Factors That Influence Nitrous Oxide Emissions from Agricultural Soils as Well as Their Representation in Simulation Models: A Review
Round 1
Reviewer 1 Report
The manuscript has been sufficiently revised. I only have minor comments which could lead to further improvements and should not require a lot of additional work from the authors.
Comments:
Table 2 : the authors could use “+” and “-" instead of “I” and “D” to show increase/decrease
Table 3 : this is now better than it was before
Figure 3 : There are many messages that the authors try to convey through this figure. It takes the reader quite some time to get the message(s) by looking at this schematic to the point that some may just skip it. I believe that a schematic that allows the reader to understand the “factor” by looking at the figure rather than reading the title will be better. In other words, it may be worth the effort to explore how the use of images (along with text) can improve this figure.
Line 596-598 : Soil N2O measurements do not influence N2O emissions but they affect the accuracy of measured N2O data, which in turn affect various modelling stages (development, parameter optimisation, validation).
Line 598-600 : this sentence adds nothing new — you can remove it
Line 606 : Table 3 does not say anything about measured data uncertainty. It shows how varied N2O EFs are across different crops, climates and soil conditions. I think the authors need to focus the text in section 2.3 to those aspects that cause soil N2O measurements to be notoriously variable i.e. intra-field variability in soil conditions (aeration, ph, N availability can vary at a micro-level and with depth), not-uniform crop growing patterns (random or caused by pests, diseases)
Line 651 : could add Arjan Hensen et al 2013 Environ. Res. Lett. 8 025022
Lines 718-720 : Would be nice to add this code to supplementary material e.g. GitHub link if possible and willing
Line 683 : the term “system dynamic” is confusing; in general, the terminologyused to describe models is. I think the authors can remove this term from the manuscript (I think it appears twice) since they differentiate between “empirical” and “process-based” models in the modelling section.
Line 939 : It should be clear (in introduction, discussion, conclusions) that measurement factors do not influence soil N2O emissions — they affect our quantitive understanding of them. Measurements are the “rather weak” link between what we know (or think we know) and how we model it.
Line 980-982 : this sentence needs rewording e.g. most model-based studies focus on regions where field measured data are available for use in model calibration and validation/testing œ
Line 983 : either remove “instead of only comparing the simulated N2O emissions to measured N2O data“ or state that you “recommend a more “holistic approach” to model calibration and validation that considers (or integrates measured data on) crop N , N leaching etc… “
Author Response
Point 1: The manuscript has been sufficiently revised. I only have minor comments which could lead to further improvements and should not require a lot of additional work from the authors.
Comments:
Table 2: the authors could use “+” and “-" instead of “I” and “D” to show increase/decrease
Response 1: Thank you for the review and for the valuable comments. We have changed “I” and “D” to “+” and “-” in Table 2 (L445).
Point 2: Table 3: this is now better than it was before.
Response 2: Thank you for the kind comment.
Point 3: Figure 3: There are many messages that the authors try to convey through this figure. It takes the reader quite some time to get the message(s) by looking at this schematic to the point that some may just skip it. I believe that a schematic that allows the reader to understand the “factor” by looking at the figure rather than reading the title will be better. In other words, it may be worth the effort to explore how the use of images (along with text) can improve this figure.
Response 3: Thank you for the suggestion. We have modified Figure 3 by adding colour to make the relationship of the factors more visible to the reader.
Point 4: Line 596-598: Soil N2O measurements do not influence N2O emissions but they affect the accuracy of measured N2O data, which in turn affect various modelling stages (development, parameter optimization, validation).
Response 4: We have elaborated this sentence to “The measurement factors do not directly influence N2O emissions (although disturbance of natural conditions may occur when taking a sample, e.g. with chambers). However, the measurements are important factors to report because they affect the accuracy of the measured N2O amount and are useful for reporting on the uncertainties of the N2O measurements. The N2O measurements are a link to our understanding of what happens in the soil and what can be modelled. The measurements therefore also influence various modelling stages (e.g. model development, parameter optimization and model validation)” (L596-L604).
Point 5: Line 598-600: this sentence adds nothing new — you can remove it.
Response 5: We have removed “Measurement factors pertain to the accounting of the N2O emissions, which is important because of the great spatial and temporal heterogeneity of N2O.”
Point 6: Line 606: Table 3 does not say anything about measured data uncertainty. It shows how varied N2O EFs are across different crops, climates and soil conditions. I think the authors need to focus the text in section 2.3 to those aspects that cause soil N2O measurements to be notoriously variable i.e. intra-field variability in soil conditions (aeration, ph, N availability can vary at a micro-level and with depth), not-uniform crop growing patterns (random or caused by pests, diseases)
Response 6: Thank you for the comment. We removed the sentence “Table 3 provides an overview of the range of N2O measurements obtained with respect to geographic regions, soils, crops and fertilizer types and amounts.” To address the uncertainties in the measurements, we added two studies that point to the uncertainties of measured N2O emissions in the same field (L653-L660).
Point 7: Line 651: could add Arjan Hensen et al 2013 Environ. Res. Lett. 8 025022
Response 7: Thank you for the recommendation. We have added Hensen et al. (2013) (L663).
Point 8: Lines 718-720: Would be nice to add this code to supplementary material e.g. GitHub link if possible and willing.
Response 8: We have added the GitHub link in the supplementary material (https://github.com/snailslowrun/N2O/blob/main/comparison.R).
Point 9: Line 683: the term “system dynamic” is confusing; in general, the terminology used to describe models is. I think the authors can remove this term from the manuscript (I think it appears twice) since they differentiate between “empirical” and “process-based” models in the modelling section.
Response 9: We have removed the term “system dynamic” and replaced it with correct technical term “mathematical” (L697) and removed it in the caption for Table 4 and on L685 changed the wording to “process-based”.
Point 10: Line 939: It should be clear (in introduction, discussion, conclusions) that measurement factors do not influence soil N2O emissions — they affect our quantitive understanding of them. Measurements are the “rather weak” link between what we know (or think we know) and how we model it.
Response 10: We have elaborated on this important point in the introduction, discussion and in the conclusion (L132-L134, L596-L604, L947-L951).
Point 11: Line 980-982: this sentence needs rewording e.g. most model-based studies focus on regions where field measured data are available for use in model calibration and validation/testing.
Response 11: We have rewritten this sentence to “Most model-based studies focus on regions where field measured data are available for model calibration and validation” (L988).
Point 12: Line 983: either remove “instead of only comparing the simulated N2O emissions to measured N2O data” or state that you “recommend a more “holistic approach” to model calibration and validation that considers (or integrates measured data on) crop N, N leaching etc… ”
Response 12: We reworded the sentence to “We recommend a more holistic approach to model calibration/validation whereby several simulated variables related to N2O emissions in the model, such as soil NO3-, soil water, or crop yields should be compared with measured data when possible, as this would improve the simulation of N2O in the soil system.” (L991).
Reviewer 2 Report
In my opinion it is improved in the great level what indicates that Authors tried to do their best and collaborated with reviewers and Agronomy Editors. I have no comments to the re-submitted manuscript and in my opinion it can be published in Agronomy.
Author Response
Thank you for the review and kind comment.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The article is useful in connecting dimensions of influence over agricultural N2O emissions and modeling. Care needs to be taken not to overstate or state without justification in places. An explanation for the selection of three models for the detailed look should be provided. While the paper could be useful, it finishes in somewhat disappointing fashion. It could be more interesting and powerful if tradeoffs of the models and evaluation of how their approaches align with the factors included in the review were presented in a more thorough and organized manner.
Comments for author File: 
Comments.pdf
Reviewer 2 Report
This paper as a review of factors which cause nitrous oxide emissions from the agricultural soils is a knowlegde compendium on this issue. Therefore in my opinion it is important contribution to science. I think that this paper will be interesting for students and scientific workers. It contains also numerous literature what would be helpful for people interested in this problem. One of my suggestion is too little real data of nitrogenous emission in dependence on various factors like for instance different doses of fertilization, different soil moisture or soil temperature. An additional advantege of this paper is collection of models which are used to simulate nitrification and denitrification in the soils.
Reviewer 3 Report
This is a nice review of measurements-based and model-based literature on soil N2O emissions. It is a useful manuscript that explains how N2O emissions arise in agricultural soils and how widely-used models describe these processes. The authors could use more figures to show how environmental factors affect emissions (e.g. see figs in https://royalsocietypublishing.org/doi/10.1098/rstb.2013.0122) This will allow them to use fewer words in describing the key factors/processes. Also, I feel that Table 2 can be expanded with more references.
Moreover, I believe that there should be more text on the uncertainty around measured N2O data, what causes it and how significant it can be; something more quantitative than simply stating the uncertainty is large. Such new text can be part of section 2.3. of the manuscript. By doing that, the authors can build a link with their N2O modelling section and discuss how models deal with uncertainty. After all it is through relying on measured data that models are developed and their credibility proved/disproved.
The last paragraph of the manuscript shows why discussing the issue of data/model uncertainty is important. Models have a theoretical structure, which has uncertainty e.g. models consider the role of soil depth for biogeochemical processes in different ways or not at all. Models have parameters, which are also uncertain; e.g some model parameters cannot be even measured in reality (e.g. biological activity related). How are model-based N2O predictions assessed against measured N2O data, what do such comparisons show and how do they inform us on how much we should trust model-based predictions (and in a way the models themselves)?
Below are some references that could help the authors navigate through the relevant literature. Please accept my apologies for some of them being only a DOI link and if some of them are already used in the text.
In general, since this is a review I would recommend to ensure that there is a decent number of recent (2015>) relevant publications among those that are cited
I would be happy to re-read this manuscript.
https://www.sciencedirect.com/science/article/abs/pii/S026974911000535X?via%3Dihub
https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2006GB002686
https://link.springer.com/article/10.1007%2Fs10705-011-9458-9
https://link.springer.com/article/10.1007%2Fs10705-006-9000-7
Ehrhardt, F., Soussana, J.-F., Bellocchi, G., Grace, P., McAuliffe, R., Recous, S., Sándor, R., Smith, P., Snow, V., de Antoni Migliorati, M., Basso, B., Bhatia, A., Brilli, L., Doltra, J., Dorich, C. D., Doro, L., Fitton, N., Giacomini, S. J.,
Grant, B., … Zhang, Q. (2017). Assessing uncertainties in crop and pasture ensemble model simulations of productivity and N 2O emissions. Global Change Biology, 24(2), e603–e616. http://doi.wiley.com/10.1111/gcb.13965
Wang, G., & Chen, S. (2012). A review on parameterization and uncertainty in modeling greenhouse gas emissions from soil. Geoderma, 170(C), 206–216. http://dx.doi.org/10.1016/j.geoderma.2011.11.009
Necpálová, M., Anex, R. P., Fienen, M. N., Del Grosso, S. J., CASTELLANO, M. J., Sawyer, J. E., Iqbal, J., Pantoja, J. L., & Barker, D. W. (2015). Understanding the DayCent model: Calibration, sensitivity, and identifiability through inverse modeling. Environmental Modelling & Software, 66(C), 110–130. http://dx.doi.org/10.1016/j.envsoft.2014.12.011
Fitton, N., Datta, A., Smith, K., Williams, J. R., Hastings, A., Kuhnert, M., Topp, C. F. E., & Smith, P. (2014). Assessing the sensitivity of modelled estimates of N2O emissions and yield to input uncertainty at a UK cropland experimental site using the DailyDayCent model. Nutrient Cycling in Agroecosystems, 99(1–3), 119–133. http://link.springer.com/10.1007/s10705-014-9622-0
Myrgiotis, V., Rees, R. M., Topp, C. F. E., & Williams, M. (2018). A systematic approach to identifying key parameters and processes in agroecosystem models. Ecological Modelling, 368, 344–356. http://dx.doi.org/10.1016/j.ecolmodel.2017.12.009
Myrgiotis, V., Williams, M., Rees, R. M., Smith, K. E., Thorman, R. E., & Topp, C. F. E. E. (2016). Model evaluation in relation to soil N2O emissions: An algorithmic method which accounts for variability in measurements and possible time lags. Environmental Modelling & Software, 84(C), 251–262. https://doi.org/10.1016/j.envsoft.2016.07.002
Li, X., Yeluripati, J., Jones, E. O., Uchida, Y., & Hatano, R. (2015). Hierarchical Bayesian calibration of nitrous oxide (N2O) and nitrogen monoxide (NO) flux module of an agro-ecosystem model: ECOSSE. Ecological Modelling, 316, 14–27. http://dx.doi.org/10.1016/j.ecolmodel.2015.07.020
Lehuger, S., Gabrielle, B., Oijen, M. van, Makowski, D., Germon, J. C., Morvan, T., & Henault, C. (2009). Bayesian calibration of the nitrous oxide emission module of an agro-ecosystem model. Agriculture, Ecosystems and Environment, 133(3–4), 208–222. http://linkinghub.elsevier.com/retrieve/pii/S0167880909001339
Snow, V. O., Rotz, C. A., Moore, A. D., Martin-Clouaire, R., Johnson, I. R., Hutchings, N. J., & Eckard, R. J. (2014). The challenges - and some solutions - to process-based modelling of grazed agricultural systems. Environmental Modelling & Software, 62, 420–436. https://doi.org/10.1016/j.envsoft.2014.03.009
