Machine Learning to Identify Three Types of Oceanic Fronts Associated with the Changjiang Diluted Water in the East China Sea between 1997 and 2021

Round 1

Reviewer 1 Report

Congratulations to the authors. This manuscript is well written and organized and I think that it will be of great interest to readers performing research in the East China East. If possible, it would be useful if the authors use wider lines for contours in corresponding figures. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Changjiang Diluted Water is critical for analyzing biogeochemical processes in East China Sea. It’s meaningful to study the long-term variability based on satellite data.

However, this study couldn’t provide us much insightful knowledge about this topic. Results are superficial. The validation of the MPNN model is too simple. I don’t think this study could meet the requirement of the Journal at this stage.

 

1.  Quantitative analysis is expected for studying the spatial distribution of Chl a concentration and its relationships with environmental factors.

2.     Tidal effect should be considered before combining isohaline locations every day.

3.     SSD data could be validated with in situ measurements.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Overall the paper is well-written and scientifically robust and of good value for the Remote Sensing scientific community and worthy or publication in my opinion. I recommend a low-end major revision to address some comments below.

In my view, the article title could be revised to be clearer.

“Oceanic Front determined by the Changjiang Diluted Water Distribution in the East China Sea during Summers over the Past 25 years (1997–2021)”

What is determined? That a front exists, or the front location?

Also, this paper studies three distinct fronts, but that is not clear in the title.

Also, since machine learning is such a hot topic, the paper might get more reads and references if that is in the title.

From the paper, it is the distribution, location, frequency, etc.

So I recommend something like this for the title:

Using machine learning to identify three types of oceanic fronts associated with the Changjiang Diluted Water in the East China Sea between 1997 and 2021”

 

Line 13: This sentence is unclear: The Changjiang Diluted Water (CDW) in the ECS propagates northeastward and longitudinally formed ocean fronts.

I would rewrite as:

“The Changjiang Diluted Water (CDW) in the ECS propagates northeastward and forms longitudinally-oriented ocean fronts. “

 

Line 36 is awkward with the wording “colored matters” – what is this? Be more specific, like “dissolved solids or pollutants that reflect light in specific wavelength” These are mentioned on lines 57-58.

 

In Section 2.2.1, there should be some discussion of the accuracy/errors associated with the various products used, both in terms of temporal consistency (can the fronts be missed due to cloud cover) and the raw accuracy needed of the variables studies, and what the native data accuracy is.  Are there any possible sources of biases? What impacts might changes in available satellite data over the period of record impact the results?

 

Before machine learning, how are these fronts typically identified? I am assuming by gradients in the various parameters. Some high-level discussion of how oceanic fronts are identified in general, and what is better about the machine learning methods of this study should be included.

 

In my view Figures 4-7 are to many plots, too small to understand what is going on well. I would pick some specific images to focus on and make them larger, and then move the remaining images to the supplemental material.

I’d like to see a bit more discussion on what future studies should look at, there is just a broad statement about how more work is needed.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

To make it easy-read, figures could be improved by unifying the size, highlighting the contour lines.

Figure 4-7 could be merged in one.

 

 

Author Response

We merged figures 4-7 and changed the figure size and contour lines for better readability.

Reviewer 3 Report

The authors have carefully addressed the reviewer comments and I believe the manuscript is ready for publication after a careful spell/english check. Good work.

E

Author Response

Thank you for your kind comments. We got the English editing after your comment.