A known adverse side effect of stratospheric aerosol modification (SAM) is the alteration of the quasi-biennial oscillation (QBO), which is caused by the stratospheric heating associated with an artificial aerosol layer.
Multiple studies found the QBO to slow down or even completely vanish for point-like injections of
Stratospheric aerosol modification (SAM) by the artificial injection of sulfur dioxide (
Besides backscattering ISR, sulfate aerosols also absorb parts of the outgoing tropospheric longwave radiation (OTLR) and the incoming near-infrared radiation (NIRR).
The absorption of OTLR and NIRR causes a significant warming of the lower tropical stratosphere
Multiple studies revealed that the QBO could also be heavily perturbed during a potential deployment of SAM (e.g.,
Together with equatorial point injections, a modification of the QBO has been also noticed for other injection strategies.
Additionally,
To overcome this limitation, in this study we investigate the QBO response to three different injection locations for the same models as used by
In Sect.
MAECHAM5 is the middle atmosphere version of the spectral general circulation model (GCM) ECHAM5
MAECHAM5 was interactively coupled to the prognostic modal aerosol microphysical model HAM
The Community Earth System Model version 2 (release 2.1) in the Whole Atmosphere Community Climate Model version CESM2(WACCM6) is a state-of-the-art fully coupled climate model, which is also used in the new CMIP6 simulations
Although the standard version of WACCM6 uses comprehensive chemistry from the troposphere to the lower thermosphere, the version used here only simulates middle atmospheric (stratosphere, mesosphere and lower thermosphere) chemistry, with 98 simulated chemical species.
Sulfate aerosols are treated using the Modal Aerosol Model (MAM4) as described in
The experimental setup of the simulations performed in this study is in accordance with the proposal of the GeoMIP6 test bed experiment
With ECHAM, three different injection strategies have been simulated for both injection species (
All simulations were performed for a period of 10 years.
If not otherwise stated, the results presented in this study are averaged over the last 8 years of the respective simulation, since
Setup of all performed simulations.
The point injections have been performed in a single equatorial grid box centered at 1.4
ECHAM simulates the QBO well in the control simulation (Fig.
Time–height cross sections of the 5
For an injection of
The dynamic mechanisms which cause the observed modification and breakdown of the QBO for an equatorial point injection of
In this section, we will investigate the reasons for the different QBO responses to the three tested injection strategies exemplarily based on an injection of
Additionally, we are aware of the fact that the QBO may also change due to a modified wave driving.
However, we found no significant changes in QBO wave driving in our simulations (not shown), which is in agreement with earlier studies (e.g.,
The artificial sulfate aerosols heat the lower stratosphere by the absorption of OTLR and NIRR, whereby the location and magnitude of this heating strongly correlate with those of the sulfate mass mixing ratio
Latitude–height cross section of the zonal mean net aerosol heating rate
This aerosol-induced heating results in a significant positive temperature anomaly centered at the Equator (Fig.
The warming of the lower stratosphere is the primary perturbation induced by the sulfate aerosols, as indicated by the good agreement of the sulfate mass mixing ratio, the net aerosol heating rates, and the temperature anomalies. All changes in dynamics – including the QBO – are obviously induced by this initial warming in a second step.
Opposite to the lower-stratospheric warming, statistically significant negative temperature anomalies are located in the middle and upper tropical stratosphere for all three injection strategies (Fig.
Following
Latitude–height cross section of the anomaly of the zonal mean residual vertical velocity
Our simulations confirm that
The reason for the increase of
In the upper stratosphere (i.e., between 20 and 3
Within the TEM framework, the characteristics of the general acceleration of the BDC can be further directly linked to the tropical confinement of the aerosol-induced temperature anomaly, as shown by
Consequently, it is ultimately the meridional shape of the aerosol-induced lower-stratospheric temperature anomaly or – simply spoken – the degree of tropical confinement of the artificial sulfate aerosols that determines the QBO response to artificial sulfate injections.
Besides the increase of the tropical upwelling in the rising branch of BDC,
Latitude–height cross section of the zonal mean temperature
Latitude–height cross section of the anomaly of the meridional zonal mean temperature gradient
Vertical profile of the 5
However, despite the fact that QBO changes due to artificial sulfur injections are frequently interpreted as a consequence of an increased residual tropical upwelling and a modification of thermal wind balance, one can not see both as two separate processes.
In contrast, the acceleration of the BDC discussed in Sect.
As discussed in Sect.
For region-so2-25, the QBO was also found to be locked in a permanent westerly phase, but the vertical extent as well as the strength of the westerlies is weaker than for point-so2-25, which is in agreement with the results of
For 2point-so2-25, the QBO was not found to be modified significantly, and it basically preserved its natural periodicity (Fig.
As discussed in Sect.
Our results clearly show that differences in the QBO response with respect to our three tested injection strategies are linked to differences in the meridional structure of the aerosol-induced temperature anomaly.
Therefore, the absolute strength of the aerosol-induced lower-stratospheric temperature anomaly does not permit a statement about the strength of the QBO modification when comparing different injection strategies.
For instance, the tropical (i.e., 5
Since it is the degree of tropical confinement of the artificial sulfate aerosols that is ultimately decisive for the observed QBO response also when explaining the observed QBO changes solely as the consequence of an increased residual tropical upwelling, we will use thermal wind balance in our argumentation throughout this study as it directly links the observed QBO changes to the observed aerosol-induced temperature anomalies.
For the point and the region injection strategies, the QBO was found to be impacted much less in our experiments with an injection rate of 5
Vertical profile of the 5
For all three tested injection strategies, the response of the QBO is in principle independent of the injection species –
However, for the point and region injection strategies, the modification of the QBO was found to be slightly stronger with respect to the strength and the vertical extent of the lower-stratospheric westerlies when injecting
Zonal mean artificial sulfate burden for the ECHAM simulations of the
Vertical profile of the 5
Global mean aerosol size distributions focusing on
Global mean top-of-atmosphere (TOA) all-sky net RF exerted by artificial sulfate aerosols as a function of injection rate.
The dotted black line marks a RF of
The reason for the higher sulfate burden obtained for an injection of
Both models simulate a reasonable QBO in the control simulation (Fig.
In the following two sections, the QBO response to the 2point and region injections will be compared for ECHAM and CESM based on the injection of
Time–height cross sections of the 5
Latitude–height cross sections of the anomaly of the meridional zonal mean temperature gradient
For the 2point injections of
Zonal mean artificial sulfate burden for the 2point injections
Nevertheless, the QBO responds qualitatively similar to a 2point injection of
Following
In contrast to the 2point injections, the response of the QBO to a region injection of
For ECHAM, the results are explained by the weakening of the usually positive poleward
However, Fig.
We assume that the significant difference in the QBO response to a region injection of
Latitude–height cross section of the ozone concentration anomaly for the simulations of an
However, based on our analysis we cannot fully explain why the QBO is locked in a strong permanent easterly phase in CESM. The lower-stratospheric ozone depletion and the upper-stratospheric ozone increase alone may only partly account for this substantial difference between both of our models. Most likely, differences in the SAM-induced changes of the resolved and parameterized wave forcing of the QBO may explain its different responses to SAM in both models. Additionally, differences in the GW parameterization of both models itself are likely to account to the observed differences, as they are tuned to represent the QBO realistically in the current climate but may react very differently to an external forcing like artificial sulfate aerosols.
For the 2point injections, changes in stratospheric ozone levels are mostly located outside the equatorial region (Fig.
The ozone changes observed in the CESM simulations are consistent with previous simulations with the older version of the same model (CESM1(WACCM); see, for instance,
Within this study, we performed several simulations with the GCMs ECHAM and CESM to comprehensively compare the response of the QBO to different SAM setups with regard to the injection strategy, the injection rate, and the injection species.
Thereby, we aimed at a deeper investigation of the reasons for structural differences in the QBO response to different SAM setups.
We identified the following key characteristics of the QBO response to SAM:
The QBO response to SAM does fundamentally depend on the injection strategy. The injection rate and species rather act to scale the strength of this response. We clearly identified the meridional structure of the aerosol-induced temperature anomaly within the lower tropical stratosphere instead of its absolute strength as the key parameter explaining the observed different responses of the QBO to our different injection setups. For the equatorial point and for the region injections, the aerosol warming peaks more or less sharply at the Equator, causing a weakening of the poleward In contrast,
Obviously, linking the QBO response to artificial sulfur injections to the meridional shape of the aerosol-induced temperature anomaly offers us the possibility to explain the fundamentally different responses of the QBO to all of our three injection strategies simulated with ECHAM in a stringent manner.
This is a clear advancement compared to earlier studies, e.g.,
Therewith, our results indicate that the modification of thermal wind balance in the lower tropical stratosphere between 40 and 80
An increase of the injection rate from 5 to 25
Compared to ECHAM, we found the QBO to be much more sensitive to artificial sulfur injections in CESM for the 2point and region injections.
Nevertheless, changes in ozone and the associated SW heating alone cannot explain the substantial differences of the QBO response to a region injection between ECHAM and CESM.
Besides differences in the representation of aerosol microphysics and in horizontal and vertical resolution, we think that differences in the GW parameterization most likely explain why the QBO responds so differently to a region injection.
Primary data and scripts used in this analysis and other supplementary information that may be useful in reproducing the author's work are available:
The supplement related to this article is available online at:
HF performed the ECHAM simulations with strong support from UN. DV performed the CESM simulations. HF wrote the paper with contributions from DV on CESM description and results. All authors discussed the idea and results of this study.
The authors declare that they have no conflict of interest.
This article is part of the special issue “Resolving uncertainties in solar geoengineering through multi-model and large-ensemble simulations (ACP/ESD inter-journal SI)”. It is not associated with a conference.
Henning Franke wants to thank Ulrike Niemeier and Stefan Bühler (University of Hamburg, Germany) for enabling him to work on this very exciting master's thesis project, for the excellent supervision, and for the possibility to publish parts of his results in this paper. We thank Yaga Richter (NCAR) for providing input data, Debra Weisenstein and David Keith (Harvard) for giving valuable input and tips on the simulation setup as well as for giving the impulse for this project, Marco Giorgetta (MPI-M) for giving very helpful comments on the first manuscript, and three anonymous reviewers for their helpful suggestions. ECHAM simulations have been performed on the computer of Deutsches Klimarechenzentrum (DKRZ). Ulrike Niemeier obtained support from the German DFG-funded Research Unit VollImpact FOR2820 subproject TI344/2-1 and DFG-funded Priority Program “Climate Engineering: Risks, Challenges, Opportunities?” (SPP 1689). The CESM project is supported primarily by the National Science Foundation. This work was supported by the National Center for Atmospheric Research, which is a major facility sponsored by the National Science Foundation under cooperative agreement no. 1852977. Support for Daniele Visioni was provided by the Atkinson Center for a Sustainable Future at Cornell University and by the National Science Foundation through agreement CBET‐1818759.
This research has been supported by the Deutsche Forschungsgemeinschaft (grant nos. FOR2820, TI344/2-1, and SPP1689) and the National Science Foundation (grant nos. CBET-1818759 and 1852977). The article processing charges for this open-access publication were covered by the Max Planck Society.
This paper was edited by Timothy J. Dunkerton and reviewed by three anonymous referees.