Assessment of structural sediment connectivity within catchments: insights from graph theory

Cossart, Étienne; Fressard, Mathieu

doi:https://doi.org/10.5194/esurf-5-253-2017

Articles | Volume 5, issue 2

https://doi.org/10.5194/esurf-5-253-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/esurf-5-253-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 5, issue 2

Research article

|

10 May 2017

Research article |

| 10 May 2017

Assessment of structural sediment connectivity within catchments: insights from graph theory

Étienne Cossart and Mathieu Fressard

Download

Final revised paper (published on 10 May 2017)
Preprint (discussion started on 08 Nov 2016)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

SC1: 'units of measurements of IC', Marco Cavalli, 10 Nov 2016
- AC1: 'Connectivity index - dimensional analysis', Etienne Cossart, 18 Nov 2016
SC2: 'Re: Linking fuinctional and structural connectivity and methodological clarifications', Rafael J. P. Schmitt, 15 Nov 2016
- AC2: 'Response to Mr. Rafael Schmitt', Etienne Cossart, 18 Nov 2016
  - SC3: 'Response: Assessment of structural sediment connectivity within catchments: insights from graph theory', Rafael J. P. Schmitt, 22 Nov 2016
    - AC3: 'Response to Mr. Rafael Schmitt', Etienne Cossart, 20 Jan 2017
RC1: 'Referee comment', Anonymous Referee #1, 06 Dec 2016
- AC5: 'Response to reviewer 1', Etienne Cossart, 20 Jan 2017
RC2: 'Referee comment', Anonymous Referee #2, 07 Dec 2016
- AC6: 'Response to reviewer 2', Etienne Cossart, 20 Jan 2017
SC4: 'Additional literature for more interesting discussion on graph theory/networks in sediment cascades', Rens Masselink, 08 Dec 2016
- AC4: 'Response to Mr. Masselink', Etienne Cossart, 20 Jan 2017

Peer-review completion

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

AR by Etienne Cossart on behalf of the Authors (20 Jan 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (23 Jan 2017) by Paola Passalacqua

RR by Anonymous Referee #2 (03 Feb 2017)

Suggestions for revision or reasons for rejection

Review of “Assessment of structural sediment connectivity within catchments: insights from graph theory” by Cossart and Fressard for Earth Surface Dynamics.

This review is of a revised manuscript following a first round of review comments. I note that I was a reviewer during the first round of review.

I commend the authors in reorganizing and revising the manuscript toward much improved clarity! Now it is more clear to me the novel contributions of this work. This manuscript presents a review of approaches for assessing structural sediment connectivity and describes relevant metrics from graph theory for measuring structural sediment connectivity. The authors point out that the connectivity index (IC) presented in Borselli et al 2008 and Cavalli et al 2013 is incapable of assessing how changes on the landscape affect sediment connectivity (delivery). Therefore, this work provides a network-based method for assessing structural sediment connectivity, by presenting a new way to compute IC, to address this problem. The manuscript concludes with an application of this new approach to a catchment in the French Alps.

My main issue is that the variable IC reported in this paper is presented with two entirely different definitions without distinguishing between the two with even a superscript or subscript. IC has been defined in the paper according to Borselli et al 2008 and Cavalli et al 2013 and the authors use the same IC variable in this manuscript but redefine how to compute it using network attributes rather than DEM attributes (e.g., slope). I suggest adding a subscript to the variable name to distinguish the two or call the new IC something else entirely. If the new and old way of reporting IC are similar, I would like to see a qualitative comparison (map) of the IC computed both ways for your study catchment. That way readers can get a sense if both methods provide similar insight and then further see that the new IC computed with network attributes offers additional benefits in that you can assess how landscape changes directly affect structural sediment connectivity. I think this comparison would strengthen the paper greatly.

This manuscript advances the field of sediment connectivity, worthy of publication in Earth Surface Dynamics. However, I believe this manuscript would benefit from some minor revisions.

Detailed manuscript comments (P, page; L, line):
P 1, L 16: I suggest adding here that this hypothesis is guiding this work. When I first read this I was wondering if this hypothesis was being tested herein. I suggest “The main hypothesis guiding this work is that the network structure…” or something similar.

P 3, L 3-7: This is a great addition that helps frame the manuscript and inform the reader where the manuscript is headed. Much appreciated!

P 3, L 12: The Sediment Delivery Ratio (SDR) is defined improperly. As it is written it represents the volume of sediment eroded (Vh) divided by the volume of sediment delivered to the outlet (Vo). Vh is > Vo, so in this context would be a number >1. The spirit of the SDR is to provide a percentage of the eroded sediments that reach the outlet. Therefore, SDR should be written as Vo/Vh to provide a number <1, consistent with the literature and your discussion on SDR (P 3, L 15-16). The results and conclusions of the manuscript are unaffected by this issue as mention of the SDR was given as contextual background.

P 6, L 10: “it can represents” should be “it can represent” or “it represents”

P 6, L 11: “arious” should be “various”?

P 7, L 8-11: “…we have developed the calculation of the IC (index of connectivity)…” Is this the same IC referred to as the “connectivity index” that was developed by Borselli et al 2008 and Cavalli et al 2013 (P 4, L 2)? When you say you are “developing the calculation of the IC” it is unclear whether you are stating that you have developed the IC or that you are developing your own method for calculating IC that is different from that proposed by Borselli et al 2008 and Cavalli et al 2013. From looking ahead, it appears that you are proposing a way to calculate the IC using network characteristics. I wonder if calling this new variable the IC is the best way to go as it adds confusion in that you have an IC that can be computed in two different ways! As an aside, I wonder if you compute the IC from the two different methods if you would obtain similar results or not. Maybe add a subscript to IC such that IC_w is the watershed assessment of IC from Borselli et al 2008 and Cavalli et al 2013 and IC_n to denote that this is the IC computed from the network. Or change the name of IC computed from the network to a different name/variable altogether. I think this was where some of my initial confusion on the reading of the previous version of the manuscript lie because I thought IC had been developed before and you were just redescribing and applying it here.

P 7, L 21: I think the denominator should be F_jo? Same in figure 1. Otherwise, why mention F_jo?

P 8, L 22: F and Shi do provide a “description of the sediment cascade skeleton” but more generally provide a description of the river-network skeleton. At this point, F and Shi are metrics characterizing directed graphs and could be used to assess cascades of other constituents through river networks, not just sediment.

P 8, L 31: The denominator is Ai and I think it should be Shi? Again, I think it is confusing to have IC essentially defined twice (equation 2 and again at equation 11). Maybe the spirit of IC is preserved in this new definition with the upstream and downstream terms and also with what it is trying to represent, but this a completely different variable.

P 9, L 14: I am not sure “reduced-complexity network” is the best term for describing your simplified network. You also use “virtual sediment cascade” (P 1, L 21) and a “fictive and simple sediment cascade” (P 3, L 4). I suggest choosing one term throughout and maybe “conceptual sediment cascade” or “conceptual network” to illustrate the concept of this work.

P 9, L 20: I suggest adding the specific variable you are referring to for clarity, both in text, in caption, and on the figures. For instance, here “…a map of potential fluxes (Fi)…” reminds the reader that by potential fluxes you are referring to Fi, and also add Fi to figure 2c. Figure 1 helps sort this out easily and is a very nice addition to the paper! Similarly, for accessibility (Shi).

P 13, L 21: At this point I am more convinced that there needs to be more clarity between the IC of Borselli et al 2008 and Cavalli et al 2013 and the network based IC calculated herein. Otherwise it will present confusion in the literature to which IC someone is referring to. Also, I would be interested in seeing if calculating IC both ways would give similar insights. This would be a nice addition and could really help showcase the strength of your new approach compared to that of Borselli et al 2008 and Cavalli et al 2013. It seems like you should already have all the information necessary to quickly perform this computation for your catchment. I do not think this further analysis is critical, but it would be a nice intermediate step for the community to see if the suggestion is to move away from the IC of Borselli et al 2008 and Cavalli et al 2013 and to using the network based IC developed herein.

References: In looking through the references, I did not see “Gran and Czuba, 2015” which upon checking the final published form is “Gran and Czuba, 2017”.

Hide

RR by Anonymous Referee #1 (08 Feb 2017)

ED: Publish subject to minor revisions (review by Editor) (19 Feb 2017) by Paola Passalacqua

AR by Etienne Cossart on behalf of the Authors (03 Mar 2017) Author's response Manuscript

ED: Publish subject to minor revisions (review by Editor) (22 Mar 2017) by Paola Passalacqua

AR by Etienne Cossart on behalf of the Authors (24 Mar 2017) Author's response Manuscript

ED: Publish as is (27 Mar 2017) by Paola Passalacqua

ED: Publish as is (28 Mar 2017) by Tom Coulthard (Editor)

AR by Etienne Cossart on behalf of the Authors (30 Mar 2017) Manuscript

Short summary

An assessment of connectivity will help in describing coupling patterns and scale dependence of erosional processes to understand and predict how policies at catchment scale may supply sediments to the river system (dismantlement of hydraulic infrastructures, changes in terms of land use, etc.).