Scenario analysis of the energy demand and CO₂ emission reduction potential of the urban transport system of Beijing through 2030

Abstract

An assessment of the energy demand and the potential for sector-based emission reductions will provide necessary background information for policy makers. In this paper, Beijing was selected as a special case for analysis in order to assess the energy demand and potential of CO₂ abatement in the urban transport system of China. A mathematical model was developed to generate three scenarios for the urban transport system of Beijing from 2010 to 2030. The best pattern was identified by comparing the three different scenarios and assessing their urban traffic patterns through cost information. Results show that in the high motorization-oriented pattern scenario, total energy demand is about 13.94% higher, and the average CO₂ abatement per year is 3.38 million tons less than in the reference scenario. On the other hand, in the bus and rail transit-oriented scenario, total energy demand is about 11.57% less, and the average CO₂ abatement is 2.8 million tons more than in the reference scenario. Thus, Beijing cannot and should not follow the American pattern of high motorization-oriented transport system but learn from the experience of developed cities of Europe and East Asia.

Keywords

scenario analysis / urban traffic pattern / energy demand / reduction potential

Introduction

The Copenhagen World Climate Conference (COP15) concluded on December 18, 2009. During the conference, there were difficult negotiations between the developed and developing countries on emissions reduction and cost-related problems. Many research institutions and nongovernment organizations have conducted their respective studies on climate change. Although these are varied, there is a common emphasis on the analysis of the potential for sector-based emissions reduction among these studies. Many researchers on climate change have focused on key sectors, including the electricity and heat sectors, transportation, building industries, agriculture, as well as major industrial sectors, such as steel, chemicals and petrochemicals, aluminum, and cement industries. Of these, the transport sector is one of the most important.

Many studies have concentrated on solving urban traffic congestion or on the relationship between traffic demand on one hand and population, urbanization, and economic factors on the other hand [1-3]. However, other studies have also analyzed CO₂ abatement measures within the transportation sector [4,5]. These studies have applied different methodologies, ranging from engineering perspectives to top-down approaches. They usually analyze the different characteristics of the transport sector, including transport energy use and greenhouse (GHG) emissions trends in specific countries or cities, international comparison of the emission trends, fuel economics, urban traffic patterns, and climate change policies.

Many domestic studies have been conducted on energy conservation and CO₂ reduction in the transport sector in China [6,7]. Hu and Jiang [8] used the AIM model to evaluate the technology and countermeasures for GHG mitigation in China. Zhou [9] depended on the LEAP platform to generate energy consumption scenarios in 2020 and analyze the emission trends under different scenarios. The transport sector served as a key component of their studies.

With the rapid growth of urban motor vehicles, vehicle exhaust emissions have become major sources of urban air pollution, even in developed countries where the technology is advanced and motor vehicle emissions are controlled rigorously. For example, among the major sources of emissions in the US, 84% of CO and 42% of NO_X come from motor vehicle exhaust. In Europe, motor vehicle exhaust contributed about 76% of CO and 36% of NO_X [10]. In 1997, motor vehicles in Beijing emitted 67% of CO and 41% of NO_X, and those in Shanghai emitted 61.8% of CO and 20.9% of NO_X. In 2000, the percentage of CO and NO_X emissions increased to 83% and 43% [11], respectively, which is close to the level recorded abroad in the 1960s and 1970s. In 2001, the daily average rate of NO_X within the 3rd Ring Road, between the 3rd Ring Road and the 4th Ring Road, and outside of the 4th Ring Road in Beijing exceeded 43.9%, 41%, and 35%, respectively; meanwhile, the daily average rate of CO exceeded 38.9%, 17%, and 10%, respectively. Vehicle exhaust emissions, accounting for 50% of gaseous pollutants in the air, have become a major factor affecting the air quality in Beijing.

The Environment and Transport Working Group points out that the atmospheric pollution contribution from motor vehicles exhaust emissions in some large cities has reached 60%. The transport sector is also one of the largest sources for CO₂, which is the most important anthropogenic GHG [12]. According to statistics, global transport energy consumption accounted for about 28% of the total energy consumption in the world, and its emission volume accounted for about 23% in 2007 [13]. Currently, the urban transport energy consumption of China accounts for about 10%-15% of the total energy consumption of the country, and this figure is projected to increase in the future. Without control, the energy consumption of this sector will reach about 25% of total energy consumption, surpassing that of industrial consumption by that time. Vehicle exhaust emissions will be the next major source of pollution in Chinese cities because they currently account for about 20% of the total emissions in the country.

The urban transport development mode of Beijing will have a great effect on those of other cities in China. The total number of motor vehicles in Beijing is predicted to reach 5 million in 2011 and exceed 6 million before 2015. In addition, the current 6 million households in the city is predicted to surpass more than 7 million in the near future. If the other cities will adopt the urban transport development pattern prevailing in Beijing, the total number of motor vehicles in the country will reach 700 million in 2030. By that time, the current oil supply will not be able to meet the transport energy demand, and vehicle exhaust will be the largest source of GHG emissions. Therefore, choosing Beijing as a case study and identifying its future urban transport sustainable development pattern will be of great significance. The case of Beijing offers valuable reference to other Chinese cities and could play an important role in response to climate change.

In this paper, the urban traffic patterns of different cities in the world are also analyzed. The high-priority development traffic pattern of the urban transport system in Beijing is identified using the scenario analysis method. This paper aims to provide relatively abundant information to support the discussion on the transport system of Beijing and its role on the entire transport sector of China.

Urban traffic in Beijing

As one of the largest capital cities in the world, Beijing is currently suffering from serious traffic congestion. With the continuing economic growth, transport and energy demands, as well as the resulting air pollution, are on the rise. Figure 1 shows the annual growth of motor vehicles and private cars in Beijing since 1995. From 2000, the number of private cars increased annually by over 10% from 855000 in 2000 to 3200000 in 2009. Traditionally, the travel demand in Beijing can be distinguished into five modes: cars, taxis, public transport (e.g., bus, light rail, and subway), bicycle, and on-foot travel. Figure 2 illustrates the modal split surveys conducted in 1986, 2000, 2005, and 2008. During the 25 years covered by the surveys, the share of public transport in total demand has only decreased slightly from 32% in 1986 to 26.5% in 2000. However, due to greater investments in subway construction as well as public transport system optimization, the share of public transportation in total travel demand has increased from 26.5% in 2000 to 29.8% in 2005. Just before the 2008 Beijing Summer Olympic Games, there were already five new subway lines in use. Presently, the total length of the Beijing subway exceeds 200 kilometers. The share of public transport has increased from 29.8% in 2005 to 34.2% in 2008. The major modal shifted from riding bicycles to driving cars, at a rate of 20% in the first 20 years. In the past five years, however, the share of private cars has been basically stable.

The travel demand figures and transportation structure in different stages can be compared based on travel survey data culled from Beijing residents (Table 1). In 2008, the per capita transport trip generation rate was 2.91, and total travel was 34 million person times, representing a 16.4% increase from 2005 (Table 1).

Methodology

Frame of model

Let D_t be the total urban traffic demand, which can be denoted as

(1)D_t=\mathrm{365}{p}_t{n}_t{d}_t,

where t\in \{\mathrm{2010},\mathrm{2011},\cdots ,\mathrm{2030}\} denotes the year, p means resident population, n means per capita trip generation rate (the number of trips for one person in one day), and d means per capita trip distance.

Let E_t be the energy demand of the urban traffic for a certain year, which is equal to

(2)E_t={\displaystyle {\sum }_m{\displaystyle {\sum }_q{D}_t{S}_{m,t}{\rho }_{q,t}}},

where\rho denotes the energy consumption for every 100 kilometers (fuel economy) and

(3)S_{m,t}=m_t{q}_t,

where m means the proportion of each of the different traffic modes. In this paper, three traffic modes, travelling by car (private vehicles, motorcycles, and taxicabs), travelling by public transport (buses and urban rail transit), and travelling by person (by walking and riding bicycles) are defined. These will be described in detail in Section 3.2. In the above, m\in \{\mathrm{1},\mathrm{2}\}; when m=\mathrm{1}, it means travelling by car, otherwise, it means travelling by public transport. Given that the third traffic mode is travelling by person, its energy demand could be excluded in the study. Finally, the symbol q means the ratio of the different travel modes to corresponding traffic pattern and q\in \{\mathrm{1},\mathrm{2},\mathrm{3},\mathrm{4}\}. The different modes and their corresponding number codes are listed in Table 2.

Urban traffic patterns

Total traffic demand can be classified using the frequency distributions of different travel modes, with reference to the urban traffic structure. A person can choose to drive a car, take a bus, take the subway, walk, or ride a bicycle; an individual can also choose a mixed mode for his travel one at a time in the city. However, the different travel modes can be mainly classified into three, namely, travelling by car (private vehicle, motorcycle, and taxi), travelling by public transport (bus and urban rail transit), and travelling by person (walking and bicycle). After conducting an investigation on 30 cities in the world, and according to the different frequency distributions of the three modes mentioned above, we arrive at four urban traffic patterns, labeled A, B, C, and D. These are described below:

A. High motorization-oriented pattern (the percentage of the travelling by car mode is greater than 50%).

B. Public transport-oriented pattern (the percentage of the travelling by public transport mode is greater than 50%).

C. Cycling and pedestrian-oriented pattern (the percentage of the travelling by person mode is greater than 50%.

D. Uniform pattern (the percentage of each of the above three modes is less than 50%).

Figure 3 shows 30 trip structures in large cities in America, Europe, East Asia, and China.

In pattern A, the ratio of travelling by car is very high because Americans prefer driving to walking. The percentage of travelling by car in Orlando, Houston, and Los Angeles exceed 90%, and that in Seattle, Washington DC, San Francisco, and Chicago exceed 80%. Although the public transportation in New York City is very modern, the share of travelling by car is still close to 70%. The share of travelling by person is less than 10%, except in Berkeley and UC Davis. As a result of bicycle policies, the share of travelling by person increased to about 22% and 40% in Berkeley and UC Davis, respectively.

In pattern B, the percentage of travelling by public transport in Dalian, a unique city in China, is just above 50%. Due to the advanced urban train system in Tokyo, and the well-developed bus and rail transit system in Hong Kong, the shares of travelling by public transport in Tokyo and Hong Kong are both higher than 60%. The share of travelling by person in the cities in pattern B are all higher than those in pattern A and are mostly concentrated in the range between 15% and 30%.

Most of the large cities in Europe belong to pattern D. Beijing, another unique city in China, has been in pattern D since 2000, but the proportion of travelling by car is higher than in Basel and Copenhagen, which reached 26%. The shares of travelling by person in Copenhagen, Amsterdam, and Berlin fall within the range of 35%-45% and that in Basel is close to 50%. Although the percentage of travelling by car is close to 50% in Paris and London, the shares of travelling by person are 15% and 24%, respectively.

Several large cities in China currently follow pattern C. Travelling by person dominates in these cities, and the percentage is mainly concentrated within the range of 60% to 80%. Even in Shanghai, the percentage is close to 60%.

Evidently, the proportion of the trips between patterns A and C are nearly opposite in Fig. 3. In order to grasp the development trend and constraints of pattern C, Beijing was selected as the special case for in-depth study for energy demand, GHG emission, and cost considerations to provide references for policy makers as they try to develop the urban transport system in China.

The trip characteristics of Beijing

In Fig. 3, the dash line denotes the trend of the urban traffic pattern in Beijing from 1985 to 2008. In the past 25 years, the traffic pattern in Beijing has changed profoundly from pattern C to pattern D. In the future, the environmental protection and the energy security issues will be major concerns.

Figure 4 presents the development of the indexes of per capita taxis, per capita public transport, per capita private cars, and per capita motor vehicles from 1986 to 2008. All of the indicators were normalized in the figure in order to show the trend of changes more clearly.

Figure 4 shows that the urban traffic pattern of Beijing experienced roughly three stages from 1986 to 2008. Details are presented below.

Stage 1 (1986-1995). The rapid growth in taxi service started the procedure of motorization in Beijing. In 1986, the number of taxis in Beijing was only 12451, transporting about 9.96 million passengers. In 1995, the number of taxis reached 59493, transporting about 64.9 million passengers, or an equivalent average of about 1.78 million taxi rides every day. The rapid growth in taxi service was due to a preferential policy for taxis issued by the municipal government of Beijing in 1984. However, the growth of public transport was slow at that time, transporting passengers of 37.16 billion in 1995, experiencing a reduction of 1.7% from 1985. In contrast, the urban rail transit transported passengers of about 55.8 billion, which is an increase of 3.29%.

Stage 2 (1996-2006). Private cars increased rapidly, and Beijing entered into a period of high motorization. Since 1996, the rapid increase in the number of motor vehicles, mainly based on the increasing number of private cars, indicates that Beijing has entered into a period of motorization. In 2006, the number of motor vehicles reached 2.88 million. Of these, more than 2.07 million are private cars, accounting for 71.8% of total vehicles; this is approximately 5.9 times that of 1996. The annual average growth is about 21%. In the past 20 years, the ratio of travelling by bicycle declined sharply, decreasing from 58% in 1986 to 30% in 2005 (Fig. 2).

Stage 3 (2006-). Since 2006, the urban rail transit system in Beijing has developed rapidly, and a decade of motorization development worsened the urban traffic system. However, before the 2008 Beijing Summer Olympic Games, five new subways were built and put into operation. According to the Beijing traffic development program, the mileage of subways will reach 300 kilometers in 2010 and increase to 500 kilometers in 2020.

Scenario analysis

Scenario description

The scenario analysis time extension covers the period of 2010 to 2030, with 2005 as the baseline year.

There are three scenarios generated in this article. These are the reference scenario (Scenario 1), high-motorization scenario (Scenario 2) and well-developed public transport scenario (Scenario 3). The differences between these are listed in Table 3.

Main assumptions and sources in scenario definition

The total urban traffic demand forecast is the first stage of the entire analysis process. Eq. (1) shows that the total demand is related to population, capita trip generation rate, and per capita trip distance. These variables shall be described separately in the following analysis. One important assumption is that all scenarios have the same demand forecast; such an assumption can avoid uncertainties when forecasting demand and help researchers focus on analyzing the differences between the scenarios.

Projecting the total medium- and long-term urban traffic demand of Beijing is difficult. Various studies give different projections on the population of Beijing residents [16, 17]). By comparing different predictions, the scenarios assume that the municipal government would strictly limit migration, that mortality and education conversion rate would change with the plan, and that fertility would be higher than the current level. The population of Beijing residents is forecasted to exceed 17.5 million in 2020 and reach 18 million in 2030.

After the year 2000, the urban areas of Beijing were extended outward, based on the four central districts, forming a ring plus radial road network structure. The spreading of the urban area has stimulated the construction of the ring-road system, and as a result, the newly built ring roads have accelerated the expansion of the city at the same time. With the construction and use of the second, third, fourth, and fifth Ring Roads and due to the soaring price of houses in urban areas, residents have already begun moving to the outskirts of the city (Fig. 5). According to the statistics, the resident population density in 2005 was 995 persons/km², representing an increase of 17.5% from 2002. Moreover, the density of the four core districts was reduced by 28.8% but that of the suburban areas increased by 31.8%. These factors stimulated an increase in the trip distance of residents. In 2008, the per capita trip distance increased to 9.7 kilometers. With this, Beijing entered the motorized travel mode, and this trend is predicted to accelerate further in the future.

Table 4 shows the projected results of resident population (p), trip generation rate (n), and trip distance (d).

Based on the projected results and Eq. (1), the projection in total traffic demand can be obtained, as presented in Fig. 6.

In Fig. 3, the red line shows the trend of changes in Beijing traffic patterns from 1986 to 2008. The detailed assumption is given in Table 5.

Scenario analysis results

The energy efficiencies and CO₂ emissions have been listed in Table 6. By Eq. (2), we could obtain the energy consumption and CO₂ emission.

Energy consumption

Rapid motorization would be the main driving force for energy consumption just before 2020 arrives, and by then, the traffic pattern is expected to play a more important role in total energy consumption (Fig. 7).

Given the different traffic patterns, the total energy consumption will be 23464.43 TJ in 2010 in Scenario 1. This figure is 1145.07 TJ lower than in Scenario 2 and 1440.3 TJ higher than in Scenario 3. The difference is expected to increase further in subsequent years. In 2030, the total energy consumption is projected to be 45599.09, 64273.39, and 29292.76 TJ in Scenario 1, Scenario 2, and Scenario 3, respectively.

The cumulative energy consumption reduction between Scenario 3 and Scenario 1 is approximately 11.57% of the cumulative energy consumption of Scenario 1 from 2005 to 2030, and that between Scenario 2 and Scenario 1 is approximately 13.94% of the cumulative energy consumption of Scenario 1 from 2005 to 2030.

CO₂ emission

CO₂ emissions in Scenario 1 are projected to increase from 10.97 million tons in 2005 to 15.06 million tons in 2010. After this, the growth rate will be fast. The annual CO₂ emissions are projected to reach 21.58 million tons in 2020 and 29.29 million tons in 2030.

A similar trend is present in Scenario 2. In 2010, the emissions are projected to reach 15.67 million tons. This figure is 0.6 million tons higher than Scenario 1. The emissions are projected to reach 25.34 and 37.27 million tons in 2020 and 2030, respectively (the differences between these two scenarios is 3.76 and 7.98 million tons). The cumulative CO₂ emission difference between these two scenarios from 2005 to 2030 is 84.05 million tons (Fig. 8).

In Scenario 3, the growth rate is predicted to be the slowest. CO₂ emissions are expected to increase from 14.4 million tons in 2010 to 22.01 million tons in 2030. Compared with Scenarios 1 and 2, the cumulative CO₂ emission reduction from 2005 to 2030 would be 69.75 million tons and 153.8 million tons, respectively.

Cost

When assessing the three different scenarios from the perspective of energy demand and emission reduction potential, it is necessary to consider the corresponding cost. The total transport cost consists of infrastructure, operating, external, and travel-time costs. Infrastructure cost includes the cost for building and fixing roads. Operating cost includes fixed cost and variable cost. External cost includes traffic accident, air pollution, noise costs, etc. [18]. In this regard, policy makers in Canada have developed a software to analyze transport cost [19, 20]. Through the investigation of operation, infrastructure, travel time, and external costs, the aggregate transport costs of Beijing are listed in Table 7.

Figure 9 indicates the total transport cost difference between Scenarios 2 and 1, as well as between Scenarios 3 and 1. In Scenario 2, the overall traffic cost is expected to reach 492 billion dollars in 2020 and 680 billion dollars in 2030, representing an increase of 12.09% and 24.2%, respectively, compared with Scenario 1. On the contrary, in Scenario 3, the overall traffic cost is expected to reach 371 billion dollars in 2020 and 438 billion dollars in 2030 and a decrease of 18.2% and 25.2%, respectively, compared with Scenario 1. Although the initial investment in public transport is greater than in other modes of transportation, its service life is much longer, and its capacity much greater. Therefore, the average cost of public transport is less than that in other travel modes. Table 7 demonstrates clearly that the external cost accounts for a big proportion of total transport cost, and therefore, a higher percentage of private car travel could inevitably lead to a significant increase in the total cost. Although the above cost forecasts still contain considerable uncertainty, the overall trend is evident.

Assessment of the urban traffic pattern

As illustrated in Fig. 3, the urban traffic patterns of large cities in North America, Europe, and East Asia are different. Apart from traffic pattern A, there are patterns D and B in developed countries. In order to grasp the direction of traffic pattern C, patterns A, B, and D will be assessed.

In terms of energy supply, environment capacity, and social traffic cost, pattern A found in the developed cities in the US should be abandoned. Currently, the per capita share of energy in China is low, but the total volume of GHG emissions in the country is very high. If all the cities in China will develop in the way of pattern A, the energy security of the country will be threatened, and the country will be placed at a disadvantageous position in response to climate change.

At present, the trend in urban sprawl in Beijing has begun, and the population is densely packed in the central areas. With rapid population growth, the traffic demand will further increase in the future. Figure 3 indicates that Beijing has been geared toward the way of pattern D since 2000. For the sustainable development of the urban transport system in Beijing and considering some constraints in energy supply and the environment, a prudent decision should be made on whether or not to continue following pattern D. If policymakers decide to continue following this pattern, measures should be taken to limit the number of cars to alleviate urban traffic congestion, as some cities of the developed countries in Europe, such as Paris and London, have already done.

After comparing energy demand forecast, pollution emissions, and social costs, this paper suggests that the optimal traffic pattern for the urban transport development of Beijing should be pattern B. Evidently, as shown in Figs. 6, 7, and 8, travelling by public transport has a competitive advantage compared with the other two modes, considering energy, environment, and social costs. In Scenario 2, the average annual reduction on energy demand, pollution emission, and social cost are projected to reach 61%, 36.8%, and 30.3%, respectively; compared with Scenario 1, the average annual reduction could reach 28.2%, 16.6%, and 18.7%, respectively.

Conclusion

In this study, some findings and corresponding recommendations are formulated. These are described below.

First, the results of the scenario analysis show that the future urban traffic demand of Beijing could increase at a constant rate, and private cars could comprise the main source of energy consumption. Beijing, a densely populated area, should execute restrictive policies on cars to alleviate problems related to urban traffic congestion. Residents from the suburbs should be encouraged to park cars in transfer center parking lots and switch to public transports as they travel to the center area.

Second, in the urban traffic system, full public transport development is the effective way to realize sustainable development of the transportation resources and environment of Beijing. Public transport can satisfy transportation demands, effectively easing the supply pressure in transport resources and improving air quality in the city. The informationization and intelligentization of public transport management, operation, and service should be positively pushed forward. Furthermore, great efforts should be exerted to improve the travelling speed and efficiency of the entire system in order to provide excellent service to the public.

Third, travelling by bicycle is a traditional mode of transportation favored by Beijing citizens. Beijing has good roads, experiences little snowfall, and has a moderate climate, providing very suitable conditions for developing bicycle traffic. Therefore, positive traffic strategies should be made in order to encourage more people to use this mode. In the city center, for example, appropriate space should be reserved for bicycles, the branch roads should be employed thoroughly, and the specific transport system should be established to form gradually a local traffic network for bicycles.

Finally, walking is another main mode of transportation, and one of the important traffic patterns connected to public transport. In 2008, walking took up over 10% of all the travelling choices in Beijing. As an environmental-friendly transport pattern, it should be one of the main traffic patterns in the future traffic system of Beijing. Walking should be further encouraged, and the walking traffic system should be improved as an important part of the transport system. These could be achieved through the creation of a safer, more convenient, and more comfortable traffic environment for pedestrians, including other vulnerable traffic groups.

References

[1] Mao H X. Research on person trip characteristics of Chinese citizens. Dissertation for the Doctoral Degree. Beijing: Beijing Polytechnic University, 2005 (in Chinese)

[2] Langel A. Sustainable transport in Davis. World Transport Policy & Practice, 2007, 13(3): 8-22

[3] Pucher J, Renne J. Socioeconomics of urban travel: evidence from the 2001 NTHS. Transportation Quarterly, 2003, 57(3): 12

[4] Ranjan K B. Energy demand and environmental implications in urban transport-case of Delhi. Atmospheric Environment, 1996, 30(3): 403-412

[5] Cooper J, Ryley T, Smyth A, Granzow E. Energy use and transport correlation linking personal and travel related energy uses to the urban structure. Environmental Science & Policy, 2001, 4(6): 307-318

[6] Huang C, Chen C H, Wang B Y, Dai Y, Zhao J, Wang H K. Urban travel modal split and its impact on energy and environment. Journal of Highway and Transportation Research and Development, 2005, 22(11): 163-166 (in Chinese)

[7] Qiu L X, Chen Y, Zhao D Q, Cai G T, Chen Y, Hu X L, Jiang K J. Scenario analysis of passenger traffic development in Guangzhou until 2020. Acta Scentlarum Naturalium Universitatis Sunyatsen, 2008, 47(Z1): 114-118 (in Chinese)

[8] Hu X L, Jiang K J. Evaluation of Technology and countermeasure for greenhouse gas mitigation in China. Beijing: China Environmental Science Press, 2001 (in Chinese)

[9] Zhou D D. China Sustainable Energy Scenarios in 2020. Beijing: China Environmental Science Press, 2003 (in Chinese)

[10] Bruee B, The Future of the US Motor Vehicle Emission Control Program, Manufactures of Emission Control Association, 1997

[11] Wang J W. Research on air environment influence of sustainable development urban transport. Dissertation for the Master Degree. Xi’an: Chang’an University, 2005 (in Chinese)

[12] IPCC Working Group III. The Fourth Assessment Report of the IPCC. London: Cambridge University Press, 2007

[13] IEA. World Energy Outlook 2009.

[14] Xiong W, Chen X H. Comparison of urban traffic patterns between domestic and overseas cities. City Planning Review, 2009, 33(3): 56-66 (in Chinese)

[15] Stephen M. Street & Patterns. New York: Spon Press, 2005

[16] Chen G, Cao G Y, Liu Y B, Pang L H, Zhang L, Ren Q, Wang H T, Zheng X Y. The Future Population of Beijing—A Projection on the Population. Human Capital and Urbanization using PDE Model, Population and Development, 2006, 12(4): 29-41(in Chinese)

[17] Ma X H, Hou Y F. A Study on Beijing’s Population Variation Trend in the Future 50 Years. Market and Demographic Analysis, 2004, 10(2): 46-62 (in Chinese)

[18] Caltrop E, Proost S. Road transport externalities. Environment and Resource Economics, 1998, 11(3,4): 335-348

[19] Litman T. Full cost accounting of urban transportation: Implication and tools. Software Review, 1997, 14(3): 169-174

[20] Bosi M, Ellis J. Exploring options for sectoral crediting mechanisms. OECD/IEA, 2005

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg