Atmos. Chem. Phys., 19, 7151–7163, 2019 https://doi.org/10.5194/acp-19-7151-2019 © Author(s) 2019. This work is distributed under the Creative Commons Attribution 4.0 License. Implication of strongly increased atmospheric methane concentrations for chemistry–climate connections Franziska Winterstein1, Fabian Tanalski1,a, Patrick Jöckel1, Martin Dameris1, and Michael Ponater1 1Deutsches Zentrum für Luft- und Raumfahrt (DLR), Institut für Physik der Atmosphäre, Oberpfaffenhofen, Germany anow at: MERPH-IP Patentanwälte PartG mbB, Munich, Germany Correspondence: Franziska Winterstein ([email protected]) Received: 16 January 2019 – Discussion started: 18 January 2019 Revised: 6 May 2019 – Accepted: 13 May 2019 – Published: 29 May 2019 Abstract. Methane (CH4) is the second-most important di- 1 Introduction rectly emitted greenhouse gas, the atmospheric concentra- tion of which is influenced by human activities. In this study, Methane (CH4) is a potent greenhouse gas (GHG), subject to numerical simulations with the chemistry–climate model strong anthropogenic emissions that contribute substantially (CCM) EMAC are performed, aiming to assess possible con- to global warming. It is not just a radiatively active gas by sequences of significantly enhanced CH4 concentrations in itself but is chemically active as well, strongly influencing the Earth’s atmosphere for the climate. the chemical composition of the atmosphere. Beyond that, We analyse experiments with 2×CH4 and 5×CH4 present- its sources are prone to temperature changes and it is gen- day (2010) mixing ratio and its quasi-instantaneous chem- erally expected that climate change (i.e. surface warming) ical impact on the atmosphere. The massive increase in will lead to enhanced CH4 emissions, accelerating the tem- CH4 strongly influences the tropospheric chemistry by re- perature rise. For instance, additional CH4 emissions are ex- ducing the OH abundance and thereby extending the CH4 pected from wetlands due to climate-driven changes (Gedney lifetime as well as the residence time of other chemical et al., 2004; Zhang et al., 2017; Ma et al., 2017). Moreover, substances. The region above the tropopause is impacted a large quantity of CH4 is stored as methane hydrate, not by a substantial rise in stratospheric water vapour (SWV). only in permafrost soil but also in the seafloor. Permafrost The stratospheric ozone (O3) column increases overall, but soil stores about a 100-fold of the current CH4 burden in SWV-induced stratospheric cooling also leads to a enhanced the atmosphere, and oceanic methane hydrates store even ozone depletion in the Antarctic lower stratosphere. Regional a 1000-fold (IPCC, 2013). Current estimates indicate that patterns of ozone change are affected by modification of GHG emissions from thawing permafrost soils could repre- stratospheric dynamics, i.e. increased tropical upwelling and sent a major terrestrial biogeochemical feedback to climate stronger meridional transport towards the polar regions. We change over the coming decades (Comyn-Platt et al., 2018). calculate the net radiative impact (RI) of the 2 × CH4 exper- At the same time, it is under debate whether a possible −2 iment to be 0.69 W m , and for the 5 × CH4 experiment to strong release of CH4 from thawing permafrost in the Arctic be 1.79 W m−2. A substantial part of the RH is contributed region could potentially force an abrupt climate change (as by chemically induced O3 and SWV changes, in line with discussed by O’Connor et al., 2010). At present, the release previous radiative forcing estimates. of methane hydrate from reservoirs is highly uncertain as To our knowledge this is the first numerical study using a well as the magnitude of future natural and anthropogenic CCM with respect to 2- and 5-fold CH4 concentrations and emissions of methane. Increasing surface temperatures it is therefore an overdue analysis as it emphasizes the im- cause enhanced CH4 emissions from thawing permafrost pact of possible strong future CH4 emissions on atmospheric soils to the atmosphere, but the amount is currently poorly chemistry and its feedback on climate. constrained (Hayes et al., 2014; Schaefer et al., 2014; Koven et al., 2015; Schuur et al., 2015). For instance, Dean et al.(2018) stated that there is basically no significant Published by Copernicus Publications on behalf of the European Geosciences Union. 7152 F. Winterstein et al.: Implication of strongly increased atmospheric methane concentrations increase in Arctic methane emissions at the moment, though ment resulted in a value near zero. This example even more they may increase towards the end of the 21st century. motivates an assessment of simulations that include chemi- Nevertheless, permafrost thaw could potentially release cally driven atmospheric adjustments to increases in CH4. trapped CH4 and transform frozen soil to wetland areas, To our knowledge, studies using data derived from which would then add to Arctic CH4 emissions. Moreover, chemistry–climate model (CCM) simulations, including ex- ongoing heating of the Arctic sea surface temperature (SST) treme CH4 emissions (i.e. beyond current and near-future will also enhance future CH4 production in the ocean, and amounts), are not available so far. A CCM is an atmospheric a reduction in sea ice concentration (SIC) may increase the global circulation model that is interactively coupled to a de- direct transfer of CH4 from the ocean to the atmosphere. tailed chemistry module. In contrast to CTMs, in CCMs the In particular, enhanced SST can increase the production of simulated concentrations of the radiatively active gases are CH4, as permafrost underlying the continental shelf begins used for the calculations of net heating rates. Changes in the to thaw (Miller et al., 2018). How a changing climate will abundance of these gases due to chemistry and advection in- impact future CH4 emissions remains a topic of debate in fluence heating rates and, consequently, variables describing atmospheric science, since emissions from the most climate- atmospheric dynamics. This creates a dynamical–chemical sensitive CH4 sources, i.e. wetlands, are difficult to quantify coupling in which the chemistry influences the dynamics and precisely. vice versa. Since CH4 influences other trace gases due to its Although there remain important knowledge gaps about oxidation products as well as the removal of the hydroxyl the magnitude of CH4 emissions, it is important to improve radical (OH), a comprehensive chemistry module is neces- our understanding of how strongly future CH4 emissions sary. In simulations with doubled carbon dioxide (CO2), in may impact our atmosphere and the environment. About contrast, the feedback on climate and chemistry is induced 90 % of the emitted CH4 is removed in the troposphere. A only by its radiative impact. Apart from accounting for the change in tropospheric CH4 concentration affects the oxidiz- direct radiative impact of CH4, the use of a CCM is strongly ing capacity of the atmosphere, modifies ozone in the tropo- desired, since the atmospheric CH4 chemical feedback is a sphere and influences the CH4 lifetime itself (e.g. Saunois key process for understanding the variations in atmospheric et al., 2016; Frank, 2018; Holmes, 2018). Additionally, it CH4 and its effects on other chemical constituents of the at- affects the stratosphere. For example, enhanced CH4 emis- mosphere (Holmes, 2018). sions will lead to an abundance of stratospheric water vapour The present work is the first study investigating atmo- (SWV) and, as a consequence, will strongly influence strato- spheric effects due to strong CH4 emissions with such a spheric ozone (O3)(Stenke and Grewe, 2005; Revell et al., CCM. Idealized simulations of significantly enhanced CH4 2016). concentrations are performed, i.e. 2-fold (2 × CH4) and To assess the direct and indirect effects of strongly en- 5-fold (5 × CH4) enhanced CH4 concentrations compared hanced CH4 emissions on atmospheric composition and to present-day condition, allowing possible future conse- Earth’s climate, numerical model studies are able to sup- quences for atmospheric composition to be assessed while port investigations such as identifying potential signatures considering chemical feedback processes. In a first step, we impacting climate change. So far only a limited amount of conducted CCM simulations without interactive ocean cou- numerical studies are available concerning the impact of very pling, i.e. the surface conditions regarding SST and SIC strong CH4 emissions. Exemplary, the effect of 2-fold CH4 are prescribed (suppressed surface temperature feedback). was investigated in a 1-D radiative–convective climate model Equivalent to the work of Smith et al.(2018), the results by Owens et al.(1982) and by MacKay and Khalil(1991). can be interpreted as rapid adjustments to a sudden CH4 en- Shang et al.(2015) used a chemistry transport model (CTM) hancement before the ocean reacts to the perturbation, which but doubled CH4 emission over China only. Other CTM stud- would occur on a far larger timescale. ies have focused on recent changes and fluctuations in the In this study we will use the ECHAM/MESSy Atmo- atmospheric CH4 concentration (e.g. Dalsøren et al., 2016) spheric Chemistry (EMAC) CCM (Jöckel et al., 2016), as- or have tried to explain CH4 trends, which is a challenge be- sessing the range of atmospheric responses by abrupt in- cause of important uncertainties in the global CH4 budget, creases in CH4 concentrations. A short description of EMAC i.e. the balance of surface sources and atmospheric and sur- is given in Sect.2, as well as an explanation of the simula- face sinks (Saunois et al., 2016). Furthermore, CTMs are lim- tion strategy. In Sect.3 the reference simulation representing ited in assessing climate-change-related issues, because they near-present-day condition is briefly evaluated with obser- do not include the feedback between chemistry and dynam- vations (Sect. 3.1) and a discussion of the impact of 2-fold ics. Smith et al.(2018) investigated the fast radiative feed- and 5-fold increased CH4 concentrations in respective sce- backs (adjustments) in a model intercomparison using simu- nario simulations is presented in Sect.