the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Jan 2026
Retrieval of the precipitable water vapor from shipborne multi-GNSS measurements in tropical cyclone-prone regions of the Northwest Pacific during the summer season in 2021
Abstract. Global Navigation Satellite System (GNSS) is useful for monitoring atmospheric precipitable water vapor (PWV) content. GNSS observations performed in the ocean are relatively rare, making PWV observations at sea difficult to achieve. We previously retrieved shipborne GNSS PWV in the Northwest Pacific and conducted a comparative study with other observation systems. In this study, we demonstrate that reliable results can be obtained for the shipborne GNSS PWV over the ocean by using observations from similar regions at different times and comparing them with another dataset. To achieve this, we introduce the retrieval and validation of PWV from shipborne GNSS observations conducted aboard the research vessel from July 30 to August 25, 2021, in tropical cyclone-prone regions of the Northwest Pacific Ocean. The shipborne GNSS-derived PWV is validated against three reference datasets - radiosonde, low Earth orbit satellite (MetOp-IASI), and geostationary satellite (GK2A-AMI) - to assess its accuracy and reliability in oceanic environments. The GNSS PWV exhibits good agreement with radiosonde measurements, with a mean bias of −0.94 mm and a root mean square (RMS) of 4.28 mm. In addition, the two variables exhibit a correlation coefficient of 0.80. A comparison between GNSS-derived PWV and PWV obtained from GK2A-AMI observations reveals a minimal mean bias of 0.08 mm, which indicates good agreement. The RMS value between the two datasets is slightly greater than that observed with the radiosonde, reaching 4.83 mm. This comparison yields a correlation coefficient of 0.79. Furthermore, the PWV derived from MetOp-IASI shows a substantial bias of 2.58 mm and a relatively large RMS error of 6.99 mm compared with the GNSS PWV, with a correlation coefficient of 0.71. These results are consistent with previous findings. We suggest that the PWV derived from shipborne GNSS observations in tropical cyclone-prone regions of the Northwest Pacific provides stable and reliable results.
- Preprint
(1700 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 20 Feb 2026)
-
RC1: 'Comment on egusphere-2025-6016', Anonymous Referee #1, 19 Jan 2026
reply
-
AC1: 'Reply on RC1', Dong Hyo Sohn, 02 Feb 2026
reply
Dear Referee #1,
We really appreciate your precious time spent on carefully reviewing our manuscript.
We have revised the content of the document based on the comments.
Please see the attached file.
Thank you for your comments again.
-
AC1: 'Reply on RC1', Dong Hyo Sohn, 02 Feb 2026
reply
-
RC2: 'Comment on egusphere-2025-6016', Anonymous Referee #2, 30 Jan 2026
reply
Review : Retrieval of the precipitable water vapor from shipborne multi-GNSS measurements in tropical cyclone-prone regions of the Northwest Pacific during the summer season in 2021
General Comments:
This paper by Sohn et al. describes an experiment conducted in the summer of 2021, which aimed at monitoring atmospheric water vapor in the northwestern Pacific Ocean using a shipborne GNSS antenna. After discussing the importance of such measurements and outlining the methodology used to retrieve water vapor data, the authors evaluate the technique using satellite radiometric measurements. The paper concludes with an analysis of a notable cyclonic event.
Numerous recent studies (cited in this paper) focus on retrieving and validating integrated water vapor content measurements from shipborne antennas. The novelty of this work may lie in its use of two operational meteorological satellite radiometers, as well as its detailed description of a cyclonic event based on GNSS measurements.
The paper is well written, with a clear structure. While some of the presented work may not be entirely original, I believe that several points deserve publication (as mentioned previously).
However, I have a few concerns:
- The methodology used to retrieve GNSS PWV is not rigorous enough, and several critical steps are either overlooked or not detailed enough (e.g., specific models for GNSS analysis, parameters for ZWD to PWV conversion).
- Some statements are not always adequately justified or supported by results.
These points will be addressed in more detail in the "Specific comments" section.
Thus, I recommend a major revision of this manuscript.
Specific comments:
- Abstract: You should mention the discussion about tracking the cyclonic event.
- Line 34: QZSS is not a global satellite positioning system; it’s a regional one (https://qzss.go.jp/en/overview/services/sv02_why.html)
- Lines 39-40: The transition between this sentence and the next paragraph needs improvement.
- Line 64: please improve the transition.
- Lines 78-86: You should highlight the contributions and advancements of this article compared to your last study (Sohn et al., 2020).
- Line 125 / Equation 2: I don’t think this is the model you’re actually using: the equation models delay correction for a zenith angle z, including a curvature correction (ΔR), but it’s not meant for zenith delay correction alone. Please update the formula (e.g., with the one proposed by Davis, 1985). Also, lines 124-125 are ambiguous: Saastamoinen (1973) does not propose a model for horizontal gradients. Please clarify this.
- Table 1, Processing: What is the temporal resolution of the tropospheric parameters you estimate? Are they estimated independently for each epoch, or do you introduce temporal correlation?
- It seems to me that not all AERAT1675 antennas are modeled in the igs14.atx database. Which model do you consider? Is it an elevation-only calibration model (more suitable for an antenna that is not permanently oriented toward the north).
- You use GPT2 / GMF: is the ship displacement taken into account when applying these models?
- Equations (3) and (4): please specify how the parameters Ps and Tm, which are essential for an accurate retrieval of PWV, are determined.
- Figure 4: are you sure about the label “Frequency”? Should it not rather be “Count”?
- Line 199: in fact, for a significant contribution to PWV, water vapor is no longer present above roughly 10–15 km altitude.
- Section 4.1: maybe you could mention some studies comparing GNSS and RS41 retrieval? Were the comparison realized during the whole cruise or only the "experiment" area?
- Sections 4.1 / 4.2: would a synthetic figure presenting the PWV time series from the different techniques not be of interest? Possibly as a conclusion to Section 5? I leave this to your judgment. Section 4.2: satellite PWV is given with respect to the sea surface, whereas your antenna must be located between 10 and 20 m above the sea level; should the GNSS PWV therefore not be vertically corrected and referred back to the sea surface? An order of magnitude is that the PWV vertical variation is roughly dPWV = −4e-4 PWV dh: for dh = 20 m and PWV = 60 mm, this gives about 0.5 mm. Could you please clarify this point?
- Line 235: the link is no more valid
- Figure 6: The are 1225 data pairs. I believe that ZTDs are estimated every hour (from fig 4): that's a lot of matching points for a 27-day cruise, isn't it? (or I missed something about the matching procedure).
- Section 4.2: have you evaluated the spatial variability of the satellite PWV over the 0.5 × 0.5 degree areas? It would be interesting to confront this variability with the observed discrepancies: is this variability stable from one area or period to another? Is a larger variability associated with larger discrepancies?
- Figure 8: at the very end of the leg, there is also a very strong decrease in PWV, following a very high precipitation rate. Is this also related to a meteorological event, or is it linked to the northward movement of the ship only?
- Figure 9: could you add the position of the ship?
- Figures 10–11: could you add the corresponding day-of-year in the caption? Is it possible to locate the depressions mentioned / the typhoon on the maps?
- Line 336: please replace “cyan” with “magenta”.
- Line 344: I am not sure this is noise; it may rather be spatio-temporal variability.
- Line 355: please use “showed” instead of “demonstrated”.
- Line 357: here you state that you have demonstrated the reliability of GNSS measurements during the typhoon. You have shown a correlation between precipitation peaks and a decrease in PWV, and highlighted a good agreement over the whole period between GNSS, RS, and satellite measurements, but not specifically during the typhoon activity period. Could you nuance this statement? Or alternatively, could you complement it with time series of the differences, to be compared in parallel with the meteorological conditions?
- Conclusion: Ultimately, would the purpose of PWV retrieved from shipborne GNSS antennas not be to calibrate and/or validate satellite water vapor measurements?
Citation: https://doi.org/10.5194/egusphere-2025-6016-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 102 | 31 | 17 | 150 | 10 | 18 |
- HTML: 102
- PDF: 31
- XML: 17
- Total: 150
- BibTeX: 10
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript "Retrieval of the precipitable water vapor ..." submitted by Sohn et al. discusses GNSS-based water vapor estimates. While the paper is well written and the validations are reasonable, I have some major questions on the GNSS processing.
l43 Please explain the "relatively insufficient" claim for GNSS observations" made in the ocean"
l95 Please specify which AT1675 antenna was used.
l111 For PPP, you "introduce" orbit, clock, and ERP products. If estimating something, you need to explain this explicitly.
l114 Skip equation 1 and the associated text; this is commonly known
. l126 Skip the Saastamoinen equation; this is commonly known and can be referenced.
Table 1: You claim that GPS, GLONASS, Galileo, and BeiDou were used in PPP based on the IGS final products. This is not possible, first BeiDou is not contained in any final IGS products at the level of the analysis centers (except MGEX solutions) and the IGS final product refers to the combined product series which contain GPS (and GLONASS in a secondary product). Please clarify which products were used and provide proper citations for them. Please provide the week number for the used igs14.atx file (please explain which antex values were used for Galileo and BeiDou). Please check which components of the IERS Conventions were applied for your ship-based processing and update the Table. The receiver clock is usually estimated epoch-wise. Please provide the intervals for tropospheric delays. This is crucial and has to be stated prominently.
l148 Please consider a kinematic coordinate time series + a comparison against a tide gauge if possible to validate the PPP.
l153 The land-based GNSS station are far away, water vapor could be very different over 30km. I guess these two days showed very stable weather conditions.
Figure 4 What is the sampling rate of these PVW estimates?
Minor:
l25 please name the variables
l40 there are also other initiatives to derive best possible water vapor products (e.g., EGVAP)