the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Mar 2022
Roots induce hydraulic redistribution to promote nutrient uptake and nutrient cycling in nutrient-rich but dry near-surface layers
Abstract. The rhizosphere is an exclusive passage for water and nutrients from the bulk soil to the whole plant. As such, its importance significantly outweighs the limited volume it represents. Numerous recent experimental and modeling studies have shown that plants invest considerable resources in modifying this region. However, roots must also balance the significant differences in resource availability in the vast soil volume they inhabit. Studies suggest that hydraulic redistribution by roots helps counterbalance the large differences in water status experienced by roots. In addition, experimental evidence suggests that hydraulic redistribution plays a role in mitigating drought effects and aiding nutrient uptake. However, whether hydraulic redistribution is a passive happy accident or a process controlled by plants remains unclear. Here, we present a novel mathematical model that integrates rhizosphere-scale modification of soil hydraulic properties by root exudation with long-distance interaction between roots that occupy disparately resourced soil regions. The model reproduces several known phenomena. First, hydraulic redistribution is proportional to the hydraulic gradient between wet and dry regions. Its magnitude substantially increases with the accumulation of hydrophilic rhizodeposits in the rhizosphere of the dry region. However, its effect on net water uptake by the whole root system is meager, negating the current hypotheses that hydraulic redistribution helps mitigate drought. Second, hydraulic redistribution facilitates nutrient uptake. We observed that periodic rewetting of nutrient-rich but dry soil layers significantly increases the active uptake of soluble nutrients. Moreover, cyclic rewetting of the rhizosphere increases the mineralization of the organic matter, thereby releasing nutrients locked in soil organic matter. The latter is another mechanism that supports the well-known phenomenon of priming of organic matter mineralization by root exudation. Overall, our model supports a hypothesis that roots faced with nutrient and organic matter accumulation in unfavorably dry soil regions facilitate hydraulic redistribution via exudation and benefit from the increased nutrient uptake and mineralization rate. These mechanisms could crucial role in determining whether plants can adapt to shifts in resource distributions under a changing climate.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(1882 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on bg-2022-52', Anonymous Referee #1, 22 Mar 2022
The manuscript by Yan and Ghezzehei presents a soil-plant water transport model including a two layer-based hydraulic redistribution (HR) mechanism. The model is used to study how a prescribed rhizosphere volume around the roots mediates HR effects, with a particularly strong effect on nutrient transport and uptake. The authors also show that the enhanced water content in the rhizosphere can promote decomposition of native organic matter. The topic is interesting and suitable for Biogeosciences, but the manuscript in my view is not always clear and would benefit from a thorough proof-reading. Most important, the proposed model now contains three components—rhizodeposits, organic matter, nutrients—that are connected conceptually but not mathematically. This makes hydrology and biogeochemistry in the model too un-coupled to support the conclusion that rhizodeposits are an adaptive response of plants to face dry conditions. My main comments are explained first, followed by other suggestions.
Main comments
- Nutrients are modelled independently of organic matter mineralization.
Nutrients are essentially regarded as a passive tracer, transported through the domain and taken up by plants. With higher hydraulic conductivity in the rhizosphere, nutrients are transported more rapidly according to the model, so the result that rhizosphere improves nutrient uptake is expected. Organic carbon decomposition rate is also enhanced thanks to the higher water content in the rhizosphere, which is also an expected result. In other words, both enhanced nutrient uptake and decomposition rates are a direct consequence of the model construction. This approach is simple and easy to understand, but has two drawbacks: i) nutrients are produced by decomposition of organic matter (now the two processes are independent) and ii) nutrients can be immobilized and re-mineralized by microbes in the rhizosphere (now only passively transported). Without this dynamic links between nutrient availability, organic matter and rhizosphere environmental conditions, it is difficult to say if nutrient availability actually improves thanks to the rhizosphere environment and HR.
- Rhizodeposits are not static and are costly for the plant.
One component of the model deals with organic matter decomposition, referring to “rhizospheric C mineralization” (L171)—yet, the rhizosphere is regarded as fixed in terms of both extent and hydraulic properties (driven by rhizodeposit mass of C content). However, one could argue that the rhizosphere function needs to be ‘maintained’ by secreting C as the rhizosphere compounds are mineralized. In the current model there is no connection between rhizosphere and organic matter decomposition, which might be a reasonable approximation given the short time scales of the investigation (individual dry-down period). However, by neglecting the dynamics of rhizosphere properties, it is not possible to estimate the costs for the plant, and thus it is difficult to conclude that rhizodeposits are adaptive and promote plant survival or growth. Is the extra nutrient uptake worth the cost of the rhizodeposits and their maintenance in the long term? Extending the model to study these long-term effects (even in a highly idealized way) is in my view necessary to “shed light on how plants can adapt to non-ideal resource distribution” (L50) and to support the conclusion that “the investment in rhizodeposits thus seems critical…” (L379). If these extensions are not feasible, the scope of the work should be adapted to avoid speculation on long-term adaptations.
- Short time frame.
Linked to the point above, the model is used to simulate a few days during a dry period, which is not enough to balance gains and costs of rhizodeposits. I wonder if at least longer simulations could be attempted to assess the behavior of the model when the soil nears the wilting point. At that point cavitation in the xylem might affect the results (see comment below). Also, longer simulations would allow studying how the deep soil water depletion affects HR occurrence.
Other comments
General: the manuscript needs proof-reading and a thorough check of sentence structure (see some examples below)
Notation: I find the notation for the soil layers and leaves counterintuitive—the deeper layer is numbered 1, then moving up we find layer 2, and finally the leaves identified by subscript 0. This numbering makes it hard to follow the text and understand the figures
L7: “happy accident” could be re-phrased in more scientific terms
L17: not clear what “roots faced with nutrient and organic matter accumulation” means
L19: missing word “could play”
L87: typo “Therefore”
L46: nitrogen is to some degree always present—do you mean that HR increased in nutrient amended plots?
L50: explain/change term “variable resourced”
Model concept: the model separates roots in two soil layers, defining them in terms of length or area per unit soil volume, but this does not imply that the model describes two long roots (Fig. 1 caption, and in L63); in fact, long roots would imply high hydraulic conductance, which has a series of consequences. I would reformulate in the caption and redraw the figure to avoid mis-interpretation; the model deals with many roots of different lengths, whose properties are defined on a per unit soil volume
L110: what are the rhizodeposits made of? Units indicate mass, but in terms of dry/wet weight, mass of C?
L118: not clear what “excluding a rhizodeposit free control” means; no rhizodeposit could mean X_0=0
L140: symbol R_b is not defined and not listed in Table A1
Eq. 15: the soil-to-leaf conductance depends on xylem water potential (cavitation curve); that is the essence of the model by Sperry et al. (1998) cited here, but this feature is not included. Given the short duration of the simulations it should not affect the results, but with longer dry-downs it could change the speed of water depletion. Small detail: I would use brackets in the order {[(…)]}
L150: why only active uptake? Passive uptake should be relatively easy to include in this framework
L151: citation should be formatted as Author (year)
L170: this section is on nutrient mineralization potential according to the title, but describes organic matter decomposition. Nutrient mineralization is related but not quite the same thing
L181: how can this model be deemed “effective in describing the role of rhizodeposition in increasing nutrient availability” since it does not describe nutrient mineralization, nor any reaction the nutrient might undergo?
L188: extra “s”?
L191: why faster water flow? If I understand correctly, water flow is driven by a water potential difference, not water content
L201: the label in Fig. 5 indicates average water potential, not water potential in zone 2
L216: is the nutrient uptake effect noticeable because of no-flow boundary conditions? Nutrients are not produced in this model, just transported and taken up by the plants
L245-246: multiple typos: space missing, “promotes”, “pronounced”
Fig. 9 caption: the model describes only active nutrient uptake, but the caption mentions passive uptake
L284: missing verb
L285: not clear link to previous sentence despite “Therefore”
L287: higher variability compared to what?
L292: longer simulations needed (see comment above)
L297: explain/change term “initialize”
L298: but if plants are so effective at promoting mineralization of native organic matter, its content will decrease through time, until it will not be able to provide nutrients; an equilibrium will be attained so the relative advantage might be temporary
L312: enhanced mineralization with respect to what control/reference state?
L326: define “good”
L332: previously it was argued that microbial activity was stimulated, not slowed down
L338: why only “often” and not always? are you referring to fertilized systems?
Table A1: check units of A, now a volume, but labelled as specific surface area
Citation: https://doi.org/10.5194/bg-2022-52-RC1 - AC1: 'Reply on RC1', Teamrat Ghezzehei, 15 Jun 2022
-
RC2: 'Comment on bg-2022-52', Anonymous Referee #2, 11 Apr 2022
In the paper entitled “Roots induce hydraulic redistribution to promote nutrient uptake and nutrient cycling in nutrient-rich but dry near-surface layers”, the authors present a mathematical model of soil hydraulics, aimed to investigate the effects of soil modification by rhizosphere deposition on water and nutrient uptake. I think the mathematical model and the analyses presented in this paper are sound and of scientific value, but I do have two main issues with the presentation of the work. First, I think the framing of the work is not consistent with the methodology, and should be reconsidered. Second, I think that the writing is not up to the standard I expect for publication, and that the paper requires a thorough proof read to improve the text.
The authors present the following knowledge gap in the abstract of the paper: “whether hydraulic redistribution is a passive happy accident or a process controlled by plants remains unclear”. In the introduction, the authors state: “The exact mechanism by which roots can induce HR is, however, not known. Here, we present a modelling study that demonstrates that alteration of rhizosphere soil by rhizodeposition facilitates HR”. I am of the opinion that this knowledge gap is not one that can be filled with the mathematical modelling approach presented in this paper. The fact that one can simulate HR by changing rhizosphere properties in a mathematical model does not provide evidence that real plants can do the same, and it will this not shed light on the mechanisms by which roots can induce HR. I am of the opinion that the paper should be reframed to focus on the results that show how changes in rhizosphere properties differentially affect water uptake, nutrient uptake, and nutrient mineralisation in different soils and soil water conditions. This can then be framed as a series of hypotheses that should be tested with experiments: namely that plants are able to modify soil properties in the way that was tested with the model through rhizosphere depositions, and that the exudation of these deposits is more common or more pronounced in soil types and soil water conditions where the model predicts the largest increase in the uptake of either water (under wet conditions and in sandy loams) or nutrients (under dry conditions and in sandy loam). However, I think that the authors currently overextend the impact of this work in sections 4.3 and 4.4. I think framing this work in the context of climate change is not appropriate in the paper’s current form, and a statement such as the one made in L. 363-365 (“Our results indicate that answering questions about plant adaptation to complex and changing soil and environmental conditions requires integrating biotic and abiotic feedback in the soil-plant-atmosphere continuum”) does not match up with the model presented in this paper. Either the modelling should be extended significantly to include more than just a hydraulic model, or the framing of this work should not go beyond the metrics simulated by the model (i.e. water and nutrient uptake). This reframing should lead to significant re-writing of the introduction and the discussion (sections 4.3 and 4.4 in particular).
I would suggest that the paper is carefully proofread to improve the text, which is currently full of grammatical flaws, especially in the introduction and discussion sections. I will mention a few examples I found in the first couple of paragraphs, but have not put the entire text to this level of scrutiny.
L. 27: I guess this is the explanation of hydraulic redistribution? Please cue the term here if that is the case.
L. 28: The current sentence structure reads as if roots modify their immediate surroundings and also modify hydraulic redistribution. Change the order of these two statements to correctly follow up on “…separate advances in our understanding of…’
L. 29: Replace ’-an’ with ‘, which is’, and follow this with a better definition of the rhizosphere.
L. 32: Rewrite by switching the order of plant water uptake and wetness of the rhizosphere.
L. 33: ‘the rhizosphere’s carbon investment’ suggests that it is the rhizosphere that is doing the investing.
L. 36: Change ‘peculiar’ to ‘specific’
Finally, I have a couple of minor comments.
In the current analysis, do the results of changes in nutrient uptake include the combined effect of increased nutrient uptake and increased nutrient mineralisation? If so, it would be insightful to decouple the two, presenting both their individual effects and their combined effect.
What is a rhizoshealth? (L. 302) I presume this should read rhizosheath, but I would still like an explanation of the term.
The reasoning behind some of the model’s design is explained in detail in the discussion, but I feel this should have been explained in the introduction: L. 305-308 – ‘Increased rhizodeposits … nutrient mineralisation.’
Citation: https://doi.org/10.5194/bg-2022-52-RC2 - AC2: 'Reply on RC2', Teamrat Ghezzehei, 15 Jun 2022
Interactive discussion
Status: closed
-
RC1: 'Comment on bg-2022-52', Anonymous Referee #1, 22 Mar 2022
The manuscript by Yan and Ghezzehei presents a soil-plant water transport model including a two layer-based hydraulic redistribution (HR) mechanism. The model is used to study how a prescribed rhizosphere volume around the roots mediates HR effects, with a particularly strong effect on nutrient transport and uptake. The authors also show that the enhanced water content in the rhizosphere can promote decomposition of native organic matter. The topic is interesting and suitable for Biogeosciences, but the manuscript in my view is not always clear and would benefit from a thorough proof-reading. Most important, the proposed model now contains three components—rhizodeposits, organic matter, nutrients—that are connected conceptually but not mathematically. This makes hydrology and biogeochemistry in the model too un-coupled to support the conclusion that rhizodeposits are an adaptive response of plants to face dry conditions. My main comments are explained first, followed by other suggestions.
Main comments
- Nutrients are modelled independently of organic matter mineralization.
Nutrients are essentially regarded as a passive tracer, transported through the domain and taken up by plants. With higher hydraulic conductivity in the rhizosphere, nutrients are transported more rapidly according to the model, so the result that rhizosphere improves nutrient uptake is expected. Organic carbon decomposition rate is also enhanced thanks to the higher water content in the rhizosphere, which is also an expected result. In other words, both enhanced nutrient uptake and decomposition rates are a direct consequence of the model construction. This approach is simple and easy to understand, but has two drawbacks: i) nutrients are produced by decomposition of organic matter (now the two processes are independent) and ii) nutrients can be immobilized and re-mineralized by microbes in the rhizosphere (now only passively transported). Without this dynamic links between nutrient availability, organic matter and rhizosphere environmental conditions, it is difficult to say if nutrient availability actually improves thanks to the rhizosphere environment and HR.
- Rhizodeposits are not static and are costly for the plant.
One component of the model deals with organic matter decomposition, referring to “rhizospheric C mineralization” (L171)—yet, the rhizosphere is regarded as fixed in terms of both extent and hydraulic properties (driven by rhizodeposit mass of C content). However, one could argue that the rhizosphere function needs to be ‘maintained’ by secreting C as the rhizosphere compounds are mineralized. In the current model there is no connection between rhizosphere and organic matter decomposition, which might be a reasonable approximation given the short time scales of the investigation (individual dry-down period). However, by neglecting the dynamics of rhizosphere properties, it is not possible to estimate the costs for the plant, and thus it is difficult to conclude that rhizodeposits are adaptive and promote plant survival or growth. Is the extra nutrient uptake worth the cost of the rhizodeposits and their maintenance in the long term? Extending the model to study these long-term effects (even in a highly idealized way) is in my view necessary to “shed light on how plants can adapt to non-ideal resource distribution” (L50) and to support the conclusion that “the investment in rhizodeposits thus seems critical…” (L379). If these extensions are not feasible, the scope of the work should be adapted to avoid speculation on long-term adaptations.
- Short time frame.
Linked to the point above, the model is used to simulate a few days during a dry period, which is not enough to balance gains and costs of rhizodeposits. I wonder if at least longer simulations could be attempted to assess the behavior of the model when the soil nears the wilting point. At that point cavitation in the xylem might affect the results (see comment below). Also, longer simulations would allow studying how the deep soil water depletion affects HR occurrence.
Other comments
General: the manuscript needs proof-reading and a thorough check of sentence structure (see some examples below)
Notation: I find the notation for the soil layers and leaves counterintuitive—the deeper layer is numbered 1, then moving up we find layer 2, and finally the leaves identified by subscript 0. This numbering makes it hard to follow the text and understand the figures
L7: “happy accident” could be re-phrased in more scientific terms
L17: not clear what “roots faced with nutrient and organic matter accumulation” means
L19: missing word “could play”
L87: typo “Therefore”
L46: nitrogen is to some degree always present—do you mean that HR increased in nutrient amended plots?
L50: explain/change term “variable resourced”
Model concept: the model separates roots in two soil layers, defining them in terms of length or area per unit soil volume, but this does not imply that the model describes two long roots (Fig. 1 caption, and in L63); in fact, long roots would imply high hydraulic conductance, which has a series of consequences. I would reformulate in the caption and redraw the figure to avoid mis-interpretation; the model deals with many roots of different lengths, whose properties are defined on a per unit soil volume
L110: what are the rhizodeposits made of? Units indicate mass, but in terms of dry/wet weight, mass of C?
L118: not clear what “excluding a rhizodeposit free control” means; no rhizodeposit could mean X_0=0
L140: symbol R_b is not defined and not listed in Table A1
Eq. 15: the soil-to-leaf conductance depends on xylem water potential (cavitation curve); that is the essence of the model by Sperry et al. (1998) cited here, but this feature is not included. Given the short duration of the simulations it should not affect the results, but with longer dry-downs it could change the speed of water depletion. Small detail: I would use brackets in the order {[(…)]}
L150: why only active uptake? Passive uptake should be relatively easy to include in this framework
L151: citation should be formatted as Author (year)
L170: this section is on nutrient mineralization potential according to the title, but describes organic matter decomposition. Nutrient mineralization is related but not quite the same thing
L181: how can this model be deemed “effective in describing the role of rhizodeposition in increasing nutrient availability” since it does not describe nutrient mineralization, nor any reaction the nutrient might undergo?
L188: extra “s”?
L191: why faster water flow? If I understand correctly, water flow is driven by a water potential difference, not water content
L201: the label in Fig. 5 indicates average water potential, not water potential in zone 2
L216: is the nutrient uptake effect noticeable because of no-flow boundary conditions? Nutrients are not produced in this model, just transported and taken up by the plants
L245-246: multiple typos: space missing, “promotes”, “pronounced”
Fig. 9 caption: the model describes only active nutrient uptake, but the caption mentions passive uptake
L284: missing verb
L285: not clear link to previous sentence despite “Therefore”
L287: higher variability compared to what?
L292: longer simulations needed (see comment above)
L297: explain/change term “initialize”
L298: but if plants are so effective at promoting mineralization of native organic matter, its content will decrease through time, until it will not be able to provide nutrients; an equilibrium will be attained so the relative advantage might be temporary
L312: enhanced mineralization with respect to what control/reference state?
L326: define “good”
L332: previously it was argued that microbial activity was stimulated, not slowed down
L338: why only “often” and not always? are you referring to fertilized systems?
Table A1: check units of A, now a volume, but labelled as specific surface area
Citation: https://doi.org/10.5194/bg-2022-52-RC1 - AC1: 'Reply on RC1', Teamrat Ghezzehei, 15 Jun 2022
-
RC2: 'Comment on bg-2022-52', Anonymous Referee #2, 11 Apr 2022
In the paper entitled “Roots induce hydraulic redistribution to promote nutrient uptake and nutrient cycling in nutrient-rich but dry near-surface layers”, the authors present a mathematical model of soil hydraulics, aimed to investigate the effects of soil modification by rhizosphere deposition on water and nutrient uptake. I think the mathematical model and the analyses presented in this paper are sound and of scientific value, but I do have two main issues with the presentation of the work. First, I think the framing of the work is not consistent with the methodology, and should be reconsidered. Second, I think that the writing is not up to the standard I expect for publication, and that the paper requires a thorough proof read to improve the text.
The authors present the following knowledge gap in the abstract of the paper: “whether hydraulic redistribution is a passive happy accident or a process controlled by plants remains unclear”. In the introduction, the authors state: “The exact mechanism by which roots can induce HR is, however, not known. Here, we present a modelling study that demonstrates that alteration of rhizosphere soil by rhizodeposition facilitates HR”. I am of the opinion that this knowledge gap is not one that can be filled with the mathematical modelling approach presented in this paper. The fact that one can simulate HR by changing rhizosphere properties in a mathematical model does not provide evidence that real plants can do the same, and it will this not shed light on the mechanisms by which roots can induce HR. I am of the opinion that the paper should be reframed to focus on the results that show how changes in rhizosphere properties differentially affect water uptake, nutrient uptake, and nutrient mineralisation in different soils and soil water conditions. This can then be framed as a series of hypotheses that should be tested with experiments: namely that plants are able to modify soil properties in the way that was tested with the model through rhizosphere depositions, and that the exudation of these deposits is more common or more pronounced in soil types and soil water conditions where the model predicts the largest increase in the uptake of either water (under wet conditions and in sandy loams) or nutrients (under dry conditions and in sandy loam). However, I think that the authors currently overextend the impact of this work in sections 4.3 and 4.4. I think framing this work in the context of climate change is not appropriate in the paper’s current form, and a statement such as the one made in L. 363-365 (“Our results indicate that answering questions about plant adaptation to complex and changing soil and environmental conditions requires integrating biotic and abiotic feedback in the soil-plant-atmosphere continuum”) does not match up with the model presented in this paper. Either the modelling should be extended significantly to include more than just a hydraulic model, or the framing of this work should not go beyond the metrics simulated by the model (i.e. water and nutrient uptake). This reframing should lead to significant re-writing of the introduction and the discussion (sections 4.3 and 4.4 in particular).
I would suggest that the paper is carefully proofread to improve the text, which is currently full of grammatical flaws, especially in the introduction and discussion sections. I will mention a few examples I found in the first couple of paragraphs, but have not put the entire text to this level of scrutiny.
L. 27: I guess this is the explanation of hydraulic redistribution? Please cue the term here if that is the case.
L. 28: The current sentence structure reads as if roots modify their immediate surroundings and also modify hydraulic redistribution. Change the order of these two statements to correctly follow up on “…separate advances in our understanding of…’
L. 29: Replace ’-an’ with ‘, which is’, and follow this with a better definition of the rhizosphere.
L. 32: Rewrite by switching the order of plant water uptake and wetness of the rhizosphere.
L. 33: ‘the rhizosphere’s carbon investment’ suggests that it is the rhizosphere that is doing the investing.
L. 36: Change ‘peculiar’ to ‘specific’
Finally, I have a couple of minor comments.
In the current analysis, do the results of changes in nutrient uptake include the combined effect of increased nutrient uptake and increased nutrient mineralisation? If so, it would be insightful to decouple the two, presenting both their individual effects and their combined effect.
What is a rhizoshealth? (L. 302) I presume this should read rhizosheath, but I would still like an explanation of the term.
The reasoning behind some of the model’s design is explained in detail in the discussion, but I feel this should have been explained in the introduction: L. 305-308 – ‘Increased rhizodeposits … nutrient mineralisation.’
Citation: https://doi.org/10.5194/bg-2022-52-RC2 - AC2: 'Reply on RC2', Teamrat Ghezzehei, 15 Jun 2022
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 897 | 289 | 63 | 1,249 | 53 | 47 |
- HTML: 897
- PDF: 289
- XML: 63
- Total: 1,249
- BibTeX: 53
- EndNote: 47
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1