Identifying PM_2.5-Related Health Burden in the Context of the Integrated Development of Urban Agglomeration Using Remote Sensing and GEMM Model

Round 1

Reviewer 1 Report

The authors used interpretable self-adaptive deep neural network model (SADNN) and global exposure mortality model (GEMM) with in-situ PM2.5 data, satellite AOD data and ERA5 reanalysis model data to analyze the correlation between air pollution and premature mortality in the Yangtze River Delta （YRD） from 2015 to 2019, the results showed that PM2.5 in the YRD region decreased year by year, but human health in large cities, especially in northern Anhui and Jiangsu Province, still faced great threats. The data used in this article are relatively substantial, the paper was well-structured, the language expression is accurate, and the results have certain guiding significance for policymakers.

However there are still some issues needed to be answered:

1) The introduction to the SADNN model is too simple, the reader cannot quickly understand how many nodes are used in the DNN model? How efficient is the SADNN model running?

2) What is the PM2.5 accuracy of SADNN simulation after adding ground-measured PM2.5, satellite AOD data and auxiliary data? Is it possible to cross-validate?

3) What is the data source for PM2.5?

4) Demographic Data Correction Section (3.3.1): The author used two datasets and mentioned that correction was carried out using China Population and Employment Statistics Yearbooks, please describe how to correct the former dataset?

5) Premature Mortality Estimation Section (3.3.3): How is premature mortality defined? Only the criteria for age greater than or equal to 25 years old are found in the article? Is this the only condition?

6) What does CI mean in line 257? What do the next two values mean?

7) In Figure 7, the results in different researches vary widely, up to or even more than 40%, how to explain such a large difference, which is more accurate?

8) The resolution of the figures is too low to see clearly the content on the picture, and it is recommended that the author change the clarity of the figures.

9) References section: Please verify the author's name, such as reference 7; Please unify the author's name; Please unify the format of the journal, full name or abbreviation; Please unify the format of ‘et al’.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

1.      Does the method of PM2.5 map come from References No. 21  Is the precise of mapping same to References No. 21

2.      Line 182-187 the data of World Pop is raster type, But data of the china population yearbook is region bype. How to correct the world population.

3.      Line 242-246 Four type as H-H L-L H-L and L-H is used to distinguish region characteristics, but what is H-H L-L H-L and L-H, how to define them.

4.      The data types is so much. How to make them consistent in both spatial and date.

5.      The impact of pollution on human health is slow. What is the scientific basis for using data from the same year in the article.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

To the Authors, 

please, see the attached file.

Comments for author File: Comments.pdf

From my point of view, the quality of English language is fine. I am not a native English speaker. therefore,  I would recommend that minor editing is required. 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thanks to the authors for the serious reply. After the revision, the article is clearer and easier for readers to understand.

The authors' reply also answered all my questions, and I personally think that the revised article satisfies the requirements of the journal and is recommended to be accepted for publication.

Reviewer 3 Report

To the authors, 

The authors answered to the comments I made on the manuscript.

The manuscript was revised by the authors following the comments and suggestions I made. I have no more suggestions/comments about this manuscript before its publication.

 

Minor editing of English language required.