Dear Amani Mahindawansha,
Thank you for providing responses to each comment made by the reviewers.
Please consider these discussed aspects and also address my minor points:
Why would the gravimetric water content be lower when using cryogenic extraction? I would expect that more water is extracted cryogenically than by oven drying.
It is still unclear where and when the ponding took place. It needs to be very clear when did you calculate evaporation from an unsaturated soil pore spaces and when the evaporation took place from open waters ponding on the fields. This would need to be accounted for when calculating evaporation fractionation as given by Gonfiantini (1986).
Looking forward to receiving the revised manuscript.
Best,
Matthias

Gonfiantini, R. 1986. Environmental isotopes in lake studies. In: P. Fritz and
J.C. Fontes, editors, Handbook of environmental isotope geochemistry.
Vol. B. The terrestrial environment. Elsevier, Amsterdam. p. 113–168.
doi:10.1016/B978-0-444-42225-5.50008-5

I thank Amani Mahindawansha and colleagues for their revisions under the difficult conditions of the situation at the University of Giessen.
I received two reviews of the revised manuscript of which one reviewer accepted the manuscript as is and another reviewer provided an extended list of feedback with a request for major revision.
I also thoroughly read the revised manuscript and have several aspects that need to be addressed before publication.
Please respond in your revision to each comment and provide examples of how the manuscript has been changed to address the feedback.

I do not think that the spatial information as shown in Figure 1 of the experimental set up is that important. One assumes that the proximity between different treatments does not affect the results. However, the temporal dynamics of the treatment are still not clear to me. For example, when was the field flooded? For how long were they flooded? Which samples were taken during flooding (you state in P4L10 that some samples were taken during flooding conditions). When did you assume n=1 and when n= 0.5 in the C-G model?
P2L14: General enrichment of soil water in heavy isotopes could also be a result of infiltration of rainwater that is enriched in heavy isotopes. The deviation of the soil water from the LMWL is an indicator (= kinetic fractionation). Yes, equilibrium fractionation might take place within the soil, were humidity is high, but the resulting effect cannot easily be distinguished from potential isotope changes due to rainwater infiltration. Please rephrase.
P2L16: Benettin is not the correct reference here. Please refer to older manuscripts (e.g. by Gat, Gonfiantini).
P3L6: Consider rephrasing to emphasize the focus on agricultural soils. Currently, for (II), it is held very general, but I suggest to rephrase (II) and (III) to include the focus on agricultural soils in both objectives.
P3L12: If these precipitation amounts are reflecting the specific study years, this needs to be stated. If they are long-term averages, this also would need to be clarified.
P3L14: Please put the study periods in perspective. Where they exceptionally dry/wet or representing long-term climate observations?
P4L18: What does “Pore water is not captured by cryogenic extraction” mean?
P4L20: How was GW sampled? At which depth is the water table?
P4L24: were
P5L7: Unclear why “This isotopic signal is then carried to deeper compartments via leaching. In deeper compartments, mixing of soil water with water transported through cracks may occur. The concept of multi-compartment transport indicates the history of the evaporation process as well as the depth and degree of isotope signal changes by the preferential flow.” is mentioned here.
P5L16: Most frequent? What does this mean? Do you mean the last event before sampling?
P5L19: (2018) twice
P6L1: Again Benetting is not the correct reference here. I believe that n=1 is giving for example in Gonfiantini or in Allison. Also, this indicates that for wet rice, there was always ponding. Please clarify when which n-value was used.
P6L2: How sensitive are the calculations to changes in the assumed n-value?
P6L13: I suggest to keep the expression of “surface ponded water”, as introduced in your methods section to prevent confusion in the results section, because “surface water” could as well be stream water for example.
P7L12: What does “pattern …. Increased” mean?
P7L13: Please add GW, irrigation and rainfall to the lc-excess plots.
P7L14: This sentence is an interpretation and not presenting results. Consider moving to discussion section.
P7L19: Precipitation and irrigation isotope ratios should be presented at the beginning of section 3.1
P7L20: This sentence seem to be not correct. “Frequent precipitation events” could also have relatively similar isotope ratios. Therefore, this sentence need to be rephrased.
P7L33: Unclear what “, and along with depth towards deep soils (0.20±0.1)” means.
P7L21: varied
P8L17/18: “separation”? You mean fractionation?
P8L20: Ploughing and puddling practices should have been introduced in the site description.
P8L26: It is not clear how they are transported downwards. You are mentioning the drying stage, but water movement during drying is minimal, since pressure differences in the soil will have been already equilibrated by then. Or do you mean the downward movement of the drying front?
P9L8: It would be very helpful to see the lc-excess of the ponding water and irrigation water in Figure 3 d,h,l,p to support the “single compartment” hypothesis. Also, in Figure 3d, the lc-excess does not support a single well-mixed compartment, since the lc-excess varies within the upper 20 cm.
P9L8: How do you derive the infiltration front. What is the depth of it that you are referring to here?
P9L11: If the conclusion is that “piston-like matrix flow” is dominant, what did you mean with “single compartment”? I guess this expression is unclear.
P9L15: The groundwater depth information should already be mentioned in the site description (i.e., it is important to know that the soil water sampling generally took place above the GW table.
P9L15: If capillary rise takes place, why is the bottom of your isotope depth profile so different from the groundwater isotope ratio? You refer to isotopically depleted GW, but your observations of GW are enriched compared to the soil.
P9L16: Not sure why hydraulic redistribution would result in a smoothing of the isotope depth profiles. If deep roots would transport water to shallow depths, one would expect a spike of deep water isotope ratio in the otherwise shallow water isotope ratio.
P9L31: Unclear why the irrigation water has an “evaporation imprint”.
P10L2: It is unclear why preferential flow would be a necessary process to reach to “gradual isotopic depletion towards deep soils”. Based on simulations assuming no preferential flow (see Fig. 9 in Sprenger et al., 2019, doi: 10.1002/2015RG000515), the mixing of fractionated soil water with non-fractionated rainwater would result in similar profiles as you show. Unclear how you can derive preferential flow from the observations.
P10L10: Here you explain the shape of the isotope depth profile with evaporation, while before you explain the shape with preferential flow.
P10L17: I suggest to not cite (Sprenger et al., 2016), but a classic paper from the first generation isotope hydrologists describing such basic concepts (e.g., Dansgaard).
P10L26: Unclear what is meant with “profiles with multiple compartments”
P11L13: leaf area
P11L16: Rephrase to clarify that you are talking about your own data and not Rothfuss anymore.
P11L18: reported for Asia? What kind of climate, soil and vegetation? 30% of F_E for Asia is not a very meaningful information.
P12L21: You mean aerobic conditions?
P12L27: errors related to F_E?
P12L30: For now, I am not convinced that your data provides sufficient info about preferential flow. Also, (III) is part of (II): preferential flow is part of “soil water movement”.
P13L2: What are “lower compartments”?
P13L4: How are these two sentences related? Why using “However”?
P13L7: Something seems to be missing here: not able to measure?
Figure 2: What is the x-axis scale? Please add the measurement frequency mm/h or mm/day? Is it correct that on a day that wet rice experienced irrigation there was no irrigation for dry rice or mize? Some seem to overlap, but
Figure 3: To my understanding, you measured gravimetric water content. Please use that instead of soil water content, which is usually used for volumetric water content. I also suggest to use the unit g g-1 in this context.
Figure 4: Description of the regression line is missing in the caption and/or legend.

Dear first author, Amani Mahindawansha, and corresponding author Lutz Breuer,
Thanks for sending the revised manuscript and I appreciate the responses provided to each of the comments.
I think that the manuscript can be published after clarification of the two following issues that came up.
1.) Now that I see can in Figure 3 that RW isotope ratios are more enriched than the soil water isotopes for Wet rice (WS) and GS2 and GS3 (Fig. 3 a&b) I do not understand how one gets the high FE values reported in Figure 5. In this case, dP will be bigger than dS in Equation (1) and that would result in a negative FE value. In general, it would not be expected that the fraction of evaporation loss is highest for the soil water samples that are least fractionated (Wet rice (WS)). Please look into your calculations and explain what is going on here – I seem to misunderstand something.
2.) Figure 2: I understand that in hydrological manuscripts, precipitation on bar plots often has an inverse axis. However, this is usually only necessary if there is also stream discharge or similar data additionally plotted in the graph. More importantly, the y-Axis for d2H is inverse, which is very confusing and should be changed. I am sorry that I did not realize in the previous round of reviews.
Best wishes,
Matthias Sprenger