Upload
others
View
6
Download
0
Embed Size (px)
Citation preview
Using mobile measurements to update on-road transportation emissions inventories
Shaojun Zhang, Assistant ProfessorSchool of Environment, Tsinghua University
2019 International Emissions Inventory ConferenceDallas, TexasJul 31, 2019
2019 International Emissions Inventory Conference
Shaojun Zhang, Hui Wang, Xinyu Liang and Ye Wu School of Environment, Tsinghua University
K. Max Zhang Sibley School of Mechanical and Aerospace Engineering, Cornell University
Parikshit DeshmukhJacobs Technology Inc.
Richard Baldauf, James Faircloth, and Richard Snow U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
We acknowledge the support from Ministry of Science and Technology (MOST) of China’s international collaborative project with U.S. EPA.
Authors and Acknowledgements
Outline
BackgroundMethods
Results and discussion
Summary
3
Official source apportionment of ambient PM2.5 pollution
Other relevant concerns : high fraction of nitrate aerosol (e.g., over 30% in Beijing), summertime O3 pollution, and exceedance of NO2 concentrations in urban areas
Vehicle emissions and air pollution problems in China
45%
Beijing (81 µg/m3, 2015) Beijing (58 µg/m3, 2017)
29%
Shanghai (53 μg/m3)
Vehicle/Mobile sources Coal combustionDust IndustrialResidential, argicultural and others
4
Data source: Ministry of Ecological Environment of China; Note: Cross-boundary transport not included
HDDV emissions control in China
Controlling air pollutant emissions from HDDVs are one of the prioritized tasks in the new “Clean Air Action Plan (2018-2010)”
Generally, the European regulatory systems have been used in China China III and earlier stages: typically no modern aftertreatments China IV and China V (implemented nationwide since 2015):
SCR as mainstream aftertreatment for NOX control (2.0 g/kWh for China V) Almost no DPF adoption
China VI: recently implemented since Jul 2019 in a few cities/regions PEMS regulations: conformity factors of 1.5 for NOX and 2.0 for PN (China VIb) OBD requirements: capability of remote on-board sensor-based monitoring
5
Technology innovation to enhance in-use surveillance
PEMS1~2 vehs per
day
Dynamometer1~2 vehs per
day
Regulated methods Real-world efficient methodsNon-intrusive
Plume chasing50~80 vehs per day2~5 mins per veh
Remote sensing1000+ vehs per dayShort snap per veh
High accuracySmall sample size
Low accuracyLarge sample size
Applicable to key fleetsNOx sensor required
Remote OBD monitoring
OBM data loggerData collection and
transmission (to cloud)
Accuracy
Fleet coverage6
Joint workshop at Tsinghua University, Beijing
Technical communication at EPA ORD NRMRL in RTP, NC
• Sponsored by Ministry of Science and Technology of China (MOST)-USEPA collaborative agreement on technology innovation for environmental protection
• A focus on developing cost-effective road emission measurement methods • Collaborative project on motor vehicle plume chasing and joint research workshops
Tsinghua-NRMRL collaborative project
7
Outline
BackgroundMethods
Results and discussion
Summary
8
Research framework
• Gridded emission input
• WRF/2D-VBS CMAQ
• Impact from updated EFs on ambient NO2, O3 and PM2.5
• Updating key EFs for HDVs
• Updating fleet-level emission inventories
3. Emission inventory
update2. Plume chasing
1. Method evaluation
4. Air quality simulations
• Concurrent PEMS-chasing tests
• Optimize EF calculation methods
• Large-scale field tests
• EF database based on plume chasing
9
Mobile chasing platform
10
Deep cycle battery set
PM2.5
CO2
NOX /NO
Sample inlet
BC
Camcorder
GPS
PN1
Computer
10
Licor 820
Eco physicsnCLD66
Magee AE33(880 nm)
TSI DustTrak 8530
TSI CPC3007
Pollutant analyzers (Resolution: 1Hz)
Concurrent PEMS-chasing tests
Experimental routes in Beijing.
• 12 heavy-duty diesel trucks recruited for concurrent PEMS-chasing tests to evaluate plume
chasing for NOX emission factors.
11
Local roads (20~40 kph)
Express roads (50~80 kph)
PEMS: SEMTECH-EcoStar
Test fleet: 4 China III (no SCR), 4 China IV (SCR) and 4 China V (SCR)
Testing procedure and EF calculations• Measure road background and exhaust plumes intended for targeted vehicles
12
• NOX EF (g NOx per kg fuel) calculations• Approach A: linear regression without BG subtraction (slope indicates fuel based NOx emissions)
• Approach B: moving average ratio with BG subtraction
Mobile chasing for HDV emissions Nearly 3000 vehicles measured by using mobile chasing since 2017
JJJ region: more than 1200 vehicles
PRD region: more than 1000 vehicles
Sichuan: more than 200 vehicles
Fen-Wei Plain & Central China: ~500 vehicles
13
Vehicle emission models in China
International models (e.g., MOBILE/MOVES; COPERT) have been used to estimate vehicle emissions in China since 1990s.
Local emission measurements are fundamental to develop emission models and reduce uncertainty in policy decisions.
− EMBEV (Tsinghua): initially sponsored by the Beijing EPB, and based on thousands of LDV dynamometer tests and hundreds of LDV/HDV PEMS tests
− China’s National Emission Inventory Guidebook (Tsinghua and VECC): developed for the MEP in 2015, largely based on the EMBEV
− EMBEV-Link: city-scale link-level emission inventory models based on real-world traffic datasets
The Guidebook for road sector will be updated soon, as supported by China MOST
14
EMBEV: Zhang et al., Atmos. Environ., 2014; EMBEV-China: Wu et al., Sci. Total Environ., 2017; EMBEV-Link: Yang et al., Atmos. Chem. Phys., 2019
EMBEV emission factors: heavy-duty diesel vehicles
Major data sources: PEMS tests for pre-China IV HDVs (and China IV urban buses) Mode-specific emission rates available (i.e., similar to MOVES) Few tests for China IV and China V trucks, and their emission factors are largely estimated
based on the reduction trends in European fleets (i.e., 30% ↓ from China III to China IV)
Estimated emission factors (GVW>12 t) under a typical driving cycle (40 km h-1)
1.46
0.55 0.440.22 0.12 0.02
0.0
0.5
1.0
1.5
2.0
Emis
sion
fact
or (g
/km
)
PM
3.31
0.730.42 0.21 0.10 0.10
0
2
4
Emis
sion
fact
or (g
/km
)
HC15.21
8.446.71 7.19
5.04 4.28
0
4
8
12
16
Emis
sion
fact
or (g
/km
)
NOx
Wu et al., 2017, STOTEN 15
Gridded emission input
Monitor siteRoad
NOXemission
16
• Real-world traffic datasets aided allocation• Total fleet emissions by vehicle category• Temporal and spatial allocations, based on specific road lengths (considered the effects from vehicle
volumes, urban/rural, city, speed)
~900 traffic sites (mostly express roads)
Air quality simulation systems
WRFv3.3 configuration− Planetary boundary layer: Mellor-Yamada-Janjic TKE− Land surface layer: Noah− Microphysics: WSM3− Cumulus scheme: G-D− Long/short wave radiation: RRTM
CMAQv5.0.1 enhanced by 2D-volatility basis set (VBS) module
− Improve the simulation performance of SOA− Gas-phase chemistry scheme: SAPRC99− Aerosol module: AERO6
• WRF-CMAQ model system− Three-layer nested simulation (36km-12km-4km)− Focus region: Beijing-Tianjin-Hebei (JJJ)− Duration: January, May, August and November in 2020
17
36 km×36 km
4 km×4 km
Outline
BackgroundMethods
Results and discussion
Summary
18
Comparison of NOX emission factors between mobile chasing and PEMS
Vehicle-specific relative errors within ±20%
Fleet-average errors: -3.8% (local roads by approach A) and -0.3% (freeways by approach B)
Note: The solid markers indicate measurements on local roads, while the hollow markers indicate measurements on freeways. 19
0 20 40 60 80 100 120 1400
20
40
60
80
100
120
140
V9V10V11V12
V5V6 V7V8
V2 V3 V4
Cha
sing
EFs
(g N
OX /
kg-fu
el)
PEMS EFs (g NOX /kg-fuel)
China Ⅲ China Ⅳ China Ⅴ V1
Local roads_Approach AFreeways_Approach B
0 20 40 60 80 100 120 1400
20
40
60
80
100
120
140
China III China IV China V V1 V9
V10V11V12
V5V6V7V8
V2 V3 V4
Cha
sing
EFs
(g N
OX /k
g-fu
el)
PEMS EFs (g NOX /kg-fuel)
Local roads_Approach AFreeways_Approach A
a.
0 20 40 60 80 100 120 1400
20
40
60
80
100
120
140
V9V10V11V12
V5V6 V7V8
V2 V3 V4
Cha
sing
EFs
(g N
OX /
kg-fu
el)
PEMS EFs (g NOX /kg-fuel)
China III China IV China V
b.
V1
Local roads_Approach BFreeways_Approach B
Concurrent PEMS-Chasing comparison(12 HDVs, 245 chasing tests)
0
20
40
60
80
100
120
140
China VChina IV
Local roads Freeways
NO
x EF
s(g/
kg_f
uel)
Veh. IDV1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12
China III
NOX emission factors on local roadsand freeways for PEMS results
Mobile chasing for HDV emissions——NOX
NOX emissions: No significant improvement in China IV diesel trucks (claimed to have adopted SCR) compared with China III trucks; limited improvements for China V diesel trucks.
− Failure to refill urea tanks and tampering with the SCR device are suspected NG HDVs: Significantly higher NOX emissions (lean-burn engines, no NOx aftertreatment) HDDVs registered in various provinces: Beijing not significantly improved compared to other provinces
China III China IV China V NG0
40
80
120
160
200
Diesel
25%-75% 5%-95% Median
NO
X E
Fs
( g/kg_fuel)
Mean
GD BJ LN SD SX HN SHX SC HB TJ Others0
20
40
60
80
100
120
NO
X E
Fs
( g/kg_fuel)
25%-75% 5%-95% Median Mean
Note: GD:Guangdong; BJ:Beijing; LN:Liaoning; SD:Shandong; SX:Shanxi; HN:Henan; ; SHX:Shaanxi; SC:Sichaun; HB:Hebei;TJ: Tianjin
Impact of emission standard and fuel Impact of registration place——HDDV
20
BC emissions: decrease with emission standards; high emitters exist for China III to China V
HDDV in various provinces: related to different vehicle emission control policies
Mobile chasing for HDV emissions——black carbon (BC)
China III China IV China V NG0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0 25%-75% 5%-95% Median
BC
EFs
( g/kg_fuel)
Mean
Diesel
GD BJ LN SD SX HN SHX SC HB TJ Others0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5 25%-75% 5%-95% Median
BC
EFs
( g/kg_fuel)
Mean
Note: GD:Guangdong; BJ:Beijing; LN:Liaoning; SD:Shandong; SX:Shanxi; HN:Henan; ; SHX:Shaanxi; SC:Sichaun; HB:Hebei;TJ: Tianjin
Impact of emission standard and fuel Impact of registration place——HDDV
21
A revisit of NOx and PM2.5 emissions for HDVs
22
0
2
4
6
8
10
2010 2012 2014 2016 2018 2020
Emis
sion
s (m
illion
tons
)
NOX
by EMBEVby Chasing results
Total emission trends
0.0
0.5
1.0
1.5
2.0
Emis
sion
fact
or (g
/km
)
PM2.5
EMBEV updated by Chasing results
0
5
10
15
20
Emis
sion
fact
or (g
/km
)
NOX
Emission factor
0
100
200
300
400
500
2010 2012 2014 2016 2018 2020Emis
sion
s (th
ousa
nd to
ns)
PM2.5
by EMBEVby Chasing results
* Average GVW: 20 t, Driving speed: 40 km/h, BC/PM2.5=0.7
High emitter20%
10%
8%
---- Medium ---- Mean ---- EMBEV
NOX PM2.5
↑ baseline
↓ updated
Air quality impacts: nitrogen dioxide
Monthly average NO2 concentrations are estimated to increase 2-4 μg m-3 (12%-15%) in Beijing-Tianjin-Hebei region when the simulations apply the chasing NOX emission factors.
2020 baseline NO2 concentration
Jan May
Aug Nov
NO2 concentration change
Jan May
Aug Nov
23
Air quality impacts: ozone
Summer: O3 concentrations increased due to increased NOX emissions Winter: O3 concentrations decreased by ~2 ppbv due to strong VOC-limited conditions.
2020 baseline 8-h maximum O3 concentration
Jan May
Aug Nov
O3 concentration change
Jan May
Aug Nov
24
Air quality impacts: PM2.5 and key components
Summer: PM2.5 is predicted to increase 1.2-2.3 μg m-3 (4%-5%) due to increased nitrate. Winter: decreased atmospheric oxidability (i.e., O3) leads to PM2.5 reduction from baseline
2020 baseline PM2.5 concentration
Jan May
Aug Nov
PM2.5 concentration change
-2.0 -1.0 0.0 1.0 2.0
-1.0 1.0 3.0
Jan May
Aug Nov
Jan
Aug
ECPOMSOMNH4NO3SO4PM2.5
ECPOMSOMNH4NO3SO4PM2.5
Changes by aerosol component
25
Summary
On-road evaluations were conducted for plume chasing tests by comparing with PEMS measurements.
Chasing tests indicate that NOx emissions have not been significantly improved with the emission standard for HDVs in China.
BC emissions for HDVs have been reduced with the emission standard; however, high emitters were still found among the fleet.
With updated NOx emission factors, which will not decline before 2020, we estimated increased ambient concentrations for annual NO2 and PM2.5 , and summer O3 in the Jing-Jin-Ji region.
26
Thank you for your attention !Contact. E. [email protected]
Acknowledgements to ongoing and former project sponsors: Ministry of Science and Technology (MOST) of China's International Science and Technology Cooperation Program (2016YFE0106300; i.e., the MOST-EPA collaboration program), and the National Key Research and Development Program of China (2017YFC0212100).
Disclaimer: The views expressed in this presentation are those of the author and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency