This manuscript derives composition-specific aerosol scavenging efficiencies (SE) through the analysis of ten storms from the Deep Convective Clouds and Chemistry (DC3) and the Studies of Emissions, Atmospheric Composition, Clouds and Climate Coupling by Regional Surveys (SEAC⁴RS) field experiments combined with process-scale modeling. The manuscript addresses an important problem—quantifying composition-dependent aerosol scavenging in deep convection—and provides a valuable observational dataset. The results presented can be used directly for evaluating cloud-scale chemistry transport model representation of aerosol scavenging.

The article is written clearly and would be a good contribution to ACP. However, the authors need to address the following comments before I can recommend publication in ACP.

 Major comments:

The largest concern is the limited evaluation of uncertainty in entrainment rates and its propagation into scavenging efficiency (SE). Given that SE is directly dependent on the entrainment, a quantitative uncertainty analysis (e.g., sensitivity tests or error propagation) would substantially strengthen the robustness of the conclusions.

• Entrainment rates vary substantially among cases (Table S2), and different tracers were used to diagnose entrainment in different storms. While this is understandable given observational constraints, a more explicit discussion of the comparability of entrainment rates derived from different tracers would strengthen confidence in the results.
• It would be helpful to show the vertical profiles of the tracers used to derive entrainment alongside the corresponding aerosol vertical profiles, to assess how similar their vertical structures are and how sensitive the entrainment correction may be to profile heterogeneity.
• More importantly, the manuscript does not quantify how uncertainty in entrainment rates and in the clear-air and outflow aerosol profiles propagates into SE uncertainty. In Table S2, only cases with multiple outflow legs include standard deviations, while other cases appear to lack uncertainty bounds.

The discussion focuses primarily on chemical processes and entrainment of mid-free-tropospheric aerosol layers in interpreting SE variability. However, the dynamical controls on scavenging are less clearly developed. Although SE is shown as a function of the SWEAT index, the analysis does not fully explore how dynamical factors (e.g., convective intensity and updraft strength) may influence the observed chemical and scavenging signatures. The connection between storm dynamics and composition-dependent SE could be developed further, particularly in the Conclusions section.

Minor comments:

Tick directions are inconsistent across figures. For example, Fig. 4 and Fig. 6 have inward ticks, while other figures use outward ticks. Please standardize formatting.

Figure 2: It would be helpful to include the sample size (e.g., number of seconds of sampling) within each 1-km bin for the clear-air vertical profiles. Additionally, the method used to define the inflow aerosol concentration below cloud base should be clarified. Was the bin closest to cloud base used, or was an average taken across all sub-cloud layers?

Line 381: It would strengthen the discussion to provide references and typical magnitudes for below-cloud scavenging efficiency for comparison.

The Conclusions section would benefit from a brief, explicit statement of the main limitations of the study (e.g., entrainment uncertainty, tracer dependence, simplified chemical interpretation).

This paper describes aerosol scavenging in deep convective storms based on DC3 and SEAC⁴RC field measurements and the entrainment model simulations. The deep convective cloud scavenging efficiencies of sulfate, nitrate, ammonium, and organic aerosols are calculated, and physical and chemical processes affecting the scavenging efficiency analyzed. The results provide valuable information and dataset for further investigating the spatiotemporal distributions of aerosols in the upper troposphere. The paper is well written in general and can be accepted for publication in ACP subject to minor revisions.

Comments in detail

In this study, the scavenging efficiencies (SEs) of sulfate, ammonium, and nitrate aerosols are calculated based on aircraft observations combined with parcel model simulations. The results show that organic nitrate SEs (~40%) are lower than sulfate and ammonium SEs (75%), and the causes of difference are attributed to the entrainment of mid-tropospheric nitrate layers formed from lightning-produced nitrogen oxides. These results appear to be sensible/reasonable from the field observation of view. On the other hand, scavenging here should be referred to wet deposition of aerosols via nucleation, impaction and Brownian diffusion, as mentioned in this paper. This wet scavenging process is clearly defined and treated separately from the lightning production of nitrogen oxides in regional or global chemistry model. The questions are:

• Could the term ‘scavenging efficiency’ and calculation method still be used appropriately in the case that there are much more aerosols entrained from mid-troposphere than from the inflow region?
• Should nitrate have a same SE level as sulfate and ammonium if there were no nitrate mid-troposphere layer or lightning nitrogen oxide source in the model simulation?

It might not realistic to address these issues by performing additional modeling work. But adding some discussions in the conclusions might be helpful for guiding future modeling research.

The article “Aerosol Scavenging in DC3 and SEAC4 RS Deep Convective Storms” by Barth et al., interpret scavenging efficiencies of speciated mass concentrations of aerosols for investigations performed during DC3 and SEAC4RS field experiments on different storms. The studies are in line with the investigations performed by the research group. The results highlight the importance of multi and intercorrelation of dynamics, physics, and chemistry used to interpret processes occurring during thunderstorms. This study add important information on more robust quantification of aerosol scavenging by studying more convective storm than existing research which was limited to a maximum of 3 storms.

Major comments:

Please add information about the dependency on pressure of instrument calibrations. Aircraft measurements handle with vertical distributions and an accurate measurement needs knowledge on the influence of pressure and temperature on measurements. If the instruments were ground base calibrated and measurements were performed at high altitude it could affect the measurements, please comment.

Even if the study consider zero initial concentration of organic acids it is evident that these concentrations can not be zero. Please comment based on the existing field measurements studies which stated various concentrations of organic acids in the atmosphere, how the model could be affected.

Figure 5S represents the scatter plot of pNO3 and the NO3 - fraction of NO3 - + HNO3 for 6 different cases. Excluding the extreme purple dot the outflow linearity change drastically and data became more representative in terms of linearity. Please discuss how this potential exclusion would affect the results.

May you better explain why the aerosols scavenging efficiency differ for all of them (sulfate, ammonium, etc.) during the 18 May 2012 and 2 June 2012 events? It could be an effect of experiment failure?

Minor comments:

Line 129: better add „using i- to n- ratios of butane and pentane isomers”

Line 148: Faloona et al., 2004 is missing from reference list. Please check all the references in this sub-section. Most probably references from Supplementary information should be added in the list of main article

Line 379-381: please revise the sentence

Line 389: Isn’t there „West”?

Line 542: „hydroxyacetone”

Line 562-563: please add „;” for reference literature separation

Figure 2 (06 June 2012) pNO3 x 5 is missing

Please add a short note to table 2S to explain the reason of outflow time representation in the form of e.g. 25:50:00.

Please be consistent with the cases representation form „12 June” or „12 June 2012”