Impact of Collaborative Agglomeration of Manufacturing and Producer Services on Air Quality: Evidence from the Emission Reduction of PM2.5, NOx and SO2 in China
Round 1
Reviewer 1 Report
Main comments:
QGIS software is used to visualize and analyze hazardous air pollutants in 29 provinces of China.
What is the full name of QGIS?
This paper explores the effect of industrial agglomeration on three major air pollution sources, i.e., SO2, PM2.5, and NOx.
These sources have complex sources and sinks,for example, atmospheric CO has significnat correlation with NOx(Metrology and Measurement Systems, XXII(1): 25-38, 2015),
and the related analysis showed a significant positive correlation of PM2.5 with SO2, NO2 and CO as opposed to a negative correlation with O3(Air Quality, Atmosphere & Health 10:725–736,2017).
Did the author considering the correlation of pollution sources in the developed algorithm model? which might be usefully for identifying pollution sources accurately.
Please discuss this issue appropriately in the manuscript.
As for Data sources used in this work, the author stated that the provincial panel data from 29 Chinese provinces (excluding Tibet, Qinghai, Hongkong, Macau and Taiwan) from 2006 to 2019 are used in this study. How and where did the author download the raw pollutant data ? Please provide the related information.
Whether the analysis results based on the developed algorithm model are compared or tested with other satellite remote sensing observation data ?
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In the presented paper, Ye ...et al. investigate the effect of industrial agglomeration on air pollution in mainland China. For this, they subdivide into agglomeration of manufacturing, of producer services and co-agglomeration of both. They hypothesize, that scale effects of co-agglomeration and establishing of producer services can effectively mitigate air pollution and provide a mediator model which entangles all those factors.
I suggest to provide additional figures which show timeline as line diagram with time on x-axis for PM 2.5, SO2 and NOx like in figures 2 to 4. All results are related to mainland China. For comparison purposes I recommend, to provide comparable numbers describing PM 2.5, SO2 and NOx in Europe as well as in the U.S. (years 2006 ... 2019) for at least the descriptive tables.
In line 370 it is mentioned, that data does not contain apparent outliers. Please describe chosen method for outlier detection and handling.
In table 2 please provide units for each variable.
For tables 3 to 5 the abbreviation FGLS is given, but abbreviation OLS (I guess "ordinary least square") was omitted. Please add.
Also in tables 3 to 5 the reason of giving ID = Yes in the bottom line of the tables remains unclear.
In section 3.3.1, please give a reference for "first phase lag test".
In section 3.4.1, it remains unclear, what the difference between equation (13) and equation (14) is, because the right side of both equations seems to be the same. Please provide a definition for "Energe". Why is the left side of equation (13) and equation (15) the same (ln PMit)?
The same applies for section 3.4.2 and equations (16) to (18).
After adding numbers for Europe and U.S. in section 2 (see my notice above), please also mention and compare those numbers in the conclusion section.
line 677: What is a "rational industrial structure"?
In the conclusion I miss an educational moment. Optimizing industrial and intensifying R&D seems expectable.
What about change in CO2 emissions entangled with described industrial structure changes? Are they high or low compared to needed reduction path?
A more general question: What about changing of energy sources to renewable energies? Please add a section to the conclusion which describes potential of supplementation of energy sources by renewables in China. Especially in the light of global climate change I'd like to see a paragraph with consequences on this topic and, if possible, with absolute numbers on CO2 emissions.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
In this article, authors used Chinese provincial panel data to systematically study the relationship between industrial agglomeration and the three major air pollution sources by combining the generalized FGLS panel econometric model with SDM. It is an interesting topic worthwhile to study, nevertheless I have some major concerns to be tackled with before possible acceptance for publication. In addition, the language needs to be further strengthened, and should pay attention to the integrity and accuracy of information expression.
1. Line 256 I would suggest to add any reference for QGIS software.
2. Figure 7. Is it possible to change a color scheme to make the picture look more distinct?
3. Data sources of “Control variables”
Line 403 “In terms of control variables, the coefficients of per capita GDP and the quadratic term of per capita GDP in Tables 3–5 are the opposite…” The problem with the regression analysis in this paper is that covariance of the different parameters is not accounted for. For example, are per capita GDP and quadratic term of per capita GDP correlated? Also the attitude seems to be that any positive correlation is indication of causality, which is not the case.
4. As a conventional air pollutant, only the emission of PM2.5 is significantly affected by the industrial agglomeration, but NOx and SO2 are not. What is the reason for this difference? It may be explained.
5. It is recommended to merge the tables with similar presentation for easier comparison. In addition, the language of the full text needs to be strengthened. The sentence is too long and there are still some incompatibilities. For example, line 274-278, “To be specific, the location entropy index [50] adopts the ratio between the proportion of the number of employees in an industry of a region of the total number of employees in the industry and the proportion of the number of employees in the industry of the total number of employees in the country.” Therefore, the language needs to be summarized and simplified to make the expression clearer and easier to read across the manuscript.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Now, I suggest acceptance of this manuscript.
Reviewer 3 Report
I’d like to thank the authors for their response and clarifications in the manuscript. The manuscript has improved substantially.
