10-06-2017, 06:55 AM
THE IMPACT OF TRANSPORTATION CONTROL MEASURES ON EMISSION CONTROL DURING THE OLYMPIC GAMES IN BEIJING
1. INTRODUCTION
1.1 GENERAL
Emissions from vehicles come from the by-products of the combustion process which takes place inside the vehicle when it moves. In addition to this, volatile organic compounds (VOC) also escape through fuel evaporation. As vehicle exhaust systems have improved, evaporative emissions have become a larger component of total vehicle VOC emissions
The combustion process results in emissions of volatile organic compounds (VOC), oxides of nitrogen (NOx), particulate matter (PM), and carbon monoxide (CO), which are released from the tailpipe while a vehicle is operating. Volatile organic compounds (VOC) also escape into the air through fuel evaporation. With today s efficient exhaust emission controls and gasoline formulations, evaporative losses can account for a majority of the total VOC pollution from current model cars especially on hot days.
The rate of emissions varies largely with the vehicular characteristics like speed, acceleration rates etc. Increase in traffic volumes and changes in travel-related characteristics also increase vehicular emissions significantly. The rates of vehicle emissions can be predicted using models. These models can be grouped into three categories, namely, emission factors models, average speed models, and modal models. The prediction capability of most developed models is relatively poor. Therefore, there is a pressing need to improve the predictability of the existing models or to develop new ones with better accuracy. In addition, different models were evaluated for the prediction capability of certain emissions such as carbon monoxide (CO), nitrogen monoxide (NO), nitrogen dioxide (NO2) and hydrocarbons (HC). These models are mainly based on traffic volume, composition, and flow.
[attachment=7076]
2. LITERATURE REVIEW
The literature review is focused on the earlier studies regarding the congestion and its impacts on vehicle emissions
Alam, Habib (2003) evaluated the effects o alternative transport options on congestion and air pollution in Dhaka city. The options examined in the study included the banning of rickshaws and auto rickshaws from the major roads of the city, improvement of bus service and introduction of rail transit system. From the analysis of future demand or transport facilities it was observed that in the next 15 years the travel demand by about 70 percent and congestion by 250 percent. By 2020, about 60 percent o the major roads will become highly congested during the peak hours with a speed less than 5km/hr. In order to keep mobility at the current level it is required to increase the number of buses and cars by about 50 percent and 25 percent, respectively.
Beevers, Carshaw (2005) have successfully implemented a congestion charging scheme in London which measurably reduces traffic flows in central London. From a comprehensive analysis of impact using detailed traffic data combined with Environment Research Group s road traffic emissions model, has obtained a number of results. They have identified that there is a significant reduction in the emissions of NOx and PM10 associated with increase in vehicle speed. There was also evidence that the speed changes, in km/hr are uniform across the whole range of average speed and therefore changes at the slower speed have a disproportionate effect on vehicle emissions. They have also found out that as a result o the scheme in changes in vehicle kilometer are large particularly in the charging zone itself.
Kun, Lei (2007) have found out that in order to develop effective strategies for vehicle emission controls, it is necessary to develop a microscopic simulation platform for estimating vehicle emissions that can capture the vehicle instantaneous modal activities. For this they have developed an integrated microscopic traffic emission simulation platform by using the microscopic traffic simulation model VISSIM and the modal emission model CMEM. A sub network selected from Haiden district of Beijing is used to build a traffic simulation network ,whose traffic and emission conditions are subsequently evaluated. They first analysed the relationship between the instantaneous emission/fuel consumption rate and the instantaneous speed/acceleration. They also analysed and calculate the emissions for a variety o vehicle types in the network. From these studies they have concluded that the optimization of signal timing plan can improve the traffic operations as well as reduce the traffic emissions. But some traffic control strategies which are used to improve the traffic operations my increase traffic emissions.
Smit, Brown, Chan (2008) examined how and to what extent models which are currently used to predict emission and fuel consumption from low traffic include the effects of congestions. They have connected the emissions to congestion by a key factor called driving pattern. Majority of emission and models explicitly incorporate congestion in the modeling process, but for one important family of emission models, namely average speed models, this could not be determined directly. Re-examination of the driving patterns on which three average speed models (COPERT, MOBILE, EMFAC) are based, shows that it is likely that congestion is represented in these patterns. Since emission factors are based on these patterns, this implies that the emission factors used in these emission models also reflect different levels of congestion. Congestion is thus indirectly incorporated in these models. They have also found out that in order to get more accurate (local) emission predictions and to achieve correct application in particular situations, it is important to improve current average speed models by including a congestion algorithm.
3 EMISSIONS AN OVERVIEW
3.1 TYPES OF EMISSIONS
Cars and trucks do not burn all of the fuel that goes into the combustion chamber. In fact, most cars only burn about 85% of the fuel and as the car gets older less and less fuel is burned. It is this unburned fuel that is responsible for a lot of the air pollution problems today. As the amount of vehicles on the roads increases, so does the amount of pollution. There are four main types of air pollution identified by the Environmental Protection Agency(EPA) and all of them are found in vehicle emissions.
The first type of air pollution is carbon monoxide. Carbon monoxide is dangerous because it reduces the amount of oxygen that can be delivered to our body's organs and tissues, which can be life threatening. A mild poisoning of this toxic gas will result in headaches and dizziness. According to EPA road vehicles are responsible for 51% of carbon monoxide pollution. Of that 51%, cars and light trucks make up 87%.
The second type of air pollution is hydrocarbons. The EPA estimates that the average American car driven 12,500 miles per year produces 10,727 lbs of hydrocarbons. These hydrocarbons damages the lung tissue and it also causes respiratory problems. Road vehicles make up 29% of hydrocarbon pollution with cars and light trucks responsible for 88% of that pollution.
Fuel burning at high temperatures will produce a third air pollutant called, nitrogen oxides. 34% of nitrogen oxide pollutants are produced by road vehicles with cars and light trucks responsible for 52% of that and diesel vehicles responsible for 42%. The combination of sunlight, hydrocarbons and nitrogen oxide results in the smog phenomenon. Smog is found in cities all over the world. Hot, humid and sunny summers increase smog..
Particle matter is the final type of air pollution. Haze is the result of this pollution due to actual particle s of solids and liquids found in the air. The majority of particle matter pollution is produced by industrial plants. Vehicles are only responsible for 10% of this air pollution, but of that 10%, diesel vehicles produce 72%.
3.2 FACTORS AFFECTING THE VEHICULAR EMISSIONS
The are various factors which influence the vehicular emissions. They are the following:
3.2.1 Traffic Parameters
Traffic parameters such as traffic volume, traffic composition, average speed of the flow etc. come under this category. Each of the above parameters significantly affects the pollution levels. As the traffic volume increases the amount of total pollutants will be increased. Traffic composition also significantly affects the pollution levels. Emission rates vary from vehicle to vehicle. Hence, the same amount of traffic volume with different vehicular composition produces different amounts of pollutants. Hence vehicle composition details are to be collected accurately on to a prescribed performa of volume count study. Speed significantly affects the amount of pollutants. As the speed increases the rate of fuel consumption will be decreased, which results in the reduction of (Suspended Particulate Matter) SPM levels. Nitrogen oxides are also vary with the speed. As the speed is affecting all these pollutants it is needed to measure the speed accurately.
3.2.2 Roadway Parameters
Roadway parameters those affecting the pollution levels are carriageway width, lateral clearance, medians, and shoulder width etc. As the road width increases the maneuverability to movement of the vehicles will be increased, which results in the reduction of pollutants. Medians reduce the obstruction caused by the opposing vehicles, which results in the reduction of number of accelerations and decelerations. So the fuel consumption will be minimized. Hence the emission levels will be decreased.
3.2.3 Environmental Parameters
Air quality also depends upon the environmental factors such as the ambient air temperature, humidity, wind speed and wind direction etc. The motor vehicle exhaust emissions are strongly sensitive to ambient temperature. In cold climate they exhaust more emissions. Wind speed and direction will not influence the emissions from vehicles. But they are very useful for air quality modeling. As the wind speed increases the rate of dispersion will be increased, which results in the reduction of pollution concentrations. Similarly wind direction also plays an important role in the pollutant concentration reduction.
3.2.4Vehicle Parameters
The age of the vehicle, its condition and servicing frequency, type of engine (2-stroke, 4-stroke), VMT (Vehicle Miles Traveled) come under this division. The older vehicles will emit more emission than a newer one, if they are not maintained properly. Vehicles with 4-stroke engine would produce lesser emissions than 2-stroke engine. Similarly, vehicles with catalytic converter will emit less pollutant.
3.2.5Type of Fuel
The type of fuel and its composition will also affects the emissions. For example, a gasoline-fueled car may emit high CO emissions than a truck using diesel as fuel. Also the Particulate Matter is mainly coming from the diesel fueled vehicles. Likewise, the composition of fuel in terms of its octane number, cetane number, sulfur content, aromatics and olefins also influence the pollution levels.
3.3PERMISSIBLE LIMITS
The permissible limits for various types of emissions are given by World Health Organisation (WHO) , National ambient air quality standards (NAAQS) for India & Association of Indian Automobile Manufacturers are shown below.
Table1: Maximum permissible limits ( g/m3) of pollutants in the air set by the WHO
Pollutant Time-weighted average Average time
Sulphurdioxide 500
350
100 to 150 a
40 to 60a
10 minutes
1 hour
24 hour
1 year
Carbon monoxide 30
10
1 hour
8 hours
Nitrogen dioxide (WHO 1987, 1977) 400
150
1 hour
24 hours
Ozone (WHO 1987, 1978) 150 to 200
100 to 120
1 hour
8 hours
Total SPM (Suspended Particulate Matter) 150 to 230 a
60 to 90 a
24 hours
1 hour
Source: WHO,1992
Table2: National ambient air quality standards (NAAQS) for India (in g/m3)
Pollutant and time-weighted average Industrial area Residential, rural, and other areas Sensitive area Methods of measurement
Sulphur dioxide
Annual average
24 hours
80.00
120.00 60.00
80.00 15.00
30.00 Improved West and Gaeke method
Ultraviolet fluorescence
Oxides of Nitrogen
Annual average
24 hours
80.00
120.00 60.00
80.00 15.00
30.00 Jacob and hochheiser modified (Na-Arsenite) method
Gas-phase chemilluminescene
Suspended particulate matter
Annual average
24 hours
360.00
500.00
140.00
200.00
70.00
100.00 High-volume sampling (average flow rate not less than 1.1 m3 per minute)
Respirable particulate matter (size less than 10 gm)
Annual average
24 hours
120.00
150.00
60.00
100.00
50.00
75.00
Respirable particulate matter sampler
Lead
Annual average
24 hours
1.00
1.50 0.75
1.00 0.500
0.750
Atomic absorption spectrometry after sampling using EPM 2000 or an equivalent filter paper
Carbon monoxide
8hours
1 hour 5.00
10.00 2.00
4.00 1.00
2.00 Non-dispersive infrared spectroscopy
Ammonia
Annual average
24 hours -- 400.00
100.00 -- --
Source: Central Pollution Control Board,
Table 3: Mass emission standards for diesel-driven vehicles (in g/kwh)
1 April 1992 1 April 1996 1 April 2000 Euro II Euro II
Gross vehicle weight >3.5 tonnes
Carbon monoxide 14.00 11.20 4.50 4.00 2.00
Hydrocarbon 3.50 2.40 1.10 1.10 0.60
Oxides of nitrogen 18.00 14.40 8.00 7.00 5.00
Particulate matter for >85 KW 0.00 0.00 0.61 0.15 0.10
Particulate matter for <85 KW 0.00 0.00 0.61 0.15 0.10
1 April 1992 (g/kWh) 1 April 1996 (g/km) 1 April 2000 (g/km)
Gross vehicle weight <3.5 tonnes
Carbon monoxide 14.00 5.0-9.0 2.72-6.90
Hydrocarbon 3.50 -- --
Oxides of nitrogen 18.00 2.4-4.0 0.97 1.70
Particulate matter 0.00 -- 0.14-0.25
Source: Association of Indian Automobile Manufacturers,1999
3.4 COMPARISON OF EMISSION RATES OF PETROL ENGINES AND DIESEL ENGINES
Diesel engines consume around 30% less fuel than petrol engines and this results in much lesser carbon dioxide emissions. The diesel engines produce virtually no carbon monoxide and are much safer than petrol engines. Tests done on car emissions revealed that Nitrogen Oxides are higher in a new diesel engine when compared to a new petrol engine. But by the time they cover 50,000 miles or so, they are the same and after that the petrol engine produces more oxides than the diesel engine. Hydrocarbon emissions contained in petrol engine emissions are considerably more than that in diesel engine emissions. However, diesel is certainly more dangerous from the point of view of Suspended Particulate Matter (SPM). SPM refers to solid particles suspended in open air, such as soot generated by combustion of various fuels. They might cause respiratory problems because of their tendency to deposit themselves in the lungs. Though much has been done to improve the fuel efficiency and reduce emissions from the petrol engine, still more needs to be done.
3.5 EMISSION TESTING
Emissions testing is a crucial part of automobile testing to ensure that the vehicle is not polluting the environment. In some states there are strict guidelines and rules to follow for emission testing that are inescapable
The aim of the emission test is to reduce the production of harmful gases and pollutants so that the air quality can be improved. The aim of the vehicle emission control is to reduce the pollutants emitted by vehicle.
The pollutants to be checked in Vehicle Emissions Testing are:
The level of hydrocarbon in emissions testing is checked as excess of these causes smog and also leads to cancer. Hydrocarbon is produced by the unburnt or partially burnt fuel. As per the specification of engine there is the regulation for non-methane hydrocarbons and total hydrocarbons.
Carbon monoxide released because of the partial combustion of fuel is then checked. When inhaled it reduces the oxygen carrying capacity of blood and excess of CO in body is fatal. During vehicle emission testing the level of CO emitted by your vehicle is checked.
Nitrogen oxide which is the resultant of combination of nitrogen in the air with the oxygen at high temperature and pressure is also chcked. Nitrogen oxide causes acid rain and smog. If extra amount of NO and NO2is inhaled then it can lead to less resistance towards respiratory infections.
Oxygen is the remaining unburned gas during combustion. Although Oxygen is not bad but is checked to find out whether you have cheated in the emissions testing or not. Also with the percentage of O2in the emission test determines the fuel ratio of engine while running. This test is also done by gas analyser.
Testing procedure
To perform the vehicle emission test, the vehicle will be accelerated on dynamometer which is a treadmill like machine. Then a probe is placed in the tailpipe. The readings on pollutants is then checked in the emission testing on to computer. The readings are then compared to the standard values and its year.
3.6 EMISSION TEST TYPES
Broadly we can categories the emissions test in the following steps
Acceleration Simulation Mode
Vehicle Model Year 1995 or older than that uses the dynamometer. Vehicle tires are set on this machine which then applied load and resistance on the vehicle to create the road conditions. This consists of two phases. These include
50/15 mode - In this phase of 50/15 mode, constant speed of 15 mph is maintained and vehicle's 50% horsepower is used to stimulate dynamometer.
25/25 mode - In second phase the speed is increased to 25 mph and horsepower is reduced to 25%.
On-Board Diagnostics
Performance of the vehicle is also monitored on the in-build computer systems. This include fuel metering system for emissions test, ignition system, emission control equipment. Any dis-balance shown in this can lead to the failure in the vehicle emission test.
Two Speed Idle
In this a probe is inserted into the tail piece of the vehicle to find out different emitted gases. This probe is attached to the computer that records the reading and compare these with standard values in emissions testing. This is also done in two phases.
1. high speed test (2200-2800 RPMs)
2. tested at idle (350-1200 RPMs.)
3.7 CONGESTION AND AIR POLLUTION
Road transport has grown rapidly over the last decades. As a consequence, traffic-related emissions and air pollution have increased substantially despite the increasing use of abatement measures such as catalytic converters. Due to the growth in road traffic, congested traffic conditions have become increasingly common and severe worldwide, particularly in major cities. A strong further increase in the demand for transport and congestion is projected around the world.
Traffic congestion has been indicated as being one of the main contributors to environmental impacts in urban areas. Congestion is the deterioration of the quality of traffic flow on a network element or in an entire road network due to increased travel demand and/or reduced capacity for traffic movement, which may be observed in terms of different interrelated measures including, but not limited to, increased traffic density, increased travel times, increased delays, lower travel speeds and increased queuing
3.8 DRIVING CYCLES AND EMISSIONS
Driving cycle is a sequence of vehicle operating conditions. It includes idle, acceleration, steady state and deceleration. There are number of studies relating driving characteristics and vehicular emissions using on board measurement and remote sensing techniques. Conducting on-board measurements would be expensive and difficult to collect large number of vehicle samples. In the case of remote sensing techniques, it can be used only at a particular location and can capture instantaneous emissions alone. Nesamani, Subramanian et al.,(2006)in Chennai estimated the emissions for different driving patterns using International Vehicle Emission (IVE) model. It is a computer model designed to estimate emissions from motor vehicles. It was jointly developed by International Sustainable Research Centre and the University of California at Riverside. The general inputs to the model are fleet characteristics, vehicle activity and emission factor based on local conditions. From the results it was observed that the emission rates vary significantly from one class of road to another and large effects was on local streets. This was mainly due to local street spending higher percentage of time in acceleration mode with low average speed. Both arterial and sub-arterial roads have almost similar emission rate except NOx which was higher in arterial. This could be attributed to sharp acceleration rate in arterial streets. CO2 emission was lower in arterial street and higher in local street.
4 ASSESSING EFFECT OF TRAFFIC SIGNAL CONTROL STRATEGIES ON VEHICLE EMISSIONS
4.1 GENERAL
The vehicle pollution and traffic noise have worsened due to the rapid increase of vehicles in Beijing, China. Several experts and scholars have actively carried out the research on vehicle emissions under different traffic management and control strategies. This paper is intended to evaluate the impact of different traffic control strategies on vehicle emissions using the microscopic traffic simulation model VISSIM.
4.2 EMISSION CHARACTERISTICS UNDER SIGNAL COORDINATION AND NON-COORDINATION
4.2.1 Data collection
The real world data were collected using portable emission measurement system (PEMS). Qianmen Avenue (a road section from Hepingmen to Chongwenmen) in Beijing was selected as coordinated signal controlled road and Garden East Road North Taipingzhuang Road (from Zhixin Bridge to Anzhen Bridge) as non-coordinated signal controlled road. Then, the emission levels of a vehicle under these two types of controls were compared (i.e., different traffic management and control strategies). These routes were selected mainly according to the average speed, while the traffic flow, congestion level, and traffic flow composition from morning to afternoon of the road were considered as the secondary criteria .
A 2000-year-manufactured Jetta CI with the mileage of 0.12 million kilometers has been selected for the experiment subject. The experiment was implemented from 9:00 a.m. to 11:00 a.m. on July 25 26, 2006. A total of 3, 133 valid emission records for the signal coordinated road on July 25 and 5, 316 valid emission records for the signal non-coordinated road on July 26, 2006 were collected. In addition, traffic surveys were conducted for two hours during the experiment period, including traffic compositions, actual road conditions, traffic flow data of both directions, number of vehicles entering into and leaving off intersections, existing signal timings, and so on. These data served s the input parameters for the microscopic traffic simulation model VISSIM to simulate the traffic operations of coordinated signal controlled road.
4.2.2 Characteristics analysis of emission data
By integrating the electronic map of Beijing city and the real-time emission data, this study established an emission analysis platform, in which MapInfo was implemented for GIS analysis and the Map Basic was selected for the second development tool.
4.2.2.1 Comparison of emission factors
Vehicle emission factor indicates emissions per unit of distance, hour or fuel consumption, which is measured in g/km, g/h, or g/kg (fuel). It is an important indicator for measuring the level of vehicle emissions, as well as the basis and foundation for studying the emission control strategy. Fig. 1 illustrates the comparison of emission factors on coordinated versus non-coordinated signal controlled roads. Compared with emission factors on the non-coordinated signal controlled road, the HC and CO emission factors on the coordinated signal controlled road have decreased by 50% and 30%, respectively; however, NOx has increased by 10%. This result shows that using signal coordination can effectively reduce the HC, CO emissions, but increase the NOx pollution.
Fig1. Comparison of emission factors on coordinated versus
non-coordinated signal controlled roads
4.2.2.2 Comparison of emissions based on driving cycles
Analyzing emissions based on different driving cycles is an effective way to study the microscopic characteristics of vehicle emissions. Fig. 2, 3, 4 shows the comparison of emissions under different driving cycles based on coordinated versus non-coordinated signal controlled roads.
Fig2. Comparison of NOx emissions under different driving cycles
Fig3. comparison of HC emissions under different driving cycles
Fig4. Comparison of CO emissions under different driving cycles
As illustrated in Fig. 2, 3, 4 the CO and HC emission rates under different driving cycles on coordinated signal controlled road are lower than those on non-coordinated signal controlled road, but the CO emissions increased slightly under idling. The NOx emission rate under different driving cycles on the coordinated signal controlled road is similar to the one on the non-coordinated signal controlled road. From
the perspective of emissions, all emission rates under acceleration are very high, and NOx emission under deceleration should not be overlooked.
4.2.2.3 Analysis of emission characteristics based on different speed intervals
The quantified relationship between traffic and vehicle emission can be generated by investigating the trend of emission factors under different speed intervals, which will provide data to support the development of the traffic management and control strategies for reducing vehicle emissions. The trend of emissions under different speed intervals is shown in Fig. 5, 6, 7.
Fig5. The trend of NOx emissions under different speed intervals
Fig6. The trend of HC emissions under different speed intervals
Fig7. The trend of CO emissions under different speed intervals
From figures 5, 6, 7 it can be seen that the HC and CO emission factors show a consistent trend of declining as the speed increases. The emissions are very low after the speed reaches 20 km/h, which indicates that the HC and CO emissions can be controlled under low speeds. NOx has just minor changes after the speed reaches 20 km/h, which illustrates that an optimization of speed has no significant effect on reducing NOx. CO and HC emission rates under different speed intervals on coordinated signal controlled road are lower than those on non-coordinated signal controlled road, but NOx has an opposite trend. The result indicates that using signal coordination can effectively reduce the HC, CO emissions, but NOx emission will become worse, which is consistent with the earlier conclusions.
4.3 ANALYSIS OF EMISSIONS BASED ON MICROSIMULATION UNDER SIGNAL COORDINATION
4.3.1 Development of microscopic traffic simulation platform
VISSIM is a microscopic model based on behavior and time-step analysis, which can simulate the traffic operation of various types of roads, such as the urban road and highway. In order to compare and analyze the impact of emissions under different traffic management and control strategies an integrated microscopic simulation platform of traffic emissions has been developed by integrating the microscopic traffic simulation model of VISSIM and the VSP based approach of emission modeling, which is used to simulate the network of Qianmen coordinated signal controlled road based on traffic surveys.
By simulating Qianmen road in Beijing with the VISSIM model, the velocity(v) and acceleration (a) can be output on a second-by-second basis, and the instantaneous VSP can be calculated according to equation (1). Then selecting 1 as the interval unit, the VSP value belonging to interval [-30, 30] can be designated into different VSP Bins using equation (2). When the emission rates are finally obtained corresponding to different VSP intervals and the VSP frequency distribution, the total vehicle emissions of certain trip can be calculated using equation (3)
VSP = v (1.1a + 0.132) + 0.000302 v3 (1)
for all VSP [n 0.5,n + 0.5], VSPBin = n (2)
where n is n integer
Q = f(p) R(pi) (3)
where Q is the total vehicle emissions of a certain trip; i is the travel time (i = 1, 2, 3 .n); pi is the VSP value of the ith second; f(pi) is the VSP distribution of the ith second when VSP equals Pi; R(pi) is the emission rate of the ith second when VSP equals pi
4.3.2 Impact of different traffic management and control strategies on emissions
4.3.2.1 Impact of different signal timings on emissions
Different traffic signal control strategies can result in different microscopic driving cycles of vehicles in the network, thus leading to different levels of emission pollution. In order to analyze the impact of emissions under different signal timings, the network of Qianmen coordinated signal controlled road is simulated based on the actual surveys of the signal timing plans of the entire road, and then the signal timing plans of simulated network of Qianmen are transferred into non-coordinated signal controlled for comparison. Furthermore, the vehicle emissions of coordinated and non-coordinated signal controlled roads can be calculated using the VSP-based approach of the emission modeling. Two simulation results are converted to emission factors, as shown in Table 4.
Table 4. Comparison of calculation results on coordinated versus non- coordinated signal controlled roads
Travel time(s) Average speed (km/hr) NOx (g/km) HC(g/km) CO(g/km) CO2
Signal coordination 466 23.12 1.061 0.0529 2.559 122.148
No signal coordination 491 21.87 1.233 0.0588 2.985 134.566
Trend +5% -5% +14% +10% +14% +9%
It can be seen from table 4 that the emission factors increase by bout 10%, the travel time increases by 5%, and the average speed drops 5% after the coordinated signal controlled roads has been converted to non-coordinated road. This indicates that the optimization of the traffic signal control plan can effectively improve the operation of road traffic and reduce vehicle emission pollution.
4.3.2.2 Impact of different traffic demands on emissions
Different traffic demands directly influence the vehicles speed and driving behavior on roads, and thus affect vehicle exhaust emissions. In order to compare the impact of different traffic demands on the emission, the network of Qianmen coordinated signal controlled road with increased versus decreased demands by 20% flows were simulated respectively. Then, the vehicle emissions of the coordinated controlled road with increased and decreased demands by 20% were calculated with the VSP based approach of emission modeling. The two simulation results were converted to emission factors as shown in table 5
Table 5. Comparison of simulation results by increased versus decreased traffic demand by 20% on coordinated signal controlled roads
Travel time(s) Average speed (km/hr) NOx(g/km) HC(g/km) CO(g/km) CO2
(g/km)
Trend -22.53% +28.29% -78.32% -62.19% -69.01% -59.08%
Decreasing 20% flow 361 29.66 0.23 0.020 0.793 49.978
Baseline flow 466 23.12 1.061 0.0529 2.559 122.148
Increasing 20% flow 514 20.93 1.466 0.063 3.629 150.278
Trend +10.3% -9.47% +36.29% +19.09% +41.81% +23.03%
From table 5, it can be seen that when traffic demands increase by 20%, the travel time increases 48 seconds and the average speed decreases by 9.47%. The emission factors are significantly increased. CO and NOx increased by 41.81% and 36.29% respectively. When the traffic demands decrease by 20%, the travel time reduces 105 second as well as the average speed improves by 28.29%. The emission factors are significantly decreased respectively, as much as over 50%, and the maximum reduction is for NOx, up to 78.32%. This indicates that the decreasing traffic demands can significantly reduce vehicle emissions and optimize traffic operations. Therefore, when selecting traffic management and control strategies to reduce emissions on roads, traffic manager can consider implementing route guidance and traffic demand management strategies on roads with high emissions.
4.4 CONCLUSIONS
The real world emissions under signal coordination and non-coordination in Beijing were collected and compared and the emission levels and distribution characteristics under these two control strategies were analyzed. It was found that in comparison with the emission factors of non-coordinated road, the HC and CO emission factors of coordinated signal controlled road decreased by 50% and 30%, but NOx increased by 10%. The results showed that using signal coordination the HC and CO emissions can be effectively reduced. All emission rates under acceleration are very high. Therefore, the emission under acceleration should be strictly controlled. In order to analyze the impact of different traffic management and control strategies on emissions an integrated microscopic simulation platform of traffic emissions has been developed by integrating the microscopic simulation model of VISSIM and the VSP based approach of the emission modeling, which was used to evaluate the impact of two traffic control strategies, alternative signal timing and different traffic demands on emissions. It was found that the emission factors increased by about 10%, the travel time increased by 5%, and the average speed drops by 5% after the coordinated signal controlled road was converted t non-coordinated road. The result indicates that the optimization of the traffic signal control plan can effectively improve the operation of road traffic and reduce vehicle pollution. When the traffic demand decreased by 20%, the average speed improved by 28.29%, and the emission factors significantly decreased respectively, as much as over 50%. The maximum reduction was for NOx, up to 78.32% which indicates that decreasing traffic demands can significantly reduce vehicle emissions. Therefore, it has been concluded that when selecting traffic management and control strategies to control emissions on roads, traffic managers can consider implementing the optimization of signal timing and at the same time try to control traffic demands.
5 THE IMPACT OF TRANSPORTATION CONTROL MEASURES ON EMISSION REDUCTIONS DURING THE 2008 OLYMPIC GAMES IN BEIJING, CHINA
5.1 GENERAL
Since 2001, when Beijing won the bid to host the 2008 Olympics, air quality for the competition was a major concern. Beijing held the 2008 Olympic Games from August 8 to August 24, 2008. In June 2008, the local government of Beijing promulgated temporary transportation control measures to be implemented during the event. The study periods were classified as before and during the Beijing Olympic Games to evaluate the effect of temporary transportation control measures adopted throughout the event. Field traffic flow monitoring and a calibrated transportation simulation platform based on geographic information system (GIS) have been applied in this study to achieve road network activity and operational speed with a 1 km x 1 km resolution. A bottom-up methodology was applied to develop motor vehicle emission inventories before and during the Games. The effectiveness of transportation control measures has been evaluated by comparing these grid-based emission inventories. The emission reduction benefit has also been evaluated by the emission intensity derived from reverse modeling of curbside air quality monitoring results.
5.2. METHODOLOGY
5.2.1. Traffic modeling
The urban area of Beijing is divided into 2055 grid cells, each of which is 1 km x 1 km. An urban transportation simulation platform based on TransCAD has been applied in this study to estimate the vehicle kilometers traveled (VKT) and the average speed in these cells. Annual traffic flow monitoring, travel demand management policies as well as personal trip mode investigation has been applied as basic inputs to the simulation platform. Simulated road page link activities are further transformed into grid-based activities. Motor vehicles identified in traffic volume monitoring and modeling have been classified into six categories including light duty vehicles (LDV), taxis, light-duty trucks (LDT), heavy duty trucks (HDT), buses and heavy-duty vehicles (HDV). Daily and peak hour VKT for each vehicle category and average speed have been estimated by the system. Information provided in each grid cell includes geographic coordinates, road length, average speed and daily VKT for each vehicle category.
The Beijing Transportation Research Center conducted a special investigation on continuous traffic monitoring on 132 road links in urban Beijing before and during the Olympic Games to estimate the variation in transportation activity on a geographic scale. The results of this investigation, as well as routine spatial and temporal origin destination flow investigations for residents in Beijing, were used to calibrate the transportation control scenarios in the simulation platform.
5.2.2. Emission factor development
MOBILE5B- China, a localized model based on US Environmental Protection Agency (USEPA). MOBILE5B and PART % was applied in this study to estimate speed- dependant vehicle emission factors. A special module has been developed for estimating the effect of transportation control measures on the average vehicle emission factor. The emission factor was mainly influenced by the control measures adopted during the Games in two ways.
First, is the vehicle operation parameter such as speed which is one of the most important factors affecting emissions. Second, is the fleet technology configuration. As yellow labeled vehicles were banned throughout Beijing and strict operational restricts were also placed on trucks. Table 6 lists the influence of the detailed transportation control measures on the major factors related to traffic emissions.
Table6. Influence of transportation control measures on the main factors o traffic emissions
Strategies Factors influencing traffic emissions
Travel restriction by odd-even license Speed, Total activity
Truck operational restrictions Fleet composition, Total activity
Removal of yellow labeled vehicles Fleet composition, Total activity
The overall vehicle fleet change has been taken into consideration to evaluate the effect of vehicle operation parameters and fleet configurations on average emission factors. The registration distribution of yellow labeled vehicles in Beijing has been investigated within the local Inspection and Maintenance database. Green Environmental labels have been issued in Beijing to gasoline vehicles meeting at least Euro I and diesel vehicles meeting at least Euro II requirements. Other vehicles were issued yellow environmental labels. The yellow labeled vehicles which are usually referred to as high emitting vehicles were mostly banned while the Olympic Games were underway. As a result of this, the average age of the light duty fleet and heavy duty fleet operating in urban area of Beijing during the Olympics was reduced by 0.48 and 3.55 years respectively.
By the end of 2007, about 3, 47000 yellow labeled vehicles were still in use in Beijing which was 11% of the total fleet. The fleet composition of HDT has been changed the most because diesel trucks can only get a label if they can meet at least Euro II emission standards in Beijing. Figure 8 shows the change of speed dependant average NOx emission factor for HDT, up to 80km/h before and during the Beijing Olympics. It should be noted that not all the trucks with yellow environmental labels were banned. The municipal government has issued special passes to some yellow labeled trucks to guarantee the progress of important project
Fig 8. Speed dependent average NOx emission factors for HDT before and during the Olympics
. The speed-dependent emission factors for Volatile Organic Compound (VOC) , Carbon monoxide (CO) and Nitrogen oxide (NOx) for each vehicle class are calculated within MOBILE5B-China. Figure9 shows the relationship between the emission correction factor and the average speed for LDV relative to the baseline speed of 30 km/h. It shows that the VOC and CO emission factors have a significant relationship with speed, and generally decrease with increasing speeds. The NOx emission factors, which are less sensitive to speed changes, follow a parabolic path and tend to decrease to 35 km/h and then increase.
Fig 9. Speed correction factor for emission factor for LDV
Real-world vehicle emission measurements on a fixed route have been carried out with PEMS in Beijing before and during the Games. PEMS measurements on 5 light-duty gasoline vehicles showed that the average reduction of their CO, HC and NOx emission factors were 26.8%, 28.7% and 0.8%, respectively, during the Games as the trip average speed increased from 26.8 km/h to 34.6 km/h. This agrees with the modeling output in this study. As shown in Figure 9, the modeled CO, VOC and NOx emission factor reductions for LDV for the same speed increment are 21.8%, 21.1% and 1.1%, respectively. The real-world measurements have also shown that the CO and VOC are sensitive to the change of vehicle speed, but the NOx is not sensitive to the change of vehicle speed. The similarity in the decreased level of the three air pollutants from both measurements and modeling work confirms the good performance of the MOBILE5B-China emission factor model.
5.2.3. Emission reduction benefit evaluation
5.2.3.1. Grid-based emission inventory development
In order to quantify the emission reduction benefit of the transportation control measures adopted, a bottom-up methodology has been applied in this study to build emission inventories for the urban areas before and during the Games. Grid-based vehicular speeds from traffic modeling are used to obtain emission factors for each vehicle type in each grid cell. This approach is a physical representation of realistic traffic conditions in each grid cell of the urban area. Emissions in each grid cell were calculated by multiplying the VKT and speed-dependent emission factors for the six vehicle categories:
Qpc = EF pi, c x VKT i,c (1)
where Qpc is the emissions of pollutant P in cell c (g) ; EF pi, c is the speed- dependant emission factor of pollutant P or vehicle category i in cell c (g/km) ; VKT i,c is the total vehicle kilometers traveled of vehicle category I in cell c (km). The emissions were summed over all grid cells to obtain total urban emissions.
5.2.3.2. Emission reduction benefit derived from curbside monitoring
For this two sampling sites were set up on the sidewalks of the North 4th Ring Road, about 26mfrom the road centerline. Measurements were taken at 2 m above the ground. The detailed location of the sites is shown in figure 10 , which is approximately 7kmfrom the Olympic Park. The monitoring was performed before the Olympics (June 22 - July 5, 2008) and during the Olympics (July 28 - August 22, 2008). Hourly traffic volume was counted throughout the monitoring periods based on continuous video taping from the crossover above the road. Hourly data for CO and NOx concentrations were provided by Thermo 48i and 42i, respectively, at both sites along the sidewalks. A portable weather station was also set up near the air quality monitoring instruments to record second-by-second ambient temperature, wind speed, wind direction, humidity and atmospheric pressure.
Fig 10. Location of curbside monitoring sites
Line-source dispersion models are usually used for the calculation of air quality based on known theoretical relationships between emissions, meteorology parameters and air pollutant concentrations. Street pollution models like OSPM and CALINE4 have been combined with curbside air quality measurements to estimate in-situ emission intensity or emission factors. For the dispersion of non-reactive or slowly reactive vehicle exhaust gases at short distances, chemical transformations can be disregarded. Equation (2) may be used for calculations of hourly emissions from traffic, provided that both receptor and background concentrations are available from in-field monitoring.
Q = Cr Cbg
Fd
where Cr and Cbg are the receptor concentration with the contribution from road traffic and the background concentration from sources other than motor vehicles in the specified road link, Q is the traffic emissions at the road link, and Fd is a factor describing dispersion processes under certain meteorological conditions. In this study, only those monitoring data with the wind direction perpendicular or nearly perpendicular to the road were selected for CALINE4 simulation. Thus, in this case the difference between the monitored concentrations at both sides of the road can be used as the indicator of the direct contribution from the road traffic. The dispersion factor Fd is given by a line-source pollution model, in our case, CALINE4. The model was originally developed for calculating CO concentrations (in ppm) from a road. A scaling factor has been applied in this study to adapt CALINE4 for NOx. The inputs of the model include road geometry, meteorological parameters (wind speed, wind direction and its standard deviation, temperature, atmospheric stability class and mixing height), background concentration, and receptor information. The model also requires the input of road page link emission intensity, which becomes the target for reverse application of CALINE4 in this study.
Considering the possible systematic errors in the line-source dispersion model, the reduction ratio instead of the absolute value of derived emission intensity has been used in this study for further analysis. The emission reduction ratio derived from curbside monitoring was compared with those achieved from the bottom-up inventory modeling to show the emission reduction benefits of the transportation control measures.
5.3 Results and discussion
5.3.1. Traffic flow effect
Figure 11 shows the reduction ratio in daily traffic flows for each vehicle category from continuous traffic monitoring at 132 road links before and during the Olympics. Vehicle categories other than taxis and buses show a significant decrease in traffic flow due to the temporary traffic control measures. HDT had the highest reduction
ratio because of the strict traffic intervention measures on its operational area. Increasing bus and taxi flow has been observed at some road links inside the downtown area or close to the Olympic venues.
It can be seen from figure 11 that the daily flow of LDV and HDT has been reduced by 30.1% and 53.8%, respectively, during the Games. The flow of taxis and buses has been increased by 29.6% and 9.8%, respectively. The monitoring sites are close to the main Olympic venues, which could be the reason for higher public transportation demand. Two Olympic-specific lanes and increasing trips to the Olympic Park have also counteracted the effect of traffic control.
Fig 11. Daily traffic flow reduction ratio monitored during the Olympics. Line within box: Median value; Top line of box: third quartile; Bottom line of box: first quartile; Top outlier: maximum; Bottom outlier: minimum
Figure 12 shows the weekday hourly flow monitored at the North 4th Ring Road. Hourly flow data has shown that most traffic reduction was achieved during the daytime. Traffic flow between midnight and 3:00 shows a temporarily slight increase because the municipal government issued a supplemental rule suspending license-based traffic control measures during these three hours.
Fig 12. Variation of weekday traffic flow at the North 4th Ring road before and during the Olympics
Daily VKT and average speed distributions derived from the transportation simulation platform are shown in figures 13 and 14. Total urban VKT has been reduced by 32.0% during the Games. The average speed weighed by grid VKT has been increased from 25 km/h to 37 km/h during the Games. It has been found that a significant proportion of the grids are located in the speed bins lower than 30 km /h in downtown areas before the Games.
Fig13. Daily VKT for urban Beijing before (a) and during (b) the Olympics
Fig 14. Daily average speed for urban Beijing before (a) and during the Olympics
5.3.2. Bottom-up emission inventory
Grid-based vehicle activities has been acquired from the calibrated transportation simulation platform and the speed dependent emission factors has been acquired from MOBILE5b-China. Based on these data the bottom-up methodology described in equation.(1) was used to develop mobile source emission inventories. The grid-based vehicle emission inventories for CO, VOC, NOx and PM10 in the Beijing urban area before and during the Olympic Games are shown in figures 15-18, respectively. From the figures it has been observed that the distribution of vehicle emissions in Beijing's urban area are more concentrated in the urban core area with higher population density and travel demand.
Fig 15. Grid-based daily vehicle emission inventory for CO in urban Beijing before (a) and during (b) the Olympics
Fig16. Grid-based daily vehicle emission inventory for VOC in urban Beijing before (a) and during (b) the Olympics
Fig17. Grid-based daily vehicle emission inventory for NOx in urban Beijing before (a) and during (b) the Olympics.
Fig 18. Grid-based daily vehicle emission inventory for PM10 in urban Beijing before (a) and during (b) the Olympics.
The radial arterials and exit highways with heavy traffic also contributed to high emission intensity in these grids. Grid-based emissions can be aggregated to calculate the total vehicular emissions. Table 7 shows a comparison between the total urban vehicle emissions before and during the Games.
Table 7. Daily urban motor vehicle emissions of Beijing in tons before and during the 2008 Olympics
VOC CO NOx PM10
Before the Olympics 371 2993 282 15.9
During the Olympics 165 1293 153 7.7
Emission reduction 55.5% 56.8% 45.7% 51.6%
Improved traffic efficiency and reduced fleet average age helped Beijing achieve a higher mobile source emission reduction than lower VKT. More significant reductions have been found in VOC and CO emissions because they are more sensitive to average speed as shown in figure9. Based on the daily grid-based inventory before the Olympics, trucks have contributed 21.6% and 40.3% of total motor vehicle NOx and PM10 emissions in urban Beijing, respectively. Because of the higher contributions of trucks to PM10 emissions, strict controls on them during the Olympic Games achieved a higher reduction in total PM10 emissions than NOx emissions.
5.3.3. Emission intensity at the North 4th Ring Road
Figure 19 shows hourly average downwind CO and NOx concentration monitored at the North 4th Ring Road site before and during the Olympics. Receptor CO and NOx concentration was reduced by 31.2 7.4% and 34.9 13.9%, respectively. The largest CO reduction was observed from 16:00 to 19:00, one of the two rush hours. The largest NOx reduction was observed during the midnight. Long before the beginning of the Olympics, the Beijing Traffic Management Bureau was already enforcing a routine traffic control regulation that trucks were only allowed to operate inside the 4th Ring Road between 23:00 and 6:00. Trucks were the most important NOx and PM contributor at night before the Olympics. The banning of yellow labeled vehicles during the Olympics mostly affected the traffic flow of diesel trucks and average emission factors among all vehicle categories, especially at night. This resulted in a higher reduction in NOx than CO during the midnight.
Fig 19. Hourly average downwind CO (a) and NOx (b) concentrations monitored at the North 4th Ring Road before and during the Olympics.
The Beijing Environmental Protection Bureau (EPB) reports air pollution level of PM10, SO2 and NO2 in Beijing through its public domain .As the Beijing EPB does not provide CO and NOx data, the air pollution index (API) of NO2 was used for comparison. The average API of NO2 in Beijing was reduced by 46.2% during this same period. The reduction level is somewhat higher than the NOx monitoring results (34.9% reduction on average); however, both results show that the NOx emission reductions during the Olympic Games are significant.
CO and NOx emission intensity has also been derived from curbside air quality monitoring with the reverse application of the CALINE4 software package. Figure 20 shows hourly CO emission intensities before and during the Olympic event derived from curbside air quality monitoring. A higher reduction was also observed in the daytime. Table 8 lists the daily emission intensities derived from the monitoring at the North 4th Ring Road site. CO and NOx emissions have been reduced by 44.5% and 49.0%, respectively. As a comparison, daily CO and NOx emission intensity calculated from grid modeling at the cell where curbside monitoring site locates was reduced by 51.0% and 50.1%, respectively, during the Games. The reduction ratio agrees well between the emission intensity derived from grid modeling and field monitoring.
Fig 20. Hourly CO emission intensity at the North 4th Ring Road
Table 8: Emission intensity derived from reverse modeling at the North 4th Ring Road before and during the Olympics.
Unit Before the Olympics
CO NOx During the Olympics
CO NOx
Daily emission intensity kg/km/day 1473.4 71.4 817.7 36.4
Daily traffic flow vehicle/day 242761 189760
5.4. Conclusions
The estimation of the grid-based emission inventories has found that motor vehicle emissions of VOC, CO, NOx and PM10 inside urban Beijing during the 2008 Olympics have been reduced by 55.5%, 56.8%, 45.7% and 51.6%, respectively, as compared to the emission inventory before the Olympics. The effectiveness of transportation control measures adopted during the 2008 Olympics has been demonstrated. The co-effects of traffic flow reduction, traffic congestion improvement and the banning of high emitting vehicles have helped to reduce the direct emissions from motor vehicles by more than one half. Curbside emission intensity derived from air quality monitoring further confirms that such co-effects have magnified the emission reduction benefit from VKT reduction. The experience gained in achieving good air quality during the Olympics suggests that besides vehicle technology improvement, the traffic system can also be improved to attain lower total emissions. Such strategies may include travel demand management and improvement of the public transportation system.
6 SUMMARY AND CONCLUSIONS
Traffic congestion is one of the major problems faced by modern transportation system. One of the chief negative impacts of congestion is the pollution caused by the emissions from vehicles. This report studied about the impact of congestion on emissions, various types of emissions, its permissible limits, factors causing emissions and emission rates at different driving cycles. From the study of various factors that influence the emissions it can be concluded that the emissions can be decreased by an increase in vehicular speed and road width. It can also be seen that the temperature greatly influences the amount of emissions.
In the first case study (Yingying, 2009) the real world emissions under signal coordination and non-coordination in Beijing were collected and compared and the emission levels and distribution characteristics under these two control strategies were analyzed. The results showed that using signal coordination the HC and CO emissions can be effectively reduced. To analyze the impact of different traffic management and control strategies on emissions an integrated microscopic simulation platform of traffic emissions has been developed by integrating the microscopic simulation model of VISSIM and the VSP based approach of the emission modeling, which was used to evaluate the impact of two traffic control strategies, alternative signal timing and different traffic demands on emissions. The result indicates that the optimization of the traffic signal control plan can effectively improve the operation of road traffic and reduce vehicle pollution.
In the second case study (Zhou,2010) grid-based emission inventories were estimated and has found that motor vehicle emissions of VOC, CO, NOx and PM10 inside urban Beijing during the 2008 Olympics have been reduced by 55.5%, 56.8%, 45.7% and 51.6%, respectively, as compared to the emission inventory before the Olympics. The co-effects of traffic flow reduction, traffic congestion improvement and the banning of high emitting vehicles have helped to reduce the direct emissions from motor vehicles by more than one half. The experience gained in achieving good air quality during the Olympics suggests that besides vehicle technology improvement, the traffic system can also be improved to attain lower total emissions.
REFERENCES
1. Alam, J. B., Habib, N. K. M, (2003), Effects of alternative transportation options on congestion and air pollution in Dhaka city , Journal of Civil engineering,31,165-175
2. Beevers, s. d., Carslaw, D. C., (2005), The impact o congestion charging n vehicle emissions in London , Atmospheric Environment, 39,1-5
3. Kun, C., Lei, Y., (2007), Microscopic Traffic Emission Simulation and case study for Evaluation of Traffic control strategies, Journal of Transportation system Engineering and Information Technology, 7(1), 93-100.
4. Nesamani, K. S., Subramanin, K. P., (2006), Impact of Real World Driving Characteristics on Vehicle emissions , JSME International Journal, 49(1), 19-26
5. Smit, R., Brown, A. L., Chan, Y. C., (2008), Do air pollution emissions and fuel consumption models for roadways include the effect of congestion in the roadway traffic flow? , Environmental Modelling and Software, 23,1262-1270
6. Yingying, Z., Xumei, C., Xiao,Z., Guohua, S., Yanzhao, H., Lei, Y., (2009), Assessing effect of traffic control strategies on vehicle emissions , Journal of Transportation System Engineering and Information Technology, 9(1), 150-155
7. Zhou, Y., Wu, Y., Yang, L., Fu, L., He, K., Wang, S., Hao, J., Chen, J., Li, C., (2010), The impact of transportation control measures on emission reductions during the 2008 Olympic Games in Beijing, China , Atmospheric Environment,44,285-293
1. INTRODUCTION
1.1 GENERAL
Emissions from vehicles come from the by-products of the combustion process which takes place inside the vehicle when it moves. In addition to this, volatile organic compounds (VOC) also escape through fuel evaporation. As vehicle exhaust systems have improved, evaporative emissions have become a larger component of total vehicle VOC emissions
The combustion process results in emissions of volatile organic compounds (VOC), oxides of nitrogen (NOx), particulate matter (PM), and carbon monoxide (CO), which are released from the tailpipe while a vehicle is operating. Volatile organic compounds (VOC) also escape into the air through fuel evaporation. With today s efficient exhaust emission controls and gasoline formulations, evaporative losses can account for a majority of the total VOC pollution from current model cars especially on hot days.
The rate of emissions varies largely with the vehicular characteristics like speed, acceleration rates etc. Increase in traffic volumes and changes in travel-related characteristics also increase vehicular emissions significantly. The rates of vehicle emissions can be predicted using models. These models can be grouped into three categories, namely, emission factors models, average speed models, and modal models. The prediction capability of most developed models is relatively poor. Therefore, there is a pressing need to improve the predictability of the existing models or to develop new ones with better accuracy. In addition, different models were evaluated for the prediction capability of certain emissions such as carbon monoxide (CO), nitrogen monoxide (NO), nitrogen dioxide (NO2) and hydrocarbons (HC). These models are mainly based on traffic volume, composition, and flow.
[attachment=7076]
2. LITERATURE REVIEW
The literature review is focused on the earlier studies regarding the congestion and its impacts on vehicle emissions
Alam, Habib (2003) evaluated the effects o alternative transport options on congestion and air pollution in Dhaka city. The options examined in the study included the banning of rickshaws and auto rickshaws from the major roads of the city, improvement of bus service and introduction of rail transit system. From the analysis of future demand or transport facilities it was observed that in the next 15 years the travel demand by about 70 percent and congestion by 250 percent. By 2020, about 60 percent o the major roads will become highly congested during the peak hours with a speed less than 5km/hr. In order to keep mobility at the current level it is required to increase the number of buses and cars by about 50 percent and 25 percent, respectively.
Beevers, Carshaw (2005) have successfully implemented a congestion charging scheme in London which measurably reduces traffic flows in central London. From a comprehensive analysis of impact using detailed traffic data combined with Environment Research Group s road traffic emissions model, has obtained a number of results. They have identified that there is a significant reduction in the emissions of NOx and PM10 associated with increase in vehicle speed. There was also evidence that the speed changes, in km/hr are uniform across the whole range of average speed and therefore changes at the slower speed have a disproportionate effect on vehicle emissions. They have also found out that as a result o the scheme in changes in vehicle kilometer are large particularly in the charging zone itself.
Kun, Lei (2007) have found out that in order to develop effective strategies for vehicle emission controls, it is necessary to develop a microscopic simulation platform for estimating vehicle emissions that can capture the vehicle instantaneous modal activities. For this they have developed an integrated microscopic traffic emission simulation platform by using the microscopic traffic simulation model VISSIM and the modal emission model CMEM. A sub network selected from Haiden district of Beijing is used to build a traffic simulation network ,whose traffic and emission conditions are subsequently evaluated. They first analysed the relationship between the instantaneous emission/fuel consumption rate and the instantaneous speed/acceleration. They also analysed and calculate the emissions for a variety o vehicle types in the network. From these studies they have concluded that the optimization of signal timing plan can improve the traffic operations as well as reduce the traffic emissions. But some traffic control strategies which are used to improve the traffic operations my increase traffic emissions.
Smit, Brown, Chan (2008) examined how and to what extent models which are currently used to predict emission and fuel consumption from low traffic include the effects of congestions. They have connected the emissions to congestion by a key factor called driving pattern. Majority of emission and models explicitly incorporate congestion in the modeling process, but for one important family of emission models, namely average speed models, this could not be determined directly. Re-examination of the driving patterns on which three average speed models (COPERT, MOBILE, EMFAC) are based, shows that it is likely that congestion is represented in these patterns. Since emission factors are based on these patterns, this implies that the emission factors used in these emission models also reflect different levels of congestion. Congestion is thus indirectly incorporated in these models. They have also found out that in order to get more accurate (local) emission predictions and to achieve correct application in particular situations, it is important to improve current average speed models by including a congestion algorithm.
3 EMISSIONS AN OVERVIEW
3.1 TYPES OF EMISSIONS
Cars and trucks do not burn all of the fuel that goes into the combustion chamber. In fact, most cars only burn about 85% of the fuel and as the car gets older less and less fuel is burned. It is this unburned fuel that is responsible for a lot of the air pollution problems today. As the amount of vehicles on the roads increases, so does the amount of pollution. There are four main types of air pollution identified by the Environmental Protection Agency(EPA) and all of them are found in vehicle emissions.
The first type of air pollution is carbon monoxide. Carbon monoxide is dangerous because it reduces the amount of oxygen that can be delivered to our body's organs and tissues, which can be life threatening. A mild poisoning of this toxic gas will result in headaches and dizziness. According to EPA road vehicles are responsible for 51% of carbon monoxide pollution. Of that 51%, cars and light trucks make up 87%.
The second type of air pollution is hydrocarbons. The EPA estimates that the average American car driven 12,500 miles per year produces 10,727 lbs of hydrocarbons. These hydrocarbons damages the lung tissue and it also causes respiratory problems. Road vehicles make up 29% of hydrocarbon pollution with cars and light trucks responsible for 88% of that pollution.
Fuel burning at high temperatures will produce a third air pollutant called, nitrogen oxides. 34% of nitrogen oxide pollutants are produced by road vehicles with cars and light trucks responsible for 52% of that and diesel vehicles responsible for 42%. The combination of sunlight, hydrocarbons and nitrogen oxide results in the smog phenomenon. Smog is found in cities all over the world. Hot, humid and sunny summers increase smog..
Particle matter is the final type of air pollution. Haze is the result of this pollution due to actual particle s of solids and liquids found in the air. The majority of particle matter pollution is produced by industrial plants. Vehicles are only responsible for 10% of this air pollution, but of that 10%, diesel vehicles produce 72%.
3.2 FACTORS AFFECTING THE VEHICULAR EMISSIONS
The are various factors which influence the vehicular emissions. They are the following:
3.2.1 Traffic Parameters
Traffic parameters such as traffic volume, traffic composition, average speed of the flow etc. come under this category. Each of the above parameters significantly affects the pollution levels. As the traffic volume increases the amount of total pollutants will be increased. Traffic composition also significantly affects the pollution levels. Emission rates vary from vehicle to vehicle. Hence, the same amount of traffic volume with different vehicular composition produces different amounts of pollutants. Hence vehicle composition details are to be collected accurately on to a prescribed performa of volume count study. Speed significantly affects the amount of pollutants. As the speed increases the rate of fuel consumption will be decreased, which results in the reduction of (Suspended Particulate Matter) SPM levels. Nitrogen oxides are also vary with the speed. As the speed is affecting all these pollutants it is needed to measure the speed accurately.
3.2.2 Roadway Parameters
Roadway parameters those affecting the pollution levels are carriageway width, lateral clearance, medians, and shoulder width etc. As the road width increases the maneuverability to movement of the vehicles will be increased, which results in the reduction of pollutants. Medians reduce the obstruction caused by the opposing vehicles, which results in the reduction of number of accelerations and decelerations. So the fuel consumption will be minimized. Hence the emission levels will be decreased.
3.2.3 Environmental Parameters
Air quality also depends upon the environmental factors such as the ambient air temperature, humidity, wind speed and wind direction etc. The motor vehicle exhaust emissions are strongly sensitive to ambient temperature. In cold climate they exhaust more emissions. Wind speed and direction will not influence the emissions from vehicles. But they are very useful for air quality modeling. As the wind speed increases the rate of dispersion will be increased, which results in the reduction of pollution concentrations. Similarly wind direction also plays an important role in the pollutant concentration reduction.
3.2.4Vehicle Parameters
The age of the vehicle, its condition and servicing frequency, type of engine (2-stroke, 4-stroke), VMT (Vehicle Miles Traveled) come under this division. The older vehicles will emit more emission than a newer one, if they are not maintained properly. Vehicles with 4-stroke engine would produce lesser emissions than 2-stroke engine. Similarly, vehicles with catalytic converter will emit less pollutant.
3.2.5Type of Fuel
The type of fuel and its composition will also affects the emissions. For example, a gasoline-fueled car may emit high CO emissions than a truck using diesel as fuel. Also the Particulate Matter is mainly coming from the diesel fueled vehicles. Likewise, the composition of fuel in terms of its octane number, cetane number, sulfur content, aromatics and olefins also influence the pollution levels.
3.3PERMISSIBLE LIMITS
The permissible limits for various types of emissions are given by World Health Organisation (WHO) , National ambient air quality standards (NAAQS) for India & Association of Indian Automobile Manufacturers are shown below.
Table1: Maximum permissible limits ( g/m3) of pollutants in the air set by the WHO
Pollutant Time-weighted average Average time
Sulphurdioxide 500
350
100 to 150 a
40 to 60a
10 minutes
1 hour
24 hour
1 year
Carbon monoxide 30
10
1 hour
8 hours
Nitrogen dioxide (WHO 1987, 1977) 400
150
1 hour
24 hours
Ozone (WHO 1987, 1978) 150 to 200
100 to 120
1 hour
8 hours
Total SPM (Suspended Particulate Matter) 150 to 230 a
60 to 90 a
24 hours
1 hour
Source: WHO,1992
Table2: National ambient air quality standards (NAAQS) for India (in g/m3)
Pollutant and time-weighted average Industrial area Residential, rural, and other areas Sensitive area Methods of measurement
Sulphur dioxide
Annual average
24 hours
80.00
120.00 60.00
80.00 15.00
30.00 Improved West and Gaeke method
Ultraviolet fluorescence
Oxides of Nitrogen
Annual average
24 hours
80.00
120.00 60.00
80.00 15.00
30.00 Jacob and hochheiser modified (Na-Arsenite) method
Gas-phase chemilluminescene
Suspended particulate matter
Annual average
24 hours
360.00
500.00
140.00
200.00
70.00
100.00 High-volume sampling (average flow rate not less than 1.1 m3 per minute)
Respirable particulate matter (size less than 10 gm)
Annual average
24 hours
120.00
150.00
60.00
100.00
50.00
75.00
Respirable particulate matter sampler
Lead
Annual average
24 hours
1.00
1.50 0.75
1.00 0.500
0.750
Atomic absorption spectrometry after sampling using EPM 2000 or an equivalent filter paper
Carbon monoxide
8hours
1 hour 5.00
10.00 2.00
4.00 1.00
2.00 Non-dispersive infrared spectroscopy
Ammonia
Annual average
24 hours -- 400.00
100.00 -- --
Source: Central Pollution Control Board,
Table 3: Mass emission standards for diesel-driven vehicles (in g/kwh)
1 April 1992 1 April 1996 1 April 2000 Euro II Euro II
Gross vehicle weight >3.5 tonnes
Carbon monoxide 14.00 11.20 4.50 4.00 2.00
Hydrocarbon 3.50 2.40 1.10 1.10 0.60
Oxides of nitrogen 18.00 14.40 8.00 7.00 5.00
Particulate matter for >85 KW 0.00 0.00 0.61 0.15 0.10
Particulate matter for <85 KW 0.00 0.00 0.61 0.15 0.10
1 April 1992 (g/kWh) 1 April 1996 (g/km) 1 April 2000 (g/km)
Gross vehicle weight <3.5 tonnes
Carbon monoxide 14.00 5.0-9.0 2.72-6.90
Hydrocarbon 3.50 -- --
Oxides of nitrogen 18.00 2.4-4.0 0.97 1.70
Particulate matter 0.00 -- 0.14-0.25
Source: Association of Indian Automobile Manufacturers,1999
3.4 COMPARISON OF EMISSION RATES OF PETROL ENGINES AND DIESEL ENGINES
Diesel engines consume around 30% less fuel than petrol engines and this results in much lesser carbon dioxide emissions. The diesel engines produce virtually no carbon monoxide and are much safer than petrol engines. Tests done on car emissions revealed that Nitrogen Oxides are higher in a new diesel engine when compared to a new petrol engine. But by the time they cover 50,000 miles or so, they are the same and after that the petrol engine produces more oxides than the diesel engine. Hydrocarbon emissions contained in petrol engine emissions are considerably more than that in diesel engine emissions. However, diesel is certainly more dangerous from the point of view of Suspended Particulate Matter (SPM). SPM refers to solid particles suspended in open air, such as soot generated by combustion of various fuels. They might cause respiratory problems because of their tendency to deposit themselves in the lungs. Though much has been done to improve the fuel efficiency and reduce emissions from the petrol engine, still more needs to be done.
3.5 EMISSION TESTING
Emissions testing is a crucial part of automobile testing to ensure that the vehicle is not polluting the environment. In some states there are strict guidelines and rules to follow for emission testing that are inescapable
The aim of the emission test is to reduce the production of harmful gases and pollutants so that the air quality can be improved. The aim of the vehicle emission control is to reduce the pollutants emitted by vehicle.
The pollutants to be checked in Vehicle Emissions Testing are:
The level of hydrocarbon in emissions testing is checked as excess of these causes smog and also leads to cancer. Hydrocarbon is produced by the unburnt or partially burnt fuel. As per the specification of engine there is the regulation for non-methane hydrocarbons and total hydrocarbons.
Carbon monoxide released because of the partial combustion of fuel is then checked. When inhaled it reduces the oxygen carrying capacity of blood and excess of CO in body is fatal. During vehicle emission testing the level of CO emitted by your vehicle is checked.
Nitrogen oxide which is the resultant of combination of nitrogen in the air with the oxygen at high temperature and pressure is also chcked. Nitrogen oxide causes acid rain and smog. If extra amount of NO and NO2is inhaled then it can lead to less resistance towards respiratory infections.
Oxygen is the remaining unburned gas during combustion. Although Oxygen is not bad but is checked to find out whether you have cheated in the emissions testing or not. Also with the percentage of O2in the emission test determines the fuel ratio of engine while running. This test is also done by gas analyser.
Testing procedure
To perform the vehicle emission test, the vehicle will be accelerated on dynamometer which is a treadmill like machine. Then a probe is placed in the tailpipe. The readings on pollutants is then checked in the emission testing on to computer. The readings are then compared to the standard values and its year.
3.6 EMISSION TEST TYPES
Broadly we can categories the emissions test in the following steps
Acceleration Simulation Mode
Vehicle Model Year 1995 or older than that uses the dynamometer. Vehicle tires are set on this machine which then applied load and resistance on the vehicle to create the road conditions. This consists of two phases. These include
50/15 mode - In this phase of 50/15 mode, constant speed of 15 mph is maintained and vehicle's 50% horsepower is used to stimulate dynamometer.
25/25 mode - In second phase the speed is increased to 25 mph and horsepower is reduced to 25%.
On-Board Diagnostics
Performance of the vehicle is also monitored on the in-build computer systems. This include fuel metering system for emissions test, ignition system, emission control equipment. Any dis-balance shown in this can lead to the failure in the vehicle emission test.
Two Speed Idle
In this a probe is inserted into the tail piece of the vehicle to find out different emitted gases. This probe is attached to the computer that records the reading and compare these with standard values in emissions testing. This is also done in two phases.
1. high speed test (2200-2800 RPMs)
2. tested at idle (350-1200 RPMs.)
3.7 CONGESTION AND AIR POLLUTION
Road transport has grown rapidly over the last decades. As a consequence, traffic-related emissions and air pollution have increased substantially despite the increasing use of abatement measures such as catalytic converters. Due to the growth in road traffic, congested traffic conditions have become increasingly common and severe worldwide, particularly in major cities. A strong further increase in the demand for transport and congestion is projected around the world.
Traffic congestion has been indicated as being one of the main contributors to environmental impacts in urban areas. Congestion is the deterioration of the quality of traffic flow on a network element or in an entire road network due to increased travel demand and/or reduced capacity for traffic movement, which may be observed in terms of different interrelated measures including, but not limited to, increased traffic density, increased travel times, increased delays, lower travel speeds and increased queuing
3.8 DRIVING CYCLES AND EMISSIONS
Driving cycle is a sequence of vehicle operating conditions. It includes idle, acceleration, steady state and deceleration. There are number of studies relating driving characteristics and vehicular emissions using on board measurement and remote sensing techniques. Conducting on-board measurements would be expensive and difficult to collect large number of vehicle samples. In the case of remote sensing techniques, it can be used only at a particular location and can capture instantaneous emissions alone. Nesamani, Subramanian et al.,(2006)in Chennai estimated the emissions for different driving patterns using International Vehicle Emission (IVE) model. It is a computer model designed to estimate emissions from motor vehicles. It was jointly developed by International Sustainable Research Centre and the University of California at Riverside. The general inputs to the model are fleet characteristics, vehicle activity and emission factor based on local conditions. From the results it was observed that the emission rates vary significantly from one class of road to another and large effects was on local streets. This was mainly due to local street spending higher percentage of time in acceleration mode with low average speed. Both arterial and sub-arterial roads have almost similar emission rate except NOx which was higher in arterial. This could be attributed to sharp acceleration rate in arterial streets. CO2 emission was lower in arterial street and higher in local street.
4 ASSESSING EFFECT OF TRAFFIC SIGNAL CONTROL STRATEGIES ON VEHICLE EMISSIONS
4.1 GENERAL
The vehicle pollution and traffic noise have worsened due to the rapid increase of vehicles in Beijing, China. Several experts and scholars have actively carried out the research on vehicle emissions under different traffic management and control strategies. This paper is intended to evaluate the impact of different traffic control strategies on vehicle emissions using the microscopic traffic simulation model VISSIM.
4.2 EMISSION CHARACTERISTICS UNDER SIGNAL COORDINATION AND NON-COORDINATION
4.2.1 Data collection
The real world data were collected using portable emission measurement system (PEMS). Qianmen Avenue (a road section from Hepingmen to Chongwenmen) in Beijing was selected as coordinated signal controlled road and Garden East Road North Taipingzhuang Road (from Zhixin Bridge to Anzhen Bridge) as non-coordinated signal controlled road. Then, the emission levels of a vehicle under these two types of controls were compared (i.e., different traffic management and control strategies). These routes were selected mainly according to the average speed, while the traffic flow, congestion level, and traffic flow composition from morning to afternoon of the road were considered as the secondary criteria .
A 2000-year-manufactured Jetta CI with the mileage of 0.12 million kilometers has been selected for the experiment subject. The experiment was implemented from 9:00 a.m. to 11:00 a.m. on July 25 26, 2006. A total of 3, 133 valid emission records for the signal coordinated road on July 25 and 5, 316 valid emission records for the signal non-coordinated road on July 26, 2006 were collected. In addition, traffic surveys were conducted for two hours during the experiment period, including traffic compositions, actual road conditions, traffic flow data of both directions, number of vehicles entering into and leaving off intersections, existing signal timings, and so on. These data served s the input parameters for the microscopic traffic simulation model VISSIM to simulate the traffic operations of coordinated signal controlled road.
4.2.2 Characteristics analysis of emission data
By integrating the electronic map of Beijing city and the real-time emission data, this study established an emission analysis platform, in which MapInfo was implemented for GIS analysis and the Map Basic was selected for the second development tool.
4.2.2.1 Comparison of emission factors
Vehicle emission factor indicates emissions per unit of distance, hour or fuel consumption, which is measured in g/km, g/h, or g/kg (fuel). It is an important indicator for measuring the level of vehicle emissions, as well as the basis and foundation for studying the emission control strategy. Fig. 1 illustrates the comparison of emission factors on coordinated versus non-coordinated signal controlled roads. Compared with emission factors on the non-coordinated signal controlled road, the HC and CO emission factors on the coordinated signal controlled road have decreased by 50% and 30%, respectively; however, NOx has increased by 10%. This result shows that using signal coordination can effectively reduce the HC, CO emissions, but increase the NOx pollution.
Fig1. Comparison of emission factors on coordinated versus
non-coordinated signal controlled roads
4.2.2.2 Comparison of emissions based on driving cycles
Analyzing emissions based on different driving cycles is an effective way to study the microscopic characteristics of vehicle emissions. Fig. 2, 3, 4 shows the comparison of emissions under different driving cycles based on coordinated versus non-coordinated signal controlled roads.
Fig2. Comparison of NOx emissions under different driving cycles
Fig3. comparison of HC emissions under different driving cycles
Fig4. Comparison of CO emissions under different driving cycles
As illustrated in Fig. 2, 3, 4 the CO and HC emission rates under different driving cycles on coordinated signal controlled road are lower than those on non-coordinated signal controlled road, but the CO emissions increased slightly under idling. The NOx emission rate under different driving cycles on the coordinated signal controlled road is similar to the one on the non-coordinated signal controlled road. From
the perspective of emissions, all emission rates under acceleration are very high, and NOx emission under deceleration should not be overlooked.
4.2.2.3 Analysis of emission characteristics based on different speed intervals
The quantified relationship between traffic and vehicle emission can be generated by investigating the trend of emission factors under different speed intervals, which will provide data to support the development of the traffic management and control strategies for reducing vehicle emissions. The trend of emissions under different speed intervals is shown in Fig. 5, 6, 7.
Fig5. The trend of NOx emissions under different speed intervals
Fig6. The trend of HC emissions under different speed intervals
Fig7. The trend of CO emissions under different speed intervals
From figures 5, 6, 7 it can be seen that the HC and CO emission factors show a consistent trend of declining as the speed increases. The emissions are very low after the speed reaches 20 km/h, which indicates that the HC and CO emissions can be controlled under low speeds. NOx has just minor changes after the speed reaches 20 km/h, which illustrates that an optimization of speed has no significant effect on reducing NOx. CO and HC emission rates under different speed intervals on coordinated signal controlled road are lower than those on non-coordinated signal controlled road, but NOx has an opposite trend. The result indicates that using signal coordination can effectively reduce the HC, CO emissions, but NOx emission will become worse, which is consistent with the earlier conclusions.
4.3 ANALYSIS OF EMISSIONS BASED ON MICROSIMULATION UNDER SIGNAL COORDINATION
4.3.1 Development of microscopic traffic simulation platform
VISSIM is a microscopic model based on behavior and time-step analysis, which can simulate the traffic operation of various types of roads, such as the urban road and highway. In order to compare and analyze the impact of emissions under different traffic management and control strategies an integrated microscopic simulation platform of traffic emissions has been developed by integrating the microscopic traffic simulation model of VISSIM and the VSP based approach of emission modeling, which is used to simulate the network of Qianmen coordinated signal controlled road based on traffic surveys.
By simulating Qianmen road in Beijing with the VISSIM model, the velocity(v) and acceleration (a) can be output on a second-by-second basis, and the instantaneous VSP can be calculated according to equation (1). Then selecting 1 as the interval unit, the VSP value belonging to interval [-30, 30] can be designated into different VSP Bins using equation (2). When the emission rates are finally obtained corresponding to different VSP intervals and the VSP frequency distribution, the total vehicle emissions of certain trip can be calculated using equation (3)
VSP = v (1.1a + 0.132) + 0.000302 v3 (1)
for all VSP [n 0.5,n + 0.5], VSPBin = n (2)
where n is n integer
Q = f(p) R(pi) (3)
where Q is the total vehicle emissions of a certain trip; i is the travel time (i = 1, 2, 3 .n); pi is the VSP value of the ith second; f(pi) is the VSP distribution of the ith second when VSP equals Pi; R(pi) is the emission rate of the ith second when VSP equals pi
4.3.2 Impact of different traffic management and control strategies on emissions
4.3.2.1 Impact of different signal timings on emissions
Different traffic signal control strategies can result in different microscopic driving cycles of vehicles in the network, thus leading to different levels of emission pollution. In order to analyze the impact of emissions under different signal timings, the network of Qianmen coordinated signal controlled road is simulated based on the actual surveys of the signal timing plans of the entire road, and then the signal timing plans of simulated network of Qianmen are transferred into non-coordinated signal controlled for comparison. Furthermore, the vehicle emissions of coordinated and non-coordinated signal controlled roads can be calculated using the VSP-based approach of the emission modeling. Two simulation results are converted to emission factors, as shown in Table 4.
Table 4. Comparison of calculation results on coordinated versus non- coordinated signal controlled roads
Travel time(s) Average speed (km/hr) NOx (g/km) HC(g/km) CO(g/km) CO2
Signal coordination 466 23.12 1.061 0.0529 2.559 122.148
No signal coordination 491 21.87 1.233 0.0588 2.985 134.566
Trend +5% -5% +14% +10% +14% +9%
It can be seen from table 4 that the emission factors increase by bout 10%, the travel time increases by 5%, and the average speed drops 5% after the coordinated signal controlled roads has been converted to non-coordinated road. This indicates that the optimization of the traffic signal control plan can effectively improve the operation of road traffic and reduce vehicle emission pollution.
4.3.2.2 Impact of different traffic demands on emissions
Different traffic demands directly influence the vehicles speed and driving behavior on roads, and thus affect vehicle exhaust emissions. In order to compare the impact of different traffic demands on the emission, the network of Qianmen coordinated signal controlled road with increased versus decreased demands by 20% flows were simulated respectively. Then, the vehicle emissions of the coordinated controlled road with increased and decreased demands by 20% were calculated with the VSP based approach of emission modeling. The two simulation results were converted to emission factors as shown in table 5
Table 5. Comparison of simulation results by increased versus decreased traffic demand by 20% on coordinated signal controlled roads
Travel time(s) Average speed (km/hr) NOx(g/km) HC(g/km) CO(g/km) CO2
(g/km)
Trend -22.53% +28.29% -78.32% -62.19% -69.01% -59.08%
Decreasing 20% flow 361 29.66 0.23 0.020 0.793 49.978
Baseline flow 466 23.12 1.061 0.0529 2.559 122.148
Increasing 20% flow 514 20.93 1.466 0.063 3.629 150.278
Trend +10.3% -9.47% +36.29% +19.09% +41.81% +23.03%
From table 5, it can be seen that when traffic demands increase by 20%, the travel time increases 48 seconds and the average speed decreases by 9.47%. The emission factors are significantly increased. CO and NOx increased by 41.81% and 36.29% respectively. When the traffic demands decrease by 20%, the travel time reduces 105 second as well as the average speed improves by 28.29%. The emission factors are significantly decreased respectively, as much as over 50%, and the maximum reduction is for NOx, up to 78.32%. This indicates that the decreasing traffic demands can significantly reduce vehicle emissions and optimize traffic operations. Therefore, when selecting traffic management and control strategies to reduce emissions on roads, traffic manager can consider implementing route guidance and traffic demand management strategies on roads with high emissions.
4.4 CONCLUSIONS
The real world emissions under signal coordination and non-coordination in Beijing were collected and compared and the emission levels and distribution characteristics under these two control strategies were analyzed. It was found that in comparison with the emission factors of non-coordinated road, the HC and CO emission factors of coordinated signal controlled road decreased by 50% and 30%, but NOx increased by 10%. The results showed that using signal coordination the HC and CO emissions can be effectively reduced. All emission rates under acceleration are very high. Therefore, the emission under acceleration should be strictly controlled. In order to analyze the impact of different traffic management and control strategies on emissions an integrated microscopic simulation platform of traffic emissions has been developed by integrating the microscopic simulation model of VISSIM and the VSP based approach of the emission modeling, which was used to evaluate the impact of two traffic control strategies, alternative signal timing and different traffic demands on emissions. It was found that the emission factors increased by about 10%, the travel time increased by 5%, and the average speed drops by 5% after the coordinated signal controlled road was converted t non-coordinated road. The result indicates that the optimization of the traffic signal control plan can effectively improve the operation of road traffic and reduce vehicle pollution. When the traffic demand decreased by 20%, the average speed improved by 28.29%, and the emission factors significantly decreased respectively, as much as over 50%. The maximum reduction was for NOx, up to 78.32% which indicates that decreasing traffic demands can significantly reduce vehicle emissions. Therefore, it has been concluded that when selecting traffic management and control strategies to control emissions on roads, traffic managers can consider implementing the optimization of signal timing and at the same time try to control traffic demands.
5 THE IMPACT OF TRANSPORTATION CONTROL MEASURES ON EMISSION REDUCTIONS DURING THE 2008 OLYMPIC GAMES IN BEIJING, CHINA
5.1 GENERAL
Since 2001, when Beijing won the bid to host the 2008 Olympics, air quality for the competition was a major concern. Beijing held the 2008 Olympic Games from August 8 to August 24, 2008. In June 2008, the local government of Beijing promulgated temporary transportation control measures to be implemented during the event. The study periods were classified as before and during the Beijing Olympic Games to evaluate the effect of temporary transportation control measures adopted throughout the event. Field traffic flow monitoring and a calibrated transportation simulation platform based on geographic information system (GIS) have been applied in this study to achieve road network activity and operational speed with a 1 km x 1 km resolution. A bottom-up methodology was applied to develop motor vehicle emission inventories before and during the Games. The effectiveness of transportation control measures has been evaluated by comparing these grid-based emission inventories. The emission reduction benefit has also been evaluated by the emission intensity derived from reverse modeling of curbside air quality monitoring results.
5.2. METHODOLOGY
5.2.1. Traffic modeling
The urban area of Beijing is divided into 2055 grid cells, each of which is 1 km x 1 km. An urban transportation simulation platform based on TransCAD has been applied in this study to estimate the vehicle kilometers traveled (VKT) and the average speed in these cells. Annual traffic flow monitoring, travel demand management policies as well as personal trip mode investigation has been applied as basic inputs to the simulation platform. Simulated road page link activities are further transformed into grid-based activities. Motor vehicles identified in traffic volume monitoring and modeling have been classified into six categories including light duty vehicles (LDV), taxis, light-duty trucks (LDT), heavy duty trucks (HDT), buses and heavy-duty vehicles (HDV). Daily and peak hour VKT for each vehicle category and average speed have been estimated by the system. Information provided in each grid cell includes geographic coordinates, road length, average speed and daily VKT for each vehicle category.
The Beijing Transportation Research Center conducted a special investigation on continuous traffic monitoring on 132 road links in urban Beijing before and during the Olympic Games to estimate the variation in transportation activity on a geographic scale. The results of this investigation, as well as routine spatial and temporal origin destination flow investigations for residents in Beijing, were used to calibrate the transportation control scenarios in the simulation platform.
5.2.2. Emission factor development
MOBILE5B- China, a localized model based on US Environmental Protection Agency (USEPA). MOBILE5B and PART % was applied in this study to estimate speed- dependant vehicle emission factors. A special module has been developed for estimating the effect of transportation control measures on the average vehicle emission factor. The emission factor was mainly influenced by the control measures adopted during the Games in two ways.
First, is the vehicle operation parameter such as speed which is one of the most important factors affecting emissions. Second, is the fleet technology configuration. As yellow labeled vehicles were banned throughout Beijing and strict operational restricts were also placed on trucks. Table 6 lists the influence of the detailed transportation control measures on the major factors related to traffic emissions.
Table6. Influence of transportation control measures on the main factors o traffic emissions
Strategies Factors influencing traffic emissions
Travel restriction by odd-even license Speed, Total activity
Truck operational restrictions Fleet composition, Total activity
Removal of yellow labeled vehicles Fleet composition, Total activity
The overall vehicle fleet change has been taken into consideration to evaluate the effect of vehicle operation parameters and fleet configurations on average emission factors. The registration distribution of yellow labeled vehicles in Beijing has been investigated within the local Inspection and Maintenance database. Green Environmental labels have been issued in Beijing to gasoline vehicles meeting at least Euro I and diesel vehicles meeting at least Euro II requirements. Other vehicles were issued yellow environmental labels. The yellow labeled vehicles which are usually referred to as high emitting vehicles were mostly banned while the Olympic Games were underway. As a result of this, the average age of the light duty fleet and heavy duty fleet operating in urban area of Beijing during the Olympics was reduced by 0.48 and 3.55 years respectively.
By the end of 2007, about 3, 47000 yellow labeled vehicles were still in use in Beijing which was 11% of the total fleet. The fleet composition of HDT has been changed the most because diesel trucks can only get a label if they can meet at least Euro II emission standards in Beijing. Figure 8 shows the change of speed dependant average NOx emission factor for HDT, up to 80km/h before and during the Beijing Olympics. It should be noted that not all the trucks with yellow environmental labels were banned. The municipal government has issued special passes to some yellow labeled trucks to guarantee the progress of important project
Fig 8. Speed dependent average NOx emission factors for HDT before and during the Olympics
. The speed-dependent emission factors for Volatile Organic Compound (VOC) , Carbon monoxide (CO) and Nitrogen oxide (NOx) for each vehicle class are calculated within MOBILE5B-China. Figure9 shows the relationship between the emission correction factor and the average speed for LDV relative to the baseline speed of 30 km/h. It shows that the VOC and CO emission factors have a significant relationship with speed, and generally decrease with increasing speeds. The NOx emission factors, which are less sensitive to speed changes, follow a parabolic path and tend to decrease to 35 km/h and then increase.
Fig 9. Speed correction factor for emission factor for LDV
Real-world vehicle emission measurements on a fixed route have been carried out with PEMS in Beijing before and during the Games. PEMS measurements on 5 light-duty gasoline vehicles showed that the average reduction of their CO, HC and NOx emission factors were 26.8%, 28.7% and 0.8%, respectively, during the Games as the trip average speed increased from 26.8 km/h to 34.6 km/h. This agrees with the modeling output in this study. As shown in Figure 9, the modeled CO, VOC and NOx emission factor reductions for LDV for the same speed increment are 21.8%, 21.1% and 1.1%, respectively. The real-world measurements have also shown that the CO and VOC are sensitive to the change of vehicle speed, but the NOx is not sensitive to the change of vehicle speed. The similarity in the decreased level of the three air pollutants from both measurements and modeling work confirms the good performance of the MOBILE5B-China emission factor model.
5.2.3. Emission reduction benefit evaluation
5.2.3.1. Grid-based emission inventory development
In order to quantify the emission reduction benefit of the transportation control measures adopted, a bottom-up methodology has been applied in this study to build emission inventories for the urban areas before and during the Games. Grid-based vehicular speeds from traffic modeling are used to obtain emission factors for each vehicle type in each grid cell. This approach is a physical representation of realistic traffic conditions in each grid cell of the urban area. Emissions in each grid cell were calculated by multiplying the VKT and speed-dependent emission factors for the six vehicle categories:
Qpc = EF pi, c x VKT i,c (1)
where Qpc is the emissions of pollutant P in cell c (g) ; EF pi, c is the speed- dependant emission factor of pollutant P or vehicle category i in cell c (g/km) ; VKT i,c is the total vehicle kilometers traveled of vehicle category I in cell c (km). The emissions were summed over all grid cells to obtain total urban emissions.
5.2.3.2. Emission reduction benefit derived from curbside monitoring
For this two sampling sites were set up on the sidewalks of the North 4th Ring Road, about 26mfrom the road centerline. Measurements were taken at 2 m above the ground. The detailed location of the sites is shown in figure 10 , which is approximately 7kmfrom the Olympic Park. The monitoring was performed before the Olympics (June 22 - July 5, 2008) and during the Olympics (July 28 - August 22, 2008). Hourly traffic volume was counted throughout the monitoring periods based on continuous video taping from the crossover above the road. Hourly data for CO and NOx concentrations were provided by Thermo 48i and 42i, respectively, at both sites along the sidewalks. A portable weather station was also set up near the air quality monitoring instruments to record second-by-second ambient temperature, wind speed, wind direction, humidity and atmospheric pressure.
Fig 10. Location of curbside monitoring sites
Line-source dispersion models are usually used for the calculation of air quality based on known theoretical relationships between emissions, meteorology parameters and air pollutant concentrations. Street pollution models like OSPM and CALINE4 have been combined with curbside air quality measurements to estimate in-situ emission intensity or emission factors. For the dispersion of non-reactive or slowly reactive vehicle exhaust gases at short distances, chemical transformations can be disregarded. Equation (2) may be used for calculations of hourly emissions from traffic, provided that both receptor and background concentrations are available from in-field monitoring.
Q = Cr Cbg
Fd
where Cr and Cbg are the receptor concentration with the contribution from road traffic and the background concentration from sources other than motor vehicles in the specified road link, Q is the traffic emissions at the road link, and Fd is a factor describing dispersion processes under certain meteorological conditions. In this study, only those monitoring data with the wind direction perpendicular or nearly perpendicular to the road were selected for CALINE4 simulation. Thus, in this case the difference between the monitored concentrations at both sides of the road can be used as the indicator of the direct contribution from the road traffic. The dispersion factor Fd is given by a line-source pollution model, in our case, CALINE4. The model was originally developed for calculating CO concentrations (in ppm) from a road. A scaling factor has been applied in this study to adapt CALINE4 for NOx. The inputs of the model include road geometry, meteorological parameters (wind speed, wind direction and its standard deviation, temperature, atmospheric stability class and mixing height), background concentration, and receptor information. The model also requires the input of road page link emission intensity, which becomes the target for reverse application of CALINE4 in this study.
Considering the possible systematic errors in the line-source dispersion model, the reduction ratio instead of the absolute value of derived emission intensity has been used in this study for further analysis. The emission reduction ratio derived from curbside monitoring was compared with those achieved from the bottom-up inventory modeling to show the emission reduction benefits of the transportation control measures.
5.3 Results and discussion
5.3.1. Traffic flow effect
Figure 11 shows the reduction ratio in daily traffic flows for each vehicle category from continuous traffic monitoring at 132 road links before and during the Olympics. Vehicle categories other than taxis and buses show a significant decrease in traffic flow due to the temporary traffic control measures. HDT had the highest reduction
ratio because of the strict traffic intervention measures on its operational area. Increasing bus and taxi flow has been observed at some road links inside the downtown area or close to the Olympic venues.
It can be seen from figure 11 that the daily flow of LDV and HDT has been reduced by 30.1% and 53.8%, respectively, during the Games. The flow of taxis and buses has been increased by 29.6% and 9.8%, respectively. The monitoring sites are close to the main Olympic venues, which could be the reason for higher public transportation demand. Two Olympic-specific lanes and increasing trips to the Olympic Park have also counteracted the effect of traffic control.
Fig 11. Daily traffic flow reduction ratio monitored during the Olympics. Line within box: Median value; Top line of box: third quartile; Bottom line of box: first quartile; Top outlier: maximum; Bottom outlier: minimum
Figure 12 shows the weekday hourly flow monitored at the North 4th Ring Road. Hourly flow data has shown that most traffic reduction was achieved during the daytime. Traffic flow between midnight and 3:00 shows a temporarily slight increase because the municipal government issued a supplemental rule suspending license-based traffic control measures during these three hours.
Fig 12. Variation of weekday traffic flow at the North 4th Ring road before and during the Olympics
Daily VKT and average speed distributions derived from the transportation simulation platform are shown in figures 13 and 14. Total urban VKT has been reduced by 32.0% during the Games. The average speed weighed by grid VKT has been increased from 25 km/h to 37 km/h during the Games. It has been found that a significant proportion of the grids are located in the speed bins lower than 30 km /h in downtown areas before the Games.
Fig13. Daily VKT for urban Beijing before (a) and during (b) the Olympics
Fig 14. Daily average speed for urban Beijing before (a) and during the Olympics
5.3.2. Bottom-up emission inventory
Grid-based vehicle activities has been acquired from the calibrated transportation simulation platform and the speed dependent emission factors has been acquired from MOBILE5b-China. Based on these data the bottom-up methodology described in equation.(1) was used to develop mobile source emission inventories. The grid-based vehicle emission inventories for CO, VOC, NOx and PM10 in the Beijing urban area before and during the Olympic Games are shown in figures 15-18, respectively. From the figures it has been observed that the distribution of vehicle emissions in Beijing's urban area are more concentrated in the urban core area with higher population density and travel demand.
Fig 15. Grid-based daily vehicle emission inventory for CO in urban Beijing before (a) and during (b) the Olympics
Fig16. Grid-based daily vehicle emission inventory for VOC in urban Beijing before (a) and during (b) the Olympics
Fig17. Grid-based daily vehicle emission inventory for NOx in urban Beijing before (a) and during (b) the Olympics.
Fig 18. Grid-based daily vehicle emission inventory for PM10 in urban Beijing before (a) and during (b) the Olympics.
The radial arterials and exit highways with heavy traffic also contributed to high emission intensity in these grids. Grid-based emissions can be aggregated to calculate the total vehicular emissions. Table 7 shows a comparison between the total urban vehicle emissions before and during the Games.
Table 7. Daily urban motor vehicle emissions of Beijing in tons before and during the 2008 Olympics
VOC CO NOx PM10
Before the Olympics 371 2993 282 15.9
During the Olympics 165 1293 153 7.7
Emission reduction 55.5% 56.8% 45.7% 51.6%
Improved traffic efficiency and reduced fleet average age helped Beijing achieve a higher mobile source emission reduction than lower VKT. More significant reductions have been found in VOC and CO emissions because they are more sensitive to average speed as shown in figure9. Based on the daily grid-based inventory before the Olympics, trucks have contributed 21.6% and 40.3% of total motor vehicle NOx and PM10 emissions in urban Beijing, respectively. Because of the higher contributions of trucks to PM10 emissions, strict controls on them during the Olympic Games achieved a higher reduction in total PM10 emissions than NOx emissions.
5.3.3. Emission intensity at the North 4th Ring Road
Figure 19 shows hourly average downwind CO and NOx concentration monitored at the North 4th Ring Road site before and during the Olympics. Receptor CO and NOx concentration was reduced by 31.2 7.4% and 34.9 13.9%, respectively. The largest CO reduction was observed from 16:00 to 19:00, one of the two rush hours. The largest NOx reduction was observed during the midnight. Long before the beginning of the Olympics, the Beijing Traffic Management Bureau was already enforcing a routine traffic control regulation that trucks were only allowed to operate inside the 4th Ring Road between 23:00 and 6:00. Trucks were the most important NOx and PM contributor at night before the Olympics. The banning of yellow labeled vehicles during the Olympics mostly affected the traffic flow of diesel trucks and average emission factors among all vehicle categories, especially at night. This resulted in a higher reduction in NOx than CO during the midnight.
Fig 19. Hourly average downwind CO (a) and NOx (b) concentrations monitored at the North 4th Ring Road before and during the Olympics.
The Beijing Environmental Protection Bureau (EPB) reports air pollution level of PM10, SO2 and NO2 in Beijing through its public domain .As the Beijing EPB does not provide CO and NOx data, the air pollution index (API) of NO2 was used for comparison. The average API of NO2 in Beijing was reduced by 46.2% during this same period. The reduction level is somewhat higher than the NOx monitoring results (34.9% reduction on average); however, both results show that the NOx emission reductions during the Olympic Games are significant.
CO and NOx emission intensity has also been derived from curbside air quality monitoring with the reverse application of the CALINE4 software package. Figure 20 shows hourly CO emission intensities before and during the Olympic event derived from curbside air quality monitoring. A higher reduction was also observed in the daytime. Table 8 lists the daily emission intensities derived from the monitoring at the North 4th Ring Road site. CO and NOx emissions have been reduced by 44.5% and 49.0%, respectively. As a comparison, daily CO and NOx emission intensity calculated from grid modeling at the cell where curbside monitoring site locates was reduced by 51.0% and 50.1%, respectively, during the Games. The reduction ratio agrees well between the emission intensity derived from grid modeling and field monitoring.
Fig 20. Hourly CO emission intensity at the North 4th Ring Road
Table 8: Emission intensity derived from reverse modeling at the North 4th Ring Road before and during the Olympics.
Unit Before the Olympics
CO NOx During the Olympics
CO NOx
Daily emission intensity kg/km/day 1473.4 71.4 817.7 36.4
Daily traffic flow vehicle/day 242761 189760
5.4. Conclusions
The estimation of the grid-based emission inventories has found that motor vehicle emissions of VOC, CO, NOx and PM10 inside urban Beijing during the 2008 Olympics have been reduced by 55.5%, 56.8%, 45.7% and 51.6%, respectively, as compared to the emission inventory before the Olympics. The effectiveness of transportation control measures adopted during the 2008 Olympics has been demonstrated. The co-effects of traffic flow reduction, traffic congestion improvement and the banning of high emitting vehicles have helped to reduce the direct emissions from motor vehicles by more than one half. Curbside emission intensity derived from air quality monitoring further confirms that such co-effects have magnified the emission reduction benefit from VKT reduction. The experience gained in achieving good air quality during the Olympics suggests that besides vehicle technology improvement, the traffic system can also be improved to attain lower total emissions. Such strategies may include travel demand management and improvement of the public transportation system.
6 SUMMARY AND CONCLUSIONS
Traffic congestion is one of the major problems faced by modern transportation system. One of the chief negative impacts of congestion is the pollution caused by the emissions from vehicles. This report studied about the impact of congestion on emissions, various types of emissions, its permissible limits, factors causing emissions and emission rates at different driving cycles. From the study of various factors that influence the emissions it can be concluded that the emissions can be decreased by an increase in vehicular speed and road width. It can also be seen that the temperature greatly influences the amount of emissions.
In the first case study (Yingying, 2009) the real world emissions under signal coordination and non-coordination in Beijing were collected and compared and the emission levels and distribution characteristics under these two control strategies were analyzed. The results showed that using signal coordination the HC and CO emissions can be effectively reduced. To analyze the impact of different traffic management and control strategies on emissions an integrated microscopic simulation platform of traffic emissions has been developed by integrating the microscopic simulation model of VISSIM and the VSP based approach of the emission modeling, which was used to evaluate the impact of two traffic control strategies, alternative signal timing and different traffic demands on emissions. The result indicates that the optimization of the traffic signal control plan can effectively improve the operation of road traffic and reduce vehicle pollution.
In the second case study (Zhou,2010) grid-based emission inventories were estimated and has found that motor vehicle emissions of VOC, CO, NOx and PM10 inside urban Beijing during the 2008 Olympics have been reduced by 55.5%, 56.8%, 45.7% and 51.6%, respectively, as compared to the emission inventory before the Olympics. The co-effects of traffic flow reduction, traffic congestion improvement and the banning of high emitting vehicles have helped to reduce the direct emissions from motor vehicles by more than one half. The experience gained in achieving good air quality during the Olympics suggests that besides vehicle technology improvement, the traffic system can also be improved to attain lower total emissions.
REFERENCES
1. Alam, J. B., Habib, N. K. M, (2003), Effects of alternative transportation options on congestion and air pollution in Dhaka city , Journal of Civil engineering,31,165-175
2. Beevers, s. d., Carslaw, D. C., (2005), The impact o congestion charging n vehicle emissions in London , Atmospheric Environment, 39,1-5
3. Kun, C., Lei, Y., (2007), Microscopic Traffic Emission Simulation and case study for Evaluation of Traffic control strategies, Journal of Transportation system Engineering and Information Technology, 7(1), 93-100.
4. Nesamani, K. S., Subramanin, K. P., (2006), Impact of Real World Driving Characteristics on Vehicle emissions , JSME International Journal, 49(1), 19-26
5. Smit, R., Brown, A. L., Chan, Y. C., (2008), Do air pollution emissions and fuel consumption models for roadways include the effect of congestion in the roadway traffic flow? , Environmental Modelling and Software, 23,1262-1270
6. Yingying, Z., Xumei, C., Xiao,Z., Guohua, S., Yanzhao, H., Lei, Y., (2009), Assessing effect of traffic control strategies on vehicle emissions , Journal of Transportation System Engineering and Information Technology, 9(1), 150-155
7. Zhou, Y., Wu, Y., Yang, L., Fu, L., He, K., Wang, S., Hao, J., Chen, J., Li, C., (2010), The impact of transportation control measures on emission reductions during the 2008 Olympic Games in Beijing, China , Atmospheric Environment,44,285-293