Parametric soil water retention models: a critical evaluation of expressions for the full moisture range

Madi, Raneem; de Rooij, Gerrit Huibert; Mielenz, Henrike; Mai, Juliane

doi:https://doi.org/10.5194/hess-22-1193-2018

Articles | Volume 22, issue 2

https://doi.org/10.5194/hess-22-1193-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/hess-22-1193-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 22, issue 2

Research article

|

12 Feb 2018

Research article |

| 12 Feb 2018

Parametric soil water retention models: a critical evaluation of expressions for the full moisture range

Raneem Madi, Gerrit Huibert de Rooij, Henrike Mielenz, and Juliane Mai

Download

Final revised paper (published on 12 Feb 2018)
Supplement to the final revised paper
Preprint (discussion started on 25 Apr 2016)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'comments on Madi et al Parameteric soil water retention models', Anonymous Referee #1, 25 May 2016
- AC1: 'Reply to referee 1', Gerrit H. de Rooij, 04 Aug 2016
RC2: 'hess-2016-168', Anonymous Referee #2, 28 Jul 2016
- AC2: 'Reply to referee 2', Gerrit H. de Rooij, 04 Aug 2016
RC3: 'review', Anonymous Referee #3, 01 Aug 2016
- AC3: 'Reply to referee 3', Gerrit H. de Rooij, 04 Aug 2016

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (08 Aug 2016) by Alberto Guadagnini

AR by Gerrit H. de Rooij on behalf of the Authors (01 Sep 2016) Author's response Manuscript

ED: Referee Nomination & Report Request started (02 Sep 2016) by Alberto Guadagnini

RR by Anonymous Referee #4 (01 Nov 2016)

RR by Anonymous Referee #5 (05 Nov 2016)

Suggestions for revision or reasons for rejection

Specific Comments

1. My impression, which I stress is only one person’s opinion, is that the manuscript is not very compelling. There are no new data, there are no comparisons with other parameter estimation techniques, and the idealized simulation does not provide clarity on which method best represents hydraulic properties across the full moisture range.
2. pg. 7. When introducing the “popular parameterizations,” include a table of the parameterizations that will be considered and evaluated. That would make this 20-page section easier to follow and set up the remaining text. The column headings might be “Parameterization” “Source” “Evaluation” and “ID.” Under evaluation, you might say “Not selected because …” or “Selected.” Under ID, you would list the labels BCO, etc., or na.
3. Pgs 7-26. Consider moving into an Appendix the parameterizations that are evaluated but not considered for the paper. You also might consider moving the entire section into an appendix and just summarizing the outcome in the main paper.
4. Pg. 7. The text defines h as the matric potential, notes that it is negative in unsaturated soils, and then says that “many of the cited papers adopt this notation for its reciprocal, the suction.” First, the suction is not the reciprocal, it is the negative. Second, state that this notation will be used for this paper.
5. Pg. 10. After Equation 11a, include a statement similar to “This equation is denoted VGA below.”
6. pg. 26. The soils are introduced in the order silt loam, silt, sand, and clay. For the reader’s benefit, keep the same order throughout the text and the figures and tables. For example, Figure 2 should be a) silt loam, b) silt, c) sand, and d) clay. Figure 3 should be a) silt loam, b) silt, and c) sand. Same goes for Tables 1 and 2.
7. Pg. 30. Vapor flow was allowed but not evaluated. State what effect, if any, it had on the results. In particular, the reader will want to know whether and to what degree it affected the comparisons.
8. Pg. 30. The bottom boundary is described as free drainage. I assumed that meant a seepage face, but the results in the Figures suggest it was a unit gradient condition. Please clarify. In either case, explain the impact of the boundary condition on the drainage predictions.
9. Figure 2 and elsewhere. The label pF is a unit of matric potential (or soil water suction) that represents the log of the potential (suction) value. Consider labeling axes as Matric Potential (-cm) or Soil Water Suction (cm).
10. Figure 3. Add the K(h) units somewhere.
11. Figures 4, 9, 10, and 12 have multiple panels that use different y-axes, making it difficult to compare across panels. Seriously consider standardizing the plot axes.
12. Table 1. Clarify whether the parameters in Table 1 are the result of fitting retention data only.
13. Include a Table of the three conductivity parameters for the three K models. This table will supplement Table 1. Combined, the two tables provided all the parameters used in the simulations.
14. The statement is made that the different parameterizations had a minor effect on evaporation (relative to the bottom flux) based on the variation among simulations. However, the differences in evaporation are just as great, or greater, than the bottom fluxes. Please reconcile this discrepancy.

Hide

ED: Reject (19 Nov 2016) by Alberto Guadagnini

ED: Reconsider after major revisions (further review by Editor and Referees) (26 Apr 2017) by Alberto Guadagnini

AR by Gerrit H. de Rooij on behalf of the Authors (18 Nov 2017) Author's response

ED: Referee Nomination & Report Request started (20 Nov 2017) by Alberto Guadagnini

RR by Anonymous Referee #4 (09 Dec 2017)

Suggestions for revision or reasons for rejection

From the story of this paper I understand that it was subjected to various revisions, but it is still not very convincing and overall it appears to have been organized more as a report than a scientific paper. For example, see the Introduction section and the sentences at L.407-421. My evaluation is that the present revision can be accepted after an additional moderate revision, but I do suggest it should be published as a technical note.

While the topic of this study is the “full” moisture range of the functions describing the soil hydraulic properties, it is clear that the authors devote more space to the wetter zone of the water retention curve and specifically to the air-entry value of this hydraulic property. Whereas more attention might perhaps be done to the 10-to-7 value of the suction head, I have one general comment about the identification of the wetter part of the retention curve. Specifically, I think that the suction table can be an appropriate equipment to get soil water content values close to saturation, but the evaporation experiment might be not. The main reason for that is the fact that the evaporation test starts from a hydrostatic condition in the soil core (if I understood well, with about zero suction head at the bottom), and therefore this test starts when the soil core is already under an average unsaturated condition nearly at, or already a bit far from, the air-entry value. Therefore, I have suspicion that in some cases the fitted parameters more related to the wet range might be poorly identified. The authors should discuss that point somehow.

P. 16, L. 479-483. The sentences at L. 479-483 are not completely clear to me. If the initial condition is that of -400 cm matric head at soil surface, then it is -300 cm at the lower end of the soil profile. This latter value of the matric head is around the “field capacity” condition of most medium-textured soils, when drainage fluxes are already relatively small. The authors should discuss that point.

P.1, Abstract; P.16, L. 480. The authors selected the “free drainage” as lower boundary condition for the numerical simulation runs. This choice has evident (and well-known) consequences when interpreting the relevant numerical results. Actually, in the abstract (L. 28-29) the authors state that “… cumulative bottom fluxes varied by up to an order of magnitude between them”, and consequently they warn the readers about the “need for a careful selection procedure for the parameterization of the soil hydraulic properties”. I would bring to the authors’ attention the fact that by imposing a “free drainage” lower boundary condition in the flow domain (i.e. the unit gradient of total hydraulic head), implies that the bottom water flux equals (apart the minus sign) the unsaturated hydraulic conductivity (K). Variable K is affected by relatively large errors and moreover it varies by several orders of magnitude from the wet to the dry range of matric pressure heads. This definitely explains the outcomes reported in this paper, and is also something one should expect.

Hide

RR by Antonio Coppola (18 Dec 2017)

Suggestions for revision or reasons for rejection

The revision of this manuscript has now a very long and troubled history and, in my opinion, was a bit too demanding for a paper whose target, limits and virtues were enough clear from the beginning of this history.
The “multistep” revision brought pro and cons to the manuscript, that I will try to shortly summarize below.
1. Yes, actually the paper still appears as a sort of hybrid between scientific and review paper, and yet in my opinion this becomes unavoidable anytime one attempts to generalize concepts. On the other hand, the title was already declaring its nature from the beginning;
2. I do not repeat all the points I raised in my previous review. However, the authors partly fulfilled them in this last version of the manuscript, at least in the premises and in the theoretical discussion of the parametric approaches available for analyzing soil water retention and conductivity data;
3. The paper became much denser (may be too denser), with a very full and comprehensive list of contributions to the topic.
4. Unfortunately, even if the authors largely discussed the issue of bimodality in the porous systems and its crucial role in describing hydraulic properties and predicting soil water (and solute) fluxes, they limited their analysis only to unimodal approaches. This was a limit of the previous manuscript version. Now, the paper appears a bit crippled, with the premises introducing an issue which is not actually dealt with in the results. In this sense, the title referring to the “full moisture range” remains a bit misleading;

In general, I keep some of my initial doubts on the paper, especially about its novelty, also because the skeleton of the paper did not change substantially, even if now the part limited to the unimodal approaches seems actually clearer than the first version.
In any case, I still do not consider these doubts so sizable to reject the paper, especially after all these efforts.

Hide

ED: Publish subject to minor revisions (review by editor) (28 Dec 2017) by Alberto Guadagnini

AR by Gerrit H. de Rooij on behalf of the Authors (04 Jan 2018) Author's response Manuscript

ED: Publish as is (08 Jan 2018) by Alberto Guadagnini

AR by Gerrit H. de Rooij on behalf of the Authors (08 Jan 2018)

Short summary

Water flows through soils with more difficulty when the soil is dried out. Scant rainfall in deserts may therefore result in a seemingly wet soil, but the water will often not penetrate deeply enough to replenish the groundwater. We compared the mathematical functions that describe how well different soils hold their water and found that only a few of them are realistic. The function one chooses to model the soil can have a large impact on the estimate of groundwater recharge.