Carbon Emission Prediction and the Reduction Pathway in Huairou District (China): A Scenario Analysis Based on the LEAP Model

Abstract

With increasingly severe global climate change, reducing carbon emissions has become an important way to promote sustainable development. However, few scholars have researched carbon emissions and carbon reduction in the Huairou district, Beijing, China. Based on the Long-range Energy Alternatives Planning System (LEAP) model, this study sets four scenarios, including a baseline scenario (BAS), an industrial structure upgrading scenario (Indus), a technological progress scenario (Tech), and a comprehensive transformation scenario (COM), to simulate the long-term annual carbon emissions of Huairou district from 2021 to 2060. The results indicate that all four scenarios could realize the target of carbon peaking by 2030. Among them, the peak carbon emissions under the Indus scenario are the highest (2608.26 kilotons), while the peak under the COM scenario is the lowest (2126.58 kilotons). Moreover, by distinguishing the carbon emissions of sectors, it can be found that the commercial sector is the largest source of carbon emissions. The proportion of carbon emissions from the industrial sector will decline, while that from the urban household sector will increase. Furthermore, the analysis of the carbon emission reduction potential of sectors reveals that the commercial and industrial sectors have the greatest potential for carbon emission reduction in the medium term. However, the focus of carbon emission reduction needs to shift towards the commercial and urban household sectors in the long term. This study could provide references for formulating carbon emission reduction pathways and realizing sustainable development.

1. Introduction

Climate change has become one of the most severe challenges nowadays. The greenhouse gas emissions caused by human activities have led to unprecedented changes in the climate system, posing a threat to global ecosystem stability and the sustainable development of human society [1]. To address this crisis, countries worldwide are taking measures to control carbon emissions. As the world’s largest developing country and carbon emitter, China announced in 2020 its commitment to “peak carbon emissions before 2030 and strive to achieve carbon neutrality before 2060”, which is of great significance for promoting global climate governance.

As centers of population and economic activity, cities generate over 70% of global energy-related carbon emissions [2]. In the rapid urbanization and industrialization process in China, cities are the main drivers of carbon emissions. Therefore, whether a city can successfully achieve low-carbon transformation is directly related to the success or failure of the country’s carbon reduction goal. As the capital and a super-large city of China, Beijing has strongly demonstrated a leading role in low-carbon development. The carbon emissions and low-carbon development path of Beijing have attracted the attention of many scholars [3,4]. For example, He et al. (2019) found that Beijing’s carbon emissions in 2030 would be 72 million tons, and net zero emissions would be achieved by 2050 [5]. Huang et al. (2022) held that enhancing energy efficiency and improving energy structure are key factors in reducing Beijing’s carbon emissions [3]. However, few scholars conduct research on carbon emissions in specific regions of Beijing. Actually, as a super large city, Beijing has a total area of 16,410.54 square kilometers and 16 districts. There are huge differences in energy consumption and carbon emissions among districts, and it is necessary to conduct scientific planning and precise policies at a fine scale.

Huairou district is one of the 16 districts in Beijing, located in the northeast of Beijing. It covers an area of 2122.8 square kilometers, making it the second largest district in Beijing. In 2024, the Huairou district had a permanent resident population of 439,000, with a GDP of CNY 56.45 billion and a GDP growth rate of 5% (data source: the website of the Huairou district government (https://www.bjhr.gov.cn/) accessed on 30 August 2025). Its industrial structure is dominated by the tertiary sector (69.4%), followed by the secondary sector (29.8%) and the primary sector (0.8%). Under the goals of carbon peaking and carbon neutrality, the Huairou district government, like other districts in Beijing, has implemented a series of measures to promote industrial upgrading and technological progress. For instance, they are shutting down energy-intensive and high-polluting enterprises, while supporting industries such as scientific research and development. However, the Huairou district also has a significant distinction from other districts. The Huairou district has abundant forest resources and water sources. It attaches great importance to ecological protection and restricts large-scale high-intensity industrialization and urbanization development. Therefore, the reduction pathway for carbon emissions in the Huairou district may differ from that in other districts. For this reason, this study explores the carbon emissions and carbon reduction pathways in the Huairou district. Specifically, we construct the Long-range Energy Alternatives Planning System (LEAP) model and set four scenarios, a baseline scenario (BAS), an industrial structure upgrading scenario (Indus), a technological progress scenario (Tech), and a comprehensive transformation scenario (COM), to simulate carbon emissions in the Huairou district from 2021 to 2060.

The marginal contribution of this study is as follows. Firstly, this study broadens the application of the LEAP model by employing it to forecast carbon emissions at the level of urban administrative districts. Existing research on carbon emissions has mainly focused on the national, provincial, and prefectural levels. This study applies the LEAP model to an administrative district in Beijing, which not only extends the LEAP model’s application to a more micro-level but also fills the gap in existing research on carbon emission in the Huairou district. Secondly, this study reveals a unique pathway to carbon reduction. Most existing research indicates that upgrading the industrial structure is a key to reducing carbon emissions. However, this is not significant for the Huairou district; what is more important is the development of energy-saving technology. Furthermore, looking at each specific sector, the urban household and commercial sectors are the key for the Huairou district to reduce carbon emissions in the future. The findings could provide a reference for the Huairou district and other similar areas.

The arrangement of this study is as follows. The Section 2 is a literature review. The Section 3 describes methodology, including an analysis framework, scenario settings, and data sources. The Section 4 presents the simulation results, including the prediction of energy consumption and carbon emissions, and a carbon emission reduction potential analysis. Section 5 is the discussion. The Section 6 is the conclusion and policy implications.

2. Literature Review

With increasing pressure to reduce carbon emissions, scholars have conducted extensive research on forecasting carbon emissions. In terms of methodology, scholars usually employ top-down models or bottom-up models to simulate carbon emissions. Top-down models primarily rely on econometric and general equilibrium theories to forecast carbon emissions, such as the CGE model [6] and the STIRPAT model [7,8]. Owing to the predominant use of economic theories, top-down models facilitate macroeconomic analysis. However, they cannot capture the impact of technological progress on carbon emissions nor reveal specific pathways for carbon reduction. Comparatively, bottom-up models operate at the micro level, predicting carbon emissions through detailed analysis of energy production, conversion, and consumption processes. Typical bottom-up models include the MARKAL model [9], the IAM model [10], and the LEAP model [11]. Compared to the MARKAL and IAM models, the LEAP model permits researchers to flexibly construct the model according to research objectives and data availability. It is well-suited for scenarios with incomplete energy data and is widely employed in carbon emission predictions and carbon reduction path analysis at the national, provincial, municipal, and industry levels. The details are as follows.

At the national level, Chen et al. (2020) employed the LEAP model to forecast China’s carbon emissions from 2020 to 2050 [12]. They found that China’s carbon emissions would stabilize at 10 billion tonnes in the future. Moreover, replacing thermal power generation with clean energy sources was pivotal to China’s carbon reduction. Song et al. (2023) carried out a similar study. They indicated that it is important for China to promote energy structure optimization and industrial restructuring to achieve carbon peaking [13]. Besides China, Nieves et al. (2023) utilized the LEAP model to analyze Colombia’s long-term carbon emissions under two distinct scenarios [14]. They found that by 2050, Colombia’s carbon emissions would reach 140.1 million tonnes under the positive scenario and 150.5 million tonnes under the negative scenario. El-Sayed et al. (2023) applied the LEAP model to analyze Egypt’s energy consumption and carbon emissions over the next three decades [15].

At the provincial and municipal levels, scholars have analyzed the carbon emissions and carbon reduction pathways of multiple provinces or prefecture-level cities. For example, Huang et al. (2022) used the LEAP model and an extended STIRPAT model to investigate Beijing’s carbon reduction pathways [3]. They found that upgrading the energy structure and enhancing energy efficiency are essential to Beijing’s carbon reduction. Cai et al. (2023) employed the LEAP model to forecast carbon emissions in Bengbu city from 2020 to 2060. The results indicated that the carbon emissions peak would range between 18.17 and 29.16 million tonnes [16]. Enhancing industrial electrification rates and reducing coal-fired power generation were identified as critical measures for promoting carbon reduction. Zhao et al. (2025) employed the LEAP model to forecast carbon emissions for Shenzhen city, observing that carbon emissions exhibit an initial rise followed by a decline, with peak emissions ranging from 48.5 to 55.0 million tonnes [17]. Miao et al. (2024) indicated that Jiangsu province is most likely to achieve a carbon emissions peak between 2025 and 2030, with the peak emissions estimated to be approximately 792–853.9 million tons [18]. Reducing energy intensity, optimizing the power production structure, and improving the terminal energy consumption structure are key to achieving carbon emission reductions in Jiangsu province. Ming et al. (2024) found that reducing energy consumption intensity and improving the power generation mix are key measures for driving carbon emission reductions in Zhejiang province [19].

Additionally, some scholars have conducted studies on specific industries. For example, some scholars have predicted carbon emissions in the transport industry [20,21,22], the power industry [23,24,25], the construction industry [26,27,28], and the iron and steel industry [29,30] under different scenarios. Specifically, Ren et al. (2024) proposed a LEAP-REP (renewable energy penetration) model to analyze the carbon emissions of China’s power sector [23]. They pointed out that the significant supply of clean energy and the development of energy storage technologies could drive China’s power sector to achieve net-zero emissions by 2050. Zhang and Luo (2023) constructed a LEAP model to predict the carbon emissions of China’s public buildings sector [27]. They found that, except for the baseline scenario, all other scenarios could achieve carbon peaking by 2030, with carbon emission peaks of 83,178 million tonnes and 92,256.3 million tonnes, respectively.

In summary, while existing studies have conducted predictive analyses of carbon emissions from multiple perspectives, there remains room for further research. Firstly, predictive analyses of carbon emissions have primarily focused on the national, provincial, municipal, and industry levels. Although some studies have focused on Beijing’s carbon emissions and carbon reduction pathways, few have broken down their research into specific regions of Beijing (such as the Huairou district). Secondly, some studies predict carbon emissions over a relatively short period and do not consider China’s latest carbon peak and carbon neutrality restrictions, which may make the predictions biased. Therefore, in the context of the carbon peak and carbon neutrality, this study uses the LEAP model to predict carbon emissions in the Huairou district, Beijing, and identify its carbon reduction pathways, so as to provide more precise energy-saving and emission-reduction solutions for the Huairou district and other similar regions.

3. Methodology

3.1. Model Framework

Based on the availability of data and the characteristics of the LEAP model, this study utilizes the LEAP model to predict long-term annual carbon emissions of the Huairou district from 2021 to 2060. The LEAP model is a scenario-based energy–environment–economy model characterized by simplicity of use, modest data requirements, and capability for specific scenario simulations [31]. It is now widely employed for stimulating carbon reduction. Its principle involves employing parameters such as population, GDP growth rates, and energy consumption structures to conduct bottom-up scenario simulations of energy consumption arising from energy conversion and final energy demand, thereby projecting carbon emissions [12]. Following Li et al. (2024), we constructed the LEAP model framework as shown in Figure 1 to simulate carbon emissions in the Huairou district [32]. It comprises four sections: parameter settings, terminal energy demand, energy conversion, and the resources section.

As shown in Figure 1, energy consumption mainly consists of two parts: terminal energy demand and energy conversion loss. Terminal energy demand refers to the total amount of energy consumed by end-demand sectors for consumption purposes. It could be calculated based on the activity levels of end-demand sectors and their corresponding energy intensity [32]. Referring to Yang et al. (2025), we construct the following formula to calculate terminal energy demand in the Huairou district [33].

$${EC}_k={\sum }_{m=1}^M{AL}_m\times {EI}_m\times {\theta }_{m,k}$$

(1)

where ${EC}_k$ is the consumption of energy k by the energy end-demand sector, comprising seven sectors: the agriculture, industry, construction, commercial, transport, urban household, and rural household sectors. ${AL}_m$ is the activity level of the energy end-demand sector m. Specifically, the activity levels of the agricultural sector, the commercial sector, the transport sector, and the construction sector are measured by their GDP, the activity level of the industrial sector is measured by the output of industrial products, and the activity level of the household sector is measured by the resident population. ${EI}_m$ is the energy intensity of the energy end-demand sector m, which is the amount of energy consumed per unit of economic activity. ${\theta }_{m,k}$ is the proportion of energy k consumed by the energy end-demand sector m to its total energy consumption.

As for energy conversion loss, it refers to energy loss accounted for during energy processing, conversion, transportation, storage, distribution, and other intermediate links [34]. Specifically, it includes energy loss incurred during electricity transmission and distribution, power generation, and heating supply [34]. Due to the availability of data, we mainly consider the loss associated with electricity transmission and distribution. Omitting this loss would result in underestimated energy consumption and carbon emissions. It could be calculated using the following formula:

$${ES}_k={\displaystyle \frac{{EC}_k}{1-{\lambda }_k}}\times {\lambda }_k$$

(2)

where ${\mathrm{E}\mathrm{S}}_{\mathrm{k}}$ is the energy conversion loss and ${\mathsf{\lambda }}_{\mathrm{k}}$ is the rate of loss in the energy k transmission process.

On this basis, the total consumption of energy k could be obtained by summing the terminal energy demand and energy conversion loss, as follows.

$$E_k={EC}_k+{ES}_k$$

(3)

where $E_k$ is the total consumption of energy k. Furthermore, carbon emissions could be calculated by combining the energy consumption and the energy carbon emission factor. The specific formula is as follows:

$$CS=\sum _{k=1}^K{F}_k\times E_k$$

(4)

where CS is the total carbon emissions and $F_k$ is the carbon emission factor for energy k.

3.2. Scenario Settings

The LEAP model is a scenario-based energy–environment–economy model. Considering that carbon emissions are usually influenced by many factors such as industrial structure and energy intensity, we set up the following four scenarios to simulate and analyze carbon emissions in the Huairou district under different scenarios.

Baseline scenario (BAS): This scenario is the most likely scenario of energy consumption and carbon emission in the Huairou district according to the economic development trend. It is guided by the government’s development planning document and the implementation of energy-saving and carbon emission reduction policies.

Industrial structure upgrading scenario (Indus): This scenario is based on the baseline scenario, and the economic growth is consistent with the baseline scenario. The difference between this scenario and the baseline scenario is that the pace of industrial restructuring is faster, and the proportion of the tertiary industry is higher during the same period.

Technological progress scenario (Tech): This scenario is based on the baseline scenario and considers accelerating the development of energy-saving and carbon emission reduction technologies in the Huairou district. Through technological progress, the efficiency of energy utilization will improve, thus achieving a reduction in energy intensity.

Comprehensive transformation scenario (COM): This scenario is based on the baseline scenario and considers both industrial structure upgrading and technological progress. It aims to explore the carbon emission reduction potential of the Huairou district when implementing multiple measures simultaneously.

3.3. Data Sources

Based on the availability of data, this study analyzes the amounts of energy consumption and carbon emissions in the Huairou district from 2021 to 2060, using 2020 as the base year.

The data used in this study includes macro and scenario parameters. Among them, the macro parameters include population, GDP growth rate, urbanization rate, the loss rate of electricity transmission and distribution, and energy consumption structure. The population is from the development planning document of the Huairou district. The GDP growth rate is based on historical GDP data and calculated using the STATA-Arima model. The loss rate of electricity transmission and distribution and the urbanization rate are set referring to the studies of Wang et al. (2020) [35] and Wu et al. (2019) [36]. The energy consumption structure is based on the energy planning of the Huairou district. The main macro parameters are shown in

Table 1. Main macro parameters.

Parameters		2020 (Base Year)	2030	2060
Population (ten thousand)		44.1	52.5	54
Urbanization rate		75.90%	80%	90%
GDP growth rate		5.90%	3.72%	1.84%
Energy consumption structure 1	Coal	4.86%	3%	2.66%
	Natural gas	26.28%	30%	32%
	Electricity	49.65%	55%	61%
	Petroleum	0.01%	0.7%	0.01%
	Other (thermal and LNG)	19.20%	11.3%	4.33%
The loss rate of electricity transmission and distribution		4.32%	3.80%	3.40%

¹ Due to limited space, the energy consumption structure shown here is that of the industry sector. The complete energy consumption structure for each sector is shown in Appendix A Table A1.

.

The scenario parameters include industrial structure and the growth rate of energy intensity. The industrial structure is based on China’s 2030 energy and power development planning and 2060 outlook. The growth rate of energy intensity is referenced to Hong et al. (2021) [37]. The specific scenario parameters are shown in

Table 2. Scenario parameters.

Key Parameters		BAS		Indus		Tech		COM
		2030	2060	2030	2060	2030	2060	2030	2060
Industrial structure	Primary Industry	1.10%	1.10%	1.10%	1.10%	Same as the BAS scenario.		1.10%	1.10%
	Secondary Industry	38.33%	24.65%	28.90%	18.90%			28.90%	18.90%
	Tertiary Industry	60.57%	74.25%	70.00%	80.00%			70.00%	80.00%
The growth rate of energy intensity	Agriculture	−1.89%	−4.15%	Same as the BAS scenario.		−3.77%	−4.15%	−3.77%	−4.15%
	Industry	−1.03%	−3.58%			−2.05%	−2.33%	−2.05%	−2.33%
	Construction	−2.30%	−5.05%			−4.59%	−5.05%	−4.59%	−5.05%
	Commercial	−1.93%	−4.24%			−3.85%	−4.24%	−3.85%	−4.24%
	Transport	−1.66%	−3.69%			−3.32%	−3.69%	−3.32%	−3.69%
	Household	−0.03%	−0.02%			−0.09%	−0.05%	−0.09%	−0.05%

.

4. Results

4.1. Prediction of Energy Consumption

Based on the LEAP model, we forecast the energy consumption of the Huairou district from 2021 to 2060 under four scenarios, and the results are shown in Figure 2. It can be found that energy consumption under all four scenarios shows an upward trend in the short term and peaks around 2030. In the long term, energy consumption under all four scenarios will decline. Specifically, under the BAS scenario, the energy consumption of the Huairou district peaks at 989.72 kilotons of coal equivalent in 2030, with an average annual growth rate of 8.64%. It declines to 645.29 kilotons of coal equivalent in 2060. However, under the COM scenario, energy consumption peaks at 836.49 kilotons of coal equivalent in 2026, which is 15.5% lower than in the BAS scenario and could save 153.23 kilotons of coal equivalent. Under the Tech scenario, energy consumption peaks at 839.15 kilotons of coal equivalent in 2030, which is 150.57 kilotons lower than in the BAS scenario. However, energy consumption in the Indus scenario peaks higher than in the BAS scenario, with a peak of 1006.46 kilotons of coal equivalent in 2030. It is 16.74 kilotons higher than in the BAS scenario.

From the above results, we can conclude that the Indus scenario has the highest energy consumption among all scenarios. This indicates that only promoting the upgrading of the industrial structure cannot reduce the amount of energy consumption in the Huairou district. This may be because the energy intensity of the tertiary industry in the Huairou district is higher than that of the secondary industry. Meanwhile, the above results also show that energy consumption in the Tech scenario (839.15 kilotons) is significantly lower than in the BAS scenario (989.72 kilotons), indicating that technology is a key to reducing energy consumption. Therefore, in pursuit of carbon emission reduction targets, the Huairou district should strengthen research and development and the application of energy-saving technologies. Furthermore, the energy consumption in the COM scenario (836.49 kilotons) is the lowest, indicating that simultaneous promotion of technological progress and industrial structure upgrading has a stronger propelling effect on reducing energy consumption.

4.2. Prediction of Carbon Emissions

The prediction results for carbon emissions in the Huairou district under the four scenarios are shown in Figure 3. It can be found that the carbon emissions under the four scenarios show a trend of first increasing and then decreasing, which is consistent with the trend of energy consumption. In the early stage, perhaps due to the expansion of population size and the growth of the economy, energy consumption increases, which in turn makes carbon emissions increase. In the later stage, under the national policy constraints of carbon peaking and carbon neutrality targets, energy consumption decreases, thereby reducing carbon emissions.

As shown in Figure 3, it can be found that carbon emissions under the BAS scenario peak at 2577.01 kilotons in 2030 and fall to 1520.59 kilotons in 2060. Carbon emissions under the Indus scenario peak at 2608.26 kilotons in 2030, which is 1.21% higher than in the BAS scenario. This indicates that promoting industrial structure upgrading could not reduce carbon emissions in the Huairou district effectively. This may be because the current energy intensity of the tertiary industry in the Huairou district is higher than that of the secondary industry. The peak carbon emission under the Tech scenario is 2187.07 kilotons, which is 15.1% lower than in the BAS scenario. This indicates that in the case of the Huairou district, promoting the application of energy-saving and clean technologies is an important measure to reduce carbon emissions. Finally, carbon emissions under the COM scenario peak at 2126.58 kilotons in 2025, which is the lowest peak among all scenarios, and the peak time is the earliest. This indicates that the reduction of carbon emissions is more strongly propelled by the dual effects of industrial structure upgrading and technological progress. Therefore, it is not difficult to conclude that technological progress could reduce carbon emissions in the Huairou district effectively, and promoting industrial structure upgrading and technological progress simultaneously could help drive carbon emissions to lower levels.

To analyze carbon emissions across sectors, we further examined the carbon emission shares of the seven major sectors under the BAS scenario and the COM scenario. The simulation results are shown in Figure 4 and Figure 5, respectively. It can be found that under the BAS scenario, the carbon emission shares of the agriculture sector and construction sector are relatively small and stable, both maintaining around 2%. The carbon emission share of the commercial sector remains around 30%, consistently the largest emitting sector. The carbon emission share of the transport sector is also relatively large and stable, maintaining around 18%. The carbon emissions share of the rural household sector may decrease slightly, dropping from 7.6% in 2020 to 4.2% in 2060. However, the carbon emission shares of the industrial and urban household sectors fluctuate significantly. Among these, the share of the industrial sector decreases from 24.3% in 2020 to 14.5% in 2060, a reduction of approximately 10%. The share of the urban household sector increases from 15.9% in 2020 to 29.7% in 2060, an increase of approximately 14%. Furthermore, the carbon emission share of the urban household sector will surpass the industry sector and become the second-largest carbon emission sector in 2045. This finding is similar to Ren et al. (2024), who took the Chinese power sector as an example and found that the share of the industrial electricity demand will decrease from 73% in 2015 to 46% in 2050, while that of residential will increase from 13% in 2015 to 26% in 2050 [23]. Under the COM scenario, the share of carbon emissions from sectors and their trends are consistent with those in the BAS scenario.

4.3. Carbon Emission Reduction Potential Analysis

To analyze the carbon emission reduction pathways for the Huairou district, this study further analyzes the carbon emission reduction potential of each sector. Specifically, we construct the following formula to calculate the carbon reduction potential of each sector.

$${Poten}_{mt}={\displaystyle \frac{{COM}_{mt}-{BAS}_{mt}}{{COM}_t-{BAS}_t}}$$

(5)

where ${\mathrm{P}\mathrm{o}\mathrm{t}\mathrm{e}\mathrm{n}}_{\mathrm{m}\mathrm{t}}$ represents the carbon emission reduction potential of sector m in year t; ${\mathrm{C}\mathrm{O}\mathrm{M}}_{\mathrm{t}}$ and ${\mathrm{B}\mathrm{A}\mathrm{S}}_{\mathrm{t}}$ represent the total carbon emissions in the COM scenario and BAS scenario in year t, respectively; and ${\mathrm{C}\mathrm{O}\mathrm{M}}_{\mathrm{m}\mathrm{t}}$ and ${\mathrm{B}\mathrm{A}\mathrm{S}}_{\mathrm{m}\mathrm{t}}$ represent the carbon emissions of sector m in the COM scenario and BAS scenario in year t, respectively. Figure 6 and Figure 7 show the carbon emission reduction potential of each sector for the years 2030, 2040, 2050, and 2060.

As shown in Figure 6a, under the BAS scenario, carbon emissions in 2030 are projected to be 2577 kilotons. Comparatively, carbon emissions in the COM scenario are projected to be 2072 kilotons, resulting in a total emissions reduction potential of 505 kilotons. Among these, the carbon emissions reduction potential of the agricultural sector and the urban household sector is relatively small, both at around 20 kilotons of carbon emissions, accounting for less than 5%. Similarly, the carbon emission reduction potential of the construction sector, transport sector, and rural household sector is also small, accounting for about 6%. The carbon emission reduction potential in the commercial sector and the industry sector are significant, at 139.9 kilotons and 229.6 kilotons, respectively, accounting for over 70% of the total. The reason for these results may be that during this stage, carbon emissions in the commercial and industry sectors are relatively high, and energy intensity is also relatively high, leaving significant room for carbon emission reductions.

As shown in Figure 6b, the carbon emission reduction potential of the agricultural sector, the rural household sector, and the construction sector in 2040 does not change significantly compared to 2030. However, the carbon emission reduction potential of the commercial sector will increase to 229 kilotons of carbon emissions, accounting for 34.74%. It will surpass the industrial sector to become the sector with the greatest potential for carbon reduction. In contrast, the carbon emission reduction potential of the industry sector decreases from 45.47% in 2030 to 26.12% in 2040. In addition, the carbon emission reduction potential of the urban household sector and the transport sector has increased significantly. Their carbon reduction potential is approximately 13.4%, with carbon emission reduction of approximately 88 kilotons. This may be attributable to the fact that, with the development of the economy and upgrading of the industrial structure, the proportion of the commercial sector increases significantly, and the living standards of urban residents also rise substantially. Consequently, their energy demands expand, thereby increasing the carbon emissions and the potential for carbon reduction.

In 2050 and 2060, it can be observed from Figure 7 that the carbon emission reduction potential of the agricultural, rural household, construction, and transport sectors remains at a relatively stable level. The commercial sector remains the most promising sector for carbon emission reduction, with its potential stabilizing at around 245 kilotons of carbon emissions, accounting for approximately 35%. The carbon emissions reduction potential of the urban household sector will further increase, reaching 28.5% and 197.4 kilotons of carbon emissions in 2060. However, the carbon emissions reduction potential of the industrial sector will further decrease, dropping to 93.7 kilotons of carbon emissions by 2060, with the proportion decreasing to 13.52%. Based on these results, it is not difficult to conclude that in the medium term, the commercial sector and the industrial sector are the most important sectors for reducing carbon emissions. However, from a long-term perspective, the importance of the urban household sector will continuously increase, eventually becoming the second largest sector for carbon reduction potential after the commercial sector. Hence, the government should pay greater attention to the commercial sector and the urban household sector in the long term.

5. Discussion

This study contributes to the literature by revealing a unique pathway to carbon reduction. Existing research has mostly centered on the pathway of upgrading the industrial structure. However, this study found that the development of energy-saving technology is more important. Furthermore, looking at each specific sector, the commercial sector and the urban household sector are the key sectors to reduce carbon emissions. This finding is not difficult to understand. Figure 4 and Figure 5 above illustrate the carbon emissions share of each sector under different scenarios, revealing that the commercial sector consistently accounts for the largest share. The share of carbon emissions from the industrial sector shows a significant decline, while that from the urban household sector exhibits a marked increase. Correspondingly, Figure 6 and Figure 7 demonstrate the carbon reduction potential across sectors, indicating that the carbon emissions reduction potential of the industrial sector steadily diminishes, whereas that of the commercial sector and urban household sector progressively increases.

The reasons for these results may be as follows. Firstly, with the upgrading of the industrial structure, the proportion of industrial output will decrease. Meanwhile, with the development of energy-saving technologies, the carbon emission intensity of the industrial sector will continuously decrease, ultimately resulting in diminishing carbon reduction within the industrial sector. The decline in carbon emissions from the industrial sector will decrease its carbon reduction potential. However, it should be noted that the upgrading of the industrial structure will increase the share of the commercial sector’s output, leading to higher carbon emissions from the commercial sector and greater potential for carbon emissions reductions. Secondly, the development of the economy will further enhance the level of urbanization, thereby expanding the size of the urban population. Meanwhile, the development of the economy will also diversify the energy demands of urban residents, ultimately leading to higher carbon emissions and greater potential for carbon reduction within the urban household sector. In short, this study highlights the critical role of the commercial and urban household sectors for carbon reduction. The government should formulate more policies targeting carbon emission reductions in these two sectors.

Another contribution of this study is to broaden the application of the LEAP model by employing it to forecast carbon emissions at the level of urban administrative districts. Existing studies on carbon emissions have primarily focused on the national, provincial, or prefectural city levels. This study employs the LEAP model to examine carbon emissions within a district of Beijing city, which could extend the application of the LEAP model to a more micro-level scale. Meanwhile, although there are some studies on carbon emissions in Beijing, few studies have examined carbon emissions from a specific district of Beijing. This paper could contribute to filling this gap to a certain extent.

However, this study also has some shortcomings. Firstly, due to the availability of data, this study takes 2020 as the baseline year. In the future, more accurate simulation analyses could be carried out based on updated data. Secondly, due to the availability of data and its relatively low proportion of carbon emissions, the energy conversion losses in this study only encompass those arising from electricity transmission and distribution. This may result in the carbon emissions predicted in this study being smaller than the actual value. Future research could be conducted based on more comprehensive data. Thirdly, this study constructs the LEAP model by integrating the economic conditions of the Huairou district. Owing to the omission of non-economic factors like cultural differences, our findings and policy implications may not apply to some regions.

6. Conclusions and Policy Implications

Based on the LEAP model, this study constructs a multi-sectoral energy consumption and carbon emission accounting framework for the Huairou district, Beijing. By setting up four scenarios, namely the BAS scenario, Indus scenario, Tech scenario, and COM scenario, we systematically simulate the carbon emission of the Huairou district from 2021 to 2060. The results show that all four scenarios could realize the target of carbon peaking by 2030. Among them, the peak carbon emissions in the Indus scenario are the highest (2608.26 kilotons), while the peak carbon emissions in the COM scenario are the lowest (2126.58 kilotons). Moreover, by distinguishing the carbon emissions of various sectors, it can be observed that the commercial sector is the largest contributor to carbon emissions, with a stable share of around 30%. The proportion of carbon emissions from the industrial sector will continue to decline (from 24.3% in 2020 to 14.5% in 2060), while the proportion of carbon emissions from the urban household sector will continue to increase (from 15.9% in 2020 to 29.7% in 2060). Furthermore, the analysis of the carbon emission reduction potential of various sectors reveals that in the medium term, the commercial and industrial sectors are the key to reducing carbon emissions. However, in the long term, the focus of carbon emission reduction needs to shift towards the commercial and urban household sectors.

Based on the above findings, this study puts forward the following policy implications. Firstly, vigorously promote the innovation and application of energy-saving and clean technologies. The government could establish a dedicated fund to support research in clean technologies such as carbon capture, utilization and storage (CCUS), smart energy management, and green building. Meanwhile, promote the adoption of clean technologies through policy and market incentives. For example, construct industrial demonstration bases and provide subsidies for first-of-a-kind equipment. Secondly, actively promote deep carbon reduction in the commercial sector. Given that the commercial sector will account for the largest share of carbon emissions (approximately 30%), governments should prioritize it as the key to carbon reduction. They could introduce dual controls on both the total energy consumption and energy intensity of the commercial sector, implement ultra-low energy consumption standards for buildings in key commercial zones, and establish incentive mechanisms for green electricity procurement. Thirdly, actively promote the low-carbon transformation of the industrial sector. Although the share of the industrial sector’s carbon emissions will decline in the future, its carbon reduction potential remains substantial in the medium term. To promote carbon emission reductions in the industrial sector, the government could establish an industrial entry and exit mechanism based on carbon intensity and strictly restrict the entry of high-energy-consuming projects. Concurrently, establish a special fund for industrial low-carbon transition, supporting existing enterprises in upgrading production lines. Fourth, guide urban households towards a low-carbon lifestyle transition. Given the fact that carbon emissions and carbon reduction potential in the urban household sector will rise in the long term (rising from 15.9% to 29.7%), the government should proactively plan for the transformation of community energy systems. This includes incorporating rooftop photovoltaic installations and charging infrastructure into new residential developments. Furthermore, encourage public transport and non-motorized travel, thereby fostering the adoption of green consumption patterns.