A Priori Solar Radiation Pressure Model for BeiDou-3 MEO Satellites
Round 1
Reviewer 1 Report
This paper deals with a timely subject of the precise orbit determination of the emerging Chinese GNSS system, i.e., the BeiDou-3. Due to the lack of precise information about the optical properties of the new generation of the BeiDou satellites, the authors formulate the adjustment of the box-wing model for BeiDou-3 satellites. However, the approach proposed by the authors is virtually the same to the approach proposed by Montenbruck et al. (2015, doi: 10.1007/s00190-014-0774-0) for the Galileo satellites with only very small differences. Thus, the methodology from this paper is not new as it has been already discussed in detail by Montenbruck et al., (2015). The authors of this paper provide the results of the application of empirical orbit model parameters and perform their validation in several ways, consistently with the methodology described by Montenbruck et al. (2015). Although this study considers the box-wing model for the emerging GNSS system, this paper should be considered as a case study, rather than an original article.
The paper includes some major drawbacks which have to be carefully addressed. Thus, the paper may be recommended to for the publication after a major revision. Please find below more specific remarks which refer to particular sections of the paper.
Section 1. The study of the literature is well written, however, it lacks some of most recent articles which raise the subject of the precise orbit determination of GNSS using box-wing models, including new GNSS, such as Galileo. The optical and geometrical parameters released by GSA are validated by (Duan et al., 2018, doi:10.1016/j.asr.2018.11.007) and results of the precise orbit determination calculated employing empirical and analytical parameters (Bury et al., 2018, doi: 10.1007/s10291-019-0860-0). Moreover, when introducing MGEX and using MGEX data please cite the official paper: (Montenbruck et al., 2017, doi: 10.1016/j.asr.2017.01.011).
Line 52: “A priori cuboid box-wing model” or “the cuboid a priori box-wing model”
Line 74: Due to the fact…
Section 2.1. The introduced formula (1) presents the classical ECOM model which is barely used in this form (by most of the IGS/MGEX Analysis Centers, no once-per-rev terms in Y and D are calculated). I would suggest presenting the ECOM2 model, or the reduced ECOM1 model which considers terms D0, Y0, B0, B1S, B1C. Moreover, it would be nice to see the results of the precise orbit determination of the BDS-3 satellites using ECOM2 model as a comparison to the ABW model.
Section 2.3 In this section the authors describe the acceleration acting on the BDS-3 satellites as a function of the β angle. A significant difference between satellite C30 and the other spacecraft is visible. However, this part is poorly described, i.e., sentences describing the magnitudes of the accelerations are not sufficient. A short explanation of those differences should be provided, i.e., the explanation, whether the discrepancies result from the higher mass of C30 or the area of the BDS satellite elements.
Section 3. In this section, authors introduce formulae (2) and (3) which express the accelerations acting on the satellite solar panels and satellite bus, respectively. However, it is not clear why do we need separate equations for the solar panels and the bus. Moreover, please provide the value of the Solar constant (S) which is used in this study.
Table 3. Here, the authors could provide also information about the mass of the particular satellites.
Line 173: The authors provide here the SLR residuals, and overwhelmingly miserable description of it. Moreover, the SLR residuals analysis should be presented in the results section, not in the methodology. Moreover, when authors finally introduce the SLR residual analysis in Line 260, they do not explain the measurement principle of SLR, and the reason, why this technique is independent and reliable for the validation of the microwave orbit solution. Please consider extending the SLR part in results as well as in the literature review giving some references to existing bibliography on SLR tracking of GNSS for the orbit determination and orbit validation.
Equation 5. If the approach for the estimation of the ABW model parameters is similar to that presented by Montenbruck et al. (2015), why the equations describing abox, D (5) in this paper and (18) in Montenbruck et al. (2015) are not consistent? They are different by a constant factor of 2.
Line 191: Saying “correlations between most of ABW parameters, and the “guess” optical parameters..” is not precise. Please provide numerous values of the correlations and the correlated parameters.
Line 223: “of” is redundant
Table 5. How accurate are the estimated ABW parameters, could authors provide the formal error of the estimated values?
Section 4. The introduction to this section should be re-written. The reference to Table 5 should not be in this part.
Line 243: Rephrase this sentence.
Line 245: Change “deg” into the symbol
Line 237: Remove the caption from the text.
Line 260: As I mentioned before, add more information about the SLR validation.
Line 260: should be “. The one-way”.
Line 262: Change “eps” into a symbol
Table 7. Could the authors explain why the offset increases for C20 and C21 when applying the cuboid model?
Line 290: Change “eps” into a symbol
Section 5.
Lines 318-325: The authors mention here the estimation of the ABW parameters for the Galileo satellites. In my opinion, this part does not fit the conclusions. If the authors would like to present the results of the parameter estimation for Galileo it might be useful to check the consistency of their model by the comparison with the model proposed by Montenbruck et al. (2015) and place It in the result section. This could be another way of the validation of your model or at least the information, that the authors’ model is correct.
Line 330: “of” is redundant; change “beta” into a symbol
Line 332: change “be a priori” into “by a priori…”
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript discusses the estimation of the BeiDou-3 cuboid coefficients of box-wing model with an adjustable box-wing model. The cuboid model was then validated by orbit overlapping, clock overlapping as well as SLR residuals. Whereas this is an import and interesting topic, the paper contains several severe major deficiencies as listed below. Furthermore, the English of the manuscript is poor. Therefore, I recommend a thorough and major revision of the manuscript.
(P = Page, L = Line)
- For the POD processing, the satellite phase center offsets (PCOs) are of crucial importance. However, these values are not given (even not mentioned at all!) by the authors throughout the manuscript.
- The LRA offsets must be given when performing the SLR validation.
- The authors stated that “the main dimensions of these two satellites are more cuboid than the BeiDou-2 satellites”, but they used the optical properties and areas of BeiDou-2 for CAST BDS-3 satellites. This is scientifically not sound!
- The English of the manuscript is poor. Several basic grammatical errors appear in the current edition. As a result, I strongly recommend the authors to find a native speaker to help the writing.
- P1 L32-33
“Since, April 2019, there are 18” may be “As of April 2019, there are …”?
- P1 L34 “operation”
The expression “operation” is confusing. At the time of writing, GEO and IGSO satellites were not yet operational (=healthy).
- P2 L78-79
“The main dimensions of these two satellites are more cuboid than the BeiDou-2 satellites”. Proof your statement with proper references.
- P2 L85
“validated from” -> may be “validated by”
- P3 Table 1
The descriptions of “Attitude model” are contradictory for BDS-2 and BDS-3.
- P4 L110-111
“the amount of data of BeiDou-3 is smaller than these of BeiDou-2, especially for the C30-C35 satellite”
Do you have any explanation?
- P6 L169-170
“For the CAST’s satellites, the initial optical properties and areas were used the BeiDou-2 values for instead”
See above
- P7 L174-175
No details on the SLR analysis are given: reference frame, correction models, data editing, ...
- P10 Figure 5
Add the slope to the figure
- P10 L246
“and found” -> “and we found”
- P10 L249
The authors attributed the linear systematic variation to “the poorer accuracy of “guess” body-dimensions than that released values of SECM satellites”. However, the body-dimensions used in the paper are the same to the values released by SECM at ION2018.
- P10 L251
“24h orbit overlapping”
Please introduce/explain the concept of "24h orbit overlapping". It's not clear what you're doing here.
- P12 L274 “for the SECM” -> For
- P12 L275 “improved for” -> “was improved by”
P13 L290
“…satellite orbit, also had a…” -> “and also had”
L291-293: wrong sentence!
P13 Table 9
Why 48h clock overlapping while 24h orbit overlapping?
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The Authors were responsive to the most of the issues from the original review. I recommend a publication of the manuscript after clarifying some minor issues listed below.
Line 132 should be „especially for the C35 satellite”.
Table 7 should not be split into two pages. The same for Table 8.
Tables 6, 8, 9 and others: “Prn” should be “PRN” – as an abbreviation from “pseudo-random noise” number
Zajdel et al. (2017) reported that the standard deviations of SLR residuals to BeiDou-2 satellites between August 2017 and September 2017 are 21, 33, 19, 54 mm for BeiDou-408, 410, 412, 417, respectively (Table A2 from: https://www.mdpi.com/2072-4292/9/10/1049 ). Thus, STD was just 19 mm for BeiDou MEO and higher for BeiDou IGSO. The Authors used the orbits based on ECOM2 models from the Center for Orbit Determination in Europe.
In this manuscript (in Table 8) the SLR residuals are much bigger. Can you explain what can be the reason for the differences in SLR residual standard devaiations and in slope values?
Author Response
Please see the attachment.Author Response File: Author Response.docx
