Black Carbon Concentration Estimation with Mobile-Based Measurements in a Complex Urban Environment
Round 1
Reviewer 1 Report
The manuscript used high-resolution mobile measured BC concentrations in West Oakland, CA. to develop land-use regression models and integrate four machine learning models. There are two issues that should be dealt with in the manuscript：
1. The model tuning part should be mentioned in the development progress.
2. Some states in this manuscript are not clear.
Some states in this manuscript are not clear.
Author Response
Thanks for the comments. Please check the attached file for our responses.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The manuscript presents a very interesting study but he manuscript is poorly written. Also, the quality of the figures needs improvement, at least 300 dpi. To get an idea of the quality of the writting, consider the first line abstract:
"Black carbon (BC) is an essential pollutant from transporation since it significantly impacts public health and climate change. Understanding its distribution in the complex urban environment"
The main problemis not the word "transporatation", the problem is the author claims that BC is essential fromroad transportation, which makes no sense.
Also, some paragraphs has twophrases and each paragraphs needs at least 3 phrases.
Resubmit
very bad
Author Response
Thanks for the comments. Please check the attached file for our responses.
Author Response File: Author Response.pdf
Reviewer 3 Report
The methodology they use is interesting, however it would be good to see the results of the different methods in more detail. There is a lack of more information about the values ​​of black carbon in the outliers, what levels they found, etc. As a summary, it's fine. The supplementary document lacks more information on the methods and results obtained.Author Response
Thanks for the comments. Please check the attached file for your responses.
Author Response File: Author Response.pdf
Reviewer 4 Report
Very inserting study where the authors describe about the integrated a land-used ,model with the four machine learning. I have fews suggestion for the authors. as follows:
Could you please describe the procedure of split data for the training and validation sets. and give the reason why you selected 80% and 20%
Please add the Taylor diagram for the comparison of models.
In your discussion of the NN obstacles, provide the citation information to support your claims.
1. The 80% and 20% data split was described in the methods section. Give more details on the method you used to divide the dataset. If the data includes a date, make a table to make the time period for the data clearer. 2. Please magnify and clear out the dot and change the red label's color in Figures 5-7. 3. For the x and y labels in Figure 8, please enlarge and replace the red dot to make the difference clearer. 4. Please separate the feature extraction from the machine learning model in Figure 1's workflow. Feature extraction comes first, followed by the model diagram. 5. I think in the model specification section, the author should separate into subsections and give a brief explanation of the model and learning algorithm. 6. Please explain the development of models and model tuning via concepts that readers may easily grasp using a flow chart or algorithm.7. For Figure 3, it should be replaced with a bar chart, so that it is easier for readers to understand.
8. Please modify Table 1 because it is unclear. Additionally, the author talks about overfitting; yet, the results in Table 1 show that the majority of the developed models are overfitting. Therefore, perhaps you might rework the model development or provide more information about the reasons why the model findings are unsatisfactory.9 Please add the Taylor diagram for the comparison of models.
10. In your discussion of the NN obstacles, provide the citation information to support your claims.
Author Response
Thanks for the comments. Please check the attached file for your responses.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
I'm fine with the changes
