Aerosols overtake greenhouse gases causing a warmer climate and more weather extremes toward carbon neutrality

Human activities since the preindustrial era have increased atmospheric greenhouse gases, leading to a 1.1 °C rise in global mean surface air temperature. Warming enhances thermal and latent energy, increasing the likelihood and severity of extreme weather such as heat waves and heavy precipitation, with substantial impacts on ecosystems, agriculture, economies, and health. Many countries have pledged carbon neutrality around mid-century to meet the Paris Agreement targets. Carbon neutrality policies reduce fossil-fuel-related emissions of both GHGs and co-emitted air pollutants (PM and O₃ precursors), promising air-quality co-benefits. However, air pollutants such as aerosols and tropospheric ozone also affect radiative forcing: black carbon warms, while sulfate cools. Thus, emission reductions toward carbon neutrality may either dampen or amplify climate responses relative to GHG mitigation alone. Prior studies have largely examined mean climate change; yet, extremes have more immediate societal impacts. Events during COVID-19 highlighted how abrupt aerosol reductions can alter regional climate, rainfall, and wildfire risk. This study addresses how individual changes in GHGs, aerosols, and tropospheric O₃ under a stringent mitigation pathway (SSP1-1.9) affect global climate and extremes toward carbon neutrality.

Previous research indicates that under intermediate future pathways (e.g., SSP2-4.5), changes in GHGs dominate projected mean temperature and precipitation responses, with aerosol-only effects being smaller. DAMIP experiments provide attribution by isolating forcing agents (GHG-only, aerosol-only, etc.), but future DAMIP experiments are limited to SSP2-4.5. Observational and modeling studies have shown that aerosol changes can rapidly impact climate, including during the COVID-19 emission reductions, influencing rainfall in China and wildfire risk in the western U.S. Aerosols exert complex direct and indirect effects, altering cloud microphysics and radiative balance, modulating precipitation via droplet size and cloud lifetime, and shifting the ITCZ. However, most prior assessments emphasized mean climate; fewer examined extremes under stringent mitigation consistent with Paris-aligned carbon neutrality. This study fills that gap by isolating individual forcing contributions under SSP1-1.9 and by analyzing extreme temperature and precipitation indices.

The study uses the fully coupled Community Earth System Model version 1 (CESM1) with CAM5 atmosphere (1.9° × 2.5°, 30 vertical levels), CLM4 land, and POP2 ocean. CAM5 includes major aerosol species (BC, POM, SOA, sulfate, dust, sea salt) in four modal sizes and simulates aerosol-radiation and aerosol-cloud interactions (cloud albedo and lifetime effects). Aerosol convective transport and wet deposition schemes are modified to enhance performance. Three-dimensional ozone concentrations are prescribed; tropospheric O₃ dominates the radiative effects considered. The carbon neutrality pathway is represented by CMIP6 SSP1-1.9, targeting ~1.9 W/m² forcing by 2100 and limiting warming to ~1.5 °C. Five experiment sets isolate driver effects under quasi-equilibrium conditions using 3-member ensembles, each integrated at least 200 years with the last 100 years analyzed: (1) Baseline: GHGs, aerosol and precursor emissions, and O₃ fixed at 2020 levels. (2) GHG2050: GHGs at 2050 levels; aerosols/precursors and O₃ at 2020, isolating GHG effects by 2050. (3) AerGHG2050: GHGs and aerosols/precursors at 2050; O₃ at 2020, isolating aerosol effects by comparing to GHG2050. (4) ALL2050: GHGs, aerosols/precursors, and O₃ at 2050, isolating O₃ effects via comparison to AerGHG2050 and giving combined 2050 impacts. (5) ALL2100: all drivers at 2100 levels, giving combined end-century impacts. GHG concentrations (e.g., CO₂ 414 ppm in 2020 rising to 437 ppm in 2050 then 400 ppm in 2100; CH₄ declines from 1884 ppb in 2020 to 1429 ppb in 2050 and 1061 ppb in 2100; N₂O increases to 344 ppb in 2050 and 353 ppb in 2100). Aerosol/precursor emissions (SO₂, BC, OC) decrease substantially from 2020 to 2050 and 2100, with large reductions over polluted regions; AOD declines by 0.08–0.2 (up to >0.3 by 2100). Tropospheric O₃ is prescribed from ScenarioMIP multi-model means for 2020, 2050, 2100, with widespread decreases (TCO reductions 10–25% by 2100; surface O₃ declines up to >20 ppb in NH mid-to-high latitudes). Effective radiative forcing (ERF) at TOA and surface is diagnosed for attribution. Regional analyses use 21 land subregions. Extreme indices computed from daily fields include: heat waves and humid heat waves (frequency HWF, duration HWD, amplitude HWA using present-day thresholds), total precipitation on wet days (Pretotal, RR ≥ 1 mm/day), consecutive wet days (CWD), and heavy precipitation days (R10mm, RR ≥ 10 mm/day). Model present-day climatology is evaluated against ERA5 (2020). To contrast with a medium pathway, multi-model DAMIP analyses under SSP2-4.5 (SSP245, SSP245-GHG, SSP245-Aer) are used to compare relative contributions of GHGs vs aerosols by mid-century.

- Under SSP1-1.9 toward carbon neutrality, reductions in aerosols and their precursors dominate future warming and precipitation changes, far outweighing GHG and tropospheric O₃ effects. - Surface air temperature: In 2050, GHG-only changes (GHG2050) yield slight warming (<0.2 °C over most regions; maxima ~0.2 °C over Greenland). Adding aerosol reductions (AerGHG2050) produces strong warming with regional increases up to ~2.0 °C over NH mid-to-high latitudes relative to 2020. Regional mean warming due to aerosol reductions is ~0.5–1.4 °C, vs GHG-induced <0.2 °C; tropospheric O₃ changes cause slight cooling over most subregions (magnitudes < −0.2 °C). By 2100 (ALL2100), warming intensifies over NH mid-to-high latitudes and weakens over the tropics relative to 2050; net-negative CO₂ after 2050 does not offset warming from continued aerosol reductions. - Effective Radiative Forcing: TOA ERF changes due to GHGs in 2050 are weakly positive (~0.5–1.0 W/m² regionally), while aerosol reductions increase TOA ERF much more (up to ~4.5 W/m² over East Asia; ~2.5 W/m² over Northern Europe, North America, East Africa). Negative ERF anomalies (to −3 W/m²) occur over bright surfaces (e.g., Greenland, deserts) due to reduced absorbing aerosol (BC). Surface ERF patterns mirror TOA. Regional mean ERF changes from aerosols (±2.0 W/m²) exceed those from GHGs and O₃ (±0.5 W/m²). ERF anomalies strengthen by 2100 driven by further aerosol declines. - Precipitation: GHG-only changes mainly affect tropical oceans (enhanced Western Pacific rainfall). Aerosol reductions (AerGHG2050) amplify precipitation changes globally, increasing precipitation over NH (notably along the tropics) and decreasing over many SH oceans, indicating a northward ITCZ shift. Land precipitation increases up to ~0.3 mm/day over Southeast, East, and South Asia; GHG and O₃ impacts are weaker (±0.05 mm/day). Patterns intensify by 2100. - Humidity: Specific humidity changes are small in GHG2050 except north of 60°N, but increase markedly in AerGHG2050, with positive anomalies up to ~1 g/kg along the tropics, consistent with increased precipitation; O₃ impacts are minimal. - Model–MME comparison under SSP1-1.9: CESM1 reproduces spatial patterns similar to CMIP6 multi-model means for ALL2050 (spatial correlations >0.8 for temperature, >0.4 for precipitation), but with larger amplitudes. Global mean changes (2050 vs 2020): CESM1 temperature +0.85 °C, precipitation +0.07 mm/day (vs CMIP6 means +0.60±0.50 °C, +0.04±0.24 mm/day). For 2100 vs 2020: CESM1 +0.92 °C, +0.10 mm/day (vs +0.40±0.60 °C, +0.04±0.24 mm/day). - Extremes – heat waves: In GHG2050, heat wave frequency (HWF) <5 days/year, duration (HWD) <4 days/event, amplitude (HWA) <0.25 °C/day over most regions. With aerosol reductions (AerGHG2050), HWF >40 days/year, HWD >20 days/event, HWA >0.75 °C/day over large areas. ALL2050 similar, with O₃ slightly damping heat waves. By 2100, mean HWF >50 days/year and HWD >28 days/event globally; HWA up to ~1.5 °C/day in NH mid-to-high latitudes. - Extremes – humid heat waves: Aerosol reductions similarly amplify humid heat waves: humid-HWF >30 days/year, humid-HWD >24 days/event, humid-HWA >0.5 °C/day in AerGHG2050 vs much smaller in GHG2050. - Extremes – precipitation: In GHG2050, annual total precipitation on wet days (Pretotal) declines over many land regions, especially tropics, Northern Europe, North Asia. Aerosol reductions increase Pretotal with significant positive anomalies >40 mm over much of the globe, especially NH mid-to-high latitudes; GHGs and O₃ contribute weak negative changes in 2050. Total wet days (CWD) show similar behavior. Heavy precipitation days (R10mm) increase mainly in the tropics due to aerosol reductions, while GHGs and O₃ exert opposite, weaker effects. - Contrast with medium pathway: Under SSP2-4.5 (DAMIP), mid-century warming is dominated by GHGs (SSP245-GHG), with aerosol-only impacts much smaller (SSP245-Aer), opposite to the dominance of aerosol reductions under SSP1-1.9.

The study demonstrates that along a carbon neutrality pathway (SSP1-1.9), large reductions in aerosols and their precursors dominate radiative forcing changes and thus warming, precipitation shifts, and extremes—contrary to medium pathways where GHGs dominate. Aerosol declines increase ERF and unmask committed GHG-induced warming, leading to amplified heat waves (dry and humid) and enhanced precipitation, particularly over NH regions and with a northward ITCZ shift. The results link enhanced radiative forcing to thermodynamic responses such as increased atmospheric moisture following Clausius–Clapeyron scaling, driving heavier rainfall and more frequent, longer, and stronger heat waves. The findings highlight that co-pollutant reductions toward clean air can exacerbate warming and extremes unless accompanied by rapid and substantial GHG and O₃ precursor reductions to meet the 1.5 °C target and mitigate adverse impacts. The comparison with DAMIP under SSP2-4.5 underscores pathway dependence: aerosol-forcing dominance emerges under stringent mitigation with large aerosol declines, while GHGs dominate under medium scenarios.

Toward carbon neutrality (SSP1-1.9), aerosol reductions overtake GHG and tropospheric O₃ changes as the dominant driver of near-term warming, precipitation changes, and increases in extreme heat and heavy precipitation. This overturns expectations from medium scenarios and indicates that achieving climate goals requires deep, concurrent reductions in GHGs and O₃ precursors to counteract warming and extreme-event amplification from cleaner air. The study provides a quasi-equilibrium, component-wise attribution of climate responses using CESM1 and highlights strong aerosol-driven ERF increases and associated hydroclimate shifts. Future work should expand multi-model intercomparisons under SSP1-1.9, better constrain aerosol-cloud interactions and aerosol composition (absorbing vs scattering) effects, include interactive chemistry and aerosol–gas interactions, and assess transient versus stabilized responses to inform adaptation and mitigation planning.

Key limitations include: (1) Equilibrium simulations may underestimate the full equilibrium response to long-lived GHG changes; transient versus stabilized responses can differ, though main conclusions about aerosol dominance likely hold. (2) Model uncertainty: CESM1 has relatively high equilibrium climate sensitivity (4.1 °C) and stronger aerosol-cloud interaction ERF than multi-model means, yielding larger responses than CMIP6 averages. (3) Large uncertainties in aerosol ERF and aerosol–cloud interactions remain. (4) Ozone concentrations are prescribed; interactions between aerosols and gas-phase chemistry (e.g., heterogeneous reactions affecting O₃) are not represented. (5) Feedbacks of climate on air pollutants (and reciprocal feedback on climate) are neglected. (6) Opposing effects of absorbing vs scattering aerosol reductions are not fully disentangled; their relative future roles merit further investigation.