Analyzing the Effect of BRT Policy Strategies on CO 2 Emissions: a Case Study of Beijing

Transcription

1 Analyzing the Effect of BRT Policy Strategies on CO Emissions: a Case Study of Beijing Xumei Chen, Ph.D. Professor (Corresponding Author) MOE Key Laboratory for Transportation Complex Systems Theory and Technology, School of Traffic and Transportation, Beijing Jiaotong University Beijing 000, P.R. China Phone: -0-0; Fax: -0-0; tcxm@.net Lei Yu, Ph.D., P.E. Yangtze River Scholar of Beijing Jiaotong University and Professor of Texas Southern University College of Science and Technology, Texas Southern University 00 Cleburne Avenue, Houston, Texas 00 Phone: --00; Fax: --; yu_lx@tsu.edu and Ying Wang, Graduate Research Assistant MOE Key Laboratory for Transportation Complex Systems Theory and Technology, School of Traffic and Transportation, Beijing Jiaotong University Beijing 000 China Phone: -0-0; Fax: -0-0; 00@bjtu.edu.cn Submitted for Presentation at the rd Transportation Research Board Annual Meeting Washington, D.C. January 0 Word Count: (Text) + 0* (Tables) + 0* (Figures) = Submission Date: July 0, 0 TRB 0 Annual Meeting

2 Chen, Yu, and Wang. 0 ABSTRACT Public transportation systems play an important role in reducing CO emissions from transportation sectors because they deliver low carbon trips per capita. Bus rapid transit (BRT) system, as a newly thriving bus service in Asia and the first, and so far the only, mass transit technology certified under the Kyoto Protocol, has been considered as a crucial solution to achieving low-carbon urban mobility. Different policy strategies in the development of BRT may result in different levels of modal shifts, which affect the amount of reductions of CO emissions from buses. However, few existing studies have been conducted to analyze the effect of BRT policy strategies on CO reductions in China. This paper presents an in-depth analysis of effects of BRT policy strategies on CO reductions based on a case study in Beijing. Potential policy strategies for BRT development are identified and analyzed, under which three scenarios are designed. A CO emissions estimation method suitable for complicated vehicle classes and operating patterns is established to assess the effect of BRT policy strategies on CO emissions. Results indicate that BRT system has a great potential to reduce CO emissions in Beijing if a positive development policy is adopted. Key words: Bus Rapid Transit; Policy Strategy; CO ; Modal Share 0 TRB 0 Annual Meeting

3 Chen, Yu, and Wang INTRODUCTION With the rapid urbanization and economic development, enhancing the environmental sustainability while meeting the demand for mobility is becoming a critical strategic goal in many Asian cities. Cumulative CO emissions are a major cause of global warming. Public transportation systems can play an important role in reducing CO emissions from transportation sectors because they deliver low carbon CO trips per capita. In public transportation systems, BRT is an innovative mode. Due to its ecofriendliness, BRT is considered as one of the best transit strategies to reduce transportation related CO emissions and mitigate the negative environmental impacts, according to a recent analysis (). It is the first, and so far the only, mass transit technology certified under the Kyoto Protocol (). In China, more and more BRT systems have been implemented and put into operations in recent years. Beijing was the first city where a BRT system was ever planned and implemented in China. One of purposes in designing a BRT system was to mitigate the greenhouse gas emissions in Beijing. However, the role of BRT in CO mitigation remains under-investigated. There are several frequently asked questions including: () how much net CO is reduced by the BRT system in Beijing from the current level of public transit services being offered; () how much additional CO reductions can be achieved if different development policy strategies for BRT system are adopted; and () what would be the significance of non-public transportation commuters being attracted to use the BRT system? In this context, this research is intended to conduct an in-depth analysis of effects of BRT policy strategies on CO reductions based on a case study in Beijing. Potential policy strategies for BRT development are identified and analyzed, under which three scenarios are designed. A CO emissions estimation method suitable for complicated vehicle classes and operating patterns is established to assess the effect of BRT policy strategies on CO emissions. Analysis results are discussed and strategies for increasing BRT's contribution to greenhouse gas reductions are suggested. EXISTING STUDIES In recent years, some studies have been conducted to assess the impact of BRT vehicles on greenhouse gases or other pollutant emissions, which provide valuable insight into the impacts of BRT development policies and strategies on the production of greenhouse gas emissions. Vincent and Jerram () examined BRT as a near-term strategy for reducing CO emissions in a typical medium-sized U.S. city. They compared expected CO emissions from three scenarios to meet the city s growth in work trips by 0 and calculated a CO emissions inventory for each scenario. They found that BRT offers the greatest potential for greenhouse gas reductions and BRT vehicles generally offer lower CO emissions per passenger mile than Light Rail Transit. They suggested that further study to enhance a methodology to estimate expected CO reductions with BRT would be valuable. The work of McDonnell et al. () analyzed the role of BRT as a tool for mitigating transport related CO emissions. A Quality Bus Corridor (QBC), implemented in Dublin, Ireland, in, was selected as a case study. BRT is in keeping with the concept of a QBC. McDonnell et al. estimated CO emissions associated with differing levels of bus priority for the period -00 and for the Kyoto commitment period (00-0). They found that, in the absence of a QBC, TRB 0 Annual Meeting

4 Chen, Yu, and Wang peak-time emissions for the sample population would have been 0% higher than in the factual scenario. Shi () analyzed the characteristics of residential trips in Beijing. He assumed three BRT development scenarios. Air pollutant emissions including HC, CO, and NO X under different scenarios were estimated and BRT s impacts on emissions were analyzed. In this study, CO emissions were not analyzed. Hook et al. () attempted to develop a common methodology for estimating CO abatement potential of BRT projects at the project level. They used three BRT systems: Bogotá, Mexico City and Jakarta in their studies. The CO estimation methodologies adopted in these three BRT systems were compared and the accuracy of different estimation methodologies was discussed. Trigg and Fulton () analyzed three different scenarios for the increased use of BRT by 00, compared to a business-as-usual scenario, to estimate impacts of BRT on CO and its cost-effectiveness as a CO mitigation intervention. The International Energy Agency s mobility model was used as an integrated modeling framework to analyze the CO benefits of the worldwide deployment of BRT. In order to properly analyze BRT, a comprehensive database of all BRT systems in the world was used. Among scenarios, cumulative reductions of CO emissions were estimated to be - % in the transportation sector by 00. Annual savings of CO emissions in the year of 00 were estimated to be in the range of -%. Studies identified in the literature review have offered insights into the quantified assessment of BRT s role in greenhouse gas reduction and factors contributing to the CO reduction. They suggested that the modal shift is a key factor. Different scenarios are designed to be concurrent with different development policy strategies. A methodology to estimate expected CO reductions from BRT is needed. However, little effort has been made to analyze the BRT s potential to mitigate the CO emissions in China, especially in Beijing. The work in this paper has expanded the previous research on assessing the role of BRT in CO reductions by using the densely populated Beijing as a case study. The methodology of the modal-share-based estimation for CO emissions for different modes is implemented. Recommendations for development strategies for the BRT system in Beijing are presented. METHOD AND APPROACH Estimation method for CO emissions The transportation system is highly interconnected. CO emissions from the transportation sector are influenced by a group of factors, such as the driving cycle, fuel category, modal structure, and passenger travel activities. The relative importance of each of factor to total changes in emissions varies with different levels of estimation (, ). At a city level, only the most important and easy-to-monitor factors should be used, including the modal share, load factor, and CO emission factor. The function used to estimate CO emissions from different modes at the city level is formulated in the following Equation (): Ei N* Si* Di* Mi () Where E i = CO emissions for the mode i (metric tons), N = residential trips (trips), S i = modal share for mode i (%), D i = average trip distance for mode i (km), and M = average CO emissions per capita per km for mode i (g). i TRB 0 Annual Meeting

5 Chen, Yu, and Wang Average CO emissions per capita per km for different modes can be derived from CO emission factor, load factor, and rated passenger capacity, as shown in the following Equation (): EFi Mi () L i* P i Where EF i = CO emissions factor for mode i (g/km), L i = load factor for mode i, which determines the average occupancy (%), and P i = rated passenger capacity for mode i (persons). Therefore, total CO emissions in a city can be further calculated by Equation (): EFi E N* Si* Di* () i L i* P i Where E = total CO emissions in a city (metric tons). Approach to determine emission rate In this research, several modes including conventional bus, BRT, subway, taxi, and car are considered. A CO emission factor can be calculated using the standard Intergovernmental Panel on Climate Change (IPCC) coefficients or other energy conversion factors used to convert fuel (or electricity) consumption to carbon emissions (). However, in cities of developing country, vehicle numbers are growing rapidly, and their patterns of usage are changing as cities sprawl outwardly. Except for well defined, centrally operated vehicle fleets, the transportation sector fuel consumption data is physically difficult to be collected due to the highly decentralized decision making process for transportation activities (). Hence, we use electricity-consumption-basedmethod to estimate the CO emission factor only for subway, which uses electricity as the power. We do not choose a fuel-consumption-based approach to estimate CO emission factor for the fuel-intensive modes of transportation (i.e., conventional bus, BRT, taxi, and car) because such data are not easily collected and not reliable either. An emission-model-based approach to determine the emission rate for fuel-intensive modes of transportation has been used in this research. This approach can capture the speed, idling, or acceleration characteristics of vehicles, which reflects real operating patterns in the roadway network. For subway, the CO emission factor is determined in following steps. To produce the electricity used by the subway operation,,00 tons of standard coal has been consumed per year recently in Beijing according to the statistics from Beijing metro system (, ).. tons of CO will be emitted when one ton of standard coal is burned (0, ). Multiplying the. by the standard coal consumption of,00 tons, we derive that Beijing subway produce,. tons of CO emissions annually. In 00, the total subway mileage was 0. billon vehicle kilometers according to 0 Beijing Transport Annual Report (). Dividing,. tons of annual CO emissions by 0. billon vehicle kilometers yields,. grams of CO per kilometer. For the conventional bus, BRT, taxi, and car, a portable emissions measurement system (PEMS) system (OEM-00) has been used to collect real time CO emission data and a GPS device to collect driving activity data. More than 00 million groups of data were collected. The vehicle specific power approach is used to establish the emission estimation model, which determines the CO emission factor (). Resulting CO emission factors for different modes are summarized in Table. TRB 0 Annual Meeting

6 Chen, Yu, and Wang TABLE CO Emission Factors for Different Modes Modes Conventional bus BRT Subway Taxi Car CO emission Factors (g/km). 0.0,.. 0. ANALYSIS OF DIFFERENT BRT DEVELOPMENT POLICY STRATEGIES Factors that influence BRT development policy strategies in Beijing Various factors affect BRT development policy strategies, which further impact CO emissions. The BRT system in Beijing only carries.% of passenger trips in the urban transportation system. BRT is underdeveloped and its role to provide large scale highquality services has not been fully realized at present. In the future, economic, political, land use, transit priority, and urban transportation goal factors may all influence BRT development policy strategies in Beijing. The first BRT line (BRT Line ) was opened in 00. From then on, sufficient financial support was available to open BRT Lines and due to Beijing 00 Olympic Games. However, currently the economic growth slows down; there will be no largescale special events equivalent to Olympic Game to be held in the near future; and the land price has increased to a considerably high level. Meanwhile, the reality of the limited roadway space and rapidly deteriorating traffic congestion has pressured the government to reconsider transit priority strategies and the license plates controlling policy. A stricter emission control standard will be implemented gradually with the 00 as the target year. In view of impacts from these factors, three possible BRT development policy strategies are identified for years up to 00 in Beijing. () Conservative development policy strategy Beijing will maintain a steady economic growth from now to 00. There will be no significant development policy strategy provided for the BRT systems and no more travel demand management measures undertaken for cars. Urban sprawl will not be effectively controlled. The development of the BRT system maintains a natural growth trend because Let it be strategy is adopted. Load factors for different modes remain same levels. Stricter control regulations for CO emissions from both BRT vehicles and other vehicles will be implemented. It is assumed that such regulations will reduce the CO emission factor by -%. () Proactive development policy strategy Due to the rapid economic growth and strong land use control, the BRT system development will be enhanced by the Beijing municipal government since it is a financially affordable system. Travel demand management measures, such as congestion toll collection and low emission zone, will be implemented. No competitive local services will be provided along BRT routes. A good publicity on BRT will encourage more usage of BRT systems. The BRT becomes a main component of the ground transportation system in Beijing and a BRT network with a comprehensive coverage is developed. Load factors for both BRT and conventional bus systems are improved because service levels increase. Compared with the first policy strategy, the CO emission factor for BRT vehicles and other vehicles is assumed to decrease by -%. () Moderate development policy strategy The economy of Beijing will develop at a moderate speed. BRT systems have been defined as a role to supply the feeder services supplemental to subway systems in TRB 0 Annual Meeting

7 Chen, Yu, and Wang the urban transportation strategy. Travel demand management measures which promote private mode users to switch to public modes will be implemented. Mixed-use and highdensity development patterns along structural axes will encourage the usage of subwayconnected-brt-services. More and more communities will be attracted to such services. Hence, load factors for both BRT and subway systems are increased. It is assumed that the impetus to maintain a sustainable and eco-friendly urban transportation system makes CO emission factor for all modes decrease by -% with better running conditions and vehicle technologies. Scenario design under different BRT development policy strategies Because of the uncertainty of BRT development policy strategies, scenario-based analysis can be a good option. This paper relies upon ridership, mode-shift, and other parameters in Equation () to design scenarios. These parameters will be estimated under different BRT development policy strategies for the year of 00. Actual operation and performance data are collected to estimate these parameters. Based on the estimated parameters, three BRT development scenarios will be designed and their resulting CO emission will be further analyzed. In Equation (), several parameters including residential trips N, average trip distance D i, and rated passenger capacity P i, are assumed unchanged values under three scenarios in 00. They need to be estimated for 00. Residential trips N will be estimated for 00 using an empirical equation, as shown in Equation (). T ' N ' N0* * a () T 0 Where N ' = residential trips for 00 (trips), N 0 = residential trips for base year 00 (trips), T ' = average trips per day per capita for 00 (trips), T 0 = average trips per day per capita for base year 00 (trips), and a = ratio between urban land area for 00 and base year 00. According to 00 Beijing Household Travel Survey and Urban Master Plan of Beijing (, ), N 0 =,,00,000 (trips), T ' =.0 (trips), T 0 =. (trips), and a =.. We can then derive residential trips for 00 N ' as. billion (trips). With the increase of the residential travel demand, the average trip distance will increase due to the extension of urban rail transit network. We analyze the average trip distance in 000, 00, and 00. From 000 to 00, the average trip distance for bus and taxi did not change significantly with only a slight fluctuation and the average trip distance for cars and subway increased by.% and.% respectively. These increasing rates have been used for estimating the average trip distance from 00 to 00. Therefore, we can obtain the average trip distance Di for all modes for 00. Rated passenger capacity P i is related to the vehicle design. We assume that this parameter will not change in the near future since no significant changes on vehicle seats or space size are predicted. Other parameters in Equation (), such as, modal share S i, CO emission factor EF i, and load factor L i, are different under three scenarios. These parameters need to be estimated for different scenarios in 00 respectively. Modal share values from 00 to 00 are analyzed to estimate the modal share S i in 00. From 00 to 00, the modal share for buses including BRT increased slowly, but decreased slightly in 00. The modal share for the subway TRB 0 Annual Meeting

8 Chen, Yu, and Wang. 0 0 changed with an average growth of %, the modal share for taxi decreased with an average rate of %, and the modal share for cars increased with an average growth of.%. Modal shares in 00 are set in three scenarios in Table. In order to better analyze effects of BRT development policy strategies on CO emissions, an increase range of to % for the BRT modal share is designed in Scenario. Modal shares for other modes in Scenario have been changed with the average change rate from 00 to 00. For Scenarios and, the model share for public transit system will increase considering the improvement of public transit infrastructure in the future. The modal share for BRT in Scenario increases within a range of to %. In Scenario, a modal share of 0% which reflects trips using subway-connected-brt-services is considered. Beijing has begun to introduce Phase emission standards recently. CO emission factor EF i under all three scenarios will be lower in consideration of the implementation of stringent emission control standards, the enhancement of fuel efficiency, and the improvement of engine technologies. Table shows the EFi values for different scenarios. Load factor L i was also assumed according to the identification of three possible BRT development policy strategies in 00. As shown in Table, the load factor under Scenario does not change. Under Scenario, the load factor for the conventional bus increases by 0% and the load factor for BRT increases by %. Load factors for both BRT and the subway are increased by % under Scenario. TABLE Parameters for Different Scenarios Unchanged parameters for three scenarios Different parameters for scenario Different parameters for scenario Modes Conventional bus BRT Subway Taxi Car N ' (trips),,0,000 D i (km) P i (persons) 0 0 S i.00% -%.%.%.% EF i (g/km) L i.00%.00%.00% 0.0%.0% S i.00% -.00%.%.% 0.00% EF i (g/km) L i 0.00%.00%.00% 0.0%.0% Different S i.00% -.00% 0.00%*.0%.0% 0.00% parameters for EF i (g/km) scenario L i.00%.00% 0.00% 0.0%.0% Note: 0.00%* represents a modal share in which the trips use subway-connected-brt-services. In such trips, the trip distances for BRT and subway are and km respectively. TRB 0 Annual Meeting

9 Chen, Yu, and Wang. 0 0 RESULTS AND ANALYSIS CO emission in has been chosen as the base year. 00 Beijing Household Travel Survey shows that Beijing residents made. billion annual trips in 00. Other parameters are listed in Table and results on CO emission are shown in Table. TABLE Parameters for Estimate 00 Beijing CO Emission in Beijing Modes Conventional bus BRT Subway Taxi Car S i.%.%.%.%.% D i (km) EF i (g/km). 0.0,.. 0. L i % % % 0.%.% P i (persons) 0 0 Note: Modal share S i and average trip distance D i are from 00 Beijing Household Travel Survey. Load factor L are from the survey by public transportation agency in Beijing. i TABLE CO Emission Results for Different Modes in 00 Modes CO emission (metric ton) CO emission per capita (g/person) Conventional bus BRT Subway Taxi Car Total,0,0 0,0,0,0,,00,, Percentage.% 0.%.%.%.% 00.00% As presented in Table, CO emissions for the transportation sector in Beijing have reached,,0 metric tons. This means an average CO emission per capita of 0.g/person. Cars emit.% of CO emissions, which are substantially higher than those from the conventional bus, BRT, and subway. CO emissions from BRT are the lowest among all modes. It can be observed that the public transportation system, especially the BRT system, has a great potential for CO emission reductions. Hence, CO emissions from BRT under different BRT policy strategies will be analyzed and the effect of BRT policy strategies on CO emissions in Beijing will be assessed in the following section. CO emissions under different development policy strategies CO emissions under different development policy strategies are calculated using Equation (). Figures (a) and (b) illustrate the results for 00 and three scenarios for 00. TRB 0 Annual Meeting

10 Chen, Yu, and Wang. CO Emission CO Emission (Metric Ton) Base Year,,0 00 Scenario,,00 (+.%),,0 (+.%),,0 (-.%),0,0 (-.%) 00 Scenario 00 Scenario,,0 (-.%),0,0 (-.%) (a) CO Emission Per Capita CO Emission Per Capita (g/person) Base Year Scenario.0 (+.%). (+.%) 00 Scenario. (-.%). (-.%) 00 Scenario. (-.%). (-.%) 0 (b) FIGURE Comparison of CO emissions for 00 and three scenarios in 00. As shown in Figure, when the conservative development policy strategy for BRT is adopted (BRT modal share increases within a range of to % and modal shares for other modes in Scenario are changed in a natural growth), which corresponds Scenario, CO emissions will increase to,,0-,,00 metric tons. Compared with CO emissions in 00,,,0-,0,0 metric tons are added with an increase of. to.%. Meanwhile, the average CO emission per capita also increases from 0.g/person to.-.0 g/person with an increase of. to.%. It can be observed that CO emissions from the transportation section in Beijing will increase significantly when no effective measures to encourage modal shift towards public transportation modes including BRT are implemented. When a proactive development policy strategy for BRT system is implemented, the BRT network can be extended at a rapid speed. With a better marketing speed and TRB 0 Annual Meeting

11 Chen, Yu, and Wang an improved reliability, the BRT modal share will increase and the load factor will be improved. Total CO emissions can reach,0,0-,,0 metric tons. This is a decrease of,0-,,0 metric tons. The percentage of decrease is. to.%. The average CO emission per capita declines to.-.g/person with a. to.% reduction. It can be concluded that when the development of BRT is paid more attention and its role in Beijing ground transportation system is defined as the main public transportation mode, there can be a decrease of CO emissions although the travel demand increases. Moreover, the average CO emission per capita can be reduced significantly, which facilitates a transition towards a low-carbon or decarbonized urban mobility essentially. If the Beijing government initiates a moderate development policy strategy for BRT, the BRT system will be considered as the supplement of the subway, in which BRT routes collect and distribute passenger flows for the subway system. Then, total CO emissions can vary from,,0 to,0,0-,,0 metric tons. Compared with CO emissions in 00,,,0-,,0 metric tons of CO emissions are reduced with a decrease of. to.%. The average CO emission per capita can decrease to.-. g/person, which represents a decrease of. to.%. It can be observed that promoting BRT as the supplement of the subway system and shifting mode to high quality multimodal and integrated public transportation systems have very high contributions to reducing CO emissions in the urban transport system. This facilitates low carbon trips for Beijing. CONCLUSIONS Low carbon mobility with low energy consumption is essential for a sustainable and competitive future for Asian cities. Different BRT development policies and strategies implemented by the governments will yield different results in emissions of carbon dioxide from the urban transportation sector. In this paper, we employed a scenario analysis approach to the case study of Beijing. A CO emission estimation method suitable for Beijing that has complicated vehicle classes and operating patterns, has been established to assess the effectiveness of the BRT policy in reducing CO emissions. Factors that influence BRT development policy strategies in Beijing are analyzed and three potential policy strategies for BRT development are identified. Accordingly, three scenarios in the target year of 00 are designed and results are analyzed. Urban travel demand in Beijing is estimated to increase by.% from 00 to 00, which brings a great challenge to mitigating CO emissions. The analysis in this paper suggested that if the conservative development policy strategy for BRT is adopted (BRT modal share increases within a range of to % and modal shares for other modes are changed in a natural growth), there will be a significant increase in CO emissions. A proactive development policy strategy for the BRT system can be implemented to control CO emissions with significant effects. Significant CO emission reductions can be resulted by a moderate development policy strategy for BRT, in which BRT systems have been defined as a role to supply the feeder services to the subway system. Most of reductions are achieved by an attractive intermodal public transit system. We can conclude that the BRT system has a great potential to reduce CO emissions for Beijing. As a result, it is recommended that positive development policies for BRT be adopted in Beijing. BRT can be deployed more quickly, and in greater quantities, than the urban rail systems. This increases opportunities to shift trips away from private modes. Whether the BRT system is considered as a main public TRB 0 Annual Meeting

12 Chen, Yu, and Wang transportation mode or a supplement of the subway system, CO emissions can be controlled and significant CO emission per capita reductions can be delivered. Such policies and related measures are comparatively at low costs. The analysis presented here is limited to a set of data collected in Beijing. It should be noted that localized factors, such as electricity generation mix, driving activities, emission and efficiency standards (), and travel demand management policies, will all affect the results in a particular city. The study attempts to develop a methodology for quantifying CO emissions for those cities facing similar problems in Beijing. Future studies should consider using better and more extensive data to determine the sensitivity of parameters on estimating CO emissions, thereby offering more practical strategies to monitor and improve the control of CO emissions. The CO emission analysis for other cities that have different development patterns should also be explored. ACKNOWLEDGEMENTS The authors acknowledge that this paper is prepared based on NSFC #0, National Basic Research Program of China (No. 0CB0), Program for New Century Excellent Talents in University (NCET--0), and the Fundamental Research Funds for the Central Universities (No. 0JBM0). This research is partially supported by the National Science Foundation (NSF) under grant #. REFERENCES. Vincent, W. and L. C. Jerram. The potential for bus rapid transit to reduce transportation-related CO emissions. Journal of Public Transportation, BRT Special Edition, 00, Vol., No., pp. -.. Alameda-Contra Costa Transit District. Why BRT. ng-focus/your-guide-to-bus-rapid-transit/why-brt/. Accessed Nov., 0.. McDonnell, S., S. Ferreira, and F. Convery. Using bus rapid transit to mitigate emissions of CO from transport. Transport Reviews, 00, Vol., No., pp. -.. Shi, L. Study of the impact of BRT on vehicular emissions in Beijing. Graduate Thesis, Beijing Technology and Business University, 00.. Hook, W., C. Kost, U. Navarro, M. Replogle, and B. Baranda. Carbon dioxide reduction benefits of bus rapid transit systems: learning from Bogotá, Colombia; Mexico City, Mexico; and Jakarta, Indonesia. Journal of the Transportation Research Board, No., Transportation Research Board of the National Academies, Washington, D.C., 00, pp. -.. Trigg, T. and L. Fulton. Bus rapid transit: cost and CO implications of future deployment scenarios. Proceedings of the st Annual Meeting of the Transportation Research Board, Washington, D.C, Jan 0.. Schipper, L., M. Cordeiro, and Ng. Wei-Shiuen. Measuring CO impacts of urban transport projects in developing countries. Proceedings of the th Annual Meeting of the Transportation Research Board, Washington, D.C. Jan 00.. Beijing Mass Transit Railway Operation Corporation. Report on energy consumption of Beijing subway, Beijing. 0. TRB 0 Annual Meeting

13 Chen, Yu, and Wang Beijing Mass Transit Railway Operation Corporation. Build an energy-saving subw ay. July, 0. Accessed Nov., Beijing Municipal Commission of Development and Reform. 00 annual evaluatio n of coal consumption in power generation enterprises is started. Aug., 00. htt p:// Accessed Nov., 0.. U.S. Energy Information Administration. Voluntary Reporting of Greenhouse Gases Appendix F. Electricity Emission Factors. U.S. Department of Energy. OMB No. 0-0, Washington, D.C Wang, G.C. 0 Beijing Transport Annual Report. Beijing Transportation Research Center, 0.. Xu, Y.F., L. Yu, and G.H. Song, Improved Vehicle Specific Power Bins for Light- Duty Vehicles in Estimation of Carbon Dioxide Emissions in Beijing. Proceedings of the th Annual Meeting of the Transportation Research Board, Washington, D.C. Jan 00.. Beijing Transportation Research Center. The fourth Beijing household travel survey. Beijing, 0.. Beijing Municipal Commission of Urban Planning. Urban Master Plan of Beijing (00-00). Beijing, 0.. Lutsey, N. Regulatory and technology lead-time: The case of US automobile greenhouse gas emission standards. Transport Policy, 0, Vol., pp. -0. TRB 0 Annual Meeting