Tapin et al.

General Comment.

This is an important paper that will provide essential data for a wide constituency of modellers. The paper is very well researched, well written and well presented. It will be highly cited. I recommend publication after very minor revisions.

We need to understand methane if we are to manage global climate change.  But there remains much uncertainty in the global methane budget, and sources and sinks are constantly changing. Measuring mixing ratio is not enough – to identify and quantify sources and sinks we need to use isotopic constraints – primarily C isotopes (though also H and clumped isotopes). To do that, we need really good databases of source signatures. That’s what this paper provides, and why it will be important.

I strongly support publication of this important work.

That said, I have some queries about the content of the paper, which could benefit from some discussion in the text. Three concerns are: 1. The lack of tropical data, and the assessment of latitudinal variation, especially from C3 and C4 plants, both in agriculture and wetlands. 2. The weighting given to ‘freshwater’ sources. 3.  The weighting given to ‘geological’ sources. 4. The chlorine sink. 5. The stratospheric sink.

The supplementary figures are of great interest.  

Specific comments. (and see reference list at end)

Line 23 – 1930 in 2024 – maybe cite NOAA (Lan et al. 2026) for end 2025.

L25. Lifetime. Maybe distinguish perturbation life (12 yr, as cited in IPCC) from burden/flux life of 9 yr.

L33. Long list of papers. Maybe add Ciais et al. 2026, Fujita et al. 2025, Riddell-Young et al. 2025,  and switch Nisbet 2019 to 2023.

L36 – add biomass burning (both natural and anthropogenic).

L64-66 – See also Line 270, and Table 1 FFG segment…Geological emissions. Petrenko et al have presented convincing evidence that the Etiope et al estimate of geological emissions is much too high, and I suspect even the downsized 21.1 Tg guess in Table 1 is still high.

L82 – Also see Fujita et al 2025.

L227 – sub-sector variability, and Sections 2.32 and 2.33. Sub sector uncertainty and aggregation. This is interesting, but perhaps there should also be a brief discussion of latitudinal uncertainty….there are many factors driving strong latitudinal gradients. In particular, C4 plants, which pre-concentrate and thus accept more 13-C in photosynthesis, are dominant in the tropical grasslands and wetlands (e.g. papyrus), while C3 rules in the boreal realm. Also in the sinks, in the Brewer-Dobson circulation the polar vortex delivers low mixing ratio, highly 13-C rich methane, while in the tropical troposphere the upward loss to stratosphere is of high mixing ratio 12-C rich methane.

L265 – Freshwater systems. I’m very sceptical here, but I realise this is a long-running puzzle and in this paper I think the best approach is simply to say briefly that fluxes are very uncertain. The re-scaling by one-third maybe should get a further note here, in addition to the reference to Saunois et al 2025. Note for example the Shaw et al. paper, which is one of the very few to study the Congo freshwater and wetlands. Shaw et al. (Fig. 1 top right) found low fluxes over the open water of large Lake Bangweulu, which is both shallow and warm, and very large fluxes over the wetlands very close by. The implication is that in the lake, much of the production was taken up by methanotrophy in the top of the aerated open water, but in the wetlands the methane egressed from papyrus and reed stems etc, or through ebullition in oxygen poor stagnant water.

L271 – Geological methane. See earlier comment – I strongly suspect this 23Tg/yr figure is too high.

L285 – Hydroxyl - See Morgenstern et al. 2025, and Ciais et al. 2026..

L292 and also Table 2 – the chlorine sink is small for methane in Tg but disproportionately large in its leverage on d13C…this is an important factor as the sink is still poorly quantified. It probably needs a somewhat longer discussion. Maybe also refer to Allen et al. 2007.

L394 to 397. also Fig. 1. Latitudinal gradient of C3 and C4 plants. There are several references that could be discussed here. Ganesan et al. is mentioned elsewhere (L482) in the paper but could be cited here too. Also see France et al. 2022, and Nisbet et al (MOYA) 2022. The C3/C4 gradient affects: 1) wetland emissions – equatorial wetlands are papyrus rich, while  outer tropics have more reeds (C3/C4) and boreal wetlands have almost no C4 plants. 2) agricultural emissions – ruminant grazers eat grasses in the tropics that are mostly C4, but in the drier outer tropics cattle and antelope are often browsers, on C3 trees. 3) biomass burns – tropical grass fires are C4 rich  but tropical bushes and tree fires are C3, so methane given off by incomplete combustion is somewhat less heavy.  Boreal forest fires are C3.

See Barker et a. 2020.

L436. AGW signature is regionally very variable. In the tropics, heavier than -60‰. In the populated northern hemisphere, breath of grassland animals (mostly beef) is probably lighter than -60‰ but intensive livestock (especially dairy) are fed C4 maize and their breath can be quite heavy.

L482 – emphasise the lack of tropical measurement? See also L501…

L523 – no aggregation uncertainty stated, but actually the aggregation involves aggregating different components of low latitude and high latitude results.

Table 5 – maybe mention the France et al (2022) and Nisbet et al. (2022) tropical wetland results?? These were from both S. America and Africa….

L560 – China is dominant – but India is also a big coal emitter..

Table 6 – Chemistry - mention the stratospheric sink?

L633 – Ethiopia – see Brychkova et al – Ethiopian cattle eat C3 above about 2500m but around 2000m many forage grasses are C4.

L671 – Landfills - maybe Nisbet et al 2020.

L687 – Caspian Sea. This isn’t clear – is the Caspian here treated as a Freshwater source? I think its about 12‰ salinity -I recall once being told there were people on an island who survived by drinking it but it’s pretty brackish. Also it has methane/smud volcanoes and there are huge petrochemical methane sources nearby – gasfields and oilfields. Again I’m most unconvinced by the overall freshwater emission estimates…see also Line 782 and 785 (Caspian).

L703 Section 4.2.3 – atmospheric chemistry – I am concerned by two problems – 1) the Cl sink with its large isotopic leverage and 2) the stratospheric sink. Both of these seem poorly understood and with high uncertainty.

Also note that OH in the tropical troposphere is changing (Morgenstern et al 2025) and likely has significant longitudinal variation around the globe.

L715 Oxidative capacity. - could mention Ciais et al. 2026 and Nisbet &Manning 2026.

L721 – references here are a bit elderly. ? Penn et al 2025, Skeie et al 2023?

L918 – targeted campaigns – see MOYA – double volume of Phil Trans Royal Soc. 2021-2022.

https://royalsocietypublishing.org/rsta/issue/379/2210

https://royalsocietypublishing.org/rsta/issue/380/2215

CONCLUSION

This is an important paper with important data that will be a most useful resource for those modelling methane budgets. That in turn has global significance of the UN Global Methane Pledge is to succeed.

The paper should be accepted with very minor revisions. It will be heavily cited.

SOME REFERENCES TO CONSIDER

Allan, W., et al. 2007. Methane carbon isotope effects caused by atomic chlorine in the marine boundary layer: Global model results compared with Southern Hemisphere measurements. Journal of Geophysical Research: Atmospheres, 112(D4).

Barker, P. A. et al.  2020. Airborne measurements of fire emission factors for African biomass burning sampled during the MOYA campaign. Atmos Chem Phys, 20, 15443-15459.

Brychkova G, et al. 2022. Climate change and land-use change impacts on future availability of forage grass species for Ethiopian dairy systems. Sci Rep. 2022 Nov 28;12:20512

Ciais, P., et al. 2026. Why methane surged in the atmosphere during the early 2020s. Science 391.6785 (2026): eadx8262.

France, J. L., et al. 2022. δ13C methane source signatures from tropical wetland and rice field emissions." Phil Trans Royal Soc A: 380.2215.

Fujita, R., et al.  2025. Global fossil methane emissions constrained by multi‐isotopic atmospheric methane histories. J. Geophy Res: Atmos, 130, e2024JD041266.

Morgenstern, O. et al. 2025. Radiocarbon monoxide indicates increasing atmospheric oxidizing capacity. Nature communications, 16, 249.

Nisbet, E. G. & M. R. Manning. 2026. What is causing the methane surge?. Science 391.6785: 556-557.

Nisbet, E. G. et al. 2022. Isotopic signatures of methane emissions from tropical fires, agriculture and wetlands: the MOYA and ZWAMPS flights. Phil Trans Royal Soc A 380.2215.

Nisbet, E. G. 2020. Methane mitigation: methods to reduce emissions, on the path to the Paris agreement. Reviews of Geophysics, 58, e2019RG000675.

Penn, E. et al. 2025 What can we learn about tropospheric OH from satellite observations of methane? Atmos Chem Phys 25,2947-2965.

Petrenko, V. V. et al. 2017. Minimal geological methane emissions during the Younger Dryas–Preboreal abrupt warming event. Nature, 548, 443-446.

Riddell-Young, B., et al. 2025. Microbial driver of 2006–2023 CH4 growth indicated by trends in atmospheric δD–CH4 and δ13C–CH4. PNAS 122, e2516543122

Shaw, J. T., et al. 2022. Large methane emission fluxes observed from tropical wetlands in Zambia. Global Biogeochemical Cycles, 36, e2021GB007261.

Skeie, R. B. et al 2023. Trends in atmospheric methane concentrations since 1990 were driven and modified by anthropogenic emissions. Comms Earth & Envir, 4, 317

Tyler, S.C., et al. 1988. Measurements and interpretation of δ13C of methane from termites, rice paddies, and wetlands in Kenya." Glob Biogeochem Cycl 2, 341-355.

Yao, P., et al. 2026. Distinct dual-isotopic signatures of major methane sources in South Asia. EGUsphere, 2026, 1-60.

This manuscript presents a comprehensive effort to develop spatially resolved δ13C-CH4 source signature maps and their associated uncertainty estimates for use in atmospheric methane inversion studies. The dataset spans 1998–2022 at 1° × 1° resolution and includes major methane source categories, including fossil fuels, agriculture and waste (AGW), wetlands, biomass burning, and other natural sources. The authors further implement these datasets within an atmospheric modeling framework (CIF coupled with LMDz) to conduct forward simulations and sensitivity analyses of CH₄ mole fractions and δ13C-CH4. A central contribution of the study lies in its systematic quantification of uncertainties in isotopic source signatures, including sub-sector variability and aggregation effects. The manuscript clearly demonstrates that uncertainties in OH kinetic isotope effects (KIE) and AGW source signatures dominate the variability in modeled δ13C-CH4 fields. The study is timely and relevant, as constraining methane sources remains a major challenge in atmospheric science, particularly in reconciling bottom-up inventories with top-down inversion results. The dataset and methodological framework presented here have clear potential to improve prior constraints in isotopic inversion systems. However, despite these strengths, the manuscript has several limitations that reduce its current impact. Most notably, the absence of model-observation comparison, the relatively coarse treatment of key source categories (especially AGW), and the limited discussion of structural uncertainties (e.g., transport and oxidation processes) weaken the robustness and applicability of the conclusions. Addressing these issues would substantially strengthen the manuscript. Overall, I recommend major revision prior to publication.

Major Comments

1. The manuscript focuses on forward simulations and sensitivity analyses without comparison to atmospheric δ13C-CH4 observations. While the authors justify this choice, the absence of validation limits confidence in the dataset and its applicability. Please include at least a first-order comparison with available surface or global δ13C–CH4 observations to demonstrate the realism of the modeled fields.
2. The isotopic source maps are largely derived from previously published datasets and literature synthesis. While this approach is necessary, the manuscript does not sufficiently address the potential propagation of systematic biases. Provide a more critical assessment of inter-dataset inconsistencies and discuss how these uncertainties may influence the final isotopic maps.
3. The agriculture and waste sector is identified as the dominant source of uncertainty, yet it is treated in a relatively aggregated manner. Where possible, further disaggregate AGW into sub-categories (e.g., livestock, rice paddies, waste management), or include sensitivity analyses demonstrating the effect of finer sectoral partitioning.
4. The results highlight OH-KIE as a dominant driver of δ13C-CH4 uncertainty, yet the treatment of this process is relatively simplified. Please expand the discussion of OH-KIE parameterization, including laboratory constraints, variability, and implications for inversion systems.
5. The manuscript primarily focuses on OH oxidation, while other sinks (e.g., Cl oxidation, soil uptake) receive limited attention. Provide a more quantitative discussion of these additional sinks and their potential isotopic impacts.
6. The uncertainty propagation methodology is rigorous but complex and may be difficult for readers to follow. Maybe include a schematic diagram summarizing uncertainty components and clarify how these uncertainties are incorporated into inversion prior error structures.
7. All simulations are conducted using a single atmospheric transport model (LMDz), without discussion of model-related uncertainties. Discuss the potential influence of transport model choice and resolution on the results.
8. The manuscript identifies regions with high isotopic uncertainty but does not explore them in depth. Include more detailed regional analyses or case studies, particularly for key methane-emitting regions.
9. The manuscript includes a forward-looking discussion on satellite observations (e.g., GOSAT, TROPOMI), but this remains largely conceptual. Clarify current technical limitations and, if possible, provide quantitative estimates of detection capabilities.
10. While the manuscript is generally well organized, some sections, particularly those describing uncertainty propagation and sensitivity analyses, are dense and difficult to follow. Please improve readability by summarizing key findings at the end of sections and reducing redundancy between the Results and Discussion.