Clean air policies are key for successfully mitigating Arctic warming

The Arctic has warmed about three times faster than the global average since 1971, raising risks of sea ice loss and regional tipping points. While net-zero CO2 commitments aim to limit warming, current trajectories fall short of Paris Agreement goals. In parallel, regulations to curb transboundary air pollution target short-lived climate forcers (SLCFs) such as particulate matter and tropospheric ozone, which affect both air quality and climate. Despite known linkages, greenhouse gases (GHGs) and SLCFs are often assessed separately. The short atmospheric lifetimes of SLCFs make them promising levers for near-term Arctic and global climate mitigation, yet their mitigation potential and associated uncertainties remain insufficiently constrained. Prior studies often used idealized SLCF scenarios disconnected from GHG pathways and have struggled to separate competing influences of CO2 and particulate components on air quality and climate. This study evaluates how realistic, policy-relevant reductions in key air pollutants and methane influence Arctic climate and human health, and how to design strategies that jointly optimize climate and air quality outcomes.

The paper situates its work within literature showing strong aerosol influences on Arctic amplification and the masking of GHG warming by sulfate aerosols. Prior modeling has indicated that reductions in anthropogenic aerosols can lead to near-term warming, complicating air quality–climate tradeoffs. Studies highlight substantial co-benefits from SLCF mitigation (e.g., methane and black carbon) for climate, health, and food security, but many employed idealized or decoupled scenarios and offered limited process uncertainty reporting. Evidence points to the high efficacy of targeting high-latitude black carbon due to snow/ice albedo effects, yet quantitative attribution separating BC, sulfate, and organic carbon contributions has been limited. This study addresses these gaps by using policy-relevant emissions pathways aligned with socioeconomic scenarios and decomposing radiative forcing pathways (radiation, cloud, and surface albedo interactions), while explicitly quantifying uncertainties.

• Scenarios: The study analyzes four Shared Socioeconomic Pathway (SSP) climate scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5) and four Arctic Monitoring and Assessment Programme (AMAP) air pollution scenarios: Current Legislation (CLE), Maximum technically Feasible Reduction (MFR), MFR plus Sustainable Development Scenario (MFR_SDS), and Climate Forcing Mitigation (CFM). AMAP scenarios were developed with the GAINS model to explore targeted SLCF policies broadly consistent with SSP CO2 trajectories.
• Emissions: Historical emissions from CEDS and AMAP were used; future emissions include sulfur dioxide (SO2), black carbon (BC), organic carbon (OC), methane (CH4), and CO2. SSPs represent differing socioeconomic and policy narratives, with generally declining sulfur and carbonaceous aerosol emissions in most scenarios, except more limited reductions in SSP3-7.0.
• Climate modeling: Temperature responses (global and Arctic) for 2015–2050 were simulated using multiple Earth System Models (ESMs), including CESM, NorESM, GISS-E2-1, and UKESM1, complemented with CMIP6 multi-model ensembles for SSPs. An emulator framework based on Absolute Global/Regional Temperature-Change Potentials (AGTP/ARTP) was used to attribute temperature responses to emission pulses by species and region. The emulator decomposes radiative forcing pathways into direct radiation interactions, cloud interactions, and surface albedo changes, with regional sensitivity to aerosols, especially BC in the Arctic.
• Air quality and health: Ten global models (including ESMs and chemistry transport models) provided PM2.5 and ozone fields. The TM5-FASST model estimated PM2.5- and ozone-attributable mortality, accounting for evolving population, age structure, and baseline disease rates. Model outputs were downscaled to 0.5° resolution using satellite-informed approaches; concentration responses were scaled linearly with regional emission changes.
• Forcing estimates: Global effective radiative forcings (ERFs) for 2015 relative to 1850 were estimated for sulfur, BC, and OC across radiation, clouds, and surface albedo interactions. Example ERFs (W m−2): Sulfur: −0.28 (radiation), −0.55 (cloud), −0.08 (albedo); BC: +0.26 (radiation), −0.05 (cloud), +0.09 (albedo); OC: −0.02 (radiation), −0.21 (cloud), −0.01 (albedo).
• Uncertainty treatment: The emulator conditioned results on multi-model forcing ranges and explored sensitivity to equilibrium climate sensitivity (ECS), removing models with ECS outside 2.5–4°C for sensitivity tests. The emulator focuses on forced responses and does not include unforced variability. Potential biases from low Arctic BC in some models and lower ECS in emulator settings were considered. Confidence intervals (typically 5–95%) were reported for temperature changes and health outcomes, with model and method uncertainties discussed in supplementary materials.

• Sulfur unmasking: Rapid sulfur emission reductions associated with clean air policies can enhance Arctic warming by about 0.8°C by 2050 relative to 1995–2014, offsetting some climate benefits from GHG reductions.
• Mitigation via BC and CH4: Prioritizing reductions of black carbon and methane can counteract a substantial fraction of the sulfur-unmasking effect:
  • BC reductions (MFR vs CLE) reduce Arctic warming by ~0.3°C (0.1–0.4°C) in 2050.
  • CH4 reductions add ~0.2°C (0.1–0.2°C) additional Arctic cooling.
  • In a CFM scenario (prioritizing CH4 and warming SLCFs while maintaining CLE for cooling species), Arctic temperatures in 2050 are ~0.4°C lower than in CLE.
  • Combined SLCF mitigation in MFR/MFR_SDS reduces Arctic warming by ~0.2°C despite concurrent sulfur declines.
• Comparative climate benefits: Ambitious BC and CH4 reductions can yield Arctic climate benefits by 2050 comparable to those from CO2 reductions alone (e.g., ~0.5°C difference between SSP1-2.6 and SSP5-8.5 in Arctic warming).
• Regional leverage: Emissions from Arctic Council countries, though only ~6% of global BC, contribute ~40% of the Arctic warming reduction achievable from BC controls due to strong snow/ice albedo effects.
• Air quality and health co-benefits:
  • PM2.5 concentrations fall markedly under MFR and MFR_SDS globally, in Arctic Council countries, and in Asian Arctic Council observer countries, with most reductions before 2030. Despite declines, many regions remain above the WHO 2021 guideline of 5 µg/m³ in 2050.
  • PM2.5-attributable mortality in 2050 (MFR vs CLE): globally −1.2 million deaths (−28%); Arctic Council −100,000 (−58%); Asian observer countries −700,000 (−31%).
  • Under CFM, weaker sulfur mitigation leads to smaller PM2.5 reductions and higher mortality than MFR.
  • Ozone mortality is much smaller than PM2.5 mortality but follows similar scenario patterns; under CFM, global ozone-related mortality increases by ~72% from a 2015 baseline of ~400,000 by 2050; Arctic Council ozone mortality rises ~11% from ~300,000 in the same period.
• Scenario comparisons: Arctic warming in an MFR with SSP1-2.6 context is projected at ~1.5°C by 2050 vs ~1.9°C with weaker SLCF controls, indicating substantial near-term benefits of stringent SLCF mitigation alongside CO2 reductions.

The study demonstrates that while clean air policies deliver large health benefits, rapid declines in sulfur emissions unmask previously hidden greenhouse warming, accelerating near-term Arctic temperature rise. Targeted mitigation of warming SLCFs—especially black carbon and methane—can substantially counterbalance this effect, yielding Arctic climate benefits comparable to ambitious CO2 reductions by mid-century. The results highlight the outsized leverage of high-latitude BC controls due to snow/ice albedo pathways and show that combining SLCF measures with CO2 mitigation slows Arctic warming more effectively than CO2-focused strategies alone in the near term. From a policy perspective, aligning air quality and climate frameworks is essential: designs that prioritize reductions of warming SLCFs while managing the pace of cooling aerosol declines can maximize co-benefits and minimize unintended warming. Health co-benefits are large and rapid under maximum feasible controls, reducing PM2.5 exposure and associated premature deaths across regions, though many areas will still exceed stringent WHO guidelines without deeper, broader emission cuts. Overall, integrated strategies that jointly address CO2, CH4, and particulate carbon deliver the most robust near-term Arctic climate and health benefits.

The paper introduces an integrated modeling approach coupling policy-relevant emissions scenarios, multi-model climate simulations, and an emulator to quantify climate and health outcomes of clean air policies. It finds that: (1) sulfur emission reductions necessary for air quality lead to notable near-term Arctic warming via unmasking; (2) prioritizing BC and CH4 mitigation can offset much of this additional warming; and (3) maximum feasible SLCF controls provide substantial health co-benefits, preventing up to 1.2 million premature deaths globally in 2050 relative to current legislation. Policy implications are clear: aggressively reduce BC and CH4, particularly in and near the Arctic, while continuing deep CO2 cuts to stabilize climate. Future research should refine regional attributions, improve aerosol–cloud and albedo process understanding, expand ensemble sizes to better capture uncertainties, and explore policy designs that coordinate the timing of cooling aerosol reductions with accelerated warming SLCF mitigation.

• Observational constraints limit direct detection of warming unmasking from declining sulfur emissions.
• Emulator simplifications: forced response only (no unforced variability), linear scaling of concentration responses, potential underestimation of Arctic warming and BC impacts due to low BC concentrations in some underlying models.
• Model uncertainty: aerosol radiative forcing, aerosol–cloud interactions, and feedbacks remain major uncertainty sources; ESM ensembles may not fully span plausible ranges.
• Sensitivity to climate parameters: results depend on emulator parameter choices (e.g., ECS and ARTP scaling). Alternative ECS selections (2.5–4°C) change SLCF climate impacts by ~20% and CO2 impacts by ~10–20%, within reported uncertainties.
• Scenario assumptions: While more policy-relevant than idealized cases, scenarios still rely on assumptions about technology adoption, legislation implementation, and socioeconomic pathways that may diverge from future realities.