Peak Assessment and Driving Factor Analysis of Residential Building Carbon Emissions in China’s Urban Agglomerations

Abstract

Urban agglomerations, as hubs of population, economic activity, and energy consumption, significantly contribute to greenhouse gas emissions. The interconnected infrastructure, energy networks, and shared economic systems of these regions create complex emission dynamics that cannot be effectively managed through isolated city-level strategies. However, these regions also present unique opportunities for innovation, policy implementation, and resource optimization, making them crucial focal points in efforts to reduce carbon emissions. This study examines China’s three major urban agglomerations: the Yangtze River Delta, the Pearl River Delta, and the Beijing–Tianjin–Hebei region. Utilizing data from 2005 to 2020 and a comprehensive evaluation model (BCPCAM), the research offers more profound insights into the socio-economic factors and collaborative mechanisms influencing emission trends, facilitating the development of targeted strategies for sustainable development and carbon neutrality. The findings indicate that (1) economic development and carbon control can progress synergistically to some extent, as economically advanced cities like Beijing and Shanghai have achieved their carbon peaks earlier; (2) natural resource endowment significantly affects urban carbon emissions, with resource-rich cities such as Tangshan and Handan, where fossil fuels dominate the energy mix, facing considerable challenges in reducing emissions; and (3) notable differences exist in the growth patterns of carbon emissions between urban and rural buildings, underscoring the need for tailored carbon reduction policies.

1. Introduction

1.1. Background

Studying urban agglomerations as distinct entities is vital due to their unique roles as economic, social, and environmental hubs that significantly impact both regional and global carbon emissions. Characterized by high population densities, concentrated economic activities, and intensive energy consumption, urban agglomerations are major contributors to greenhouse gas emissions, particularly from the building sector. Unlike individual cities, these clusters often share interconnected infrastructure, energy networks, and economic systems, leading to complex emission dynamics that cannot be effectively addressed through isolated city-level analyses. Moreover, urban agglomerations present both challenges and opportunities for carbon mitigation. While they face considerable pressures from rapid urbanization and resource consumption, they also function as centers of innovation with more significant and more tremendous potential for adopting advanced technologies, promoting green policies, and achieving economies of scale in emission reduction strategies. By concentrating on urban agglomerations, research can better capture the interplay of regional socio-economic factors, resource allocation, and collaborative governance, ultimately providing more targeted and impactful solutions for sustainable urban development and carbon neutrality goals.

Globally, urban agglomerations are responsible for 50–60% of the world’s total carbon emissions. Mega-urban agglomerations, which consist of core and peripheral cities—such as the New York metropolitan area, the Pearl River Delta, the Yangtze River Delta, and the Tokyo metropolitan area—exhibit heightened levels of economic activity, building density, energy consumption, and population concentration, leading to an even greater share of carbon emissions. In response to the challenge of carbon emissions, countries worldwide are actively implementing various measures, with China notably proposing the “30.60” dual carbon target [1]. Within urban agglomerations, buildings serve as major sources of carbon emissions due to their high energy use and population density, making them crucial for conserving energy and reducing emissions efforts [2,3]. According to 2020 data, carbon dioxide emissions from operating buildings reached 2.16 billion tons, representing 21.70% of China’s total carbon emissions for that year (CABEE, 2022). Notably, carbon emissions from residential buildings comprise over 60% of total building emissions [4].

As global urbanization continues and living standards improve, the energy consumption and carbon emissions associated with urban agglomerations are anticipated to rise further [5,6]. In fact, carbon emissions from mega-urban agglomerations account for approximately 75% of total emissions, positioning them as critical areas for emission reduction initiatives [7,8]. The carbon emission challenges faced by urban agglomerations are expected to intensify. Thus, fostering the high-quality development of these regions and achieving a decoupling of economic growth from carbon emissions is essential for realizing global carbon peak and carbon neutrality objectives.

Currently, China’s three prominent city clusters—the Yangtze River Delta, the Pearl River Delta, and the Beijing–Tianjin–Hebei region—contribute nearly half of the nation’s GDP and play a crucial role in global economic growth and environmental governance [8,9]. As key players in the globalization process, the reduction in carbon efforts of these city clusters will significantly influence the green transformation of the world economy. Their approaches to carbon emission management can offer valuable insights and pathways for other rapidly developing city clusters globally, while also showcasing a low-carbon development model that may serve as a reference for other nations. Therefore, conducting in-depth research and promoting carbon emission peak assessment in the construction sectors of China’s major city clusters is not only essential for China to achieve its “dual carbon” goals but also holds critical importance for advancing global sustainable development [10,11].

1.2. Literature Review

The literature on assessed peak levels of carbon emissions has expanded significantly, typically categorized into historical peak assessments and future peak predictions. Generally, academic research tends to focus more on predicting the trajectory of future CO₂ emissions. Regarding historical peak assessments, the World Resources Institute (WRI) has established criteria for analyzing carbon emissions trends. This analysis stipulates that regional carbon emissions must peak within the last five years and exhibit a stable downward trend over a relatively extended period [12]. Additionally, emissions should decrease by more than 10% within five years following the peak. While some recovery post-peak is permissible, it cannot surpass the peak levels of carbon emissions [13].

To implement these assessments, researchers primarily utilize the Mann–Kendall (MK) trend test in statistics [14,15], along with decoupling models in economics [16]. For instance, Qian et al. and Huo et al. applied both the Mann–Kendall trend test and Tapio decoupling analysis to investigate the spatiotemporal relationship between CO₂ emissions and the effect of socio-economic development on emissions in China [4,17]. This assessment criterion has been referenced in numerous studies [18,19,20].

Numerous studies have been conducted to evaluate forecasts of peak operational carbon emissions within the building industry on global [21], national [22,23,24], and provincial [18] levels. Cui et al. [25] employed a federated learning algorithm based on seasonal autoregressive integrated moving averages to predict carbon emissions across thirteen countries and regions. Ma et al. [26] utilized a federated learning algorithm based on seasonal autoregressive integrated moving averages to forecast carbon emissions across thirteen countries and regions. Li et al. [27] developed a provincial-level model for building carbon emissions in China by integrating both top-down and bottom-up approaches. Additionally, Tang et al. [28] created a framework to project CO₂ emission trends and potential reductions in urban residential buildings in Jiangxi Province, extending to 2060, underscoring the significance of systematic perspectives in achieving urban carbon neutrality goals.

The CO₂ emission peak prediction model primarily analyzes and forecasts emissions from both a top-down macroeconomic perspective and a bottom-up micro-specific activity perspective. The top-down models encompass the STIRPAT model [29,30,31], Kaya identity [24,32], EKC model [33], Index Decomposition Analysis (IDA) models [34], and various econometric models [35,36], which are all instrumental in examining the socio-economic factors that contribute to carbon emissions. Researchers frequently integrate top-down models with conducted scenario analysis to forecast the peak of building carbon emissions. For instance, Huo et al. [37] investigated the effect of urbanization on urban building carbon emissions in China, emphasizing the importance of recognizing the intertwined effects of urbanization on carbon output. Similarly, Ke et al. [38] underscored the potential for energy conservation and emissions reduction in public buildings. Typical bottom-up models include the Long-range Energy Alternatives Planning (LEAP) model [39] and the MARKAL model [40]. Notably, the LEAP model has been employed to predict both the peak carbon emissions and the timing of that peak in China’s public buildings across various scenarios [41].

Through a comprehensive review of both the domestic and international literature, we found that in urban agglomerations undergoing rapid urbanization, the socio-economic system and government policy mechanisms not only influence spatial structures but also play a crucial role in affecting carbon emissions. This influence manifests across both temporal and spatial dimensions, revealing unique patterns. Specifically, we observe the following: (1) While previous studies have evaluated carbon emissions, a unified model at the urban agglomeration level that systematically assesses the emissions from the residential building sector at the city level is still lacking. (2) In the current research, although there has been investigation into the carbon peak state, the interplay between peak cities and economic development remains unclear—specifically, whether achieving this peak occurs actively or passively, and what underlying factors drive this process. (3) At the regional level, existing research has its limitations; it often focuses on individual cities or countries while neglecting the intermediate level of urban agglomerations. Moreover, comparative studies on the mechanisms influencing carbon emissions across different urban agglomerations remain insufficient, suggesting that this area warrants further exploration.

This study introduces an innovative model designed to address the issue effectively. It presents a comprehensive Building Carbon Peak Assessment Model (BCPCAM) that integrates the Mann–Kendall trend test, decoupling analysis, and a carbon peak assessment framework, aiming to overcome the limitations of existing methods.

Firstly, this model accurately identifies the peak processes of cities within three major urban agglomerations and determines whether each city’s carbon emissions are reaching their peak actively or passively. It also examines the variations in carbon emissions across different urban agglomerations and explores their underlying causes, thereby providing robust support for the comprehensive promotion of overall building carbon emissions peaking.

Secondly, in utilizing this comprehensive assessment model, we consider the distinctions between urban and rural areas, allowing for a thorough evaluation of the carbon emission peak conditions and causes in both rural and urban buildings. This establishes a solid foundation for predicting the future trajectory of carbon emissions in urban agglomeration buildings and for the effective allocation of carbon quotas.

Ultimately, this model offers decision-makers a detailed overview of carbon emissions at the urban agglomeration level while providing valuable new perspectives for developing more efficient regional carbon emission reduction strategies.

2. Methods

2.1. Research Framework

Historical trends in building carbon emissions across the three urban agglomerations present substantial challenges in the effort to achieve peak carbon emissions. Thus, evaluating these historical trends at the provincial level is essential for understanding the conditions surrounding peak carbon emissions. Analyzing the peak carbon emissions associated with buildings in these agglomerations will provide authorities with valuable insights into emissions at various stages, which is vital for developing effective carbon-neutral strategies for the residential construction sector. In light of this, this study proposes a comprehensive model to assess peak building carbon emissions in the three urban agglomerations, as detailed in Figure 1.

2.2. Assessment of Carbon Peak Situation

According to international standards, the peak of historical carbon emissions is defined as the point at which emissions in a region reach their highest level over a period of five years or more, compared to the most recent emission inventory. If this condition is not met, it cannot be considered a long-term trend. In this study, the historical evolution trends of BCE, BCEA, and BCEP in urban and rural areas are classified into four categories: peaked, plateau, fast growth, and slow growth [42]. The peaked category indicates a consistent downward trend following the peak, while the plateau category reflects only minor fluctuations after reaching the peak. The fast and slow growth categories denote stable growth rates, either rapid or gradual. Figure 2 illustrates the evaluation process of the peak situation. Below are the comprehensive steps involved in conducting the MK test and decoupling test:

Step 1: Begin by assessing the volume of time series data (denoted as N) for each city within the three major mega-city agglomerations following the peak year of carbon emissions (BCEs). If the data volume N < 5, this suggests that the emissions peak has not been sustained long enough for detailed analysis, signaling that the city has yet to reach its peak emissions.

Next, compute the average annual growth rate (AAGR) of BCE from the past five years for each city and juxtapose this with the national AAGR. Classifying cities with AAGR > national AAGR as “peak pressure provinces” (faster growth) and those with AAGR ≤ national AAGR as “peak momentum provinces” (slower growth) provides an initial categorization based on growth pressure.

Step 2: Applying the Mann–Kendall (MK) test for N > 5, For cities where N > 5, your approach to apply the Mann–Kendall test is appropriate. Using Z for trend direction and p for statistical significance at alpha = 0.05 is standard. Here is the breakdown of classifications based on MK test results:

-

Significant Downward Trend (Z < 0, p < 0.05). If emissions decline post-peak with Z < 0 and p < 0.05, indicating a statistically significant downward trend, you categorize cities based on the percentage decline.

-

Non-Significant Downward Trend (Z < 0, p ≥ 0.05). The trend is not statistically significant, so classifying the city as a “plateau” is reasonable since emissions have not shown a meaningful decline.

-

Non-Significant Upward Trend or Flat Trend (Z ≥ 0, p ≥ 0.05). A lack of a statistically significant upward trend implies stability. “Plateau” classification is appropriate as emissions are neither rising nor falling significantly.

-

Significant Upward Trend (Z ≥ 0, p < 0.05). A statistically significant upward trend post-peak is identified. Following guidance from earlier steps, further classification into “fast” or “slow” growth based on AAGR comparisons would then provide insight into the growth pattern [43].

Step 3: For those cities that have confirmed the peak of carbon emissions according to the MK trend test, the next stage involves expanding the decoupling model to explore the complex relationship between BCE and various social development indicators. If the decoupling index is below 0.8 in the short term and falls under 0 in the long term (three years or more), it signifies a strong decoupling relationship. This suggests that proactive urban planning and strategies have effectively contributed to reaching the emission peak. Conversely, failing to meet these thresholds implies that the city has experienced a more passive decline in BCE, reflecting a lack of decisive action or systemic changes.

2.2.1. The Mann–Kendall Trend Test Model

The Mann–Kendall (MK) trend test is a well-known statistical method used to analyze trends and identify abrupt change points in time series data, especially for climate variables like precipitation, runoff, carbon emissions, temperature variations, and water quality indicators [44,45]. The test holds significant advantages: it operates without the need for data to conform to a specific distribution and is adept at managing instances of missing data, which is a common issue in environmental studies [15,46]. Its reliability has led to recommendations for its use by esteemed organizations like the World Meteorological Organization.

Hypotheses, Null Hypothesis (H₀): There is no trend in the data (the data are independent and randomly ordered). Alternative Hypothesis (H₁): There is a monotonic trend (increasing or decreasing) in the data over time. The Mann–Kendall test calculates a statistic S. It is the cumulative sum of the signs of the differences between every pair of observations in the dataset:

$$S={\sum }_{i=1}^{n-1}{\sum }_{j=i+1}^n sign(x_j-x_i)$$

(1)

where

$$sign(x_j-x_i)=\left\{\begin{array}{c}1,\\ 0,\\ -1,\end{array}\right.\begin{array}{cc}& if x_j-x_i>0\\ & if x_j-x_i=0\\ & if x_j-x_i<0\end{array}$$

(2)

Interpretation of S, S > 0: Indicates an increasing trend. S < 0: Indicates a decreasing trend. S ≈ 0: Implies no trend. The test also calculates a variance Var (S), which is used to standardize S and obtain a Z-score. The Z-score helps assess the significance of the trend

$$\mathrm{Z}=\left\{\begin{array}{c}{\displaystyle \frac{\mathrm{S}-1}{\sqrt{\mathrm{V}\mathrm{a}\mathrm{r}\left(\mathrm{S}\right)}}},\\ 0, \\ {\displaystyle \frac{\mathrm{S}+1}{\sqrt{\mathrm{V}\mathrm{a}\mathrm{r}\left(\mathrm{S}\right)}}},\end{array}\right.\begin{array}{cc}\mathrm{i}\mathrm{f} \mathrm{S}>0& \\ \mathrm{i}\mathrm{f} \mathrm{S}=0& \\ \mathrm{i}\mathrm{f} \mathrm{S}<0& \end{array}$$

(3)

The Z-score follows a standard normal distribution, and its absolute value can be compared to critical values of the normal distribution (e.g., 1.96 for a 5% significance level) to test for significance. The two-tailed p-value associated with the Z-score indicates the probability of observing the calculated trend due to random chance. If the p-value is below a chosen significance level (e.g., 0.05), the null hypothesis of no trend is rejected, suggesting a statistically significant trend. This study takes the significance level p-value as 0.05, with its corresponding Z-score being 1.96.

2.2.2. Decoupling Analysis Model

The Tapio decoupling model has emerged as the preferred framework for assessing the interactive relationship between economic growth and environmental protection. Its high accuracy, minimal data requirements, and distinct advantages in representing the relationship between the economy and the environment make it particularly valuable. This model is versatile and can thoroughly analyze the decoupling status of resource utilization efficiency and economic development across various key sectors, including the construction industry [47,48], transportation [49], industrial structure [50], international trade [51], and ecological environment [52].

This study centers on three fundamental socio-economic factors: population size, residential building area, and per capita disposable income. Its objective is to investigate the decoupling relationship between these factors and building carbon emissions through quantitative analysis methods. The specific calculation formula is as follows:

$$\begin{array}{cc}{\phi }_{\mathrm{I}}& ={\displaystyle \frac{\Delta \mathrm{C}/{\mathrm{C}}^0}{\Delta \mathrm{I}/{\mathrm{I}}^0}}={\displaystyle \frac{({\mathrm{C}}^{\mathrm{T}}-{\mathrm{C}}^0)/{\mathrm{C}}^0}{({\mathrm{I}}^{\mathrm{T}}-{\mathrm{I}}^0)/{\mathrm{I}}^0}}\\ {\phi }_{\mathrm{P}}& ={\displaystyle \frac{\Delta \mathrm{C}/{\mathrm{C}}^0}{\Delta \mathrm{P}/{\mathrm{P}}^0}}={\displaystyle \frac{({\mathrm{C}}^{\mathrm{T}}-{\mathrm{C}}^0)/{\mathrm{C}}^0}{({\mathrm{P}}^{\mathrm{T}}-{\mathrm{P}}^0)/{\mathrm{P}}^0}}\\ {\phi }_{\mathrm{P}\mathrm{F}}& ={\displaystyle \frac{\Delta \mathrm{C}/{\mathrm{C}}^0}{\Delta \mathrm{P}\mathrm{F}/\mathrm{P}{\mathrm{F}}^0}}={\displaystyle \frac{({\mathrm{C}}^{\mathrm{T}}-{\mathrm{C}}^0)/{\mathrm{C}}^0}{(\mathrm{P}{\mathrm{F}}^{\mathrm{T}}-\mathrm{P}{\mathrm{F}}^0)/\mathrm{P}{\mathrm{F}}^0}}\end{array}$$

(4)

The variables φI, φP, and φPF represent the decoupling elasticity of carbon emissions from residential buildings in relation to per capita residential disposable income, population, and per capita residential floor space, correspondingly. The indicators ΔC, ΔI, ΔP, and ΔPF denote the changes in Residential Building Carbon emissions alongside per capita residential disposable income, population, and per capita residential floor space over the period T. Figure 3 offers a general categorization of decoupling states based on the Tapio decoupling model, while

Table 1. Regional division.

Regional Division	Cities
Yangtze River Delta urban agglomerations (YRD)	Shanghai, Nanjing, Changzhou, Nantong, Yancheng, Yangzhou, Zhenjiang, Wuxi, Hangzhou, Ningbo, Jiaxing, Shaoxing, Jinhua, Zhoushan, Anqing, Taizhou, Huzhou, Taizhou, Chuzhou, Ma’anshan, Wuhu, Xuancheng, Tongling, Chizhou, Suzhou, Hefei.
Pearl River Delta urban agglomerations (PRD)	Guangzhou, Dongguan, Zhuhai, Foshan, Zhaoqing, Jiangmen, Huizhou, Shenzhen, Zhongshan.
Beijing–Tianjin–Hebei urban agglomerations (BTH)	Beijing, Tianjin, Shijiazhuang, Zhangjiakou, Chengde, Qinhuangdao, Tangshan, Langfang, Baoding, Cangzhou, Hengshui, Xingtai, Handan.

presents a detailed summary of the explanatory and dependent variables.

3. Data Sources

This study encompasses 48 cities across three major urban clusters in China: the Yangtze River Delta, the Pearl River Delta, and the Beijing–Tianjin–Hebei region. It employs the China Building Energy Consumption Measurement Method (CBECM), established in prior research, to assess the energy consumption of both urban and rural residential buildings from 2005 to 2020. Data on building area and energy consumption are sourced from the Urban and Rural Construction Energy Consumption and Carbon Emission Database (CBEED), which has been extensively utilized in previous studies. Additional economic and population data are obtained from the China Statistical Yearbook. The locations of the carbon emissions situation in urban agglomerations and their case cities are shown in Figure 4.

4. Results and Discussion

4.1. The Historical Development of Carbon Emissions from Residential Buildings in Three Mega-City Agglomerations

This study, following a comprehensive analysis of the CBEED data, presents the trends in carbon emissions from urban residential buildings (UBCs), per capita carbon emissions (UBCPs), carbon emissions per unit of building area (UBCF), as well as carbon emissions from rural residential buildings (RBCs), per capita carbon emissions (RBCPs), and carbon emissions per unit of building area (RBCF) across 48 cities from 2005 to 2020. These trends are illustrated in Figure 5a,b.

The evolution of carbon emissions from residential buildings in urban areas between 2005 and 2020 is depicted in Figure 5a. Despite some fluctuations, the agglomeration, with an increase of 1.91 times, and then the Beijing–Tianjin–Hebei (BTH) urban agglomeration, has risen by 1.64 times. These trends in UBC reflect China’s economic growth, rapid urbanization, and improving living standards. The overall trend in urban building carbon (UBC) emissions for the three urban agglomerations shows a consistent upward trajectory. Notably, the Yangtze River Delta (YRD) urban agglomeration has experienced significant growth, increasing by 2.55 times over the past 16 years. This is followed by the Pearl River Delta (PRD) urban important to highlight that the UBC in the Beijing–Tianjin–Hebei urban agglomeration is substantially higher than that of the other two regions, climbing from 72.27 Mt. CO₂ in 2005 to 118.28 Mt. CO₂ in 2020. This increase is closely linked to the region’s geographical conditions, as the northern location necessitates winter heating.

The UBCP and UBCF of the BTH urban agglomeration are significantly higher than those of the other two regions. Since 2008, the UBCF has shown a steady decline, while the UBCP has fluctuated downwards after peaking at 1.53 t CO₂·person⁻¹ in 2011. The PRD urban agglomeration follows suit, with both UBCP and UBCF exhibiting a slight downward trend marked by volatility after their peak in 2011. In contrast, the YRD urban agglomeration has experienced a general upward trend, reaching its peak in 2018. This phenomenon is closely linked to the city’s growing permanent population, increased governmental emphasis on ecological civilization, and enhanced awareness of energy conservation among residents.

The variations in UBCF trends across urban agglomerations point to disparities in energy efficiency improvements. For instance, the steady decline in UBCF in regions like the BTH urban agglomeration reflects the adoption of advanced heating systems and stricter building energy codes. In contrast, regions like YRD still exhibit rising trends, indicating the need for targeted policy interventions that address local energy challenges.

Figure 5b illustrates the historical trends of Residential Building Carbon (RBC), Residential Building Carbon Per Capita (RBCP), and Residential Building Carbon Footprint (RBCF) from 2005 to 2022. With the exception of RBCP and RBCF in the Pearl River Delta (PRD) region, the overall trend indicates an upward trajectory. Among the three major urban agglomerations, total carbon emissions from rural buildings are lower than those in urban areas. Notably, the RBC in the Beijing–Tianjin–Hebei (BTH) urban agglomeration is the highest, rising from 28.97 Mt CO₂ in 2005 to 59.98 Mt CO₂ in 2020. Following this, the Yangtze River Delta (YRD) urban agglomeration saw its emissions reach 51.3 Mt CO₂ in 2020, marking an increase of 58.08%. The Pearl River Delta (PRD) urban agglomeration shows a stable upward trend. It is worth noting that the RBCP and RBCF of the PRD urban agglomeration are higher than those of the other two urban agglomerations, and have gradually declined after reaching a peak in 2011. This trend is closely tied to the effects of urbanization. Over the past 16 years, the rural population in the Pearl River Delta urban agglomeration has experienced positive growth, increasing by 658,000 individuals, while both the BTH and YRD urban agglomerations have seen declines—specifically, the BTH urban agglomeration’s population decreased by 13.22 million and the YRD’s by 14.23 million. Concurrently, the per capita living area in rural regions has steadily increased, contributing to the ongoing growth in RBCP and RBCF.

4.2. Historical Trends in the Evolution of Carbon Emissions from Residential Buildings in Three Mega-City Agglomeration

This study further examines the changes in carbon emissions from the residential building sector at the urban level in the three major urban agglomerations, as illustrated in Figure 6.

Among the 48 cities, UBC generally exhibits an upward trend. Notably, cities such as Tianjin, Shijiazhuang, Shanghai, Suzhou, Ningbo, Guangzhou, and Shenzhen demonstrate particularly robust growth momentum. Tianjin and Shanghai have experienced remarkable growth rates exceeding 60% from 2005 to 2020, specifically 69.86% for Tianjin and 61.34% for Shanghai. Conversely, cities like Tongling, Chizhou, Ma’anshan, Anqing, and Xuancheng have shown relatively slow growth, with rates of less than 3% during the same period. This disparity is closely linked to the dynamics of urbanization, rapid social and economic development, the continual expansion of urban residential space, and the influence of climatic conditions on the demand for heating energy.

Carbon emissions trends exhibit significant variation across different cities. Prior to 2015, only a handful of cities, including Beijing, Shijiazhuang, Foshan, Zhongshan, and Ma’anshan, reached the Urban Building Carbon (UBC) highest levels. Interestingly, half of the cities peaked in UBC emissions in 2020, a phenomenon closely tied to the home quarantine measures enacted by local governments during the COVID-19 pandemic, as well as the widespread adoption of remote work by companies [53]. The widespread adoption of remote work and reduced industrial activities led to lower energy demand. However, this temporary decline does not reflect structural changes, necessitating long-term strategies for sustainable emission reductions.

In 2020, the top three cities in the UBC emissions ranking were Beijing (31.71 Mt. CO₂), Tianjin (25.59 Mt. CO₂), and Shanghai (19.52 Mt. CO₂). All three are Chinese municipalities characterized by high levels of economic development, dense populations, and energy structures predominantly reliant on fossil fuels. Additionally, the cities with the highest average annual growth rates over the study period were Zhoushan (10.93%), Yancheng (9.61%), and Shaoxing (9.2%), mainly due to the thriving construction industries in those areas.

In comparison to UBC, the trends for UBCP and UBCF differ significantly. UBCP generally displays a fluctuating upward trend across most cities, whereas UBCF tends to show a fluctuating downward or a consistent downward trend. According to the UBCP data, 14 cities peaked before 2015, with Xingtai, Tongling, and Dongguan recording their highest UBCP values in 2005. It is worth noting that a significant portion of the increased UBCP in densely populated cities such as Shanghai and Beijing can be attributed to the urban heat island effect, which drives higher energy demand for cooling. This highlights the importance of incorporating green infrastructure and reflective building materials in urban planning to mitigate heat absorption.

Similarly, the UBCF data indicate that 20 cities peaked before 2015, including 14 cities such as Tianjin, Tangshan, and Baoding, which also reached their highest UBCP values in 2005. When comparing levels from 2005, Zhaoqing, Zhoushan, and Yancheng exhibited the most significant increases in UBCP and UBCF. By 2020, UBCP had risen by 689%, 291%, and 226%, respectively, while UBCF had increased by 315%, 178%, and 107%, respectively. In contrast, Ma’anshan experienced a notable decline, with its UBCP and UBCF decreasing by 29% and 65%, respectively, in 2020. The growth in Zhaoqing, Zhoushan, and Yancheng can be credited to population growth and economic development. Conversely, Ma’anshan’s sharp decline resulted from the growth rate of carbon emissions lagging behind the increases in urban population and building area. The evolution trends of RBC, RBCP, and RBCF between 2005 and 2020 are illustrated in Figure 7.

Among the three major urban agglomerations, the Resource-Based Consumption (RBC) of the Beijing–Tianjin–Hebei (BTH) urban agglomeration is notably high, with Beijing, Baoding, Tangshan, and Shijiazhuang ranking in the top four. This trend is likely linked to the region’s climate characteristics, architectural styles, and energy infrastructure. The architectural layout of rural buildings in northern regions, characterized by single or double-story structures, and building layouts tend to be more dispersed, significantly contributing to higher energy consumption due to inefficient heating. While some progress has been made in adopting clean energy, a significant portion of the region still relies on coal as its primary energy source. Modernizing these layouts with centralized heating systems or renewable energy solutions could substantially reduce RBC emissions.

From 2005 to 2016, cities other than Beijing and Shanghai experienced an average RBC growth rate of 5%, surpassing the national average of 2.58%. Notably, 19 cities, including Guangzhou, Langfang, Ningbo, Hangzhou, Tianjin, Dongguan, Nanjing, Shaoxing, and Zhuhai, demonstrated a consistent growth trend. This increase is closely tied to the development of local urban economies and the rising foreign population. Particularly significant is the trend in Shanghai and Beijing, where RBC peaked in 2012 and declined at an average annual rate of 6%. This shift is closely associated with local government policies, optimizing energy structures, and changing lifestyles. For instance, Shanghai has introduced a “rural net zero carbon path” and is promoting energy-saving renovations of buildings, while Beijing has pursued a “coal-free” strategy through the promotion of clean energy sources.

With the exception of the four cities—Shanghai, Beijing, Guangzhou, and Tongling—RBCP and RBCF displayed a fluctuating growth trend in other urban areas. Most cities reached their peak population growth before 2015, indicating that the influence of the three urban agglomerations on RBC growth is gradually diminishing. This is particularly evident in Foshan, Zhongshan, Zhuhai, and Dongguan within the Pearl River Delta urban agglomeration, where fluctuations are especially pronounced. These trends are closely tied to the rapid development of the local economy, accelerated urbanization, frequent construction activities in rural regions, and shifts in energy structure. The growth trend of RBCF underscores the improvement in living standards for rural residents. In contrast, the RBC fluctuations in Shanghai, Beijing, Guangzhou, and Tongling have shown a downward trajectory, with RBCP in Guangzhou and Shanghai peaking in 2005. Notably, data from the “Shenzhen Seventh National Census Bulletin” indicates that Shenzhen, a major city within a special economic zone, has achieved complete, urbanization, with its urban population now at 100%, signifying the absence of a traditional rural population.

4.3. Peak Carbon Emissions from Residential Buildings in the Three Major Urban Agglomerations

Utilizing the assessment method outlined in Section 2.2, Figure 8 provides a detailed overview of carbon emissions from buildings across the three major urban agglomerations. Upon comparing and analyzing the peak characteristics of carbon emissions in these regions, it became evident that, except for seven specific cities, the Urban Building Carbon (UBC) levels in all other urban agglomerations exhibited a rapid growth trend. Notably, both the Urban Building Carbon Peak (UBCP) and the Urban Peak Carbon Footprint (UPCF) of many cities reached their historical highs before 2015.

According to UBC’s carbon emissions assessment, Beijing is the only city to have achieved its peak emissions thus far. Other cities, such as Foshan, Zhongshan, Ma’anshan, and Shijiazhuang, display a plateau and a declining trend, indicating significant potential for them to quickly reach their peak carbon emissions. Conversely, Zhaoqing, Handan, Zhoushan, Jiangmen, Huizhou, Nantong, Anqing, Taizhou, Baoding, Wuhu, Chizhou, Tangshan, Hangzhou, Changzhou, Nanjing, and Zhejiang—41 cities in total—are experiencing rapid growth in carbon emissions. This swift increase places considerable pressure on the construction industry’s efforts to reduce emissions. To meet the carbon peak targets for these cities, a more significant investment of resources is necessary. Additionally, Shanghai and Shenzhen are currently experiencing a slowdown in growth but still need to intensify their efforts to achieve their peak targets.

In Beijing, a deeper decoupling analysis was performed to explore the relationship between urban population (P_u), urban per capita disposable income (I_u), and urban residential building floor space (PF_u), in conjunction with overall changes in UBC. The analysis showed that although Beijing’s population, per capita disposable income, and per capita residential building area are all increasing, Urban Building Carbon emissions have peaked and begun to decline. This trend suggests that Beijing may have implemented adequate measures to lower carbon emissions. These measures could include enhancing energy efficiency, adopting renewable energy sources, and promoting the development of green buildings.

There are notable distinctions between UBCP and UBC peak evaluations. Specifically, the number of cities in the rapid growth phase of UBCP has declined from 41 to 32, with no cities identified as experiencing slow growth. It is essential to highlight that eight cities—Tangshan, Zhuhai, Langfang, Ningbo, Zhongshan, Wuhu, Ma’anshan, Zhangjiakou, and Dongguan—reached the peak of UBCP before 2015, subsequently entering a plateau stage. As UBCP peaked, seven cities—Jinhua, Beijing, Shenzhen, Shijiazhuang, Foshan, Xingtai, and Tongling—exhibited a clear downward trend, indicating their transition into the peak stage.

Further decoupling research was conducted on cities where UBCP peaked, and the results are detailed in

Table 2. Decoupling assessment of UBCP and socioeconomic development metrics.

City	ΔUBCP/UBCP	ΔI/I	ΔP/P	ΔPF/PF	φI	φP	φPF	Stage
Beijing	−0.40	1.35	0.14	0.19	−3.38	−0.34	−0.48	Actively peaked
Shijiazhuang	−0.45	1.86	0.75	0.32	−0.24	−0.61	−1.43	Actively peaked
Xingtai	−0.18	3.27	0.71	0.41	−0.06	−0.25	−0.44	Actively peaked
Jinhua	−0.25	0.32	0.33	0.04	−0.78	−0.76	−7.12	Actively peaked
Tongling	−0.28	3.32	0.47	0.66	−0.08	−0.58	−0.42	Actively peaked
Shenzhen	−0.36	1.42	0.76	0.00	−0.25	−0.47	−78.26	Actively peaked
Foshan	−0.39	1.72	0.58	0.15	−0.22	−0.67	−2.56	Actively peaked

. The findings reveal that all these cities are in the “Actively peaked” stage, meaning that carbon emissions have reached their peak and are now on the decline. These data illustrate that these cities have reduced carbon emissions while continuing to advance socio-economically, achieving a decoupling of economic growth from carbon emissions.

In the UBCF sector, a total of 16 cities, including Jiangmen, Nantong, Zhaoqing, Nanjing, Zhongshan, and Shaoxing, are currently experiencing significant rapid growth stages. Meanwhile, 10 cities, such as Guangzhou, Shanghai, Ningbo, Hengshui, and Zhuhai, have entered a plateau phase. Additionally, 22 cities, including Jinhua, Qinhuangdao, Shenzhen, Beijing, and Tianjin, reached their highest UBCF levels in history before 2015. Further decoupling research was conducted on cities where UBCF had peaked. Notably, all these cities successfully achieved peaks in UBCF and realized a decoupling of economic growth from carbon emissions. The results can be found in

Table 3. Decoupling assessment of UBCF and socioeconomic development indicators.

City	ΔUBCF/UBCF	ΔI/I	ΔP/P	ΔPF/PF	φI	φP	φPF	Stage
Beijing	−0.51	1.58	0.21	0.22	−0.32	−2.42	−2.32	Actively peaked
Tianjin	−0.13	1.65	0.29	0.44	−0.08	−0.45	−0.30	Actively peaked
Shijiazhuang	−0.59	1.67	0.75	0.32	−0.35	−0.79	−1.86	Actively peaked
Zhangjiakou	−0.43	1.43	0.39	0.39	−0.30	−1.10	−1.11	Actively peaked
Chengde	−0.28	1.55	0.43	0.46	−0.18	−0.65	−0.61	Actively peaked
Qinhuangdao	−0.24	1.06	0.40	0.48	−0.23	−0.62	−0.51	Actively peaked
Tangshan	−0.27	3.23	0.48	0.37	−0.08	−0.57	−0.75	Actively peaked
Langfang	−0.28	1.84	0.87	0.31	−0.15	−0.32	−0.92	Actively peaked
Baoding	−0.50	2.71	1.02	0.60	−0.19	−0.49	−0.84	Actively peaked
Cangzhou	−0.14	1.61	0.37	0.22	−0.09	−0.39	−0.64	Actively peaked
Xingtai	−0.52	3.27	0.67	0.41	−0.16	−0.78	−1.26	Actively peaked
Handan	−0.25	2.84	0.68	0.45	−0.09	−0.36	−0.55	Actively peaked
Jinhua	−0.27	0.32	0.33	0.04	−0.83	−0.82	−7.64	Actively peaked
Hefei	−0.52	3.99	1.99	0.53	−0.13	−0.26	−0.98	Actively peaked
Ma’anshan	−0.62	3.34	1.12	0.64	−0.20	−0.58	−1.02	Actively peaked
Wuhu	−0.46	3.63	1.47	0.47	−0.13	−0.31	−0.97	Actively peaked
Tongling	−0.70	3.32	0.47	0.66	−0.21	−1.47	−1.05	Actively peaked
Chizhou	−0.51	3.53	0.43	0.53	−0.14	−1.19	−0.95	Actively peaked
Shenzhen	−0.40	1.42	0.76	0.00	−0.28	−0.52	−85.36	Actively peaked
Foshan	−0.47	1.98	0.58	0.15	−0.24	−0.81	−3.11	Actively peaked
Dongguan	−0.44	1.79	1.02	0.12	−0.25	−0.43	−3.79	Actively peaked

.

The decoupling analysis demonstrates that cities like Beijing have successfully decoupled carbon emissions from economic growth, achieving “actively peaked” status. This was facilitated by a combination of policy-driven renewable energy adoption, public transportation expansion, and smart city initiatives. Other cities, particularly in the PRD region, should consider replicating these strategies to accelerate their decoupling processes.

Figure 9 illustrates the distribution of peak assessment results for UBC, UBCP, and UBCF across various geographical locations. This map highlights the effectiveness of China’s efforts in controlling and reducing carbon emissions in different regions. Most UBC cities within the three major urban agglomerations are currently in a rapid growth phase. The Beijing–Tianjin–Hebei (BTH) urban agglomeration has already reached the UBCF peak, with its UBCP primarily in a plateau or peak stage. The decoupling of UBC, UBCP, and UBCF among peaking cities in these urban agglomerations has also reached their maximum levels. The three indicators reach their peaks in the following sequence: carbon emissions per floor space peak first, followed by carbon emissions per capita, and finally the total carbon emissions. This conclusion may further support the findings of Li et al. [42] and Huo et al. [43].

4.4. Carbon Emission Peaks from Rural Residential Buildings in the Three Major Urban Agglomerations

Within the 48 cities encompassed by the three major urban agglomerations, there is a notable disparity in carbon emission peaks between the rural and urban residential construction sectors. Specifically, Figure 10 illustrates the detailed assessment results for the carbon emission peaks associated with rural residential buildings.

From 2005 to 2020, the number of cities with rural residential buildings (RBCs) exhibited a notable growth trend compared to urban buildings (UBCs), experiencing only a brief decline lasting less than five years, ultimately stabilizing at 31 cities. Among these, five cities—Tianjin, Suzhou, Qinhuangdao, Chengde, and Zhangjiakou—have entered a phase of slow growth. Conversely, Wuxi, Tongling, Chizhou, Anqing, and Zhongshan have reached a plateau, as their growth rates have significantly slowed. Notably, five cities—including Hengshui, Handan, Beijing, Shanghai, and Tangshan—saw a pronounced decline after reaching their peak carbon emissions. However, it is essential to highlight that the reasons behind these peaks may vary.

We conducted a detailed decoupling analysis to explore the relationship between the rural population (P_u), per capita disposable income of rural residents (I_u), the building area of rural residences (PF_u), and variations in total rural building carbon emissions. For specific results, please refer to

Table 4. Decoupling assessment of RBC and socioeconomic development metrics.

City	ΔRBC/RBC	ΔI/I	ΔP/P	ΔPF/PF	φI	φP	φPF	Stage
Beijing	−0.54	1.19	−0.02	0.17	−0.46	25.28	−3.18	Passive decline
Tangshan	−0.12	0.80	−0.21	0.15	−0.15	0.55	−0.79	Passive decline
Hengshui	−0.22	0.50	−0.16	0.09	−0.45	1.37	−2.39	Passive decline
Handan	−0.16	0.81	−0.19	0.14	−0.19	0.85	−1.13	Passive decline
Shanghai	−0.62	2.79	0.20	0.24	−0.22	−3.14	−2.58	Actively peaked
Ma’anshan	−0.20	1.12	−0.27	0.22	−0.18	0.75	−0.92	Passive decline
Xvancheng	−0.21	0.41	−0.21	0.11	−0.51	1.00	−1.90	Passive decline
Chizhou	−0.32	0.40	−0.22	0.09	−0.81	1.45	−3.55	Passive decline

. Observations indicate that, except for Shanghai, carbon emissions in other cities showed a natural downward trend. In contrast, Shanghai demonstrated increased carbon emissions despite active policy regulations, while its RBC and three socio-economic development indicators maintained a robust decoupling state.

In cities such as Beijing, Tangshan, Hengshui, Handan, Ma’anshan, Xuancheng, and Chizhou, there exists a significant negative decoupling relationship between rural building carbon (RBC) emissions and the rural population. This trend is primarily a result of the passive decline in RBC in these areas, which has been exacerbated by a substantial population loss from its peak year to 2020. Notably, while previous research has indicated that Beijing’s total building carbon emissions have peaked [45], there remains a lack of in-depth exploration into the specific factors contributing to this phenomenon. To address this gap, this study introduces the Building Carbon Peak Comprehensive Assessment Model (BCPCAM), which aims to thoroughly analyze the intricate relationship between the decline in rural population and the associated decrease in Beijing’s RBC. The goal is to inform future policy formulation and urban development strategies effectively.

The peak evaluation results of the RBCP across the three urban agglomerations are clearly illustrated in Figure 10b. Over the past five years, the average annual growth rate of the RBCP has been 5.81%, with 30 cities surpassing the national average of 2.58%. In contrast, 12 cities experienced slightly lower development speeds, indicating a trend of decelerating growth. Notably, 36 cities are currently in a phase of rapid development, while five cities—Tongling, Chizhou, Zhongshan, Foshan, and Guangzhou—have not exhibited significant signs of decline after reaching their peaks and have entered a plateau phase.

Through the MK time series test, we identified a marked downward trend for both Shanghai and Beijing, indicating they have reached a peak stage. Furthermore, the decoupling results presented in

Table 5. Presents the decoupling analysis of RBCP and socio-economic development.

City	ΔRBCP/RBCP	ΔI/I	ΔP/P	ΔPF/PF	φI	φP	φPF	Stage
Beijing	−0.53	1.19	−0.02	0.17	−0.45	26.5	−3.12	Passive decline
Shanghai	−0.69	3.18	0.29	0.27	−0.22	−2.37	−2.55	Actively peaked

reveal a substantial negative decoupling relationship between the RBCP of Shanghai and Beijing and three key economic development indicators, which exhibit a consistent downward trend. In the case of Beijing, this decline can primarily be attributed to a reduction in the rural population, which has contributed to the RBCP reaching its peak level.

The results of the RBCF peak assessment are illustrated in Figure 10c. The study reveals that 28 cities are experiencing rapid growth, while another eight cities, including Anqing, Tianjin, Handan, and Langfang, have entered a phase of relatively slow growth. Before 2015, the RBCF for 12 cities had reached its maximum value, indicating that they were at a plateau or peak stage. Additional MK trend test findings suggest that eight cities, such as Guangzhou, Chizhou, Tongling, Qinhuangdao, and Tangshan, have transitioned into the plateau period. Notably, Foshan, Zhongshan, Beijing, and Shanghai have entered the peak period of RBCF.

Further analysis on decoupling shows a substantial and significant decoupling effect between the RBCF of Shanghai and Zhongshan and three key economic development indicators, facilitating their active peaking. In contrast, Beijing and Foshan exhibit a passive downward trend (

Table 6. Decoupling assessment of RBCF and socioeconomic development factors.

City	ΔRBCF/RBCF	ΔI/I	ΔP/P	ΔPF/PF	φI	φP	φPF	Stage
Beijing	−0.60	1.19	−0.02	0.33	−0.50	30.00	−1.82	Passive decline
Shanghai	−0.75	3.18	0.29	0.26	−0.24	−2.57	−2.88	Actively peaked
Foshan	−0.25	1.41	0.24	0.11	−0.18	−0.18	−2.27	Actively peaked
Zhongshan	−0.51	1.19	0.53	0.17	−0.43	−0.43	−3.00	Actively peaked

). This study establishes that the indicators of RBC, RBCP, and RBCF in Shanghai and Beijing have reached their peak status. These findings effectively validate the comprehensiveness and depth of the BCPCAM model employed in this research.

Analyze the spatial distribution of peak carbon emissions from rural buildings across the three urban agglomerations (refer to Figure 10). The analysis reveals that both the Rural Building Carbon Peak (RBCP) and Rural Building Carbon Footprint (RBCF) in these agglomerations are generally experiencing rapid growth. Notably, most cities within the Beijing–Tianjin–Hebei (BTH) urban agglomeration have either reached their RBC peak or are in a plateau phase. In economically advanced cities like Beijing and Shanghai, RBC, RBCP, and RBCF have all attained peak levels, although the factors driving these peaks differ. Following its peak, Beijing saw a downward trend in RBC, RBCP, and RBCF, while Shanghai experienced an active peaking in both RBC and RBCF.

This observation aligns with prior research findings indicating that carbon emission peaks can manifest in regions with varying levels of economic development and energy structures [42]. The study further highlights that due to differences in social and economic progress, the peak carbon emission scenarios in various urban agglomerations and cities exhibit significant diversity. Urban agglomerations possess unique economic, social, climate, and environmental advantages. Collaborative efforts in carbon reduction can foster a joint approach to meeting emission reduction targets, ultimately leading to a comprehensive peak. This insight provides a valuable reference for other sectors and economies as they evaluate their peak carbon emission strategies. Figure 11 shows the spatial distribution of carbon peak in urban residential buildings (RBC; RBCP; RBCF).

5. Conclusions

This study provides a comprehensive assessment of carbon emission peaks in urban and rural residential buildings across three major urban agglomerations in China: the Yangtze River Delta, the Pearl River Delta, and the Beijing–Tianjin–Hebei region from 2005 to 2020. This study innovatively proposed a comprehensive assessment model for building carbon peak (BCPCAM) in identifying emission trends, quantifying peak characteristics, and evaluating decoupling stages. The key conclusions of this study are as follows:

(1)

Key findings emphasize substantial variations in emission trends across cities and urban agglomerations. These trends reflect differences in socio-economic development, energy structures, and policy effectiveness. Overall, 41 cities are experiencing rapid increases in carbon emissions, Beijing emerged as the only city that had reached its carbon emission peak and was actively pursuing emission reductions. Beijing has actively peaked its emissions through effective policies and structural adjustments, while resource-intensive regions like Tangshan face challenges due to reliance on fossil fuels. By studying these patterns, the BCPCAM model has demonstrated its ability to guide evidence-based interventions and provide valuable insights for emission management.

(2)

Regarding Urban Building Carbon Peaks (UBCPs), 32 cities are witnessing significant growth, while eight cities, including Tangshan, Zhuhai, Langfang, and Ningbo, are in a plateau stage. Seven cities—Jinhua, Beijing, Shenzhen, Shijiazhuang, Foshan, Xingtai, and Tongling—have entered the “actively peaked” stage, meaning their carbon emissions have peaked and are starting to decline. Regarding the Urban Building Carbon Footprint (UBCF), ten cities—including Guangzhou, Shanghai, Ningbo, Hengshui, and Zhuhai are currently in a plateau phase. In contrast, 22 cities, such as Jinhua, Qinhuangdao, Shenzhen, Beijing, and Tianjin, are experiencing an active peak stage. Most cities within the Beijing–Tianjin–Hebei (BTH) urban agglomeration have either reached their Residential Building Carbon (RBC) peak or have entered a plateau phase. An assessment of the carbon emission peaks in rural and urban building sectors reveals significant differences. The growth rates of carbon emissions in Wuxi, Tongling, Chizhou, Anqing, and Zhongshan have notably slowed and are now in a plateau state. Except for Shanghai, which has actively reached its carbon emission peak, Hengshui, Handan, Beijing, and Tangshan are in a passive peaking mode. Among the 36 cities studied, the Residential Building Carbon Peak (RBCP) is still in a phase of rapid growth. Tongling, Chizhou, Zhongshan, Foshan, and Guangzhou are also experiencing plateau states. Furthermore, the RBC, RBCP, and RBCF of Beijing and Shanghai have all reached their peak levels. Notably, Beijing demonstrates a passive downward trend after hitting its peak, whereas Shanghai has successfully achieved active peaking in both RBC and RBCF.

(3)

The research conclusions underscore the scientific value of this research, mainly its focus on statistical modeling and pattern identification. This approach ensures that the findings are data-driven and widely applicable. The BCPCAM framework’s adaptability makes it a valuable tool for analyzing emission trends in other high-emission sectors, such as transportation, industrial manufacturing, and agriculture. Focusing on emission patterns and their implications, this study provides policymakers with actionable, evidence-based recommendations for achieving carbon neutrality and sustainable urban development.

6. Policy Recommendations

Using tailored strategies based on urban agglomeration characteristics, each urban agglomeration has unique socio-economic and energy-related challenges. For instance, regions like Beijing–Tianjin–Hebei, characterized by higher heating demand, should focus on improving building insulation and adopting clean heating technologies. In contrast, the Pearl River Delta should emphasize retrofitting urban buildings with energy-efficient systems to address its higher urban density and cooling demands.

At the city cluster level, establishing a regional coordination mechanism is essential for optimizing resource allocation and facilitating technology sharing. Strengthening internal collaboration among city clusters will enhance resource efficiency and promote the exchange of technologies. For instance, the Beijing–Tianjin–Hebei, Yangtze River Delta, and Pearl River Delta city clusters could set up a regional carbon emission management platform or a data-sharing platform to collaborate on monitoring, resource allocation, and carbon management emissions.

Introduce dynamic monitoring and evaluation mechanisms for building sector carbon emissions and mitigation measures to adjust and optimize policy implementation. For example, real-time monitoring of building energy consumption and carbon emissions using big data and smart technologies can provide targeted recommendations based on data analysis.

Encourage a societal shift toward low-carbon lifestyles by increasing public participation in the planning, design, and operation of low-carbon buildings. Educational campaigns and demonstration projects should be used to raise awareness of the significance of low-carbon buildings for sustainable development.