Review Reports

Round 1

Reviewer 1 Report

The paper proposes an ensamble modeling method for satellite coverage trafic volume prediction  and the results are promising.  There are a few things that I would like the authors to consider:

Line 298, How is the value for the control parameter \delta found? what does it mean "small" / "enlarged"? Same question for the value in table 1. The novelty of the proposed method should be better empathized, with respect to the current state of art (how were the existing methods combined to achieve the optimal result). What tools are required to generate the experimental data and perform the experiments? Can the results be generalized for a network of satellites (using more satellites to cover the same area).

 

Minor remarks:

line 115, for clarity, write the complete name, and the acronym (PRS) in the subsection title. line 127, eq (3), the third term appears incorrect; in the second sum, j=1 and i<j results i=0, which is not correct. I think that it should be   i \neq j. Double check this equation. figure 1, the nadir point site is not marked. line 130, the notation for x, y with superscript in round brackets is not defined.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors propose an effective surrogate ensemble modeling named partial weighted aggregation method (PWTA). They propose to use it for better prediction in satellite coverage traffic volume (SCTV).

Perfect writing and understanding.

Line 191 BesTGMSE T capital letter.

Suggesting:

Figures 2 and 3 could indicate the source.
As you do in table 4, to use bold for ease analysis in table 3
To use the same row for each method in the tables: 3, 4 and 6 for ease comparison between them
Perhaps, After table 6, it could include a analysis or final comparation to highlight you contribution.

Clarification

In table 6 appears WTA and BestGMSE; and in tables 3 and 4 no. Perhaps, a clarification could be added.

Question:

Line 365.
What decision were it used to these values of points size?.
Less than 30 or upper than 80 not valid?.
Should be interesting to indicate some commentary.

Regards

Author Response

Please see the attachment.

Author Response File: Author Response.pdf