Bank erosion processes measured with UAV-SfM along complex banklines of a straight mid-sized river reach

Duró, Gonzalo; Crosato, Alessandra; Kleinhans, Maarten G.; Uijttewaal, Wim S. J.

doi:https://doi.org/10.5194/esurf-6-933-2018

Articles | Volume 6, issue 4

https://doi.org/10.5194/esurf-6-933-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/esurf-6-933-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 6, issue 4

Research article

|

25 Oct 2018

Research article |

| 25 Oct 2018

Bank erosion processes measured with UAV-SfM along complex banklines of a straight mid-sized river reach

Gonzalo Duró, Alessandra Crosato, Maarten G. Kleinhans, and Wim S. J. Uijttewaal

Download

Final revised paper (published on 25 Oct 2018)
Preprint (discussion started on 16 Mar 2018)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'On surveying long river corridors with SfM and drones.', Patrice Carbonneau, 05 Apr 2018
- AC1: 'Author's response to Referee #1', Gonzalo Duro, 25 Jun 2018
RC2: 'cComment on «A low-cost technique to measure bank erosion processes along middle-size river reaches» by Duró et al.', Anonymous Referee #2, 13 May 2018
- AC2: 'Author's response to Referee #2', Gonzalo Duro, 25 Jun 2018

Peer-review completion

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

AR by Gonzalo Duro on behalf of the Authors (25 Jun 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (13 Jul 2018) by Richard Gloaguen

RR by Anonymous Referee #2 (14 Aug 2018)

RR by Patrice Carbonneau (23 Aug 2018)

Suggestions for revision or reasons for rejection

Second review of ' A low-cost technique to measure bank erosion processes along middle-size river reaches '

The authors have produced a new manuscript with much more analysis and more informative figures. In a nutshell, I remain of the opinion that this work does not represent a great innovation in UAV photogrammetry. And I question why this should constitute a methodological contribution.

At the outset, the authors have not significantly changed the title. I think it is questionable to call this method 'low-cost' since it relies on GCPs acquired with an RTK-GPS which is a tool at the high end of the spectrum of terms of field equipment cost. This paper has however clarified the problems with the data. In figure 7 the authors now show a detailed plot of SfM-ALS errors. This says much about the problems with the data. Figure 7b shows a scaling problem in the domain, not noticed by the authors. The lower banks is too low but the upper bank is too high. This is caused by a mis-calibration of the focal length. This in turn is caused by the fact that all the GCPs are at the same datum. Added to that is a slight downstream doming effect. Figure 7 actually shows that there is no significant rotation problem. The authors have sadly devoted great effort to prove this with a complicated analysis, but figure 7 does the job on its own, even if it reveals another problem. A simple analysis of the horizontal span of the GCPs in comparison to the vertical GCP accuracy would have demonstrated rotational stability and so figure 9, with the complex analysis, is not needed. The methods clarifications also now reveal that the authors used the GCPs in stages with some at the alignment phase and others later at the densification stage. I have never done this or seen it, it needs justification or else would seem overly complicated.

There are a many things the authors could have done to tailor their work to the specific challenges of the problem. First, they should have set GCPs in a wider footprint and tried to set some on the bottom bank. Second, they should have flown lines in the cross-stream direction. Third, they could have considered a contribution of direct georeferencing, the recent paper by James et al 2017 gives very clear indications on what precision to expect. I we consider figure 12, then it would seem that the authors needed to detected metric changes over a year. The DG methods shown in Carbonneau and Dietrich 2017 combined with a prediction of data quality from James et al 2017 could have sufficed and the authors did not even need any GCPs to get the quality required to monitor these large changes. The overwhelming impression of the work is that they went out to the field before consulting any published works on how to address their problem. Now the authors are trying to publish sub-optimal data collected with a sub-optimal method even when it is clear that the most signification improvements to be made are with the field data acquisition.

The authors have improved their knowledge of photogrammetry but the work is still characterised by incorrect statements. Aside from the fact that they did not see the implications of figure 7B, they make minor incorrect statements about theory. These are highlighted in my annotated manuscript.

In the end, I think existing UAV-SfM literature can support the geomorphology work presented here. The geomorphology is interesting but its study does not require a new UAV method. The authors could simply show figure 7 in a geomorphology paper and state that errors are within 5-10cm and are therefore small in relation to the scale of the changes seen on the scarps. In terms of the SfM problem (long straight reaches with sub-vertical scarps), the data collection methods here is simply not suited to the problem. The authors have not taken into account some newly emerging literature before they designed their survey. Unfortunately I cannot recommend publishing a technical note based on this data.

A potential path for improvement would be to use the ALS data to aid in the calibration of the camera in terms of distortion and focal length. This would be innovative and depart from the existing literature.

Patrice Carbonneau

Referee Report: PDF

Hide

ED: Publish subject to minor revisions (review by editor) (03 Sep 2018) by Richard Gloaguen

AR by Gonzalo Duro on behalf of the Authors (13 Sep 2018) Author's response Manuscript

ED: Publish as is (28 Sep 2018) by Richard Gloaguen

ED: Publish as is (30 Sep 2018) by Tom Coulthard (Editor)

AR by Gonzalo Duro on behalf of the Authors (09 Oct 2018) Manuscript

Short summary

The challenge to measure three-dimensional bank irregularities in a mid-sized river reach can be quickly solved in the field flying a drone with ground-control points and later applying structure from motion photogrammetry. We tested a simple approach that achieved sufficient resolution and accuracy to identify the full bank erosion cycle, including undermining. This is an easy-to-use and quickly deployed survey alternative to measure bank erosion processes along extended distances.