the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Aug 2026
Impact of GNSS zenith total delay and tropospheric gradient Assimilation on severe weather: a case study of Super Typhoon Koinu
Abstract. In this study, we investigate the impact of assimilating Global Navigation Satellite System (GNSS) Zenith Total Delays (ZTDs) and tropospheric gradients (TGs) on the simulation of Super Typhoon Koinu (2023) using the Weather Research and Forecasting (WRF) model. Four data assimilation experiments are conducted: a control experiment assimilating conventional observations only, experiments separately assimilating ZTDs and tropospheric gradients on top of conventional observations, and a combined experiment assimilating both ZTDs and tropospheric gradients on top of the conventional observations. A two-month assimilation experiment using observations from approximately 250 GNSS stations demonstrates the overall benefit of GNSS data, with reductions in root-mean-square error (RMSE) of up to 60 % in ZTD when both ZTDs and tropospheric gradients are assimilated. Assimilation of tropospheric gradients alone also reduces the ZTD RMSE, demonstrating that gradients provide independent and complementary information beyond the vertically integrated moisture constraint of ZTDs. The impact of Typhoon Koinu is then examined as it approaches Taiwan between 3 and 6 October 2023. Continuous assimilation of GNSS observations improves the representation of the atmospheric moisture field surrounding the cyclone, leading to changes in the spatial distribution of integrated water vapor and the azimuthal moisture asymmetry around the storm. These moisture adjustments produce small but systematic improvements in the environmental deep-layer steering flow and, consequently, modest reductions in typhoon track error after landfall over Taiwan. The combined assimilation of ZTDs and tropospheric gradients also provides the most accurate simulation of cyclone intensity. The results demonstrate that tropospheric gradients complement ZTDs by improving the representation of horizontal moisture variability, thereby enhancing the simulation of the tropical cyclone environment. This study highlights the potential of assimilating multiple GNSS tropospheric products to improve tropical cyclone prediction in numerical weather prediction systems.
- Preprint
(2171 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-4285', Anonymous Referee #1, 22 Sep 2026
-
RC2: 'Comment on egusphere-2026-4285', Anonymous Referee #2, 07 Nov 2026
General comments
This manuscript investigates the impact of assimilating GNSS Zenith Total Delays (ZTDs) and tropospheric gradients (TGs) on the simulation of Super Typhoon Koinu using WRFDA 3DVAR. In addition to the typhoon case study, the authors perform a two-month cycling experiment with a large amount of GNSS stations. The study addresses a timely topic, namely the exploitation of tropospheric gradients as an additional source of atmospheric information beyond traditional ZTD observations. The subject is relevant in absolute terms, but also with a view to making tropospheric gradients available for numerical weather prediction models (which is not yet fully the case).
The manuscript is well written and clearly structured. The scientific motivation is sound, the methodology is described in sufficient detail, and the conclusions are largely consistent with the presented results. I find the demonstration that tropospheric gradients contribute information complementary to ZTD observations to be an important aspect of the paper. The two-month statistical experiment substantially strengthens the manuscript by showing that the observed effects are not limited to a single tropical cyclone event.
Overall, I consider the manuscript suitable for publication after minor revisions. However, I believe that some aspects of the interpretation and positioning of the results could be strengthened.Major comments
• Clarification of statistical significance
Throughout the manuscript, the authors frequently refer to "small but systematic improvements", "modest reductions" and similar expressions. While the reported impacts are consistent, it is often difficult to assess whether the differences between experiments are statistically significant. This issue is particularly relevant for steering flow errors (Section 4.5), track errors (Section 4.6) and intensity metrics (Section 4.6).
Although not strictly essential, I encourage the authors to discuss the statistical significance and robustness of these differences more explicitly. This could involve a brief analysis of uncertainty estimates, confidence intervals, or expected sampling variability.• Distinction between demonstrated results and inferred causal chain
One of the strengths of the manuscript is the effort to establish a physically coherent pathway linking TG assimilation to improved typhoon prediction: TG assimilation → moisture distribution → moisture asymmetry → steering flow → track/intensity improvements.
While this interpretation is plausible and generally supported by the diagnostics presented in Sections 4.3 to 4.6, I believe the manuscript could more clearly distinguish between what is directly demonstrated by the analyses, and what remains a physically motivated interpretation. This would be particularly helpful for non-specialists in these topics.• Positioning with respect to modern operational NWP systems
The manuscript convincingly demonstrates the added value of tropospheric gradients within the WRFDA framework used in this study. However, modern operational NWP systems utilize large volumes of satellite observations.
Although it falls outside the scope of this study, the discussion would benefit from more explicitly noting that the incremental contribution of tropospheric gradient observations in such observation-rich environments remains an open research question. This would not diminish the present results but rather help to situate them more effectively within the broader context of operational data assimilation.Minor comments
• Relative importance of the two-month experiment
In my opinion, the two-month cycling experiment provides some of the strongest evidence presented in the paper. The manuscript could place a bit more emphasis on this point, which confirms previously obtained results. Showing that assimilation of TGs alone reduces ZTD RMSE is particularly interesting because it strongly suggests that TGs provide information that is not merely redundant with ZTD observations.• Discussion of observation density
The typhoon case benefits from an unusually dense GNSS network over Taiwan. Although it would arguably be difficult to assess without conducting a test that removes a portion of the observations, the authors could briefly comment on the extent to which similar effects might be expected in regions with a lower density of GNSS stations.• Conclusions
The conclusions are generally well balanced and appropriately cautious. However, as previously suggested, the final paragraph could explicitly distinguish between the demonstrated ability of TGs to provide complementary atmospheric information, and their prospective operational value, which still requires assessment in a wider range of assimilation systems and meteorological contexts.The manuscript presents a technically sound and carefully executed study. The evidence supporting the complementary information content of tropospheric gradients is convincing and represents a valuable contribution to the GNSS data assimilation literature. The suggested revisions mainly concern clarification of significance, interpretation, and positioning of the results rather than any methodological issue
Citation: https://doi.org/10.5194/egusphere-2026-4285-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 204 | 125 | 88 | 417 | 102 | 88 |
- HTML: 204
- PDF: 125
- XML: 88
- Total: 417
- BibTeX: 102
- EndNote: 88
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Specific comments
Lin 20_ correct spaces “azimuthal moisture”
Lin31_ typo “crustal”
Lin 33_ What is “GNSS meteorology”
Lin 43: spaces “heavy rainfall prediction”
Lin 62_ explain a bit what is an observation operator
Fig1_ the left figure could have more clear the station points it they were in another colour, for example red
Lin141_ Why to place this “Future work will investigate the impact of GNSS observations using convection -permitting simulations.” here? : You are doing a description of the model you are using now. This sentence would be placed elsewhere
Lin 145_ you say the simulation starts on 1 sep 2023, until when? How long forecast? And when the typhoon occurred exactly? Please explain this a bit more
Lin 168_ but what was the sigmao value used in this study?
FIG3 _ I find the legend too long. The explanations should be given in the txt , not there.
Lin 228_ “mo ving “ typo
Lin 228 and 230_ Specify more the dates, and the phases of the cyclone
Lin 261_ Could a real radar reflectivity image be provided so to compared with the simulated ones? And this for the different experiments?
Anyhow, which of the runs is this figure 4a from?
Lin267- A more clear explanation of how the statistics were calculated may be needed. Because if what you do is compare O-B of ZTD of CTRL that (had not ZTD in the analysis) with the others that have it, it is logical that the RMS reduces due to you are approaching the analysis to this obs now. So a more detailed explanation should be needed here.
Lin_309_ Do this image and the simulated reflectivity one are plot over the same area so to make easy the comparison of the structures? It is hard to see what it is said in the text. So if not better to remove or change this comparison.
Lin 365_ Space in “spatial IWV”
Fig 4,5,6,7,8_again I find the legend too long. The part of the explanations should be given in the text , not there.
Lin 514; space at “environmental steering”
Lin 530_ space “complem entary”
General comments.
I have found many times the idea of “ZTD observations strongly constrain the column-integrated moisture field, whereas tropospheric gradients provide additional
information on its horizontal organization” in many of the sections, so some redundancy of this idea can be found along the text.
This work where 4 experiments have been run and compared is very useful to study the possible complementary information that the gradients give to the ZTD vertical profiles of humidity, and in this case in a severe case study that occurred in the tropics.
It would be interesting to see its impact in an operational environment from a model that apart of conventional observations, assimilates other satellite observations, because in that situation where ZTD by itself is beneficial too, as it has been proven and written in many papers, the addition of gradients and its positive impact could be demonstrated.