Dry deposition of atmospheric dust is a key process controlling the atmospheric lifetime of dust, source-region removal, and long-range transport. This is particularly important in the post-dust-storm low-wind phase, where its behavior directly affects the accuracy of global dust cycle simulations and climate radiative forcing estimates. For a long time, dust deposition processes under low wind speeds have remained difficult to quantify accurately, and most models adopt monotonically wind-speed-dependent parameterization schemes. This study utilized a closed-circuit recirculating wind tunnel and employed a two-stage experimental protocol (high-wind loading followed by low-wind deposition). Combined with laser diffraction particle size analysis and a first-order exponential kinetic model, the study systematically quantified the deposition kinetic parameters (a, k, τ, J₀) for different particle sizes under wind speeds of 0–6 m s⁻¹. The results revealed a non-monotonic response in the net deposition of fine-to-medium particles, with a peak at approximately 2 m s⁻¹, leading to the proposal of a potential “low-wind window”. This work provides important process-based experimental constraints for improving the parameterization of dry deposition in atmospheric models.

The manuscript is generally well written and logically structured. The experimental design is systematic, and the kinetic model fits the data well, providing valuable measured data and physical interpretations for low-wind dust deposition research. However, I have the following suggestions for improvement regarding the experimental design, particle physics processes, and the physical interpretation of net deposition.

Major Comments：

In practical scenarios, the dry deposition of dust is almost always accompanied by secondary resuspension. For this reason, previous studies on dust dry deposition mechanisms, especially experimental research, have generally suppressed or eliminated the influence of secondary resuspension to investigate the pure mechanism of dust dry deposition. Since no measures were adopted in this experiment to inhibit secondary resuspension, the final results reflect the combined effect of two physical processes: deposition and re-entrainment. The non-monotonic variation of dust dry deposition with wind velocity observed in this study arises from the competition between these two mechanisms, which is not inconsistent with existing physical models of dust dry deposition. It is suggested that the author revise the relevant statements to avoid misleading readers.

Furthermore, it should be noted that since the experimental results incorporate the effect of secondary resuspension, which is closely correlated with surface properties, the shape and size of the trays used in the experiments will exert a direct impact on the findings. The author is advised to conduct in-depth analysis and discussion on the rationality and representativeness of the experimental results.

Specific Comments:

1) It is suggested to define u* and delta in the caption of Figure 1 to help readers quickly understand the information presented.

2) It is recommended to specify the measuring position for the wind velocity profile.

The trays, with a height of 3 cm, are obviously higher than the surface roughness length z0 provided in the paper. Thus, the trays act as surface roughness elements under experimental conditions and exert a notable influence on the near-ground wind field. Given that the development of a turbulent boundary layer requires a certain fetch length, the measuring location of the wind velocity profile should be clearly stated. In addition, it needs to be clarified whether the trays were reset after each test run, so as to ensure consistent layout of the trays across all experimental tests.

3) With a diameter of 10 cm, the trays are prone to particle resuspension. Please clarify whether small beads or water were placed at the bottom of the trays to suppress resuspension.

4) Please specify the installation position and height of the dust concentration monitor. Given that the gradient of dust concentration is key environmental data for this experiment.Why relevant measurements were not conducted？

5) Line 171: When the boundary layer thickness is 0.48 m, there appear to be only three valid wind velocity measurement points. Since the adopted fitting function contains three unknown parameters, the fitting results are not meaningful. Please provide the coefficient of determination for each wind velocity profile fitting.

6) It is recommended to number all equations for ease of citation.

7) Line 230: Generally, a monotonic relationship between deposition velocity and wind velocity is only recognized when solely considering dry deposition and ignoring the effect of resuspension.

8) The critical wind velocity for dust emission from the surface can be readily obtained. Please discuss its relationship with the three wind speeds adopted in the experiment, as this will effectively support the analysis of resuspension effects in the paper.

9) The y-axis label of Figure 3b is missing a closing parenthesis.

10) Line 274: Please set a in italic.

11) In lines 329–330, “peak” should be corrected to “Peak”. Please check the entire manuscript and ensure consistent formatting and capitalization of related symbols and terms for better readability.

12) In Figure 3(a), the k value for PM2.5 at 4 m s⁻¹ is shown as 4.03, whereas the text describes that k reaches or approaches its maximum at 2 m s⁻¹. Please check the data for consistency with the description and make corrections or provide clarification.

The present study investigates the quantification of size-resolved dust deposition kinetics in a large closed-circuit wind tunnel. Overall, the manuscript is well written, well organized, and strongly motivated, addressing a genuine and insufficiently constrained problem: the behavior of dry dust deposition under weak, post-storm background wind conditions, where existing parameterizations show large discrepancies and typically assume a monotonic response. The experimental facility is impressive, the kinetic framework provides a sensible and concise approach for summarizing the experimental runs, and the size-resolved analysis represents a notable strength of the study. The finding that deposition rates at 2 m s⁻¹ exceed those observed at 0, 4, and 6 m s⁻¹ for fine- to medium-sized particles is particularly interesting and may prove valuable for the development and improvement of deposition parameterizations. I have only a few minor comments that should be addressed by the authors before the manuscript is considered for publication.

1. Dust is generated through erosion of the soil source at a wind speed of 14 m s⁻¹, while the source bed remains present on the tunnel floor during the subsequent deposition phases conducted at 2, 4, and 6 m s⁻¹. Under the higher deposition wind speeds, the source bed may continue to emit particles or undergo re-erosion. Consequently, the observed suppression of net deposition at 4–6 m s⁻¹ may partly reflect ongoing source emissions rather than the re-entrainment of previously deposited particles. The manuscript attributes this behavior to “resuspension and horizontal export”; however, it does not clearly distinguish between continued emissions from the source bed and the re-entrainment of deposited material. Could the authors please comment on the relative importance of these two processes and discuss how they may affect the interpretation of the results?
2. A deposition velocity is defined as V_d = J/C. However, the quantities M(t), a, and J₀ appear to be reported as absolute values without dividing by the suspended concentration C(t), which both decays during a run and may differ between runs. The manuscript states that C₀ was recorded "to calibrate the initial-condition comparability across wind-speed treatments" (Lines 155–157) but does not state whether the fitted quantities are normalized by C₀. Could the authors please clarify this point and discuss its implications for the interpretation and comparison of the reported deposition kinetics?
3. In the wind tunnel, a wind-speed-dependent fraction of dust may be lost to the walls, ceiling, turning vanes, and fan assembly through inertial impaction, while recirculated air may continue to resupply dust to the test section. At higher wind speeds, deposition may plausibly shift toward vertical or internal tunnel surfaces, meaning that the observed reduction in floor deposition could partly reflect a redistribution of deposited material rather than a reduction in total deposition. The authors are therefore encouraged to address, at least qualitatively, the potential influence of wall losses and recirculation on the reported deposition patterns and their interpretation.
4. The experiments were performed using a cropland/agricultural soil. How different might the results be if other dust soil types had been used? The authors acknowledge this limitation in Section 4.4; however, it would be useful to include some further discussion, or at least qualitative estimates, of how soil type, mineralogy, texture, particle-size distribution, and aggregation state may influence the reported deposition kinetics.
5. I was not able to fully understand whether, and to what extent, the role of soil moisture was considered in the experimental design and analysis. Since soil moisture can influence particle cohesion, threshold friction velocity, dust emission rates, and potentially deposition behavior, its treatment may have important implications for the interpretation of the results. Could the authors please clarify whether soil moisture was measured or controlled during the experiments and discuss its potential impact on the reported deposition kinetics?
6. Lines 166 – 178: I suggest rephrasing this section and providing a more detailed explanation to help the reader better understand the methodology and interpretation of the results.
7. Figure 2: How important is the shape of mineral particles for the sedimentation process? Was particle shape considered in the analysis? Since particle shape can influence aerodynamic properties and settling velocities, a brief discussion of its potential impact on the results would be helpful.Lines 193–197: Are the PM classes cumulative or differential?
8. Figure 4: I may have misunderstood the calculation, but it appears that there could be an issue with the reported median value. Could the authors please check the calculation and clarify this point?
9. The authors should briefly discuss how the results of this study could be exploited in numerical simulations, particularly with respect to the development or improvement of dust deposition parameterizations.