Landscape structures regulate the contrasting response of recession along rainfall amounts

Lee, Jun-Yi; Yang, Ci-Jian; Peng, Tsung-Ren; Lee, Tsung-Yu; Huang, Jr-Chuan

doi:https://doi.org/10.5194/hess-27-4279-2023

Articles | Volume 27, issue 23

https://doi.org/10.5194/hess-27-4279-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/hess-27-4279-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 27, issue 23

Research article

|

06 Dec 2023

Research article |

| 06 Dec 2023

Landscape structures regulate the contrasting response of recession along rainfall amounts

Jun-Yi Lee, Ci-Jian Yang, Tsung-Ren Peng, Tsung-Yu Lee, and Jr-Chuan Huang

Download

Final revised paper (published on 06 Dec 2023)
Supplement to the final revised paper
Preprint (discussion started on 27 Apr 2022)
Supplement to the preprint

Interactive discussion

Status: closed

RC1:
'Comment on hess-2022-88', Anonymous Referee #1, 04 May 2022

The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2022-88/hess-2022-88-RC1-supplement.pdf

Citation: https://doi.org/10.5194/hess-2022-88-RC1
- AC1: 'Reply on RC1', Jr-Chuan Huang, 26 Oct 2022
  
  The reply was uploaded in the form of a supplement.
  
  Citation: https://doi.org/10.5194/hess-2022-88-AC1
RC2:
'Review of hess-2022-88', Anonymous Referee #2, 16 Aug 2022
This overall well written paper intends to relate the classical a and b recession parameters to stream network, rainfall and antecedent moisture conditions. As discussed in the review by Anonymous Referee #1, there is a methodological problem: the presented analysis investigates the relationship between the marginal distributions of the parameters and possible explanatory variables, i.e. the analysis omits that a and be are not independent; a solution would be to first model the relation between a and b but as far as I see from fig. 5, there is no evident relationship between a and b.

Besides, it is unclear what the main contribution of the paper is beyond a state-of-the-art case study (which is probably no enough to justify publication in HESS). A clear presentation of what we could learn from a case study in the selected hydroclimatic area would be of key importance. The paper would also strongly benefit from a concise synthesis of known factors influencing recession properties and a better justified selection of the potential explanatory variables that are retained.

For all above reasons, I suggest rejecting the paper.

Detailed comments:

There is a lack of references for the theoretical aspects of how recession properties depend on landscape properties

There is no discussion of active drainage density (the actual drainage network can vary strongly seasonally)

the literature review should be improved; the previous findings are summarized but not yet synthesized; we also do not know where the previous work has been done (catchments size, climate, region etc); is this study the first in a tropical area?

When talking about travel times, it is important to be more specific wether this is in the channeled or the unchannelled state (i.e. in-stream or in the hillslopes), (e.g. Rinaldo et al., 2006)

1: attention some units are wrong, the same units should be on both sides of the equation

There are not enough details on how the explanatory variables of Table 1 are computed for the 260 events (what is total precip, what is Qtot (including or excluding baseflow?), how is peak flow identified if there are several peaks etc. etc.)

Rinaldo, A., Botter, G., Bertuzzo, E., Uccelli, A., Settin, T., and Marani, M.: Transport at basin scales: 1. Theoretical framework, Hydrology and Earth System Sciences, 10, 19-29, 2006.
Citation: https://doi.org/10.5194/hess-2022-88-RC2
- AC2: 'Reply on RC2', Jr-Chuan Huang, 26 Oct 2022
  
  The reply was uploaded in the form of a supplement.
  
  Citation: https://doi.org/10.5194/hess-2022-88-AC2

Peer review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (11 Nov 2022) by Theresa Blume

AR by Jr-Chuan Huang on behalf of the Authors (23 Dec 2022) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (18 Jan 2023) by Theresa Blume

RR by Anonymous Referee #1 (12 Feb 2023)

Suggestions for revision or reasons for rejection

Landscape structures regulate the contrasting response of recession along rainfall amounts

GENERAL COMMENTS

This a much-revised version of the paper I previously reviewed entitled “Landscape structure and rainstorms swing the response of recession nonlinearity”.

My main criticism of the original paper was of the interpretation of the causes of the variability in the recession parameter “a”, which I said was flawed due to variability of “a” being unit-dependent as a consequence of the dependence of “a” on the power-law exponent “b”. To address this issue, the authors applied the “decorrelation” method recommended by Dralle et al. (2017). I am skeptical that the decorrelation really solves the underlying problem such that causal mechanisms can confidently be attributed to variability in “a” when “b” varies. Biswal (2021) describes how the decorrelation does not completely dissociate “a” from “b” and concludes that the method of fixing “b” as a constant is preferable. I think the question that remains is, for the specific cases that the authors analyze, to what degree does the decorrelation method not solve the underlying issue but yet does improve the situation sufficiently that the interpretations that the authors make are generally valid, albeit with some uncertainty. This topic I believe is still an open research question, therefore I am open to having this paper be published. However, the authors should both acknowledge the arguments made by Biswal (2021) and acknowledge that there is methodological uncertainty in their analysis.

Overall, I am impressed with additional work that has been done. The introduction is much more thorough. The review summarized in Table 1 is, by itself, is a valuable contribution to the literature.

I think the English does not meet the standard of a HESS article. I sometimes guessed at the authors intended meaning. Improvement is needed throughout the entire paper, so I did not attempt to suggest edits. I am recommending major revisions primarily largely on the need to improve the English but also on some undeveloped arguments.

LINE-BY-LINE COMMENTS

45-47: The transition from 3 to 1.5 is not due to groundwater being “vertically sourced from different hydraulic properties”. The transition is due to the influence of the upstream boundary condition becoming a factor as the aquifer drains.

211 and 229: Should this be Table 2?

255: See also Roque et al. (2022) who discuss the role of contrasting shallow and deep geologic layers on the recession parameters.

301-302: I do not see why, necessarily, the “inverse relationship between H and ‘a’ confirms that the hydraulic parameters vary markedly with depth”. H is the vertical distance between the highest and lowest points in the basin, correct? It is a measure of the surface topography, not of the underlying geology. This statement requires further explanation.

325-326: I do not follow the explanation of why “a” would decrease with large storms. Why would drainage be slower from a major typhoon? What is meant by “overwhelm the effect of flow velocity?” Could not expect the opposite? During a heavy storm, there is more drainage from the upper layers, which are likely to be more conductive. There will also be more surface storage being drained, which I would think would be associated with a high “a”. Table 1 also show more studies found negative than positive relationships of “a” with measures of initial storage. How did those studies interpret the negative relationship? I think Section 4.3.1 requires additional explanation.

Table 1: It would be helpful to the see the authors’ study placed in this table, too.

Figure 9: It is not clear to me what the arrows are for in the lower three panels.

References

Biswal, B. (2021). Decorrelation is not dissociation: there is no means to entirely decouple the Brutsaert-Nieber parameters in streamflow recession analysis. Advances in Water Resources, 147, 103822. https://doi.org/10.1016/j.advwatres.2020.103822

Dralle, D., Karst, N., & Thompson, S. E. (2015). a, b careful: The challenge of scale invariance for comparative analyses in power law models of the streamflow recession. Geophysical Research Letters, 42(21), 9285-9293. https://doi.org/10.1002/2015GL066007

Roques, C., Rupp, D. E., De Dreuzy, J. R., Longuevergne, L., Jachens, E. R., Grant, G., Aquilna, L., Selker, J. S. (2022). Recession discharge from compartmentalized bedrock hillslopes. Hydrology and Earth System Sciences, 26(16), 4391-4405. https://doi.org/10.5194/hess-26-4391-2022

Hide

RR by Anonymous Referee #2 (21 Feb 2023)

ED: Reconsider after major revisions (further review by editor and referees) (08 Mar 2023) by Theresa Blume

AR by Jr-Chuan Huang on behalf of the Authors (18 Apr 2023) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (29 Apr 2023) by Theresa Blume

RR by Anonymous Referee #1 (14 May 2023)

Suggestions for revision or reasons for rejection

Landscape structures regulate the contrasting response of recession along rainfall amounts

GENERAL COMMENTS

The third version of this paper, while improved, still contains some unclear statements, and requires additional editing. I provided some editorial suggestions but did not attempt to be thorough.

I recommend this paper be returned to the authors for minor revisions.

LINE-BY-LINE COMMENTS

9-19: The abstract does not sufficiently summarize the study and the significance of the results. It should, for one, give the physical significance of the relationships between the recession parameters (a, b) and the various factors (e.g., L/G, catchment area). For example, it is not enough to only state that the value of b decreases as L/G increases. What is the significance of L/G? What does it represent with regards to what influences the flow of water through the catchment What does it imply in terms of subsurface flow that b decreases as L/G increases?

15-16: Flow-path length and gradient should be defined. Subsurface flow path length? River channel flow path length? Gradient of the river channel? Mean gradient of the catchment? Mean hillslope gradient?

17: Non-linearity of what? “Non-linearity” needs to be defined.

21-33: The English usage in the first two paragraphs is often poor and the information is presented in a manner that is difficult to follow. These paragraphs provide very important context for the authors’ study so I think more care should be given towards clearly relaying the information. Currently, it reads as a somewhat unorganized list of results from 4 – 5 studies. Separate paragraphs for effects of landscape structure and rainfall/antecedent storage conditions might help. Separate paragraphs discussing the recession coefficient and non-linearity might also help.

25: That “the recession coefficient, a, approximates the recession rate” is not a generally true statement.

27-33: Which of these relationships were determined from theory and which were shown empirically?

51: Not yet clear at this point what “point-cloud” means. Please define.

51-54: I don’t think it is correct to say that recession responses are dependent on “a” and “b”. “a” and “b” are a convenient way for us to characterize recession curves. Real-world behavior does not depend on these artificial values.

54-55: This statement appears to be very relevant to the authors’ study. The authors should elaborate: Why do Biswal and Nagesh Kumar conclude that “b” may respond to peak flow differently depending on the structure of the drainage networks?

80-83: It is not perfectly clear to me what exactly the flow-path length (L) is, which makes it harder for me to assess the physical significance of L/G. Is L the length of the hillslope, as in the distance from a drainage divide to a stream? How is it calculated for an entire catchment?

Is G = H/L? Is L/G, therefore, simply L/(H/L) = L^2/H? If so, what does L^2/H say about landscape structure that simply G (or 1/G = L/H) alone does not?

Given how much of the discussion on the results centers on L/G, it is important that its geomorphic/hydrologic/hydraulic significance be stated.

How does a large value of L/G imply a “short-and-gentle” hillslope (as stated on line 364)? I understand that the “gentle” part comes from a low value of G leading to a high value of L/G. However, a “short” hillslope would have a lower L, which leads to lower value of L/G.

107: Actually, “b” is the slope in a plot of log(-dQ/dt) vs log(Q).

179: “This contrasting response coincided with a difference in drainage area and was relatively consistent across all catchments”. I don’t fully understand this sentence. What exactly was consistent across all catchments?

192-193: Why make a distinction for catchments with drainage area larger than 800 km2?

200-201: The Methods section should describe how each of these variables was calculated.

206-207: What might cause the opposite response of “b” to antecedent flow and peak flow? This seems to be an inconsistent result if both factors are related to storage at the beginning of the recession event. This topic should be revisited in Section 4.3.2.

228: I would rephrase this to say the point-cloud estimates are distinctly different from the estimates from the individual recessions. Both may have systematic biases since we don’t know the “true” values.

228-230: I don’t think it is a given that a skewed distribution of flood peaks is the primary cause. Jachens et al. (2020) provide an example where the distribution of peaks is not skewed and the point-cloud value of b is still less than those of the individual events.

235: Sharma and Biswal (2022) is missing from the References section.

237-239: I don’t think the authors’ ranges of the recession coefficient can be compared directly to other studies because the units are not the same. Moreover, if the other studies did not apply the decorrelation method, the comparison may not be valid.

253: Roques et al. (2022) is missing from the References section.

281-284: I do not follow the authors’ logic here. A high value of H does not necessarily lead to slower drainage. On the contrary, a steep hillslope implies a stronger hydraulic gradient and faster drainage.

SUGGESTED EDITS

50: Replace “treatments” with “analysis methods” and delete the parenthetical phrase that follows.

51: Delete “centrality of recession”. It’s a strange term and may cause confusion.

60: Replace “period typhoon invasions” with “periodic typhoons”.

64-65: Check for consistency in verb tense through paper. For example, “document” (present tense) and “discussed” (past tense) are used in the same sentence.

109: “unit” should be “units”.

111: Replace “manipulate” with “affect”.

150: Replace “was” with “is”.

154: The material that is covered beginning with “Secondly,” should be its own paragraph, as it is a distinct issue from data resolution. The new paragraph could begin with “An important concern in recession parameter estimation is the dependence between…”

155: Replace “blurs” with “confounds”.

188: I would replace “presents a vague pattern…”, with something like “shows no clear connection to large-scale landscape features on the island”.

190: Replace “fluctuated” with “differed”.

195: By “irrelative of drainage area”, do the authors mean “not correlated with drainage area"? The latter phrase would be clearer.

263: Replace “vague” with “weak”.

319: Replace “prevalently” with “prevalent”.

346: Replace “hydrological” with “hydraulic”.

355-359: I suggest deleting the entire part of this paragraph beginning with “Despite the power-law…” This paper is not about the impact of methodological choices.

360: Replace “In our cases” with “In these catchments…”

360: Rearrange the sentence to read that the recession coefficient is moderately correlated to landscape structure while nonlinearity is only weakly correlated to landscape structure. Otherwise, it reads as if, for example, landscape structure is dependent on the recession coefficient. I would look for other places in the paper where the order should be reversed.

360, 361: It would be clearer to always write “recession coefficient” and not simply “coefficient”.

369: Delete “Further”. It is redundant to write both “further” and “also”.

369: Rewrite the first sentence. Rainfall amount effects the recession coefficient, it does not affect the “estimating” of it.

373: Replace “In sum” with “In summary”.

REFERENCES

Jachens, E. R., Rupp, D. E., Roques, C., & Selker, J. S. (2020). Recession analysis revisited: impacts of climate on parameter estimation. Hydrology and Earth System Sciences, 24(3), 1159-1170.

Hide

RR by Anonymous Referee #2 (26 May 2023)

ED: Publish subject to minor revisions (review by editor) (11 Jun 2023) by Theresa Blume

AR by Jr-Chuan Huang on behalf of the Authors (21 Jun 2023) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (27 Jul 2023) by Theresa Blume

AR by Jr-Chuan Huang on behalf of the Authors (16 Aug 2023) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (30 Sep 2023) by Theresa Blume

AR by Jr-Chuan Huang on behalf of the Authors (10 Oct 2023) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (20 Oct 2023) by Theresa Blume

AR by Jr-Chuan Huang on behalf of the Authors (24 Oct 2023) Author's response Manuscript

Short summary

Streamflow recession, shaped by landscape and rainfall, is not well understood. This study examines their combined impact using data from 19 mountainous rivers. Longer, gentler hillslopes promote flow and reduce nonlinearity, while larger catchments with more rainfall show increased landscape heterogeneity. In small catchments, the exponent decreases with rainfall, indicating less landscape and runoff variation. Further research is needed to validate these findings across diverse regions.