University of California Transportation Center
UCTC Research Paper No. UCTC- 2010- 09
A Preliminary Analysis of the Environmental Impacts of the Clean
Truck Program in the Alameda Corridor, CA
Roberto Ayala, Soyoung ( Iris) You, Jean- Daniel Saphores,
Stephen Ritchie, Gunwoo Lee, and Mana Sangkapichai
University of California, Irvine
March 2010
1
1 A Preliminary Analysis of the Environmental Impacts of the Clean Truck
2 Program in the Alameda Corridor, CA
3
4 Roberto Ayala
5 Ph. D. Student
6 Transportation Science, Institute of Transportation Studies
7 University of California, Irvine, CA 92697
8 TEL: ( 949) 824- 7334; FAX: ( 949) 824- 8385
9 E- MAIL: ayalar@ uci. edu
10
11 Soyoung ( Iris) You
12 Ph. D. Candidate
13 Department of Civil & Environmental Engineering
14 University of California, Irvine, CA 92697
15 E- MAIL: soyoungy@ uci. edu
16
17 Jean- Daniel Saphores
18 Corresponding Author, Associate Professor
19 Department of Civil & Environmental Engineering
20 University of California, Irvine, CA 92697
21 TEL: ( 949) 824- 7334; FAX: ( 949) 824- 8385
22 E- MAIL: saphores@ uci. edu
23
24 Stephen Ritchie
25 Professor
26 Department of Civil & Environmental Engineering
27 University of California, Irvine, CA 92697
28 TEL: ( 949) 824- 4214; FAX: ( 949) 824- 8385
29 E- MAIL: sritchie@ uci. edu
30
31 Gunwoo Lee
32 Ph. D. Candidate
33 Department of Civil & Environmental Engineering
34 University of California, Irvine, CA 92697
35 E- MAIL: gunwool@ uci. edu
36
37 Mana Sangkapichai
38 Ph. D. Candidate
39 Transportation Science, Institute of Transportation Studies
40 University of California, Irvine, CA 92697
41 E- MAIL: msangkap@ uci. edu
42
43
44 Word Count: 4,932 + 9* 250 ( 6 figures and 3 tables) = 7,182 words.
2
1 ABSTRACT
2 The San Pedro Bay Ports ( SPBP) of Los Angeles and Long Beach in Southern California is one
3 of the largest port container complexes in the world, and the largest one is the United States. To
4 decrease the air pollution associated with port operations, a number of measures have been
5 adopted, including the Clean Trucks Program, which was introduced in 2008 to clean up the fleet
6 of drayage trucks serving the SPBP. The objective of this paper is to quantify the reduction in
7 emissions attributable to the Clean Trucks Program, with a focus on Nitrogen Oxide ( NOx) and
8 Particulate Matter ( PM2.5). Our approach is innovative as it relies on micro- simulation to capture
9 the link between congestion and pollutant emissions. We find that the Clean Trucks Program
10 could contribute significantly to the emissions of NOx (~ 27%) and PM2.5 (~ 25%) for all the
11 freeway traffic in our study area. These preliminary results suggest that the Clean Trucks
12 Program is promising, but its cost- effectiveness should be analyzed.
3
1 INTRODUCTION
2 The San Pedro Bay Ports ( SPBP) of Los Angeles and Long Beach in Southern California
3 comprise one of the largest container port complexes in the world. Although the SPBP’s
4 contribution is essential to both the state’s and the nation’s economies, increasing congestion and
5 air pollution have been affecting the daily lives of those who reside, work, and attend school near
6 the SPBP and along the freeways used by drayage trucks. According to the 2007 emission
7 inventory, SPBP related Heavy Duty Vehicles contribute 35% of PM2.5 ( particulate matter with a
8 diameter ≤ 2.5 μm), 35% of NOX ( nitrous oxides), and 57% of CO ( carbon monoxide) of port-9
related emissions in the South Coast Air Basin and within the SPBP boundary [ 1].
10 Widespread concerns about air pollutants have lead state and local government
11 organizations including the California Air Resource Board ( CARB) and the Southern California
12 Association of Governments ( SCAG), along with the SPBP, to propose strategies for reducing
13 air pollution generated by the movement of goods in and out of the SPBP complex. One of these
14 strategies is the Clean Trucks Program ( CTP), which started in October 1, 2008, with a ban on all
15 pre- 1989 trucks. Its goal is to reduce air pollution from harbor trucks by more than 80 percent by
16 2012 by replacing older and damaged trucks, retrofitting truck engines, or installing Diesel
17 Particular Filter ( DPF) [ 2].
18 The main question this paper addresses is: “ How effective has the Clean Truck Program
19 been at reducing air pollutants?” In particular, we focus on Nitrous Oxide ( NOx) and Particulate
20 Matter ( PM), which are major sources of concern for the health of local residents. NOx, which in
21 our case is mostly NO2, is an ozone precursor; PM can cause premature respiratory and
22 cardiovascular deaths as well as hospital admissions for a variety of ailments [ 3]. Exposure to air
23 pollutants results in both long- term and short- term health effects. Short- term effects include eye,
24 nose and throat irritation, bronchitis and pneumonia [ 4, 5]. Long- term health effects include lung
25 cancer, chronic respiratory and heart disease; it can also cause premature death and degrade the
26 health of elderly people.
27 The amount of air pollutants emitted into the atmosphere by trucks depend on vehicle
28 speed, miles traveled, acceleration and deceleration rates, which are ignored by planning models
29 such as EMFAC or MOBILE. We therefore rely on micro- simulation to model traffic in the
30 Alameda corridor, between the SPBP and downtown Los Angeles ( see Figure 1), to obtain more
31 realistic estimates of traffic congestions and the resulting air pollution.
32 This paper first summarizes some air pollutant reduction programs that are relevant to the
33 Clean Truck Program. Next, we provide some background information about the Clean Truck
34 Program before discussing our assumptions and our methodology. We then summarize results
35 from our analyses before presenting concluding remarks. To the best of our knowledge, this is
36 the first paper that attempts to examine the environmental benefits of the Clean Truck Program
37 using micro- simulation
38
39 REVIEW OF SOME RELEVANT PAPERS
40 A review of the transportation literature did not turn out any other paper that relies on traffic
41 micro- simulation to analyze the environmental benefits of a measure like the Clean Truck
42 Program. Micro- simulation ( we used TransModeler) allows us to quantify the impact of vehicle
43 interactions ( such as lane- changing, accelerations and decelerations) on the emissions of various
44 pollutants.
45
4
1
2 Figure 1. Map of study area.
3
4 Similar to papers dealing with changes in fuel efficiency standards, such as the Corporate
5 Average Fuel Economy ( CAFE) standards, this paper addresses the change in emissions from a
6 change in policy. The policy, in this case, is stricter emissions standards for trucks that service
7 the San Pedro Bay Ports. Critics may argue that implementing stricter emission standards is not
8 a first best approach and that implementing higher fuel taxes instead would reduce emissions
5
1 more by changing travel behavior [ 6]. However, increasing the fuel tax in this case, will likely
2 not affect port truck traffic as the demand for transporting goods is largely inelastic. In addition,
3 the primary goal of the Clean Trucks Program is to reduce the emission of air pollutants from
4 drayage trucks and not to tackle port truck congestion. Implementing emission standards by
5 government intervention provides an incentive that would otherwise not exist under free market
6 conditions [ 7].
7 A number of published papers study the benefits of reducing emissions in other parts of
8 the world. Using a simulation model ( MIMOSA), Schrooten et al. [ 8] analyze scenarios for
9 reducing air pollutants ( including PM) in the Belgian region of Flanders. The most intensive
10 scenario in their paper decreases PM emissions by 44% up to the year 2010. The marginal costs
11 for a number of emission reduction standards are relatively high, and it appears that equipping
12 heavy- duty trucks with particulate filters is the most cost- effective approach. Retrofitting buses
13 is less effective than retrofitting heavy- duty trucks because buses travel relatively fewer miles
14 each year. Replacing older heavy- duty trucks and buses by hybrids or biodiesel vehicles have
15 higher costs than benefits. Previous studies performed in London show approximately the same.
16 Jansen and Denis [ 9] on the other hand compare four policies that aim at reducing carbon
17 emissions: ( 1) an emissions cap, ( 2) a CO2 purchase “ feebate,” ( 3) a combination of “ feebate”
18 with a CO2 tax, and ( 4) instruments directed at other policies, such as road pricing and a NOx tax.
19 They rely on the software EUCARS to conduct a welfare analysis. They find that any policy that
20 involves a fuel tax has a greater impact on reducing fuel emissions. A fuel tax is also welfare
21 improving since it reduces emissions while raising revenues. The authors then evaluate an
22 equivalent tax on NOx in combination with other policies. Results are similar: the combination of
23 a tax on NOx emissions with a “ feebate” has the greatest impact at reducing emissions.
24 Also of interest, Miraglian and El Khouri [ 10] performed a cost benefit analysis of the
25 addition of stabilized ethanol/ diesel blend into the bus and truck fleet of the Greater Metropolitan
26 Area of Sao Paulo. Their paper incorporates the health benefits from a reduction in air pollutants
27 using health benefit parameters published by the EPA. They find that adding the blend improves
28 air quality and yields a net benefit of approximately US$ 2.85 billion.
29
30 BACKGROUND INFORMATION ON THE CLEAN TRUCK PROGRAM
31 The Clean Truck Program is a joint project of the Ports of Long Beach and Los Angeles. Its goal
32 is to reduce air pollution from drayage trucks by 80% by year 2012, mostly by modernizing the
33 fleet of drayage vehicles [ 11]. It is well known that drayage trucks are a large contributor to air
34 pollution in the Southern Basin [ 12]. The Clean Truck Program bans pre- 1989 trucks, as well as
35 trucks that do not meet 2007 Emission Standards from servicing the port area. It also provides
36 incentives in the form of grants, subsidies and leases, to buy newer and cleaner drayage trucks
37 that meet both state and federal emission standards. Funds from both Proposition 1B and the
38 Clean Truck Fee are used to help finance vehicle fleet upgrades. For compliance purposes, all
39 truck operators are required to install radio frequency identification tags ( RFID) on their trucks.
40 They must also register their vehicles with the Drayage Truck Registry, a database that keeps
41 information on truck age, model, year, engine year, and fuel type. Thanks to RFID tags, the
42 Drayage Truck Registry also collects information about movements in and out of the port area.
43 The Clean Truck Program is divided into two phases. The first phase bans any truck
44 equipped with a pre- 1989 model year engine from entering the ports after December 31, 2009.
45 Trucks engines that are newer than 1989 but that are not in compliance with the 2007 Emission
46 Standards, established by the California Air Resource Board and the Environmental Protection
6
1 Agency, are subject to the Clean Truck Fee, which fee is currently set at $ 35 per twenty- foot
2 equivalent container. Trucks that meet the 2007 Emission Standards are exempt from that fee if
3 they were purchased with private funds ( as opposed to funds from Proposition 1B).
4 The second phase of the Clean Truck Program starts after December 31, 2013; it bans
5 from the SPBP any truck equipped with a pre- 1994 model year engine, or any truck that is not in
6 compliance with the 2007 Emission Standards.
7 The Clean Truck Program began collecting the Clean Truck Mitigation Fee on February
8 18, 2009, despite legal challenges from both the American Trucking Association and the Federal
9 Maritime Commission that delayed its implementation ( its original starting date was October 1,
10 2008 [ 13]). On July 28, the American Trucking Association filed a lawsuit about this fee but the
11 judge dismissed the case and the program was allowed to move forward. Then in the fall of
12 2008, the Federal Maritime Commission ( FMC) issued an injunction on parts of the program
13 with the Supreme Court, Washington D. C. District. Under the Federal Shipping Act of 1984, the
14 FMC has the right to intervene when it thinks unfair competitive restrictions or unduly expensive
15 mandates have been placed on international commerce.
16 The FMC claimed that by requiring truck drivers to register their trucks through a
17 trucking company, the program would reduce competition among owner- operated truck drivers.
18 The FMC also argued that implementing a Clean Truck Fee and subsequently requiring truckers
19 to change their vehicles to cleaner- burning trucks would substantially increase transportation
20 costs and drive trucking companies out of business. The Clean Truck Program is working closely
21 with trucking companies to ensure that truck drivers are eligible for up to 20% of the cost of
22 replacing their vehicle. Proposition 1B allows $ 50,000 per truck in the form of grants, loans, and
23 leases to partially fund upgrades to new and cleaner trucks. With changes to the clean truck fee
24 and with the appointment of Joseph E. Brennan under the new administration, in June 2009 the
25 FMC decided to drop the lawsuit against the Clean Truck Program.
26
27 The 2007 California Emission Standards
28 It is useful at this point to briefly review the 2007 California emission standards. These criteria
29 were established according to engine year by the California Air Resource Board. Note that the
30 toughest restrictions after 2007 apply to both NOx and PM emissions.
31
32 Table 1. Regulations for Heavy- Duty Diesel Truck Engines [ g/ bhp* hr]
Year HC CO NOx PM
1988 1.3 15.5 10.7 0.60
1990 1.3 15.5 6.0 0.60
1991 1.3 15.5 5.0 0.25
1994 1.3 15.5 5.0 0.10
1998 1.3 15.5 4.0 0.10
2007 and later 1.3 15.5 0.20 0.01
33
34 On and after 2007, the limit on non- methane hydrocarbons is 0.14 g/ bhp- hr In terms of
35 grams per mile, this translates approximately to the following emission limits:
36
37
7
1 Table 2. Weight- based emission limits for heavy- duty diesel trucks [ grams/ mile]
Vehicle weight HC NOx PM
8,500- 10,000 lbs 0.20 g/ mi 0.02 g/ mi 0.195 g/ mi
10,000- 14,000 lbs 0.40 g/ mi 0.02 g/ mi 0.230 g/ mi
2
3 Vehicle Fleet Distribution
4 To estimate the benefits from the Clean Truck Program, it is essential to know key characteristics
5 such as fuel and engine age. These data come from two sources. The first source is a study
6 produced by the California Air Resource Board. The second source is the Port of Long Beach
7 Drayage Truck Registry. Figure 2 ( top panel) depicts the age distribution for port trucks
8 predicted up to the year 2014 by the 2006 CARB study ( 13). As a baseline, this CARB study also
9 estimated the 2005 vehicle fleet distribution based on available data. This case is labeled as
10 “ Baseline 2005.” Note that for the 2005 baseline, most trucks servicing the SPBP are from
11 model years 1994 to 2002. Older trucks ( pre- 1993) make up approximately 28 percent of the
12 total, while newer ones ( 2003 and more recent) make up less than 10 percent of the total.
13 The truck age distributions for 2010, 2011, and 2014 assume that both phases of the
14 Clean Truck Program have been implemented and that most truck drivers will install particulate
15 filters instead of purchasing new trucks to meet the Clean Truck Program requirements.
16 For the year 2010, which is denoted in blue, 87.7 percent of trucks servicing the port area
17 are assumed to have model year 1994- 2002 engines retrofitted to meet the 2007 emission, 10.4
18 percent are model year 2003- 1006, 1.7 percent are model year 2007- 2009, and the balance ( less
19 than 1 percent) are model year 2010. As the Clean Trucks Program matures ( red and green
20 bars), a slightly higher percentage of trucks is assumed to be model engine year 2010 and above.
21 Although knowing the truck age distribution is informative, it is more important to know the
22 distribution of engine years for analyzing emissions.
23 As shown on the bottom panel of Figure 2, in 2005 ( Baseline Scenario) approximately 28
24 percent of port trucks were equipped with pre- 1993 model year engines, roughly 63 percent had
25 1994- 2002 engines, and slightly less than 10 percent had 2003- 2006 model year engines. As the
26 Clean Trucks Program progresses, dust particulate filters are installed, engines are retrofitted,
27 and old trucks are replaced with newer trucks, so by the year 2010 ( denoted in blue), most trucks
28 are assumed to meet 2007 emission standards. Hence, for modeling purposes, trucks that are
29 CARB 2007 compliant are assumed to be engine model year 2007. In particular, in year 2010,
30 2003- 2006 model year engines are assumed to be 2007 emission compliant as they were
31 retrofitted or equipped with dust particulate filters; moreover, slightly less than two percent of
32 trucks will have 2007- 2009 model year engines and less than one percent will be brand new
33 trucks ( model year 2010 and above). As the Clean Trucks Program moves forward, a higher
34 percentage of trucks is assumed to have model year engines 2010 and above. For the year 2014
35 ( denoted in green), 99 percent of trucks is assumed to be engine model year 2007- 2009, and
36 slightly less than 1 percent is assumed to have 2010 model year ( and above) engines.
37
38
8
1
2
3 Figure 2. Truck Age and Engine Model Year Distributions
4
5 Figure 3 depicts the predicted age distribution of trucks from the CARB 2006 study, and
6 the actual truck age distribution servicing the ports in May of 2009 based on the Drayage Truck
7 Registry Database. The 2007- 2009 spike suggests that a surprisingly higher amount of new
9
1 trucks were ordered and replaced than was originally predicted. This suggests that a slightly
2 higher than expected proportion of truck operators opted to replace their old trucks for new
3 trucks, as opposed to installing dust particular filters or retrofitting engines. This may be due to
4 the expected compensation from Proposition 1B funds for replacing trucks.
5
6
7
8 Figure 3. Engine Age Distribution ( Baseline versus Actual)
9
10 For simplicity, we assume that trucks whose engine was retrofitted to meet the 2007
11 standards perform just like 2007 model year trucks from the point of view of emissions. This is
12 likely a lower bound on emissions as truck emissions also depend on truck aerodynamic
13 characteristics, for example.
14 For future reference, it is also useful to summarize the characteristics of the current fleet
15 of drayage trucks serving the SPBP based on information contained in the Drayage Truck
16 Registry. In May 2009, there were approximately 15,000 trucks registered with the Port Drayage
17 Registry. Approximately 22 percent of these trucks were in compliance with the 2007 Emission
18 Standards; trucks that are not in compliance with the 2007 Emission Standards are subject to the
19 Clean Truck Program fee, which helps finance newer clean trucks. Table 3 shows the
20 distribution of port trucks by year and reported engine fuel type, which is useful for estimating
21 emission rates for the actual 2009 vehicle fleet. It is important to note that the Clean Truck
22 Program does not mandate a specific fuel type; this is left to truck operators. Due to the high cost
23 of natural gas trucks, and the unavailability of hybrid- diesel trucks, most trucks servicing the port
24 have diesel engines; most of these are in compliance with the 2007 emission standards.
25
26
27
10
1 Table 3: Port Truck Numbers by Fuel Type
Fuel Type Count Percent
BioDiesel 64 0.41
Diesel 15,208 97.83
Natural Gas 259 1.67
Other Alt. Fuel 13 0.08
Unleaded 2 0.01
Total 15,546 100.00
2
3 To estimate the benefits of replacing older engines with newer clean engines, we rely on the
4 information contained in the tables above. Traffic levels are held constant to 2005 levels since
5 predicted traffic levels for future years are very uncertain. Hence, our emission reduction
6 estimates should be taken as the benefits from implementing the Clean Truck Program in 2005.
7
8 Study Area
9 The San Pedro Bay Ports ( SPBP) of Los Angeles and Long Beach in Southern California is
10 serviced by several freight corridors. Although port trucks also use surface streets ( e. g., to
11 access local railyards), we focus on emissions from the freeway network. Our study area is
12 centered on the I- 710 the I- 110 freeways, over a 22 miles stretch extending from the SPBP to
13 downtown Los Angeles; these two freeways are crossed by four major freeways ( the I- 5, 1- 10,
14 SR- 60, and SR 91). Our study area is presented in Figure 1.
15
16 METHODOLOGY
17 Our approach is similar to the one used in ( 15). An overview of our methodology is presented in
18 Figure 4. It hinges on two types of models: a microscopic traffic simulation model and a model
19 to generate emissions of various pollutants. For this work, we selected respectively
20 TransModeler and EMFAC ( 16, 17).
21
22 Tools
23 Microscopic traffic simulators are now widely used in traffic management, traffic
24 operation/ control, traffic impact studies, and assessments of Intelligent Transportation Systems
25 ( ITS) strategies. They are also starting to be used for evaluating vehicle emissions. They rely on
26 a series of mathematical models, including for example, car- following models and lane changing
27 models. They generate split- second- by- split- second movement of each network vehicle and their
28 interactions, and produce trajectories essential for better modeling emissions.
29 For this study, we selected TransModeler because it is a leading representative of a new
30 generation of microscopic traffic simulators. Moreover, vehicle trajectory in TransModeler can
31 be easily processed to estimate emissions without any additional programming and
32 TransModeler interfaces easily with Geographic Information System ( GIS) data, which is useful
33 for understanding emission dispersion ( this is left for future work).
34 To estimate emissions, we rely on the EMFAC model, which was developed by the
35 California Air Resources Board to calculate emission rates from on- road vehicles, from light-36
duty cars to heavy- duty trucks. EMFAC was chosen for its ability to incorporate results from
37 micro- simulation, such as VMT and VHT and speeds, as well as for incorporating the
38 characteristics of the fleet distribution of both passenger and heavy- duty vehicles. To fully
39 exploit the information provided by microscopic traffic simulation, it would have been better to
11
1 use a microscopic emissions model such as CMEM ( 18), but CMEM cannot calculate PM
2 emissions, and it cannot estimate heavy duty truck emissions after the 2002 model year.
3
4
5 Figure 4. Microscopic- level emission analysis framework
6
7 Data
8 To mathematically represent our network in the traffic simulator, we first extracted coordinates
9 for our basic freeway layout from a GIS layer provided by Caltrans and obtained basic freeway
10 characteristics ( such as the number of lanes and speed limits) from the Performance
11 Measurement System ( PeMS) ( 19). For additional details, we relied on Google Earth.
12 For traffic simulation, traffic OD ( Origins and Destinations) demand inputs were
13 obtained from the 2000 Southern California Association of Governments ( SCAG) traffic study,
14 which is the most comprehensive available for Southern California. To obtain OD demand
15 specifically for our network, sub- area analyses were performed in TransCAD: the sub- area
16 network was extracted from the 2000 SCAG data and OD demand was re- assigned.
17 The OD demands were then adjusted to match traffic flow data every hour as measured
18 from PeMS loop detectors. When traffic flow data from PeMS were missing, we used AADT
19 data provided by Caltrans. For O- D estimation, a path- based algorithm was utilized ( 20), and the
20 commonly- accepted GEH statistic was selected for assessing goodness of fit:
21
( ) 2 ,
0.5( )
GEH M S
M S
−
=
+
22
23
24 where M measures traffic flow and S is simulated traffic flow; both are in vehicles per hour.
25 To obtain an accurate good representation of network traffic conditions, we iterated until
26 the GEH statistic was below 5 ( 10) for at least 50% ( 85%) of our loop detectors.
27 Obtaining reliable simulations of truck activities every business day of 2005 would be
28 very impractical, so after analyzing speed contours and total traffic volumes for 2005, we
12
1 determined that Wednesday, March 9th, 2005 was representative of weekday traffic conditions at
2 the SPBP complex. We therefore focused on obtaining calibrated simulation results for that day.
3 Based on overall traffic and SPBP truck traffic volumes, traffic conditions on our
4 network were classified as follows: 1) morning ( from 7: 00 AM to 9: 00 AM); 2) midday ( from
5 9: 00 AM until 3: 00 PM); and 3) afternoon ( from 3: 00 PM until 7: 00 PM). These three categories
6 also correspond to the time periods adopted by SCAG in its OD estimation procedures. Night
7 traffic was not considered because during March of 2005, the SPBP was operating only from
8 8: 00 AM until 6: 00 PM. We considered the first hour ( 7: 00 to 8: 00 AM) to catch the early SPBP
9 truck traffic; likewise, we kept the last hour ( 6: 00 to 7: 00 PM) to capture the last flow of trucks
10 leaving the SPBP complex for the day.
11 Then for each time period we simulated the busiest and the least busy hour in order to
12 obtain upper and lower bounds for congestion and emissions. A sum of the emissions for the
13 three busiest hours weighted by the number of hours in each period gives an upper bound for
14 traffic emissions during the 12 hours during which port trucks are operating; likewise, the sum of
15 emissions for the three least busy hours weighted by the number of hours in each period ( 2 for
16 the morning period, 6 for midday, and 4 for the afternoon period) gives a lower bound for traffic
17 emissions during the 12 busiest hours of the day
18 Vehicles were categorized as light- duty vehicles ( LDV), light- duty trucks ( LDT),
19 medium- duty trucks ( MDT), heavy- duty trucks ( HDT), and port trucks ( PORTS). Each hour
20 was simulated 30 times in TransModeler to obtain reasonable estimates of mean. Emission
21 estimates and fuel consumption rates were then calculated using EMFAC 2007 for each of the 30
22 trials.
23
24 RESULTS
25 Figure 5 ( top panel) shows the change in NOx emissions compared to the baseline 2005 scenario
26 for port trucks only. Based on information from the drayage truck registry, NOx emissions are
27 reduced between 761 kg [ upper bound] and 726 kg [ lower bound]. Assuming that drayage trucks
28 operate year- round for 5 days a week, this represents between 189 and 198 metric tons for 2009.
29 Annual gains are maximized for 2014, where they ranged between 600 and 621 metric tons
30 respectively. Note that a lot needs to be accomplished between 2009 and 2014, as NOx
31 emissions from Port trucks need to decrease by a factor larger than 3.
32 The other pollutant of interest is PM2.5, which shows a reduction ranging between 19.2
33 and 19.5 kilograms per day under current conditions ( 61.3 to 62.2 kg per day by 2014); see the
34 top panel of Figure 6. This translates into an annual decrease roughly equal to 5 metric tons for
35 2009 and 16 metric tons for 2009, which is quite substantial. Note that the difference between
36 the upper and the lower bounds are much tighter for PM2.5 than for NOx.
37 It is also instructive to look at results in percentage terms to assess the contribution of
38 Port trucks to overall traffic emissions on the freeways of our study area ( see the bottom panels
39 of Figures 5 and 6). We see that emissions changes are very substantial for trucks alone: they
40 should reach approximately 78% for NOx and a remarkable 96.5% for PM2.5 by 2014. A lot
41 needs to be accomplished between now and 2014 ( current reductions are approximately 25% for
42 NOx and 30% for PM2.5); it will require a transformation of the fleet of drayage trucks serving
43 the SPBP complex.
44
13
1
2
3 Figure 5. Change in NOx Emissions
4
5
14
1
2
3 Figure 6. Change in PM2.5 Emissions
4
5
15
1 CONCLUSIONS
2 To our knowledge, this paper is the first to rely on micro- simulation in order to analyze the
3 environmental performance of the Clean Trucks Program, which was launched by the SPBP to
4 reduce emissions from drayage trucks and improve regional air quality. Our results indicate
5 already substantial decreases in the emissions of NOx and PM2.5 ( in the order of 8.5% and 8%
6 respectively). The fleet of drayage trucks will need to undergo a radical transformation,
7 however, in order to achieve the stated goals of the Clean Trucks Program.
8 Although our methodology is general, these results are preliminary. Future work will
9 extend our micro- simulation to account for a change in the hours of operation of the SPBP
10 complex that took place in July of 2005. In addition, it would be of interest to analyze how the
11 Clean Truck Program is affecting pollutant emissions from arterials in the vicinity of the SPBP
12 complex, and to consider weekends, when fewer trucks on the road. Indeed, Lawson et al. [ 23]
13 show that although NOx emissions are lower with fewer commercial trucks on the road, ozone
14 concentrations may actually be higher on weekends. Finally, the health impacts of the Clean
15 Trucks Program and the cost- effectiveness of this program should be explored.
16
17 ACKNOWLEDGEMENTS
18 Support for this research from the University of California Transportation Center ( Award
19 65A016- SA5882) is gratefully acknowledged. We would also like to thank Eric Shen and
20 Shashank Patil from the Port of Long Beach for their very helpful assistance.
21
22
23 REFERENCES
24 1. Port of Long Beach Air Emissions Inventory, 2005. CD- ROM. 2007. The Port of Long
25 Beach.
26 2. Cannon, James S. U. S. Container Ports and Air Pollution: A Perfect Storm. Available at
27 http:// s3. amazonaws. com/ energy- futures. com/ port_ study_ ef. pdf
28 3. Air Resources Board, 2008. Summary of Adverse Impacts of Diesel Particulate Matter.
29 Available at www. arb. ca. gov/ research/ diesel/ diesel_ health_ effects_ summary. pdf
30 4. Pope, C Arden; et al., 2002. Cancer, cardiopulmonary mortality, and long- term exposure to
31 fine particulate air pollution. Journal of the American Medical Association 287: 1132– 1141.
32 PMID 11879110. Available at: http:// jama. ama- assn. org/ cgi/ reprint/ 287/ 9/ 1132.
33 5. Environmental protection agency. How nitrogen oxides affect the way we live and breathe.
34 Available at http:// www. epa. gov/ air/ urbanair/ nox/ noxfldr. pdf.
35 6. Austin, David and Terry Dinan, 2005. Clearing the Air: The Costs and Consequences of
36 Higher CAFE Standards and Increased Gasoline Taxes. Journal of Environmental Economics
37 and Management, Vol. 50, Issue 3, pp. 562- 582.
38 7. Greene, David L., 1998. Why CAFE worked. Center for Transportation Analysis. Energy
39 Policy, Vol. 26, No. 8. Pp. 595- 614.
40 8. Davidson, K., Hallberg A., McCubbin, D., Hubbell, B., 2007. Analysis of PM2.5 Using the
41 Environmental Benefits Mapping and Analysis Program ( BenMAP). Journal of Toxicology
42 and Environmental Health, Part A. Vol. 70. pp. 332- 346.
43 9. Schrooten, L., De Vlieger, I., Lefebre, F., Torfs, R., 2006. Costs and Benefits of an Enhanced
44 Reduction Policy of Particular Matter Exhaust Emissions from road traffic in Flanders.
45 Atmospheric Environment 40, pp 904- 912.
16
1 10. Jansen, H., Denis, C., 1999 A Welfare Cost Assessment of Various Policy Measures to
2 Reduce Pollutant Emissions from Passenger Road Vehicles. Transportation Research, Part A,
3 Volume 4, pp. 379- 396.
4 11. Miraglia, S., El Khouri, G., 2007. Health, Environmental, and Economic Costs from the
5 Use of Stabilized Diesel/ Ethanol mixture in the city of Sao Palo, Brazil. Cadernos De Saude
6 Publica. Vol 23, pp S559- S569.
7 12. Clean Truck Program Fact Sheet. Available at
8 http:// www. portoflosangeles. org/ CAAP/ CTP_ Fact_ Sheet. pdf
9 13. Evaluation of Port Trucks and Possible Mitigation Strategies
10 April 2006. California Air Resource Board. Available at
11 http:// www. arb. ca. gov/ msprog/ onroad/ porttruck/ documents. htm
12 14. Baker, Arley, Court Ruling Allows Mayor Villaraigosa and Port of Los Angeles to Move
13 Forward with October 1 Launch of Clean Truck Program.
14 http:// www. portoflosangeles. org/ newsroom/ 2008_ releases/ news_ 091108ctp. pdf
15 15. Gunwoo Lee, Soyoung ( Iris) You, Stephen G. Ritchie, Jean- Daniel Saphores, Mana
16 Sangkapichai, and R. Jayakrishnan, forthcoming. Environmental Impacts of a Major Freight
17 Corridor: A study of the I- 710 in California. Transportation Research Record.
18 16. TransModeler User’s Guide: Traffic Simulation Software. Caliper. MA., 2008.
19 17. EMFAC 2007 user’s guide. California Air Resource Board, Sacramento, CA, 2006.
20 18. Scora, G., and Barth, M., 2006. Comprehensive Modal Emission Model ( CMEM), Version
21 3.01 User’s Guide. University of California, Riverside, Center for Environmental Research
22 and Technology.
23 19. Freeway Performance Measurement System ( PeMS) https:// pems. eecs. berkeley. edu/
24 Accessed March 20, 2008.
25 20. Choi, K., Jayakrishnan, R., Kim, H., Yang, I., and Lee, J., 2009. Dynamic OD Estimation
26 using Dynamic Traffic Simulation Model in an Urban Arterial Corridor. To be presented at
27 the 88th TRB Annual Meeting ( CD- ROM), Washington, DC.
28 21. Zito, Kelly. Oakland Port Program to Clean up Trucks, Air. Available at
29 http:// www. sfgate. com/ cgi- bin/ article. cgi? f=/ c/ a/ 2009/ 07/ 27/ BALB18TVHV. DTL. Accessed
30 July 30, 2009.
31 22. Port Rolls Out Another 100 Clean Trucks. Port of Long Beach. Available at
32 http:// www. polb. com/ news/ displaynews. asp? NewsID= 576& TargetID= 1
33 23. Chinkin, L. R., Coe, D. L., Funk, T. H., Main, H. H., Roberts, P. T., Ryan, P. A., and Lawson,
34 D. R., 2003. Weekday versus weekend activity patterns for ozone precursor emissions in
35 California's South Coast Air Basin Source. Journal of the Air & Waste Management
36 Association. Vol. 53. Issue: 7 pp: 829 - 843.