Impact of urban residences on energy consumption and carbon emissions: An investigation in Nanjing, China

Sustainable Cities and Society 7 (2013) 52–61

Contents lists available at SciVerse ScienceDirect

Sustainable Cities and Society journal homepage: www.elsevier.com/locate/scs

Impact of urban residences on energy consumption and carbon emissions: An investigation in Nanjing, China Z.H. Gu a,c,∗ , Q. Sun b,c , R. Wennersten c a b c

School of Architecture, Southeast University, PR China Institute of Thermal Science and Technology, Shandong University, PR China Division of Industrial Ecology, Royal Institute of Technology (KTH), Sweden

a r t i c l e

i n f o

a b s t r a c t

Keywords: Energy consumption Carbon emissions Nanjing

This paper analysed the energy consumption of urban households in Nanjing and the influencing factors in this energy consumption. The households studied were located in three urban districts of Nanjing: the city centre, a spontaneous residential area around the old city, and a planned satellite town. A questionnaire was used to obtain information on building characteristics, household characteristics, use of domestic appliances, and fuel oil consumption. Energy use was analysed by conversion into CO2 emissions. The study found that household use and transport were the two main contributors to domestic energy consumption. Household electricity consumption showed obvious seasonal characteristics (higher in summer than in other seasons), while transport energy consumption showed geographical characteristics (the old town had lowest transport energy consumption). Highly efficient devices may not render buildings more energy-efficient, so architects should seek to reduce the need for such devices. Energy consumption and income were generally positively correlated. Family structure also influenced energy consumption, with high-income families and small families consuming more energy per capita. Economic and social factors were found to be equally important to technical factors for energy efficiency. Based on the findings, some possible policies are recommended. © 2012 Elsevier B.V. All rights reserved.

1. Introduction

China’s energy consumption and CO2 emissions have increased very rapidly in the past 30 years and especially in the past decade, making China the second largest energy consumer and the largest CO2 emitter in the world (Chinanews.com, 2011; IEA, 2010). Responding to growing challenges on energy demand and CO2 emissions, the Chinese government has been trying to develop strategies on sustainable energy in recent years (Xu, Sun, Wennersten, & Brandt, 2010). For the first time, the Chinese urban population exceeded the rural population in 2011. In that year the urban population reached 690.79 million, representing 51.27% of the total population, while the rural population was 656.56 million, representing 48.73% of the total (National Bureau of Statistics of China, 2012b). In general, energy consumption in urban cities is an important part of strategies on sustainable energy, since urban cities are currently the main energy consumers worldwide (Crompton and Wu, 2004). In addition to architectural design and spatial planning, the pattern of energy consumption in households due to human activities is important for the total energy demand and associated CO2 emissions in urban cities (Chen, Yoshino, & Li, 2010; Mihalakakou, Santamouris, & Tsangrassoulis, 2002; Hickman and Banister, 2007). Therefore, it is necessary to analyse household energy consumption in urban areas of Chinese cities and apply this knowledge in devising strategies on sustainable energy and climate change.

Sustainable strategies for energy can be considered from two sides – energy production and consumption. Fig. 1 shows that from the production side, the focus is on alternatives to fossil fuel, while on the consumption side the focus is on energy efficiency. The strategies from energy producers are more influential than those dealing with energy consumption, because energy providers are much more centralised compared with thousands of consumers. However, there are two reasons why we cannot overlook energy efficiency in energy consumption. One is that existing technologies for renewable energy are not yet sufficiently economically efficient, which prevents them from replacing fossil fuels. It is impossible to generate enough energy by solar technology at the local site in Asia’s high-density cities (Close, 1996). Another reason is that the amount of energy produced is determined by energy demands. Thus the effect of reducing energy demands is marked in terms of carbon emissions reduction, especially in the coming decades.

∗ Corresponding author at: School of Architecture, Southeast University, Nanjing, Jiangsu 210096, PR China. Tel.: +86 133 05153058; fax: +86 25 83792370. E-mail addresses: [email protected] (Z.H. Gu), [email protected] (Q. Sun), [email protected] (R. Wennersten). 2210-6707/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scs.2012.11.004

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

Generation (Energy resource) Fossil fuels: Coal

Distribution (Energy carrier)

53

Utilisation (Energy consumer)

Fuel oil (Petrol, diesel fuel,

Oil

kerosene)

Natural gas Nuclear power

Transportation

Coal gas Natural gas Hydrogen fuel Buildings

Renewable energy: Solar power Hydropower

Biogas, alcohol

Wind power Biomass energy Tidal energy

Electricity Industry

Geothermal energy …

Heat

Fig. 1. Diagram of energy flow from generation to utilisation, showing that fossil fuels cannot be quantified at the consumption side of the chain. Source: Made by Zhenhong Gu).

Roughly a billion Chinese (or more than 90% of the population) live in only a little more than 30% of China’s land area (Heilig, 1999). Fig. 2 illustrates the geographical concentration of the Chinese population and the increasing concentration of population in the eastern portion of the country since its rapid growth began in the mid-20th century (Wang & Wei, 2010). Apart from a few megacities such as Beijing, Shanghai, Chongqing and Guangzhou with a population of over 10 million, the majority of China’s major cities, i.e. provincial capitals and cities specially designated in the state plan, have a population of 5–10 million. With 8 million permanent population on 11 November 2010, Nanjing, the capital city of Jiangsu Province and located in the Yangtze River delta region, is a representative large city (Nanjing Statistic Bureau, 2010). Nanjing City is located at 32◦ 02 38 N, 118◦ 46 43 E, in a region where the coldest monthly temperature in January is 2.4 ◦ C and the hottest monthly temperature in July is 27.8 ◦ C, with an annual average of around 15.5 ◦ C. The average relative humidity is 77% (CDC, 2011). Fig. 3 shows mean monthly temperature in Nanjing during the period 1971–2001. In general, China can be divided into seven zones according to climate characteristics (Fig. 4). The climate in Nanjing is characterised by hot summers and cold winters, the typical weather in Zone III in Fig. 4 (Ministry of Construction of China, 1993). The coldest month of the year is January, with an average temperature of 2.4 ◦ C, and the hottest month is July, with an average temperature of 27.8 ◦ C. As in Zone II, both indoor cooling in summer and heating in winter are needed in Nanjing according to national standards (Ministry of Construction of China, 1993). Before China’s reform and opening-up policy in 1978, most residential buildings were constructed in the former Soviet Union style

to meet basic living needs (Ma, 2002). However, most of these old buildings have now been reformed or replaced and it is difficult to find any surviving examples. In the period 1976–1990, new residential buildings were constructed to meet higher living standards, but such development was mainly restricted to the old city, an area of 44.65 km2 within the Ming Dynasty city wall (Nanjing Urban Planning Bureau, 2006a). In the 1990s, some large residential communities, e.g. Longjiang and Zhongbao, were developed outside Nanjing’s old city without careful planning. The lack of business and commercial buildings means that the residents of these areas have to work in Nanjing downtown (Yeh and Yuan, 1986). After 2000, new satellite towns and residential areas began to be developed outside the old city and urban planning was applied appropriately in this development (Nanjing Local Chronicles Compilation Committee, 2011). Hexi is a new town that was planned to be the second centre of Nanjing due to its location close to the old town. The first two of Nanjing’s subways were constructed across this area, and many business and commercial buildings were planned for the central area of Hexi (Nanjing Urban Planning Bureau, 2006b). A large quantity of residential communities were planned and constructed in Hexi during the first decade of the 21st century. Our survey on energy consumption was carried out in three different urban areas, representing the three phases of residential development in Nanjing since 1978. These areas were: Zhujiang Road (Site A) in the old city, Longjiang area (Site B) in the unplanned residential area, and Hexi area (Site C) in the planned satellite town (Fig. 2). Basic information about the households and their energy consumption was collected in the three areas. Various

Fig. 2. Changes in population density in China showing an obvious trend of population concentration in eastern coastal regions (Wang & Wei, 2010,). (a) Density in 1949; (b) density in 2000; and (c) density in 2020 (projected).

54

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

Temperature (ºC) 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 -5.0 -10.0 -15.0

Average Temperature (ºC) Highest Temperature (ºC)

Lowest Temperature (ºC)

Jan.

Feb. March April

May June

July

Aug.

Sep.

Oct.

Nov.

Dec.

Fig. 3. Mean monthly temperature in Nanjing, 1971–2001 (CDC, 2011).

Fig. 4. Climate zones I–VII in China. Source: Ministry of Construction of China (1993)

Fig. 5. Location of study sites A–C and examples of streets and buildings at each site. (a) Site locations; (b) Zhujiang Road, Site A; (c) Longjiang area, Site B; and (d) Hexi area, Site C. Source: Made by Zhenhong Gu.

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

55

Table 1 Comparison of per capita CO2 emissions.

Site A Site B Site C Average

CO2 emissions from electricity (kg)

CO2 emissions from natural gas (kg)

CO2 emissions from petrol (kg)

Total CO2 emissions (kg)

1144.5 1100.9 1192.5 1144.7

48.7 57.7 58.6 54.9

540.1 610.1 1188.2 768.7

1733.3 1768.7 2439.3 1968.3

characteristics relating to urban living, such as climate, geographical, architectural, economic and social factors, and their impacts on energy consumption, were analysed (Fig. 5). 2. Aims and objectives The aims of the present study were to survey the energy use of households in Nanjing, and to identify the factors influencing urban residential energy consumption. The factors examined were climate, geographical, architectural, economic and social. Climate and geography factors are technical factors, while the others are non-technical factors. Since energy issues are of a technical nature, it is obvious that technical factors are being discussed widely at present. However, the influence of non-technical factors is also important. This study therefore analysed both types of factors. The intention was to use the findings to assess the consequences of some actions and recommend possible strategies promoting residential energy efficiency. Although the strategies may be political, social, economic and technical, the study focused mainly on architectural and urban planning strategies. 3. Methodology The survey was carried out by researchers and students from the Architectural School, Southeast University, in 2011. In all, 1500 questionnaires were randomly distributed in the three areas (500 in each) and the total number of valid returns was 77, representing 204 inhabitants (Site A 26 valid returns, representing 70 inhabitants, Site B 28 valid returns, representing 74 inhabitants, Site C 23 valid returns, representing 60 inhabitants). The response rate was thus 5.1% in total. These valid answers were used in this study. Three sample answers to the questionnaire, one for each site, are provided in Table 4. The questionnaire consisted of two sections, asking for basic information about the household and details of its energy consumption. Thus the first section contained questions on type of housing, number of household members, their age and income, distance to work and means of transport, and main household appliances. The second section of the questionnaire explored the main use of energy. In terms of fuel categories many types of energy carriers were considered, but only three types are actually used by respondents: electricity, natural gas and petrol. Electricity was the chief energy carrier, data on which were acquired from their bi-monthly electricity bills (metre reading occurs every two months in Nanjing) from January to November 2010. Electric energy use was then calculated in terms of kilowatt-hours per household over a period of one year. Coal and liquefied petroleum gas were once the main fuels for domestic cooking, but today they are usually used in commercial boilers. Natural gas is currently the main source of energy for domestic cooking in Nanjing and data on natural gas consumption were obtained from the east area of Sichuan province (Nanjing Local Chronicles Compilation Committee, 2011). Petrol is the dominant fuel used for private vehicles in China, while diesel, compressed natural gas and liquefied petroleum gas are more often used for public transport in Nanjing (Zhu, 2010). The amount of petrol

consumption was estimated by multiplying the average petrol consumption per kilometre by the total driving distance. For those using public transport, energy consumption was calculated using the average fuel efficiency of public transport and travel distance (IPCC, 2006). In order to relate the data to the discussion on climate change, the amount of energy consumed was further converted into CO2 emissions using the IPCC carbon emissions calculation formula (2006 edition) (IPCC, 2006). It was concluded that the IPCC formula was the best available option for this study, although a field-tested formula would have been more accurate. A problem is there is no one-to-one correspondence between energy consumption and CO2 emissions, since they are the result of many factors that act to increase emissions together with factors that act to reduce emissions. Much work has been done on the break-down of changes in energy use or emissions using indices (Liu, Ang, & Ong, 1992; Greening, Davis, Schipper, & Krushch, 1997; Schipper, Haas, & Sheinbaum, 1996; Shorrock, 2000). However, they relate to a given time and place. Here, the IPCC carbon emission calculation formula was used to convert three types of energy carriers (IPCC, 2006). Although they are not field test data, the carbon emissions from energy consumption are generally accurate. 4. Results and discussions Many attempts have been made to model energy consumption at residential and household level (Permana, Perera, & Kumar, 2008). Unfortunately, the models usually describe and forecast electricity demand rather than total household energy consumption. Inclusion of transport energy and cooking energy in a household’s energy consumption is too complicated to be modelled. However, in this study we tried to include all of these and analyse the dominant features. 4.1. General state of energy consumption and CO2 emissions The CO2 emissions per capita of the respondents for household and transport amounted to 1.97 t in 2010, of which the CO2 emissions from electricity consumption comprised 1.15 t (58%) (Table 1). Per capita CO2 emissions from natural gas consumption were 0.05 t (3%) and from petrol consumption 0.77 t (39%). The lowest per capita annual electricity consumption of a family was 572.1 kWh and the highest was 2369.8 kWh, which was more than 4.1-fold greater than the lowest. The lowest per capita annual natural gas consumption of a family was 4 M3 and the highest was 43.5 M3 , 10.9-fold greater than the lowest. The lowest per capita annual petrol consumption of a family was 59 L and the highest was 750 L, about 13-fold greater than the lowest. Overall, the lowest per capita annual CO2 emissions of a family was 0.85 t and the highest was 3.94 t, 4.6-fold greater than the lowest. Thus the difference in petrol and natural gas consumption between the highest and lowest consumers was much greater than the difference in electricity consumption. The Human Development Report 2007/2008 (UNDP, 2008) reported that per capita energy-related CO2 emissions in China were 4.3 t in 2007 and it forecast that by 2015, these emissions

56

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

kWh

1000.0

893.1

900.0 800.0 700.0 600.0

761.5 644.5 593.0 584.5 550.0

500.0

637.3

502.2 433.9

487.6 452.4 450.7 418.8

10 Mar. - 9 May

10 May - 9 July

442.5 391.5

400.0

754.0

478.6 469.0

610.3 551.8 547.2 498.0

446.2 390.9

300.0

Average Site A Site B Site C

200.0 100.0 0.0 10 Jan. - 9 Mar.

10 July - 9 Sep.

10 Sep. - 9 Nov.

10 Nov. - 9 Jan.

Fig. 6. Electricity consumption (kWh) per household in different seasons (kWh). Summer = July–September.

would have increased to 5.2 t per capita. Considering the low carbon emissions (2 t CO2 per capita) in rural areas, the CO2 emissions in urban areas in that year must have been at least 6.6 t per capita, as China has approximately equal numbers of rural and urban residents (National Bureau of Statistics of China, 2012a). Therefore, about 30% of total urban CO2 emissions in Nanjing were from domestic appliances in households and transport by residents. For reference, in 2007 CO2 emissions per capita were 9.95 t for Shanghai (not including ship and air transportation) (Liang, Liu, & Peng, 2010). In the same year, the CO2 emissions per capita were 23.73 t in Wuxi city (including all industry and transportation),1 6–10% of which were from goods transport, 40% from passenger transport and 3–5% from households (Wang, Zhang, & Bi, 2011). Hence total carbon emissions from residences were 5.4–9% or 1.28–2.13 t in Wuxi, which matched our data for Nanjing. 4.2. Climate factors Climate is one of the most important factors influencing household electricity consumption. As a sub-tropical city, Nanjing’s electricity consumption shows clear season changes. Fig. 3 shows bi-monthly electricity consumption per household. In line with the mean monthly temperature in Nanjing, electricity consumption showed accompanying seasonal trends. Electricity consumption in summer (July–September) was almost twice that in spring and autumn, and also considerably higher than that in winter (Fig. 6). It was obvious that electricity consumption over the normal level was used for cooling in summer and heating in winter. It can be inferred that energy consumption for cooling and heating as a proportion of total electricity consumption was about one-seventh (14%), which was far below the proportion in developed countries (50–60%) (GU, Vestbroc, Wennerstena, & Assefaa, 2009). However, with an improved standard of living, heating and cooling will consume more electricity, and the potential energy reduction for air-conditioning will exceed that of other household appliances. Thus reducing the use of air-conditioning and improving the performance of air-conditioning equipment should be key concerns in designing energy-efficient buildings. Although climate is the main factor in the seasonal variation in household electricity consumption, it cannot explain the distinct rise in consumption in the satellite town (Site C) in summer, whereas it was less than in the spontaneously formed residential area (Site B) in other seasons. By rights, the buildings in the satellite town should be energy-efficient owing to their better insulation and air-tight windows. In other words, to achieve an acceptable indoor climate, the electricity consumption at Site C should be less

than that at Site B. A plausible explanation is that the households at Site C used more appliances to get a more comfortable living climate. A detailed analysis of this is provided in Sections 4.4 and 4.5. 4.3. Geographical factors Location of the home in relation to work was clearly an important factor influencing a household’s transport energy consumption. Fig. 7 shows CO2 emissions per capita in the different areas. Site A had the lowest transport-related emissions, about 71% of those at Site C, which had the highest transport emissions. However, electricity and natural gas consumption showed only small differences between the different sites. The main difference was in petrol consumption, where that at Site C was 220% of that at Site A and 194.8% of that at Site B. This shows that location has an important influence on residents’ energy consumption for transport. Energy use for transport was greater in the satellite town (Site C) than in the old town (Site A). The CO2 emissions from transport accounted for 31% of total CO2 emissions at Site A, 35% at Site B, and 49% at Site C. The greater the distance from the city centre, the higher the proportion of transport energy consumption. Although significant local services and subways were planned and installed in Hexi, the long distance to the downtown area still required more vehicle use than for residential areas in, or close to, the old city. Mixed function is a frequently used planning strategy to reduce transport requirements. Theoretically, an area with mixed residential, business, commercial and service functions is able to radically reduce transport energy consumption. This study found that there were almost zero transport energy requirements for travelling to work places in the downtown area. However, this is a special or even exceptional case. Although all respondents wanted to live near their workplace, in most cases they had to live in the new town because of the high price of properties in the city centre, where

kg CO2 3000.0 2500.0

2000.0

Petrol Natural gas Electricity

1500.0 1000.0

500.0 0.0

Average 1

Wuxi is an industrial city near Nanjing. It has a similar level of economic development and climate to Nanjing.

Site A

Site B

Site C

Fig. 7. CO2 emissions per capita with respect to location for the different energy carriers studied.

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61 Table 2 Comparison of per capita electricity consumption (kWh) according to dwelling floor area and construction period. Floor area (m2 )

Before 2000

2001–2005

2006–2011

Total

<30 31–60 >60 All

1005.0 1090.8 – 1062.2

1403.4 1184.4 1671.6 1343.5

1140.8 906.4 1507.3 1167.5

1124.2 1087.2 1589.4

most companies and businesses are located. At the same time, with the increase in land value in the city centre, more and more residential functions are being replaced by commercial and business functions. Since 2005, there have been very few new residential projects built in the old town, and most new development projects there have been commercial and business buildings. These more single-function districts will aggravate the transport problems. In the 1990s, some city planning experts presented the concept of a ‘compact city’ to counteract infinite expansion by the intensive use of urban space (Jenks et al., 1996). The original intention of the compact city was to solve the problem of urban sprawl arising from rapid population growth. Today the compact city theory is being developed into a paradigm of sustainable urban form. The right path may be approached from two aspects: on the one hand, reversing the trend of removing residential functions from the old town; on the other hand, supplying commercial and business function in the residential communities in the new town. 4.4. Architectural factors Since the energy consumption of residences was the main object of the study, the architectural design of buildings must be mentioned. The relationship between thermal performance of a building and electricity consumption is a common problem encountered by architects (Gu et al., 2009). Most buildings at Site A were constructed before 2000, when China had no regulations on the thermal performance of buildings in this climate zone. There were no heat insulation measures used in those buildings. Once the Ministry of Construction of P.R. China (2001) issued its “Design standard for energy efficiency of residential buildings in Hot Summer and Cold Winter Zone (JGJ 134-2001)”, the first residential buildings with insulation were built in Nanjing in November 2003 (Zhang, 2006). Architects were involved in designing more energy-efficient buildings. Most buildings at Site B constructed during that period had basic heat insulation measures. After 2006, almost all the new residential buildings had thicker external wall insulation and double-glazed windows in order to meet the requirements on thermal performance in new building regulations (Nanjing Municipal Government, 2006). Most buildings at Site C were constructed during that period. Table 2 shows per capita electricity consumption according to dwelling floor area and construction period. No dwellings were built after 2005 at Site A and none before 2000 at Site C. As a result, the dwellings at Site A were smaller than those at Site C. Before 2000, residential units larger than 60 m2 per capita were so rare that the data on these are not statistically representative. Generally, larger units consume more electricity. Therefore, limiting the per capita floor area may be as important as improving building thermal performance. However, the smallest units were not the most electricity-efficient because of the threshold effect – only one or two householders lived in these dwellings but they had to operate the entire electrical system. Almost all the buildings built before 2000 were compact because of the narrow plots in the downtown area, while the buildings built after 2001 were more spacious as land was not as restricted, which explains the greater electricity consumption after 2001. The buildings built after 2006 were more

57

somewhat energy-efficient owing to the new thermal construction, but this brought an improvement of only 10% electricity saving. The reason is the residents’ self-discipline in using cooling/heating systems. According to JGJ 134-2001, the CDD26 (cooling degree-days based on 26 ◦ C) of Nanjing is 175, and the annual cooling electricity consumption is 24.9 kWh/m2 . However, the highest cooling electricity consumption of the respondents in this study was about 9.5 kWh/m2 , which was far below the standard. This does not mean that the indoor climate was poor, but rather that the residents usually only turned on their air-conditioning when the temperature exceeded 28 ◦ C. The energy-saving performance due to thermal insulation of external walls usually depends on the reference value and only when the reference value is large will the energy savings be effective. Because the residents paid for space heating and cooling according to their actual energy usage, they tended to avoid turning on the heating and cooling devices when it was not very necessary. Hence the actual energy consumption for heating and cooling was much lower than the reference value and the thermal insulation of external walls was not as effective as the theoretical value. However, while electricity consumption for heating and cooling was not the major contributor to total electricity consumption in the households surveyed, it is increasing rapidly in Chinese cities (Gu et al., 2009). If measures to improve the thermal performance of all residential buildings are not introduced, heating and cooling will consume more electricity in the future. Residential buildings usually do not have central heating and cooling systems in the hot summer and cold winter area of China. There are a few so-called high-end residential communities with central heating and cooling systems in Nanjing. They installed central heating and cooling systems that were designed not for energy efficiency, but for thermal comfort. For instance, in Landsea International Block, a famous ‘green’ residential community in Hexi, the cooling energy consumption in 2010 was 27 kWh/m2 (Meng, Zhang, Yang, & Yang, 2011), which was almost three times that of our respondents. Although central systems have higher coefficient of performance (COP), they cannot adapt to the radical energy fluctuations in residential buildings (Zhang, Wang, & Yuan, 2007). In many cases, the heating and cooling systems are idling, which counteracts their high COP. In comparison, household air-conditioning units have lower COP, but they have the flexibility to adapt to all kinds of requirements in residential buildings. One problem of household air-conditioning is that the architectural design must set aside space for the outdoor component of air-conditioning units (Fig. 8). Energy efficiency of buildings does not rely on the excessive pursuit of high COP building devices, but on reducing the use of energy-consuming devices within the buildings, which should be the main focus of architectural design. 4.5. Economic factors As mentioned above, climate cannot explain the higher electricity consumption in the satellite town (Site C) in summer. By rights, the buildings in the satellite town should be electricity-efficient, with their better insulation and air-tight windows. That is to say, to get the same indoor climate, electricity consumption at Site C should be less than that at Site B. A plausible explanation for the unexpected higher electricity consumption is that the households at Site C used more appliances to get a higher level of indoor comfort. Income is an important factor for energy consumption. As income rises, energy consumption tends to increase, as numerous studies have confirmed (Chern, Ishibashi, Taniguchi, & Tokoyama, 2003; Lahiri, Babiker, & Eckaus, 2000). Table 3 shows total energy consumption with respect to per capita income for the residents

58

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

Fig. 8. Heating and cooling systems are prominent features on the fac¸ade of residential buildings. (a) This flat building in the Landsea community has a clean fac¸ade. (b) A flat building without central heating and cooling has to install blinds to shield the outdoor components of air-conditioning units.

Table 3 Per capita income and energy consumption in different forms by the residents surveyed in Nanjing. Annual income (USD)

Electricity (kWh)

Natural gas (M3 )

Petrol (L)

<5000 5001–7500 7501–10,000 10,001–12,500 12,501–15,000 >15,000

760.2 827.5 1163.6 1435.5 1831.5 1211.4

20.0 31.5 29.4 28.9 23.9 26.3

112.0 172.4 274.0 342.4 391.7 288.7

kg CO2 3000.0 2500.0 2000.0

Petrol Natural gas Electricity

1500.0 1000.0 500.0

USD/a >1 50 00

<5 00 0

50 01 ~7 50 0 75 01 ~1 00 00 10 00 1~ 12 50 12 0 50 1~ 15 00 0

0.0

Fig. 9. CO2 emissions per capita with respect to income of the residents surveyed in Nanjing.

surveyed in Nanjing, while Fig. 9 shows the CO2 emissions distribution. The average per capita annual disposable income was 4400 USD in Nanjing in 2010 (Nanjing Statistic Bureau, 2011). Only three respondents (13 inhabitants) were lower than this standard,2 and all these respondents were not tenants but owned their property, so they have to pay for electricity themselves. As Table 3 shows, higher economic capability supported higher energy consumption (Table 4).

2 The official data did not calculate folk capital flow, which was an important part of private income. The real income data obtained by the questionnaire were usually higher than the official statistical data.

The decline found in the energy consumption of the highest earning people was unexpected. One reason may be that high-income people are at home for less time, which decreases household energy consumption. They value time very highly and cannot bear wasting time for commuting, so they often live near the office, which also decreases their energy consumption for transport. Another reason may be higher environmental awareness, causing them to restrict energy consumption. A problem that should be noted in this regard is energy transfer, i.e. highincome people may have consumed less electricity in the household because they consumed more energy at other places, e.g. by spending more time in hotels, restaurants, pubs, spas, game rooms, gymnasiums, etc. Some simply regard the home as a dormitory, which leads to low energy consumption in the household, but they do not necessarily consume less energy for living. Thus, the energy is not saved, but transferred. In general, increasing wealth induces more energy consumption, as confirmed by the data for low to middle income groups in this study. However, it is impossible to achieve energy efficiency by slowing down economic development. Other methods must be identified to decrease energy consumption even when the economy is growing. 4.6. Social factors It is difficult to quantitatively examine the influence of social factors on energy consumption. This study focused on the relationship between energy consumption structures, family structures and living habits arising from these. On the whole, different energy carriers maintained a relatively constant proportion in energy consumption structures (Mi, Nie, Li, & Li, 2011). However, for some concrete cases, they were

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

59

Table 4 Three examples from the questionnaire sheets. Location

Site A

Site B

Site C

Floor area Built time Number of household members Family members

84 m2 1998 2 Man, 31 Wife, 30

98 m2 2001 3 Man, 36 Wife, 35 A child, 3

Family annual income Distance to duty and means of transport

30,000 USD Man: 6 km, bus Wife: 5.5 km, bicycle

21,000 USD Man: 9 km, car Wife: 8 km, bus

139 m2 2006 5 Man, 35 Wife, 31 Man’s father, 67 Man’s mother, 63 A child, 2 55,000 USD Man: 13 km, car Wife: 10 km, subway

Main household appliances

Air conditioning: 1 Elec. space heater: 2 Refrigerator: 1 Washing machine: 0 Computer: 2 Elec. water heater: 0 Gas water heater: 1 Electric lighting: 5 sets and 4 point lighting sources LCD TV set: 1 Solar heater: 0 Electric bicycle: 1

Air conditioning: 2 Elec. space heater: 1 Refrigerator: 1 Washing machine: 0 Computer: 2 Elec. water heater: 0 Gas water heater: 1 Electric lighting: 5 sets and 4 point lighting sources LCD TV set: 1 Solar heater: 1 Electric bicycle: 1

Air conditioning: 4 Elec. space heater: 1 Refrigerator: 1 Washing machine: 1 Computer: 3 Elec. water heater: 1 Gas water heater: 1 Electric lighting: 7 sets and 10 point lighting sources Plasmon TV set: 2 Solar heater: 0 Electric bicycle: 1

Elec. consumption 10 January–9 March 10 March–9 May 10 May–9 July 10 July–9 September 10 September–9 November 10 November–9 January

4719 kWh 780 kWh 577 kWh 622 kWh 629 kWh 942 kWh 1169 kWh

2611 kWh 725 kWh 651 kWh 403 kWh 476 kWh 356 kWh 1234 kWh

5214 kWh 743 kWh 613 kWh 681 kWh 1508 kWh 685 kWh 984 kWh

Natural gas Petrol

24 M3 250 L

46 M3 410 L

140 M3 560 L

kWh 1500

1387

1200

1059

900 600

820

520

456 289

300

632

625

603

231

496 290

335

325 307 227

567

Peak electricity Off-peak electricity Total electricity

269

0 10 Jan. - 9 Mar.

10 Mar. - 9 May

10 May - 9 July

10 July - 9 Sep.

10 Sep. - 9 Nov.

10 Nov. - 9 Jan.

Fig. 10. Actual electricity consumption (kWh) by one resident surveyed.

interrelated. Electricity and household appliances, natural gas and cooking and hot water, petrol and transport usually had corresponding correlations. If some function used a different energy type, the energy consumption would be different, for instance, gas water heater or electrical water heater, gas stove or electromagnetic oven, motorcycle or electric bicycle, etc. Electrical appliances are usually the most convenient option, but the generation of electricity should also be considered. Actually, electricity is not an energy source but an energy carrier. Its environmental impact depends on how it is produced and today electricity is mainly produced in the world by hydropower, nuclear power and fossil fuels. A feature in common for the different engine technologies available is low efficiency, as the majority of the primary energy is dissipated as heat, which is often not utilised. It is thus obvious that transport planning in cities has to develop in another direction, namely to reduce the use of private cars. In any case, petrol cannot be recommended as an energy carrier because of its high carbon emissions and environmental impact. Although natural gas is an available alternative energy source to petrol in the

near future in terms of carbon emissions and reserves, electricity is likely to be the only available energy carrier when fossil fuels are exhausted in the future. Nanjing is a city with four distinct seasons, which brings both challenges and opportunities for energy consumption in space heating and cooling. Fig. 10 shows the bi-monthly electricity consumption of one resident that is far below the distribution of overall electricity consumption, where cooling consumed more energy than heating (see Fig. 3). It is surprising that the electricity consumption of the resident in Fig. 7 was higher in winter than in summer. Another unusual finding is that off-peak electricity consumption was more than peak electricity consumption3 in most months. Further investigation showed that the householders were

3 Nanjing Power has a policy of peak/off-peak electricity pricing. From 8 am to 9 pm is peak time and from 9 pm to 8 is off-peak time. At the time of study, the price of peak electricity was 0.56 Yuan/kWh and the price of off-peak electricity was 0.36 Yuan/kWh.

60

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

kg 2500.0

CO2 Emissions

2000.0 1500.0

Petrol Natural gas Electricity

1000.0 500.0

0.0 1

2

3

4

>5

household populaons Fig. 11. CO2 emissions per capita with respect to number of members per household.

accustomed to working at night and because of this working pattern they needed little air-conditioning in summer, but much space heating in winter. Therefore, if people were to make use of daytime to work in winter and night-time to work in summer, energy consumption patterns would better match the seasonal temperature pattern and electricity consumption would remain low. Besides the temperature pattern, transport also followed a familiar pattern. During the morning and evening rush hour, vehicles take two or three times longer to travel their normal route. Time and petrol are wasted during the traffic jams. If people were to avoid the rush hour, the commuting time would be shortened dramatically. Saving time means saving energy. Since petrol is mainly used for commuting transport, changing travel patterns has special significance in saving petrol. For China’s one-child policy, it is rare for a family to exceed five members.4 The traditional extended Chinese family, with tens of members, has disappeared in today’s cities. The mean number of household members among the questionnaire respondents was 2.84 per household, while as an average for Nanjing this number decreased from 2.92 per household in 2000 to 2.77 per household in 2010 (Nanjing Statistic Bureau, 2010). The questionnaire households can be divided into two sets by number of generations represented: small families with one or two generations, and large families with three or four generations. The different habits within different age brackets induced different energy consumption. Fig. 11 shows the CO2 emissions per capita with respect to number of household members. The overall trend was for CO2 emissions per capita to decrease when household size increased, which indicates that large families are more energy-efficient. For retired people, the household is the main area of activity. They cook and eat at home and their social activities are usually limited to the local community. Thus they consume very little energy for transport. Natural gas was used almost only for cooking in the households surveyed. There was clear trend in natural gas consumption for household size. The large families, including the elderly and children, usually cooked at home and therefore their natural gas consumption was higher. The small families without elderly people usually ate fast food or at restaurants, which led to very low natural gas consumption in the home. Private vehicles began to be popular at the beginning of the 21st century and although per capita car ownership is still at a low level, it is increasing rapidly. There were more than 1.3 million vehicles

4 Actually, people can have more than one child for some reasons, e.g. husband and wife are both ‘one-child’, remarriage, twins, paying for social support fees, etc. Hence there were still some families with more than five members during the survey. One household surveyed had 10 members: a man and his wife, the man’s two children, the man’s parents, and 4 tenants.

in Nanjing by June, 2011 (Zhu, 2011). Young people surveyed here preferred to drive the car to work when there was no convenient bus or subway line connecting the home and the workplace. However, it was rare for an elderly person to be able to drive a car. The main means of transport for the elderly were buses or bicycles. Therefore even in a large family, no more than two members drove cars, which led to minor petrol consumption per capita in this category. Family structures and life habits can be changed through many approaches, e.g. education, economic incentives, social security policies, etc. Urban planning is another possible approach. Urban planners should consider these issues more carefully in future and bear in mind that city planning based on aesthetics is not necessarily suitable for optimising urban energy consumption.

5. Conclusions This study in Nanjing revealed that household use and transport were the two main contributors to domestic energy consumption in the city. Household electricity consumption showed obvious seasonal characteristics, being higher in summer than in the other three seasons. Transport energy consumption showed geographical characteristics, with the old town nearest the business area having the lowest transport energy consumption. Household devices with a high coefficient of performance (COP) did not make buildings more energy-efficient, so architectural design should seek to reduce the use of such devices. Energy consumption generally increased with increasing income, especially at lower levels. Family structure also influenced energy consumption, with high-income families and small families consuming more energy per capita. The results showed that economic and social factors were equally important to technical factors for energy efficiency. Based on the findings above and considering sustainable urban development, some possible policies that could indirectly affect energy consumption are:

1) Urban sprawl cannot solve developing city problems. Making the most of existing city land to develop a compact city is the only way to create an energy-efficient city. Promotion of mixed residential and commercial activities without creating land use conflicts is the most important strategy to decrease energy consumption for transport. The CO2 emissions from a fully mixed-function city area are only about 70% of those of a wholly residential area. Urban planning should reverse the trend of removing residential functions from the city centre and provide more commercial and business functions in suburban areas. 2) Thermal performance should be improved not only by constructing new buildings, but also by refurbishing old buildings. An equally important consideration is to restrict per capita floor area. However, small households containing only one or two members are less energy-efficient per capita than households with more members. Unfortunately, modern families are becoming smaller in China. The government should promote large households containing several generations. 3) Satellite towns should be developed more carefully. Land finance is currently a major driving force in the development of new satellite towns in China. Even if new satellite towns are unavoidable, they should be as close to downtown areas as possible. The provision of adequate public transport to connect satellite towns and central city is essential. While local services may meet most requirements in new towns, transport between satellite towns and city centre is unavoidable. Vehicles are negative for energy efficiency and environmental impact.

Z.H. Gu et al. / Sustainable Cities and Society 7 (2013) 52–61

4) Energy issues should be considered from a systems perspective. Despite high performance by a single element, e.g. central air-conditioning, the system may not be efficient. The highest technology also has a scope of application. The best adapted is the best measure. It is important to explore and develop interactions between systems at different levels within the city and between the city and its surroundings. There is a need for more intelligent design of buildings and better systems for energy distribution, where locally produced energy can be complemented with centrally produced energy in new smart grids for electricity and low temperature district heating and cooling systems. By working with neighbourhoods, the interactions between systems can be examined from a technical but also from a social and economic perspective. Acknowledgements This study was supported by the National Natural Science Foundation of China (Grant No. 50908043). The authors would also like to thank Prof. Dongqing Han (Southeast University, China), Yinyan Xu (Nanjing Maternity and Child health Care hospital, China), Dr. Zheng Wang, Dr. Hua Liu, Dr. Bin Tang, Dr. Rong Fang, Yuan Meng, Danhong Ma, Yiming Kong, Qiuping Tang, Siyuan Han, Feifan Ma and Jingxian Zhu (Southeast University, China). References CDC (Climatic Data Center), National Meteorological Information Center, China Meteorological Administration, 2011. China Meteorological Data Sharing Service System, available in http://cdc.cma.gov.cn/index.jsp Chen, S. Q., Yoshino, H., & Li, N. P. (2010). Statistical analyses on summer energy consumption characteristics of residential buildings in some cities of China. Energy and Buildings, 42, 136–146. Chern, W. S., Ishibashi, Kimiko, Taniguchi, Kiyoshi, & Tokoyama, Yuki. (2003). Analysis of food consumption of Japanese households. Rome: Food and Agriculture Organizations of the United Nations. Chinanews.com, 2011. Expert: China’s CO2 emissions account for 23% of the world’s total emissions. Chinanews.com, Beijing, China. http://www.chinanews. com/ny/2011/12-04/3506192.shtml (accessed on 04.12.11). Close, J. (1996). Integration of photovoltaics within high rise buildings in the dense urban environments of SE Asia, consideration of legislation to promote it and to maintain solar energy access. Renewable Energy, 8(1–4), 471–474. Crompton, P., & Wu, Y. (2004). Energy consumption in China: past trends and future directions. Energy Economics, 27(2005), 195–208. Greening, L. A., Davis, W. B., Schipper, L. J., & Krushch, M. P. (1997). Comparison of six decomposition methods: application to aggregate energy intensity for manufacturing in ten OECD countries. Energy Economics, 19(3), 375–390. Gu, Z. H., Vestbroc, D. U., Wennerstena, R., & Assefaa, G. (2009). A study of Chinese strategies for energy-efficient housing developments from an architect’s perspective, combined with Swedish experiences and game theory. Civil engineering and environmental systems, 26(4), 323–338. Heilig, G. K. (1999). Can China feed itself? A system for evaluation of policy options. Laxenburg: IIASA. (CD-ROM Vers. 1.1) Hickman, R., & Banister, D., 2007. Transportation and reduced energy consumption: what role can urban planning play? Working Paper no. 1026, Transport Studies Unit, Oxford University Centre for the Environment. IEA. (2010). CO2 emissions from fuel combustion highlights (2010 ed.). Paris, France: International Energy Agency. http://www.iea.org/co2highlights/ CO2highlights.pdf (accessed on 02.01.12) IPCC. (2006). 2006 IPCC guidelines for national greenhouse gas inventories. In H. S. Eggleston, L. Buendia, K. Miwa, T. Ngara, & K. Tanabe (Eds.), Prepared by the National Greenhouse Gas Inventories Programme. Japan: IGES. Jenks, M., Burton, E., & Williams, K. (1996). The compact city a sustainable urban form? Oxford Brookes University, Oxford, UK: Spon Press.

61

Lahiri, S., Babiker, M., & Eckaus, R. S. (2000). The effects of changing consumption patterns on the costs of emission restrictions. In MIT Joint Program on the Science and Policy on Global Changes, Report no. 64. Liang, Z. H., Liu, X. Y., & Peng, W. B. (2010). The long term view of Shanghai low carbon city development. Blue book of Shanghai: Annual report on resources and environments of Shanghai 2010. Beijing: Social Sciences Academic Press (China). Liu, X. Q., Ang, B. W., & Ong, H. L. (1992). The application of the Divisia index to the decomposition of changes in industrial energy consumption. Energy Journal, 13(4), 161–171. Ma, L. J. C. (2002). Urban transformation in China, 1949–2000: a review and research agenda. Environment and Planning, 34(9), 1545–1569. Meng, C., Zhang, L., Yang, C. H., & Yang, Y. X. (2011). Assessment and analysis of energy efficiency for Landsea international Block, Nanjing. Construction Science and Technology, (14), 46–49. Mi, L. Y., Nie, R., Li, H. L., & Li, X. H. (2011). Empirical research of social norms affecting urban residents low carbon energy consumption behavior. Energy Procedia, (5), 229–234. Mihalakakou, G., Santamouris, M., & Tsangrassoulis, A. (2002). On the energy consumption in residential buildings. Energy and Buildings, 34(7), 727–736. Ministry of Construction of China, 1993. Thermal design code for civil building (GB 50176-93). Nanjing Local Chronicles Compilation Committee, 2011. Nanjing Almanac 2010. Nanjing Municipal Government, 2006. The options about improving the energy efficiency of buildings in Nanjing. Nanjing Urban Planning Bureau, 2006a. Regulatory plan of Nanjing old town (Deepening version in 2006). Nanjing Urban Planning Bureau, 2006b. City master plan of Nanjing (2007–2030). Nanjing Statistic Bureau, 2010. Communiqué of Nanjing on the major figures of the 2010 population census. Nanjing Statistic Bureau, 2011. Communiqué of Nanjing on the 2010 National Economic and Social Development. National Bureau of Statistics of China, 2012.China’s total population and structural changes in 2011. National Bureau of Statistics of China, Beijing, China. http://www.stats.gov.cn/was40/gjtjj en detail.jsp?channelid=1175&record=9 (accessed on 10.02.12). National Bureau of Statistics of China, 2012. Communiqué of the National Bureau of Statistics of People’s Republic of China on Major Figures of the 2010 Population Census (No. 1). Permana, A. S., Perera, R., & Kumar, S. (2008). Understanding energy consumption pattern of households in different urban development forms: A comparative study in Bandung City, Indonesia. Energy Policy, 36(11), 4287–4297. Schipper, L. J., Haas, R., & Sheinbaum, C. (1996). Recent trends in residential energy use in OECD countries and their impact on carbon dioxide emissions: a comparative analysis of the period 1973–1992. Journal of Mitigation and Adaptation, 1, 167–196. Shorrock, L. D. (2000). Identifying the individual components of United Kingdom domestic sector carbon emission changes between 1990 and 2000. Energy Policy, 28(3), 193–200. UNDP. (2008). Human Development Report 2007/2008, fighting climate change: human solidarity in a divided world. New York: Palgrave Macmillan. Wang, M., & Wei, D.Y., 2010. Population & natural resources case study: what are the challenges of meeting the resource needs of a very large population? In Solem, ˜ M., Klein, P., Muniz-Solari, O., & Ray, W., eds., AAG Center for Global Geography Education. Available from http://globalgeography.aag.org Wang, H. K., Zhang, R. R., & Bi, J. (2011). Carbon accounting for Chinese cities: a case of Wuxi City. China Environmental Science, 31(6), 1029–1038. Xu, B., Sun, Q., Wennersten, R., & Brandt, N. (2010). An analysis of Chinese policy instruments for climate change mitigation. International Journal of Climate Change Strategies and Management, 2(4), 380–392. Yeh, A.G.-o., Yuan, H.-q., 1986.Satellite town development in China: problems and prospects. Working Paper no. 18, Centre of Urban Studies & Urban Planning, University of Hong Kong. Zhang, Y. Z. (2006). Ju Fuyuang in Nanjing: energy efficiency and green building design and innovation. Construction Science and Technology, (7), 42–43. Zhang, Q., Wang, Y. H., & Yuan, D. L. (2007). Residential energy efficiency for Landsea International Block. Heating Ventilation and Air Conditioning, 37(9), 123–126. Zhu, S. L. (2010). Comparison of transportation energy consumption and greenhouse gas emission between Beijing and Shanghai. Urban Transport of China, 8(3), 58–63. Zhu, J.J., 2011. Registered number of automobiles has exceeded 1.3 million. Modern Express of Jiangsu reported July 20.