Root growth dynamics and allocation as a response to rapid and local changes in soil moisture

Ceolin, Samuele; Schymanski, Stanislaus J.; van Dusschoten, Dagmar; Koller, Robert; Klaus, Julian

doi:https://doi.org/10.5194/bg-22-691-2025

Articles | Volume 22, issue 3

https://doi.org/10.5194/bg-22-691-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/bg-22-691-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 22, issue 3

Research article

|

06 Feb 2025

Research article |

| 06 Feb 2025

Root growth dynamics and allocation as a response to rapid and local changes in soil moisture

Samuele Ceolin, Stanislaus J. Schymanski, Dagmar van Dusschoten, Robert Koller, and Julian Klaus

Download

Final revised paper (published on 06 Feb 2025)
Supplement to the final revised paper
Preprint (discussion started on 19 Sep 2024)
Supplement to the preprint

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2024-2557', Anonymous Referee #1, 21 Sep 2024

Review of ‘Root growth dynamics and allocation as a response to rapid and local changes in soil moisture’.

Overview
The authors use a novel experimental approach to describe root growth over small spatial and temporal scales. They find that roots ‘forage’ for changing water resources on almost daily timescales. This research adds to a growing body of research demonstrating rapid root foraging. Studies on aquaporins have shown changes in water uptake over timescales of hours. Tracer, rhizotron and root ingrowth studies have shown changes in days to weeks, but this study provides an independent technique for assessing root responses. The study is timely as there is a growing appreciation for the role of rapid root foraging and it’s implications for plant growth and coexistence.

General comments
Figure 3 seems to capture your results. I’m not sure figures 4 and 5 are necessary. Figures 4 and 5 are also a little busy and difficult to interpret – the legend could be clearer. I would recommend moving to appendices or extracting the critical data to show. The legend colors could be larger to make them easier to distinguish. The legend also says ‘growth rate of plants’ but I think you mean growth rate of plant roots. On a related note, I think the results could be presented more succinctly to provide a clearer story for the reader (e.g., .like Fig. 3). For example, it may be clearer to combine T1 and T2 into one ‘wetted’ treatment.

Specific comments
Figure legends are hard to interpret: change ‘(b) L2 of T1 plants’ to a more understandable sentence / title. I’m not sure about requirements in this journal, but I prefer some interpretation in the figures.
Figure 5 legend is mislabeled as Figure 4.
L258: ‘suggests’
L342: remove ‘on’
L350: responses in denser soils and in response to other nutrient patches would also be interesting.

Some relevant citation that should be considered.
Karlova, R., Boer, D., Hayes, S. and Testerink, C., 2021. Root plasticity under abiotic stress. Plant Physiology, 187(3), pp.1057-1070.
Schneider, H.M. and Lynch, J.P., 2020. Should root plasticity be a crop breeding target?. Frontiers in Plant Science, 11, p.546.
Kühnhammer, K., Kübert, A., Brüggemann, N., Deseano Diaz, P., van Dusschoten, D., Javaux, M., Merz, S., Vereecken, H., Dubbert, M. and Rothfuss, Y., 2020. Investigating the root plasticity response of Centaurea jacea to soil water availability changes from isotopic analysis. New Phytologist, 226(1), pp.98-110.
Beyer, M., Koeniger, P., Gaj, M., Hamutoko, J.T., Wanke, H. and Himmelsbach, T., 2016. A deuterium-based labeling technique for the investigation of rooting depths, water uptake dynamics and unsaturated zone water transport in semiarid environments. Journal of Hydrology, 533, pp.627-643.
Kulmatiski, A., 2024. Water matching: an explanation for plant growth and coexistence in water-limited systems. Discover Soil, 1(1), p.2.

Citation: https://doi.org/10.5194/egusphere-2024-2557-RC1
- CC1:
  'Reply on RC1', Samuele Ceolin, 25 Oct 2024
  
  Dear Reviewer 1,
  Thank you so much for your helpful and constructive review. We really appreciate the recognition of the timeliness of our research and the acknowledgment of the added value that the study brings to the understanding of root foraging strategies.
  We will take into account the specific comments provided and we will definitely consider and include the suggested literature in the manuscript. As regards the general comments, we would like to clarify some points brought up and to provide rationale and context for the choices we made in the manuscript.
  - “Figure 3 seems to capture your results. I’m not sure figures 4 and 5 are necessary. Figures 4 and 5 are also a little busy and difficult to interpret – the legend could be clearer. I would recommend moving to appendices or extracting the critical data to show”. It is indeed much better to move Fig. 5 to the SI section. This figure only provides a different perspective of what is already shown in Fig. 4, so its presence is not crucial to support the points discussed later. However, we still believe Fig. 4 is key for answering Question 2 (vertical responsiveness) and to provide evidence needed to back up some important points discussed in section 4.2. The figure allows in fact to understand what happens in L1 and L2 separately in each treatment. Additionally, the figure illustrates how root growth rates varies with soil moisture. Such data visualization is needed to support our claim that water uptake increased faster than root growth, and that soil moisture in a layer influences root growth in a different layer. Therefore, we propose to keep Fig. 4 in the main text, but in a modified version to make it easier to interpret and to better highlight its main messages and complementarity to Fig. 3: for example, this is what the main plot would look like for T1 in Fig. 4:
  And we would do the same for T2. We will also improve the clarity of the legend and make the legend colours larger. This would be the new caption of Fig. 4: Medians of scaled growth rates (top panel), volumetric water content (VWC %, middle panel) and water uptake rates (ml 24h−1, bottom panel) throughout the phases in L1 and L2 in plants belonging to the Treatment 1 group (left) and the Treatment 2 group (right). Each median was calculated from a range of 5-20 data points per day for L1 and 2-5 data points per day for L2. Note that during the pulse VWC was increased to 15% and that the values of VWC in the plots are lower because the measurements were taken right before the watering. Only single data points of scaled growth rate were available on Day 10 and Day 12 in L2 of T1 plants (brown line on top left panel) and on Day 1, Day 10 and Day 16 in L2 of T2 plants (brown line on top right panel). These values were excluded from the plot.
  - “I think the results could be presented more succinctly to provide a clearer story for the reader (e.g., .like Fig. 3). For example, it may be clearer to combine T1 and T2 into one ‘wetted’ treatment”. We will work on improving the clarity of the results presented, for instance by clarifying the importance of differentiating between the growth rates in L1 and L2 in T1 and T2. Combining T1 and T2 works for Fig. 3 (as the goal of that figure was to show the responsiveness to the water pulse in general), but in order to answer Question 2 (vertical responsiveness) combining T1 and T2 into one "wetted" treatment would not help, as it would not allow to compare the growth rates in different layers. In order to see whether roots at different depths responded similarly when subjected to local changes in soil moisture, we need to look at each layer and at the effects of each treatment separately, because T1 and T2 differed by the order of the pulses application in the two layers (see Fig. 1). Reporting the growth rates separately for each layer and for each treatment also allows to reveal other important insights and make some deductions, for example the fact that the lower responsiveness to the water pulse in L2 might have been given by a less intense change in soil moisture between before and after the pulse. For this reason, we propose to keep a simplified version of Fig. 4 in the text, as described above, and explain more clearly in the text why it is important to consider T1 and T2 separately and how separate evaluations of growth rate evolution in L1 and L2 answers Question 2.
  We thank you again for taking the time to review our manuscript and for your constructive insights. Looking forward to any additional feedback.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2557-CC1
  - CC2: 'Reply on CC1', Samuele Ceolin, 25 Oct 2024
    
    I realized that the figure was accidentally uploaded twice, appearing side by side. Please disregard the duplicate and consider only a single instance of the figure for review.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2557-CC2
- AC2: 'Reply to RC1', Stan Schymanski, 13 Nov 2024
  
  Dear Referee,
  As the first author's responses were labelled as a "Community Comment", I will copy and paste the response below, in order to get the record straight. I apologise for the confusion and duplication.
  Dear Reviewer 1,
  Thank you so much for your helpful and constructive review. We really appreciate the recognition of the timeliness of our research and the acknowledgment of the added value that the study brings to the understanding of root foraging strategies.
  We will take into account the specific comments provided and we will definitely consider and include the suggested literature in the manuscript. As regards the general comments, we would like to clarify some points brought up and to provide rationale and context for the choices we made in the manuscript.
  - “Figure 3 seems to capture your results. I’m not sure figures 4 and 5 are necessary. Figures 4 and 5 are also a little busy and difficult to interpret – the legend could be clearer. I would recommend moving to appendices or extracting the critical data to show”. It is indeed much better to move Fig. 5 to the SI section. This figure only provides a different perspective of what is already shown in Fig. 4, so its presence is not crucial to support the points discussed later. However, we still believe Fig. 4 is key for answering Question 2 (vertical responsiveness) and to provide evidence needed to back up some important points discussed in section 4.2. The figure allows in fact to understand what happens in L1 and L2 separately in each treatment. Additionally, the figure illustrates how root growth rates varies with soil moisture. Such data visualization is needed to support our claim that water uptake increased faster than root growth, and that soil moisture in a layer influences root growth in a different layer. Therefore, we propose to keep Fig. 4 in the main text, but in a modified version to make it easier to interpret and to better highlight its main messages and complementarity to Fig. 3: for example, this is what the main plot would look like for T1 in Fig. 4:
  And we would do the same for T2. We will also improve the clarity of the legend and make the legend colours larger. This would be the new caption of Fig. 4: Medians of scaled growth rates (top panel), volumetric water content (VWC %, middle panel) and water uptake rates (ml 24h−1, bottom panel) throughout the phases in L1 and L2 in plants belonging to the Treatment 1 group (left) and the Treatment 2 group (right). Each median was calculated from a range of 5-20 data points per day for L1 and 2-5 data points per day for L2. Note that during the pulse VWC was increased to 15% and that the values of VWC in the plots are lower because the measurements were taken right before the watering. Only single data points of scaled growth rate were available on Day 10 and Day 12 in L2 of T1 plants (brown line on top left panel) and on Day 1, Day 10 and Day 16 in L2 of T2 plants (brown line on top right panel). These values were excluded from the plot.
  - “I think the results could be presented more succinctly to provide a clearer story for the reader (e.g., .like Fig. 3). For example, it may be clearer to combine T1 and T2 into one ‘wetted’ treatment”. We will work on improving the clarity of the results presented, for instance by clarifying the importance of differentiating between the growth rates in L1 and L2 in T1 and T2. Combining T1 and T2 works for Fig. 3 (as the goal of that figure was to show the responsiveness to the water pulse in general), but in order to answer Question 2 (vertical responsiveness) combining T1 and T2 into one "wetted" treatment would not help, as it would not allow to compare the growth rates in different layers. In order to see whether roots at different depths responded similarly when subjected to local changes in soil moisture, we need to look at each layer and at the effects of each treatment separately, because T1 and T2 differed by the order of the pulses application in the two layers (see Fig. 1). Reporting the growth rates separately for each layer and for each treatment also allows to reveal other important insights and make some deductions, for example the fact that the lower responsiveness to the water pulse in L2 might have been given by a less intense change in soil moisture between before and after the pulse. For this reason, we propose to keep a simplified version of Fig. 4 in the text, as described above, and explain more clearly in the text why it is important to consider T1 and T2 separately and how separate evaluations of growth rate evolution in L1 and L2 answers Question 2.
  We thank you again for taking the time to review our manuscript and for your constructive insights. Looking forward to any additional feedback.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2557-AC2
RC2:
'Comment on egusphere-2024-2557', Anonymous Referee #2, 24 Oct 2024
The authors of this manuscript utilized a novel method (MRI technology) to analyze the responses of vegetation root system to changes in water availability with a very high temporal resolution (48 hours).
The manuscript is concise and easy to follow, providing the clear description of main steps, results and main implications. It provided very interesting data and relevant implications for our understanding on the plasticity of the root system.
However, I have some concerns, which are described below.
My main concern is whether this manuscript falls within the scope of the journal (Biogeosciences). I assumed that Biogeosciences should provide meaningful information about the interactions between biological and geographical inspects. This paper mainly described biological research without strong implications for geographical things. I am probably wrong, so I would like to leave this concern to the editor and I am looking forward to reading the authors’ explanation about it.
My second concern is about the statistical description. The authors conveyed the idea mainly by Fig. 3 with the median values and percentiles in table 1. However, given that the number of measurements at each point is not large (n=5-29), how about using mean ± std to represent the general behavior? Do you think the mean value is more representative? Can you please also include the percentile or standard deviation in Fig. 3?
Here are some minor concerns:
Lines 31-34: The main idea is about “Hydromatching”. So do you think it is necessary to talk a lot about hydrotropism, hyrdopatterning and Xerobranching?

Line 75: I would explain to use the acronym more carefully. How about changing VMC to VSM? it would be difficult to link VWC to volumetric soil moisture. If you want to define for volume water content, please replace volumetric soil moisture with that.

Lines 90-91: Can you please provide a reference about the previously tested plants?

Line 124: As for the root length measurement, is it an output of software NMRooting? If not, can you please specify how to do that? It is very valuable for the future work.

4: Each panel has 3 y-axis lines, making it difficult to interpret. I would suggest to change the color of each y-axis line to be paired with the corresponding plots. You also can change the symbols to line with circles/dots/triangles/whatever. I do not think you need to provide all dates by x-axis. You probably can use a 2-day or 5-day interval to make the figure clearer. In addition, can you please merge 4 panels into an individual figure? It would help the readers to follow the caption.

5: The median values in each panel (white points) are too small to follow. Can you please make it larger or mark it as a line?

Line 258: It should be ‘suggests’.

Line 303: Vegetation growing/development models?

Line 305: Do you think it is possible to merge this section with the “Materials and methods” section?

Overall, this is a very nice contribution. If it overlaps with the scope of the journal, it deserves to be published. Thank the authors for their endeavor to carry out this research and draft the manuscript to disseminate it.
Citation: https://doi.org/10.5194/egusphere-2024-2557-RC2
- AC1:
  'Reply on RC2', Stan Schymanski, 05 Nov 2024
  Dear Referee 2,
  
  We are very thankful for your helpful and engaging feedback on our manuscript. We greatly appreciate your recognition of the strengths and relevance of our study. Below, we first address your main concerns and provide additional context to clarify our choices, before responding to your minor concerns one-by-one.
  
  Your point about the fit of our study into the scope of Biogeosciences is important and definitely merits discussion. We chose Biogeosciences because the journal guidelines state that Biogeosciences covers “all aspects of the interactions between the biological, chemical, and physical processes in terrestrial or extraterrestrial life with the geosphere, hydrosphere, and atmosphere”. They also state that the journal covers the field of plant-soil interactions. We believe that our study strongly belongs to the field of plant-soil interactions, as it involves interactions between biological processes (root development) and the geosphere and hydrosphere (local soil moisture availability dynamics) at a smaller scale (individual plant level). On top of that, root foraging abilities are known to be deeply involved in biogeochemical processes, biomineralization and microbial weathering (processes of interest mentioned in the journal guidelines). For these reasons, we believe this manuscript falls well within the scope of Biogeosciences.
  
  About your second concern, we used medians instead of means as our data showed skewness and a moderate presence of outliers. In this case, the median is more reliable to understand the central tendency of the data. Since adding the standard errors to median values requires bootstrapping and it is not as straightforward as for mean values, we calculated the 25^th and 75^th percentiles of the medians and put them in the tables to provide an idea of the median spread. We hope this explanation clarifies our rationale behind the decision of using median values and percentiles. We will update Fig. 3 and add the percentiles in the plot.
  
  Below, you find our responses to your minor concerns:
  
  Before introducing the term Hydromatching, we decided to describe previously documented processes of root morphological adjustments to soil moisture heterogeneity (Hydropatterning, Xerobranching), in order to provide context and differentiation. It is important to underline that, while the latter processes apply to the individual root scale, Hydromatching involves whole portions of root systems.
  
  We will change the instance “volumetric soil moisture” in line 75 into “volumetric water content”.
  
  We do not really have a reference for the previously tested plants. Those were tests we performed in our lab in order to optimize the timing of the official experiment. These only consisted in growing a group of plants to assess the height they would reach in a period of approximately 3 weeks, in order to organize the timing and produce plants that were grown enough (but not too much) by the time we started our experiment in FZJ. This will be clarified in the revised manuscript.
  
  NMRooting is able to reconstruct the root system in 3D according to the MRI signal, from which the Software can segment and isolate root tissues from the soil background. The reconstructed root system can then be converted into quantitative metrics such as root length: this operation is indeed a direct output of the Software. We will make this clearer in the manuscript.
  
  After R1’s suggestions, we decided to change Fig. 4 to make it easier to interpret and to better highlight its main messages and complementarity to Fig. 3 (see our response to R1’s review). We decided to have two panels in Fig. 4, one for T1 and one for T2. In each panel there would be three plots, one showing the growth rates in L1 and L2, one showing the VWC in L1 and L2 and one showing the water uptake in L1 and L2. This setup would allow to display the data more clearly and solve the problem of the multiple y-axes that you pointed out. This display would also address your suggestion to use different symbols for each line: given that there will only be two lines per plot in the new Fig. 4, a simple colour distinction between them will suffice to maintain clarity. The two panels will be merged into one individual figure, as you recommended. We will also use a larger interval between days on the x axis, as you suggested.
  
  Thank you, we will make the median point larger.
  
  Thanks, we will correct this.
  
  How about “dynamic vegetation models”?
  
  While we understand your suggestion of moving this section into the Materials and Methods (since many of the limitations indeed arose from the methods), we still believe this section should be placed after the discussion, as there we provide some directions on what to change in future studies to avoid the problems that we faced. Some of these problems were, for example, the lack of responsiveness in L2 and the presence of artifacts in L1 (which led us to use of two different root growth detection methods in L1 and L2). We believe that listing these limitations and potential improvements for future studies in the Materials and Methods would be premature and introduce some confusion. Discussing the limitations and potential improvements after the results have been discussed and interpreted is in our opinion more helpful, as it allows for a clearer understanding of the rationale for proposed changes to the methodology in future research.
  
  We thank you again for the time you dedicated and for the constructive feedback you provided while reviewing our manuscript. We believe that it will help further improve our manuscript and we hope that our responses satisfy the concerns you raised.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2557-AC1
- AC3: 'Reply to RC2', Stan Schymanski, 13 Nov 2024
  
  Publisher’s note: this comment is a copy of AC1 and its content was therefore removed on 15 November 2024.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2557-AC3

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to minor revisions (review by editor) (14 Nov 2024) by David McLagan

AR by Stan Schymanski on behalf of the Authors (28 Nov 2024) Author's response Author's tracked changes Manuscript

ED: Publish as is (02 Dec 2024) by David McLagan

AR by Stan Schymanski on behalf of the Authors (11 Dec 2024)

Download

Article (1469 KB)
Full-text XML

Short summary

We investigated if and how roots of maize plants respond to multiple abrupt changes in soil moisture. We measured root lengths using a magnetic resonance imaging technique and calculated changes in growth rates after applying water pulses. The root growth rates increased in wetted soil layers within 48 hours and decreased in non-wetted layers, indicating fast adaptation of the root systems to moisture changes. Our findings could improve irrigation management and vegetation models.