Air quality and health co-benefits of China's carbon dioxide emissions peaking before 2030

The study addresses when China’s CO2 emissions peak and the associated social and economic implications, particularly air quality and health co-benefits of earlier and more ambitious climate mitigation pathways. Against the backdrop of the IPCC 1.5 °C report and China’s pledges to peak CO2 before 2030 and reach carbon neutrality before 2060, the authors examine how mitigation policies that reduce greenhouse gases can also reduce air pollutants from shared sources, thereby producing health benefits. Prior global work indicates health co-benefits can offset mitigation costs by mid-century, but cost-benefit ratios differ across regions, with potentially large gains in developing countries like China. With evidence suggesting China’s emissions may peak before 2030, existing analyses based on an “around 2030” peak may underestimate co-benefits. This study quantifies how different socio-economic pathways and climate policy stringencies influence CO2 trajectories, air pollutant emissions, PM2.5 exposure, health outcomes, and the monetized balance of costs and benefits, to inform policy design.

Prior research shows that reducing GHGs often co-reduces air pollutants, yielding health benefits that can offset mitigation costs by 2050 globally. Studies report higher benefit-cost ratios in developing countries (e.g., 3–9 times in China and India) compared to the EU and USA. Recent analyses suggest China could peak CO2 emissions before 2030, earlier than its Paris NDC, implying potential underestimation of air quality and health co-benefits in older literature tied to later peaks. However, comprehensive assessments that integrate diverse socio-economic futures (SSPs), multiple climate target stringencies (RCPs), and explicit health impact valuation for China have been limited. The present work addresses these gaps by coupling IAM outputs with atmospheric chemistry and epidemiological models to quantify co-benefits and economic trade-offs under early peak scenarios.

The study builds an integrated assessment framework combining socio-economic scenarios, an integrated assessment model (IAM) for emissions and costs, a chemical transport model for air quality, an epidemiological model for mortality, and economic valuation of health benefits. - Scenarios: Nine SSP–RCP combinations: SSP1, SSP2, SSP5 each with REF (no climate policy), RCP2.6 (≈2 °C), and RCP1.9 (≈1.5 °C) radiative forcing targets by 2100. The RCP constraint fixes cumulative CO2 emissions to meet the RF target, shaping energy mix, technology deployment (including CCS), and emissions. - Emissions and mitigation costs: GCAM provides projections for CO2 and co-emitted air pollutants and approximates mitigation costs as welfare loss via theoretical carbon tax revenue (direct abatement costs such as new power plants and CCS, not full macroeconomic GDP/consumption losses). Emission factors evolve with income and policy (including MAC curves for non-CO2 GHGs). Demographic and economic inputs follow SSP assumptions. - Air quality modeling: PM2.5 concentrations simulated with WRF-Chem v3.6.1 for 11 experiments: base year 2015 and 10 future cases (SSP1_REF, SSP1_RCP2.6, SSP1_RCP1.9, SSP5_REF, SSP5_RCP2.6 in 2030 and 2050). Meteorology from 2015 is held constant across scenarios to isolate policy effects. Four representative months (Jan, Apr, Jul, Oct) are simulated over Greater China at 20 km resolution. 2015 emissions are from the MIX inventory; future emissions are derived from GCAM. Biogenic emissions from MEGAN. Model performance is evaluated against ~1,500 ground stations; annual mean PM2.5 correlation ~0.81; satellite-derived PM2.5 is used for calibration and exposure estimation. - Health impact assessment: The Global Exposure Mortality Model (GEMM) estimates PM2.5-attributable mortality for noncommunicable diseases and lower respiratory infections among adults ≥25 years. Population and age structures are from LandScan (2015) and SSP projections. Attributable deaths are computed using GEMM concentration–response functions; uncertainty is propagated using 95% CIs of GEMM parameters. Health co-benefits are differences in attributable mortality between RCP scenarios and corresponding REF. - Valuation: Avoided deaths are monetized using Value of Statistical Life (VSL). Primary VSL estimates are from local Chinese stated-preference studies (e.g., Hammitt et al.), scaled over time via income-based marginal VSL; alternative scaled international VSL and US EPA approaches are explored in sensitivity analysis. Net benefits compare monetized health benefits to GCAM mitigation costs. - Focused simulations: While nine SSP–RCP combinations are conceptually considered, detailed air quality and health simulations emphasize SSP1 and SSP5 under REF and RCP2.6, plus SSP1_RCP1.9, at 2030 and 2050, with a 2015 base.

- CO2 peaking: Under all mitigation scenarios (RCP2.6 and RCP1.9), China’s CO2 emissions peak at 10,423–15,500 Mt CO2/yr in or before 2030, meeting the NDC “around 2030” target. Under RCP1.9 (≈1.5 °C), peaks occur earlier, around 2020 at 10,645–11,901 Mt CO2/yr. - Carbon intensity: China achieves the NDC carbon intensity target of 0.47 t-CO2/1000 USD by 2030 in all scenarios considered. - Energy structure: Meeting more stringent targets requires rapid shifts from fossil fuels to biomass, nuclear, and non-biomass renewables, plus large-scale CCS deployment (e.g., by 2050 in SSP2/SSP5, ~98% of fossil-fuel and ~92% of bioenergy CO2 in power and industry captured and stored). - Air pollutant emissions: CO2 mitigation co-reduces SO2 and NOx substantially due to shared energy sources (especially coal). VOC, BC, and OC see smaller co-reductions given different sectoral sources. NH3 emissions increase in climate policy scenarios because of rising fertilizer use associated with agriculture and bioenergy expansion. - PM2.5 exposure: Population-weighted PM2.5 (PWC-PM2.5) improves more under stricter policies and over the long run. In SSP1_RCP1.9, PWC-PM2.5 is 50.9 µg/m3 in 2030 (2.1 µg/m3 lower than SSP1_REF) and 43.4 µg/m3 in 2050 (5.1 µg/m3 lower). In 2050, SSP1_RCP1.9 is 3.5 µg/m3 lower than SSP1_RCP2.6. However, under SSP1_RCP1.9, about 778 million people remain exposed to >35 µg/m3 in 2050, indicating climate policy co-benefits alone are insufficient to meet China’s air quality goals. - Local trade-offs: In 2030, SSP1_RCP2.6 shows slightly higher PWC-PM2.5 than SSP1_REF (+0.4 µg/m3 nationally); some agricultural provinces (Henan +0.7, Shandong +0.5, Jiangsu +0.5 µg/m3) see increases due to higher NH3 and insufficient NOx/SO2 reductions, enhancing secondary ammonium nitrate/sulfate formation. - Health burden trajectory: PM2.5-attributable deaths are estimated at 2.44 million (95% CI: 2.05–2.81 million) in 2015, rising to 3.6–3.9 million in 2030 and 6.4–7.5 million in 2050 across scenarios, driven chiefly by population aging. - Avoided mortality (health co-benefits): Relative to SSP1_REF, SSP1_RCP1.9 avoids about 118,000 deaths (3.2%) in 2030 and 614,000 deaths (8.7%) in 2050. Compared with SSP1_RCP2.6, SSP1_RCP1.9 has 3.7% and 6.1% fewer PM2.5-related deaths in 2030 and 2050, respectively. - Economic balance: By 2050 under RCP2.6, monetized health co-benefits can fully offset mitigation costs, yielding net benefits of about $393 billion (SSP1) and $3,017 billion (SSP5) in 2017 USD, equivalent to roughly 0.45% and 2.77% of China’s GDP in 2050. Additional cost savings from reduced pollution control expenditures are not included, implying further net gains.

The analysis shows that earlier and more ambitious CO2 mitigation in China accelerates the emissions peak and yields sizable air quality improvements and health co-benefits that grow over time. These co-benefits, particularly under stringent targets and cleaner socio-economic pathways, are large enough by mid-century to offset or exceed the direct abatement costs captured by GCAM, supporting the case for ambitious climate action on economic grounds. However, due to persistent PM2.5 exposure and population aging, climate policy co-benefits alone do not prevent increases in PM2.5-attributable mortality before 2050, nor do they achieve national air quality goals such as ≤35 µg/m3 everywhere. Moreover, expanded biomass and agricultural activity can increase NH3 emissions, limiting PM2.5 reductions or even causing local increases via secondary particulate formation. Therefore, synergistic policies are needed: rapid decarbonization (including CCS) combined with strengthened conventional air pollution controls and targeted NH3 management. Leveraging these coordinated strategies maximizes joint benefits for climate, air quality, public health, and the economy, while managing trade-offs arising from bioenergy and agricultural transitions.

China can meet and likely surpass its NDC timeline for peaking CO2 emissions, especially under 1.5 °C-consistent pathways, delivering substantial long-term improvements in air quality and health. By 2050, monetized health co-benefits can offset or exceed direct mitigation costs, producing significant net economic gains. Nonetheless, to achieve air quality targets and curb the rising PM2.5-related health burden amidst demographic aging, climate mitigation must be complemented by stronger air pollution control measures, including rigorous NH3 emissions management and urban and infrastructure design supporting low-emission lifestyles. Future research should refine integrated assessments by representing climate-driven meteorological changes, evolving baseline mortality, pollutant mixtures and PM composition, sector-specific controls (especially agricultural nitrogen), and macroeconomic feedbacks to better capture total costs and benefits.

- Health exposure limited to PM2.5 mass; does not account for O3 or NOx health effects, PM composition, sources, or size distribution. While PM2.5 dominates the air pollution health burden, omission may understate total impacts and mask composition-specific toxicity differences. - GCAM-related uncertainties: technology representations, parameter choices, base-year calibration, and scenario assumptions can affect emissions and cost estimates. - Air quality policy assumptions: Dedicated future air pollution controls beyond climate policies are not modeled; stronger agricultural nitrogen management or other controls could further improve air quality. - Meteorology held constant (2015 fields) to isolate policy impacts; climate-change-induced meteorological shifts are excluded. - Baseline mortality rates assumed constant at 2015 levels due to lack of projections, potentially misrepresenting future health baselines. - Valuation uncertainties in VSL estimates and methods remain substantial, though local estimates were prioritized for applicability.