the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 Mar 2026
Alluvial river-channel width: transient adjustment and dynamic equilibrium
Abstract. Alluvial river channels naturally widen and narrow as large floods scour banks and smaller ones supply sediments that help build bars and channel margins. Despite substantial advances into the controls on the equilibrium width of river channels, relatively little theory underpins our knowledge on transient river-channel-width evolution. Such a knowledge gap inhibits us from predicting the impacts of present-day nonstationary hydrology on river-channel stability and geomorphic change. Here we present a unified approach to model transient channel widening, via erosion of cohesive banks and mobilization of noncohesive clasts, and narrowing, via lateral diffusion of sediment that attaches to the banks. The resultant model can take a full hydrograph as input, allowing the hydraulic geometry and associated "geomorphically effective" water discharge to emerge dynamically. Stable widths develop via the inverse relationship between channel width and flow depth, and therefore, shear stress on the channel margins. Equilibrium solutions closely approximate data and theory on channels with both gravel and mud banks, and we compare transient solutions to the rapidly widening Minnesota and Cannon Rivers (Minnesota, USA) and the narrowing Green River and Diamond Fork (Utah, USA). Documented Python source code to run these computations is available from GitHub and Zenodo via the "OTTAR" package, and may be installed via pip from PyPI.
- Preprint
(4508 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-1018', John Shaw, 13 Apr 2026
-
RC2: 'Comment on egusphere-2026-1018', Anonymous Referee #2, 14 Apr 2026
Summary
This paper presents a new numerical model for transient channel width adjustments in alluvial rivers. The model is a novel and important contribution to the scientific community that moves beyond equilibrium theory to capture dynamic channel width adjustments over short timescales, using daily hydrographs as input. The authors hypothesize that a model for transient width adjustment can explain the scatter found in width-discharge relationships observed in natural rivers compared to the predictions made by equilibrium theory. In addition to presenting the model, the authors apply it on four case studies both of sand and gravel bedded rivers in the USA to demonstrate the model’s validity. The key contribution of this model is an explicit mechanism of channel narrowing, which is currently absent from near-threshold channel theory. The paper is very thorough in its presentation of the model equations and its assumptions, which are well explained and justified.General Comments:
My main suggestions relate to (1) exploration of model behavior and (2) clarification of key assumptions.
I think this is a very interesting model and I'd like to know more about how it behaves. In particular, I'd like to see more exploration of parameter space and model behavior beyond the best-fit run. Are some parameters more important than others for achieving a good fit to the data? I'm also curious whether the general patterns of widening and narrowing are consistent across near-optimal model runs, or whether there are alternative well-fitting solutions that produce qualitatively different behavior. It would also be informative if figures 10–13 included shaded envelopes indicating the spread of predictions from near-optimal runs rather than the single best fit (see also my figure comments on 10-13).
I would also like to see some expanded discussion on the implications of using a fixed bed elevation. This choice is a reasonable one, but I think there could be a clearer distinction between whether deposition is the only mechanism for narrowing in this model vs. in real rivers (see my comment below on 792-793 for a more detailed discussion of this point). In which conditions or timescales might the fixed elevation assumption break down?Specific Comments:
161-162: “The obvious solution to the above thought exercise belies the two features of existing equilibrium-width theory that it illuminates.” I found this sentence hard to parse, consider revising for clarity.555: A map showing the locations of the case study sites would help contextualize these locations for readers outside of the USA.
617-618: It would be intereseting to see the ratio of bank stress: critical stress over time (similar to how it’s shown in figure 9d) for this case study and the others that follow. Consider adding a secondary y-axis to 10a, or a supplementary figure.
625: It would be helpful to have a map showing where the study locations for the Minnesota vs. the Cannon river are relative to one another. I can’t really gauge this from the text and it would be useful to have a general sense of far these two sites are from one another, since they are similar in the sense that they are sand-bedded rivers in Minnesota, but I am not very familiar with the geography of the state.
688-694: Was there a dominant mechanism of narrowing in this case, or were there roughly equal components of narrowing via deposition of suspended sediment vs. bedload?
792-793: Is the observed narrowing at intermediate flows necessarily representative of natural rivers, or is it primarily a feature of the model design? In non-rectangular channels, vertical incision can alter channel width without requiring sediment deposition. For example, if one models the channel as a trapezoid as in Lague (2010), widening vs. narrowing is a function of the relationship between the horizontal vs. vertical incision rates and the bank angle. Lague’s work highlights the potential importance of vertical incision in driving changes in channel width through geometric relationships. It would be helpful to acknowledge that if vertical incision outpaces width adjustments, narrowing could also occur at higher flows in such channels. To be clear, I think the authors’ decision to focus on width adjustments while holding bed elevation fixed is a reasonable simplification, but the discussion could expand on the implications of this assumption.
Figure Comments:
Figures 10-13: Similar to my comment on lines 617-618, I think you have an opportunity to illustrate more of the model behavior on these plots – either by adding a secondary y-axis to plot shear stress ratios, or adding a swath showing the range of results from model runs that weren’t the best fit, but close to it.Figure 10: Is this plot missing error bars or were they intentionally excluded? It would be helpful to clarify this in the caption, I wasn’t able to identify an explanation in the text.
Figure 12: I found the gray points representing measured widths hard to visually distinguish from the line for modeled width underneath on this plot in particular. I suggest making the points a bright color or otherwise adjusting the symbology of the plot to make them easier distinguish and adjusting figs. 10-13 to match.
Technical Corrections:
603: Typo, Kohout citation is missing a space786: Typo, missing parentheses around Chen and Duan citation
Citations:
Lague, D. (2010). Reduction of long‐term bedrock incision efficiency by short‐term alluvial cover intermittency. Journal of Geophysical Research: Earth Surface, 115(F2).Citation: https://doi.org/10.5194/egusphere-2026-1018-RC2
Model code and software
OTTAR A. D. Wickert https://github.com/MNiMORPH/OTTAR
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 194 | 103 | 17 | 314 | 14 | 18 |
- HTML: 194
- PDF: 103
- XML: 17
- Total: 314
- BibTeX: 14
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript proposes a process-based model for predicting river width evolution. Processes of widening and narrowing are quantified, for both bedload and suspended load. The result is a model with many equations and many parameters but presents a coherent picture of what is happening in evolving channels. The referencing and notation are thorough and meticulous. I expect that this will be a very impactful contribution.
Is the manuscript too long? It is indeed very long and I spent a lot of time lost navigating back and fourth between sections. The organization is very good, and I can see why all parts are justified. On the other hand, it probably could be cut by 5-10% in the thorough description of previous studies, and there are other parts of the manuscript that could use that space to really dig into essential points.
The study makes a profound point that much less is known about narrowing than widening. I think that this would be valuable to say in the abstract. I think it is equally important to point out (maybe in Discussion) that the narrowing terms f_sigma and k_n seem to control the resulting behavior more than other terms. Having 6-8 orders of magnitude variation (Table 2) is suspicious. I think that this is a publishable and boundary-pushing result, but I think it should be circled as a red flag for future researchers.
Relatedly, the sediment available to narrow the channel is controlled by shear stress at the cross section, which dismisses the real possibility of differences in sediment availability. I don’t mean to relax the quasi-equilibrium or near-threshold assumptions, but I suggest that it is worth mentioning that net sediment deposition at the cross section, non-local effects in suspended sediment, or maybe sediment transport hysteresis, might really affect these processes and the estimated parameters in the case studies.
My eyes raised when the equilibrium width is “close to” predictions of threshold-based theory (L700 and Figure 9 caption). What is the discrepancy (for example, within 5%)? It looks like the model would recover threshold theories by definition. If there was some discrepancy, it could use some discussion.
Line Edits
L230: roll downhill(?) toward the higher-shear-stress channel thalweg
A2: might it be possible to list the equation numbers that each variable appears in. I am spending a lot of time scrolling between sections lost.
Figure 11: the x axes on the subplots aren’t aligned.
L727: The claim is that equilibrium widths depart from equilibrium width theory. For this, the equilibrium width theory of Parker or Dunne and Jerolmack should be plotted in Figure 9 I think.
Table 2’s caption is cut off in the pdf.