Air Pollution Scenarios towards 2030 in China: Emissions and Air Quality Prof. Kebin HE Department of Environmental Science and Engineering, Tsinghua University, Beijing 100084, China

Workshop on Hemispheric Transport of Air Pollution: Emission Inventories and Future projections Beijing, China October 18th - 20th, 2006

Outlines „ „ „ „ „ „

Background Working framework Scenario development Modeling methodology Modeling results Discussion

Outlines „ „ „ „ „ „

Background Working framework Scenario development Modeling methodology Modeling results Discussion

Historical emissions Activities

Energy Consumption

100000

5000

St eel Cement Chemi cal Fer t i l i zer

80000 70000

4500 4000 3500 3000

2500

50000

2500

Hydr o NG 1000 oi l 500 coal

40000

2000

2000

30000

1980 82

84

86

88

90

92

94

96

MTCE

0

2500

1500

20000 10000

3000

1500

2000

Tg

104 tons

60000

104 tons fertilizer

90000

0

98 2000 2002 2004

1500

1000

3000

1000

2500

500

2000

0

1500

1978 80

1000 500 0 1980 82

84

86

88

90

92

94

96

SO2 PM

500

Vehi cl e Popul at i on

4

10 vehi cl es

Emissions

98 2000 2002 2004

82

84

86

88

90

92

94

0 96 98 2000 1997 1998

02

04

1999

2000

2001

2002

2003

2004

2005

Why emission projection is needed

„ „

Rapid development in “10th five years”, but Only Environment target is not met 25490Gg

¾ SO2：

19950Gg

(Emitted in 2005)

(2000)

18000Gg (Target in 2005)

„ „

Aggressive environment target for “11th five years” Projection of emissions is needed to support policy making

Outlines „ „ „ „ „ „

Background Working framework Scenario development Modeling methodology Modeling results Discussion

Work Scope „

Purpose: To simulate energy consumption and emission of air pollutants up to 2030

„

Spatial Scope: China National Level

„

Time Step: 2000 (Base Year), 2005, 2010, 2020 and 2030

„

EI sectoral scope: power, industry, domestic, transportation and biomass burning

„

Pollutants: SO2, NOx, CO, VOC, BC and OC

Framework Energy

Environmental

Technology

Technology

Options

Options

Activities/Energy Use

Emissions

Air Quality

Tool: LEAP model

Tool: Updated Trace-P EI

Tool: Models-3/CMAQ

Energy

Environmental

Policy

Policy

Options

Options

Tech based EI methodology Energy consumption

Environmental Key of emission From LEAP projection Policies / tech

Tech character Tech split Activities

Fuel Quality

Unabated Emission Factor Abatement Efficiency Implementation Rate Emission Factor

Emissions

Outlines „ „ „ „ „ „

Background Working framework Scenario development Modeling methodology Modeling results Discussion

Scenario Definition „ BAU ¾ “Sustainable Energy Scenarios towards 2020” developed by ERI ¾ Long term environmental program by government ¾ Policies and regulations to ensure sustainable development

Scenario Definition „EP Policies Households energy saving

Industry energy saving

Building energy saving

Vehicle energy efficiency improvement

Indicators

Indicators in BAU

Energy saving lamp increases to 45% in 2010, and 70% in 2030 in urban households. 24% in 2010, and 53% in 2030 in rural households.

Urban: 34% in 2010, and 45% in 2030. Rural: 20% in 2010 and 40% in 2030

Energy intensity decreases: Iron and steel: 1.72% per year, Nonmetal minerals: 3.2% per year, Chemical products: 3.5% per year, Manufacturing and processing: 3.5% per year, etc.

Iron and steel: 1.62% per year, Nonmetal minerals: 2.9% per year, Chemical products: 3.25% per year, Manufacturing and processing: 2.5% per year, etc.

Public building: Terminal heating loading (W/m2) decreases to 54.4% of current value in 2010, and 32.8% in 2030. Residential building: Terminal heating loading (W/m2) decreases to 55.5% of current value in 2010, and 36.1% in 2030.

Public building: 68.4% in 2010 and 51.8% in 2030. Residential building: 70.6% in 2010 and 54.0% in 2030.

Energy efficiency of light buses and cars increase 87% before 2030

Energy efficiency increase 46% before 2030

Scenario Definition „PCP Policies Improvement of rural cooking condition

Indicators

Indicators in BAU

Biomass consumption in cooking decreases to 50% in 2010 and 19% in 2030

53% in 2010 and 26% in 2030.

Urban: 24% of heating boilers uses natural gas in 2010, and 50% in 2030. Rural: Biomass stoves contribute 50% in 2010, and 10% in 2030

Urban: 20% in 2010, and 45% in 2030. Rural: Biomass stoves contribute 45% in 2010, and 5% in 2030

Implement EURO IV in 2010, and EURO V in 2015

EURO IV in 2012, and EURO V in 2018

Two control zone policy

New power plants install FGD (flue gas desulfurization), and eliminate power plants over 30 years old. New power plants install SCR (Selective Catalytic Reduction) from 2012.

New power plants install FGD. New power plants install SCR (Selective Catalytic Reduction) from 2015.

PM control in industry

Bag house installed with 20% of CFB (Circulating Fluidized Bed) boiler in 2010, and 75% in 2030; with 2% of grate furnace in 2010, and 30% in 2030

With CFB: 15% in 2010 and 60% in 2030; with grate furnace: 1% in 2010, and 20% in 2030

Switching heating boilers and stoves Vehicle emission standard

Scenario Design

„

Scenario Definition ¾ Business As Usual (BAU) ¾ Scenario I (BAU+EP) ¾ Scenario II (BAU+EP+PCP)

Outlines „ „ „ „ „ „

Background Working framework Scenario development Modeling methodology Modeling results Discussion

Development of EI „

Tech/Fuel options in energy consumption ¾ Power：

Pulverized coal

¾ Industry:

Coal/kiln, Coal/boiler (CFB, auto grate furnace, handfed grate furnace), coke, heavy oil, diesel

¾ Transportation: LDGV, LDDV, LDGT1, LDGT2, LDDT, HDGV, HDDV, MC ¾ Residential: Coal (Raw coal, briquette), biomass, LPG, NG „

Options of abatement tech ¾ Power: FGD, SCR ¾ Industry boiler: Fabric, ESP, wet scrubber, cyclone, none ¾ Transportation: EURO-I ~ EURO-V

9 8 7 6

700000 600000

BAU Scenar i o I Scenar i o I I

5 4 3 2

700000

Coal - f i r ed pl ant s

500000

FGD i n BAU and Scenar i o I

600000

FGD i n Scenar i o I I

500000

Capaci t y( MW)

10

Capaci t y( MW)

El ect r i ci t y by coal - f i r ed pl ant s ( Bi l l i on gi gaj oul e)

Example of Emission projection: SO2 and NOx from electricity

400000 300000 200000 100000

1 0 2005

2010

2020

2030

2000 2001

2005

SO2 By Power

13000

2010

12000

Gg 12000

11000

11000

10000

10000

9000 8000 7000 6000

SCR i n Scenar i o I I

400000 300000 200000

2020

2030

0

2000 2001

2005

2010

Nox by Power BAU S- I S- I I

9000

BAU S- I S- I I

8000 7000 6000

5000 4000 2000

SCR i n BAU and Scenar i o I

100000

0

2000 2001

Coal - f i r ed pl ant s

5000 2005

2010

2015

2020

2025

2030

2000

2005

2010

2015

2020

2025

2030

2020

2030

Emission process strategy

Location of Large Point sources

Regional Inventory

Urban & rural population

Top-down method Province-based

Road network

Gridding based on GIS

… ...

Spatial distribution

High resolution gridded inventory

Chemical speciation

Temporal allocation

Air quality model

Gridding

Landcover

Urban Pop. by Regions Rural Pop. by Regions

Road Network

Landcover by Regions

Shiplanes

Road by Regions

Population

Region bndries

LPSs

Shiplane by Types LPSs by Regions

Emis. Estimate Region

Sector

Chem. Species

Area Emis. by Fine Grid Road Emis. by Fine Grid Ship Emis. by Fine Grid LPSs Emis.

Volcano

Volcano by Locations

Vocanic Emis.

GIS Info.

Alloc. Factor

Emissions Figure from J.Woo

Chemical speciation VOC Source profile: ¾ Based on EPA-SPECIATE & Trace-P database ¾ Searching for local data

PM ¾

Local source profile except BC & OC

¾

BC & OC: Trace-P inventory

¾

BC & OC emission from combustion: an

ongoing project in our group

Temporal allocation

¾ Local temporal profile were used whenever possible „

Routinization of emission inventory process ¾ Finished the routine: from gridding results to model acceptable ASCII file

¾ Change from manual work in Excel to module & standard program

CMAQ Model ready emissions Showing differences in release height (SO2) Upper-layer sources (power plants)

Middle-layer sources (industry)

Lower-layer sources (residential and transportation

Models-3/CMAQ System Framework Meteorology Processor Emission Processor

Gridding Process

Emission Inventory

Air Quality Model

PAVE

Domain: 36km (164*97)

Outlines „ „ „ „ „ „

Background Working framework Scenario development Modeling methodology Modeling results Discussion

Sectoral Distribution of 2001 EI 100% others

80%

Agriculture Biomass Burning

60%

Power Generation 40%

Transport Domestic

20%

Industry

N H 3

H4 C

M VO C

O

N

C

O C

BC

ox N

SO 2

0%

The distribution characteristics varies a lot among the different type of primary air pollutants.

Spatial Distribution of 2001 EI

SO2

NOx

VOC

CO 0.3°×0.3°，t/(km2•year)

Emission Trend: SO2 By Power

13000 12000 11000

SO2 Emi ssi

10000 9000 8000 7000 6000 5000 4000 2000

BAU 28000 S- I S- I I26000

24000 2005

2010

22000

2015

20000

By I ndust r y

13000

18000

12000

BAU16000 S- I S- I I14000

11000 10000

2000

9000

2020

2025

2030

2005

2010

2015

By Domest i c

BAU S- I S- I I 2005

2010

BAU S- I S- I I 2005

2020

2025

2015

2020

2025

2030

15% less electricity demand (2030) leads to less emission

2010

8000 7000 2000

3500 3300 3100 on2900 2700 2500 2300 2100 1900 1700 1500 2000

2030

30% more FGD 2015 2020 leads 2025 installation to less emission

2030

Ignore SO2 Emission from transport sector

Emission Trend: NOx by Power

Gg 12000

BAU S- I S- I I

11000 10000 9000 8000

1700 1600

NOx Emi ssi 1500 on

Gg 24000 22000

7000 6000

20000

5000 2000

2005

2010 2015 18000

2020

BAU S- I S- I I 2025

1400 1300 1200

2030

by Domest i c

Gg

BAU S- I S- I I

1100 1000 900 2000

2005

2010

2015

2020

2025

2030

2025

2030

16000 7000

by I ndust r y

14000

Gg

BAU S- I S- I I

6500 6000 5500

2500

12000 10000

2000

2000

2005

2010

5000

by Tr anspor t

Gg

2015

1500

BAU S- I S- I I

2020

2025

2030

4500 4000 2000

2005

2010

2015

2020

2025

2030

1000

2000

2005

2010

2015

2020

Emission Trend: CO and VOC CO Emi ssi on

Gg 170000 160000 150000 140000 130000 120000 110000 100000 2000

BAU S- I S- I I

2005

2010

2015

2020

2025

2030

VOC Emi ssi on

Gg 20000 19000 18000 17000 16000 15000 14000 13000 12000 11000 10000

BAU S- I S- I I 2000

2005

2010

2015

2020

2025

2030

Emission Trend: BC and OC BC Emi ssi on

Gg 1200 1000 800 600

BAU S- I S- I I

400 200 0 2000

2005

2010

2015

2020

2025

2030

OC Emi ssi on

Gg 4000 3500 3000 2500 2000 1500

BAU S- I S- I I

1000 500 0

2000

2005

2010

2015

2020

2025

2030

Sectoral distribution 30000

25000

5000 0

SO2

2030 BAU S1 S2

2030 BAU S1 S2

2020 BAU S1 S2

2010 BAU S1 S2

2005 BAU S1 S2

0

2001

5000

10000

2020 BAU S1 S2

10000

15000

2010 BAU S1 S2

15000

Bi omass Bur ni ng Tr anspor t Domest i c Fossi l Fuel s Domest i c Bi of uel s I ndust r y Power

2005 BAU S1 S2

kt

20000

20000

2001

Ot her s Bi omass Bur ni ng Tr anspor t Domest i c Fossi l Fuel s Domest i c Bi of uel s I ndust r y Power

kt

25000

NOx

2030 BAU S1 S2

2020 BAU S1 S2

2010 BAU S1 S2

CO

Bi omass Bur ni ng Tr anspor t Domest i c I ndust r y Power

2005 BAU S1 S2

2030 BAU S1 S2

2020 BAU S1 S2

2010 BAU S1 S2

40000 20000 0

2005 BAU S1 S2

120000 100000 80000 60000

20000 18000 16000 14000 12000 10000 8000 6000 4000 2000 0

2001

Ot her s Bi omass Bur ni ng Tr anspor t Domest i c Fossi l Fuel s Domest i c Bi of uel s I ndust r y Power 2001

kt

180000 160000 140000

kt

Largest emission reduction

VOC

Sectoral distribution 4000

1200

3500

1000

400 200

Bi omass Bur ni ng Tr anspor t Domest i c I ndust r y Power Gener at i on

2500 2000 1500 1000

BC

„

2030 BAU S1 S2

2020 BAU S1 S2

2010 BAU S1 S2

2005 BAU S1 S2

2001

0

2030 BAU S1 S2

2020 BAU S1 S2

2010 BAU S1 S2

500

2005 BAU S1 S2

0

2001

kt

600

3000 kt

Bi omass Bur ni ng Tr anspor t Domest i c I ndust r y Power

800

OC

Largest emission reduction

Most effective abatement technology/policy: ¾ SO2: FGD ¾ NOx: SCR ¾ CO, BC & OC: advanced stoves or gasification (especially, in rural)

Emission Summary „

Emissions in 2001 SO2 (kt)

NOx (kt)

BC (kt)

OC (kt)

CO (kt)

NMVOC (kt)

20385

11347

1049

3385

155566

17441

„

Emissions Trends: SO2

NOx

BC

OC

CO

NMVOC

BAU/2001

1.13

1.23

0.94

0.85

1.01

1.08

S 1/2001

1.06

1.18

0.82

0.76

0.96

1.04

S 2/2001

1.03

1.15

0.78

0.73

0.93

0.98

BAU/2001

1.38

1.54

0.66

0.47

0.83

0.97

S 1/2001

1.15

1.33

0.49

0.38

0.75

0.87

S 2/2001

0.79

1.11

0.40

0.34

0.72

0.72

2010

2030

CMAQ result: SO2 2001

2010BAU

2010S1

2010S2

2030BAU

2030S1

2030S2

CMAQ result: NO2 2001

2010BAU

2010S1

2010S2

2030BAU

2030S1

2030S2

CMAQ result PM2.5 2001

2010BAU

2010S1

2010S2

2030BAU

2030S1

2030S2

Outlines „ „ „ „ „ „

Background Working framework Scenario development Modeling methodology Modeling results Discussion

Discussion

¾ Technology based EI ¾ Evaluation of enforcement ¾ Regional differences of technologies

Acknowledgement „

U.S.EPA for founding support (ICAP, IES)

„

ERI’s 2020 China Sustainable Energy Scenario program for energy data

„

U.S.DOE’s Dr. David Streets for TRACE-P Emission Inventory

„

U.S.EPA’s Dr. Carey Jang for technical support on Medels-3/CMAQ

Thanks

Approaches to Emission Factors

„

Chemical balance

¾ Raw gas factors (SO2, Primary PM) „

Technical reports and papers

¾ Fuel characteristics, Operating practice, Control equipments „

Field measurement

¾ Final emission factors, Removal efficiency „

International EF database

¾ When lack of local measurement data (BC/OC, VOC)

Technical Split for Coal Combustion Devices

Good Efficiency Lime

Moderate Efficiency Poor Efficiency