Research on UT1-UTC and LOD Prediction Algorithm Based on Denoised EAM Dataset

Round 1

Reviewer 1 Report

 

This paper promotes the UT1-UTC and LOD predictions based on the denoised EAM dataset, which is an improved method of LS+AR+EAM proposed by Dill et al (2019), the improvement is inspiring. However, let’s forget other details, the aurthors need to firstly explain the following doubt.

 

As the final results of the 2nd EOP PCC have been reported and the organizers are preparing a manuscript for UT1-UTC and LOD predictions, so it should not present result with contradiction. The UT1-UTC MAE results from the 2nd EOP PCC report are shown in this figure. As can be seen, the IERS bulletine A (ID 200 , the dash blue cuvre) performes the best for 1-30 days UT1-UTC predictions, this is totally different from the results shown in figure 7 in this paper.

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

During the 1st and 2nd EOP PCC campaign, the LS+AR method has been proved as one of the best method for EOP prediction. Another common sense for EOP prediction is the use of EAM dataset, especially for the short-term, 1-30 days, EOP predictions. For EOP predictions, in some sense , the quality (accuracy) of input dataset are more important than method. They author developed a method to use EOP series to suppress the noise of EAM dataset, and demonstrated that the de- noised EAM can effectively improve the accuracy， especially for the 1-6 days prediction, which is improved by ~20 %. They also demonstrated that modeling the 29.9 and 91.3 days signals in the GAM, and GAM-EAM dataset can improve the long-term (30-60 days) predictions, which means for long-term EOP prediction, may need different strategies.

Generally, the paper is a well organized and with clear conclusions, while, there is still some minors, that are presented bellow.

Major comments

On page 2, last paragraph of Introduction, line 95-107, better to combine (b) and (c) as one. Since all of them are methods to improve the EAM dataset. For organization of this paragraph, there is no need to give too much techniqual details in introduction parts. Thus line 91-95, line 96-107 should be re organized.

Minor comments:

(1) 1st page, line 17: axial components, should be component.

(2) 1^st page, line 35: UT1-UTC, better to use UT1.

(3) Page 8, line 251, better to give more details, of case 1-4 in caption of Figure 7.

(4) Path 10, line 304, better to give more details, of case 1-4 in caption of Figure 8.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

I have had the opportunity to review your paper titled "Research on UT1-UTC, LOD prediction algorithm based on denoised EAM dataset" and I appreciate the effort you have put into this research endeavor. Your study addresses a crucial topic related to the forecasting of UT1-UTC and LOD, and it presents an interesting approach involving the use of the Effective Angular Momentum (EAM) dataset. While your research shows promise and interesting, I believe there are areas where further critical analysis, improvement, and engagement with recent literature could enhance the quality and impact of your paper.
In conclusion, your research on UT1-UTC and LOD forecasting using EAM and AR modeling presents a promising avenue for future developments in the field. Addressing the critical questions, incorporating recent research, and discussing both advantages and limitations will significantly enhance the quality and relevance of your paper. I encourage you to continue refining your work and contributing to the advancement of Earth rotation and geodetic studies.

How do you envision that the Ray and Erofeeva tidal model could improve EOP predictions, and what specific tidal components does it address?

Could you provide a critical analysis comparing the performance of different tidal models, including the existing IERS tidal model, to support your claim of the need for an improved model?

While you've used the Mean Absolute Error (MAE) to evaluate your forecasts, have you considered other metrics commonly used in geodetic and Earth rotation studies, such as the Root Mean Square Error (RMSE) or the correlation coefficient?

You mention the necessity of estimating uncertainties in the EAM dataset. Could you elaborate on how uncertainties were considered in your study, and do you have plans for a more detailed uncertainty analysis in future work?

To strengthen your paper, consider including a comparative analysis with recent studies that have addressed UT1-UTC and LOD forecasting. Evaluate the advantages and disadvantages of your approach in comparison to these studies, highlighting where your methodology excels.

 

While you have discussed the advantages of your approach, it is equally important to address the limitations. A comprehensive discussion of potential pitfalls or scenarios where your idea may not perform optimally will enhance the paper's credibility.

Could you please confirm whether the figures (Figure 10 and Figure 11) display absolute prediction errors, and if so, consider explicitly mentioning this in the captions for clarity?

 

Can the authors provide an interpretation of the figures, particularly Figure 10 and Figure 11? Have you observed any discernible patterns or impacts on LOD and dUT1 related to other geophysical or meteorological parameters?

If there are indeed observed impacts, could you elaborate on the potential significance of these findings in the context of Earth rotation and EOP predictions?

This clarification and interpretation would help readers better understand the figures and the broader implications of the research in relation to geophysical and meteorological factors affecting LOD and dUT1.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

 

I think the explanation is reasonable and it need to be added to the manuscript in some appropriate place, to declare the different MAE statistical strategies compared with the 2nd EOP PCC, then the reader can remove the doubts.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

I would like to express my appreciation for addressing my previous comments and suggestions in your revised manuscript. Your responses have clarified many aspects of your work and improved the overall quality of the paper. I commend your dedication to enhancing the accuracy of Earth Orientation Parameters (EOP) predictions and your commitment to addressing reviewer feedback.

 

However, I have one additional question related to the data sources you used in your study. In your paper, you mentioned using the IERS C04 time series data, but I noticed that this dataset has a delay of 30 days. To ensure operational accuracy in EOP prediction, it is often necessary to include data from Bulletin A to fill in the gaps created by this delay.

 As an example, I refer to a study in Journal of Geodesy ( https://link.springer.com/article/10.1007/s00190-020-01354-y ) where the authors used only IERS C04 data. While their results were valuable for research purposes, it is important to consider that operational scenarios may require a more consistent dataset. Differences between Bulletin A and IERS C04, as well as increased formal errors, could potentially impact prediction accuracy in an operational setting.

 Could you please provide more details on the data sources you used for your study? Did you rely solely on IERS C04, or did you also incorporate Bulletin A data to address the 30-day delay? Understanding this aspect of your methodology will help clarify the applicability of your approach in operational scenarios.

 Thank you once again for your diligence and for considering this question. Your contributions to improving EOP predictions are highly commendable, and I look forward to your response.

Author Response

Please see the attachment.

Author Response File: Author Response.docx