General comments:

In this work, “Volatility of aerosol particles from NO3 oxidation of various biogenic organic precursors,” the authors examine analytically combining the use of different experimental setups, as a thermodenuder, an isothermal chamber and a FIGARO-CIMS system, formula-based parameterizations and an evaporation model in order to try to estimate volatility from the NO­3 oxidation of a-pinene, isoprene and b-caryophyllene, a topic that is very understudied and scarce information exists about these volatility estimations. Volatility estimation is a topic that requires a lot of effort, the combination of different experimental and modeling techniques and sometimes the results about it can be failing or varied. This happens because of the complexity of the mechanisms of aging and reactions that happen in the atmosphere that makes the fate of the precursors highly uncertain. Also using modeling techniques there are many assumptions that have to be made in order to infer volatility that make the problem even more complex. The authors manage to use rigorously a combination of methods and finally provide us with a descriptive analysis of the volatility distributions and composition of SOA formed from nitrate oxidation for the three precursors. In the end the formula-based parameterizations were re-evaluated and showed that improvements in their description have to be incorporated in the future. I find this work very valuable and almost ready to be published while a few comments could be addressed a bit better and some technical corrections could be performed in order to be clearer.

Specific comments:

In Table 1 are presented the experiments performed for this work and while most of the results of them are presented experiment 4 is not used in the analysis. It is only referred to figure S2 when trying to show the sensitivity of different initial mass loadings from the nitrate oxidation of a-pinene. Is there any specific reason to not in the main figures as well? Also in the table there are many missing values, this could be explained in the bottom why is it happening as a footnote.  Also while the nitrate experiments of a-pinene are 5, is there any reason there are 3 for isoprene and 2 for b-caryophyllene. Probably some description for the decision of the design of these experiments could be helpful in the main article. Finally for the isothermal evaporation experiments while most experiments finish at 240 min some stop after 150 min, this is probably because the experiment had to stop or there was equilibrium and no extra evaporation was noticed after that?

The evaporation model that is used for the analysis from Riipinen et al. (2010) depends on many parameters and some work has been done to investigate the sensititivity of them as it is done in Figure S3 where there is used a different value for the vaporization enthalpy of 70kJ/mol, different accommodation coefficients of 0.01 and 0.1 compared to unity of the base case as well as a mass-dependent diffusion coefficient. This analysis has been focused only on the ozonolysis of a-pinene though that is not the centre of interest in this work and in any case it showed to have better comparisons with the measurements. The b-caryophyllene nitrate oxidation products seem to have a more bimodal volatility distribution that make it more difficult to capture its “fingerprint” with the VFR in the thermodenuder and isothermal evaporation chamber, it could be probably fruitful to try to see for the three different parameterizations used how changing dHvap (with values of 70 kj/mol but also higher even 150 kj/mol), am (0.01 and 0.1) and mass-dependent diffusion coefficients would change these thermograms. This could be repeated for the isoprene case for a more broad and complete picture for the sensitivity to these parameters.

In Figure 3 the authors show the volatility distributions in the form of VBS as derived from the three different parameterizations that are examined, while someone can see the composition in ELVOCs, LVOCs and SVOCs it would be useful here to also add in the figure somewhere what is the average C* from each parameterization for each case of precursor. This would make it clearer understanding which parameterization shows higher or lower volatilities for each case.

Technical corrections:

Table 1: In footnote change S5 to S4 (there is no figure S5)

Line 227: Change 240 s to 240 min.

Line 228: model inputs instead of input

Line 229: Change concentrations estimated to concentration estimates to be clearer

Line 267: change to “agree with the VTDMA data”

Line 327: A bit complicated sentence, rewrite clearer

In Figure 4 I think the shading should change because right now it is not very easy and clear to see the differences, especially in the green shades.

Line 378 Change Though to Although

Line 387 something is missing, rewrite “and experimental approach developed by the authors”

Line 399: Change to “ an hydroxyl”

Line 408: Change to “ of the observations”

Line 450: Taken together, change the expression, suggest that the SOA … contain, rewrite a bit the sentence.

In Supplementary:

Figure S1: Change Narve, 2019 to NArVE campaign 2019 as it is mentioned in the article or This work

Figure S3: Change blue to cyan (you use cyan color)

Figure S3: Describe a,b,c,d,e,f in legend.