Interactive comment on “Bimodal distribution of size-resolved particle effective density in a rural environment in the North China Plain” by Zhou et al.

General Comments:

In this study, a combined DMA-CPMA-CPC system was applied to characterize the size-resolved particle effective density in Multiphase chemistry experiment in Fogs and Aerosols in the North China Plain (McFAN) in autumn 2019. They identified a frequent bimodal distribution of particle effective density, and a unique low-density mode (named sub-density mode) accounted for ~20-30% of total observed particles. The diurnal variations of particle effective density and the influence of pollution and secondary aerosols were discussed. They concluded that the influence of BC on the effective density is even stronger than SIA.

Overall, the paper is well-written and is appropriate for ACP. The results clearly indicate the factors that govern the variations of particle effective density. The size-resolved particle effective density shown in the manuscript is interesting and would have implications for further studies. Some minor comments are still needed to be addressed before the manuscript can be published.

Specific Comments

1) The authors directly linked the sub-density mode to fresh black carbon (BC) emissions. Some organics might also have very low densities, which might lead to ambiguous conclusions. Previous measurements have also indicated that organics dominated in smaller size ranges. This is a key requirement when clarify the significance of BC in such mode.

2) It would be better to include uncertainty data when expressing the mean density in the abstract.

3) abstract: “…for the sub-density mode (𝜌̅_eff,sub) ascribed to the agglomerate effect.” Does it refer to the agglomerate effect of BC?

4) Line 113: “A combined DMA-CPMA-CPC system was employed to measure the size-resolved effective density of particles with mobility diameter of 50, 100, 150, 220, and 300 nm” what is the uncertainty for the size selection?

5) Line 142-: It is necessary to show the uncertainty during the peak fitting with a flexible Gaussian fit algorithm, and thus potential contribution to the overall uncertainties.

6) Line 205-: “The remarkably high occurrence of the sub-density mode in our study indicates a frequent influence of local BC emission.” Can these sub-density mode be matched to the variations of BC concentration?

7) Line 309: “It indicates that photochemical aging process is very efficient in transiting particles from fractal to compact morphology.” In my opinion, the conclusion can only be obtained when the pollution during daytime and nighttime is at the same level. As discussed in the previous section, the pollution level over the study is highly varied, and thus the authors should compare the increase rate of Df under the similar conditions.

8) Section 4.4: Is it possible to assume a diurnal variation of BC density for the test in Figure 6, according to the source’s strength of BC?

9) The conclusions should be shortened to be more concise, in particular, there are several numbers that are not really important.

Dear Zhou et al.,

thank you for the interesting study regarding effective densities of ambient aerosol particles. The manuscript “Bimodal distribution of size-resolved particle effective density in a rural environment in the North China Plain” has been written very well and it is based on experiments conducted with state-of-the-art methods. The study presented in the manuscript aims to describe the effective density of ambient particles but also link it to the sources of particles, especially in case of observation of low effective densities. The figures of the manuscript are clear and mostly very informative and tables serve very well the structure of the manuscript.

One relatively important issue in the presented study is the duration of the measurement campaign. I think the experimental period is not long enough to generalize the results. Regarding to that, is it possible to modify the title and abstract so that this is brought out to readers already in the beginning of the paper? Mentioning that the study is “case study” or “short campaign” would be enough for that purpose.

As we all know, weather conditions have crucial role in aerosol formation, emission transportation and emission ageing, and they affect the ambient concentrations significantly. I propose that the authors include much more detailed weather data to the paper and investigate how the weather affect the effective densities of the particles. I think the affiliations of the authors enable the access to local weather data if it was not measured directly at particle measurement site. In addition, inclusion of the weather data into the paper enables better comparisons to other studies made later in same place or in other places by other researchers.

The authors mentioned some of the possible aerosol sources that can affect the aerosol measured in their site. I would like to see those on map (e.g. roads, factories, power plants). In addition, to study their role in the measured aerosol, I propose that the authors analyze wind directions (if available) and add discussion about it to the manuscript. E.g. quite recent studies for coal combustion emissions have reported effective densities >2 g/cm3 for particles, and it could be interesting to know if that kind of emission sources are near the measurement site possibly contributing to aerosol measured.

In the experiment descriptions the authors write that they measured particle number size distributions also. It is not presented and analyzed in the manuscript. Why so? Could it be included into the analyses of size-resolved densities? In some previous studies made using SMPS-ELPI method the different densities have been connected to modes in particle number size distribution. It could be interesting and also important to see if this kind of results can be drawn from these experiments also.

The effective density of the particles can be affected by sampling method and treatment of the aerosol before the actual measurement. I would like short discussion in the manuscript regarding how the sampling and treatment used in this study possibly influence on the particle measured.

In the manuscript, the data have been divided to “polluted” and “clean” based on the PM0.7 results. How this PM0.7 was measured? Why these two were defined again in Figure 7 but now based on PM1 and with different threshold value? If PM0.7 is based on SMPS measurements, what are the limitation regarding that (e.g. knowledge of fractal dimension)? And in my opinion, the data labeled as “clean” is not very clean air, and I propose using “less polluted” and “more polluted” instead of current terms. Furthermore, it would be interesting to know how these less polluted and more polluted periods exist in timeline of the campaign. Are they from diurnal variation of concentrations or from changes in general pollution level?

Technical questions and comments:

1. There are no clear descriptions of the meaning of colour code used in figure S3. In addition, the units for the colour axes are needed (#/cm3?).
2. Why the colours changed from fig 1 (main density mode indicated by blue) to fig 2 (main density mode indicated by black) and to fig 3 (main density mode indicated by red)? In general, please check the uniformity of the article (text, definitions, figures)
3. line 51: exits -> exist

Finally, thank you for the study which was made experimentally very well and which offered new insights to the characteristics of ambient aerosol. I hope that this kind of studies are made in future also in urban environments as well as directly for emission sources so that the the whole picture of aerosols affecting our health and climate can be understood better.