The Impacts of City Size and Density on CO₂ Emissions: Evidence from the Yangtze River Delta Urban Agglomeration

Abstract

With rapid urbanization, cities plays an increasingly key role in addressing CO₂ emission-related issues. This study aims at analysing the spatial relation between city size and CO₂ emissions in the case of Yangtze River Delta Urban Agglomeration (YRDUA) in China over the years from 2006 to 2016, making a contribution to the existing body of knowledge of the relationships between urban form and CO₂ emissions. This paper identified the following main findings: (1) There is a U-shape relationship between population size and CO₂ emissions in YRDUA; (2) There is a negative sublinear relationship between CO₂ emissions and city density in YRDUA; (3) The ideal urban form for low CO₂ performance in YRDUA is 2.716 million people living in high population density; (4) Increasing population size is an effective but not a long-term approach for CO₂ emissions reduction, because for every marginal increase of city density, the marginal reduction of CO₂ emission will decrease. (5) A demographic change in YRDUA from low-density cities to high-density cities would benefit CO₂ emission performance. These findings confirm the important roles of population size and density for CO₂ emissions reduction in urban agglomeration and so help shape current policy debates.

Appendices

Appendix 1

The Calculation Steps of CO₂ Emissions

This study has referenced a reasonably top-down method that was developed by Jing et al. (2018) to calculate CO₂ emissions at city level, mainly including the following steps.

Firstly, the EBTs of Shanghai city, Jiangsu province, Zhejiang province, Anhui province between 2006 and 2016 from China Energy Statistical Yearbook are available and transparent. In the EBTs, there are 17 categories and 27 types of fuels as shown in Table

Table 5 Energy consumption categories of EBT and distribution indicators

5 and Table

Table 6 Types of fuels and emission factors (EF_j)

6 respectively. For Shanghai city, there are existing EBTs at the city level, while for the other 25 cities, the EBTs were transferred from the provincial level to the city level by the appropriate distribution indicators which were selected considering the following two principles. The indicators were available at a city level continuously over the years. The meaning of the indicators can almost represent the categories in the provincial EBTs. Considering the above principles, several distribution indicators were chosen from the China Statistical Yearbook, China City Statistical Yearbook, or China provincial statistical yearbooks for the corresponding categories as shown in Table 5.

After the preparation of provincial EBTs data and the city-level distribution indicators, it was possible to obtain the EBTs for each city by using the following three equations:

$$\alpha_{i} = \frac{{O_{i}^{C} }}{{O_{i}^{P} }}$$

(5)

$$AD_{i,j}^{C} = AD_{i,j}^{P} *\alpha_{i}$$

(6)

$$\mathrm{AD}\overset{\mathrm C}{\mathrm j}={\textstyle\sum_{\mathrm I=1}^{17}}\mathrm{ADi},\overset{\mathrm C}{\mathrm j}$$

(7)

In Eq. (5), i represents the categories in provincial EBTs; ${\alpha }_{i}$ is the distribution coefficient in i; ${O}_{i}^{C}$ refers to the city-level distribution indicator of i in city C; ${O}_{i}^{P}$ refers to the sum of the distribution indicators of i which is all of the cities in province P.

In Eq. (6), j refers to the fossil fuel types shown in Table 6; ${AD}_{i,j}^{P}$ refers to the consumption of fossil fuel j in category i in province P, which is available in provincial EBTs; ${AD}_{i,j}^{C}$ is the consumption of fossil fuel j in category i in city C.

In Eq. (7), ${AD}_{j}^{C}$ is the total consumption of fossil fuel j with all 17 categories in city C.

Then, with the city-level EBTs, emission factors (Table 6) for each consumption of fossil fuel were taken into account to calculate the CO₂ emissions. The following Eq. (8) is the IPCC recommended approach (Paustian et al., 2006).

$${CE}^c={\textstyle\sum_{j=1}^{27}}{AD\overset cj\;\ast\;{EF}_j}$$

(8)

where ${EF}_{j}$ refers to the emission factor of fossil fuel j; ${CE}^{c}$ represents the total CO₂ emissions in city C, which is related to the total consumption of 27 types of fossil fuel.

Additionally, the CO₂ emissions data as dependent variables of 26 cities in YRDUA from 2006 to 2016 were put into a new dataset with independent and control variables.

Appendix 2

List of Acronyms

EBT	Energy balance tables
FE	Fixed effects
GHG	Green house gas
GDP	Gross domestic product
IPS	Im-Pesaran-Shin
IPCC	Intergovernmental panel on climate change
LLC	Levin-Lin-Chu
RE	Random effects
SAM	Spatial autoregression model
YRDUA	Yangtze river delta urban agglomeration