Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This research presents a global assessment of atmospheric dust as an air-quality and health hazard by combining satellite observation, reanalysis product, population data, and exposure–response function. The work is generally well-presented and interesting to readers. While the framework is useful, I have identified several issues in this manuscript. The work needs to be revised before it can be further considered publication. My specific comments are:

1. The manuscript tends to over-cite references in several places. For instance, Lines 124–127 cite 13 references [12–24] for the general statement that dust particles can be transported over long distances before being removed by wet deposition or gravitational settling. Similarly, Line 139 cites 22 references [36–57] for the statement that fine mineral particles are linked to cardiovascular and respiratory outcomes. These broad statements do not require such extensive citation lists. The authors should critically screen the references in the whole manuscript and retain only the most relevant and directly supportive studies.

2. I am concerned that there is an inconsistency in the baseline concentration used for PM10 relative-risk calculation. The PM10 baseline concentration in Eq. 1 is stated as the WHO annual mean safety concentration of PM10, but gives it as 10 μg m-3 (Line 395) . However, Table 1 lists the WHO annual PM10 guideline as 15 μg m-3. This also applies to PM2.5. Please clarify the exact number used in this study.

3. In section 2 and section 3, the manuscript should explain more explicitly how the PBL-constrained dust concentration was calculated from the LIVAS vertical dust profiles. As I understand it, the authors used ERA5 PBL height as the upper boundary and averaged the LIVAS dust mass concentration profiles between the surface elevation and the PBL height to obtain PBL-mean dust concentrations. However, the exact calculation should be described more clearly. For example, the authors should specify whether the averaging is layer-thickness-weighted, how partial layers at the PBL top are handled, whether monthly PBL heights are matched with monthly LIVAS profiles before long-term averaging.

4. This research uses LIVAS total dust and fine-mode dust as inputs to PM10 and PM2.5 health-risk equations. Yet the authors explicitly state that satellite-derived dust mass concentrations are not directly comparable to WHO AQG PM metrics, because LIVAS fine-mode dust represents particles smaller than 1 μm, whereas PM2.5 includes particles up to 2.5 μm. Similarly, LIVAS total dust may include particles larger than PM10. This size-definition mismatch is a major source of uncertainty in the exposure and risk estimates. he authors should be more cautious throughout Section 3 and avoid presenting the results as strict WHO guideline exceedances unless additional size-conversion or sensitivity analyses are provided.

5. Table A1 includes several non-sovereign territories. Therefore, the terms “nation” and “country” are not fully accurate. Please revise the table caption and variable names to use “countries/regions” instead.
6. Wording issues:

• Title, Lines 23, 39 and 895: as air quality hazard -> as an air quality hazard
• Lines 20,45, and 900: epidemiological-established  -> epidemiologically established
• Line 47: high industrialized countries -> highly industrialized countries
• Line 201: broader perceptive -> broader perspective
• Line 333: word -> world
• Line 453: Fig.5c . Figure 5 contains only panels 5a and 5b.
• Figure 9 panel labels: The upper-row panels in Figure 9 appear to be labeled (a) twice.

Author Response

The authors would like to thank the reviewer for his time, comments and suggestions. We did our best to incorporate the proposed changes and corrections in the revised manuscript, aiming at improving the presented paper. Following, you will find our responses, one by one to the comments addressed. 
Please see the attachment.

Kind regards,
Emmanouil Proestakis et al.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Please see the attached Word document for my detailed review comments.

Comments for author File: Comments.docx

Author Response

The authors would like to thank the reviewer for his time, comments and suggestions. We did our best to incorporate the proposed changes and corrections in the revised manuscript, aiming at improving the presented paper. Following, you will find our responses, one by one to the comments addressed.
Please see the attachment.

Kind regards,
Emmanouil Proestakis et al.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This study assesses the health risk of atmospheric dust to the global population using satellite-derived dust data, population estimates, and epidemiological models, revealing that ~33.5% of the world’s population is exposed to fine-mode dust concentrations exceeding the WHO annual PM2.5 guideline. This is a interesting study, with professional writing and a rigorous structure. To better present it to readers, the following suggestions are offered for the authors' consideration.

General comments

(1) The satellite fine-mode dust is defined as <1 μm, while WHO PM2.5 includes <2.5 μm. Although the authors acknowledge this discrepancy and treat WHO limits as indicative benchmarks, they still directly apply the satellite data to PM2.5 health formulas. This may systematically underestimate exposure, as 1–2.5 μm dust is also hazardous. Meanwhile, the risk coefficients are based on total PM2.5(including anthropogenic sources), while pure mineral dust is less toxic, introducing directional bias. Therefore, the overall health impact is uncertain, and a sensitivity analysis is recommended.

(2) The discussion lacks depth; a large portion of it is either only loosely related to the study or simply repeats the results. There is no comparison with similar studies, and a deeper discussion is missing.

Specific comments

L18-28: Aren't all the sentences a bit long?

L39: as air quality hazard ——>as an air quality hazard

L46: "broader regions" is ambiguous and it is not clear exactly where they refer to.

L47: high industrialized——>highly industrialized

L111: but also introduce mechanisms to account for……

L129-130: It should be emphasized, however, that……

L272-279: The sentence is too long.

L434-435: natural aerosol component ——> aerosol component primarily natural

L740: with the number going as high as to——>with values reaching up to

L751-755: Inconsistent tenses. It is suggested to unify the tenses.

L805-847: The extensive and detailed description of the responsibilities and websites of international organizations such as WMO, UNCCD, UNEP, and CAMS has nothing to do with the specific findings of this study.

L894-901: This is the research objective and should not be included in the conclusion section.

Author Response

The authors would like to thank the reviewer for his time, comments and suggestions. We did our best to incorporate the proposed changes and corrections in the revised manuscript, aiming at improving the presented paper. Following, you will find our responses, one by one to the comments addressed.
Please see the attachment.

Kind regards,
Emmanouil Proestakis et al.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Atmospheric Dust as Air Quality Hazard to the World Population

The paper uses CALIOP LIVAS dust (fine and coarse) data and a PM10/PM2.5 based epidemiological relationship to compute health risk factors at grid cell and country levels. The paper reports that dust poses significant mortality risk if the concentration is higher than the WHO recommended baseline. However, there are three major concerns about the method, which need to be addressed before the paper can be considered for further review. 

Representativeness of the epidemiological relationship & CALIOP LIVAS data

1. This paper uses a single value for the coefficient beta (in Eqns 1 and 2) for every region/grid cell. However, the literature-based relationship from certain regions may not be applicable to others due to differences in living standards, socioeconomic status, exposure patterns, baseline mortality rates, etc. Applying the same relationship everywhere seems not justifiable.
   
   
2. In addition, the size range of the particles (LIVAS fine and coarse modes) is not matching with the original epidemiological studies (PM10 and PM2.5) for use in Eqns 1 and 2. This caveat is not stated clearly in the paper except at one point “For the needs of the present study, the ESA LIVAS-based PMtotal_dust and PMfine-mode dust are considered as input in Eq.1-3 as PM10 and PM2.5, respectively.” What is the purpose (‘For the needs of the present study …’) of a research paper? It is to report a good science. 
   
   
3. I think CALIOP data are not provided (or are uncertain) for the lowest 0.5–1 km. As a result, the data may be irrelevant and/or underestimate aerosol concentrations close to the ground, where human exposure is most relevant. This also breaks the basis of the epidemiological relationship used (which most likely used concentration data from ground level).

Minor comments

1. Abstract: include information that the epidemiological relationship is an approximate one (see comments above)
2. L141: I think Valley Fever is not reliably linked to dust, despite some papers indicating so as they did not account for population increase and changes in reporting procedures of the valley fever incidences.
3. L230: provide the size range of these fine and coarse mode particles. This information is critical for the paper, because the epidemiological information is based on certain size range.
4. Fig 1, and as part of the method: explain the sampling bias in the LIVAS AOD regarding missing values in time and space (along vertical and horizontal directions) over the averaging period.
5. Sect 2.1.2: Explain how the population was harmonized as different countries carry out census at different times, and what epochs mean (2010,2015,2020).
6. L359-362: provide this statement specifically referring to dust aerosol, which is the focus of the paper: does uniform mixing within PBL hold true for dust aerosols? Later in this para and the next, clarify if such uniform mixing assumption is utilized or not (in this study).
7. Fig 3 (PBL height map) seems unnecessary. How is PBL used in this study? The method does not explain it. Method should be explained well for reproducibility, especially regarding how the CALIPSO product was processed; if it was simply used from the already existing product, it should be stated explicitly.
8. L453: There is no Fig 5c.
9. L533-548: When discussing the anthropogenic dust sources it is quite appropriate to make use of Joshi (2021) (https://doi.org/10.1016/j.atmosenv.2021.118658)
10. L719-21: but figures are shown only for PM10?
11. L729: what is short term vs long-term exposure?
12. L803-808: It is an overstatement - the study did not ‘clearly’ highlight the impact – due to the caveats of the method (see the three major comments)
13. L809-834: This is a common knowledge and does not add anything significant or insightful.
14. L836-884: There is no scientific discussion of the present topic, but just some general information on what others are doing or have done.
15. Conclusions: These are not new findings - it has been widely known that dust poses health hazards, and that dust is heterogeneously distributed in space. Also, limitations of the study are not presented (e.g., see the cited reference Giannadaki et al., 2014).

 

Author Response

The authors would like to thank the reviewer for his time, comments and suggestions. We did our best to incorporate the proposed changes and corrections in the revised manuscript, aiming at improving the presented paper. Following, you will find our responses, one by one to the comments addressed.
Please see the attachment.

Kind regards,
Emmanouil Proestakis et al.

Author Response File: Author Response.pdf

Reviewer 5 Report

Comments and Suggestions for Authors

A very useful overview and analysis of dust and its health impacts over extensive domain (60 deg N and S) for extensively averaged time frame and high resolution. So the study does not include high latitudes – what happens in these regions? No dust?

It is well known PBL varies over diurnal and seasonal basis, can finer time scales be accessed? What is precise time of day of satellite determinations?

Surprised to see western USA hardly any effects or data on dust when several studies show impacts there. Is it due to resolution?

How would impending future climate scenarios impact results? If natural dust dominates and health is harmful what are the adaptation avenues?

Author Response

The authors would like to thank the reviewer for his time, comments and suggestions. We did our best to incorporate the proposed changes and corrections in the revised manuscript, aiming at improving the presented paper. Following, you will find our responses, one by one to the comments addressed.
Please see the attachment.

Kind regards,
Emmanouil Proestakis et al.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

No comments.

Author Response

The authors would like to thank the reviewer.
Kind regards,
Emmanouil Proestakis

Reviewer 3 Report

Comments and Suggestions for Authors

The overall quality of the revised manuscript has been significantly improved. I have one final minor suggestion: would the main conclusions remain robust if the 1–2.5 μm dust fraction contributed 20–50% of the fine-mode particle mass? Furthermore, how would the risk estimates change if the toxicity coefficient of pure mineral dust were assumed to be 0.5–1.0 times that of the total PM2.5 mixture? I suggest that the authors add a brief discussion of these uncertainties to further strengthen the paper.

Author Response

Atmospheric Dust as Air Quality Hazard to the World Population

 

The authors would like to thank the reviewer for his time, comments and suggestions. We did our best to incorporate the proposed changes and corrections in the revised manuscript, aiming at improving the presented paper. Following, you will find our responses, one by one to the comments addressed.

 

Kind regards,

Emmanouil Proestakis et al.

 

 

Reviewer’s Comments

 

The overall quality of the revised manuscript has been significantly improved. I have one final minor suggestion: would the main conclusions remain robust if the 1–2.5 μm dust fraction contributed 20–50% of the fine-mode particle mass?

 

We thank the Reviewer for the positive evaluation of the revision. Regarding the first comment, we consider that there is a negative bias in the estimated RRs for the fine fraction as the formulations require PM2.5 concentrations, i.e. all particles up to 2.5 μm. It simply indicates that the RRs estimated using the fine dust fractions are lower compared to ones if the cut-off was at 2.5 and not at 1 μm. Since this is the case in the present application and considering that the bias is equitable in all cases (as for all areas the 1 μm fraction was evaluated) the results presented are relative to each other. This is why we added the discussion regarding the different size fractions between LIVAS data and epidemiological expressions.

Based on the above, we have modified the discussions in lines 473-478.

 

Furthermore, how would the risk estimates change if the toxicity coefficient of pure mineral dust were assumed to be 0.5–1.0 times that of the total PM2.5 mixture? I suggest that the authors add a brief discussion of these uncertainties to further strengthen the paper.

 

The type of data evaluated in this work, i.e. mass concentrations, cannot support any solid conclusion based on the question raised by the Reviewer. However, toxicity of fine airborne particles is driven by different compounds including typical crustal elements (Franklin et al 2008). In addition, exposure to mineral dust has been associated with several health effects (Fubini and Areán 1999). In any case, the chemical content variation of airborne particles and the impact it poses to human health is quite complex due to the numerous contributing factors. Epidemiological studies come to fill the gap and provide insights into the adverse health effects by providing concentration-response functions.

A relative discussion was added in lines 463-469.

 

Franklin et al. 2008, The Role of Particle Composition on the Association Between PM_2.5 and Mortality.

10.1097/EDE.0b013e3181812bb7

Fubini and Areán 1999, Chemical aspects of the toxicity of inhaled mineral dusts.

https://doi.org/10.1039/A805639K

Author Response File: Author Response.pdf