Advancing Sea Surface Height Retrieval through Global Navigation Satellite System Reflectometry: A Model Interaction Approach with Cyclone Global Navigation Satellite System and FengYun-3E Measurements
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsAll comments are in the attached file
Comments for author File: 
Comments.pdf
Some sentences\paragraphs are difficult to understand, especially in the introduction and abstract.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsPaper deals with investigation of possibilities to obtain the Sea Surface Heights model, applying the innovative technique of GNSS-R for remote sensing.
Authors use the dataset from spaceborne GNSS-R platforms,
Cyclone Global Navigation Satellite System (CYGNSS) and FengYun-3E (FY-3E), as the primary source of data for retrieving sea surface height (SSH).
The goal of the authors is to develop a method based on artificial neural networks. Method and data allow to detect the ocean surface
height with a precision of meter-level accuracy. Usage of short period data (about half month) enabled to achieve the Root Mean Square Error of 1.03 meter.
It is clear, that authors did achieved the main goal by developing the algorithm for data analysis and calculations.
However accuracy of detected sea surface heights is under question, and the coverage is very limited. Even authors compare achieved accuracy against the DTU21 model, which is, first of all global, and secondly, much more accurate. Reviewer just looked into it making comparison against quasigeoid model in the Baltic sea region, and accuracy of DTU21 model is about 10 cm. Question is, why we should produce the model, which is much more worse?
Tables and figures are fine.
List of references is very good, however the paper on DTU21 should be added:
Ole Baltazar Andersen, Stine Kildegaard Rose, Adili Abulaitijiang, Shengjun Zhang, and Sara Fleury. The DTU21 global mean sea surface and first evaluation. Earth Syst. Sci. Data, 15, 4065–4075, 2023
https://doi.org/10.5194/essd-15-4065-2023
Scientific soundness is high, practical findings are important and promising.
In general I like presented paper, it gives a broad introduction on achievements of other authors, presents the theoretical formulas and algorithms, and gives the convincing practical results.
Paper is written in good manner, well structured and could be printed in "Remote sensing".
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe paper explores an innovative approach to sea surface height (SSH) measurement using GNSS-R technology with data from CYGNSS and FY-3E, particularly through the use of artificial neural networks (ANN) and an Interactive Multiple Model (IMM) combined with a Kalman Filter (KF). The study holds significant implications for oceanographic research. As a reviewer, I appreciate the work done and offer the following comments for your consideration:
I have observed that while you have made a commendable effort to cite relevant literature, there is room for improvement regarding the inclusion of foundational and seminal works in the field, particularly concerning soil moisture retrieval and sea ice inversion using GNSS-R technology. The references provided include several conference papers, which, although they may contribute to the state of research, are not always considered the most authoritative sources for establishing the theoretical and empirical basis of a study.
The paper presents a novel model interaction technology and IMM-KF method that significantly contribute to the precision of SSH data fusion. It is recommended that the authors further highlight the advantages of this method over existing technologies and potential application scenarios.
The paper mentions the use of CYGNSS and FY-3E datasets but does not elaborate on the data preprocessing steps. The reviewer suggests that the authors provide more information on data filtering, outlier treatment, and data quality control measures.
Regarding the robustness of the model, the reviewer recommends that the authors discuss the model's performance under various oceanic environmental conditions (e.g., wind speed, wave conditions) and provide corresponding sensitivity analysis.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
