Balanced Synergistic Control Plans Needed for China’s Carbon Emissions

The sources of health-threatening air pollutants and CO₂ emissions differ significantly, despite both mainly originating from fossil fuel combustion. A 2017 analysis of emission inventories reveals that 86% of anthropogenic CO₂ emissions were from the energy and industrial sectors, with coal combustion contributing 97% and 50% of these emissions respectively (Fig. 1a). However, coal combustion in these sectors accounted for only 17% of primary PM_2.5 emissions (Supplementary Fig. 1).

The domestic sector, relying on coal and solid biomass, contributed 38% of PM_2.5 and 22% of SO₂ emissions. Poor air supply, mixing, and emission control in household stoves using low-quality fuels are significant factors (Ref 20). Despite contributing only 4% to CO₂ emissions, the domestic sector was responsible for over a quarter of premature deaths from ambient PM_2.5 exposure (Fig. 1b). In contrast, the energy generation sector, a major CO₂ emitter, accounted for only 10% of these deaths. This aligns with previous studies (Ref 21, Ref 22).

Certain subsectors contribute to health damage without significant CO₂ emissions, such as non-combustion industrial processes, agricultural fertilizer application, and crop residue burning (Supplementary Table 1). Although crop residue combustion emits CO₂, it is considered carbon neutral (Ref 23). These “non-synergistic subsectors” account for 23% of premature deaths attributable to the five sectors assessed.

a Source contributions to CO₂ emissions in China in 2017 based on the coordinated CO₂ emission inventory with the air pollutants. The lower bar represents the contributions of different sectors, and the upper bar represents the contributions of different fuel/process categories. b Source contributions to premature deaths attributable to long-term ambient PM_2.5 exposure in China in 2017 based on the Adjoint simulation of the Community Multiscale Air Quality modeling. The left bar represents the contributions of different sectors, and the right bar represents the contributions of different fuel/process categories. c Comparison of impacts on health damage and CO₂ emissions for each subsector. The corresponding monetized impacts are also shown on the minor axis based on a uniform statistical life value (1.33 million US dollars per premature death) and social cost of carbon (100 US dollars per ton of CO₂ emission). Source data are provided as a Source Data file.

Seven subsectors were identified as major contributors to health damage, each responsible for over 100,000 premature deaths (Fig. 1c). These include bituminous coal combustion in the domestic, energy, and industrial sectors; diesel emissions; hydraulic cement and iron production; and agricultural activities. Notably, the health impact of agricultural activities is significant due to NH₃ emissions, even though they are not major CO₂ contributors.

The ratio of health damage to CO₂ emissions varies significantly across subsectors, from 0.094 in industrial dry natural gas combustion to 22 in domestic unorganized waste combustion. This highlights the need for coordinated strategies to balance air pollution reduction and decarbonization.

Spatial Heterogeneity in Contributions to Health Damage and CO₂ Emissions

High-resolution data integration reveals spatial disparities in contributions to health damage and CO₂ emissions. Densely populated areas like the eastern and central regions and the Sichuan Basin show higher contributions to health damage than CO₂ emissions (Fig. 2a). Hubei had the highest ratio (3.9), followed by Henan and Chongqing. In contrast, Inner Mongolia had a lower health impact despite significant CO₂ emissions.

a Ratio between the gridded contribution to PM_2.5 exposure-related health damage and to CO₂ emissions. The provincial boundary shapefile is obtained from Harvard Dataverse (https://doi.org/10.7910/DVN/DBJ3BX) and is publicly available under the Creative Commons CC0 Public Domain Dedication. b Comparison of provincial contributions to health damage and CO₂ emissions in each sector. The red dots represent the ratio of provincial contributions to health damage versus to CO₂ emissions. The provinces are ranked from large to small in population size. c Correlation between population density (pop) and the ratio of each city’s contributions to health damage versus to CO₂ emissions (ratio). The dot color represents the domestic solid fuel usage fraction, which is defined as the fraction of energy consumed as solid fuels in the domestic sector out of the total province-level energy consumption. The solid line and dashed line represent the correlations for cities with domestic solid fuel consumption above and less than 1%, respectively, of the total provincial energy consumption. Source data are provided as a Source Data file.

Population density significantly influences the spatial distribution of the health damage to CO₂ emissions ratio. A log-linear regression at the city level shows a positive correlation between population density and the ratio of each city’s contribution to nationwide PM_2.5-related health damage versus CO₂ emissions (Fig. 2c). This correlation is stronger in cities with higher reliance on solid fuels for domestic energy.

Monetized Social Costs of CO₂ Emissions and PM_2.5 Exposure-Related Health Damage

Monetizing health damage and climate impacts using a uniform value of statistical life (VSL) of $1.33 million per death and a social cost of carbon (SCC) of $100 per ton of CO₂ (Methods) shows that for 36 synergistic subsectors, health damage costs often exceed climate impact costs for half of the subsectors (Fig. 1c). Higher VSL estimates would further emphasize the near-term health benefits of CO₂ emission reductions (Ref 12, Ref 29).

Comparing monetized health damage and climate impacts with sectoral GDP reveals that integrated costs of 42 economic sectors equal 20% of total GDP. In most sectors, these costs are below 5% of GDP, but exceed GDP in four sectors including electric power production and metal processing (Supplementary Fig. 7). This highlights the need for decarbonizing energy-intensive industries.

Integrating Costs from Health Damage Modifies Control Priorities

We propose a unified indicator that combines social costs from CO₂ climate impacts and PM_2.5 health damage to harmonize control strategies. In 2017, integrated costs for seven subsectors exceeded $100 billion, with bituminous coal combustion in energy generation being the highest. Other significant contributors include industrial coal use, diesel vehicles, and cement and steel production (Supplementary Table 1).

Spatial analysis shows that regions with high integrated costs are densely populated areas with high pollutant and CO₂ emissions. Cities like Chongqing, Zhengzhou, and Shanghai are major contributors, with health costs comprising 51%-84% of the integrated costs.

a Spatial distribution of PM_2.5 exposure-related health damage in China at a 36 km by 36 km resolution in 2017. b Spatial distribution of social costs from CO₂-related climate change. c Spatial distribution of the integrated costs, which equal the sum of the monetized health damage and social cost from CO₂-related climate change. The provincial boundary shapefile used in a-c is obtained from Harvard Dataverse (https://doi.org/10.7910/DVN/DBJ3BX) and is publicly available under the Creative Commons CC0 Public Domain Dedication. d Quantile‒quantile plot of the social cost distribution against the population distribution. The three lines represent comparisons of social costs from CO₂-related climate change, PM_2.5-related health damage, and integrated costs. Source data are provided as a Source Data file.

Provinces like Jiangsu, Guangdong, and Inner Mongolia rank high in social costs from climate change but lower in integrated costs due to less health damage. In contrast, densely populated inland provinces rank higher in integrated costs. Chongqing shows the largest gap between these rankings.