The present work provides an evaluation of NO₃ radical fates in a semi-rural site thanks to direct NO₃ reactivity measurements during the TO2021 campaign in summer 2021. A Flow-Tube Cavity Ring Down Spectrometer (FT-CRDS) setup was used to measure the NO₃ total reactivity and to estimate the contribution of BVOCs to this total reactivity. During this campaign, a number of other relevant measurements (NOx, O₃, actinic flux, VOCs, …) were performed to allow for a comprehensive interpretation of the observations.

This study is fully relevant and the FT-CRDS is a very interesting technique to better understand the role of NO₃ in the night-time chemistry. The paper is well written and provides detailed information on the experimental setup as well as a very thorough interpretation of the observations, and it is very much appreciable. In general, the scientific quality of this work is very good and once the authors have addressed the following minor points, I would be happy to recommend its publication in ACP.

Specific comments:

• 61: more detailed reactions should be provided to better explain the formation of RONO₂ from VOC+NO₃ reactions
• 175: it is not clear why the PTR-MS (VOCUS) was not calibrated with the standard used for the other PTR-MS (Ionicon). Could the authors provide an explanation?
• 170 and 290: the authors mention that sesquiterpenes were measured but no data/plot have been provided. If available, please provide these data in Figure 3 or in SI. Were sesquiterpenes monitored during previous campaign using GC techniques? Even though sesquiterpenes mixing ratios are expected to be very low, they are suspected to significantly contribute to NO₃ fate due their high reactivity. More information about the role of sesquiterpenes on NO₃ loss would be useful.
• The authors do not consider the role of RO₂ radicals in the NO₃ Do they consider that it is negligible? Previous field studies (e.g. Sommariva et al, 2007) suggest that reactivity with RO₂ radicals is not negligible even though RO₂ concentrations are very low. This point should be discussed and arguments should be provided for not considering these reactions.
• 324: the authors cannot conclude that NO₃ significantly contributes to the BVOC oxidation during the daytime just because the reactions with BVOCs have been shown to significantly contribute to the NO₃ total reactivity. To state that, the BVOCs lifetimes due to NO₃ oxidation should be compared to those estimated for OH chemistry (using typical OH concentrations).
• 340-387: A very detailed discussion on the NOx budget is provided but does not seems to be fully relevant here, in my opinion. It’s not clear for me what the authors want to demonstrate here. As a minimum, this should be provided with a clearer objective and in a dedicated section. It has nothing to do in the section “fractional contribution of VOCs to NO₃ losses”.

The authors report measurements of NO3 reactivity in a summer campaign at Kleiner Feldberg in Germany. They analyse the measurements in terms of different contributions of reactants to the total NO3 loss and draw conclusions about emissions and losses of nitrogen oxides at that place. Results are compared with results from previous campaigns. The manuscript is overall well written and within the scope of the journal. There are some open questions and simplifications that need to be further explained and discussed before the manuscript can be published.

L29: NO2 is only a net ozone production, if emitted as NO2. The majority of net O3 is produced from peroxy radical reactions with NO.

L140: What could have been the reason for the higher loss rate in the large tube?

L143: It would be good, if numbers for the NO3 concentrations that is used in the reactivity instrument were given and compared to ambient concentrations that are expected.

L161: Did you test if the calibration gas standard used to calibrate the CLD gave the correct concentration in the CRDS instrument?

L175: Why was the VOCUS PTR-MS not calibrated with the same gas standard as the other PTR instrument? Was the scaling factor that needed to be applied to the VOCUS instrument constant for a specific mass for the period, when both instruments measured together?

L195: How was the zero-value determined of the CLD? How often was this done and how stable was the zero?

Fig. 1: Why is there only a limited period of monoterpene measurements shown, if the VOCUS PTR was used to complete the measurements as shown in Fig. S4?

L253: Was the height of the vegetation below the tip of the inlet?

Fig. 4: It would be useful to indicate the inlet height and the height of the vegetation.

L 297: It does not make sense to give 3 counting digits for the fractional distribution, if the accuracy of measurements does not provide this accuracy.

L289: Can you justify, why you expect the same contributions of monoterpenes like in the other campaign? Seasonality, meteorological conditions, changes in the vegetation may highly impact the mix of emissions. This should be further discussed and not neglected as indicated in the in the text.

L290: Why is beta-caryophyllene a suitable proxy for the measurement of the sum of sesquiterpenes?

Fig. 5b: It is not clear, what the grey boxes are.

Fig. 5c: Was there no other (unaccounted) NO3 reactivity on average?

L 294 and Fig. 5a: The figure gives the impression that monoterpene species can explain the NO3 reactivity. However, it would be easier to judge this if the x-scale was wider and/or the time period was split into 2 panels.

L 303ff and Fig. 6: It looks as if there are more data points than shown in the figure. The large symbol size and using also black colour for the error bars makes it is hard to see details. What is the correlation coefficient? The distribution is very wide and shows that there are also a high number of points where numbers are not the same. A plot of the time series of the difference between calculated and measured NO3 reactivity could give more insights if this is due to statistically or systematic differences during specific periods of the campaign.

L314ff and Fig. 7: It should be emphasized / defined that NO3 reactions with VOCs are meant, if you say “fractional contribution F”.

L319: What are the reasons for the increase of the contribution of NO3 + VOC reactions?

L324: Can you give an estimate how the reaction rate of VOCs with OH and O3 were during daytime to support your statement about the importance of NO3 reactions for the oxidation of BVOCs during the day?

L327: Do you want to say that local anthropogenic emissions existed only during daytime? Why would this be the case?

L339ff: As discusses a bit later, you may expect a strong gradient of NO concentrations with height also within the surface layer due to the rapid reaction with O3 unless the mixing is fast, which may not be expected specifically in the night. Does your estimate of the NO concentration consider such a gradient, if the inlet of the NO instrument is at a certain height? I assume that the NO source from soil would need to be significantly higher, if this is taken into account.

L366ff: Does the model also include O3, NO3, N2O5 deposition? If so, this should be mentioned, if not it needs to be justified, why no deposition is assumed. These loss processes would all contribute to the loss of odd oxygen, and it may not be easy to distinguish between the loss for the different species. How can you justify that a 0-D box model is applicable for modelling measurements made close to the ground in night-time conditions, when mixing is poor?

L456ff and Table 1: It would be good to see the comparison of NO measurements and have this discussed in more detail the text. If the explanation for the high NO in this campaign is soil emissions, what for example could be reason, why this was not observed in the other campaigns?

Technical comments:

General technical comment: It makes it easier to read and follow the manuscript if less abbreviations are used in the text.

L36: “OH reactions being most important” instead of “OH reactions most important”

L61: subscript RONO_2

Fig. 2: You may need to increase the font size if these figures become 1-column figures.

L312: comma missing after (j_NO3).

L401: Units of productions rates are pptv s-1 and not s-1.