We thank this reviewer for providing us with feedback and for assisting us in improving the manuscript. We also thank the reviewer for noting that this is a “valuable global analysis and synthesis”.

1. “The authors have not fully considered other important soil C loss mitigation strategies or a more targeted approach for the most vulnerable regions given their capacity for soil C gains or GHG emissions. Current frameworks for soil carbon management are not fully explored”. Response: We agree that we have not examined this in detail, but nor is this the focus of our study. Rather, before it is possible to develop appropriate and targeted C management approaches, it is first imperative that the contribution of soil to emissions of the various greenhouse gases be fully quantified. Without this information, it is not possible to develop appropriate carbon management strategies, or indeed, ensure that our focus is on the most pressing needs. For example, as illustrated in the present study, greenhouse gas emissions are currently much more substantial for nitrous oxide than for carbon dioxide, and in terms of reducing ongoing emissions, the focus should actually be on nitrous oxide rather than carbon – this is the value of our current approach in subsequently enabling targeted management approaches. This is described in the Introduction, for example, where we state: “in order to improve management practices and to inform better decision-making processes, it is imperative that we quantify the precise sources of greenhouse gases and understand the factors causing their emissions”.
2. “The authors discuss the reason for excluding uncertainty estimates, but I am not convinced that this wasn’t possible given that the main data they reference does include confidence bounds. Sanderman (2017) reports uncertainty estimates, so why not use those in the current modeling effort?” Response: The data originally published by Sanderman et al (2017) erroneously lists the value as being 133 Pg C, but this was modified in the “corrected” version of their manuscript to be 116 Pg C (DOI: 10.1073/pnas.1800925115). In this corrected version, we are unable to see any confidence bounds and hence we are unable to include it in our present study. Furthermore, even in the original study (DOI: 10.1073/pnas.1706103114), we are unable to see any confidence intervals in the original text for the value of 133 Pg C, nor any confidence intervals for the historical values shown Figure 2 (for example, also refer to Table S3 and Figure S10 of Sandermann et al 2017). If we are mistaken, we would appreciate the Reviewer assisting us with this information.
3. “They suggest a multifaceted approach to land management to grow the food needed to support the global population, but have failed to enumerate any innovative or targeted approaches and instead suggest a few broad, sweeping needs that were not investigated by the present study and that have not been critically assessed for their feasibility of implementation”. Response: Please see detailed response above. Specifically, it was the aim of our study to first quantify the emissions of greenhouse gases from soil – this information is essential if we are to subsequently develop innovative or targeted approaches.
4. “Abstract should include a statement about the data sources used for this analysis”. Response: We will modify the Abstract to the extent possible, but we note that it is against the rules of the journal to include references in the Abstract.
5. Crow & Sierra (2022). Response: We thank the Reviewer for highlighting this important study, which we will now cite in our manuscript. However, the focus of Crow and Sierra (2022) is different to ours and we are not sure how their approach would benefit in our study. This is because we are not seeking to model changes over time. Rather, in our study, we take previously-computed values for greenhouse gas concentrations emitted from soil (with these previous models already examining changes over time), and based upon these values, we determine the contribution to radiative forcing.
6. “L165: C outputs as CO2 efflux only or outputs to biomass vegetation too?”. Response: We will clarify the text to avoid this confusion. Specifically, the increase in outputs is both as the increased mineralization to CO2 (heterotrophic respiration) but also in the export of products for human consumption.
7. “L178-180: Clarification needed regarding land-use change decreasing in terms of acreage of land affected globally or intensity of the change?”. Response: We will clarify the text as suggested.
8. “Fig. 2: Add to the figure caption a brief note of the ecosystem/vegetation and features”. Response: We will modify Figure 2 as suggested by the Reviewer.
9. “L280 suggests that a multifaceted approach is needed that includes landowner incentives and points to Fig 3, but the figure does not include any description of how these changes might be accomplished. They are very broad recommendations and do not include any novel or innovative approaches”. Response: The focus of the present study was not to develop novel management approaches. Rather, our focus was to “quantify the substantial quantities of anthropogenic greenhouse gas emissions from soil and their contribution to climate change”, with this information being critical in subsequently developing targeted management approaches. Indeed, if we do not first understand the nature of greenhouse gas emissions from soil, we are unable to develop and implement appropriate management approaches.
10. “L 286: Regarding the recommendation to cease land-use change (from native vegetation), including for bioenergy production. Is there enough evidence from the current study to support this recommendation?” Response: We contend that there is enough evidence to support this – widespread land-use has caused profound release of greenhouse gases historically (Figure 1), and so we need to avoid rates of land-use change increasing again which would be associated with a concomitant release of greenhouse gases.
11. “Fig 3. For the recommendation to reduce waterlogging, how feasible is this in rice paddy production systems?”. Response: We will modify the manuscript to discuss this in more detail. In addition, we will modify the arrows in Figure 3 as suggested by the reviewer.
12. “L 310-311: Consider also referencing the publication: Bailey, V. L., Pries, C. H., & Lajtha, K. (2019).”. Response: We will reference this publication as suggested.
13. “L 316: This argument about intensifying agricultural production has been made frequently before”. Response: We will modify the manuscript according to the comments from the Reviewer to note that this is a balance and that it depends upon the system, environment, and land-use conditions.
14. “What about studies that have considered the diversion of food waste and argue that we already produce enough food globally”. Response: The Reviewer makes an important point which we will incorporate into the revised manuscript.
15. “What about arguments that a portfolio of approaches are necessary to address the magnitude of our current problem?”. Response: We thank the Reviewer for noting this, which we will incorporate into the revised manuscript.
16. “L418-422: I appreciate this perspective, but given the spatiotemporal variability of soil C losses and gains, it is important to consider the inputs and outputs when possible”. Response: We will modify this statement to avoid confusion. Specifically, the approach we have taken does indeed account for existing gains in C in some regions (for example, Boreal forests) as this is part of approach used by Sandermann et al in their study. Rather, we were attempting to convey that our focus was on quantifying the historical and current net emissions of greenhouse gases rather than developing novel approaches that could be used in the future for sequestering carbon in soils.
17. “L170. Remove error message”. Response: We will fix this error, and we thank the Reviewer for noting this problem.
18. “L 235. considering the atmospheric life of N2O is 109 y”. Response: We will fix this grammatical error.

We appreciate the time that the reviewer has taken to read our manuscript and provide us with this valuable feedback.

1. “The Introduction should clearly show the novelty and importance of the study”. Response: We thank the Reviewer for this suggestion. We will revise the manuscript to more clearly articulate the novelty of this work, including a clearer delineation of ‘soil’ and ‘land’ given that previous studies have focused on land but here we distinguish here between ‘soil’ and ‘land’ as is required for understanding, valuing, and managing soil as a discrete component of land.
2. “The Methods section is not clear, please describe how to obtain and analyze the data in detail to support the robust results”. Response: We apologize if something is not clear. If the Reviewer is able to provide specific information, we will make changes accordingly. However, it is not apparent to us what information in the Methods section is not clear.
3. “The Results are your results, not the others (e.g. lines 149-150). There are many references cited in this part”. Response: As suggested by the Reviewer, we will remove citations in the Results and move them to the Discussion as appropriate.
4. “The Discussion should be the discussion of results from this study by comparing the published literature”. Response: We agree with the reviewer although we are unsure exactly what is being requested. If the Reviewer is able to provide specific details, we will modify the text as appropriate.
5. “Spelling and grammatical issues should be carefully checked (e.g. Line 170)”. Response: We will check the text carefully to fix any spelling and grammatical errors.
6. “While I think this study is interesting, however, the manuscript should be well-organized and presented, and revisions are needed to make it more readable, logical, and credible”. Response: We agree with the reviewer and we will check the manuscript carefully. However, we are unsure why the reviewer states that revisions are required to make it more “credible” and without specific details we are unable to make revisions in this regard.