| 10 Jul 2026
Do the same river engineering works cause the same morphological adjustments? Context matters!
Abstract. Rivers in the Anthropocene have undergone significant geomorphological changes due to engineering interventions such as channelization and hydroelectric infrastructure. These modifications have disrupted hydrological and sedimentary processes, reducing hydrological connectivity and habitat diversity. Evaluating their impacts is crucial for river management and restoration. We studied four channelized and bypassed sections of the French Rhône River (>60 km; ~13 % of its length). We analyzed spatio-temporal trajectories of channel narrowing and margin terrestrialization using GIS-based analysis of historical maps, aerial photographs, and satellite imagery. Historical hydrosedimentary data, including longitudinal profiles and water-level records, were used to assess the contribution of successive development phases. Beyond qualitative observations, such as the simplification of the river pattern toward a single-thread channel, we quantified the variable contributions of channelization and flow diversion. Active-channel narrowing ranged from −43 % to −17 % following channelization and from −32 % to −17 % following flow diversion. Margin terrestrialization increased by 27–44 % and 41–66 %, respectively. Where channelization had limited effects on terrestrialization, flow diversion played a greater role via water-level changes and reshaping the alluvial mosaic dependent on hydrological gradients. Although bed incision is generally expected to be limited in bypassed reaches, it reached −4.1 m in the most affected section (−5 cm·y⁻¹ during the 20th century), largely due to gravel mining associated with dam construction. This four-site comparison displays that engineering pressures interacted with local hydrosedimentary conditions to produce contrasting, legacy-dependent trajectories. Engineering margins, as ecotonal zones, are key indicators of geomorphic adjustment, documenting the system’s response to successive development phases.
This paper utilises a rich set of empirical data to isolate the responses of four different reaches of the French Rhône River which have been subjected to various engineering modifications over the course of almost the last 100 years. The combination of the detailed data (utilising flow gauging records in concert with planimetric evolution analysis, the latter derived mainly from aerial photos) and its relatively long duration offers a way of achieving new insights into how the four reaches respond in different ways to the two main perturbations experienced, namely lateral constriction and damming/flow regulation. Whilst the central finding of the work is perhaps not surprising (namely that the four reaches respond in different ways to similar perturbations, reflecting their different network contexts and boundary conditions), the outcomes are nevertheless therefore a valuable contribution – such detailed case studies of river response to anthropogenic perturbation, over multi-decadal timescales, remain relatively rare. The paper is executed in a robust manner and therefore my discussion below focuses on three different elements: (i) a few relatively minor reflective points, which are not essential to the contribution; (ii) a specific suggestion to revise the discussion, that whilst is not essential, might conceivably help to further improve the focus of the work, and (iii) identification of a few formatting and typographical errors.
Reflections/Queries
Figure 1: I think it could be helpful to define the reach acronyms PBN, PDR, etc somewhere on the legend or caption and I also wonder if it might be helpful to provide higher resolution insets of the 4 reaches as well, such that Figure 1 introduces both the large-scale context and specific details of each reach.
L188-189: It would be helpful here to detail as far as possible the extent to which the acquisition date for each photo series affects the value of the flow discharge observed on that date and therefore the extent to which flow variability between each photo series introduces inconsistencies into the inter-comparison of the digitized water and land boundaries. This point propagates through (potentially) into the analysis method in section 3.2.1
Figure 2: I spent quite a bit of time trying to figure out if the colours in each map and legend simply reflect the different dates of analysis or if the shadings reflect the evolution of a physical parameter (such as the water extent). I think it is the latter (this is implied as such at L360)? If it is then the legend should be adjusted to highlight this much more clearly. If I am incorrect (quite possible!) then I do wonder if it would be helpful to present the data in the way in which I thought it has been
L315-320 and elsewhere: There is a lot of reference throughout to Phases 1 and 2. Whilst these are defined at the outset I wonder if it would be helpful to remind readers of the actual dates/chronology of each Phase when they are referred to in this way, to aid clarity.
Sections 5.1.6 and 5.1.7. I felt that both of these sections were rather concise and a little bit more speculative and could potentially be deleted without significant loss. That said, this reflection likely falls into the same category as the specific suggestion below.
Specific Suggestion
The authors present quite a lengthy discussion. Sections 1 (5.1) and 2 (5.2) of this discussion focus on discussing the empirical findings of the work. In contrast, Section 3 (5.3) introduces a comparison of the new findings from the Rhône with other heavily modified European rivers, such as the Rhine, Danube and Po. I want to stress that I found the attempt to contextualise the study’s findings against other river systems that have also been heavily engineered interesting and I am confident that some readers would also find this material useful. That said, I also found myself wondering if this material – which adds another 3 pages to an already long discussion – is strictly necessary. I do think it could potentially be deleted without limiting the impact of the paper and doing so might offer the advantage that it would enable some readers to access the critical findings more easily. I recognise that there is therefore a trade-off: deleting the section 5.3 offers the advantage of focus and concision, but at the consequence of losing the contextualising material. For me at least the balance of the trade-off was determined in that the central thesis of the paper is that responses to engineering works are context dependent, so I felt that searching for similarities and differences between the different river systems is a little bit futile. But, other readers (and perhaps the authors) may disagree and this is why I frame the idea to delete section 5.3 as a suggestion only.
Formatting and Typographical Errors
L78 – incomplete phrase ‘few inter-site provided’?
L99-102 – text incorrectly appears in manuscript. Delete
L112 – phrase ‘objectives outlined’ needs revision for clarity
Table 1, row 4 – phrase ‘Mean annual discharge when total’ needs revision for clarity
L282 and 2877 – References to Tables 3 and 4 should be to Tables 2 and 3? Please check following references to Tables for correct sequence also.
Figure 3 – title of legend at bottom right of figure should be ‘ASSOCIATED DISCHARGE INFORMATION’
Figure 6 caption – correct reference formatting at end of caption
L553 – should be ‘Regardless of the….’
L630 and 631 - correct reference formatting
L686 – as above
L829 – text should be in separate ‘acknowledgements’ section?