Dear reviewer, thank you! We greatly appreciate your comments. Please find below our answers and what we intend to revise / add / remove in our upcoming revised manuscript version.

Answer to point 1:

Indeed, this is a very important point and for this experiment we took several precautions to make sure that soil water redistribution between layers did not occur during the time window of the experiment. Before the experiment, we waited for the sunflowers to dry out the soil sufficiently such that a) an injection would lead to a considerable change of the local soil water content (SWC) and b) that the injection of limited size would not be constrained too much by the soil being too dry, or inversely, too wet such that the pulse would diffuse over too much of a region. Based on experience and simulations using Hydrus-1D this would require that the prior SWC is above 0.05 and below 0.12 cm³ cm^-3 such that we can be assured that the upper and lower depths of pulse expansion are met. As the plant takes up the added water relatively quickly, this also reduces the risk of the injected water diffusing too far from the point of origin, something that cannot be replicated in a column without a plant. The second step we performed to assure that no water would cross between layers was to adjust the extend of the layer in such a way that we could not observe an increase of SWC at the boundary of two layers by observing the SWC profiles as they were measured by the SWaP (information is contained within the data series). Here, we also changed the boundary position of the layers by a cm up and down and found no significant changes, indicating that water did not exchange between layers. This can also be seen in supplemental Figure A1 where the boundary between the layers is indicated and where one can observe that the water of the pulses do not cross these boundaries over time. The lower boundary at about 35 cm depth is not strictly necessary for our analysis but ensures that the effect of the deeper pulse is not diluted by the water at the bottom of the column. The water in the lowest layer does redistribute within that layer as can be observed from Hydrus simulations, without the added pulses. For SWC below 0.12 cm³ cm^-3 redistributive flows are very slow and local and do not cross the second, lower boundary within the timeframe of the experiment. It should be noted too, that the matric potential gradients are in fact quite low, in the order of 10 hPa cm^-1 so redistributive flows are well contained.

--> We propose to add a Hydrus-1D simulation in an Appendix in our revised manuscript to support the present answer.

Anwer to point 2:

The reviewer addresses a valid issue, but the sunflower cultivar we used here is a very small (<50cm) growing cultivar, the “Dwarf yellow spray”, and under our conditions was no higher than 30 to 40 cm. Whereas for this experiment we do not have the leaf area, the typical leaf area of these sunflowers at this growth stage is in the order of 500 to 1000 cm².

--> We propose adding 1) the cultivar name in the text – this was indeed missing ! – and a picture of the plants of about a week before the experiments in a new Appendix, in which this small sunflower size can be observed. The calculation of the reviewer is perfectly in line with our measurements, which is also expected from the published performance of the SWaP.

Answer to point 3:

First concern:

As the reviewer points out, Eq. (4) is basically a simple volume-weighed mixing model to calculate δ_soil at time step (t) in one layer from the value at time step (t-1), namely δ_ant and that of the added, labelled water (δ_add). As the latter is known, Eq. (4) only needs to be initialized, which we explain in lines 203-208 of the original manuscript:

“Finally, values for the initial soil water isotopic composition, i.e., prior to the first addition of isotopically enriched water in the upper soil layer on DoE 1 16:00-17:00, was determined – since it was not measured in situ during the experiment – and was assumed to be the same for both soil columns (plant #1 and plant #2). More precisely, it was computed to fall onto an evaporation line with a slope of 4‰ /‰ (Rothfuss et al., 2015) and passing through the isotopic composition value of local tap water (δ²H=–50.4‰ and δ¹⁸O=–7.4‰) in a dual isotope (δ²H vs. δ¹⁸O) plot.”

The obtained soil water δ²H (δ¹⁸O) initial values in the top, middle, and bottom layers of both soil columns (in which plant #1 and #2 are grown) were -47.7, -50.4, and -50.4 ‰ (-6.7, -7.4, and -7.4 ‰) and are reported in Appendix E.

--> In our revised manuscript version, we will significantly expand the explanations of our calculations and provide these numbers in the text.

Second concern:

Yes, that is correct, δ_soil is affected by water addition only, that is, remains constant within each layer until the next water addition. We acknowledged in our manuscript that this is a simplification (see line 403-404), yet we believe it to be valid at the layer scale, since we made sure that there was 1) no (convective) water transport across the defined layers (please refer to our answer to your previous comment) and 2) no or negligible diffusion of the labeling pulse across layers during the time window of the experiments. We also believe that there was no significant evaporation and therefore changes in δ_soil in the top layer during the experiments.

Taking destructive measurements for determination of δ_soil in such conditions (small size of the column) would have been very challenging without affecting the plants themselves. The best option would be to monitor non-destructively the changes in δsoil (e.g., following the method of, e.g., Rothfuss et al. 2013). This would require to significantly modify the setup. We hope to perform such a follow-up experiment soon.

--> We agree that this needs further justification and propose to incorporate the present answer to our discussion (section 4.2).

Refs:

Rothfuss, Y., Vereecken, H., and Brüggemann, N.: Monitoring water stable isotopic composition in soils using gas-permeable tubing and infrared laser absorption spectroscopy. Water Resour. Res., 49(6), 3747-3755, https://www.doi.org/10.1002/wrcr.20311, 2013.

Dear Reviewer, thank you very much for your comments! Please find below our answers and details on how we propose to address your comments in the revised version of our manuscript.

Answer to point 1:

Thank you for this very important point! Our intention was not to draw definitive conclusions on the physiology and water transit time of sunflower plants, rather to present a proof of concept and a first application of the simple isotope monitoring method. We do not think that our estimate of capacitance should be representative of the tested cultivar: we just highlight the importance of its capacitance. There is naturally no wish to overinterpret the differences between the two plants investigated - there are indeed more possibilities, including biology.

--> We propose to explicitly mention the low n in the discussion section (4.3).

Answer to point 2:

The calibration functions obtained in hydroponic solutions would not be transferable to the case of a soil-grown plant only if a sunflower behaves differently from an isotope standpoint, i.e., discriminates differently against heavier stable isotopes during uptake from a soil than from pure liquid water. We however do not think it is case.

--> We will therefore modify that section (currently L385-388) and formulate it in a positive manner, that is: we are confident that the calibration functions are species-specific, rather than influenced by the environment in which the plant is grown, therefore are transferable to a soil-grown plant.

On a side note, calibrating in soils poses other challenges, e.g., prevention from soil evaporation to conserve the δsoil value while growing the plant in aerobic conditions.

Answer to point 3:

There are currently no non-destructive alternatives to our method. This can only be done by sampling the stem destructively and by extracting its water in the laboratory using one of the available techniques, e.g., the cryogenic vacuum distillation, centrifugation, direct vapor equilibration (Ceperley et al., 2024). Also, the extent to which the drilling affected the plant was observed with MRI pictures and discussed in the text (please see Figure 5 and text L454-L457, Appendix Fa and b).

Answer to point 4:

Thank you. It is not entirely clear which section this comment refers to. In case it refers to Eq. (5) (replacement model), it is correct that we cannot proof our assumption, but this is not the purpose for providing this equation. It serves only as a rough estimate of how large the fraction of water is that exchanges with the stationary water within the stem. It serves as an indicator of order of magnitude only.

Also we do think that we tested the sensitivity of the replacement model to the exchange fraction parameter in L424-429 (section 4.3)

“If f_ex=1 is set in the simple replacement model (Eq. 5), that is, all water flowing in the xylem is isotopically exchanging with water from the (non-conducting) stem tissues, stem water would reach a constant δ_stem in roughly 2 h. On the other hand, if f_ex=1/6, it would take 10 h for stem water to be replaced nearly fully with the xylem water, which would be roughly in line with our observation for plant #1.”

Answer to point 5:

Thank you for this important point! Our method did not “fail”, rather:

“[f]ailure to [insert the tubing (i) only a couple of millimeters inside the stem and (ii) close to the root crown] resulted in invalid data during the experiments”.

The installation of the tubing inside the stem always worked and always yielded an isotope response. It could be observed quite fast (<1 min) from the raw readings if the tubing was inserted correctly (water vapor mixing ratio values equal to the saturated values at the temperature of the stem).

--> As for Point 2 (please see our answer above), we will phrase this positively in our revised manuscript to avoid any misunderstandings about the replicability of our method.

Answer to point 6:

This statement in L199 should be seen more like a standard, conservative phrasing, because we did verify that this is substantially correct. As commented to Reviewer 1, the layers were carefully selected. In fact, we tried several boundary positions over a range of a few cm from within the SWaP data. Up to 1 cm up or down showed no significant changes in our results, giving more than sufficient confidence that no relevant cross layer exchange occurred between the upper and middle soil layers.

--> We will rephrase and replace “assuming” with “after verifying” in our revised manuscript. In addition, we will add a Hydrus-1D simulation in an Appendix to support the present answer.

Technical corrections:

References are reported in accordance with the journal guidelines (https://www.hydrology-and-earth-system-sciences.net/submission.html#references). We did find two articles (Ceperley et al., 2024; Fabiani et al., 2022) which were missing their article numbers (no page ranges). Thank you. This will be amended in the revised version of the manuscript.

Refs:

Ceperley, N., Gimeno, T. E., Jacobs, S. R., Beyer, M., Dubbert, M., Fischer, B., Geris, J., Holko, L., Kuebert, A., Le Gall, S., Lehmann, M. M., Llorens, P., Millar, C., Penna, D., Prieto, I., Radolinski, J., Scandellari, F., Stockinger, M., Stumpp, C., Tetzlaff, D., van Meerveld, I., Werner, C., Yildiz, O., Zuecco, G., Barbeta, A., Orlowski, N., and Rothfuss, Y.: Toward a common methodological framework for the sampling, extraction, and isotopic analysis of water in the Critical Zone to study vegetation water use, Wiley Interdisciplinary Reviews-Water, 11(4), e1727, https://www.doi.org/10.1002/wat2.1727, 2024.

Fabiani, G., Penna, D., Barbeta, A., and Klaus, J.: Sapwood and heartwood are not isolated compartments: Consequences for isotope ecohydrology, Ecohydrology, 15(8), e2478, https://www.doi.org/10.1002/eco.2478, 2022.