1.    Does the paper address relevant scientific questions within the scope of BG?

The paper present results of a modified version of DNDC-CMMM to better represent the NH3 volatilisation of paddy rice field and upland crop. Considering the importance of N cycles in soil biogeochemistry and of NH3 emissions in global change the paper seems within the scope of BG.

2.    Does the paper present novel concepts, ideas, tools, or data?

The paper improve existing model of N emissions for two types of cultivation and performed sensitivity analysis to several environmental factors. In this way, tools and findings are novel.

3.    Are substantial conclusions reached?

Yes, however it is a bit difficult to follow wich simulations are considered as valid and which one are considered as non relevant. Maybe a table or few clear sentences to sum up might be helpful.

4.    Are the scientific methods and assumptions valid and clearly outlined?

The methods are well detailed, however the authors combined 3 different models (DNDC, CNMM and Jayaweera-Mikkelsen) while only Jayaweera-Mikkelsen model is accompagnied by a schema. In my opinion, taking into account the numerous processes involved in each model and the successiv optimisations performed by the authors, a complete scheme of the full model including the parameters used by default or optimized could be helpful.

5.    Are the results sufficient to support the interpretations and conclusions ?

The results are well detailled but the difference in the fig 3 to 5 is not very clear for me. Also, a comparison of the NH3 volatilisation sensitvity to environmental factor between rice paddy field and upland crops might be interesting. I found that the 4.1 title does not precisely correspond to the paragraph and that the 4.3 is a bit long and difficult to follow 

6.    Is the description of experiments and calculations sufficiently complete and precise to allow their reproduction by fellow scientists (traceability of results)?

I think all the data and results are provided, however, as noticed above the numerous processes and step involved in modelisation make it difficult to evaluate without a complete scheme (especially because several parameters were took from other model). Also, it is not very clear if the data of re-calibration was the same than used in the last version of DNDC of Li et al., 2019 (l.117). Several parameters have been recalibrated (l. 168 to l.176) without mentioning how they were recalibrated (algorithms and data). Also, I don’t understand why the time conversion factor is 0.75 (so not a ratio of the two initial models time step) while one model is based on 3 hours and the other on 24 hours.

7.    Do the authors give proper credit to related work and clearly indicate their own new/original contribution?

For the major parts of the manuscript their is enough references, however the 4.4 lack of some references

8.    Does the title clearly reflect the contents of the paper?

Yes

9.    Does the abstract provide a concise and complete summary? 

Yes but maybe the description of the sensitivity analysis could have been a bit more longer. Also, the validation part is presented after the sensitivity analysis in the abstract while in the results it came aftewards.

10.    Is the overall presentation well structured and clear?

The results are relatively clear except that in my opinion the fig 6 should be citted earlier in the text while comparing the model performance.. Also, large part of the results focused on ABC treatment which is only 2 sites, in my opinion the results for others amendment are more robusts. The material and methods part is long and, as mentionned above the model and model parameter fitting is complicated to follow. The material and methods sections on data analysis are much clearer.

Is the language fluent and precise?

yes

11.    Are mathematical formulae, symbols, abbreviations, and units correctly defined and used?

Yes

12.    Should any parts of the paper (text, formulae, figures, tables) be clarified, reduced, combined, or eliminated?

As mentionned above the fig 3 and 5 are similar and a conjungated presentation of these two figures would be easier. Also, I there is several subplot in each of them but the whole figures seems somehow similar. Also, the text police is not the same in the whole manuscrit, please adjust.

13.    Are the number and quality of references appropriate?

yes

14.    Is the amount and quality of supplementary material appropriate?

I would have appreciated if the model the authors build was available for re-use

The authors describe an update to the CNMM-DNDC model to improve ammonia emission simulations. First, they performed a calibration and parameterization across upland agriculture sites. Second, they added a new algorithm to simulate ammonia emissions from rice paddy fields with flood irrigation. Finally, the authors conduct a sensitivity study to analyze the contributions to ammonia emissions over a variety of environmental factors (including atmospheric conditions, soil properties, and management of irrigation and fertilizer).

The manuscript is novel in that it brings a new management practice, flood irrigation, into a biogeochemistry model for calculation of ammonia emissions. This is a valuable addition to the field and should be considered in future models because of the differences in volatilization from management practices. I think the paper is well written and interesting and fits in the Biogeosciences scope. However, the paper could benefit from some additional clarification before it is ready for publication.

General comments:

Line 47-50: Might also want to mention efforts in ESMs such as Riddick et al., 2016 and Vira et al., 2020.

Line 63: “dose and application methods…” The type of fertilizer used (e.g., manure vs. synthetic fertilizer) will also have an effect on volatilization of NH3.

Line 102-104: The algorithms were already in the model, my understanding from reading the manuscript is they calibrated and parameterized the model further. However, this is never fully explained in the text under the materials and methods section.

Materials and methods section, I suggest organizing this differently. The authors start with a description of the sites, but I think it would be better after the model description, but this is optional

Line 155: What regional scale simulations were performed?

Section 2.2.2:

1. The authors indicate recalibration and parameterization, but don’t indicate how this was done or which components were included in the process. In fact, since the authors don’t introduce the original model, it is difficult to understand the significance of the modifications to upland volatilization. Perhaps the authors can expand this section or include a supplement that provides more detail on the original and modified algorithms. Also, how the f values are calculated is also missing in the documentation.
2. Were NH3 emissions calculated differently from different fertilizer applications (e.g.., manure vs. synthetic fertilizer or urea vs. ABC) or was only synthetic fertilizer applied at the sites?
3. For eq 1, how does the depth of the fertilizer application play a role in the NH3 flux?

Line 267-272: As noted previously, can the authors provide a description of the method used to calibrate the model? More explanation is needed for the process used to calibrate parameters. Because there isn’t a good introduction to the initial model, I am left to wonder about the role the parameters play in the emissions process and how sensitive the model is to parameter modification. This would be another good opportunity for a sensitivity analysis. Also, are the final parameter values used for all the sites or are they site/environment specific. Perhaps include a table of the parameters changed and their initial and final values would help.  

Section 2.4: I think it would also be interesting to look at a sensitivity of the model parameters as well. This would provide an indication of which parameters have the largest influence in the model simulations of emissions and which parameters are

Section 4:

1. The section begins with discussion of the factors affecting emissions and moves into why the model performs poorly for certain conditions. Perhaps split into two sections or separate in different paragraphs.
2. Section 4.2 begins by unnecessarily repeating methods (line 473-483), and again discusses model limitations rather than influences on emissions as the section heading suggests.
3. I suggest a separate section for model limitations or changing the header of 4.1 and 4.2.
4. Section 4.4 reads more like a conclusion than the actual conclusion in the paper.

The conclusion feels like more of an abstract and repeats methodology.

Data availability: Providing model output is useful but does not allow duplication of this effort. A suggestion would be to provide the actual model used by the authors.

Technical comments:

Line 16-18 (and elsewhere): “…evaluated and modified using NH3 volatilization observations from 44 and 19 fertilizer application events in cultivated upland areas and paddy rice fields in China, respectively.” The wording is slightly confusing. I suggest “evaluated and modified using NH3 volatilization observations from fertilizer application events in 44 cultivated upland areas and 19 paddy rice fields in China.”

Section 2.1: The authors should reference Table S6 when discussing the upland sites.

Line 451: is missing a space between placement and 5.

Text is awkward, particularly Line 452-454.

Section 4.3 is missing; the titles jump from 4.2 to 4.4

References:

Riddick, S., Ward, D., Hess, P., Mahowald, N., Massad, R., and Holland, E.: Estimate of changes in agricultural terrestrial nitrogen pathways and ammonia emissions from 1850 to present in the Community Earth System Model, Biogeosciences, 13, 3397–3426, https://doi.org/10.5194/bg-13-3397-2016, 2016.

Vira, J., Hess, P., Melkonian, J., and Wieder, W. R.: An improved mechanistic model for ammonia volatilization in Earth system models: Flow of Agricultural Nitrogen version 2 (FANv2), Geosci. Model Dev., 13, 4459–4490, https://doi.org/10.5194/gmd-13-4459-2020, 2020.