the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Jun 2026
Revisiting Water Vapor Transport and Water Isotopes in the Dongting Lake Basin Across Two Centennial Extreme Events
Abstract. Water vapor source tracing of extreme weather events is essential for water vapor diagnosis, relating to accurate assessments of the water cycle and associated isotopic effects. Based on our prior climatic-scale study and using a consistent methodological framework, this study identifies the atmospheric circulation patterns and water vapor transport pathways for a once-in-a-century extreme rainstorm (June 2017) and freezing disaster (January 2008) events in the Dongting Lake Basin. Despite occurring in distinct seasons, both events exhibited remarkable similarities in circulation situation and water vapor transport: at low latitudes, a deepened South Branch Trough and westward-extended Western Pacific Subtropical High dominate warm, moist maritime air transported from low to high latitudes; at mid-to-high latitudes, a strong, stable Blocking High and East Asian Trough controlled the southward surge of cold air. The Dongting Lake Basin was situated within the saddle field where cold and warm air masses converge. Southwest transport contributed the largest water vapor share for both extreme events, setting historical records for their respective months. Constrained by relatively invariant atmospheric conditions, low-latitude oceanic vapor isotopes exhibited minor seasonal variation; however, because June precipitation significantly exceeded January amounts, June water vapor and resulting precipitation isotopes were substantially depleted relative to January values. Within frontal systems formed by strong cold-warm air interactions, temperature, humidity, wind direction, and speed varied considerably at different altitudes, as clearly reflected in water vapor transport differences at these representative vertical levels. Together, these two studies form a systematic understanding of climatic regularity to extreme-event behavior.
- Preprint
(3233 KB) - Metadata XML
-
Supplement
(1070 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Referee Comment', Luke He, 23 Aug 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-2903/egusphere-2026-2903-RC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2026-2903-RC1 -
RC2: 'Comment on egusphere-2026-2903', Alexandre Cauquoin, 04 Sep 2026
General comment
Using a δ18Op in daily precipitation record in Changsha for the period 01/2010 to 12/2022 as well as IsoGSM model results, ERA5 reanalyses and meteorological data, Xiao et al. investigate the water vapor transport during two extreme events in the Dongting Lake Basin: one extreme rainfall event in June 2017 and one freezing disaster in January 2008. A daily isotope record is presented, and several analyses and research methods were performed to analyze the changes in water vapor source related to these extreme events. In its topic, this paper fits the scope of the journal ACP. However, several points in the methodology and the data used are misleading, and more systematic analyses and figures of extreme events vs. climatological background are missing. Also, I am not sure to see the added value of water isotopes for better describing extreme events in the region in terms of water vapor transport and meteorological changes in the current version of this manuscript. Moreover, there are issues related to data transparency, the quality of the figures, and misleading terms, which need to be addressed. I hesitate between rejection and major revision considering I have doubts on the use of this δ18Op record for the extreme events topic. This record, as the authors said, is more adapted for seasonal and interannual variabilities, and large-scale climate processes.
Specific comments
- The authors present a rather long record but finally focus only on two short events. While the precipitation event in July 2017 is quite clear in figure 2, the January 2008 freezing event does not seem so prominent compared to the rather similar low temperature in January 2019. Moreover, there is clearly no impact on δ18Op (see next point). In almost all the figures, there is no context of these events related to the climatology. Are these events really exceptional? Is there any change in water vapor transport and precipitation patterns compared to the mean state? In the current version of the paper, this is not clear. Moreover, there is a confusion between the mean water vapor transport all along these events (sections 3.2 and 3.3) and the more dynamic way using HYSPLIT.
- The interest of using water isotopes and their temporal resolution is not clear in this study. First, the authors focus on 3-4 days corresponding to the maximum amplitude of the events, which means only 3-4 data points. I am not sure this is really sufficient to describe such event. Secondly, there is no clear impact on modeled and observed δ18Op (figures 2, 4, 6, 8) and no clear objective why we should use them. To check the δ18Op variations just to check their variations is not enough for such a paper. To get the water vapor source in this context, there is no need of water isotopes (HYSPLIT does the job) and there is, again, no clear change in stable isotope composition of precipitation. So, I am wondering is this record is really adapted for the extreme events topic.
- It is not clear which IsoGSM simulation was used by the authors. Are the outputs from the simulation from Bong et al. (2024), nudged to the u and v components of ERA5 winds at T62L28 spatial resolution? Or are they from another dataset? The reference https://datadryad.org/stash/dataset/doi:10.6078/D1MM6B is clearly not appropriate because it corresponds to monthly mean outputs until the year 2017! Such dataset needs to be cited appropriately and, if necessary (i.e., for temporary dataset), an agreement for the use of these data needs to be asked to the authors. Moreover, there is no information about how the modeled values have been extracted from the dataset (e.g., nearest grid point from the locations, spatial resolution…).
- The comparison of the observations to IsoGSM and ERA5 is also misleading. If one wants to explain the observed δ18Op values with IsoGSM δ18Op modeled values, one has to use also the modeled precipitation and temperature from IsoGSM, and not the ERA5 reanalyses. Please note the precipitation can be quite different between ERA5 and IsoGSM. Also, precipitation amount from ERA5 is mainly from models, contrary to the temperature. Moreover, extreme precipitation value for the July 2017 event is probably not modeled in IsoGSM as well, which as a clear impact on modeled δ18Op (so the analyses based on modeled values is biased).
- It is also not clear how the HYSPLIT model has been run. No experimental setup is described. Which input data have used? From which dataset (ERA5? Other?)?
- The term centennial is very misleading because it refers more to paleoclimates. Instead, use term such “once-in-a-century”.
Technical comments
- References to the water isotopes are written from the fourth line of the introduction (lines 32-33), without introducing the concept of water isotopes and in what way they can be useful for the study.
- Lines 79-96: the quoted terms are misleading (stable water isotopes are not only chemical fingerprints) and simplification is needed.
- Lines 102-103: just cite the paper.
- Lines 191-192: this is wrong. IsoGSM was nudged to u and v components of winds, and sea surface boundary conditions (sea surface temperature, sea ice area fraction) were provided. Air temperature and water vapor fields are not driving factors, they are simulated.
- Line 205: the reference to ERA-interim is not necessary.
- Line 207: the proper reference for ERA5 are Hersbach et al. (2020, https://doi.org/10.1002/qj.3803) and Soci et al. (2024, https://doi.org/10.1002/qj.4803).
- Lines 239-245: it’s a bit strange to talk about the extreme events before introducing them. They should be properly introduced in the data section.
- Figure 2: the comparison should be made with IsoGSM outputs. Highlight the 2 events and make some air between the curves. Why there is missing data in panel b? ERA5 and IsoGSM outputs are available for this time period.
- Line 296: The extreme rainstorm event occurring from June 22 to July 2, 2017 (referred hereafter as Event A).
- Line 307-309: remove this sentence.
- Figure 3: no unit for the precipitation, make the maps in colors.
- Lines 326-328: precipitation in ERA5 and IsoGSM are not the same. If one wants to explain the d18O signal, you have to use only IsoGSM outputs.
- Figures 4 and 6: use IsoGSM outputs. The figures are not readable, the red squares are not visible, no units, water vapor transport pathways are not explained or described (how you got them?). No comparison to the climatological mean.
- Lines 392-393: not clear, please rephrase.
- Figure 5 and 7: there is no clear explanation about the numbers in the x-axis. A reader cannot locate the serial number 15 on a map.
- Figure 8: legend is missing. Equations for the 3 colors should be shown and statistical significance of the difference between the equation should be made. To say just blue is a bit above and red a bit below the LMWL is not enough.
- Figure 9: again, no comparison to the mean state.
- Code/Data availability (line 705): see my comment about IsoGSM outputs.
- Code/Data availability (lines 705-707): this is not acceptable. If the paper is accepted, the data should be stored in an open access repository such as Zenodo.
Citation: https://doi.org/10.5194/egusphere-2026-2903-RC2 -
RC3: 'Comment on egusphere-2026-2903', Anonymous Referee #3, 11 Sep 2026
Review of "Revisiting Water Vapor Transport and Water Isotopes in the Dongting Lake Basin Across Two Centennial Extreme Events" by Xiao et al.This study investigates water vapour transport during two extreme events in the Dongting Lake Basin: an extreme precipitation and an extreme freezing event. By combining ERA5 and isoGSM data, the large-scale atmospheric forcing and the dominant moisture transport pathways are identified. Further, for one time step, radiosoundings and backward trajectories are used. The authors further describe observed and modelled stable water isotopes during the events.
While this study uses promising datasets to investigate the circulation and water vapour transport for two extreme weather events, many aspects of the study remain unclear, as listed in the following:1) Use of stable water isotopes: Observed and simulated isotopic composition of precipitation and water vapour are shown and described for the two extreme event, but no further investigation is done that makes any use of these data. For example, it is hypothesised that Rayleigh fractionation is important in shaping the simulated total column water vapour isotopic composition, but an application of a Rayleigh fractionation model is missing. Further, it is shown that the two events have distinctly different behaviour with respect to the local meteoric water line without further investigation or discussion. The additional value of using stable water isotopes in this study remains unclear.
2) Meteorological description of event: The analysis is lacking a thorough description of the meteorological evolution of the events. Even though data from ERA5, isoGSM and observations is available at a temporal resolution up to 6h, the description is based on 4-day means or one single time step. It is not clear why these perspectives have been chosen in describing the events. Further, a description of observed precipitation isotopic composition, temperature and precipitation amount during the events in missing (e.g. zoom-in of Fig.2). Further, the role of the frontal passage described in Section 4.2 based on the radiosounding is not clear. Is this a stationary front leading to continuous rainfall, or is moving over the observations sites leading to a peak in precipitation? Where is it positioned in the large-scale flow?
3) Combination of datasets: ERA5 and isoGSM data are used in combination to describe the events. But no comparison of the event between the two datasets is given: Are the events represented in a similar way in the two datasets? What is the effect of the (supposedly - not stated in the data description)) coarser isoGSM dataset on the evolution of the events? In the supplement, a comparison between observations and modelled data is given from a climatological view, missing a comparison of observed and modelled fields for the two events of this study.
4) Water transport pathways: Three major water vapour transport pathways were identified for each event "by identifying systematic water vapor currents in the Q field" (lines 237-238). The manuscript does not explain how this is done. Is this a subjective decision or based on objective criteria? By additionally using air parcel trajectories, these pathways could have been compared to an objective method, but (as described in point 5) the trajectory analysis itself has major weaknesses.
5) Trajectory analysis: The trajectory analysis is based on a too low number of trajectories. Three trajectories for one time step are not enough to describe 4-days events. As the authors correctly show, the air parcel movement can differ strongly between two grid points. Similar differences are possible between time steps and starting height. In its current form, the trajectory analysis is inadequate to describe air parcel movements for moisture transport during a 4-days event. Note, the reference Sodemann et al. (2008) used in the methods section on HYSPLIT does not fit.
6) Data availability: The isoGSM link leads to a monthly isoGSM dataset until 2017, which does not correspond to the used isoGSM dataset, where data until 2022 are available. This has to be corrected. Further, a statement that isotope observations are available upon reasonable request does not follow the ACP data policy (see https://www.atmospheric-chemistry-and-physics.net/policies/data_policy.html - "In rare cases where the data cannot be deposited publicly (e.g., because of commercial constraints), a detailed explanation of why this is the case is required. The data needed to replicate figures in a paper should in any case be publicly available, either in a public database (strongly recommended), or in a supplement to the paper").
7) Language: in several places, inadequate (non-scientific) language is used, e.g. "super-strong" (lines 31, 439,559) "abnormally deep" (line 338, 443), "abnormally strong" (line 340, 448)
8) Conclusions: The conclusions remain vague or do not give any new insight. E.g. "Water vapor is a necessary condition for precipitation production, but not a sufficient condition. The production of precipitation requires not only water vapor input but also dynamic conditions to lift the water vapor and thermal conditions for water vapor condensation" is well known and does not provide any new insight.
In its current form, I do not recommend this study for publication in ACP. There are major gaps describing the data and method, in the analysis of the events themselves and in the data availability. Further, it is not clear how this study contributes new insight in water vapour transport for extreme precipitation or freezing events.
Citation: https://doi.org/10.5194/egusphere-2026-2903-RC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 94 | 58 | 20 | 172 | 19 | 11 | 11 |
- HTML: 94
- PDF: 58
- XML: 20
- Total: 172
- Supplement: 19
- BibTeX: 11
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
