Dear Authors,

Thank you for the responses to the reviewer’s comments. The discussion showed that your work can become a valuable contribution to our scientific field, provided major revisions of the manuscript as requested both by the two referees. You have addressed many of the critical points, but I will finally leave it to the referees to judge whether this is satisfactory.

Below I allowed myself to highlight some aspects that I ask you to very carefully consider in the revision. Please follow also carefully the BG guidelines, when preparing the revision.

With regard to the comments of referee 1,

I agree that one can easily think of the evaluation of the upstream processes or the full production chain. Therefore, it must be made very clear what the (limited) scope of this manuscript is, but in the discussion, it the restricted scope should be mentioned again and briefly discussed what else would be needed to evaluate the whole chain. This is necessary for the readers not to draw wrong conclusions. The conclusions should also refer very clearly to the limited scope of the study to avoid any misinterpretation.

Ref “P4, line 18: Please add the name of the model and give a referenceref”

In your response you imply that a satisfactory description of the model may not exist. Am I right?

The argument that the model has been used in other published work does not substitute the apparent lack of the model description. Using a model in science requires the description of the theory and how it is implemented in the model code. Furthermore you need to refer to how the model was tested and whether or not it was supported by the tests. If you are not able to refer to a published description, you need to describe the model yourself or use another model. Consider including the model description in the appendix .- basic equations, drivers, parameters, outputs, consideration of uncertainty, tests.

Ref: “P12 line 32: How do you justify to set negative values to zero?”

Consider using a better suited function that doesn’t force predicted data into the negative domain.

Ref: “P13 line 7-14: Please define the three sectors. I suggest to insert (<0.27, 27-33, >0.33) after “. . .by three different VWC sectors.” It would help to show this in a graph.”

Consider adding a graph to an appendix


Referee 2:

Ref: “L19 to 20 – “the fertilization rate was in the order of 120 kg N ha-1 per year between 2007 and 2015”. Can you please specify the fertilization timing?”

Please reconsider you answer. Are the background emissions during the campaign not depending on the timing of previous fertilizations? Sometimes a small amendment of information can make the article relevant for other work, e.g. reviews, which will increase its scientific value.

Ref: “L21 to 22 – “The chamber was also equipped with a GMP343 (Vaisala, FL) CO2 probe to measure the soil respiration.” Do you show this CO2 data? If not, I suggest excluding this sentence”

Both referees point to this. You can avoid this criticism by including a subordinate clause mentioning the purpose. BTW avoid the tautology “CO2 soil respiration”

Ref: “L9 – Does this sonic anemometer infers the air temperature based on the sonic temperature or it has its own temperature sensor?”

Does this mean that you have recalculated the air temperature from the sonic temperature? This is prone to errors, because some sonic anemometers have obvious problems to give accurate temperature readings. Did you check your sonic anemometer fir such errors?

Ref:” Page 8 L1 – “In order to minimize the effect of non-stationarities in the time series, the 30 min flux was finally calculated as average over six 5 min subinterval flux values.”. I wonder what would be the effect of this averaging approach on the low frequency spectral losses of their EC system. Furthermore, if you are already screening the data for non-stationarity (page 8 L24) why to estimate fluxes for these short time intervals?”

In your answer you mention that the use of 5 minutes measurement intervals lowered the effects of non-stationarity. Did you test this?

And if yes, which of the 5 minutes intervals did you choose to represent the 30 minute average?

If you used them all and averaged them, can you explain why that still reduces the effect from non-stationarity at the 30 minute time scale?

Wouldn’t you rather need to test whether the 5 minute interval was stationary and then give high-pass corrected flux values from all 5 minute covariances?

A small self-critical discussion of this procedure is advised.

Ref: “L7 – “half-hourly damping factors”. Do you mean dampening factor?”

It looks as if damping was the more accurate term.

Ref.: “L6 – In this section, it would be important to provide the spatial resolution of the grid used for upscaling the chamber fluxes. More details are also necessary on how the authors went from the output of Eq. 2 to the scaled fluxes. Did you generate digital maps of source emissions and then overlapped these maps with a footprint map? What was the software used to do these calculations?”

You forgot to mention the software.

Ref.: “Page 17 L23 – “emission optimum”. What does the word “optimum” mean here? Low N2O emissions?

No, we meant emission maximum. The term “optimum” is often used in this context (e.g. Butterbach-Bahl et al., 2013).”

Your answer does not allow to judge whether you will adopt writing what you mean or what Butterbach Bahl et al. (2013) wrote, i.e. contrary to what you mean. Please be accurate in your responses.


With kind regards,
Andreas Ibrom

Dear Authors

I am very pleased to inform you about the positive evaluation of your revised manuscript and that both the referee and I were very happy with your explanations.

With kind regards,
Andreas Ibrom