General Comments:

This manuscript is a thorough and insightful exploration into the drivers of soil N2O emissions. Building on a well-researched field, the paper employs strong and novel methods to demonstrate this idea through use of spatio-temporally integrated flux measurements and frequent measurement of crop growth dynamics to estimate soil nitrogen dynamics, plant nitrogen uptake, and N2O emissions over time. The experiment appears well conducted and to have yielded a wealth of data, to which the authors add investigative statistical methods which contribute a strong picture of the drivers of N2O and how these drivers shift over time, which I found particularly insightful. The manuscript is also clearly written, if overly descriptive and verbose at times. 

However, I have key concerns regarding the methodological rigor and the resulting interpretations in three key areas. First, there is significant opacity regarding the data handling process during machine learning, including a high risk of temporal data leakage during model validation. Second, the methods used to classify 'background' vs. 'hot-moment' emissions are tenuous and lack a mechanistic basis. Third, while the authors identify Gross Primary Productivity (GPP) as a preeminent negative driver of emissions in their discussion of results, the exclusion of soil mineral N concentrations from driver analysis creates a risk of omitted variable bias, in which GPP may act as a proxy for N exhaustion. To justify their major claim that GPP is a suppressor of N2O emissions and that N synchronicity is a path toward N2O mitigation, the authors should make efforts to decouple GPP from simple substrate limitation in their statistical investigation of drivers.

With a tighter text and these improvements to statistical methods, I am confident this manuscript will represent a strong and novel contribution to understanding pathways to agricultural GHG mitigation.

Major Comments:

L55-56: Is this sentence referring to nitrogen cycling or crop growth?

L80-84: This is true, strictly speaking. However, it is also possible to infer the dominant N-transforming process based on the rate of N2O production, as hot moments have been shown to be overwhelmingly the result of denitrification. Relatedly, a recent work used ML to differentially model nitrification-dominant and denitrification-dominant emissions (Lussich et al., 2026) https://doi.org/10.1002/jeq2.70126

L84-90: I’ll use this passage to illustrate a broader observation that characterized much of this manuscript. The authors frequently provide detailed explanations, such as the principles and mechanics behind stable isotope analysis. While thorough, these explanations are often characterized by a poor economy of words, include details which are not directly pertinent to the narrative at hand, and add up to a manuscript of considerable length (I count nearly 13,000 words). This work would significantly benefit from greater concision, illustrating only the most pertinent details and taking advantage of the expected level of familiarity of a biogeoscientist audience.

Table 1: slurry was applied the day after mineral N fertilizer was applied. Is this typical practice in the region? This creates perfect C and N chemistry to produce N2O. A study focused on understanding N2O mitigation opportunities, this is not an efficient fertility management decision. Can slurry be applied in fall, before wheat planting?

L162-163: Why were two separate outlier detection methods used to filter erroneous fluxes?

L173-184: Regarding the RF gap-filling: was there any analysis of error propagation? Or an assessment or justification of the accuracy of RF as a gap-filling method? Accurate gap filling N2O emissions is still a topic of much research, and has resulted in mixed success. One of the papers cited here reports R2 of gap-filling for N2O between 0.6 and 0.76, using only 15% of ‘missing values’ as the test set, in comparison to ~47% of the time period missing in this study. Other work has reported lower R2 values from 0.42 (Taki et al., 2019, 10.1139/cjss-2018-0041) to 0.66 (Goodrich et al., 2021, 10.1016/j.agrformet.2020.108280), also using just 15% of the data as ‘missing values’ to be filled. The other paper the authors cited here performed no analysis on the accuracy of the RF gap-filling method used. The success of ML gap-filling has also been shown to be related to the length of gaps (Taki et al., 2019), yet there is no discussion of typical gap length or the impact of lengths beyond reporting the aggregate percentage of missing values. While I acknowledge that flux gap-filling is a major challenge which has yet to be solved, and that work must go on in the meantime, nevertheless I feel it is important to acknowledge the limitations of gap-filling high resolution N2O flux data and the effect that these limitations might have on this study.

L183-185: It seems here like the authors estimated the background flux level by just excluding 30-day post-fertilization flux data with assumption that fertilization effect lasts only for 30 days. First of all, I don’t think this assumption is correct and your own data did not support the assumption. Fertilization can have long-lived impacts on N2O emissions, particularly during dry periods, offseason, etc. As well, not all hot moments are fertilizer-driven, as those like Claudia Wagner-Riddle’s group have shown. Moreover, not all fertilizer-driven N2O emissions are caused directly by fertilizer at all: the excess N added  to the soil system by fertilization may be temporarily captured by plant or microbial biomass and mineralized months or even years later, thus driving emissions that would not occur in a natural, unfertilized system but which nevertheless occur distantly from any fertilization event. The authors’ data shows this: the large peaks, as big as post-fertilization, from mid-June to August is well beyond 30-day post-fertilization period and unlikely to happen in a truly unfertilized control treatment. Moreover, the SHAP analysis (Fig 4) also shows that fertilization effect was a dominant factor for almost two-months after fertilization. 

I have serious reservation about this method of distinguishing between HM and BG emissions by simply excluding fluxes within 30 days of fertilization. This potentially inflated the background emissions and perhaps underestimated emission factors. Authors might consider alternative methods of contextually distinguishing background emissions from hot moment emissions based on outlier detection. Please see Ackett et al., 2025; doi.org/10.1029/2025JG008953.

L199-201: Spatially aggregated flux measurements are captured across a field using EC, yet a single point measurement is used for soil moisture content and temperature? Soil moisture is also highly spatially heterogenous. This seems a potentially noteworthy limitation.

L303-305: Again, the persistence of elevated N availability is highly variable and can be much longer than 30 days.

L318-325: A more complete description is needed about the data splitting process is needed in order to verify its validity. In this custom time-block method, how large were the time chunks? Was the model trained on data further in the future than the test data? When working with a single time series, the most correct method of cross validation is to use a method similar to that employed by the TimeSeriesSplit function in scikit-learn. By this method, the timeseries is split into n+1 chunks, and the model is trained on all prior data within the timeseries, including an initial runup chunk (i.e. an expanding window). This ensures that the model is never trained on future data, which would constitute a form of data leakage.

L323-325: RMSE and R2 both give disproportionately large influence to hot moments by way of heavily weighing large residuals. Consider using an evaluation metric that evenly weights residuals of all sizes, like MAE, to give a more balanced evaluation of model performance across the distribution of values.

L328-331: I have searched through the document and could not find a description of what the authors here call the “final model,” nor what the “test set” was. There seem to be a lot of mixing of terminologies regarding data handling and model evaluation, leading to a very opaque picture of the actual methods used. Starting at the beginning of this section:

“Following variable selection, model hyperparameters were optimized using 10-fold cross-validation. To account for temporal autocorrelation and avoid overfitting, while also providing representative coverage of the measurement period, we employed a custom time-block strategy. This approach involved an 80/20 split between training and validation…”

This seems plainly contradictory. If a 10-fold cross validation were used, then across each fold 90% of the data would be used for training and 10% for validation, yet the authors claim an 80/20 split. My best guess might be that the 10-fold cross validation might be a separate process exclusively for hyperparameter tuning, for which no details were provided, and then an abrupt shift to a new data handling process involving an 80/20 split of some kind takes place? Yet descriptions of each process are incomplete and poorly differentiated in their purposes, leading to a jumbled passage.

“with the validation set comprising randomly selected, non-overlapping time blocks that together represented 20% of the available data.”

The authors here claim that time blocks were randomly selected, but a timeseries should not be randomly split. The authors seem to acknowledge this idea with their reference to temporal autocorrelation, but their description of their methods here does not give confidence that they have properly dealt with this challenge. 

“This splitting strategy was consistently applied throughout the modeling workflow, including during cross-validation and final model evaluation.”

This does not make sense to me. There have been no clear definitions of what constitutes the “cross-validation” and the “final model evaluation.”

“Cross-validation results showed a R2 of 0.60 and a RMSE of 1.1 nmol N2O m-2 s-1 on the validation set, while the training set showed a R 2 of 0.98 and a RMSE of 0.29 nmol N2O m -2 s-1. The final model, trained with early stopping (10 rounds) to prevent overfitting, achieved a R2 of 0.70 and a RMSE of 1.14 nmol N2O m-2 s-1 on the test set (Fig. A2).”

This passage invites more confusion. Beyond the unclarity regarding CV vs final evaluation, the authors describe model performance on “the test set,” “the validation set,” etc. Was there a singular test set, as the language here suggests, or are the authors instead referring to the average model performance on holdout data across the 5 or 10 folds?

L333-349: In contrast to the preceding passage, this description of SHAP values is exhaustive, and may be excessive.

L360-384: This passage is a lengthy textual description of figure 2, which I’m not sure is necessary given the length of the manuscript.

L452-453: The average background flux estimate of 0.44 nmol m⁻² s⁻¹ is equivalent to ~11 g N2O-N/ha/d, which is substantially (3-4 times) larger that globally estimated background emissions of ~2.5 to 3 g N2O-N/ha/d. This could be due to the reasons I explained above.

L462-466: “In contrast, the cover crop acted as a net CO 2 source, emitting 108 g CO2-C m-2”

The cover crop phase was only measured for 2 months, which was influenced by cultivation after wheat harvest. Is this the typical cover crop growth duration? If not, then this is misleading information that provides a snapshot of the rotation. Cover crop active growth phase is often known to sequester C (NEE sink) and negligible N2O emissions.

As the authors correctly point out in their discussion, it is not the cover crop per se that is acting as an emitter of CO2, but rather this is the response to harvest of the winter wheat and the bare soil. It is tricky, but I wonder if there is a better way to convey that the cover crop is actually reducing emissions, not magnifying them, even though these emissions are taking place during the cover crop’s establishment. 

L480-483: in L452, the authors mentioned background flux of 0.44 nmol m⁻² s⁻¹. This is confusing as it seems different subset of data are being used in places to estimate background flux.

L486-509: SHAP values are best interpreted by relating the direction/magnitude of impact alongside the magnitude of the factor value. For example, the authors write: “N2O emissions were mainly suppressed by WFPS.” The fact that soil moisture had a large negative impact on flux predictions, as described here, leaves a lot of room for interpretation. Were these negative influences related to high or low soil moisture?

L510-515, Figure 5: The binning strategy used here to quantify the relationships between WFPS, GPP, and flux, is statistically fragile. The bins create artificial step functions in what is likely a linear relationship, and also reduces the number of data points for statistical analysis. It also is able to demonstrate an interaction exists, but does not quantify that relationship in terms of strength or behavior. A better approach might be to model a multiple linear regression with an interaction term. Partial r2 could also be used to more precisely determine the relative explanatory power of each factor and the interaction.

L559-560: True, but could this also be due to increasing contribution of nitrification to the net N2O emissions? How could you be so sure about Fig 7B secondary vertical axis?

L565-566: Relatedly, could not the increased role of nitrifier denitrification also indicate a decreased rate of denitrifier denitrification overall?

L646-650: The N2O offset effect on net CO2 is based on NEE and not actually on long-term C sink, which is soil C stabilization. Not all of the NEE sink from a growing season is translated into soil C gain that is stabilized for longer period beyond the growing season. The studies using long-term experiments showed much higher N2O offset effects in fertilized systems in other regions (see Dhaliwal et al., 2025; doi.org/10.1002/jeq2.70046). Therefore, the short-term nature of the study may have underestimated N2O effects.

L656-661: Mineral N sampling also showed modestly high N availability during this period, likely related to low uptake and mineralization from residue and microbes

Section 4.4, L705-708: I will take this line as an opportunity for a discussion on the idea of synchronicity between N supply and demand as a key feature in determining N2O emissions, and as a key idea in this manuscript. The authors place much significance on this idea throughout their introduction, use of methods, and the discussion here. In my view, the evidence for nitrogen synchronicity as the preeminent determinant of N2O emissions, rather than simply N availability, is a bit scattered. For example, Figure 4 excellently illustrates flux and GPP, yet mineral N availability is missing as the third key component needed to make this case. From Fig 4 there certainly does seem to be an inverse relationship between flux and GPP, but I wonder the extent to which GPP is also negatively correlated with N availability. Given the inverse relationship between GPP and mineral N (the nitrogen is moving directly from the soil and into the plant biomass), more evidence is needed to prove that GPP is independently negatively affecting N2O emissions, beyond soil N.

This leads me to my concern about the authors’ interpretation of GPP’s role in the SHAP analysis. Because soil mineral N concentrations were not included as input features in the Random Forest model, the model likely suffers from omitted variable bias. While 'days since fertilization' is included as a predictor, it is a very rough and linear proxy that fails to capture the dynamic, non-linear depletion of the soil N pool. In the absence of a direct N-substrate variable, the machine learning algorithm quite possibly utilizes GPP as a better mathematical proxy for the diminishing N pool. Consequently, the high importance attributed to GPP in the SHAP analysis may simply reflect N exhaustion rather than a mechanistic suppression of N2O production by plant uptake (synchronicity). To truly substantiate the claim that synchronicity is key to emissions, the authors would ideally show that for a given level of soil mineral N, higher GPP results in lower emissions. Without including measured soil N data in the modeling workflow, the current results show correlation with the crop’s growth phase, but do not sufficiently decouple plant demand from simple substrate limitation to make strong claims such as this one.

ReviewEGUsphere2026

General and specific comments

The study reports GHG fluxes by ECT in winter wheat following 2-3 year grass clover cropping and measurement of vegetation dynamics, soil moisture,  and mineral N to (i)  quantify GHG fluxes, .(ii) identify drivers of N2O fluxes (iii) and determine  N2O pathways.

A novel machine learning (ML)-based modeling approach was used to identify drivers for N2O peak events and to model N2O fluxes.

The ECT approach was state of the art and together with the soil and vegetation parameters determined  the data-set is probably unique and well suited for model testing. Isotopic fluxes were determined during 3 weeks after early spring fertilization with mineral and organic fertilizer to determine N2O pathways and reduction to N2.

While the evaluation of ECT data and the measured control factors using the KI model is certainly innovative and helpful to identify drivers of N2O in that study. But the study presents only data one site and one under soil conditions and fertilization regies which are quite special (one-year,  one soil,  without replicates, special case of grass-clover as preceding crop with very high expected N-mineralisation and mixed organic/mineral fertilization, in a  clay loam soil which is quite prone to intense denitrification), the study thus does not expand our knowledge on the processes and their control. It only confirms with several observations the well-known regulation of N2O fluxes by the interplay of fertilization, soil moisture, mineral N and crop growth. Similarly, the determination of emission factors and of the GHG budget must be seen as individual observation of that site and year that add little to regional or global estimates of these values.  

The new ML-model was used to identify drivers of N2O fluxes. Traditionally, this is done by conventional statistical  models or using process-based biogeochemical model like, e.g.,  DNDC. There were no attempts to compare the new model to traditional approaches. Moreover, the model is not addressed in the discussion. From my view the data-set and the evaluation reported in the paper are suitable for a methodical paper that focusses on the new model and how it relates to previous models. But the discussion largely neglects the new model and focusses on the question (1.) how the fluxes were regulated, (2) how the magnitude of N2O fluxes relates to previous studies and  (3.) what we can learn from the data on N2O mitigation strategies. While the first aspect, as said before, does  not add any new insights, the study does ono add any basis for the third aspect.

The paper is thus not suitable for publication in the present form. I suggest major review which should include   model comparison as explained above and removal of the current discussion on emission factors, flux regulation and N2O mitigation.

There are some further shortcomings in the paper regarding the GHG budget calculation, the determination of emission factors, the calculation and interpretation of the isotopic data, the presentation and discussion of the N budget and of NUE. These points are addressed in the detailed comments below.

Details:

Line 45f pH as control of denitrification product ratio missing

Role of N2 related to total N loss, yield and   GHG cost for replacing loss of fertilizer N and role of N2O reduction to N2 to understand N2O regulation is missing in introduction and also in the evaluation of the N2O drivers

L90 relevance for isotopocules to determine rN2o missing

L91-96 need to stress that process identification with stable isotope was only for a few weeks.

L103 supply full soil info: pH, bulk density, total C, total N, soil classification.

L110-118 give more details on tillage technique and straw management after wheat as this is a key control of N2O

Table 1: unit of Nin/out missing

Intersting that NH4NO3 and slurry were applied simultaneously. Consider that labile C of slurry might create hot spots of denitrification from NO3 of min. fertilizer. And that O2 consumption during nitrification of NH4 might also cause short-term anaerobic hot spots.

More details on slurry needed(volume Norg and NH4 content,  and “+18” is not clear to me

2.2 skipped, not my expertise

L185-190 background flux calculation not clear. I think it is not possible to get a robust estimate of N2O fluxes with zero fertilization from the EC data of one field. Please discuss uncertainty arising from this simplification and clarify in all parts of the paper that the reported EF is an approximation.

L199-207 one moisture profile is not enough for one field due to variability. Not adequate to use old data for BD in top soil. Please discuss these limitations

L 243 need to calculate uncertainty of soil emitted isotopes ((Wu et al., 2019). SD of isotope analysis should be reported

L254 need to report results of all scenarios and uncertainty given by Frame

L423 better keep consistent wording (N2O fluxes) to avoid misunderstanding

L436 keep consistent wording. This sentence suggests that net ecosystem flux is different from emission, but you use both expression for the same flux. So what do you mean by “indicating”

Section 4.3.4: why was there continuous positive CH4 flux? The soil was mostly rel. low in WFPS, eg clearly above 0.8 and N2o flux was low. Thus it would be expected that the soil acts mostly as CH4 sink. This questions validity of EC results. Or was there evidence for positive CH4 fluxes from the chambers? An explanation might be the decay of the grass-clover sward in the clay loam leading to CH4 production. But this would indicate strong reducing conditions which would be highly relevant for the N2O production and reduction. So this should be discussed.

L450 please check units: 0.8 nM/m2/sec gives  5.6 g N/ha/282 days so your numbers seem to be mMol/m2/sec

L278 need to explain how the root simulators are different in estimating plant available N, since Nmin is plant available. Also explain why 0-0.1 m was chosen as rooting depth is much deeper

L 303 not clear why N fertilization was used as a proxy for N availability instead of measured  Nmin in cores or accumulators. To train models of N2o, typically Nmin is measured in top soil at each gas sampling event.

L 308 cumulative N is a rough predictor of Nmin. Because this was strongest parameter it is not clear why measured Nmin was not included. Please explain or remove.

Unit nM/s2/sec: please refer to identical units: fertilized  area is kg /ha.N2o should be reported accordingly. t

L326,327 not clear what these variables mean

L450: cumulated flux in kg N/ha does not fit avg value in nM/m2/s. apparently the data shown are in µMol.

L453 – 450  Calculation of EF based on estimation of unfertilized N2O fluxes from phases of background flux is not adequate as this is affected by seasonal effects (temp, moisture, Nmin of unfertilized from  mineralization-nitrification-immobilisation. Please delete estimation or explain that a very rough approximation was used to get an EF estimate.

L 466 CO2 uptake during cropping is not a climate benefit as the fixed C is not stable unless grain and straw were long-term deposited which is quite improbable. Thus it does not make sense to be compared with GHG effect of N2o flux. Clearly, not NEE but NEE minus  net C export must be used in GHG budgets.

Please correct or delete paragraph and delete last line of Table  with  the GHG budget and report only the fluxes. Please double check the CH4 fluxes. Positive flux in wheat is not plausible. Please explain reasons  for net source (shallow groundwater, stagnic soil properties?)

L512 avoid unclear interpretations in results. PS is not a direct regulator of N2O. It is not PS itself but the multitude of plant growth effects (exudation, rhizosphere respiration, water uptake, Nmin uptake,  etc.), please address in discussion accordingly

Section 3.5: Stable isotopes also yield info on N2O reduction, thus this aspect should also be covered by the heading.

Please state and discuss whether and why  you use the mixing-reduction  or reduction -mixing scenario in FRAME and how results would change if you assumed the other scenario. Please also discuss the rising uncertainty in isotopes of soil emitted   with decreasing flux. As far as I know, the FRAME model does not yet add this effect to the reported uncertainty. But the effect is addressed in detail in Wu et al 2019. rN2O of march 17 might  be highly uncertain due to the low flux.

Fig. 7 Not clear what is shown here residual N2O fraction or unreduced N2O fraction?  r explained in the methods is: “The contributions of different N₂O production pathways and the residual fraction of unreduced N2O (r) were estimated with the FRAME software”

The  residual N2O fraction is equivalant to the N2O/N2+N2O ratio, ie lowering values show increasing N2o reduction to N2.

But in Fig 7ba it can be seen from the smaller circles along the reduction line that rN2O sjould be is lowering over tie . Hence it seems you show 1-r which corresponds to the expression in the caption “reduced fraction”.

L556: to be clear better state: d18O and SP of emitted N2O

L 557 better name it “heterotrophic bacterial denitrification” since AOB are also bacteria

L 583 not plausible that GPP is a direct driver of N2O but rather its relation to N uptake, exudation and water uptake

4.1 This descriptive discussion does not yield any new insights but just summarizes that N2O fluxes were controlled as known since long.

L 640-644 not clear which numbers are compared current study to Maier 2022? High NUE in this study is probably not related to wheat cropping but to previous crop  (grassland conversion) delivering high amounts of Nmin from sward mineralization. Please note that 3 year grass-clover is often used in organic agriculture to feed 2 follow-up harvest of winter wheat without adding mineral fertilizer, which illustrate the high delivery of plant available N from this previous cropping season.  

L646 revise CO2 in GHG budget (see above)

L 654 role of vegetation not new at all

L 700-703 only one peak event investigated thus not representative for the cropping period

L703f not adequate to derive mitigation strategies based on a  one year study with only treatment. A multitude of field plots studies has investigated this before. The current study may be in agreement with the previous ones, but this single observation can’t be used to confirm the previous results, and certainly not to propose N2O mitigation strategies.

Di W. et al. 2019. Quantifying N2O reduction to N2 during denitrification in soils via isotopic mapping approach: Model evaluation and uncertainty analysis. Environmental Research, 108806.