This manuscript presents observations of nitryl chloride (ClNO2), along with several precursor compounds, measured over several seasons at 50m above the surface at a rural site in western central Europe. ClNO2 is an important source of the powerful atmospheric oxidant atomic chlorine, and has been shown to demonstrate significant spatial variability due to its relatively complex production mechanism. Although many measurements have been made of ClNO2 at surface locations, the reduced concentrations close to nitric oxide (NO) emissions mean it is likely to be more efficiently produced in the nocturnal residual layer, from where subsequent mixing will allow it to influence surface photochemistry on the following day. The observations presented here are predominantly from within the nocturnal residual layer, and thus represent a significant contribution to the growing body of data on mid-continental ClNO2.

The authors present the ClNO2 data, and use co-located measurements of ClNO2 precursors to calculate a ClNO2 production efficiency to compare across the measurement period. This is a useful parameter on which to focus, as the complex nature of ClNO2 production results in significant variability, making it often difficult to constrain in models. The authors then use a chemical box-model to explore the effects of ozone (O3), nitrogen dioxide (NO2) and temperature on ClNO2 production across the experienced parameter space. This analysis is insightful; however, I feel the authors need to do more to demonstrate the sensitivity of their analysis and conclusions to other important parameters that control ClNO2 production. In particular, the sensitivity to the loss rate of the nitrate radical (NO3) to reaction with volatile organic compounds (VOC) warrants a more detailed sensitivity analysis than that presented in Fig. S9. The correlation between measured particulate chloride and calculated ClNO2 production efficiency should also be shown to support the argument made that this is not a limiting factor. Overall, the manuscript is well written and represents a valuable contribution to the field, and warrants publication in ACP once the following comments have been addressed.

1. As particle surface area and chloride content are key factors in the production of ClNO2, it would be useful for the reader if these data were presented somewhere in the paper or supplement and discussed in more detail. In section 3.4 the authors argue that ClNO2 production efficiency is not limited by particle chloride content, but I feel this statement would be better supported if the particle data were shown. Multiple factors can influence both the uptake of N2O5 to particles and the subsequent yield of ClNO2, such as chloride molarity and liquid water content. The authors acknowledge that they do not have sufficient data to fully characterise the particle phase, however, more could be done to demonstrate the sensitivity of the system to these parameters (e.g., McDuffie et al. 2018).
2. As with comment 1, the conclusions of the work would be better supported if the sensitivity to the gas phase loss of NO3 to VOC reaction was demonstrated (beyond that shown in Fig. S9). In the modelling work presented in Sect. 3.5 the authors assume a constant NO3 reactive loss rate (kNO3) of 0.001 s-1. As this work is carried out across both summer and winter seasons, and due to the strong biogenic control of the k NO3, it seems unlikely that this constraint is valid. Observations of kNO3 at another site in Germany have shown kNO3 values approaching 0.3 s-1 (Liebmann et al. 2018). Although the authors do carry out a set of simulations with a value of kNO3 = 0.0005 s-1, a more thorough assessment of the model sensitivity to this parameter would better support the authors assumption that it plays a minor controlling role.
3. It would be useful if the observations overlaid on the model isopleths in Fig. 6 (a) and (b) showed the observed ClNO2 mixing ratios to compare with the model values. Although the purpose of the modelling is not to recreate the observations, rather to investigate the chemical sensitivities of the system, it would provide confidence in the model’s ability to accurately represent the chemistry if the general observational trends were recreated.

References

McDuffie, E. E., Fibiger, D. L., Dubé, W. P.,Lopez Hilfiker, F., Lee, B. H., Jaeglé, L.,et al. (2018). ClNO2 yields from aircraft measurements during the 2015 WINTER campaign and critical evaluation of the current parameterization. Journal of Geophysical Research: Atmospheres,123,12,994–13,015. https://doi.org/10.1029/2018JD029358

Liebmann, J. M., Muller, J. B. A., Kubistin, D., Claude, A., Holla, R., Plass-Dülmer, C., Lelieveld, J., and Crowley, J. N.: Direct measurements of NO₃ reactivity in and above the boundary layer of a mountaintop site: identification of reactive trace gases and comparison with OH reactivity, Atmos. Chem. Phys., 18, 12045–12059, https://doi.org/10.5194/acp-18-12045-2018, 2018.

Tan et al. report the measurements of ClNO₂, NO₂, O₃ and related parameters for three seasons in 2019, obtained during the Jülich Atmospheric Chemistry Project (JULIAC) campaign in Germany. An important result of this study is the variations of ClNO₂ production efficiency in different seasons, which are most sensitive to the availability of NO₂ and increase with the decreasing temperature. This finding is valuable as it enhances our understanding on the dependence of ClNO₂ formation on the availability of NO₂ and O₃ in Europe. Overall, the manuscript is well presented, however, I feel that the importance of the study and discussion of results can be further strengthen and improved. My comments are as below.

1. Line 19: Delete the word ‘ion’ (same for line 52)
2. Line 22: Please specify the date instead of using ‘one night in September’
3. Line 58: The yield for ClNO₂ (𝜑) can be equal to 0 or 1, therefore, it should be ≤
4. Line 62–63: ‘The forward and back reactions constitute a fast thermal equilibrium between NO₃ and N₂O₅ that is established within a minute at room temperature.’ Revise this sentence by justifying how the equilibrium can be establish within a minute. Is this base on the authors’ calculation or from the literature? The concentration of NO₂ can also influence the equilibrium of NO₃ and N₂O₅
5. Line 75: ClNO₂ usually present at night but not always is the case. Suggest to delete the word ‘only’
6. Line 123: The authors should highlight in the introduction or conclusion why investigation on the seasonal variation of ClNO₂ concentrations and its formation are scientifically important
7. Line 168: The concentration of Cl₂ in the cylinder used for calibration is 5 ppmv (±5%). As we know Cl₂ is a very reactive gas that can loss on surfaces. Is this ± 5% a reliable value? The authors should provide details on whether they have quantify the output concentration of Cl₂ from the cylinder and/or consider the potential loss of Cl₂ in the calibration system, e.g. the loss on the regulator of the cylinder or tubing? This is crucial for the determination of ClNO₂ calibration factor and estimation of measurement uncertainty, which can affect the reported levels of ClNO₂ and maybe the conclusions of this study
8. Line 279: Specify the humidity (RH) of the humidified chamber air
9. Line 300: ‘no corrections are needed for the interpretation of ClNO₂ measurements’. I am wondering if this variation has been considered in the estimation of the measurement uncertainty.
10. Line 367: As shown in this figure, the ClNO₂ and related parameters are separated into long-range transport and region transport. The classification of long-range and regional has been described in the text. A lacking information here is the ‘age’ of different air masses. My question is will the ‘age’ of air masses play important role in the observed levels of ClNO₂? I think this should also be addressed in the discussion since the ‘age’ of air mass may affect the NO₂ and O₃ concentrations
11. Line 387: Section 3.3 describes the nocturnal vertical stratification and summarize that the JULIAC inlet (50 m) is most often located within the nocturnal boundary layer and on top of the surface layer. What does it means by most often? At this point, I am not so convince yet that the ClNO₂ are often measured above the nocturnal boundary layer with the discussion and provide only one day example (Fig.4). Please provide more evidence (of different seasons) and discussion in the main text or supporting info to support this argument. This is an important information for the validity of the calculation made from Eq7 (Line 465)
12. Line 541–542: The measured aerosol surface area is an essential parameter for the calculation. This should be included in the supporting info. Can the authors justify why setting the aerosol surface area to constant value in the model since they have measurement data?
13. Line 577–578: Temperature also plays an important role for the value of the ClNO₂ production efficiency due to the shift of the equilibrium between NO₃ to N₂O₅. The temperature shift may also affect the humidity which has been shown in previous studies to promote N₂O₅ uptake and production of ClNO₂. How can the authors separate the effect of humidity with the effect of temperature?
14. Line 691: Please provide a proper reference here rather than citing the general website of IUPAC
15. Supporting Information Figure S2: Explain why the response of ClNO₂ decrease with H₂O concentration (a)? Show the correlation coefficient for this linear fitting (b) as the points are spreading wide from the fitted-line.