heat plants: Burner with optimized combustion

g/GJ 0.9 0.9 0.9

Autoproducer electricity, heat and CHP plants, and public heat plants: Conventional installation

g/GJ 4.2 4.2 4.2

Other Bituminous Coal & Anthracite

Autoproducer electricity, heat and CHP plants, and public heat plants: Conventional ESP (removal efficiency equal or better than 98%)

g/GJ 68 280 560

Autoproducer electricity, heat and CHP plants, and public heat plants: Efficient 3 or 4-field ESP only (removal efficiency about 99.5%)

g/GJ 108 225 315

Autoproducer electricity, heat and CHP plants, and public heat plants: FGD or fabric filter; <20 mg/Nm3 stack PM concentration; BAT

g/GJ 50 60 60

Wood and wood waste Public electricity and CHP plants: Uncontrolled conventional g/GJ 880 1120 1600

Tinus Pulles

Data sourcesEmission Factors from CEPMEIP: technology selected

Power Plants on Other Bituminous Coal & Anthracite

• Per country one technology selected

• Based on the CEPMEIP 1995 inventory

• Country data needed to improve:

• EF measurements?• Technologies used?

Sum of Emission(kTon)

Technology

CRF/NFRCode CountryName

Autoproducer electricity, heat and

CH

P plants, and public heat plants:

Conventional E

SP

(removal

efficiency equal or better than 98%)

Autoproducer electricity, heat and

CH

P plants, and public heat plants:

Efficient 3 or 4-field E

SP

only (rem

oval efficiency about 99.5%)

Autoproducer electricity, heat and

CH

P plants, and public heat plants:

FG

D or fabric filter; <

20 mg/N

m3

stack PM

concentration; BA

T

1.A.1.a Austria 0.19Belgium 1.28Czech Republic 1.37Denmark 0.77Finland 0.43France 3.17Germany 4.52Greece 0.00Ireland 0.78Italy 3.04Netherlands 1.07Poland 15.81Portugal 1.61Slovak Republic 0.52Spain 7.78Sweden 0.07United Kingdom 13.32

Grand Total 17.70 30.98 7.05

Tinus Pulles

Particulates emitted in EuropeNational totals

• Total emissions in the EU25: 1300 kTon.

• The other countries in this study add about 500 kTon

• Total PM2.5 emissions are about 40% of the PM10 emissions

• Larger countries contribute more to the total emissions than smaller countries.

• Some discrepancies with RAINS estimates

Emission(kTon) PollutantCountryGroup CountryName PM2.5 PM10EU25 Austria 23.92 46.66

Belgium 34.67 65.59Cyprus 1.95 4.46

Czech Republic 35.99 121.55Denmark 16.71 50.79

Estonia 9.48 27.25Finland 21.35 51.16France 228.72 525.60

Germany 161.77 359.90Greece 37.28 88.20

Hungary 26.73 98.41Ireland 12.60 34.16

Italy 140.07 318.11Latvia 15.53 39.38

Lithuania 15.69 53.18Luxembourg 3.77 7.96

Malta 0.68 1.35Netherlands 23.45 48.13

Poland 141.06 454.58Portugal 36.30 78.58

Slovak Republic 12.98 42.09Slovenia 7.78 15.25

Spain 140.13 354.64Sweden 23.44 61.75

United Kingdom 109.53 234.94EU25 Total 1 281.57 3 183.69

CC Bulgaria 37.02 123.38

Romania 86.75 262.85Turkey 281.37 804.87

CC Total 405.14 1 191.10

AC Albania 3.97 12.40

Bosnia-Herzegovina 13.24 45.17Croatia 15.83 41.80

Federal Republic of Yugoslavia (Serbia&Montenegro) 49.27 174.82Former Yugoslav Republic of Macedonia 6.66 17.17

AC Total 88.98 291.37

EFTA Iceland 0.41 0.88

Norway 13.58 28.41Switzerland 8.02 16.49

EFTA Total 22.01 45.78

Grand Total 1 797.70 4 711.93

Tinus Pulles

Particulates emitted in EuropeContributions by fuel

• Non-combustion sources contribute• more than half of the PM10

emissions• one third to the PM2.5 emissions.

The very large share to PM10 emissions is due to the inclusion of wind blown dust in agriculture.

• Solid fuels are by far the largest combustion source for PM10, followed by Liquids and Biomass.

• For PM2.5 the largest combustion source is Liquids, followed by Biomass and Solids.

Emission(kTon) FuelGroup SectorGroupName PM2.5 PM10

Solids Combustion in Energy Industry 184 622

Manufacturing Industry 67 167 Non-industrial combustion 23.1 46.2 Mobile sources 0.0568 0.152

Solids Total 274 836

Liquids Combustion in Energy Industry 42.7 63.5

Manufacturing Industry 56.9 75.6 Non-industrial combustion 25.6 27.4 Mobile sources 335 338

Liquids Total 460 504

Gaseous Combustion in Energy Industry 7.16 7.16

Manufacturing Industry 2.58 2.58 Non-industrial combustion 1.47 1.47 Mobile sources 0.038 0.038

Gaseous Total 11.2 11.2

Waste Combustion in Energy Industry 9.85 12.7

Manufacturing Industry 1.93 2.77

Waste Total 11.8 15.4

Biomass Combustion in Energy Industry 15.8 18.5

Manufacturing Industry 0.703 1.03 Non-industrial combustion 383 408

Biomass Total 399 427

None Combustion in Energy Industry 6.14 10.3

Fugitive emissions from Energy 38.6 135

Industrial Processes & Product Use 205 413

Mobile sources 32.8 87.2 Agriculture 356 2270 Waste treatment 3.1 3.1

None Total 641 2920 Grand Total 1800 4710

Tinus Pulles

Comparison our national totals with national reports

PM2.5 y = 1.1672x

R2 = 0.7931

0.1

1.0

10.0

100.0

1000.0

0.1 1.0 10.0 100.0 1 000.0

National reports

Ou

r es

tim

ate

PM10 y = 0.7068x

R2 = 0.9027

0.1

1.0

10.0

100.0

1000.0

0.1 1.0 10.0 100.0 1 000.0

National reports

Ou

r es

tim

ate

• Horizontally: ETC-ACC gap filled national reported data for 2000• Vertically: our estimate for 2000

• Major differences occur, but generally not too bad…• Are these differences within our uncertainties?

Tinus Pulles

Uncertainty analysis

• Monte Carlo simulation• Probability Distribution Functions

• Activity data: • normal distributions, • standard deviation 10%

• Emission factors:• Lognormal distributions• Standard deviation based on expert judgement, derived

from the between technology variations in CEPMEIP• For a few sample countries:

• Main sector uncertainties• Compare national report with uncertainty range from Monte

Carlo

Tinus Pulles

Uncertainties: example for United Kingdom

• Dots: probabiliuty distribution for our estimate

• Red line: UK’s report

• Seems to be all right!

pm2.5 emissions in United Kingdom

0%10%

20%30%

40%50%60%

70%80%

90%100%

- 20

00

0

40

00

0

60

00

0

80

00

0

10

0 0

00

12

0 0

00

14

0 0

00

16

0 0

00

United Kingdom Value 2.5% 97.5% 50.0%Combustion in Energy Industry 24 619 14 778 58 152 26 824 Fugitive emissions from Energy 3 632 1 846 7 182 3 665 Manufacturing Industry 2 680 2 346 3 340 2 779 Non-industrial combustion 6 121 4 378 9 821 6 462 Mobile sources 41 352 37 516 51 542 44 120 Industrial Processes & Product Use 15 200 11 981 23 773 16 124 Agriculture 15 752 11 178 25 488 16 326 Waste treatment 256 44 1 513 256 National Total 109 611 99 945 151 456 119 420

80 115 150 185 220

5% 90% 5% 103.5816 143.4457

Mean=120684

Distribution for pm2.5 emissions in United Kingdom/P10

Va

lue

s in

10

^ -5

Values in Thousands

0.000

0.500

1.000

1.500

2.000

2.500

3.000

3.500

4.000

Mean=120684

80 115 150 185 220

Regression Sensitivity for pm2.5 emissions in United Kin...

Std b Coefficients

EF: EURO II - 1996 heavy d.../N276 .064

EF: Conventional refinery;.../N107 .073

EF: EURO I - 1992 heavy du.../N265 .073

EF: Conventional installat.../N73 .074

EF: Conventional stove or .../N116 .093

EF: Conventional installat.../N71 .098

EF: No control - light dut.../N334 .099

EF: All activities: (RAI;.../N13 .111

EF: All countries: (SIE; .../N19 .114

EF: EURO II - 1996 heavy d.../N278 .121

EF: EURO II - 1996 heavy d.../N277 .125

EF: All activities: (UFW;.../N17 .148

EF: Uncontrolled; optimize.../N262 .153

EF: Conventional installat.../N77 .231

EF: No control - Arable la.../N343 .265

EF: Efficient 3 or 4-field.../N129 .707

-1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1

Tinus Pulles

Uncertainties: example for Germany

• Dots: probability distribution for our estimate

• Red line: Germany’s report

• Our estimate is significantly higher

pm2.5 emissions in Germany

0%10%

20%30%

40%50%

60%70%

80%90%

100%- 5

0 0

00

10

0 0

00

15

0 0

00

20

0 0

00

25

0 0

00

Germany Value 2.5% 97.5% 50.0%Combustion in Energy Industry 21 268 14 848 38 772 22 945 Fugitive emissions from Energy 5 713 3 260 11 362 6 031 Manufacturing Industry 1 520 1 338 1 831 1 554 Non-industrial combustion 31 369 19 515 55 741 31 675 Mobile sources 54 145 47 642 69 876 57 668 Industrial Processes & Product Use 17 715 13 084 39 376 20 178 Agriculture 29 722 20 564 49 290 30 604 Waste treatment 324 53 1 953 325 National Total 161 773 148 739 212 867 175 663

120 155 190 225 260

5% 90% 5% 152.4633 206.0978

Mean=177070.9

Distribution for pm2.5 emissions in Germany/P10

Val

ues

in 1

0^ -

5

Values in Thousands

0.000

0.500

1.000

1.500

2.000

2.500

3.000

Mean=177070.9

120 155 190 225 260

Regression Sensitivity for pm2.5 emissions in Germany/P10

Std b Coefficients

EF: FGD or fabric filter; .../N145 .076

EF: No control - heavy dut.../N318 .08

AR: 1.A.4.b; Wood and wood.../W247 .082

EF: Fabric filter, high ef.../N142 .086

EF: All activities: (RAI;.../N13 .099

EF: EURO II - 1996 heavy d.../N277 .113

EF: All countries: (SIE; .../N19 .114

EF: No control - Arable la.../N342 .119

EF: No control - heavy dut.../N321 .14

EF: FGD or fabric filter; .../N144 .151

EF: All activities: (UFW;.../N17 .17

EF: EURO II - 1996 heavy d.../N278 .19

EF: FGD or fabric filter; .../N143 .264

EF: High efficiency ESP or.../N148 .274

EF: No control - Arable la.../N343 .413

EF: Modern emission optimi.../N184 .543

-1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1

Tinus Pulles

Comparison for National Total PM2.5 emissions

• For many countries the reported value is within the 95% confidence interval of our estimate

• For quite a few the deviation is larger.

Reported within 95% confidence intervalReported below 95% confidence intervalReported above 95% confidence interval

Country National Total Estimate 2.5 %-ile 97.5 %-ile 50 %-ile ReportedAustria National Total 24 022 20 187 35 288 25 832 25 881 Belgium National Total 34 674 32 950 51 506 38 606 35 241 Denmark National Total 16 763 14 684 23 175 18 170 22 924 Finland National Total 21 458 18 574 30 438 23 284 37 663 France National Total 228 745 193 215 331 830 243 751 341 592 Germany National Total 161 773 148 739 212 867 175 663 115 091 Greece National Total 37 276 32 191 52 443 40 047 48 578 Hungary National Total 26 736 23 077 38 006 28 922 25 720 Ireland National Total 12 604 11 788 16 682 13 816 12 237 Italy National Total 140 141 131 431 198 949 153 812 - Latvia National Total 15 576 11 229 25 265 16 227 10 996 Lithuania National Total 15 753 12 083 23 996 16 638 8 798 Netherlands National Total 23 446 22 421 31 100 25 716 28 806 Norway National Total 13 629 11 893 18 503 14 378 58 416 Poland National Total 114 731 95 572 176 777 125 825 131 459 Portugal National Total 36 311 30 392 59 069 39 479 95 062 Slovak Republic National Total 12 979 11 779 18 334 14 339 26 000 Slovenia National Total 7 780 5 788 12 279 8 263 6 501 Spain National Total 140 132 126 653 184 012 139 481 Sweden National Total 23 561 20 107 34 871 25 379 46 053 United Kingdom National Total 109 611 99 945 151 456 119 420 108 236

Tinus Pulles

Conclusions

• We have produced an PM2.5 and PM10 emission inventory for Europe (2000)

• The emission estimates in this inventory show considerable uncertainties

• For more than half of the countries the national reported value is within our 95% confidence interval

• We might have underestimated (residential) biomass use

• We did not detect a significant source that might be missed in national reports.

• The PM inventories might be as good as the ones used by Dick Derwent! (Thank you Dick)

• We do not know what we don’t know

Tinus Pulles

Thank you

Tinus Pulles

TNO’s TEAM model