A New Coupling Method for PM_2.5 Concentration Estimation by the Satellite-Based Semiempirical Model and Numerical Model

Round 1

Reviewer 1 Report

The paper has improved since its previous iteration. I have a few minor outstanding comments listed below.

Line 44: The WHO standard has recently been changed to 5 micrograms per cubic meter.

Line 63: “attained” should be “obtained”.

Lines 69-72: The meaning of this is a bit unclear. In particular, does the first part of the sentence mean that currently only ground observations provide the aerosol characteristics? Also, is “the model” the SEM model?

Figure 3: Suggest denoting the sites used for calibration and for validation using different colors.

Line 165: “brief” should be “briefly summarize”.

Line 169: “driving” should be “driven”.

Line 181: “indicate” should be “indicates”.

Line 228: “resulting” should be “results”.

Equation 4: I think that the “gamma prime” term should appear outside the natural logarithm in the second part of the equation.

Lines 232-233: Might be better to explicitly state the properties were derived for each grid cell and for each month.

Line 313: What is meant by “statistical evaluation” here?

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

This revised manuscript is much clearer than the original.  However, it now raises two issues that should be clarified.  The discussion of the seasonal differences suggests that desert dust and wind blown soils from  agricultural activity are not included. Is that correct? It would be helpful to list the aerosol types that are included and those that are excluded.

The relative humidity dependence/hygroscopic growth in Section 2.4 SEM is really based on a functional fit to the observations (SEM) rather than aerosol physics based on aerosol composition correct?

Lines 165-167: the English could be improved.  It might be better to say "...we briefly describe the key equations..." "...The details of the derivation can be found in LIn et al..."

Lines 253-256: the English could be improved. I think that you mean that in addition to the RH dependence through hygroscopic growth you are also including the RH dependence on fine mode fraction and mass extinction efficiency. 

Line 183 "grided" should be "gridded"

In Figure 5 through Figure 10 the upper and lower right-hand panels (d and h) "OTC" should be "OCT"

"Situ" should be replaced with "In Situ"

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Please check the attachment.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The manuscript is well revised following my comments and I suggest to accept it.

Author Response

Thank you for your positive feedback.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The manuscript "A new coupling method for PM_2.5 concentration estimation by satellite-based semi-empirical model and numerical model" describes an improved treatment of the aerosol vertical distribution using an aerosol scale height to better estimate the PM_2.5 concentration from satellite measurements.  The paper is of interest and should be published.  However, the paper needs editing and proof-reading to improve the clarity of the presentation.

For example, the sentence on lines 19 and 20 of the abstract "The numerical model can provide better spatial distribution information of aerosol properties in high spatial resolution at a large scale based on emission driven meteorology." is confusing.  Do you mean to say that the numerical model can provide large scale, high spatial resolution aerosol property information based on emission data driven by large scale meteorology?

Throughout the paper, the aerosol distributions are described as "in high spatial resolution" where I think it would make more sense to be "at high spatial resolution."

LIne 67: "...the updated technology is suitable for operation." Makes no sense.  Do you mean "...the updated methodology is suitable for operational use?"

Line 71: "grided" should be "gridded"

Sentence on lines 78-80 needs to be rewritten.  Do you mean to say that the simulated aerosol properties should be more reliable based on the use of higher spatial resolution emission data combined with realistic meteorology?

Line 305 should maybe be "Statistical evaluation"

LIne 327 maybe you would like to add the word comparison to the table caption...such as Comparison of the CSEN and SEM Models.

Line 355 "Overall, Significant" should be "Overall, significant..."

Last two lines (385 and 386) of section 3.2.3 are not a paragraph and also reminded me that there are some problems with tenses throughout the papers.  So this maybe should be "The overall results obtained with the CSEN method provide more realistic values when compared with the ground-based monitoring network data."  Then add at least one other sentence to make it a paragraph.

LIne 392...might be better as "Large scale high resolution measurements of PM_2.5 are fundamental..."

Line 397 "combing" should be "combining" (also in LIne 224).

LIne 401: might be better as "The physically-based SEM model described in our previous study..."  The discussion and conclusion section doesn't really summarize what is different about the CSEN model and why it is better.

Author Response

Dear Reviewer,

 

We thank you for the overall positive feedbacks and the numerous valuable and scrutinizing comments which helped us to improve the manuscript. The major modification includes the following items.

1. Add the model performance of the WRF-Chem in Table 2 and corresponding comparison analysis.
2. Update figures 1b
3. Use consistent notation across equations and make scientific methodology clearer.
4. Polish the text following the reviewer’s suggestion
5. Check through the unclear expression, grammatical errors and typos throughout the MS.

 

We also prepare a point-by-point response to the reviewer’s comments. Texts in Black are the reviewers’ comments, and those in normal black are our responses. The blue texts are revised sentences in the revised manuscript. All the line numbers in blue are referred to the track-change version. We hope that you and the referee will find the changes satisfactory and we are looking forward to hearing from you soon.

Author Response File: Author Response.docx

Reviewer 2 Report

General Comments

Overall, the paper presents a good overview of the methodology and shows compelling results for its application.

The introduction does a good job of explaining the relevance of a semi-empirical model and motivating the use of CTM information in your method. However, some further improvements could be made. You may want to describe further how aerosol properties like hygroscopicity affect the relationship between AOD and PM_2.5 concentrations, how these properties are currently estimated for use in models (i.e., through surface observations at discrete locations), and how the CTM could improve this. This will make it very clear for the reader why you have chosen to pursue this path.

For the description of the methodology, especially in terms of the equations and terms used, some additional description should be included. In particular, the use of consistent notation across equations is important (for example, several different symbols are used to represent the extinction coefficient). Inconsistencies in notation between some equations may also lead to confusion (for example, in deriving equation 4 from equation 2, an exponential or logarithmic term is missing). It should be made clear what the sources for each of these terms are in the final CSEN model; that is, which terms are taken from the CTM model, which are obtained from the satellite retrievals, and which (if any) are obtained from other data sources. Finally, some distinction should be made between terms which are considered time-dependent (for example, humidity and AOD will change hour-to-hour) and other terms which are assumed to be relatively static in time (if I understand correctly, the aerosol properties are assumed to be constant for each month, although I may have misunderstood this).

If possible, it would be interesting to also include the performance of the WRF-Chem model itself for estimating surface PM_2.5, alongside the SEM and SEN methods. It is indicated in the text that such an approach has the potential for high bias; it would be interesting to verify this using the simulation results you used for the derivation of the aerosol properties. It would also help to characterize the additional benefits of the CSEN method, namely the inclusion of AOD data, versus what is possible to obtain with the model alone.

Specific Comments

Line 16: It is unclear what the “limited information on the spatial resolution of observations” is referring to. Consider rephrasing this statement.

Line 22: “including” is redundant here since you mention that there are two parameters and you list both of them. Consider removing this word.

Line 29: Rather than saying that the pollution concentration is “caused by” hygroscopic growth, it is more the case that hygroscopic growth affects the relationship between AOD and pollutant concentrations and should therefore be accounted for. Consider rephrasing this statement.

Line 42: The “3” in the concentration units should be in super-script.

Line 45: The word “inversion” here is unclear; consider using a different word like “estimation”. Similar comment for Line 49.

Line 60: Consider replacing “which” with “but instead”; the point to be made here is that the semi-empirical model explains the physical mechanism of the relationships, whereas other models which rely on (for example) geographical data do not.

Line 62: The word “connection” here is vague. You may want to explain in more details how this “connection” is made.

Line 74: I think the word “mass” is superfluous here; “aerosol size distribution” covers this concept.

Line 80: The word “simulated” is repeated an extra time here and can be removed.

Line 84: To be clear, I would suggest ending the sentence with “…required by the PM2.5 estimation from SEM.”

Lines 98-99: I would suggest rephrasing “providing daily sustainable monitoring aerosol properties of the earth.” as “providing sustained daily monitoring of aerosols around the Earth.”; note that “Earth” should be capitalized.

Line 100: Suggest removing “is” here; it is redundant with another “is” later in the sentence.

Line 111: “average deviation” should be defined. Also, I would expect “MB” to stand for “Mean Bias”.

Figure 1b: It is not immediately clear what each “dot” in the plot represents; is it the relationship between surface and satellite AOD for on overpass at some site? This should be stated. I would also suggest possibly color-coding and/or shape-coding the points (for example, different colors of the points could represent different months of the year and different shapes could be used to represent the different AERONET sites) this could provide some additional insight into the AOD retrieval accuracy and what factors affect this.

Line 135: I don’t believe “GTS” has been defined yet; please consider defining it here.

Line 138: “s” should be “a”.

Line 143: It is not clear what “this map” refers to. I believe you are referring to the red boxes in Figure 2a. It should be noted explicitly what these boxes refer to.

Lines 149-150: “pollutions emitted” should be “pollution is emitted” or “pollutants are emitted”.

Lines 151-153: Suggest rephrasing “In this study, two key parameters integrated reference value (K) (and the integrated humidity coefficient (γ′) of aerosol characteristics used in estimating PM2.5 from AOD).” as “In this study, two key parameters are taken from the WRF-Chem model outputs and used to support the estimation of PM2.5 from AOD. These are the integrated reference value (K) and the integrated humidity coefficient (γ′) of the aerosol characteristics.”.

Lines 153-155: I would suggest referring to specific equations used later to describe how these properties can be derived from the model outputs.

Lines 169-172: It seems that the same symbol is used to represent the near ground aerosol extinction coefficient and the mass extinction efficiency. Are these necessarily the same?

Line 180: First “was” should be “of”.

Line 186: “e.g.” should be “i.e.”.

Line 214: There is an extra “.” at the end of the sentence.

Equation 4: based on this equation there may be a missing “exponential” term in Equation 2, or there is a missing natural logarithm term in this equation.

Line 221: Linear regression would require at least two datapoints. I am assuming that, for a single grid cell in the model, you are using multiple time-steps of the simulation to obtain multiple points for use in this regression. You should describe this process a little more. For example, are the aerosol properties derived once of each grid cell using the entire simulation periods, or are they derived for each month, or for each day, etc.

Line 227: The assumption that H varies smoothly at a regional scale could be explained in more detail. Does this mean that you are assuming that H is constant across your domain?

Line 238: “possible” should be “how”.

Line 258: “reasons” should be “regions”.

Line 284: “and or” should be “and/or”.

Line 294: “friction” should be “fraction”.

Line 306: 1996 monitors are indicated here, whereas on line 141, 1695 monitors are indicated as being used for validation.

Line 308: “mess” should be “mass”.

Line 314: The “3” in the concentration units should be in super-script.

Figure 6: For July, I think that the figures used for CSEN and SEM results are the same; this might be a mistake. The text in the figures is also very small; for the final publication I would suggest that larger figures be used if possible.

Lines 319-326: The exact numerical values of the results are listed in Table 2; there is no need to repeat them in the text.

Line 334: “and” is redundant here.

Line 351: “mess” should be “mass”.

Figure 8: The relative deviation should be defined. Is this a kind of normalized bias?

Line 367: “mess” should be “mass”.

Lines 371-373: The “3” in the concentration units should be in super-script.

Figure 9: The text in the figure is very small; suggest using larger text or a larger figure in the final publication.

Line 396: “variability” should be “variability of”.

Line 397: “marching” should be “machine”.

Lines 413-414: “the overall comparable spatial patterns with each other” should be “their overall spatial patterns are comparable”.

Lines 418-419: The “3” in the concentration units should be in super-script. Also, instead of repeating the numerical outcomes here, I would suggest summarizing these with a brief statement, for example “RMSE was reduced by between 1 and 7 µg/m³”.

Line 421: “completer” should be “more complete”.

Line 422: “practical” should be “realistic”.

Lines 429-430: There do exist methods for “gap filling” in satellite retrievals, for example using more complete long-term average maps or using other data sources (especially CTM).

Author Response

Dear Reviewer,

 

We thank you for the overall positive feedbacks and the numerous valuable and scrutinizing comments which helped us to improve the manuscript. The major modification includes the following items.

1. Add the model performance of the WRF-Chem in Table 2 and corresponding comparison analysis.
2. Update figures 1b
3. Use consistent notation across equations and make scientific methodology clearer.
4. Polish the text following the reviewer’s suggestion
5. Check through the unclear expression, grammatical errors and typos throughout the MS.

 

We also prepare a point-by-point response to the reviewer’s comments. Texts in Black are the reviewers’ comments, and those in normal black are our responses. The blue texts are revised sentences in the revised manuscript. All the line numbers in blue are referred to the track-change version. 

Author Response File: Author Response.docx