Spaceborne GNSS-R Soil Moisture Retrieval: Status, Development Opportunities, and Challenges

Round 1

Reviewer 1 Report

The authors have created an article which seeks to provide a brief review of the current status and limitations of the remote sensing of soil moisture using the GNSS-R technique. It also uses modeling and simulation tools to provide answers to subjects such as the coherence / incoherence problem, vegetation contributions, and the effect of azimuth variation. The paper takes the noble task of identifying and providing solutions to many issues in GNSS-R.

The thesis statement of the paper (Lines 104 – 109) is very broad and confusing. It states that CYGNSS data is used, but no figures seem to make use of it. It is clear, however, that the paper hopes to be a review of CYGNSS inversion techniques and a simulation study.

The CYGNSS review is incomplete, inconclusive, and incorrect at times. If this paper means to serve as a review, Table 1 is incomplete. More than 6 references should be used to summarize the works performed in the literature. The authors state that, “Chew’s work [30] does not specify whether the coherent or the non-coherent data is used.” This is incorrect. [30] establishes that they assume complete coherence from CYGNSS data.

The authors cite references for the model used in the simulations of this paper, but some of the relevant structures of the model should be made clear. I do not know what soil moisture dielectric model is used in this paper, for example. The reader should not be required to search other references to learn about simulation assumptions used in the current paper. I am uncertain of many assumptions used in this paper. It is unclear what kind of surface is assumed in many simulations. This paper needs a well-defined methodology section which outlines the common assumptions made from simulation to simulation.

Perhaps the worst offense is that the authors conclude that coherence is greater than incoherence. The wording is declarative and absolute, but many simulations over the last year have visualized that topographic effects can produce dominantly incoherent pixels across delay-Doppler maps. If this paper serves to be a review paper, this cannot be neglected.

Many of the simulations are inconclusive and confusing. I am not sure how a simulation of the GNSS-R response towards Aspen plays into the paper’s thesis of “expound[ing] its [GNSS-R soil moisture retrieval] current research status and existing problems, and discuss its development opportunities and challenges in combination with microwave scattering models.” I do not recall any of the references stating that Aspen in particular was an existing problem, development opportunity, or challenge for CYGNSS-based soil moisture retrieval. I must ask how this simulation plays into this paper’s thesis.

I would highly recommend the reviewers considering a major restructuring of this paper. The most important thing for the authors to do is address the technical writing of this paper to make their intentions and methodologies extremely clear. It is currently an incomplete review paper and an inconclusive simulation paper. I suggest the authors either make this exclusively a review paper without modeling and simulation or create a focused simulation paper that addresses a particular question from the literature in detail.

Author Response

Dear Reviewer,

   Thank you very much for your comments and suggestions.

   For more detail, please see the attachment.

Best regards

Xuerui

Author Response File: Author Response.docx

Reviewer 2 Report

The explanations are sometimes ambiguous or incomplete especially for readers who are not familiar with GNSS-R technique. For example, the information given in Table-1 is misleading. Comparing the methods based on correlation coefficients or RMSE using few test-sites leads to wrong conclusion. The test sites used in these studies are not global and differ from one study to another. Therefore, the results cannot be used to compare the retrievals skill.

Some of the statements are inaccurate, e.g. line 142 "Clarizia and Al-Khaldi [32, 33] assume that the energy is mainly non-coherent" except [33], all other studies assume that reflections from land surface are mainly coherent.

The limitations of the retrieval algorithms were not fully discussed. The methods should be evaluated from different aspects: e.g. special resolution of the final product, number of required auxiliary data, applied corrections and approximations, ability to be implemented in a near-real-time operation, …

Author Response

Dear Reviewer,

   Thank you very much for your comments and suggestions.

   For more detail, please see the attachment.

Best regards

Xuerui

Author Response File: Author Response.docx

Reviewer 3 Report

In recent years, GNSS-R has gradually become a research hotspot of the intersection of satellite geodetic survey and remote sensing. It is very interesting to use satellite geodetic technology to do remote sensing research, which is also my research direction. Previous research mainly focused on sea ice, snow and sea level. In recent years, research on soil moisture in foundations has gradually increased, but research on soil moisture on satellite-based platforms is still in the exploratory stage. This Manuscript reviews the current research on soil moisture in satellite-based soils, and summarizes the work and contributions of the existing literature in detail.

Through the comparison of existing literature, some problems in the retrieval of satellite-based soil moisture were put forward and analyzed in detail (Polarization, Coherent and non-coherent scattering components, scattering zenith angle and azimuth angle, Surface roughness, Vegetation, Effective isotropic radiated power) (EIRP), Radio frequency interference (RFI)). The Manuscript has certain reference significance for related researchers, but there are some small problems in the article, which are worthy of discussion.

 

1. line 73: The backscattering coefficient and soil moisture are nonlinear, and the influence of soil roughness on the backscattering coefficient is even greater than that of soil moisture. The expression here is not rigorous..
2. line 98: It should be Bufeng-1 not Fenfeng-1.
3. line 261-264: After the direct signal is reflected from the ground, the polarization mode changes, from circular polarization to elliptical polarization. Elliptical polarization includes left-hand circular polarization and right-hand circular polarization, not just left-hand circular polarization. You can refer to the article published by Kegen Yu in TGRS in 2019.

Author Response

Dear Reviewer,

   Thank you very much for your comments and suggestions.

   For more detail, please see the attachment.

Best regards

Xuerui

Author Response File: Author Response.docx

Reviewer 4 Report

Manuscript ID: remotesensing-979574

 

Type of manuscript: Review

 

Title: Spaceborne GNSS-R Soil Moisture Retrieval: Status, Development Opportunities, and Challenges

 

Authors: Xuerui Wu, Wenxiao Ma, Junming Xia, Weihua Bai, Shuanggen Jin, and Andres Calabia

 

The manuscript presents a review of the current research status of soil moisture retrieval using the innovative approach of GNSS-R (Global Navigation Satellite Systems Reflectometry) methodology, highlighting the limitations of existing research in some aspects like; observation geometry, polarization, and coherent and non-coherent scattering.

 

Some minor changes are necessary to fit MDPI-Remote Sensing high quality standard and to be suitable for publication.

 

In particular:

As a general remark I suggest authors to follow the indication of the journal for text and in particular formulas format.

• For example, from row 126 to row 128 the position of symbols is as superscripts.
• Authors should explain better the concept expressed from row 129 to row 131.
• Row 140 - authors should pay attention to the format of the text, moreover in table 1 the RMSE seems to be dimensionless.
• From row 158 to 168 and equation (9) needs to be correctly formatted.
• Row 288 – The axis labels of Figure 1 need to be improved cause are difficult to be read.
• Row 399 as for Figure 1.
• From row 402 to 404 – The caption of the Figure 6 needs to be expanded and to be more explicit.
• Row 425 – Table 3 – Aspen densities in the table needs to be settled as exponents (Ex.: m^-2).
• Row 480 – The quality of Figure 10 can be improved.
• Rows 498 to 505 The points in the conclusions, in my opinion, can be integrated into the speech.

 

Author Response

Dear Reviewer,

   Thank you very much for your comments and suggestions.

   For more detail, please see the attachment.

Best regards

Xuerui

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I thank the authors for considering my comments. Unfortunately, I do not believe the paper should be accepted. The authors intend to “give a review of CYGNSS inversion technique and carry out corresponding simulations according to the existed problems during the CYGNSS soil moisture inversion.” The goal of this paper is lofty: to provide simulations on GNSS-R that can help future inversion techniques. In my previous review, I suggested that the authors make this paper either a review paper or a simulation paper. Understandably, this is now primarily a simulation paper. The simulations are not comprehensive enough to provide generalized insights for GNSS-R soil moisture retrieval. I believe that the authors should write a new paper that is solely dedicated to answering specific research questions with their modeling and simulation tools. The current simulations are too narrow in scope to provide generalized SM inversion suggestions under spaceborne conditions.

Author Response

Dear reviewer,

   Thank you very much for your valuable suggestions.

   The response for your suggestion and comments are given in the attachment.

   For more detail, please see the attachment.

Best regards,

Xuerui Wu

Author Response File: Author Response.docx