The author conducted a one-year VOCs observation in Beijing, attempting to illustrate NO3 chemistry, especially the oxidation of VOCs by NO3. Then, a parameterization method was established to attempt to construct the long-term reactivity of NO3 using VOCs data. It was found that since 2011, the NO3 reactivity of VOCs in Beijing has been decreasing year by year, and there is a significant correlation with VOCs emission reduction. The overall research objectives of the article are clear, the research methods are appropriate, and the research conclusions are relatively reliable. However, there are still issues that need to be clarified, as follows.

1. Correlation diagram, meaning of horizontal and vertical coordinates. For example, Figure S1. Why use simulated values as the abscissa? Generally, the abscissa is the reference value, or the true value. In this case, it is obvious that observation values should be used as the abscissa. Please review the entire article by the author.
2. Why is temperature and humidity linear between April and May? Is there no observed data? If not, please delete it. The uptake coefficient of N2O5 also has this issue.
3. The red lines of d and e in Figure S2 are both NO3 losses, with the former being the total loss and the latter being the loss in reaction with VOCs. It seems that there is a significant difference between the two orders of magnitude, is it due to the participation of heterogeneous processes?
4. Figure S7 shows that current parameterization methods can capture changes in monthly and daily averages, but cannot capture changes in hourly averages. Why? After adding terpenes in Figure S10, the fitting results actually look worse. Why?
5. The article discusses the NO3 reactivity of VOCs, but there is no VOCs concentration sequence diagram in the article, especially the proportion of different components, seasonal changes, daily changes, etc. In addition, the author needs to explain the detection method of VOCs and what are the pre freezing stages? Detect which species and so on, for example, indicate 56 PAMS species.
6. The article emphasizes that the simulation results of terpenes come from WRF-CMAQ. What specific simulation scheme is used? Given the important contribution of terpenes, it is recommended that the author provide a detailed explanation.
7. The author used the ratio of isoprene to terpene to calculate the concentration of terpene, which may require supporting evidence. Firstly, the emission of terpenes from biological sources is related to temperature, while the emission of isoprene from biological sources is related to temperature and radiation. There may be a relationship between the two at night, but the relationship may not be significant during the day. Secondly, the author also emphasizes that motor vehicles emit a large amount of isoprene, and if the ratio is used to determine terpenes, it may lead to an overestimation of terpenes.
8. Since WRF-CMAQ can simulate OH concentration, why use J1D to calculate OH concentration?
9. Isn't Equation 11 the method for calculating NO3 reactivity? What is the difference between this equation and equation 1? It seems that the difference is only slightly fewer species.
10. Since terpenes have made significant contributions, I think it is meaningless without adding the explanation of terpenes in sections 3.1 and 3.2. I suggest deleting section 3.3 and merging it into 3.1 and 3.2 for discussion and explanation. The author can use the simulation results of long-term terpenes to illustrate the interannual trend of NO3 reactivity of VOCs in Beijing.

The chemistry of NO3 with VOC affects the budget of nocturnal SOA, and regulates regional photo chemistry indirectly. This study presents a detailed analysis about the nitrate radical reactivity towards VOC based on the one-year VOC measurement in a typical urban site. The level, compositions and seasonal variation of NO3 reactivity are well characterized. The results showed isoprene and styrene dominated NO3 reactivity on average, and proposed a month-resolved parameterization scheme to predict NO3 reactivity by using one or several VOCs species data. They tried to rebuilt the dataset of NO3 reactivity by using the scheme and collecting the historical VOC measurement data, and showed an overall decrease trend in recent years. Although this result may be highly uncertain, it provided a new and interesting perspective of the nighttime chemistry in a long-term period. This topic certainly within the scope of ACP and the manuscript is overall well-written. I would like to recommend it be accepted after the authors address the following comments listed below:

1. In section 3.2, the authors constructed a parameterization to estimate NO3-VOCs reactivity using one or several VOCs concentration based on one-year data in 2019. For robustness, it is better to evaluate this parameterization on datasets of other years, since you have collected the historical data of VOC concentrations in Beijing for several years before 2019.
2. 5: The title ‘Regulation of nighttime VOC oxidation’ is somewhat vague for readers to follow. Maybe it is better to use ‘The relationship between NO/Ox and nighttime VOC oxidation by NO3 radical’. I also found that the dependence of RNO3 on NOx has shown similar messages with the dependence on NO, which could be confusing in this figure as it is not related to the regions defined. Since the RNO3 has a relatively good exponential correlation with NO, I am wondering if Fig 6 can first focus on NO dependence and divide it into three regions, which might be regarded as NO-limited, transition and NO-saturated. In NO-saturated region, RNO3 is closed to zero and shows no dependence on both NO2 and O3. In the other two regions, RNO3 can reach up to 80%, and then it could be of interest to look into how the attribution of Ox influence the RNO3 variation.

Technical corrections:

1. Line 25: Please change ‘the’ to ‘that’
2. Line 49: This equilibrium reaction can also take place during the day.
3. Line 62: Please delete ‘type’
4. Line 86: Please change ‘;’ to ‘.’.
5. Line 102: Please change ‘varies’ a verb to an adjective.
6. Line 117: Suggest adding ‘newly’ before ‘proposed’.
7. Line 118: The word ‘regulation’ is a bit of vague here.
8. Line 138~139: Does this factor vary temporally? If so, please provide the specific values
9. Line 140: Please delete the first ‘is’.
10. Line 190: Please add the ‘reason that’ before ‘the calculated’.
11. Line 205: Please change ‘of’ to ‘among’.
12. Line 207: What is the ‘another individual VOC’? I guess it means other?
13. Line 213~214: I think the conclusion sentence here is not so necessary here. Or the authors could just put it at the beginning of this paragraph.
14. Fig 1: In my opinion, the information in this figure is a little bit overlapped by presenting NO3-VOC reactivity levels in both months and seasons of a year. I think the months of year style is enough to demonstrate the temporal variations of reactivity. And the seasonal variations can be just provided in the text. In addition, it is not easy to identify daytime and nighttime reactivity with thin frames in black and blue. Shadow padding could be better.
15. Line 251: Fitting equations are not displayed in Fig 2.
16. Eq 10: If the author chose only one species, which has the strongest correlation coefficient with total NO3 reactivity, to develop parameterization, then what does the subscript i mean? Is it month? Please specify it because it could be a little bit misleading after comparing to Eq 11.
17. Fig 2: It is called heat map not thermodynamic diagram. And the description in the figure legend is not so clear.
18. Line 308: Please also mention the method used here to account for MNT reactivity which has been described in methodology section.
19. Line 317: Is this sentence trying to say ‘to compare the monoterpene and the total’? It could be confusing here because it is followed by the fraction of MNT-NO3 reactivity.
20. Line 352: Please change ‘highlight’ to ‘highlighting’.
21. Line 356~371: Suggest moving Fig S11 to the main text. I think the seasonal variation of VOC reactivity attribution is more suitable to be presented here.
22. Line 374: Please specify the R_NO3 and add the statement ‘see method 2.3 for its calculation’ here. Furthermore, it is not clear how the authors derive the Fig 6. It means the values of each bins?