Climate change modulates the stratospheric volcanic sulfate aerosol lifecycle and radiative forcing from tropical eruptions

The study addresses how ongoing climate change will modulate the lifecycle and radiative forcing of stratospheric volcanic sulfate aerosols from explosive tropical eruptions. While volcanic eruptions are known to drive significant short-term climate variability via stratospheric sulfate aerosols, the feedback of a warmer climate on eruption plume dynamics, aerosol microphysics, transport, and resulting radiative impacts is less understood. Prior work has largely focused on how climate affects eruption frequency/magnitude or how background climate alters climate response to prescribed volcanic aerosol perturbations. This study aims to quantify, using interactive modeling, the impact of climate change on both the altitude of volcanic SO₂ injection and the subsequent aerosol lifecycle and radiative forcing for moderate- and large-magnitude tropical eruptions, comparing present-day (1990–2000) and a high-end future warming scenario (2090–2100, SSP5-8.5).

The authors categorize climate–volcano interactions into: (1) effects of climate on eruption occurrence/magnitude, with evidence for increased activity post-deglaciation but on millennial timescales; (2) modulation of climate response by background state, with mixed findings on whether future warming amplifies or dampens volcanic cooling and limitations due to prescribed aerosol properties; (3) direct effects on the sulfate aerosol lifecycle (transport, microphysics), which are largely unexplored. Earlier studies showed injection altitude sensitivity to climate but did not quantify radiative forcing or temperature impacts in warmer climates. Advances in interactive stratospheric aerosol modeling (e.g., GLOMAP-mode in UKESM1) and eruptive column models enable explicit simulation of aerosol lifecycle and radiative effects, motivating this integrated analysis.

The study couples a one-dimensional eruptive column (plume) model (Degruyter & Bonadonna, 2012) with the UK Earth System Model (UKESM1) in an atmosphere-only configuration (UM-UKCA with GLOMAP-mode interactive stratospheric aerosols). Two eruption cases at the Mount Pinatubo location (15.1°N, 120.4°E) are examined: a moderate-magnitude case injecting 1 Tg SO₂ with an eruption intensity of 1.3×10⁷ kg/s, and a large-magnitude case injecting 10 Tg SO₂ with intensity 2.7×10⁸ kg/s. For each case, ensembles of ten 3-year simulations are run for two climate states: historical (1990–2000, HIST) and future (2090–2100, SSP5-8.5, SSP585). A third set (SSP585_HIH) uses future climate but historical injection heights to isolate aerosol lifecycle effects from plume height changes. The eruptive column model computes SO₂ injection heights using UKESM1 meteorological profiles on the eruption day; SO₂ is subsequently injected in UKESM1 following a Gaussian vertical distribution centered at the computed height. Time-slice AMIP-style runs prescribe SST/sea ice and forcing agents; 15-year spin-ups precede 20-year control runs for anomalies. Key diagnostics include stratospheric aerosol optical depth (SAOD at 550 nm), top-of-atmosphere (TOA) and surface net radiative flux anomalies, sulfur species burdens, aerosol effective radius, nucleation rates, and Brewer-Dobson circulation proxies. Significance is assessed via Mann-Whitney U-tests (80% and 95% levels). E-folding times for SO₂, total S (SO₂+H₂SO₄), and tropical S burdens are obtained from exponential fits. To estimate surface temperature response, annual global-mean TOA forcing anomalies from UKESM1 are input to the FaIR simple climate model under preindustrial backgrounds to compute volcanic-induced surface temperature anomalies.

- Injection heights: For moderate-magnitude eruptions, SO₂ injection height remains near 16–17 km in both climates, but the tropical tropopause rises by ~1.5 km in SSP585, placing injections ~2 km below the tropopause. For large-magnitude eruptions, SO₂ injection height increases by ~1.5 km (from ~21 km in HIST to ~22.5 km in SSP585) due to reduced stratospheric stratification in the warmer climate. - Moderate-magnitude eruption radiative effects: Peak global-mean SAOD (550 nm) anomaly is reduced by a factor of 4 (≈75%) in SSP585 versus HIST. Ensemble mean 3-year integrated TOA radiative forcing is smaller by a factor of ~3 in SSP585 and becomes positive in SSP585_HIH. Mechanistically, a smaller fraction of injected sulfur reaches the stratosphere: ~60% after self-lofting in HIST vs ~25% one month post-eruption in SSP585. - Large-magnitude eruption radiative effects: Peak global-mean SAOD increases by ~10%; peak global-mean TOA net radiative forcing magnitude increases by ~30%; peak global-mean surface net radiative forcing increases by 18% (SSP585) to 35% (SSP585_HIH). Time-integrated SAOD increases significantly in year 1 but not over years 1–3 due to faster decay. Time-integrated TOA forcing increases by ~21% in year 1. Surface forcing increases by 16–27% for year 1 and years 1–3 across future scenarios. - Aerosol lifecycle changes (large eruptions): SO₂ e-folding time decreases by 23% (from 40±1 days in HIST to 31±2 days in SSP585/SSP585_HIH), likely due to higher OH from increased stratospheric water vapor. Total stratospheric S burden e-folding time decreases from 15.8±1.7 months (HIST) to 14.2±1.4 months (SSP585) and 12.2±1.0 months (SSP585_HIH). Effective radius peaks at ~0.41 µm (HIST) and is smaller by up to 11% (SSP585) and 18% (SSP585_HIH), increasing scattering efficiency and hence SAOD/RF in early months. Faster meridional transport (accelerated Brewer-Dobson circulation) speeds aerosol removal at mid-high latitudes, shortening lifetime; higher injection heights partly offset this by slower BDC with altitude. Early nucleation rates are 2–3× higher in warmer climate (colder stratosphere), yielding more numerous, smaller particles. - Climate response (large eruptions): Global-mean stratospheric warming (10–50 hPa) is amplified by ~55% in year 1 for both future scenarios and remains 55% larger over 3 years in SSP585 (with higher injection heights). Mid-tropospheric (400–700 hPa) cooling is amplified by ~80% in year 1 and by ~30% (SSP585_HIH) to ~50% (SSP585) over years 1–3 (marginal significance). Estimated global-mean surface cooling over years 1–3 increases by 7% (not significant) in SSP585_HIH and 15% (significant) in SSP585; year-1 surface cooling is amplified by ~21% in both. Ozone depletion is ~3× stronger in HIST than SSP585 due to higher CFCs historically.

The results demonstrate that climate change distinctly modulates volcanic sulfate aerosol forcing depending on eruption magnitude. For moderate eruptions, a higher future tropopause without increased plume heights leads to reduced stratospheric sulfur injection, strongly damping SAOD and radiative forcing. For large eruptions, accelerated Brewer-Dobson circulation and, to a lesser extent, higher injection heights decrease aerosol size and enhance first-year SAOD and radiative forcing, despite shorter aerosol lifetimes. Consequently, post-eruption stratospheric warming and tropospheric/surface cooling are amplified in a future climate. These findings emphasize the need to account for both plume dynamics and aerosol lifecycle changes when assessing volcanic impacts under climate change, challenging approaches that prescribe fixed aerosol properties or constant volcanic forcing in projections. Implications extend to geoengineering efficacy, where circulation-driven changes in aerosol size and lifetime can modulate forcing. A back-of-the-envelope assessment suggests that, on centennial scales, enhanced forcing from large eruptions (≈−0.03 W/m²) may be outweighed by reduced forcing from moderate eruptions (≈+0.08 W/m²), though this latter decrease is likely overestimated; with more realistic assumptions, the moderate-eruption forcing reduction could be ≈−0.005 W/m², implying a net increase in volcanic forcing from large events. Fully resolving the net effect requires realistic eruption statistics and coupled ocean–atmosphere feedbacks, as ocean stratification changes may further amplify surface cooling.

The study integrates eruptive column modeling with an interactive stratospheric aerosol–climate model to quantify how future warming alters volcanic sulfate aerosol lifecycle and radiative forcing. Key contributions are: (1) moderate tropical eruptions will produce much smaller stratospheric aerosol burdens and forcing in a warmer climate due to increased tropopause height without higher plume tops; (2) large tropical eruptions will yield enhanced first-year SAOD and radiative forcing, and amplified stratospheric warming and surface cooling, primarily via smaller aerosol sizes from an accelerated Brewer-Dobson circulation and higher injection heights. These insights underscore that climate–volcano feedbacks depend on eruption magnitude and that future projections should not assume constant volcanic forcing. Future work should include co-injection of ash/halogens/water, fully coupled ocean–atmosphere simulations to capture ocean feedbacks, expanded analysis to extratropical and effusive eruptions, inter-model comparisons of aerosol microphysics, and embedding plume dynamics within climate models to better resolve injection height variability.

- Only SO₂ is injected; co-emission of ash, halogens, and water—known to affect SO₂ lifetime, nucleation, and self-lofting—are omitted. - Atmosphere-only simulations with prescribed SST/sea ice preclude coupled ocean feedbacks; prior work suggests ocean stratification changes could substantially modify surface temperature responses. - Eruptive column heights are computed with a 1D model subject to uncertainties in entrainment and water phase change parameterizations; sub-daily atmospheric variability and feedbacks of the early plume on ambient conditions are not represented. - Single eruption location (Pinatubo) and fixed eruption date (July 1) limit sampling of seasonal and geographic variability. - Results may depend on aerosol microphysics representation; effective radius responses vary across models. - Tropospheric aerosol background changes are small in the chosen scenarios but could differ in other models/scenarios.