The article "A Diagnostic Intercomparison of Modeled Ozone Dry Deposition Over North America and Europe using AQMEII4 Regional-Scale Simulations" by Hogrefe et al. is another of the Air Quality Model Evaluation International Initiative model comparisons, this one focused on dry deposition of ozone. The paper analyzes the ozone dry deposition fluxes and velocities from simulations over North America and Europe for 12 different model configurations.

The article is well written and presented and should be published with only very minor revisions.  In fact, this reviewer had a difficult time finding much of anything to comment on. Corrections or suggestions for minor improvements are listed below:

- Throughout the article there are several places where "O3" occurs rather than the "O₃".

- p. 3, line 75: "A companion paper in this issue .." ... I believe this refers to the previously cited Schwede et al., 2018, but it might also refer to a companion paper in the current issue.  This ambiguity should be resolved.

- p. 17, line312: should be "... averaged over only those grid cells where a given model ..."

- pp. 22-24: The discussion here concerning Figure 10 is somewhat confusing because of the shorthand names given to the regional-scale simulation results vs the single point model results.  For example, on p. 22, lines 369-370, single point models are referred to as "GEM-MACH Zhang" and "GEM-MACH Wesely", while in Figure 10 these models are labeled "SP GM Zhang BF/SP GM Zhang HF" and "SP GM Wesely BF/SP GM Wesely HF". Some additional thought should be given to making the names more consistent between the text and Figure 10.

- p. 36, line 587: should be "between 0.4 and 1.0 cm s^-1 ..." (i.e., superscript "-1").

- Section " 4 Summary":  The authors have done an excellent job of disentangling the effects of different LU types and LU datasets on the deposition results in this work. However, given the importance of the underlying LU data, I would think a general call for improved, commonly available high-resolution LU datasets would be appropriate.  With the plethora of satellite datasets available today, a concerted effort to create better, publicly available LU datasets would reap significant benefits for air quality, weather and land surface models.

- Figures S12 an S13 in the supplement are of very poor quality and should be improved.

This paper provides a well-written and useful comparison of dry deposition from several models. The authors deal with a wealth of information in a logical and generally well-thought out way. The paper is worthy of publication after dealing with the comments made below.

Major comments:

The term Vd is used, and many plots are provided and compared, but no mention is made of which height the values refer to. A Vd calculated at 1m can be very different to one calculated at e.g. the center of model grid boxes. Are these 1m values, or Vd from grid-center, or something else?  Are the results across models really comparable?

The term effective conductance is also used throughout, but this is not a widely-used term and is not defined here. The manuscript refers to Paulot et al 2018 and Clifton et al 2020a for definition, but Paulot does not use this term at all, and Clifton doesn't really explain the difference between conductance and effective conductance. My comment about Vd above would also apply to this conductance terminology, for the aerodynamic part at least.

p7, paragraph starting on L173: A few factors are mentioned here for differences in O3 deposition (e.g vehicle speeds), but some likely more important factors are not mentioned, not least that both the anthropogenic and biogenic emissions have large uncertainties in magnitude as well as location. Indeed, there is little discussion of the fact that different models produce different amounts of O3. I think the authors should provide maps of the near-surface concentrations from the different models, and also provide references to some of the  other multi-model inter-comparisons that illustrate the magnitude of model to model differences.

p7, L177: Figures 3-4 are introduced here, but the important results shown in this Figures are  not discussed at all; instead the text proceeds to comparisons between Vd and fluxes. Some relevant discussion is found later in the paper, but the reader deserves some comment here and some guidance as to where the large differences in Vd between models will be discussed.

p17, 3.2 and elsewhere. The large importance attached to landuse is a good feature of this and other AQMEII studies. However, differences in height, roughness length displacement height and LAI are present even for the same land-cover. Although Fig. 10 and S6-S8 are interesting in connection with this, a table to compare values for some of the more common LU would be useful, along with some discussion of the implications.

p37, concerning "Data availability", I thought is was ACP policy to avoid such "available upon request" approaches, and instead to ensure data are available though zenodo or equivalent. This is always the better approach.

Minor comments

The abstract should mention the number of CTMs used in this study.

L27 and elsewhere. O3 should have 3 as a subscript

Table 1: Be more explicit with regard to the LU scheme, preferably with references. For example, which MODIS dataset? What is

DEPAC. What is AQMEII4 LU? (This one is discussed above, but a reference would be appropriate in such a summary table.)

Connected to this, L457 states that the LOTOS/EUROS model uses the official LU data of the European Union, but that is not called DEPAC as given in Table 1.

L94 and elsewhere. Define "Europe", and use a different acronym (e.g. EUR) unless you really mean the European Union (pre or post Brexit).

Actually, define North America too. All of USA + Canada, or that domain shown in Fig. 1.

p13, Fig. 5. The colors in the legend should match those of the figure.

p19, and elsewhere. Again, the "effective conductance" term, but what is this?

In SI, is "Soil" really soil, or undergrowth?