The author's appear to have adequately considered responses to referee comments. Please make sure, as you revise the manuscript, that the content of the responses to referees is also reflected in the paper text not just in the response (Other readers may well have similar questions as they read the paper). It is not clear if this is always the case in the response as currently written.

To the extent practical, please submit a tracked change version of the manuscript (I see that there will be some signifiant re-structuring - that part won't be amenable to tracked changes.)

I've reviewed the revised paper and responses to referee comments. The paper is improved, however is still difficult to understand. The methodology is complex, combining forward modeling, inverse modeling, and scale factors, and needs to be more clearly explained.

The first major point of confusion regarding the methodology is the overall structure of the calculation with respect to how emissions, concentration, and forcing are calculated. This should be clarified in one place for the reader. This statement, Line 105, " Using a combination of calculated and prescribed time series of annual radiative forcings, we derive the atmospheric CO2 radiative forcing and hence its concentration, taking account of any land and ocean feedbacks." is unclear, but is central to the calculations done in the paper. I've searched through the paper but cannot find a succinct description of the methodology on this point. This should be added either in this section or soon after to clarify this for the reader (some specific comments below relate to this as well).

For example I infer from various parts of the paper that the following are exogenous: anthropogenic CH4 emissions and CH4 mitigation (from IMAGE), land-use patterns (except that later the paper discusses optimization of land-use between natural and biomass crops, which seems a bit in-consistent with this statement), and the assumed global mean temperature pathway. Land-use feedbacks on CO2 and CH4 emissions are endogenous to the models used in the paper. Its unclear how radiative forcing and concentrations of CO2 and CH4 are determined. It seems that some adjustment to the IMAGE CH4 concentrations is calculated, which leaves a residual CO2 forcing pathway that is inverted to obtain an anthropogenic fossil CO2 budget? These issues are central to the methodology of the paper and need to be better summarized for the reader. (see also comments below on Figure 3) Is total forcing also fixed to one time path, or is the inverse calcuation focused only on the temperature pathway?


A second point that is difficult to follow is the methodology for how uncertainty is treated in the paper. One issue is that the paper alternates between discussing uncertainty in some sections, and then discussing factorial runs in others. It might be useful to be more consistent in terminology. The overall approach to addressing uncertainty needs to be more clearly stated early on in the paper (e.g. first part of section 2). There is a clear statement at the beginning of the conclusion section, but this needs to be made much sooner.

On a specific point, the paper mentions two values of Q10 used as uncertainty bounds (~line 145). Are only the bounding cases used (the text implies this as currently written)? In that case, what do the central values presented in later figures (for example bars figure 9) represent?

The results of the factorial experiments do not really seem to be fully explored in the discussion of results overall. It would strengthen the paper substantially if these could be more fully examined. For example, what is the range in contributions of each specific component? For example the modeling of methane from wetlands is discussed, but what is the contribution of wetland methane to the allowed carbon budget? Same for ozone damage, etc. What uncertainty factors (and assumed BECCS efficiency) impact these the most? There are some aggregate results shown in Figure 8, but no detailed results are presented or discussed in the paper or the supplement.

The way that uncertainty in BECCS is considered is difficult to understand since, by the way the calculations are structured, some uncertainties are addressed with a factorial design (natural or "BECCS" as an option), while uncertainty in BECCS potential is addressed by setting the value if "k" (Kapa). There are a few results presented with different values of "K", but most results seems to be based on one value of this scale factor.


There is also some inconsistency in nomenclature:

Table 1 describes a scenario with "Land-based mitigation, including BECCS" and "Land-based mitigation with no BECCS". This is clear, although there is no mention of the assumed BECCS K value, which should be added.

However then in Figure 8, there are a number of other options presented. Such as "Land-based mitigation (Optimized)" and "Land-based mitigation (Natural)" - these should both be identified in Table 1. Then here are "Coupled" scenarios shown that don't correspond with any of the abbreviations given in Table 1.

Then, further, in Figure 9 there is no BECCs scenario presented, only "Land based mitigation", but evidently land-baed mitigation that also contains BECCS (with K =1 and K = 3), which also does not correspond to any label in Table 1. (It seems that these are two variants of "Land-based mitigation including BECCS"?)

The authors need to throughly go through the paper, tables, figures, and supplement and harmonize nomenclature to avoid confusing readers.


Further specific comments:

Abstract - I presume these results are specific for the IMAGE SSP2 scenario. This should be mentioned.

Line 90
"and methane climate feedbacks) in a climate/Earth System modelling framework to quantify the unrealised potential from the mitigation of land-based options and anthropogenic CH4 sources."

I don’t see how the authors can claim here (and elsewhere) that methane mitigation has not been examined. There is substantial literature on methane migration in the context of temperature and forcing targets. Further, all the IAM scenarios such as the SSPs contain representations of anthropogenic methane mitigation, including the IMAGE scenarios from which this paper draws. Clarify this and make sure previous work is properly cited.

For example also the statement here “Although the primary challenge remains mitigation of fossil fuel emissions, these results highlight the unrealised potential of these mitigation options to make the Paris climate targets more achievable” does not seem to be warranted. Most of the scenarios in the literature, and certainly the SSPs, all include methane mitigation.


Line 141 not clear what "(the depth of zero annual amplitude)" means.

Line 205 This " The increased/reduced atmospheric CH4 concentration will have a corresponding faster/slower atmospheric decay rate than the prescribed concentration pathway. We account for this following the approach of Cubasch et al. (2001). Related changes in atmospheric radiative forcing, in response to altered atmospheric 210 CH4 concentrations, are calculated using the formulation from Etminan et al. (2016). We also include the indirect effect of these CH4 emission changes on the forcing by tropospheric ozone and stratospheric water vapour by multiplying the CH4 forcing by 1.65, based on Myhre et al. (2013)."

is unclear and seems to involve quite a number of assumptions and simplifications that should be further detailed and tested (likely largely in the supplement). First, it is not clear what it means to say "following the approach of Cubasch et al." given that this is a reference to the entire WG I IPCC report. More details should be provided. Changes in wetland emissions are, according to the author's own equations, non-linearly related to temperature. It is not clear at all that this can be represented as a "faster/slower atmospheric decay rate", whatever that means.

It is not clear that "tropospheric ozone and stratospheric water vapour" can be accurately represented by a single multiplicative factor applied to CH4 forcing. Methane concentrations and tropospheric ozone are linked through a number of mechanisms, with tropospheric ozone forcing in Meinshausen etal 2011, for example, being tied more closely to methane emissions not forcing, and with methane and tropospheric ozone forcing linked through changes in atmospheric oxidation capacity. Given that methane is a central focus of the paper, the representation of methane in the atmosphere should be better described. Use of a constant multiplier might introduce some distortions in the results.

The discussion above seems to contradict the statement in line 243 that "We use future projections of atmospheric CH4 concentrations and LULUC from the IMAGE SSP2 projections (Doelman et al., 2018) for both the methane and land-based mitigation strategies. ". From what I'm understanding from the paper it appears that the CH4 concentration is later adjusted for feedback effects - that should be made clearer.

Also, if CH4 mitigation is from IMAGE this statement in the abstract "Globally, mitigation of anthropogenic CH4 emissions has large impacts on the anthropogenic fossil fuel emission budgets, potentially offsetting (i.e. allowing extra) carbon dioxide emissions of 188-212 GtC" is, therefore, not a result of this work but a conclusion just taken from the IMAGE results. This needs to be made clearer if this is the case.

Line 310 - Is this "the temperature pathway (1.5° versus 2°C warming) having a minor effect." the case even for the high Q10 sensitivity? Presumably this also depends on the temperature pattern used as well?

Section "2.4.1 Anthropogenic Fossil Fuel Emission Budget and Mitigation Potential"
It would be useful here to clarify that this is only the CO2 emissions budget (not total GHG emissions).

It would also be useful here to add some comment on what is driving each of the terms in the budget. (e.g., which models in Figure 1)


Line 378 I assume that the calcuation of H* depends on the assumed loss from farm to final storage? If so the assumed for farm to final storage loss should be stated. (line 445 seems to imply that minimal losses are assumed? 13%).

Line 391 This line is confusing "We find that there is increased uptake of atmospheric CO2 in the land-based mitigation scenarios, although there is a reduction in land carbon from the land-use changes in these scenarios. " Aren't these just are two sides of the same effect? (and see next comment as well)

Line 395 "which is greater than the land carbon lost through land- use changes. " Not clear what this means. How do land-use changes factor into this, given that LULUC is exogenous from IMAGE and therefore constant?

Line 446 - "yields of > 30 ton DM ha-1 yr-1 would be more difficult to realise". Not clear what this refers to given that earlier in that sentence H* is estimated to be up to 10-20 tons DM/ha/year..

Line 462 and table 3. The meaning of Table 3 and the associated text is not clear given that cumulative CH4 emissions have no physical meaning since CH4 does not accumulative in the atmosphere on timescales considered here. What is the "regional scale factor" presented in the table (the one line description in the table is not sufficient to understand what this means)? That scale factor also doesn't appear to be used otherwise in this work.

Figure 2b - what do the error bars represent? In figure 2a, there is a high/low Q10 bound defined, but Figure 2b has a central point and an error bar for the simulation results. What do each of these represent? Clarify in caption and in paper text.

Figure 3. This figure is confusing. There are two temperature paths in 3a. But then there are two forcing paths in panel b that do not appear to be related to the temperature paths in panel a (what does CTRL and CH4 mitigation relate to the 1.5 and 2 degree pathways?). For a given set of climate parameters (climate sensitivity, etc.), wouldn't each forcing path uniquely relate to a temperature pathway?

Also, this seems to be a pivotal figure, but does not appear to be referenced in the text.

Figure 8 - This is a central figure, but is unclear. Describe in the legend the difference between the colored bars and the pink/red open bars? It is not clear what the text "Note that the gain in the land carbon store for the CH4 scenario is shown as a reduction from -70.8 GtC in the control run to -1.4 GtC in the “CH4” mitigation option (median of ensemble)." means, particularly given that there is no land (green) bar in the CH4 mitigation bar in this figure.

Figure 10 - I assume this "The spread of the functions " should be "The width of the lines "?

One of the scientific contributions of this paper is the analysis of BECCS within this larger context, but there's not much detail in the results given. It would enhance the value of the paper to provide in the appendix some tables for the different scenarios of the regional land areas devoted to BECCS and the assumed biomass production rates. This would make the work more readily usable.


The title and abstract of the paper focus on the combination of CH4 and land-based mitigation, but the conclusion section doesn't integrate these nor give context from the literature. The abstract indicates that methane mitigation offsets 88-212 GtC, while land-based mitigation offsets 51-100 GtC, which is much smaller. The methane mitigation analysis is exogenous to this paper, but is this range robust across models that do include this? Does the methane range include the offsetting effects from wetland emissions?

For the land-based mitigation, it would be useful if the conclusion section could provide some context for these results. What factors impact this range (e.g., under what conditions is this low, and under what conditions is this high?). How does this compare with what is assumed in the literature from IAM scenarios? Seems this might be much smaller than commonly assumed, why is that?

Are both of these ranges across 1.5 and 2.0 targets? Does a 1.5 or 2.0 degree target shift either of these ranges significantly?

I noticed that the upper end of land mitigation given in the abstract is lower than the high end of the line provided in Figure 10. (Presumably you haven't allowed the BECCS scale factor to be as high as 6 in the main results?).

Note that the journal's data policy requests that data and code associated with articles to be deposited in public data repositories. See:
https://www.earth-system-dynamics.net/policies/data_policy.html. The paper does not currently follow best practices here. Please consider depositing key input and output data in a public data repository.

The revisions to the paper are well done and have significantly enhanced the readability and depth of analysis. I have only a few minor editorial comments that could probably be addressed in page proofs.


Line 114 "reflecting the different thermal sensitivities of existing climate models"

I assume the authors mean climate sensitivity? (thermal sensitivities is not a standard term). Suggest revision to make clear to reader.

Line 116
Some words appear to be missing here
"activity, and for scenarios taken from the IMAGE integrated assessment model"

Line 117
"for CH4, we use an understanding of its atmospheric lifetime to translate methane emissions into their atmospheric concentrations"

"use an understanding" is a bit vague, but it would be sufficient to revise with a reference to section 2.2.1 that contains the new Equation 3 and text there that describes the methane calculations.)