ARTICLE IN PRESS
Atmospheric Environment 39 (2005) 6843–6856 www.elsevier.com/locate/atmosenv
On-road remote sensing of diesel vehicle emissions measurement and emission factors estimation in Hong Kong T.L. Chan, Z. Ning Research Centre for Combustion and Pollution Control, Department of Mechanical Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong Received 22 September 2004; received in revised form 7 July 2005; accepted 26 July 2005
Abstract In the present study, the real world on-road diesel vehicle emissions of carbon monoxide (CO), hydrocarbons (HC) and nitric oxide (NO) were investigated at nine sites in Hong Kong. A regression analysis approach based on the measured vehicle emission data was used to estimate the on-road diesel vehicle emission factors of CO, HC and NO with respect to the effects of instantaneous vehicle speed and acceleration/deceleration profiles for local urban driving patterns. The results show that the diesel vehicle model years, engine sizes, vehicle types and driving patterns have a strong correlation with their emission factors. A comparison was made between the average diesel and petrol vehicle emissions factors in Hong Kong. The deviation of the average emission factors of aggregate diesel vehicles reflects the variability of local road condition, vehicle traffic fleet and volume, driving pattern, fuel composition and ambient condition etc. Finally, a unique database of the correlation of diesel vehicle emission factors (i.e., g km1 and g l1) on different model years and vehicle types for urban driving patterns in Hong Kong was established. r 2005 Elsevier Ltd. All rights reserved. Keywords: On-road vehicle emissions; Remote sensing measurement technique; Emission factors; Regression analysis
1. Introduction Motor vehicle emissions are the major source of air pollution problem in most urban cities (Mayer, 1999) including Hong Kong (HKEPD, 2005). For the protection of atmospheric environment, many measurement, and control and management methods of vehicular exhaust emissions (VEEs) in urban environments have been carried out to provide criteria for determining the emission reduction and evaluating the effectiveness of vehicular emission control strategies and regulations in order to meet the clean/better air quality goals. A Corresponding author. Tel.: +852 2766 6656; fax: +852 2365 4703. E-mail address:
[email protected] (T.L. Chan).
summary of diesel vehicle emission standard is listed in Table 1 (Tsang and Ha, 2002). On-road vehicle exhaust emissions survey using remote sensing technology offers a quick and effective method of monitoring exhaust emissions from in-use petrol vehicles under the normal driving operation. The application of remote sensing vehicle exhaust emissions testing system has been used in the polluted areas of the United States, Canada, Mexico, Australia, Taiwan, Hong Kong and many other parts of the world to achieve different tasks such as the traffic fleet and volume characterisation for the low and high emitter profiling, dirty screening programme, clean screening credit utility programme and emission inventories/factors development (Cross, 2000; Walsh, 2001; Pokharel et al., 2002; Schifter et al., 2003, 2005; Chan et al., 2004).
1352-2310/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2005.07.048
ARTICLE IN PRESS 6844
T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
EFi
Nomenclature a
instantaneous acceleration/deceleration rate of diesel vehicle in km h1 s1 b, c, d, e, fy constants of regression equations as defined in Eqs. (1)–(3) c1 ; c2 ; c3 ; . . . c004 ; c005 constants of regression equations as defined in Eqs. (1)–(3) Dfuel the density of the fuel (i.e., diesel) in kg l1 Ei,j mass emission concentration of individual emission species, i (i.e., CO, HC or NO) and vehicle type, j (i.e., diesel) in g l1 fuel burned as defined in Eqs. (4)–(6)
Gj Mfuel Q Q0 Q00 V
emission factor of individual emission species, i (i.e., CO, HC or NO) in g km1 as defined in Eq. (7) fuel consumption in l 100 km1 of vehicle type, j (i.e., diesel) as defined in Eq. (8) the molar mass of the fuel (i.e., diesel) in kg mol1 ratio of CO to CO2 in volume concentration basis as defined in Eq. (1) ratio of HC to CO2 in volume concentration basis as defined in Eq. (2) ratio of NO to CO2 in volume concentration basis as defined in Eq. (3) instantaneous velocity of diesel vehicle in km h1
Table 1 Summary of Hong Kong diesel vehicle exhaust emission standards (Tsang and Ha, 2002) Implementation date/ vehicle class
1 April 1995
1 April 1997
1 January 2001
1 October 2001
Free acceleration smoke (light absorption coefficient K, m1) Private carb
K 1.20
K 1.00
K 1.00
K 0.8a
Euro I or US 88 or Japan 94 Euro I or US 88 or Japan 94
Euro I or US 88 or Japan 94 Euro I or US 88 or Japan 94
US California LEV
US California LEV
Euro I or US 91
Euro II or US 94
Banned diesel; LPG/ petrol only; EU Phase 2 or US 96 or Japan 78 Euro II or US 98
Banned diesel; LPG/ petrol only; EU Phase 2 or US 96 or Japan 78 Euro III or US 98
Taxic
Goods vehicle and buses over 3.5 tonne a
Smoke determined by ELR test for over 3.5 tonne diesels. Private Car—US California 94 as of April 1998. c Taxi—Euro II as of 1 July 1999. b
In order to estimate the emission contribution from on-road vehicles to urban air pollution, many vehicle emission factor models based on vehicle exhaust emission measurements have been developed and with features updated continuously such as EMFAC2002 (USCARB, 2004), MOBILE6 (USEPA, 2004), etc. Although these popular emission factor/rate models have been used widely, they could not reflect reasonably the on-road vehicle emissions of modal traffic events, real-time, site-specific emissions especially for the local vehicle exhaust emissions impact assessment due to the variations of local road condition, vehicle traffic fleet and volume, driving pattern, fuel composition and ambient condition, etc. In addition, they cannot be used effectively to evaluate the traffic control and management strategies for targeting the reduction of vehicle
emissions on particular vehicle group, and neither for supporting human exposure studies in roadway environments. In recent years, many research efforts have been put in developing the new methods for the emission factor models in order to provide a better understanding of the characteristics of on-road vehicle emissions under the real-world driving patterns. The emission factors for different types of vehicles coupled with the average vehicle speed were determined under the laboratory conditions by Jost et al. (1994) and Joumard et al. (1995a, b, 2003). Sjodin et al. (1998) estimated the emission factors of many exhaust toxic gases emitted from on-road motor vehicles inside a traffic tunnel using Fourier transform infrared spectroscopy and conventional analysers. The on-board instrumentation was also used to determine the characteristics of vehicle emission
ARTICLE IN PRESS T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
factors by Kelly and Groblicki (1993), Vanruymbeke et al. (1993) and Zhao et al. (1999). Although some of these developed emission factor models have been evaluated by Sturm et al. (1997) and Zachariadis and Samaras (1997), they still have their own inherent limitation. Recently, Saija and Romano (2002) have used the Computer Programme to calculate Emissions from Road Transport (COPERT) III model to estimate the atmospheric emissions of different pollutants (i.e., NOx, CO and CO2, etc.) and the road transport emissions in the urban areas of Italy. The detailed descriptions of COPERT III model can be found in Ntziachristos and Samaras (1999) and Kouridis et al. (2000). Joumard et al. (2003) showed the influence of the average cycle speed, the load and the vehicle category on emissions from the measured private light duty goods vehicles (LDGVs). They concluded that the measured vehicle exhaust emissions from LDGVs are quite different from passenger car emissions, and from emissions calculated using the European Meet/COPERT III model. Kean et al. (2003) studied the emissions from on-road vehicles. They found that the emissions of CO and NOx are functions of both vehicle speed and specific power; neither parameter alone captures all the relevant effects on emissions. Schifter et al. (2003) used the remote sensing technology to measure the emissions of CO, CO2, HC and NO from the on-road vehicles in the metropolitan area of Mexico city. Chan et al. (2004) have recently investigated the real-world petrol vehicle emissions of CO, HC and NO in Hong Kong using the on-road remote sensing vehicle exhaust emissions testing system, and have estimated the on-road petrol vehicle emission factors. Their results have showed that the petrol vehicle model years, engine sizes and driving patterns have a strong correlation with the on-road petrol vehicle emission factors. The present study intends to establish a unique database of the correlation of diesel vehicle emission factors for urban driving patterns in Hong Kong based on the vehicle types and model years.
2. Experimental procedures A typical on-road remote sensing vehicle exhaust emissions testing system setup at the Cross Harbour Tunnel (Kowloon entrance) site of Hong Kong is shown in Fig. 1. The concentrations of HC, CO, CO2 and NO along the trajectory of vehicle exhaust plume in real time were measured using a remote sensing vehicle exhaust emissions testing system (ESP AccuScan RSD3000, USA). In this system, the source detector module and the vertical transfer mirror units are positioned on the opposite sides of a single traffic lane under a bi-static arrangement. As the vehicles drive by, the infrared and ultraviolet beams pass through the exhaust plume from
6845
Fig. 1. Typical on-road remote sensing vehicle emissions testing system set-up at the Cross Harbour Tunnel (Kowloon entrance) site in Hong Kong.
the vehicle tailpipe, which absorbs some of the light. The molar ratios of HC, CO and NO to CO2 are then measured simultaneously from each vehicle. A device for measuring the speed and acceleration/deceleration of vehicles driving past the remote sensor, which includes an emitter bar and a detector bar, is also used. The vehicle license plate can also be captured by the digital colour video camera system and the vehicle information can be obtained at a later stage. A large amount of vehicle emission concentrations, vehicle speed and acceleration information, and the impact of highemitting vehicles can be obtained and identified within a short period of time from a single traffic lane. Based on the measurement data, the average emission factors of CO and HC were calculated as described by Stephens et al. (1994) and Pokharel et al. (2001). The criteria of site selection is based on the road type, number of lanes, freeway entrance, traffic volume and vehicle mix, traffic speed and acceleration mode, road grade, traffic signal location, geographical distribution and representation. The field measurement should be initiated right after the completion of system calibrations at the site. Owing to the change of local environmental conditions (i.e., surrounding vehicle emission concentrations, wind speed and direction, etc.) from time to time and site to site, and the unavoidable road vibration, the system calibration should be carried out at least every 2 h in order to obtain the best results. In the present study, the real world on-road diesel vehicle emissions were measured at nine sites in Hong Kong during late 2001, namely Cotton Tree Drive, Kwun Tong Highway Bypass, Man Kam To, Sha Tau Kok Road, Tai Po Road, Yuen Long Highway, Aberdeen Tunnel, Cross Harbour Tunnel and Lion Rock Tunnel. The vehicle speed, traffic volume and road condition varied from around 10 to 75 km h1, 31 to 553 vehicle h1 and 1/15 to 1/10 road gradient for road pavement surface in tar, respectively at these nine local sites (Chan et al., 2002). A total of 9057 diesel
ARTICLE IN PRESS T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
6846
Table 2 Registered vehicles in Hong Kong (HKTD, 2005) Type of vehicle
Fuel used
Vehicle number
Private car Motor-cycles Taxi Public light buses Private light buses Public buses Private buses Goods vehicle Government vehicles Others
Petrol Petrol Mainly LPG Diesel Diesel Diesel Diesel Diesel Diesel/petrol
384,900 43,950 18,138 4350 1935 13,251 494 120,948 6591
Diesel
Total
975
Zachariadis and Samaras, 1997; USCARB, 2004; USEPA, 2004). However, the measured vehicle emissions are highly dependent on the instantaneous vehicle speed profiles and the atmospheric disturbance on the vehicle exhaust plume under its local road and ambient conditions. Hence, an emission factor model based on the measured emission data from the on-road remote sensing vehicle exhaust emissions testing system and a correlation of on-road vehicle emission rates of CO and HC and the instantaneous vehicle speed profiles were developed (Yu, 1998; Pokharel et al., 2001). 3.1. Procedures of calculating the aggregate diesel vehicle emission factors
595,532
vehicle emission data were captured at these nine sites. Among these captured data, 6321 vehicle emission data were considered to be valid in terms of its velocity, acceleration/deceleration and emission concentrations profiles, and identifiable vehicle license plate number. Hence, the overall data capture rate for diesel vehicles at these nine sites is around 70%, which is dependent on a number of factors. The average diesel vehicle speed and acceleration of the valid measured data were 52.61 km h1 and 1.25 km h1 s1, respectively. These factors include site selection (vehicle specific power as it passes the system), meteorological conditions (humidity, wind, dust etc.), traffic flow, the system alignment due to the traffic lane vibration and levelling as well as the field calibration. In general, the data capture rate is based on the ratio of the number of valid measurement data (i.e., measurement with identifiable vehicle license plate number, and valid emission data for all three ratios of CO/CO2, HC/CO2 and NO/CO2) to the number of measurement data taken (i.e., measurement with identifiable vehicle license plate number) (Chan et al., 2002). About 25% of the total vehicle population is the diesel vehicles in Hong Kong, as shown in Table 2 (HKTD, 2005). Over 10 million passenger journeys are made on the public transport system every day in Hong Kong and there are about 274 licensed vehicles of total vehicle fleet for every kilometre of road.
3. Results and discussion Many vehicle emission factor models based on the emission measurements have been used widely and with features updated continuously in order to estimate the emissions contribution from motor vehicles to our urban air pollution (Jost et al., 1994; Joumard et al., 1995b;
Based on the measured emission data from on-road diesel vehicles at the nine local sites in Hong Kong using the remote sensing vehicle exhaust emissions testing system, the emission factors of CO, HC and NO were calculated under the real-world vehicle driving conditions in respect of instantaneous vehicle speed and acceleration/deceleration profiles. The reported CO, HC and NO concentrations were then converted into the basic measured parameters, namely the ratios of the individual volume emission concentration with regard to the CO2 in volume concentration basis (i.e., Q ¼ CO/ CO2, Q0 ¼ HC/CO2 and Q00 ¼ NO/CO2). The ratio of volume emission concentration in Q, Q0 and Q00 along the vehicle exhaust plume would not change during the remote sensing measurement (Bishop and Stedman, 1996). Two mathematical forms of the regression analysis were established to correlate the on-road diesel vehicle emissions with the driving conditions on the road lane based on the calculated regression coefficient, R2. They were similar to the prior research work (Yu, 1998) which was incorporated into the volume emission concentrations as follows: Q ¼ c1 þ c2 V þ c3 V 2 þ c4 a þ c5 a2 ;
(1)
Q0 ¼ c01 þ c02 V þ c03 V 2 þ c04 a þ c05 a2 ,
(2)
ln Q00 ¼ c001 þ c002 V þ c003 V 2 þ c004 a þ c005 a2 . c1 ; c2 ; c3 ; . . . c004 ;
(3) c005
The constant coefficients of are determined by the regression analysis. The conversion equations in fuel based emission factors, Ei (Holmen and Niemeier, 1998; Singer and Harley, 2000; Pokharel et al., 2001, 2002; Chan et al., 2004; Schifter et al., 2003, 2005) can be expressed as follows: E CO ðg l1 Þ ¼
28 %CO=%CO2 %CO=%CO2 þ 1 þ ½3 ð%HC=%CO2 Þ=0:493
1 Dfuel , ð4Þ M fuel
ARTICLE IN PRESS T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
44 ð%HC=%CO2 Þ=0:493 %CO=%CO2 þ 1 þ ½3 ð%HC=%CO2 Þ=0:493
ð5Þ
30 %NO=%CO2 %CO=%CO2 þ 1 þ ½3 ð%HC=%CO2 Þ=0:493
1 Dfuel , ð6Þ M fuel
where Mfuel is the molar mass of fuel; Dfuel is the density of fuel. For the diesel fuel, the carbon to hydrogen ratio is typically 1.85 and Mfuel is 0.01385 kg mol1, while the fuel density is 0.85 kg l1 (API, 2001). In the present study, the value of 0.493 used in Eqs. (4)–(6) is the conversion unit based on the total carbon mass basis factor which is assumed to be the equivalency factor of propane for the measurements of HC gas detector (Holmen and Niemeier, 1998; Singer et al., 1998). The individual gaseous emission factor, EFi, can be calculated in terms of the following equation: EFi ðg km1 Þ ¼ E i;j G j =100.
(7)
In the urban traffic condition, the fuel consumption depends on the instantaneous vehicle speed. A curvefitted formula for the fuel consumption of diesel vehicles in respect of the instantaneous vehicle speed has already been established according to the work of Tong et al. (2000) in Hong Kong. They have studied the fuel consumption of on-road diesel vehicles in Hong Kong, and have provided the useful equation on the relationship between the instantaneous vehicle speed and the fuel consumption: Gj ðl:100 km1 Þ ¼ 319:95 V 1:1131 .
(8)
EFCO (gkm-1)
E NO ðg l1 Þ ¼
Regression Eqn. of EFCO EFCO data
1.2 1.0 0.8 0.6 0.4 0.2 0.0 0
20
40
60
80
-1
Vehicle Speed (kmh )
(a)
0.10 Regression Eqn. of EFHC EFHC data
0.08 EFHC (gkm-1)
1 Dfuel , M fuel
1.4
0.06 0.04 0.02 0.00 0
20
40
60
80
Vehicle Speed (kmh-1)
(b)
1.2 Regression Eqn. of EFNO EFNO data
1.0 EFNO (gkm-1)
E HC ðg l1 Þ ¼
6847
3.2. Relationship between the emission factors of aggregate diesel vehicles and driving patterns
0.8 0.6 0.4 0.2 0.0
The constant coefficients in Eqs. (1)–(3) can be obtained by the regression analysis of the total diesel vehicle emission data:
(c)
Q ¼ 3:7137 103 5:0582 106 V þ 2:0647 108 V 2
Fig. 2. The EFCO, EFHC and EFNO of the aggregate diesel vehicles as a function of instantaneous vehicle speed profiles in Hong Kong.
4:7862 105 a 5:4472 106 a2 ,
ð9Þ
Q0 ¼ 2:9705 104 þ 2:4140 108 V 9:2766 1010 V 2 4:8714 106 a 1:5633 106 a2 ,
ð10Þ
ln Q00 ¼ 5:9549 þ 1:2520 102 V 8:7038 105 V 2
9:9292 103 a 2:3990 103 a2 .
ð11Þ
The effect of driving behavior on the average emission factors of EFCO, EFHC and EFNO for the aggregate
0
20
40
60
80
Vehicle Speed (kmh-1)
diesel vehicles as a function of instantaneous speed profiles in Hong Kong is shown in Fig. 2. A strong correlation between the average emission factors of the aggregate diesel vehicles and vehicle speed profiles can be found. Figs. 2a and b show that the emission factors of EFCO and EFHC for the aggregate diesel vehicles decrease rapidly with increasing the instantaneous vehicle speed till 60 km h1. Beyond the vehicle speed of 60 km h1, the emission factors of EFCO and EFHC
ARTICLE IN PRESS 6848
T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
respectively, as shown in Fig. 3 and the average fuelbased emissions factor of ECO, EHC and ENO vary from 12% to 44%, 29% to 62%, and 20% to 40%, respectively, as shown in Fig. 4. The variations of gaseous emissions reduction are mainly because some of the diesel vehicles in a particular model year are high emitters due to the multitude of vehicular engine problems and malfunctions, poor maintenance on vehicular engines and the cross-border diesel vehicles which use rather poor quality of high sulphur diesel fuel from the Chinese Mainland. If the 10% of the on-road vehicles (so called gross polluters) which may contribute to half of the vehicle emitted pollutants for each emission pollution species as described by Stedman (2002) are removed, the overall reduction trend of average EFCO, EFHC and EFNO can be improved from 10% to 44%, 19% to 54%, and 20% to 43%, respectively, as shown in Fig. 5 and the average ECO, EHC and ENO can be improved from 15% to 44%, 25% to 58%, and 21% to 43%, respectively, as shown in Fig. 6, if the baseline of diesel vehicle model year of 1990 and the last 5 years from 1997 to 2001 are compared. It has also led to the introduction of more stringent diesel vehicle emission control methods and incentive programmes by the government of HKSAR, China, since late 2000 (HKEPD, 2005). Furthermore, the regression equations of average EFCO, EFHC or EFNO and ECO, EHC or ENO of the aggregate diesel vehicles in terms of instantaneous vehicle speed profiles ranging from 10 to 75 km h1 for different diesel vehicle types are listed in Tables 7 and 8, respectively.
decrease slightly. Fig. 2c also shows that the emission factors of EFNO for the aggregate diesel vehicles decrease rapidly with increasing the instantaneous vehicle speed up to 55 km h1. Beyond the vehicle speed of 55 km h1, the emission factors of EFNO decrease slightly. However, there is no strong correlation between the average EFCO, EFHC or EFNO and ECO, EHC or ENO of the aggregate diesel vehicles and instantaneous acceleration/deceleration profiles can be determined in the present study. The regression of average EFCO, EFHC or EFNO and ECO, EHC or ENO of the aggregate diesel vehicles in terms of instantaneous vehicle speed profiles for 13 km h1 s1pap11 km h1 s1 is shown in Tables 3 and 4, respectively. 3.3. Relationship between the average emission factors of the aggregate diesel vehicles and model years and diesel vehicle types The regression equations of average EFCO, EFHC or EFNO and ECO, EHC or ENO of the aggregate diesel vehicles in terms of instantaneous vehicle speed profiles ranging from 10 to 75 km h1 for different model years are listed in Tables 5 and 6, respectively. From 1995 to 2001, four diesel vehicle emission standards for diesel vehicles were implemented in Hong Kong, as shown in Table 2 (Tsang and Ha, 2002). If the baseline of diesel vehicle model year of 1990 and the last 5 years from 1997 to 2001 are compared, the overall reduction trend of average emissions factors of EFCO, EFHC and EFNO vary from 2% to 13%, 17% to 52%, and 8% to 39%,
Table 3 The regression equation of average EFCO, EFHC or EFNO of the aggregate diesel vehicles in terms of instantaneous speed for 13 km h1 s1pap11 km h1 s1 in Hong Kong Emission species Regression equation
R2
Constant coefficients b
CO HC NO
c
d
e
f
0.9860 4.938 102 1.145 102 3.706 102 3.206 102 8.768 102 EFCO ¼ b+cV+d/ln V+e/V0.5+f/V2 EFHC ¼ b+cV+dV ln V+e ln V+f ln V/V 0.8338 3.830 5.954 102 9.095 103 1.147 7.236 EFNO ¼ b+cV+dV1.5+eV2 ln V+fV0.5 0.9842 7.302 6.511 101 5.063 102 2.833 104 3.462
Table 4 The regression equation of average ECO, EHC or ENO of the aggregate diesel vehicles in terms of instantaneous speed profiles for 13 km h1 s1pap11 km h1 s1 in Hong Kong Emission species
CO HC NO
Regression equation
ECO ¼ b+c ln V+d/ln V+e/V EHC ¼ b+cV+dV ln V+eV1.5 ENO ¼ b+cV ln V
R2
0.5183 0.7837 0.7851
Constant coefficients b
c
d
e
1.479 102 6.264 101 5.309
1.352 102 9.078 102 8.298 103
3.983 102 4.134 102 —
6.306 102 1.025 102 —
Table 5 The regression equation of average EFCO, EFHC and EFNO of the aggregate diesel vehicles as a function of model year for 10 km h1 to 75 km h1 in Hong Kong Model year
Regression equation
R2
Samples
Constant coefficients b
c
d
e
f
146 259 505 485 407 397 737 1184 746 570 441 236
1.156 101 1.401 101 6.138 101 8.587 102 1.386 5.914 101 7.864 7.093 1.003 102 3.228 101 4.908 101 1.318
4.466 102 2.037 102 3.386 1.917 101 5.456 102 4.982 101 1.454 101 1.042 2.529 102 3.990 101 7.148 102 5.622 102
4.367 101 3.243 9.623 9.156 102 8.436 104 3.241 1.361 103 1.588 101 1.083 101 6.087 102 1.824 102 1.105 103
8.825 101 1.135 102 8.846 2.250 103 4.525 1.928 102 5.053 2.078 101 2.605 102 8.932 9.365 101 6.355 104
— 6.574 102 3.723 101 3.236 103 7.574 101 3.463 102 3.152 1.391 102 3.592 102 7.058 — 1.604 105
(b) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
EFHC ¼ b+c ln V+d/ln V+e ln V/V2 EFHC ¼ b+cV+d ln V/V+e/V+f/V2 EFHC ¼ b+cV+dV2+eV2 ln V+fV3 EFHC ¼ b+cV+d ln V/V+e/V1.5+f ln V/V2 EFHC ¼ b+c/ln V+d ln V/V2+e/V2 EFHC ¼ b+cV+d/ln V+e/V+f/V2 EFHC ¼ b+cV+dV3+e ln V EFHC ¼ b+cV+dV ln V+e ln V+f ln V/V EFHC ¼ b+cV+d ln V+e/ln V+f/V EFHC ¼ b+cV3+d/V+e/V2 EFHC ¼ b+cV+dV ln V+eV0.5 ln V+fV/ln V EFHC ¼ b+cV+dV2+e/ln V+f ln V/V2
0.7281 0.9415 0.8165 0.9162 0.7994 0.9266 0.8806 0.9738 0.9568 0.9747 0.9537 0.8361
146 259 505 485 407 397 737 1184 746 570 441 236
2.487 3.761 101 2.799 101 2.404 101 1.401 101 1.251 5.745 101 3.148 1.947 101 2.466 102 1.121 102 5.726
2.859 101 1.338 103 2.705 102 7.665 104 5.724 101 1.102 103 1.259 102 4.414 102 4.903 103 5.781 5.258 101 4.423 103
5.352 1.542 102 2.374 103 7.820 3.468 102 6.135 5.637 107 6.734 103 2.103 1.098 1.167 102 2.945 105
3.022 102 4.642 102 5.393 104 4.699 102 5.164 1.845 102 2.375 101 9.104 101 5.059 101 1.407 101 1.636 1.832
— 1.812 102 3.598 5.649 102 — 6.132 101 — 6.320 7.135 101 — 6.655 6.989
(c) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
EFNO ¼ b+cV+dV2+eV2 ln V+f/V EFNO ¼ b+cV+d ln V+e/V1.5+f ln V/V2 EFNO ¼ b+cV+dV ln V+eV1..5+fV2 EFNO ¼ b+cV+dV/ln V+e ln V+f ln V/V2 EFNO ¼ b+cV+d/ln V+e/V1.5+f/V2 EFNO ¼ b+cV+dV/ln V+e/ln V+f/V EFNO ¼ b+cV+dV ln V+eV/ln V+f/V2 EFNO ¼ b+cV+dV ln V+e/V2 EFNO ¼ b+cV+dV ln V+eV/ln V+f/V2 EFNO ¼ b+cV+dV ln V+e ln V+f ln V/V EFNO ¼ b+cV+d ln V/V+e ln V/V2+f/V2 EFNO ¼ b+cV+dV2+eV/ln V+f ln V/V
0.7446 0.9840 0.9513 0.7757 0.9335 0.9530 0.9906 0.9966 0.9980 0.9980 0.9685 0.9895
146 259 505 485 407 397 737 1184 746 570 441 236
2.695 2.317 101 7.570 9.321 5.861 2.965 3.475 1.194 3.628 2.566 101 2.551 102 2.812 101
1.534 101 4.381 102 1.484 4.504 101 1.346 102 2.342 101 5.720 101 1.051 101 5.631 101 3.631 101 8.733 104 1.474
6.764 103 5.535 0.736 3.049 2.220 102 1.538 5.076 102 2.408 102 4.79 102 5.590 102 4.810 1.356 103
1.197 103 3.961 103 0.266 6.611 3.689 102 9.852 1.714 3.523 101 1.740 7.387 9.677 102 7.362
3.189 102 5.049 103 6.831 103 2.072 102 6.074 102 1.893 102 4.593 101 — 9.386 5.352 101 2.183 102 5.863 101
ARTICLE IN PRESS
0.8738 0.9118 0.9739 0.9713 0.9712 0.8677 0.9376 0.9798 0.9965 0.9976 0.8392 0.8953
6849
EFCO ¼ b+cV+d/ln V+e/V EFCO ¼ b+cV+d ln V+e/V+f/V2 EFCO ¼ b+cV+dV3+eV0.5 ln V+fV/ln V EFCO ¼ b+cV+d ln V+e/ln V+f/V EFCO ¼ b+cV+dV2+eV3+f/V2 EFCO ¼ b+cV+dV/ln V+e/ln V+f/V EFCO ¼ b+cV+dV2+eV3+f lnV EFCO ¼ b+cV+dV ln V+e ln V+f ln V/V EFCO ¼ b+cV+d ln V+e/ln V+f/V EFCO ¼ b+cV+dV ln V+e ln V+f ln V/V EFCO ¼ b+cV+dV ln V+e ln V/V2 EFCO ¼ b+cV+dV2 ln V+eV2.5+fV3
T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
(a) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
ARTICLE IN PRESS T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
6850
Table 6 The regression equation of average ECO, EHC and ENO of the aggregate diesel vehicles as a function of model year for 10 km h1 to 75 km h1 in Hong Kong Model year Equations
R2
Samples Constant coefficients b
c
d
e 6.164 102 2.028 103 1.734 102 1.122 107 7.261 101 2.839 103 1.211 103 3.476 103 8.324 101 6.638 101 1.095 103 8.779 102
(a) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
ECO ¼ b+cV2.5+dV3+e/V2 ECO ¼ b+cV3+d/V+e/V2 ECO ¼ b+cV0.5+d ln V+e ln V/V ECO ¼ b+cV2+dV2 ln V+eeV ECO ¼ b+cV1.5+d ln V+e ln V/V ECO ¼ b+c(ln V)2+d/ln V+e/V ECO ¼ b+c ln V+d/V+e ln V/V2 ECO ¼ b+c ln V+d/ln V+e/V ECO ¼ b+cV2+deV+e/V2 ECO ¼ b+cV+dV3+e(ln V)2 ECO ¼ b+cV+dV2+eV2.5 ECO ¼ b+cV2 ln V+dV2.5+eeV
0.5763 146 0.5526 259 0.5040 505 0.5990 485 0.5128 407 0.7527 397 0.5662 737 0.7183 1184 0.9348 746 0.9669 570 0.5850 441 0.5418 236
6.186 1.431 101 9.453 101 9.508 4.533 101 4.561 102 2.654 101 8.325 102 5.259 1.231 101 1.306 4.229
1.452 103 1.481 105 4.402 8.022 103 8.045 10 3 6.076 6.641 8.110 101 1.119 103 1.194 102 4.721 101 1.093 103
6.938 105 2.516 102 2.676 101 1.772 103 9.431 1.617 103 4.661 2.265 103 4.291 1034 1.341 105 1.212 102 1.489 103
(b) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
EHC ¼ b+c ln V+d/ln V+e/V EHC ¼ b+cV1.5+d/V+e/V1.5 EHC ¼ b+c ln V+d/ln V+e/V EHC ¼ b+cV0.5+d/ln V+e/V EHC ¼ b+cV/ln V+d/V+eeV EHC ¼ b+c ln V+d/V0.5+e/V EHC ¼ b+c ln V+d/ln V+e/V EHC ¼ b+cV2+deV+e/V EHC ¼ b+cV3+deV+e/V2 EHC ¼ b+cV+dV3+e/V0.5 EHC ¼ b+c(ln V)2+dV/ln V+eV0.5 EHC ¼ b+cV2.5+dV3+e/V2
0.7341 146 0.5285 259 0.5353 505 0.5015 485 0.5872 407 0.7326 397 0.5684 737 0.8181 1184 0.5111 746 0.6805 570 0.5324 441 0.5955 236
5.780 102 2.067 1.095 102 1.373 101 1.561 101 2.744 7.650 2.162 101 4.387 101 1.723 102 1.453 101 5.170 101
5.545 101 1.005 103 1.018 101 3.508 101 5.092 102 4.038 7.299 5.771 105 1.482 107 2.156 103 2.901 8.174 105
1.606 103 2.447 103 6.248 101 1.773 102 2 3.062 10 4.770 102 4.744 101 8.713 101 9.986 6.839 9.651 101 1.333 102 2 2.102 10 3.223 102 7.139 1035 3.304 1.802 1034 1.056 101 1.222 106 3.001 2.112 1.205 101 9.397 106 1.453 101
(c) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
ENO ¼ b+c ln V+d/ln V+e/V ENO ¼ b+cV2+d/ln V+e/V0.5 ENO ¼ b+ceV+d(ln V)2+e ln V ENO ¼ b+c ln V+d/ln V+e/V ENO ¼ b+cV2+dV3+eeV ENO ¼ b+cV+d ln V+e ln V/V ENO ¼ b+cV ln V+dV3+eeV ENO ¼ b+cV2+dV3+e/V ENO ¼ b+cV2.5+deV+e(ln V)2 ENO ¼ b+cV2.5+dV3+e/V2 ENO ¼ b+cV+dV2+eeV ENO ¼ b+cV+deV+eV0.5
0.4771 146 0.8815 259 0.4726 505 0.4880 485 0.5031 407 0.7143 397 0.8769 737 0.8809 1184 0.6516 746 0.9595 570 0.8980 441 0.5281 236
1.666 103 1.577 101 2.360 101 3.255 103 6.451 8.037 101 3.899 5.544 6.635 4.572 2.248 2.232 101
1.570 102 9.010 106 9.114 1034 3.123 102 1.183 103 1.940 101 1.512 102 1.039 103 2.861 105 1.278 103 1.485 101 4.042 101
4.634 103 2.050 102 1.643 8.991 103 1.312 105 2.155 101 6.061 106 4.119 106 3.631 1034 2.051 105 1.099 103 2.797 1033
3.4. Comparison of the average emission factors of petrol and diesel vehicles According to our previous work of Chan et al. (2004), the average emission factors of EFCO, EFHC and EFNO from petrol vehicles have a strong correlation to the instantaneous vehicle speed profiles. The comparison between the average emission factors of petrol and diesel
7.149 103 1.968 102 1.010 101 1.376 104 1.225 103 1.692 102 4.467 1034 1.607 101 0.119 1.915 102 3.652 1034 5.067
vehicles for different vehicle speeds is shown in Fig. 7. It has shown that the petrol vehicles emit 3.4, 6.0, 9.2 and 15.5 times of CO pollutants higher than the measured diesel vehicles for the vehicle speeds of 10, 30, 50 and 70 km h1, respectively. Similarly, the petrol vehicles emit 2.2, 2.7, 2.7 and 3.3 times of HC pollutants higher than the diesel vehicles for the vehicle speeds of 10, 30, 50 and 70 km h1, respectively. However, the diesel
ARTICLE IN PRESS T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
0.41
Average EFCO (gkm-1)
0.40
0.39 0.37 (0.23) (0.15)
(0.23)
9
0.42 (0.27) 0.39
8
(0.20)
0.37 (0.25)
0.35 (0.21)
0.35
0.30 0.29
0.30
(0.21)
0.28
(0.09)
(0.17) 0.25 0.24
0.25
(0.15)
Average ECO (gl-1)
0.45
6851
7.53 7.41 7.21 (1.83)(0.76) 7.25 (0.83) (1.58)
7.04 (0.90)
7
6.08 (0.80) 5.83 (1.11)
6
6.07 (0.96) 5.25 (1.48) 4.57 (0.59) 4.05 (0.93)
5 4
4.48 (0.84)
(0.18)
3 0.20 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year 0.055
Average EFHC (gkm-1)
0.050
2
0.051 0.048 (0.04)
1.4
(0.02) 0.043 0.042 (0.02) (0.01)
0.045
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
(a)
1.2 0.041 0.039
(0.02) 0.037
(0.02)
0.040
0.035
(0.02)
(0.02)
0.035
0.031 0.029
(0.01)
0.026 (0.02)
0.030
Average EHC (gl-1)
(a)
(0.01)
1.0
1.08 (0.50)
0.96 (0.13) 0.85 0.82 (0.17)(0.10)
0.70 (0.16) 0.61 0.65 (0.10) (0.10)
0.8 0.6
0.77 (0.08) 0.70 (0.17) 0.53 0.43 0.43 (0.08) (0.06) (0.05)
0.4
0.023
0.025
(0.01)
0.2
0.020 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
(b) 0.55
0.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
(b) 12
0.50 (0.21) 0.45
Average EFNO (gkm-1)
(0.16)
0.45
0.47 (0.20)
0.45 (0.16) 0.41 0.36 (0.14)
0.35
0.34 0.32 0.31 (0.11)
0.30
10
(0.17)
(0.21)
0.40
10.14 (1.33)
0.42
0.29 (0.12)
(0.10)
(0.16) 0.27 (0.12)
Average ENO (gl-1)
0.50
8.92 (2.11)
8
9.16 (1.18) 8.72 (1.93) 7.23 7.10 (0.91) (0.91)
7.11 6.69 (1.09) (0.99)
5.78 5.51 5.44 (0.56) (0.32) (0.16) 5.06 (0.53)
6
0.25 0.20
(c)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
4
(c) Fig. 3. Average and standard deviation of EFCO, EFHC and EFNO of the aggregate diesel vehicles as a function of model years in Hong Kong.
vehicles emit 9.1, 7.3, 5.6 and 2.9 times of NO pollutants higher than petrol vehicles for vehicle speeds of 10, 30, 50 and 70 km h1, respectively. The results are mainly attributed to the characteristics of different engine types and operation conditions, and the fuel properties of petrol and diesel.
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
Fig. 4. Average and standard deviation of ECO, EHC and ENO of the aggregate diesel vehicles as a function of model years in Hong Kong.
4. Conclusions In the present study, on-road diesel vehicle gaseous exhaust emissions of CO, HC and NO were investigated using a remote sensing vehicle exhaust emissions testing system at nine sites in Hong Kong. Compared with the
ARTICLE IN PRESS T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
6852 0.40
9 0.35 0.35 0.34 (0.15) (0.17) (0.11) (0.14) 0.34
8
0.31 (0.15)
0.30
7 0.27 (0.07) 0.25 (0.13)
0.24 (0.12)
0.25
0.22 (0.09) 0.19 (0.11)
0.20
Average ECO (gl-1)
Average EFCO (gkm-1)
0.35
0.34 0.34 (0.13) (0.15)
7.34 6.87 6.97 (0.50) 7.07 (0.67) (1.13) (1.04)
6.94 (0.89) 5.96 (0.75) 5.64 (0.96)
6
5.83 (0.78)
4.47 (0.54)
5 4
4.96 (1.18) 3.83 (0.66)
4.35 (0.80)
3 0.15 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
(a)
2
0.050
Average EFHC (gkm-1)
0.045
1.4 0.043 (0.02)
1.2 0.038
0.040
(0.01)
0.039 (0.01) 0.036 (0.01)
0.034
0.034
(0.01)
0.035
0.028
0.030
(0.01)
0.031 (0.01)
(0.01) 0.025
0.025
0.022 (0.01)
Average EHC (gl-1)
0.045 (0.02)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
(a)
0.023 (0.01)
1.0
1.00 0.93 (0.46) (0.07)
0.83 0.81 (0.16) (0.10)
0.8
0.68 0.63 (0.16) 0.59 (0.09) (0.09)
0.6
0.75 (0.07) 0.66 (0.14) 0.52 0.42 0.42 (0.07) (0.04) (0.05)
0.4
(0.01)
0.2
0.020 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
(b)
Model Year
0.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
(b)
0.50
0.40 0.35 0.30
0.42 (0.13)
12 0.42 0.41 (0.13) (0.13)
9.87 (1.11)
10
0.36 (0.14)
0.33 0.31 (0.12) (0.08)
0.33 (0.09) 0.29 0.28 0.28 (0.08) (0.08) (0.11) 0.24 (0.08)
0.25
Average ENO (gl-1)
Average EFNO (gkm-1)
0.45
0.45 (0.12)
8.61 (1.97)
8
6
8.96 (1.07)
8.40 (1.75) 7.10 6.98 (0.88) (0.89)
6.83 6.56 (0.65) (0.96) 5.71 5.44 5.40 (0.54) (0.21) (0.11) 4.92 (0.33)
0.20 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
(c)
4
Model Year
Fig. 5. Average and standard deviation of EFCO, EFHC and EFNO of the aggregate diesel vehicles without the gross polluters as a function of model years in Hong Kong.
aggregate diesel vehicle model years at the Hong Kong sites, an overall significant reduction trend of CO, HC and NO emissions was observed especially when the 10% of high emitters are not taken into the consideration. The advancement of engine technology and diesel oxidation catalytic converter has kept the diesel vehicle
(c)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Model Year
Fig. 6. Average and standard deviation of ECO, EHC and ENO of the aggregate diesel vehicles without the gross polluters as a function of model years in Hong Kong.
emissions down in recent model years. A regression analysis approach based on the measured diesel vehicle emission data was also used to estimate the on-road diesel vehicle gaseous emission factors (i.e., g km1 and g l1) of CO, HC and NO with respect to the effects of instantaneous vehicle speed and acceleration/deceleration
Table 7 The regression equation of average EFCO, EFHC and EFNO of the diesel vehicles as a function of different vehicle types for 10 km h1 to 75 km h1 in Hong Kong Vehicle type
Regression equation
R2
Samples
Constant coefficients c
d
e
f 1.485 102
EFCO ¼ b+cV+dV/ln V+e ln V+f/ln V EFCO ¼ b+cV+dV ln V+e ln V EFCO ¼ b+cV+d/V1.5+e ln V/V2+f/V2 EFCO ¼ b+cV+dV3+e ln V EFCO ¼ b+cV ln V+dV2 EFCO ¼ b+cV2+dV2 ln V+eV/ln V EFCO ¼ b+cV+dV/ln V+e/ln V+f/V
0.8839 0.9952 0.9646 0.7513 0.9887 0.8166 0.9945
86 1636 3486 479 8 330 296
1.092 102 7.561 1.092 101 7.186 8.004 101 4.115 4.986 102
1.683 3.503 101 6.737 104 1.446 101 4.894 103 0.004 4.506 101
1.135 101 6.131 102 3.665 102 1.008 105 3.078 101 6.882 104 2.878
33.745 3.383 5.670 102 2.894 — 5.548 101 1.605 102
3.819 102 — — — 2.591 102
(b) Passenger Taxi Light duty Middle duty Heavy duty Light bus Bus
EFHC ¼ b+cV+d/ln V+e ln V/V2+f/V2 EFHC ¼ b+cV+dV ln V+eV/ln V+f ln V EFHC ¼ b+cV+dV2+e ln V/V2+f/V2 EFHC ¼ b+cV+dV ln V+eV2 ln V EFHC ¼ b+c ln V/V+d/V EFHC ¼ b+cV+d ln V+e/ln V+f/V EFHC ¼ b+cV+dV2+eV3
0.8977 0.9610 0.7921 0.9215 0.8484 0.8679 0.9856
86 1636 3486 479 8 330 296
4.144 103 8.559 101 7.723 102 2.265 101 6.045 102 2.955 102 2.455 101
8.688 104 3.438 101 1.124 103 5.537 2.328 6.290 102 2.043 102
3.784 103 2.728 102 5.353 106 5.655 101 7.819 3.135 101 1.630 103
7.226 102 1.157 1.781 101 1.492 101 — 7.765 102 2.278 106
1.962 102 5.256 101 5.701 101 — — 1.125 103 —
(c) Passenger Taxi Light duty Middle duty Heavy duty Light bus Bus
EFNO ¼ b+cV/ln V+d/V+e ln V/V2 EFNO ¼ b+cV+d ln V+e/V+f/V2 EFNO ¼ b+cV+d ln V+f/ln V EFNO ¼ b+cV+dV2+eV3+f ln V/V EFNO ¼ b+cV0.5+d/V0.5+e ln V/V EFNO ¼ b+cV+d/ln V+e/V+f/V2 EFNO ¼ b+cV+dV ln V+eV2+fV3
0.8813 0.9700 0.9578 0.9216 0.9968 0.9807 0.9848
86 1636 3486 479 8 330 296
1.016 2.445 1.393 101 4.003 1.075 102 2.697 101 1.037 101
3.375 102 6.771 103 3.524 102 9.749 102 3.952 2.697 102 1.871
5.533 101 6.227 101 2.466 1.123 103 1.325 103 1.294 102 5.371 101
2.081 102 2.921 101 2.364 101 4.554 106 1.373 103 3.727 102 1.026 102
— 1.148 102 — 2.577 102 — 1.221 103 3.340 105
ARTICLE IN PRESS
(a) Passenger Taxi Light duty Middle duty Heavy duty Light bus Bus
T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
b
6853
6854
Table 8 The regression equation of average ECO, EHC and ENO of the diesel vehicles as a function of different vehicle types for 10 km h1 to 75 km h1 in Hong Kong Vehicle type
Regression equation
R2
Samples
Constant coefficients c
d
e
ECO ¼ b+c ln V/V+d/V+e/V2 ECO ¼ b+cV+d ln V+e ln V/V ECO ¼ b+cV3+d/ln V+e ln V/V ECO ¼ b+cV+dV3+eeV ECO ¼ b+cV1.5+d/V0.5+e ln V/V ECO ¼ b+c ln V+d/V+e/V1.5 ECO ¼ b+cV ln V
0.5635 0.9342 0.5346 0.6055 0.5942 0.5413 0.9411
86 1636 3486 479 8 330 296
2.271 101 1.621 102 2.792 101 3.492 2.499 103 2.149 102 9.831
1.344 102 4.617 101 9.096 107 2.694 101 4.985 101 4.032 101 2.249 102
4.931 103 0.403 102 1.904 102 3.848 105 4.217 104 5.043 103 —
2.524 104 2.891 102 1.967 102 2.533 1033 4.663 104 1.322 104 —
(b) Passenger Taxi Light duty Middle duty Heavy duty Light bus Bus
EHC ¼ b+cV3+deV+e/V1.5 EHC ¼ b+cV3+d/V2+eeV EHC ¼ b+c(ln V)2+d ln V+e/V EHC ¼ b+cV0.5 ln V EHC ¼ b+cV3+d/V0.5+e ln V/V EHC ¼ b+cV+d/ln V+e/V EHC ¼ b+cV+d/V0.5+e/V
0.6980 0.6191 0.5108 0.5791 0.5413 0.5876 0.6395
86 1636 3486 479 8 330 296
1.022 3.157 101 1.384 101 1.864 1.425 102 6.215 101 2.902
1.529 106 6.472 107 6.250 101 3.378 102 2.278 105 1.254 101 1.078 102
3.557 1034 4.671 101 5.959 — 2.681 103 2.639 102 1.505 101
6.852 101 2.305 105 4.565 101 — 3.304 103 5.431 102 2.652 101
(c) Passenger Taxi Light duty Middle duty Heavy duty Light bus Bus
ENO ¼ b+cV2.5+dV3+e/V2 ENO ¼ b+cV2+dV3+e/V2 ENO ¼ b+c ln V+d/ln V+e/V ENO ¼ b+cV+dV ln V+eV ln V ENO ¼ b+cV/ln V+d/V+e/V2 ENO ¼ b+c ln V+d/V+e/V2 ENO ¼ b+cV2.5+dV3+e/V2
0.6273 0.5424 0.5354 0.6516 0.6606 0.6049 0.5582
86 1636 3486 479 8 330 296
4.445 4.076 2.202 102 2.352 102 5.687 102 1.146 102 9.943
1.023 103 1.459 103 1.928 101 5.229 102 2.348 101 2.029 101 1.254 103
2.205 105 9.932 106 6.171 102 5.008 1.544 104 1.337 102 4.752 105
2.067 102 1.077 102 9.924 102 1.452 102 1.493 105 9.352 103 5.684 102
ARTICLE IN PRESS
(a) Passenger Taxi Light duty Middle duty Heavy duty Light bus Bus
T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
b
ARTICLE IN PRESS T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
6 Average EFCO (gkm-1)
5.05
EFCO of Diesel Vehicles
5
EFCO of Petrol Vehicles
4 3 2
2.50
2.48
2.21
6855
vehicle gaseous emission factor (i.e., g km1 and g l1) database on different model years and vehicle types for urban driving patterns in Hong Kong has been established which can be used for predicting the dispersion of real-world on-road diesel vehicle emissions at the specific site and supporting human exposure studies in urban roadway environments.
1.50
1
0.42
0.24
Acknowledgements
0.16
0 10
50
Vehicle Speed (kmh ) 0.198
0.20
EFCO of Diesel Vehicles EFCO of Petrol Vehicles
0.10
References
0.093
0.092
0.071
0.05
0.034
10
0.026
30
0.070
0.021
50
70
Vehicle Speed (kmh-1)
(b)
1.4
1.27
EFCO of Diesel Vehicles
1.2
EFCO of Petrol Vehicles
1.0 0.8 0.6
0.44
0.4 0.2
0.28 0.20
0.14 0.06
0.05
0.07
0.0 10 (c)
This work was supported by the grants from the Research Grants Council of the Hong Kong Special Administrative Region, China (RGC Project No. PolyU 5292/03E) and the Central Research Grants of The Hong Kong Polytechnic University (Project Nos. BQ738 and 143-B1-9719).
0.15
0.00
Average EFNO (gkm-1)
70
-1
(a)
Average EFHC (gkm-1)
30
30
50
70
Vehicle Speed (kmh-1)
Fig. 7. The comparison of average emission factors of EFCO, EFHC and EFNO from petrol and diesel vehicles in Hong Kong.
profiles for urban driving patterns in Hong Kong. The results show that the diesel vehicle model years, vehicle types and driving patterns have a strong correlation to their gaseous emission levels. The deviation of the average emissions of aggregate diesel vehicle reflects on the variability of local road condition, vehicle traffic fleet and volume, driving pattern, fuel composition and ambient condition. Finally, a unique on-road diesel
API, 2001. Alcohols and Ethers, A Technical Assessment of Their Application as Fuels and Fuel Components, third ed. Publication 4261, American Petroleum Institute, Washington, DC. Bishop, G.A., Stedman, D.H., 1996. Measuring the emissions of a passing car. Accounts of Chemical Research 29, 489–495. Chan, T.L., Dong, G., Ning, Z., Hung, W.T., Cheung, C.S., Leung, C.W., 2002. On-road remote sensing of petrol vehicle emissions measurement and emission factors estimation for urban driving patterns in Hong Kong. In: Proceedings of the Regional Workshop on Better Air Quality in Asian and Pacific Rim Cities (BAQ2002), vol. III, Hong Kong, December 16–18, pp. 291–298. Chan, T.L., Ning, Z., Leung, C.W., Cheung, C.S., Hung, W.T., Dong, G., 2004. On-road remote sensing of petrol vehicle emissions measurement and emission factors estimation in Hong Kong. Atmospheric Environment 38, 2055–2066, 3541. Cross, T., 2000. Remote sensing technology. In: Proceedings of Better Air Quality—Motor Vehicle Control & Technology Workshop, Session 2—In-use Vehicle Testing Technology. The Hong Kong Polytechnic University, Hong Kong. Hong Kong Environmental Protection Department (HKEPD), March 2005. Environmental Protection Department of Hong Kong SAR, http://www.epd.gov.hk/epd/english/ environmentinhk/air/air_maincontent.html. Hong Kong Transportation Department (HKTD), April 2005. Trend of registration and licensing of vehicles by class of vehicles. Transportation Department of Hong Kong SAR, http://www.info.gov.hk/td/eng/transport/regist1_index.html. Holmen, B.A., Niemeier, D.A., 1998. Characterizing the effects of driver variability on real-world vehicle emissions. Transport Research Part D 3, 117–128. Jost, P., Hassel, D., Joumard, R., Hickman, A.J., 1994. Vehicle emissions and fuel consumption modelling based on continuous measurements. SAE Technical Paper No. 945127. Joumard, R., Jost, P., Hickman, J., 1995a. Influence of instantaneous speed and acceleration on hot passenger car
ARTICLE IN PRESS 6856
T.L. Chan, Z. Ning / Atmospheric Environment 39 (2005) 6843–6856
emissions and fuel consumption. SAE Technical Paper No. 950928. Joumard, R., Jost, P., Hickman, J., Hassel, D., 1995b. Hot passenger car emissions modeling as a function of instantaneous speed and acceleration. The Science of the Total Environment 169, 167–174. Joumard, R., Andre, M., Vidon, R., Tassel, P., 2003. Characterizing real unit emissions for light duty goods vehicles. Atmospheric Environment 37, 5217–5225. Kean, A.J., Harley, R.A., Kendall, G.R., 2003. Effects of vehicle speed and engine load on motor vehicle emissions. Environmental Science & Technology 37, 3739–3746. Kelly, N.A., Groblicki, P.J., 1993. Real-world emissions from a modern production vehicle driven in Los-Angeles. Journal of the Air and Waste Management Association 43, 1351–1357. Kouridis, C., Ntziachristos, L., Samaras, Z., 2000. COPERT III. Computer programme to calculate emissions from road transport. User’s Manual (Version 2.1), EEA, Technical Report No. 50, Copenhagen, 46pp. Mayer, H., 1999. Air pollution in cities. Atmospheric Environment 33, 4029–4037. Ntziachristos, L., Samaras, Z., 1999. COPERT III. Computer programme to calculate emissions from road transport. Methodology and emission factors (Version 2.1). EEA, Technical Report No. 49, Copenhagen, 86pp. Pokharel, S.S., Bishop, G.A., Stedman, D.H., 2001. Fuel-based on-road motor vehicle emissions inventory for the Denver metropolitan area. Presented at the International Emission Inventory Conference, Denver, CO, USA, http://www. feat.biochem.du.edu/reports.html. Pokharel, S.S., Bishop, G.A., Stedman, D.H., 2002. An on-road motor vehicle emissions inventory for Denver: an efficient alternative to modeling. Atmospheric Environment 36, 5177–5184. Saija, S., Romano, D., 2002. A methodology for the estimation of road transport air emissions in urban areas of Italy. Atmospheric Environment 36, 5377–5383. Schifter, I., Diaz, L., Duran, J., Guzman, E., Chavez, O., Lopez-Salinas, E., 2003. Remote sensing study of emissions from motor vehicles in the metropolitan area of Mexico city. Environmental Science & Technology 37, 395–401. Schifter, I., Diaz, L., Mugica, V., Lopez-Salinas, E., 2005. Fuelbased motor vehicle emission inventory for the metropolitan area of Mexico city. Atmospheric Environment 39, 931–940. Singer, B.C., Harley, R.A., 2000. A fuel-based inventory of motor vehicle exhaust emissions in the Los Angeles area during summer 1997. Atmospheric Environment 34, 1783–1795. Singer, B.C., Harley, R.A., Littlejohn, Ho, Vo, T., 1998. Scaling of infrared remote sensor hydrocarbon measurements for
motor vehicle emission inventory calculations. Environmental Science & Technology 32, 3241–3248. Sjodin, A., Persson, K., Andreasson, K., Arlander, B., Galle, B., 1998. On-road emission factors derived from measurements in a traffic tunnel. International Journal of Vehicle Design 20, 147–158. Stedman D.H., January 2002. Remote sensing: a new tool for automobile inspection & maintenance. Independent Institute Paper, http://www.feat.biochem.du.edu/links.html. Stephens, R.D., Giles, M.T., Groblicki, P.J., Gorse, R.A., McAlinden, K.J., Hoffman, D.B., James, R., Smith, S., 1994. Real world emissions variability as measured by remote sensors, SAE Technical Paper No. 940582. Sturm, P.J., Almbauer, R., Sudy, C., Pucher, K., 1997. Application of computational methods for the determination of traffic emissions. Journal of the Air and Waste Management Association 47, 1204–1210. Tong, H.Y., Hung, W.T., Cheung, C.S., 2000. On-road motor vehicle emissions and fuel consumption in urban driving conditions. Journal of the Air and Waste Management Association 50, 543–554. Tsang, M., Ha, K., 2002. The implementation of stringent emission standards and fuel specifications, a Hong Kong example. In: Proceedings of the Regional Workshop on Better Air Quality in Asian and Pacific Rim Cities (BAQ2002), vol. III, Hong Kong, December 16–18, pp. 418–422. US California Air Resources Board (USCARB), April 2004, http://www.arb.ca.gov/msei/msei.htm. US Environmental Protection Agency (USEPA), February 2004. MOBILE6 vehicle emission modeling software, http:// www.epa.gov/otaq/m6.htm. Vanruymbeke, C., Guitton, J.P., Vidon, R., Pruvost, C., 1993. Testing of an alternative method of pollutant emission measurement for passenger cars. The Science of the Total Environment 134, 197–209. Walsh, M.P., 2001. High polluting petrol fuelled vehicles—an approach to reducing emissions. Report to the Environmental Protection Department of Hong Kong SAR, China. Yu, L., 1998. Remote vehicle exhaust emission sensing for traffic simulation and optimization models. Transportation Research Part D 3, 337–347. Zachariadis, T., Samaras, Z., 1997. Comparative assessment of European tools to estimate traffic emissions. International Journal of Vehicle Design 18, 312–325. Zhao, H., Cheung, C.S., Hung, W.T., 1999. Investigation of motor vehicle on-road exhaust emissions behavior. Journal of Combustion Science and Technology 5, 100–107 (English version).
