Chemistry-climate feedback of atmospheric methane in a methane emission flux driven chemistry-climate model

Stecher, Laura; Winterstein, Franziska; Jöckel, Patrick; Ponater, Michael; Mertens, Mariano; Dameris, Martin

doi:https://doi.org/10.5194/egusphere-2024-2938

Preprints

https://doi.org/10.5194/egusphere-2024-2938

Preprints

01 Oct 2024

| 01 Oct 2024

Chemistry-climate feedback of atmospheric methane in a methane emission flux driven chemistry-climate model

Laura Stecher, Franziska Winterstein, Patrick Jöckel, Michael Ponater, Mariano Mertens, and Martin Dameris

Abstract. The chemical sink of atmospheric methane (CH₄) depends on the temperature and on the chemical composition. Here, we assess the feedback of atmospheric CH₄ induced by changes of the chemical sink in a warming climate using a CH₄ emission flux driven setup of the chemistry-climate model EMAC, in which the chemical feedback of CH₄ mixing ratios can evolve explicitly. We perform idealized perturbation simulations driven either by increased carbon dioxide (CO₂) mixing ratios, or by increased CH₄ emission fluxes. The CH₄ emission flux perturbation leads to a large increase of CH₄ mixing ratios. Remarkably, the factor by which the CH₄ mixing ratio increases is larger than the increase factor of the emission flux, because the atmospheric lifetime of CH₄ is extended.

In contrast, the individual effect of the global surface air temperature (GSAT) increase is to shorten the CH₄ lifetime, which results in a significant reduction of CH₄ mixing ratios in our setup. The corresponding radiative feedback is estimated at -0.041 W/m²/K and -0.089 W/m²/K for the CO₂ and CH₄ perturbation, respectively. The explicit adaption of CH₄ mixing ratios leads to secondary feedbacks of the hydroxyl radical (OH) and ozone (O₃). Firstly, the OH response includes the CH₄-OH feedback, which enhances the CH₄ lifetime change, and, secondly, the formation of tropospheric O₃ is reduced. Our CH₄ perturbation induces the same response of GSAT per effective radiative forcing (ERF) as the CO₂ perturbation, which supports the applicability of the ERF framework for CH₄.

Received: 23 Sep 2024 – Discussion started: 01 Oct 2024

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 1557 KB)

Supplement (2830 KB)

Download & links

Laura Stecher, Franziska Winterstein, Patrick Jöckel, Michael Ponater, Mariano Mertens, and Martin Dameris

Status: final response (author comments only)

CC1: 'Comment on egusphere-2024-2938', Zosia Staniaszek, 02 Oct 2024

Thanks for a really interesting read and an important step in developing more models with methane emissions. The section on tagging the ozone response is particularly novel in a methane emissions driven system and really illustrates the wide ranging impact of changes in methane (that are often excluded by using a lower boundary condition).

The increase in mixing ratio above the level of increase in emissions is as expected due to the chemical feedbacks in the system and the lengthening of methane lifetime, albeit here the feedback factor is higher than most models (1.73 as a crude estimate using delta conc/delta emissions (Holmes et al 2018)). I would be interested in whether you could calculate a feedback factor for EMAC using these simulations. Also, a note that more recent feedback factors than those quoted here can be found in Sand et al 2023 Supplementary Table 2 (https://www.nature.com/articles/s43247-023-00857-8).

I would also be interested in the timescale of the change in methane mixing ratio, and the perturbation lifetime. In the methods section you mention a long spin up time. In UKESM-ems we found that with a large methane emissions decrease you get a much faster than expected change in mixing ratio due to the rapid increase in OH and decrease in methane lifetime, and I would expect the opposite effect in an increase such as done here.

Best wishes,

Zosia Staniaszek

Citation: https://doi.org/10.5194/egusphere-2024-2938-CC1
RC1: 'Comment on egusphere-2024-2938', Anonymous Referee #1, 05 Nov 2024

Stecher et al. (2024) assess the methane feedback on chemistry and climate by conducting chemistry-climate model simulations driven by the methane emissions. As the second most important greenhouse gas after CO2, it is critical to understand the methane cycle from the climate impact and mitigation perspectives. Several sets of model simulations are performed to assess the radiative and chemistry feedback in this work. The results show larger sensitivity of the methane lifetime towards climate change compared to previous studies. The impacts on O3 chemistry are also discussed thoroughly in a tagging approach. The manuscript is well written and organized. I recommend accepting the manuscript after minor revision.

General comments:
Ocean model coupling

A mixed layer ocean model is coupled in the REF-SSTvar simulations. How representative are the SSTs in these runs compared to the prescribed SSTs? This would have impacts on air-sea interactions and thus affecting atmospheric chemistry and composition in addition to the GSAT. The biases in reference cases may propagate into the perturbed cases, which would affect your assessment in the climate response. Do you think the biases could be canceled out between the reference run and perturbed run?

Methane radiative feedback
The simulations between reference cases and perturbed cases suggest larger sensitivity of the methane lifetime towards climate change compared to previous studies and the authors attribute such difference to the different methane representations in the models, e.g., emission driven vs prescribed lower boundaries. But the authors also point out in the manuscript, the model-specific parameterizations, mechanisms, etc would lead to the model-dependent results. For example, using different radiative transfer models give very different radiative feedback as shown in the manuscript. How to better quantify the sensitivity to different model setup or model schemes? How to improve the model confidence in assessing methane radiative feedback?

Ozone

There are a lot of discussions in the manuscript on the impacts on tropospheric and stratospheric ozone. What about impacts on surface O3, which is more relevant to the health effects.

Specific comments:
Page 9, Section 3.1, line 265-267, How sensitive of OH levels to lightning NOx? Do changes in lightning NOx play a role here?
Page 13, line 375-380: How do you treat N2O in the model? Does your model read N2O emissions or prescribe N2O at lower boundaries?
Figure S2 shows the difference in the specific humidity. Is specific humidity in your model also affected by chemistry? In other words, is water vapor a prognostic chemical tracer explicitly involved in the chemical reactions?

Citation: https://doi.org/10.5194/egusphere-2024-2938-RC1
RC2:
'Comment on egusphere-2024-2938', Anonymous Referee #2, 02 Dec 2024
General Comments:
This study by Stecher et al. makes use of novel methane flux-driven capability in the atmosphere-only and coupled chemistry-climate model EMAC to quantify the impact of a carbon dioxide concentration perturbation and a methane flux perturbation on atmospheric composition, taking account of feedbacks on methane via changes in methane lifetime. The study also examined the radiative effects from these impacts.
I would like to congratulate the authors on the new and exciting capabilities in the EMAC model that are exploited in this study. The methane flux-driven capability is a major step change in capability, which is only available in a small number of models worldwide. The study also makes use of ozone tagging, to attribute the changes in ozone to different sources (lightning vs stratosphere vs biogenic, etc.). And I also particularly liked the analysis in which the authors compared the fast and slow responses in atmosphere only simulations versus coupled simulations.
Overall, I found the manuscript to be very well written. It provides unique insight into atmospheric composition responses – both fast and slow – due to both co2 and methane perturbations, and their radiative effects. The manuscript would provide a positive contribution to the scientific literature and is an excellent fit for Atmos. Chem. Phys. Therefore, I fully recommend publication, subject to minor changes in response to the specific and technical comments documented below.
Specific Comments:
In Section 2.1, you state that chlorine and bromine halogen chemistry is included and that oxidation of methane by chlorine is considered. Can you include some clarification on whether the chlorine sink is only relevant in the stratosphere and/or whether the methane sink in the marine boundary layer through tropospheric halogen chemistry is included?

In Section 2.1, what criteria were used to determine whether “a quasi-equilibrium is reached”?

Line 240: You state that the radiative effects of ozone and water vapour are calculated separately for the troposphere and stratosphere – can you include what definition you use for the tropopause?

Section 3.1, line 262: You mention that the lifetime with respect to OH oxidation is reduced when the model is allowed to respond to the CO2 perturbation. Although stratospheric oxidation is a more minor sink for methane than tropospheric oxidation by OH, I wondered whether you could also diagnose the stratospheric lifetime

Section 3.1, line 269: Given the lack of significant differences in the oxidants, you hypothesize that the change in methane concentrations in the lower stratosphere are due to transport, i.e., because of reduced tropospheric concentrations and a more efficient Brewer Dobson circulation. Do you have any mass flux diagnostics that can support that statement?

Section 3.1: It is interesting that the sensitivity of the methane lifetime to climate change in EMAC appears to be stronger than in Voulgarakis et al. and Thornhill et al. I'm not convinced that it is due to methane being more fully interactive here, and it would make an interesting follow-up study to try to unpick the reasons behind these model differences.

Section 3.1, lines 303-305: You state that “, the explicit treatment of the CH4 feedback in our set-up allows for a subsequent feedback of OH and correspondingly for a self-feedback on the CH4 lifetime, which can explain the enhanced sensitivity of the CH4 lifetime towards climate change.” Have you verified in this setup that if methane was driven by concentration-based boundary conditions that the sensitivity of methane lifetime to temperature would be more comparable to that in other models?

Section 3.1, lines 306-310: Here, you argue that the model response in EMAC is more consistent with f=1 than estimates of f = [1.2, 1.4]. I wonder how representative the range of 1.2-1.4 is for the EMAC model. Do you know what the feedback factor from EMAC is in concentration-driven simulations or even from your ch4 flux perturbation simulation?

Section 3.3: Here you state that Table 4 shows the total SARF, ERF, ∆GSAT and the associated climate sensitivity parameters λ, as well as individual radiative effects corresponding to the composition changes of CH4, O3 and stratospheric H2O. You then go on to compare SARF and ERF for the co2 and ch4 perturbation simulations. In the case of the CH4 perturbation simulation, the ERF is a factor of 3 larger than the SARF. I think it’s important to state upfront that the SARF here is only capturing the direct radiative effect of CH4 alone, whereas the ERF captures the radiative effects from the ch4-driven chemical adjustments (e.g., ozone, stratospheric water vapour).

Section 3.3: Here, the radiative effect of stratospheric water vapour from the methane flux perturbation experiment seems to be nearly comparable in magnitude with that from ozone. This doesn’t appear to be consistent with the relative contributions from water vapour and ozone in the present-day forcing from methane. Can you comment further on its radiative effect?

Technical Comments:
Abstract, line3: Change “we assess the feedback of atmospheric CH4” to “we assess the feedback on atmospheric CH4”; similar comment for line 12: “secondary feedbacks of the hydroxyl radical”

Abstract, line 11: I’m not certain that the word “adaption” is appropriate here. In fact, I ended up looking it up and found this: Adaption vs. Adaptation - What’s the Difference? I’m not convinced that adaptation is the right word either but you could use words such as “evolve”, “respond”, “change”, “adjust”

Introduction, line 22: Change “most important sink of atmospheric CH4 is the oxidation with” to “most important sink of atmospheric CH4 is oxidation by”

Introduction, line 23; Change “underlying the CH4 oxidation” to “underlying CH4 oxidation”

Introduction, line 25: Change “indirect contributions of ozone” to “indirect contributions from ozone”

Introduction, line 28: Change “involving CH4 emission reduction” to “involving CH4 emission reductions”

Introduction, line 30: Change “effects on O3 and H2O, the CH4 oxidation” to “effects on O3 and H2O, CH4 oxidation”; Please make similar changes for other occurrences (e.g., line 34, etc.)

Introduction, line 41: Change “Up to date, only a limited number” to “To date, only a limited number”

Introduction, line 45: Change “which is endorsed” to “which was endorsed”

Introduction, line 46: Give full name for IPCC

Introduction, lines 49-51: The phrase “. On the one hand, the resulting CH4 response in turn alters the atmospheric CH4 lifetime, which leads to subsequent adaptions of the CH4 mixing ratios.” seems awkward and should be re-phrased. I’m wondering whether something like “When CH4 mixing ratios are free to evolve, the resulting CH4 response alters the atmospheric CH4 lifetime, which in turn leads to further changes in the CH4 mixing ratios.” would be better but open to other suggestions too!

Introduction, lines 64-65: change “The magnitude of this feedback, however, has been highly model dependent.” To “The magnitude of this feedback, however, is highly model dependent.”

Introduction, line 67: Change “resulting from changes in the chemical sink.” To “resulting from changes in the chemical sink and/or methane emissions.”

Introduction, line 71: Change “the change of CH4 mixing ratios” with “the change in CH4 mixing ratios”

Introduction, line 71: Change “Next to the CH4 feedback” to “In addition to the CH4 feedback”

Methods, line 118: Give full names for NOAA/ESRL

Methods, line 133: Change “NMHC” to “NMHCs”

Methods, line 135: Change “exchange of chemical species between atmosphere and ocean” to “exchange of chemical species between the atmosphere and ocean”

Methods, line 151: Change “the top of the atmosphere” to “the top of atmosphere”

Methods, line 163: Change “we assess the so-called fast response as difference” with “we assess the so-called fast response as the difference”; Likewise on line 164

Section 2.3, line 207: Replace “are accounted as part of the forcing” with “are accounted for as part of the forcing”

In Section 2.3, lines 232-240: On the implementation of the diagnostic radiation calls, I think this section is not quite as clear as it could be. I would suggest that you say upfront that you will explain the double call method by using the calculation of the radiative effect from water vapour, as an example. And then go on to say that the same approach can be used for other constituents, by swapping the reference climatology for a perturbation climatology. It could read something like “To calculate the radiative effect of H2O, for example, the second, i.e. first diagnostic …”. One could even use numbers to specify the calls. For example, P is the prognostic call, and D1 and D2 are two diagnostic calls, which differ only by a single climatology, i.e., the reference climatology in D1 is replaced in D2 by a climatology from the perturbation experiment for whichever constituent. In the same section (lines 238-239), you again mention replacing the climatology for water vapour, which seems odd if I’ve understood the methodology correctly.

In Section 2.3, line 245: Replace “served as reference” with “served as the reference”

Section 3.1, line 255: Replace “the cooling to be expected from CO2 increase” with “the cooling to be expected from the CO2 increase”

Section 3.1, line 260: Change “any feedback of natural CH4 emissions” to “any feedback on natural CH4 emissions”

Section 3.1, line 262: Change “with respect to the oxidation with OH” with “with respect to OH oxidation”

Section 3.1, line 265: change “by the increase of tropospheric humidity” with “by the increase in tropospheric humidity”

Section 3.1, line 282: Change “were set to RCP8.5 conditions of the years” to “were set to RCP8.5 conditions for the years”

Section 3.1, line 300: change “except of the GISS-E2-R” to “except for the GISS-E2-R”

Section 3.1, line 304: Change “subsequent feedback of OH” to “subsequent feedback on OH”

Section 3.1, line 329: Change “from CO2 perturbation” to “from a CO2 perturbation”

Section 3.1, line 352: change “GHG concentration” to “GHG concentrations”

Section 3.1, line 353: Change “category O3 stratosphere contributes also most to the strong reduction” to “category O3 stratosphere is also the largest contributor to the strong reduction”

Section 3.1, line 355: change “with an increase of lightning NOx emissions” to “with an increase in lightning NOx emissions”

Table 3 caption: Change “are scaled by a factor of 0.6 before added” to “are scaled by a factor of 0.6 before being added”

Section 3.1, line 357: change “In addition, also biogenic emissions” to “In addition, biogenic emissions”

Section 3.2, line 394: Change “increase of the CH4 surface mixing ratio” to “increase in the CH4 surface mixing ratio”

Section 3.2, line 398: Change “the increase of CH4 emissions” to “the increase in CH4 emissions”

Section 3.2, line 441-442: Change “the increase relative to the total reference O3 is with up to 15% most pronounced in the lower NH” to “the increase of up to 15% relative to the total reference O3 is most pronounced in the lower NH troposphere”

Section 3.3, line 509: Change “of stratospheric O3 it is 0.16 W m−2” to “of stratospheric O3 is 0.16 W m−2”

Section 3.3, line 520: Change “induced by composition changes of the climate response” to “induced by changes in composition changes due to the climate response”

Section 4, line 600: change “allows for secondary feedbacks of OH and O3” to “allows for secondary feedbacks on OH and O3”

Section 4, line 604: Change “substantial differences of the climate response of tropospheric O3 between this study and previous work” to “substantial differences in the climate response of tropospheric O3 between this study and previous work”

Section 4, line 610: change “increases of tropospheric O3” to “increases in tropospheric O3”

Section 4, line 627: Change “caused by changes of C5H8 emissions” to “caused by changes in C5H8 emissions”

Section 4, line 683: Change “increases from wetlands as response to a 4×CO2 perturbation” to “increases from wetlands in response to a 4×CO2 perturbation”

Section 4, line 684: Change “effect of feedbacks of the gas-phase chemistry and of natural emissions” to “effect of feedbacks on the gas-phase chemistry and on natural emissions”

Section 4, line 687 and 688: Change “feedback of CH4” to “feedback on CH4”
Citation: https://doi.org/10.5194/egusphere-2024-2938-RC2

Laura Stecher, Franziska Winterstein, Patrick Jöckel, Michael Ponater, Mariano Mertens, and Martin Dameris

Supplement

https://doi.org/10.5194/egusphere-2024-2938-supplement

Laura Stecher, Franziska Winterstein, Patrick Jöckel, Michael Ponater, Mariano Mertens, and Martin Dameris

Viewed

Total article views: 502 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
259	52	191	502	20	4	5

HTML: 259
PDF: 52
XML: 191
Total: 502
Supplement: 20
BibTeX: 4
EndNote: 5

Views and downloads (calculated since 01 Oct 2024)

Month	HTML	PDF	XML	Total
Oct 2024	175	29	58	262
Nov 2024	60	11	90	161
Dec 2024	24	12	43	79

Cumulative views and downloads (calculated since 01 Oct 2024)

Month	HTML	PDF	XML	Total
Oct 2024	175	29	58	262
Nov 2024	60	11	90	161
Dec 2024	24	12	43	79

Viewed (geographical distribution)

Total article views: 497 (including HTML, PDF, and XML) Thereof 497 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 13 Dec 2024

Short summary

Methane, the second most important anthropogenic greenhouse gas, is chemically decomposed in the atmosphere. The chemical sink of atmospheric methane is not constant, but depends on the temperature and on the abundance of its reaction partners. In this study, we use a global chemistry-climate model to assess the feedback of atmospheric methane induced by changes of the chemical sink in a warming climate, and its implications for the chemical composition and the surface air temperature change.


Total:	0
HTML:	0
PDF:	0
XML:	0