General comments

In their manuscript titled “Simulating soil atmosphere exchanges and CO₂ fluxes for a restored peatland”, the authors used the CoupModel to simulate soil atmosphere exchanges, with a focus on CO₂ fluxes, for a peatland in Quebec, Canada that was restored 14 to 16 years prior to the study period. To evaluate the model performance, the authors compared the simulated energy fluxes, soil temperatures, water table depth and NEE to in-situ measurements performed at the site during the study years. To simulate and test the effect on NEE of further time passing after restoration, the authors tested the sensitivity of the simulated NEE to changes in acrotelm thickness. They concluded that the model is well-suited to simulate soil atmosphere exchanges of a restored peatland. To assess the effect of interannual climate variability on CO₂ uptake by the restored peatland, the authors quantified the variation in annual NEE using meteorological data from 28 years as climate forcing for the model. Comparing the results to the CO₂ fluxes measured over 15 years at a nearby pristine peatland showed that at the restored site CO₂ uptake was more strongly reduced by dry conditions. In an additional simulation the authors tested different climate change scenarios affecting air temperature and/or precipitation and showed that restored peatlands will likely maintain their function of taking up CO₂ under a changing climate.

As explained by the authors, restoration of peatlands can return them to carbon sinks and thereby counter climate change. Considering the large areas of drained and extracted peatlands worldwide, investigating the current and predicting the future C uptake function of restored peatlands is a highly relevant scientific question with important implications for future management of restored peatlands and that is well within the scope of BG.

Besides simulating the CO₂ exchange for a specific restored peatland, the authors increase the relevance of their study by evaluating the general applicability of the CoupModel to restored peatlands and by providing the model code for use by other researchers at other peatland sites. The scientific methods applied in the study are clearly outlined either within the manuscript or in earlier publications that the authors refer to. The authors thus ensure that their study results are reproducible.

The model used in the study has already been developed and published before. The novel part is that the authors showed that the model is also applicable to restored in addition to drained and pristine peatlands and can therefore now be used to simulate the full succession of land use change in peatlands.

The second main conclusion of the manuscript is that restored peatlands are more vulnerable to dry conditions than pristine sites and confirms the findings from earlier studies that are cited in the introduction of the manuscript.

The conclusions drawn from the study results are relatively broad considering that the results were derived using a model setup for one restored peatland and one point in time after restoration only. While the technical evaluation of CoupModel for restored peatlands is sufficiently supported by the results, the content-related conclusions on the resilience of restored peatlands to climate variability and on the emission factors of restored peatlands should be verified in future studies based on data sets from more study sites and years. Is it valid to call it a “long-term model simulation” if you use climate forcing for 28 years but assume that the peatland remains in a state of 14 to 16 years after restoration?

The overall presentation of the study is very clear and easy to follow as it uses a clear structure defined by the research objectives as well as fluent and precise language. The figures are very clear and easy to understand and therefore nicely support the study results.

Specific comments

Title:

ll. 1-2 Please consider revising the title to make it more precise. Which “soil atmosphere exchanges” do you mean? CO₂ fluxes are a “soil atmosphere exchange” too.

Introduction:

In line 264 of the results section you briefly mention that the “thickness of the newly grown mosses […] partly represents the time since restoration”. To motivate your research objective 2) (testing the model sensitivity to changes in acrotelm thickness) it would be helpful if you could elaborate the structural changes in a peatland with time after restoration in one paragraph in the introduction.

Methods:

You state in your study objectives (l. 123) that one of your aims is to adapt the CoupModel for restored peatlands. Please expand a bit on lines 155-157 to elaborate the changes you made to CoupModel compared to the model setups for pristine and extracted sites. A schematic showing the layers of the soil profile that were considered in the CoupModel for pristine, extracted, and restored peatlands, respectively, would be helpful.

Results:

At the end of each paragraph of the model evaluation in section 3.1 you speculate about the reasons for deviations of model results from observations. As this is already an interpretation of your results, I think that it would be good to move combine the respective sentences to form an additional paragraph in the discussion section of your manuscript. I think that your finding of model deficiencies in simulating the time of phase changes in spring is a relevant discussion point. Maybe you could even add a sentence on how the model performance for this time of phase changes could be improved in the future?

l. 267 How did you choose the new acrotelm thicknesses of 20 cm, 40 cm, and 100 cm? At what times after restoration would you expect to find such acrotelm thicknesses? Please elaborate.

l. 271 In the introduction you mention that the main difference between pristine and restored peatlands that also leads to the higher vulnerability of restored peatlands to dry conditions is the missing mesotelm collapse layer in restored peatlands. If I understand correctly, by testing the sensitivity of modelled CO₂ fluxes to the acrotelm thickness you aim to estimate how much the CO₂ fluxes in a restored peatland can be expected to change with time after restoration. Do you think that your assumption that a mesotelm collapse layer does not develop in a restored peatland with increasing moss layer thickness is valid? Or could it be that with this model setup you artificially introduce a difference between pristine and restored peatlands that in reality disappears with time after restoration?

l. 278 Is this something that you would expect to happen in reality? Does this also happen in pristine peatlands?

l. 301 In the introduction and in the methods section you mention that the CoupModel has previously been used to simulate CO₂ fluxes at Mer Bleue. Could you potentially add these simulated fluxes to the comparison of CO₂ fluxes between the pristine Mer Bleue and the restored BDB site? I am wondering how comparable the model results from BDB and the observations from Mer Bleue are. Comparing model results from both BDB and Mer Bleue could help to rule out the possibility that model deficiencies introduce a higher sensitivity of CO₂ fluxes to climate variability. This could be especially beneficial as the model results were evaluated against EC measurements only for years where climate conditions did not deviate much from the long-term mean (l. 289).

l. 305 Maybe mention that the average rate of CO₂ uptake reported here is the same as the emission factor for BDB that you refer to in the discussion to avoid confusion about the terms.

Discussion:

l. 380 I think that this is an important conclusion that should be supported by data added to the results section. Can you estimate around what time after restoration a new acrotelm thickness of 40 cm could be reached?

Do you think that the remnant ditches will have a significant effect on the C balance of restored peatlands over long time periods after restoration? Should their effect be considered in future studies when evaluating the differences in the C uptake function between pristine and restored peatlands?

Conclusion:

You could consider adding your finding that restored peatlands will likely maintain their C uptake function under future climate change.

Figure 3:

Would it be more correct to display GPP and NPP with a negative sign similar to NEE? Consider defining the abbreviation NPP, as NEE, GPP and ER have also been explained in the main text.

Figure 7:

Would it be more correct to display GPP with a negative sign similar to NEE?

Technical corrections

Please note that I am not an English native speaker - therefore please consider the following as "suggestions" rather than "corrections".

l. 24 “to” instead of “of”?

l. 31 “in dry years” or “under dry conditions”?

l. 122 The word “fluxes” can be left out as CO₂ is also exchanged between soil and atmosphere.

l. 183 Do you mean “C dynamics”?

l. 278 “cannot” instead of “can’t”

ll. 280, 315, 316 To me these sentences are all missing an “influence/impace/effect of” e.g. l. 280 “Effect of interannual climate variability on …”

l. 341 “… shows that CoupModel …”

ll. 382 – 383 Is the unit correct or should it be -100 mb as in line 265? 

Please see the attached file.