Title: Numerical experiments of in situ particle sampling relationships to surface and turbulent fluxes through Lagrangian coupled large eddy simulations

 Authors: Park et al.

Summary:

Aerosol particle fluxes measured from past field campaigns or recorded in the literature show significant variations and inconsistent results. Different methods of flux calculations add additional uncertainty to the retrieved aerosol particle fluxes. Given those highly-varying results, one following question is: How representative are our point /moving-in situ measurements for the true aerosol particle fluxes over the sampled areas? This manuscript tries to tackle this fundamental problem by quantifying the uncertainty introduced by different sampling strategies, instrumentation, retrieval methods using the LES outputs as truth. Specifically, the authors investigate the following questions: Does the approach we adopted to sample the aerosol particle fluxes (i.e.., sampling areas, sampling instruments, retrieval methods, and stability of the PBL) have a significant impact on the final retrieved surface aerosol particle fluxes? If so, how large uncertainty is introduced by each step to the final retrieved surface fluxes? The author found that all these factors influence the final retrieved aerosol particle fluxes, with their associated uncertainties varying substantially.

There are so many exciting findings in their results. For example, they found that different directional alignments for flux sampling (stream-wise, span-wise, and diagonal direction, Figure 6) introduce different uncertainties to the estimated aerosol particle fluxes (Figure 8). Such information is extremely useful when planning a future field campaign.

Besides the authors’ comprehensive answers to the fundamental questions mentioned above, the more significant merit of this manuscript is that they provide a general framework to test new measuring techniques, sampling strategies and their related uncertainty quantification. Such a framework is of great value for future field campaigns, both in achieving the scientific goal of those campaigns and reducing human resources and cost.

It is a pleasure to read such a comprehensive study, I recommend a minor revision for this paper.

The only moderate problem that I found is that the URL to their code repo seems not correct. The authors might need to replace it with the correct one.

Below is a list of minor to moderate comments identified in the current manuscript.

Recommendation:

Minor revision

 General comments:

1. Codes on Github is not accessible: The URL for the codes used in this study jumps to the webpage: https://github.com/RichterL, and I cannot find the NTLP model there. If you decide to use Github as the code repo, please also attach a license (e.g., GPLv2/v3) to it.

Specific comments:

1. Interpretation of Figure 9:

In Figure 8, we generally see the uncertainty grows with height. For example, Figure 8(c)-(d) shows shapes of upside-down triangles. Figure 9 shows the inferred surface fluxes from Figure 8 by using Equations (6)-(7). In Figure 9(c)-(d), the uncertainty shapes are no longer upside-down triangles. In Figure 9(c), the level with maximum uncertainty is now at \frac{z}{z_{inv}} = 0.7. Can you explain why this is the case? Same question for Figure 9(d).

2. The schematic figure for your deployment of stationary and moving probes (L390-400): It would be great if you can provide a schematic figure showing the locations of the probes. For the moving probe, you can draw it as a moving line with its arrow head indicating its moving direction. Please also mark the direction of flow.

3. Calculation related to ogive probes.
   1. L 394: “at a user-defined speed of 50m/s”. Is there any justification for using this speed 50m/s? What is the normal speed of planes in the field campaign?
   2. In Figure 10, you show the ogive curves for the moving probe in stream- and span-wise directions. I assume those calculations are based on a speed of 50m/s. Is it possible to calculate another two lines using a speed either higher or lower than 50m/s?. This will enable readers to know how moving speed affects the convergence rate.

4. Figure 13: the highly varying predictive ranges of flux at high levels (\frac{z}{z_{inv}}>0.9): Compared to the low level where the oscillation of flux range is of small amplitude, the flux range changes dramatically at high levels. Is it due to your LES setting, e.g., too few vertical levels in high altitudes, or low model top? Any explanation for this feature?

Technical corrections:

L383: “A common technique in field measurements, an ogive curve O_g_{wc}(f_0) represents a running integral…”: broken sentence?

This study shows deep research on the surface flux estimates from different

numerical approaches and compares the observation, eddy covariance, simulated

instrumentation and the theoretical flux-profile method results have been provided,

turbulent transport impact on aerosol particles found has estimated. This article could

be accepted after minor revisions.

L14: MABL(marine atmospheric boundary layer) first-time appearance should add

the full name.

L31: Coarse mode aerosols have many influences focusing on lighting and also

growth larger contribute to air pollution, related reference also includes the following

articles:

Pan, Z., Mao, F., Rosenfeld, D. et al. Coarse sea spray inhibits lightning. Nat

Commun 13, 4289 (2022). https://doi.org/10.1038/s41467-022-31714-5.

Lee, S.‐H., Gordon, H., Yu, H.,Lehtipalo, K., Haley, R., Li, Y., &Zhang, R. (2019).

New particle formation in the atmosphere: From molecular clusters to global

climate.Journal of Geophysical Research:Atmospheres,124,

7098–7146. https://doi.org/10.1029/2018JD029356

Wu, Hao & Li, Zhanqing & Jiang, Mengjiao & Liang, Chun-Sheng & Zhang,

Dongmei & Wu, Tong & Wang, Yuying & Cribb, Maureen. (2021). Contributions of

traffic emissions and new particle formation to the ultrafine particle size distribution

in the megacity of Beijing. Atmospheric Environment. 262. 118652.

10.1016/j.atmosenv.2021.118652.

L53：Atmospheric stability is a key parameter impact on particle transport, using

Monin–Obukhov stability theory (MOST) has many progress the related reference:

Irwin JS and Binkowski FS. Estimation of the Monin-Obukhov scaling length using

on-site instrumentation. Atmos Environ 1981; 156: 1091–4.

Srivastava P and Sharan M. An analytical formulation of the Monin–Obukhov

stability parameter in the atmospheric surface layer under unstable conditions.

Bound-Layer Meteor 2017; 165: 371–84.

L58: “other field-based studies” many launched in the atmospheric boundary layer

found that interaction between aerosol exists in the atmospheric boundary layer, could

relate to：

Li Z, Guo J and Ding A et al. Aerosol and boundary-layer interactions and impact on

air quality. Natl Sci Rev 2017; 4: 810–33.

Lauros J, Sogachev A and Smolander S et al. Particle concentration and flux

dynamics in the atmospheric boundary layer as the indicator of formation mechanism.

Atmos Chem Phys 2011; 11: 5591–601.

L101：direct numerical simulation(DNS) and other models also can simulate flux

measurements has a high correlation to the aerosol turbulence interaction(ATI).

Chen S, Yau MK and Bartello P et al. Bridging the condensation–collision size gap: a

direct numerical simulation of continuous droplet growth in turbulent clouds. Atmos

Chem Phys 2018; 18: 7251–62.

Eaton JK and Fessler JR. Preferential concentration of particles by turbulence. Int J

Multiph Flow 1994; 20: 169–209.

Li D,Wei A and Luo K et al. Direct numerical simulation of a particle-laden flow in a

flat plate boundary layer. Int J Multiph Flow 2016; 79: 124–43.

WeiW, Zhang H andWu B et al. Intermittent turbulence contributes to vertical

dispersion of PM2.5 in the North China Plain: cases from Tianjin. Atmos Chem Phys

2018; 18: 12953–67.

L153: “while K(xp) is the average subgrid momentum diffusivity obtained from the

LES model, interpolated to the particle location”, this parameter should provide more

methods or pathways to explain how to get it.

L214: the concentration vertical distribution of aerosol has rare research, but we can

find some evidence based on some UAV measurements, such as:

Mehta, Manu & Khushboo, Richa & Raj, Rahesh & Singh, Narendra. (2020).

Spaceborne observations of aerosol vertical distribution over Indian mainland

(2009-2018). Atmospheric Environment. 117902. 10.1016/j.atmosenv.2020.117902.

Kemppinen, Osku & Laning, Jesse & Mersmann, Ryan & Videen, Gorden & Berg,

Matthew. (2020). Imaging atmospheric aerosol particles from a UAV with digital

holography. Scientific Reports. 10. 16085. 10.1038/s41598-020-72411-x.

L338: “The disaggregation technique employed here demonstrates the importance of

areal coverage and directional sampling when calculating aerosol mass flux”, the

aerosol mass flux method and parameter setting in L286, and the reference?

L495: “a horizontal average over the entire domain” how to deal with the surface

layer and the ABL?