Sustainable energy and urban form in China: the relevance of community energy management

Energy Policy 29 (2001) 55}65

Sustainable energy and urban form in China: the relevance of community energy management Bryn Sadownik, Mark Jaccard* School of Resource and Environmental Management, Simon Fraser University, Vancouver, Canada V5A 1S6 Received 19 August 1999

Abstract Chinese cities are experiencing major environmental e!ects from fossil fuel-based energy consumption for mainly residential and, increasingly, urban transportation uses. Community energy management (CEM) is a sustainable energy strategy which looks at how purposely shaping the built environment and designing urban services in consideration of energy production, distribution and use could a!ect both the long-term demand for energy and the type of energy supplied. This study explores what CEM is in a Chinese context by analysing trends in land-use planning, urban transportation and residential energy, and then suggests CEM strategies that would be appropriate in directing urban development towards a more sustainable energy path. A spreadsheet model is used to evaluate aggregate energy-related emissions in the year 2015 that result from two alternative scenarios of urban growth throughout China. The model focuses on how energy demand, residential energy technology penetration and transportation mode choices are a!ected by factors of density and mix of use in neighbourhood development. Results from this exercise suggest that China can achieve urban residential and transportation emissions reductions of approximately 14% for CO , 10% for SO , 40% for NO and 14% for   V particulate emissions in 2015 by adopting certain aspects of CEM. Issues around the implementation of CEM are also addressed in this study by examining key institutional and policy issues involved in land-use planning, site and building design, alternative energy supply and transportation management. Recommendations and implementation strategies are suggested.  2000 Published by Elsevier Science Ltd. Keywords: China; Energy; Urban form; Urban planning

1. Introduction In seeking ways to minimize the environmental impacts of energy use, countries have begun shifting the focus in energy development towards planning in relation to the end use demand for energy, primarily at a technology use rate, technology and building scale. A further focus has been to examine how purposely shaping the built environment and designing urban services in consideration of energy production, distribution and use, could a!ect the long-term demand for energy and type of energy supplied. This approach has loosely been termed, &community energy management' (CEM). It is based on the premise that a signi"cant proportion of future energy consumption is predetermined when land

* Corresponding author. Tel.: #1-604-291-4219. E-mail address: [email protected] (M. Jaccard).

use and urban form is designated. It is directed at residential, commercial and urban transportation energy use. This energy policy perspective may be particularly relevant to China, since that country is experiencing rapid urbanization. One study in the early 1990s estimates that at least one-half of all the urban structure in China, measured in terms of square meters, was added during the 1980s, alone (World Bank, 1993). Given these trends, consideration of CEM is crucial in addressing China's energy challenges. In this study we explore the bene"ts and viability of a CEM approach in Chinese urban areas, and in particular three questions. First, what CEM is in a Chinese context, speci"cally in terms of land-use planning, transportation management, site and building design, and energy supply. Second, we explore what the aggregate energy-related air emissions would be of alternative scenarios of future Chinese urban development and residential energy planning. Third, we examine how present institutional and economic structures may facilitate or

0301-4215/01/$ - see front matter  2000 Published by Elsevier Science Ltd. PII: S 0 3 0 1 - 4 2 1 5 ( 0 0 ) 0 0 1 0 5 - 1

56

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

limit the application of CEM, and discuss, given these constraints, how CEM could be implemented * outlining institutional and economic reforms that would be relevant to successfully implementing CEM.

E E E

2. CEM strategies for China E CEM studies have been primarily undertaken in an industrialized country context, where the nature and make-up of residential, commercial and urban transportation energy, as well as urban spatial form, are considerably di!erent. Unique features of urban energy and spatial form in China that need to be re#ected in this study include: E The dominance of dispersed coal burning in urban residential and commercial energy use. E A relatively low, but increasing, level of energy use. E A very low level of residential energy e$ciency (Zhong Xiang Zhang, 1998). E Rapidly changing urban transportation patterns due to increasing personal incomes and signi"cant changes in urban form. Chinese urban areas are experiencing growing numbers of cars, trucks, and buses. E Changing urban land-use patterns. Urban areas are developing specialized neighbourhoods, resembling increasingly what is seen in the West. Changes to land use patterns include the emergence of downtown retail and business centers, residential districts and targeted development zones. Residences are becoming less spatially linked with the workplace (Gaubatz, 1995). E Rapid-paced and piecemeal development of outlying urban areas. CEM strategies that may be appropriate in China include: E Encouraging a greater diversity of energy sources and encouraging renewable technology appropriate to the end use task. E Replacing decentralized and uncontrolled coal combustion for individual apartment blocks and dwellings with community facilities with environmental controls. E Developing co-generation and district energy further, emphasizing larger-scale, integrated systems. District energy/co-generation could be made more economical by also supplying hot water to public bath houses and

 Unlike urban areas in North America and Europe, the specialization of districts*central business core, industrial parks, suburban residential * has, in general, not signi"cantly taken place in China. During the Maoist period, urban planning was limited, and work units provided not only employment, but housing, and other services to its workers, resulting in a mixed-use urban form within walled compounds.

E

E E

private dwellings as that market develops, and by providing both district cooling and heating. Encouraging gaseous fuels. Encouraging land-use development that will make district energy most economical. Facilitating the continuing use of bicycle transportation. Improving the quality of public transportation in conjunction with land-use planning. Encouraging greater mixed land use and density in suburban developments. The dispersal of activity centers and increase in local neighbourhood services need to follow outward extensions of the city. Encouraging the interaction of industrial energy provision with community energy through waste heat utilization. Maintaining mixed land use in central areas.

3. Measuring the environmental bene5ts of CEM A spreadsheet model is used to broadly estimate the aggregate energy-related environmental e!ects of alternative scenarios of future Chinese urban development and residential energy planning. Two alternative scenarios of urban development in the year 2015 are evaluated. The "rst scenario, `development of current trendsa (DCT) outlines a future which re#ects the trends which have occurred since China began its pattern of rapid economic growth, and the examples of newly industrialized countries. The second scenario, `community energy managementa (CEM) incorporates a number of CEM strategies into future urban development. The CEM strategies are discussed in more detail in Table 1. Environmental e!ects addressed in this study are limited to present and future atmospheric air emissions * CO , NO , SO and particulate emissions which  V  emanate from residential and urban transportation energy uses. The exercise also limits energy end uses to those uses which are in#uenced by CEM strategies * more speci"cally, those uses a!ected by factors of density, mix of use in neighbourhood development, and policies which in#uence the level of development control, enforcement of building codes, and new technology penetration. 3.1. Methodology The methodology used to evaluate the future scenarios described above consists of two overall steps: 1. The present and future urban population of China are divided into &development classes' * groupings which characterize di!erent residential building patterns (i.e. mix of use, type of building). The CEM and DCT

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

57

Table 1 CEM scenario Area

Strategies

Land-use planning

E Land-use planning control resulting in more coordinated development. E A tendency towards mixed land use and the maintenance of dispersed business centres. E A tendency towards maintaining a relatively high density, but not to the detriment of local environmental quality.

Transportation management

E E E E

Site and building design

E Building to maximize the shape coe$cient. E Ensuring that buildings are built so that they can be easily and economically retro"tted for district heating and/or cooling.

Energy supply and delivery systems

E The replacement of decentralized and uncontrolled coal combustion in individual apartment blocks and dwellings. E Encouraging the interaction of industrial energy provision with residential uses. E A faster introduction of new fuels and technologies (such as district cooling, waste heat). E Increased gas penetration for cooking and heating.

A greater emphasis on public transportation development. The facilitation of bicycle and pedestrian transportation. The implementation of transportation management strategies to discourage automobile growth. The development of employer sponsored commuting services and other high occupancy vehicle travel.

Table 2 Description of development classes Development class

De"nition

Example

L1 L2 M1 M2

Low rise, mixed use Low rise, non-mixed use Medium rise, mixed use Medium rise, non-mixed use

H1 H2

High rise, mixed use High rise, non-mixed use

1}3 storeys * mainly old pre-1950s buildings 1}3 storeys * new villas, exclusive suburbs 4}10 storeys * mainly 1950s-1980s development, and some new development 4}10 storeys * mainly new, 1980s#development in rapidly changing cities, as well as older development in areas of low &non-productive' investment such as satellite cities 11#storeys * mainly new development which maintains mixed neighbourhood 11#storeys * new development in rapidly changing urban areas

scenario di!er in their relative distribution of development classes. 2. Total emissions for each scenario are determined by multiplying the population in each respective development class by the per capita emissions for each development class. (&development class emission rates') Each step is developed in greater detail below. 1. Development classes: In each future scenario, the projected urban population is divided between di!erent types of &development classes'. These are described in Table 2. The type of development class guides the type of end use technologies that are adopted and the level of energy consumption, which in turn impacts the level of air emissions. Speci"c development classes have been advanced which characterize the nature of building types and the mix of land use in Chinese communities, based on urban spatial and urban form patterns discussed in

the literature (Gaubatz, 1995; Feng Liu, 1993; World Bank, 1993). 2. Development class emission rates: Determining the total per capita emissions for each development class involves the following steps: (A) End uses, and the technologies (and modes in the case of transportation) that provide energy for those end uses, are speci"ed: Four categories of energy end use are considered * space heating, space cooling, cooking and transportation. The choice of end uses and the technologies that are associated with them have been limited to those that relate to broad community infrastructure development, and the e!ect of urban form on that development. Thus, micro-site decisions such as solar design and landscaping are excluded. Hot water heating is not considered as a separate end use because existing

58

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

household hot water uses are included as part of cooking, and a large percentage of hot water use takes place outside of the home (i.e. public bathhouses). In this study, the technologies considered for each type of end use are as follows: E Space heating * coal stoves, central heating (coal), district heating (coal), district heating co-generation (coal), electric heaters and gas (coal gas and natural gas); E Cooling * individual air conditioning, central air conditioning, and district cooling (co-generation). Although electric fans are currently the dominant form of cooling, they are not included because they are not considered to be equivalent to air conditioning as an end use. E Cooking * coal, piped gas, lique"ed petroleum gas (LPG) E Transportation * transit, high occupancy vehicles, pedestrian, cycling, motorcycle and automobile. (B) Emission rates for CO , NO , SO , particulates  V  (t/GJ) are determined for each technology. (C) For the present and for the competing future scenanrios, technology market shares are described which indicate the relative technology penetration in each development class for each end use. The penetration rates of the di!erent technologies used to accomplish the end uses considered in this model are estimated for each development class. Shares are given for both the initial period and the two future scenarios. Market shares total 100%; thus it is assumed that a development class will have some type of technology penetration for each end use. The determination of technology shares is based on the scenarios outlined in Table 1. For example, the technology share of natural gas for cooking is greater for the CEM scenario. (D) Overall emission rates are determined for each development class (&development class emission rates') by multiplying each technology share by its respective technology emission rates (t/GJ) and summing them for all technologies in that development class. (E) Present and future per capita yearly energy consumption (GJ/capital/year) values are calculated for each end use. (F) An &energy consumption multiplier' is calculated for each development class, which estimates the di!erence of development class on overall energy consumption due to nontechnology factors (i.e. the e!ects of shared walls; changes in trip generation). (G) The total per capita emissions (GJ/capita/year) for each development class are "nally determined by multiplying &development class emission rates' by the &energy consumption multiplier' and the &per capita energy consumption values'.

3.2. Archetypes and zones In order to re#ect regional and economic di!erences in energy use and supply, the urban population is grouped into six &zones', and is subsequently divided into 2 &Archetypes'. The zones re#ect per capita GNP and climate, and the archetypes re#ect city size. Distinctions between zones are integrated into steps 1 and 2(C)}(F) of the model; archetype distinctions are made in step 1. The number of factors used to classify zones and archetypes have been limited to keep the model relatively simple. Climate was chosen because of its e!ects on the demand for heating and cooling energy. This model adopts the o$cial categorization of climatic regions for residential energy purposes stipulated in the National Standard Design Code for Heating, Ventilation and Air-conditioning which divides China into the central heating zone, transition zone and the non-heating zone. Per capita GNP is incorporated because regional di!erences in wealth are signi"cant in China and rising incomes have been a key factor in the surge of urban residential energy use. Per capita GNP is chosen to re#ect both the ability of urban areas to a!ord certain types of investment, such as public transportation, and di!erences in personal consumption between urban areas. Aspects of per capita residential and urban transportation energy use (i.e. level of energy use, technology adoption) generally relate with measurements of &wealth'. City size is used as a factor in breaking up the urban population in step 1, because cities over and under 1 million are quite di!erent in terms of urban form. 3.3. Data sources The projections are based on numerous data sources, including considerable information found in published, refereed studies in this subject area. O$cial Chinese data are used in some cases, though often in the case where internationally veri"ed data was not available. In some instances, we have made quantitative interpretations of qualitative trends, particularly in the case of land-use characterization, where the availability of hard data is limited. Key data sources are listed below: Population: Base population data } China statistical yearbook (1992), Population projections * United Nations (1995).  The central heating zone * where the sum of the days in which the average daily temperature is steadily less than or equal to 53C, is more than 90 days in the whole year. Presently, central heating is only allowed and provided for in this Zone. The transition zone * where the sum of the days, in which the average daily temperature is steadily less than or equal to 53C is between 60 and 89 days; or in which the average daily temperature is steadily less than or equal to 83C in less than 60 days but is more than or equal to 75 days. The non-heating zone * everywhere else.

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

Residential energy use: Energy shares, energy consumption * Feng Liu (1993), Qiu Daxiong et al. (1994), Energy e$ciency * Fuqiang Yang (1997), Peters (1997). Urban land use: Housing stock projections * Lang Siwei and Fan Youchen (1990), Current land-use trends * Leaf (1995), Gaubatz (1995), Existing buildings structure characterizations * Lang Siwei et al. (1992), New building construction * Treister (1987). Transportation: Projected vehicle growth * Stares and Liu Zhi (1996), Mode shares * Qing Shen (1997), Rate of vehicle possession * China statistical yearbook (1992). Land-use energy relationships: Transportation * Dunphy and Fisher (1996), California Energy Commission (1993), Building heating * Keyes (1976) cited in Owens' (1987).

59

Fig. 1. Total annual CO emissions for the initial period and future  scenarios (2015).

3.4. Results of model run The development class emission rates are multiplied by the energy consumption multipliers and the composite energy values to come up with total annual air emissions for the two scenarios and for the initial time period. The results are shown by emission type in Figs. 1}4. Contributions of each end use to total emissions are also shown in these "gures in order to give a sense of the growth of each end use from the initial period to 2015, and to show what end uses are contributing most to reductions in emissions between the DCT and CEM scenarios. Emission reductions that occur from the DCT scenario to the CEM scenario are described in Table 3. Overall, the results of this modelling exercise broadly suggest that CEM strategies may bring about signi"cant reductions in air emissions, particularly those a!ecting local air quality as Chinese urban areas become more and more motorized. The CO savings of 14.4% between  the CEM and the DCT future scenarios fall in a similar range to those found in the Failing's (1995) study for British Columbia which found that CEM measures would save approximately 17%. It is worth noting that CEM measures have an equally signi"cant e!ect on particulate emissions (14.0% savings), a major health concern, and an even stronger e!ect on NO emissions V (40.4% savings). SO emissions are relatively less a!ected  (9.7% savings). The e!ects of CEM in this modelling exercise are midterm. Given the even longer lifespan of infrastructure, the environmental bene"ts from the CEM measures would increase into the long term. As can be seen in Figs. 1}4, transportation is a major future source of NO emissions, but a relatively smaller V portion of the other emission types. Transportation is

 This study's methodology is similar.

Fig. 2. Total annual SO emissions for the initial period and future  scenarios (2015).

Fig. 3. Total annual NO emissions for the inital period and future V scenarios (2015).

strongly a!ected by the CEM initiatives modelled in this exercise, and major emission reductions result from a relatively small reduction in automobile acquisition from transportation management and land-use planning. CO emission reductions result from savings in each end  use, although heating contributes the most overall because of the larger size of this end use. Reductions in particulate emissions in cooking contribute the most to

60

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

Fig. 4. Total annual particulate emissions for the initial period and future scenarios (2015).

Table 3 Di!erences in total emissions between the CEM and the DCT scenarios Emission

Savings between CEM and DCT scenarios (%)

CO  SO  NO V Particulates

14.4 9.7 40.4 14.0

total particulate reductions. Coal burning is displaced quite strongly by LPG and piped gas. Heating is the end use that is the least a!ected by the CEM measures evaluated in the model. A breakdown of heating technology shares reveals that the share of coalbased central and stove heating is not very di!erent between the DCT and CEM scenarios, although the share of these technologies is substantially reduced compared with the initial technology share distribution in 1991. The DCT scenario is already achieving many &CEM' technology penetration aims. New building stock in the DCT scenario is already multi-storey for which central heating is the standard technology. The Chinese government already has strong goals for district heating and the multi-family residential development that occurs in China is favourable to district heating (particularly compared to building trends in North America). The CEM scenario only increases this slightly. Technology share di!erences that are more apparent are the greater portion of cogenerated heating and the smaller share of electric heating in the CEM scenario. The latter has little e!ect on particulate emissions, because particulate removal in electricity generation is reasonably high, even though energy e$ciency levels are relatively low for that end use. The other emissions are cut because of e$ciency di!erences. It would be useful as a follow-up to this model to explore how site speci"c design and penetration of passive and active solar heating may reduce heatingrelated emissions.

Cooling does not "gure strongly in the future scenarios for China as a whole. This end use accounts for a very small percentage of all emissions. Although cooling in the non-heating zone makes up close to one-quarter of the per capita annual energy use that heating does in the heating zone, the non-heating zone makes up a much smaller portion of the total urban population. The e!ects of CEM on cooling should not be overlooked since all emissions besides CO a!ect the local region foremost.  The portion of these emissions from cooling in the nonheating zone are signi"cant where cooling comprises approximately a third of SO emissions. The CEM scen ario reduces these emissions by 16%. This occurs even with a very small introduction of district cooling in the CEM scenario. The model results should be interpreted as an exploration of the magnitude of bene"ts to air quality and CO  emissions that could occur if CEM strategies are adopted. The model is grounded in data and observable trends; however, the model has simpli"ed many of the complexities found in the actual system. The key instances include: E Assuming no energy consumption for cooling in the central heating zone and no energy consumption for heating in the non-heating zone. E Assuming that the comfort level of di!erent development classes (i.e. low storey compared to high storey) is identical and that the average per capita living space is the same across all development classes. E Not taking into account many of the behavioral aspects of energy use. For example, in centrally heated apartments, inhabitants will often open windows to control the heat in their apartments because they have no control over the temperature setting. E Simplifying urban form in a limited number of "xed development classes which may not represent all Chinese communities. E Limiting future changes in coal mix, technological end-use e$ciency and emission control. In addition to these simpli"cations, uncertainty is also present in the model due to data limitations. This manifests in di!erent ways. First, some values are problematic. For example, it is rarely explicit when a study describes a value that relates to urban area (i.e. total urban energy consumption), what de"nition of urban is described by this value * towns, cities, countries, agricultural, nonagricultural. Second, in some instances data is derived from a site speci"c Chinese study or a study from another country because of a lack of available Chinese national data. In certain cases, the analysis is performed using  The reality is more complex. For example, apartments generally have more living space, and therefore have a higher energy heating/cooling requirement. This is interesting considering the reverse is true in North America and Europe.

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

&educated' assumptions (quantitative interpretation of qualitative trends) rather than on hard data. Furthermore, there are some parameters over which there is a range of speculation (motorization growth rate, population growth rate), and others where there are divergent study results (e!ects of urban form on transportation). A wide range of parameters in the model were tested for the sensitivity of the results to changes in their values, including the assumed relationship between urban form and energy consumption, assumptions in technology energy e$ciencies and coal mixes, energy service demands. It was found that the results are most sensitive to transportation parameters and in particular, the projection of motor vehicle penetration. However, re"ning this parameter further does not a!ect the fact that CEM structuring of urban form will have a positive impact on air quality, but simply when this e!ect will occur. The gains of structuring urban land form favourably now, even if motorization is slower, will still be realized.

4. Implementation of CEM Current residential and transportation energy practices have led to critical and deteriorating air quality conditions in an increasingly larger set of Chinese cities. In many respects, local environmental pollution problems could be weighted by government much more than global problems, and e!orts to address these problems will gain a strong public support (Gan, 1998). This is a compelling basis for the adoption of CEM strategies. However, as a whole, the successful implementation of CEM will rely on the ability to integrate urban planning and energy planning at a state and local level, and on creating policies that can "t into the new political economy relationships that are developing. The emergent nature of the market system is favourable to incorporating, from the start, tools and processes that can implement CEM objectives, such as regulation, market incentives, education/information and targeted public investment. The ability to implement CEM, and realize the bene"ts suggested in the modelling exercise, also relates speci"cally to the institutional structures that have developed around land-use planning, energy planning and transportation planning. Given China's huge population, regional (and local) dynamics, and uneven pace of change, existing and developing institutional structures defy generalized descriptions. Nevertheless, this should not hinder one from drawing on a considerable body of literature to o!er a few insights into how CEM might succeed in being implemented. 4.1. Land-use planning Land use planning in China needs to be integrative and holistic in order to in#uence urban development

61

including urban growth; industrial, commercial and residential siting; and infrastructure development. Currently, the dominant means of control of urban form is through the use of &master plans' which are strategic 20 year plans for urban construction and development which de"ne a city's identity and its future direction and scale of development. When the plan is approved it becomes a statutory document * all development is to coincide with this plan. However, new development can be disjointed and poorly planned, while inner-city development scattered and unconnected, all of which would suggest that actual development di!ers from the planning process outlined. Key institutional aspects which contribute to this include: E Development control may be relaxed in order to increase the attractiveness of particular sites for development, so that economic development and short-term growth can be accelerated (Gar-on Yeh and Fulong Wu, 1995). E A reliable framework for the protection of planners against arbitrary decisions and executive interference is absent (Hok Lin Leung, 1989). E Broad zoning guidelines and lack of understanding about property rights to a certain piece of land (Gar-on Yeh and Fulong Wu, 1995). E The structure of economic management and development has caused economic activities in a spatial unit to be compartmentalized according to structure of the bureaucracies involved. E Land is often developed on the periphery of urban areas because acquisition of land for leasing or the development of commercial housing in the city centre involves much time-consuming bargaining. E Urban planning practices are slow to reconcile with market changes, especially now that industries and enterprises have more discretion in choosing their location than before (Gar-on Yeh and Fulong Wu, 1995). A revised master planning concept, which still involves coordination between the state and municipal levels of government, would be important in bringing together energy policy and urban planning policy. However, it could be reformed so that it can be better integrated into actual municipal planning and development processes. This would involve changing its frequency, #exibility and its content. For example, removal of the requirement of State Council approval would discourage static plans. Nevertheless, the state level role in setting objectives to be incorporated into the urban planning process would be valuable. Furthermore, the incorporation of adaptive management principles into urban planning would be helpful in coordinating land-use development. A mechanism could be created that allows actual development to be &fed back' into the plan including the nature, location, and phasing of infrastructure investments, and the

62

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

requirement for new planning to be done to re#ect these changes. Hong Kong and Singapore, for example, have #exible macrostructure plans that allow for such feedback (World Bank, 1993). In order to reduce arbitrary planning decisions, a stronger legal framework could be applied to planning practices. For example, information about a land development proposal as well as the decisions and their justi"cations could be well advertized (Hok-Lin Leung, 1994). In addition, the better enforcement of existing laws and regulations would allow land use plans to be more e!ectively implemented. CEM land use objectives need to be encouraged through policy instruments that are appropriate to the government's evolving role towards the emergent market economy. Instruments include regulatory (legal) mechanisms and market incentives. The current development of a leasehold land system rather than a system based on private land ownership is favourable for the implementation CEM site and building objectives because the state can specify terms of the land lease. Terms could stipulate development requirements favourable to CEM objectives, such as building restrictions and design, height, arrangement of structures and permitted land use. Existing building approval processes could also be expanded to include consideration of the use of microclimate, landscaping, and location of paved surfaces. 4.2. Energy planning Energy planning needs to be able to identify energy e$ciency and alternative supply investment opportunities appropriate to local conditions, that make sense from an economic, social and environmental perspective, and that encourage the e$cient development of these opportunities. Many Chinese national energy policy goals are in keeping with the CEM strategies * for example, China has stated policies of encouraging cogeneration, district heating systems and clean coal technologies such as coal cleaning and coal gasi"cation, as well as equally weighting consideration energy exploitation and conservation projects. Nevertheless, the actual implementation of these policies has not fully realized their intent. For example, the balance of energy project have been exploitation dominated (Zhong Xiang Zhang, 1995). CEM strategies require the #exibility to comprehensively evaluate energy projects, which is a challenge in the existing institutional arrangement. Di!erent ministries have di!erent processes and guidelines to assess projects; there is no single intra-ministry process regarding project assessment. The government is constrained in its ability to coordinate policies across subsectors. Sustainable energy supply could be encouraged by the adoption of consistent and transparent procedures across government ministries and by evaluating options within an integrated resource planning framework. In considering projects, it is important that the bene"ts and costs of all

available supply and demand-side alternatives and their respective impact on the local environment are thoroughly studied. Greater coordination of policies and investment across energy subsectors is very important to implementing CEM objectives. This could be accomplished through the addition of integrative mechanisms or through institutional restructuring. For example, provincial and local energy bureaus could be created that combine existing separate bureaus. Coordinating bodies free from the competitive dynamics that often characterize interagency communication could be introduced within the bureaucratic structure. Plan allocations of energy at below market prices (in e!ect acting as energy subsidies) still account for a substantial portion of energy procurement, providing weak incentives for investment in energy conservation. In order to encourage future energy conservation, current subsidies for energy consumption should be eliminated (Zhong Xiang Zhang, 1998). Financial incentives, which have been used in the past to encourage e$cient and renewable energy supply, also need to re#ect the signi"cant changes to energy development and "nancing. For example, banks, which have gained considerably more control in "nancing, have little incentive to invest in energy e$ciency (Zhong Xiang Zhang, 1995). Gan (1998) argues that energy conservation is unlikely to be favoured by most investors, unless strong policy support is in place. Policy measures could include: E Encouraging commercial banks to o!er preferential interest rates for energy e$cient and alternative energy technology investment. E Reinstituting the tax exemption for delivered heat. E Adjusting the tax treatment of energy equipment expenditures to recognize that most renewable resource have no fuel costs to be deducted from revenues for tax purposes (Byrne et al., 1996). E Using "nancial incentives to encourage technical upgrade investments as an e!ective way to solve capital constraint problems. For example, investing in bottom-cycle cogeneration from existing power stations is less capital intense than new cogeneration development. 4.3. Transportation management Motorized transportation needs to be actively managed through a holistic planning process which fully considers all alternatives including demand management. China currently has a window of opportunity to establish e!ective transportation policies before motorization is too far advanced. Faced with considerable pressures of motorization and population growth, cities are active in transportation planning although this is made di$cult by fragmented responsibilities for municipal transport.

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

Speci"c engineering and technical means have been emphasized in domestic tra$c research. Chinese master plans frequently have major outlines for urban road networks which, according to Stares and Liu Zhi, go well beyond the scale of expressway construction in existing developed motorized cities across the world (Stares and Liu Zhi, 1996). The development of inter-agency teams at the municipal level would allow more comprehensive and holistic planning that includes consideration of demand-side management and land-use planning, as well as tra$c management and engineering tra$c solutions. It is important that transportation planning develop a wide range of possible strategies that are evaluated using multiple criteria to determine the most suitable strategy. Public transportation planning for both the short and long term would allow public transportation improvements to be achieved while pursuing longer term options. For example, Allport (1996) suggests that both a long term (15 years) transport strategy, and shorter term action plans (5 years) could be formulated and funded. Less expensive means of public transportation development could be explored that can be upgraded as funds permit. For instance support could be given to the improvement of ground public transport which could lead into the development express buses, busways and light rail transit. China is unique compared to other countries in the role of the work unit in securing housing for its employees. However, this role has undergone signi"cant changes in the past decades, so that work units no longer determine directly where their sta! and workers live, but instead purchase housing for workers in comprehensively developed housing projects which are developed by the municipality's development companies (Gar-on Yeh and Fulong Wu, 1995). The focus is turning to municipal control over business siting. It is critical that urban planning recognize the relationships between comprehensively developed housing projects, where work units purchase housing for their workers, and the location of the workplace since that link is no longer directly made as part of economic development. Approaches that could be taken might include encouraging development companies to coordinate with work units that have a demand for housing at the initial development stages of their projects, including siting decisions. Or, if projects are already being established, preferential treatment could be given to work unit housing demands that are in closest proximity to the development.

5. Conclusion The way urban form unfolds will orient energy patterns far into the future. It is therefore critical to shape urban form and infrastructure patterns so that more

63

sustainable energy paths can be followed. CEM should be an important component of a sustainable energy strategy in China, and can complement conventional energy analyses and strategies. Reductions in local air emissions is a particularly strong rationale for applying CEM in China, where the direct local environmental e!ects of residential energy consumption are felt much more than in North America. Energy conservation in 1990 has also evolved as a higher priority issue (Gan, 1998). In many ways, China is well suited to adopting this perspective, particularly in its current position in being able to shape market forces before practices become entrenched. The ability for land control is much stronger given China's more &conditionally structured' property rights. Other advantages include the closer ties that currently exist in China between work and residence location due to the continuing organizational role of the work unit. The work unit is also critically placed to implement transportation strategies. China already approaches many energy issues with a CEM angle, for instance through its strong focus on district energy and growing interest in renewable energy development. Furthermore, the municipality is already part of the institutional structure that oversees energy provision, which can complement its role as urban planner. Nevertheless, to adopt CEM, some key challenges need to be overcome. Economic opportunities are short term and immediate, and are di$cult to reconcile with longer term, sustainability concerns. And even when policy and legislation are formulated, they need to be supported and enforced, and di!erent government agencies and levels of government coordinated, so that what unfolds is indeed guided by policy objectives. Central authority in China is weakening with market reform. While decentralized decision making can be a good thing for CEM implementation, this will not be the case if local authorities lack resources, are ine!ective or vulnerable to corruption. This will be a continuing implementation challenge. However, though implementation challenges are neither straightforward or simple to overcome, China is not positioned any worse, relative to other countries, in adopting CEM principles though the speci"c nature of implementation challenges are nonetheless unique. The countries where CEM analysis has developed are the same countries where entrenched planning regimes reinforce sprawled land uses, making CEM implementation particularly di$cult. It is more fruitful to shape and mould urban planning and transportation towards sustainable ends, than to undo 50 years of urban structure which reinforces unsustainable energy use. Likewise it is easier to hinder acquisition or use of the car that has not yet been acquired, than to alter an already existing high per capita ownership pattern (as in North America). China has a much more recent urban land use pattern on which to look to, as well as fairly active environmental directives in#uencing energy policy making. Because

64

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

Table 4 Suggested research questions Rank

Aspect

Suggested Research Questions

1.

Level of motorization

2. 3. 4.

Urban gas penetration Urban transportation e$ciency Urban form and transportation

5. 6.

Travel demand Heating demand

7.

Housing stock growth

8.

District heating penetration

9.

Cooling demand

10.

Particulate emission control technologies

What level of motorization ownership can be expected in the future? At what point will the growth rate slow because of limited road capacity? When will this occur? How will the growth in motor ownership a!ect mode shares? What conditions can maximize penetration of urban gas? What are the characteristics of Chinese urban travel? How does urban form a!ect travel demand and mode choices in a Chinese context (for example, where travel is less home-work based)? How much will travel demand characteristics change in the near future? How much will heating demand grow, particularly given the strong demand growth in the transitional heating zone? What is the future projection of housing stock? Is it expected di!er from the present growth rate? Will the expansion of the district heating system be able to meet government projections? How does urban form a!ect the economics of district heating in China speci"cally? How much will heating demand grow, particularly given the strong demand growth in the Transitional Heating Zone? To what extent are there e!ective particulate emissions control technologies in electric generation, district heating and central heating? How will this change in the future?

urban design measures are simultaneously supportive of other community objectives, there are strong motives for policy makers to follow this path, despite its admittingly challenging and complex implementation. A very important consideration in pursuing CEM strategies is cost, and in particular, cost in relation to alternative strategies that would also reduce greenhouse gas and local air emissions. Though costs are not explicitly measured in this study, some speculations are o!ered. Costs are likely to be relatively low per tonne of CO ,  NO , SO or particulates reduced. There are likely to be V  many instances where moving to a more coordinated, mixed land use will lead to total cost reductions in providing infrastructure and will lower total energy costs because of e$ciency improvements. Cost evaluation would also be a!ected by issues such as prices (i.e. the price of coal; whether steam/heat prices are reformed), and from who's perspective cost is evaluated (i.e. social or private). Furthermore, CEM strategies will most likely lead to favourable non-cost bene"ts such as a relatively greater employment e!ect from pursuing locally based energy projects compared to megaprojects, as well as reducing the loss of agricultural land to urban sprawl. This paper can be viewed as a &departure point' for exploring the relevance of CEM to China. The modelling, though it may be simple in its method and data sources, is illustrative of the magnitude of bene"ts that could be realized in a country that is in China's position of major increases in urban growth and energy use, and provides a suitable basis for further study. Many of the issues raised in this paper need to be taken up in greater detail in future research. In conducting this study, it was found that there are areas in which the knowledge base can be

improved in order to have a clearer understanding of the potential e!ects of CEM on energy-related air emissions. These aspects are presented in Table 4 and are based on parameters in the modelling exercise that were shown to cause the savings between the DCT and CEM scenario to be the most sensitive. They are presented from the strongest to the weakest. In addition to the informational needs outlined in Table 4, future CEM research directions are also suggested. China has been treated in this research as a homogenous entity, overlooking the unique variations in urban planning and development experiences of di!erent urban centres. A study exploring the detailed application of CEM to a speci"c urban community would be an important &next step' in investigating this issue, both in terms of quantitative analysis and in terms of the social, political and implementation issues. It would also be useful to apply the model's method on a smaller scale and with more speci"c data; alter the model to address other CEM strategies (such as facilitating renewable energy penetration); and calculate detailed costs estimates of reducing emissions through CEM measures and comparing these costs to alternative measures such as coal washing.

Acknowledgements The authors wish to acknowledge the helpful comments of anonymous referees and Chinese and

 Colleagues at Tsinghua University in Beijing are in the process of doing just this.

B. Sadownik, M. Jaccard / Energy Policy 29 (2001) 55}65

international colleagues on the Energy Strategies and Technologies Working Group of the China Council for International Cooperation on Environment and Development. References Allport, R., 1996. Investment in mass rapid transit. China City Planning Review 12(1), 27}41. Byrne, J., Bo Shen, Xiuguo Li, 1996. The challenge of sustainability: balancing China's energy, economic and environmental goals. Energy Policy 24(5), 455}462. California Energy Commission, 1993. Energy Aware Planning Guide. China Statistical Information and Consultancy Service Centre, 1992. China Statistical Yearbook. Beijing, People's Republic of China. Dunphy, R., Fisher, K, 1996. Transportation, congestion, and density: new insights. Transportation Research Record (1552), 89}96. Failing, L., 1995. Energy, sustainability and communities: assessing the potential for community energy management in British Columbia. Master of Resource Management Research Project, School of Resource and Environmental Management, Simon Fraser University, Burnaby, Canada. Feng Liu, 1993. Energy use and conservation in China's residential and commercial sectors: patterns, problems, and prospects. Energy Analysis Program, Energy & Environment Division, Lawrence Berkely Laboratory, Berkely, California, USA, LBL-33867. Fuqiang Yang, 1997. Personal communication. Lawrence Berkeley National Laboratory, Energy and Environment Division, Berkeley, CA. Gan, L., 1998. Energy development and environmental constraints in China. Energy Policy 26(2), 119}128. Gar-on Yeh, A., Fulong Wu, 1995. Internal structure of Chinese cities in the midst of economic reform. Urban Geography 16(9), 521}554. Gaubatz, P.R., 1995. Urban transformation in post-Mao china: impacts of the reform era on China's urban form. In: Davis, D., et al. (Ed.), Urban Spaces in Contemporary China: the Potential for Autonomy and Community In Post-Mao China. Woodrow Wilson Center Press, New York, pp. 28}60.

65

Hok Lin Leung, 1989. Strengths and weaknesses of the planning profession in China. China City Planning Review 5(3), 59}71. Hok-Lin Leung, 1994. The mess of land development in China. Paper presented to the Learned Societies Conference Calgary, Alberta, 9}11 June. Lang Siwei, Fan Youchen, 1990. Design of energy-e$cient residential building in transition zone of china. Energy and Buildings 14, 225}229. Lang Siwei, Yu Joe Huang, Levine, M., 1992. Energy conservation standard for space heating in Chinese urban residential buildings. Lawrence Berkeley Laboratory, Berkeley, California. Leaf, M., 1995. Inner city redevelopment in china: implications for the city of Beijing. Cities 12(3), 149}162. Owens, S., 1987. Energy, Planning and Urban Form. Pion Ltd, London. Peters, R., 1997. Personal communication. Based on work by the Saskatchewan Energy Conservation and Development Authority for the Canadian International Development Agency, 1995. Qing Shen, 1997. Urban transportation in Shanghai, China: problems and planning implications. International Journal of Urban and Regional Research 21(4), 589}606. Qiu Daxiong, Ma Yuqing, Lu Yingyun, 1994. Household energy consumption in Beijing and Nanning China. Energy 19(5), 529}538. Stares, S., Liu Zhi, 1996. Motorization in urban China: issues and actions. China City Planning Review 12(1), 13}26. Treister, K., 1987. Chinese architecture, urban planning, and landscape design - - a series of essays. Florida Architecture and Building Research Center Monograph, University of Florida. United Nations, 1995. Compendium of Human Settlements Statistics, 1995. Fifth Issue. Department of Economic and Social Information and Policy Analysis, Statistics Division, New York. World Bank, 1993. China: Urban Land Management in an Emerging Market Economy. Washington, DC, USA. Zhong Xiang Zhang, 1995. Energy conservation in China: an international perspective. Energy Policy 23(2), 159}166. Zhong Xiang Zhang, 1998. The Economics of Energy Policy in China. Edward Elgar Publishing, Inc. Massachusetts, USA.