General comments:

Through isotope observation data and HYSPLIT model simulation, this study reveals the close relationship between isotope changes and different water vapor sources, and clearly shows the rapid movement of water vapor source area through d-excess, which provides a certain theoretical basis for in-depth analysis of water vapor source path and water vapor supply in extreme precipitation area. Although there are still some shortcomings in this paper, I suggest that this paper be revised and published on HESS.

Specific comments,

1）L.73 ‘above sea level’ can be abbreviated as ‘a.s.l.’.

2）L. 78-80 How to define extreme precipitation in this paper? What standards are used?

3）There are six extreme precipitation indexes (WMO). Which indexes are studied in this study? The reviewers believe that the relationship between precipitation events, precipitation intensity and stable isotopes should be analyzed respectively;

4）How long did the HYSPLIT backward trajectory mode simulation track?

5）The typical reason for the formation of cold wave in winter is the complex weather in 2018 / 2019. In addition to sufficient water vapor conditions and circulation field, atmospheric stability is also a necessary condition for the formation of extreme precipitation events. The analysis of atmospheric stability is lack in this paper

6）The variations in δ18Ov and dv values of water vapor have a good correspondence with wind direction, but what is the relationship with the vapor transport simulated by the HYSPLIT model?

7)  L.118-119 ‘(c) January 7−11, (d) February 16−22, and (e) January 27−31, 2019. ‘ should be ‘(c) January 7−11, (d) January 27−31, and (e) February 16−22, 2019.

Extreme precipitation events can lead to great disasters to society. It is very essential to study the mechanism of extreme precipitation events with multiple approaches. With the development of continuous measurements of water vapor isotopes with high temporal resolution, tracing moisture source and water vapor transport through water vapor isotopes becomes an important tool for studying precipitation process (especially the extreme precipitation events). However, the complex variation patterns of water vapor isotopes and the atmospheric processes behind extreme precipitation events are unclear due to the lacking of observation studies. This study successfully reveals the complex variation patterns of water vapor isotopes and their controlling factors during the winter extreme precipitation events in Nanjing, which is important for further understanding the complex variation patterns of water vapor isotopes and the underlying mechanism. However, some issues listed below need to be addressed to improve the paper.

Lines 44-59: In this paragraph, I suggest adding some case studies on precipitation process (especially the extreme events) through water vapor isotopes observations, aiming to demonstrate the effectiveness of the water vapor isotope approach for studying the extreme precipitation.

Lines 99-100: What’s the basis for choosing 8 days and 1500 m for the HYSPLIT simulations?

Lines 115-171: In the Results section, you combined the results of isotopic variations during precipitation events and the discussion of controlling factors for isotopic variations. It is difficult for readers to follow. I suggest separating the results and the discussion and adding a new subsection for discussing the influencing factors of isotopic variations.

Fig.2: Array the sub-figures in the same way as Figs. 3-5.

Line 101: Why the resolution of the CWT field is 0.5°×0.5°? But the resolution of the GDAS dataset you used is 1°×1°.

Line 111: What kind of reanalysis data do you use？ERA5 or ERA-Interim?

The authors presented an continuous observation of near surface vapor isotope (δ¹⁸O and d-excess) for 5 winter raining events in Nanjing, China. Although precipitation isotopes have been used to diagnose different moisture sources, the vapor isotopes are less observed and more rare for moisture source identification of winter precipitation. This manuscript presented the vapor isotopes abrupt shifts during the 5 raining events and related them to the moisture transportation and moisture shifts.

My concern about this work is that the measurement is the near surface vapor, not the free atmosphere vapor that formed the precipitation observed. In fact, the vapor isotope shifts are caused by the re-evaporation of precipitation, which is much lower than the normal vapor isotope without precipitation events. Even slight precipitation can produce the lower δ¹⁸O and d-excess in the observed near surface vapor. Therefore, to trace the vapor source, we usually remove the period apparently influenced by rainfall events. This have been found by previous publications in comparing the concurrent vapor and rainfall δ¹⁸O and d-excess.

I suggested the author to compare the vapor isotopes data with parallel precipitation isotopes and you can find the trick. I also suggest the authors to refer more references for similar observations.