Along a reach of the Upper Tualatin River, Oregon, USA, the bed abruptly changes from gravel to sand. Field data show that this gravel-to-sand transition (GST) is co-located with a valley-slope break, an abrupt drop in median bed-sediment size (D50D_{50}), an increase in channel width, and an increase in bank-material cohesiveness, but the data cannot separate the relative causal weights of these covarying controls (i.e. cannot rank the controls by the strength of evidence that each determines the transition) or identify a single primary trigger. The unresolved problem is to determine the relative causal roles of the valley-slope break, the increase in bank cohesion (with which riparian vegetation covaries), and episodic sediment-supply discontinuities in fixing the position and abruptness (the along-channel distance over which the bed changes from gravel- to sand-dominated, i.e. over which D50D_{50} falls across the transition) of this transition, and to establish whether the observed empirical relationships suffice for a coupled mechanistic model — hydraulics, sediment transport, and bank-failure mechanics — that predicts the transition under altered flows, sediment supply, or management interventions.

Background

Rivers that drain mountainous terrain commonly show downstream fining: the median bed-sediment diameter (D50D_{50}) decreases in the downstream direction as particles abrade and are selectively sorted. In many alluvial rivers this fining is not gradual: over a short distance the bed changes abruptly from gravel-dominated to sand-dominated. Such a gravel-to-sand transition (GST) is a widespread feature whose cause has been debated for decades, and it coincides with a well-known bimodality of river-bed sediments — the grain size gap, i.e. the near-absence of bed material in the coarse-silt-to-fine-gravel range between about 1 and 10 mm. Proposed controls on where a GST forms include breaks in valley or water-surface slope, declines in sediment supply and flow competence (the ability of a flow to move particles of a given size), reductions in bed shear stress (the frictional force per unit area that flowing water exerts on the bed) from channel widening, and increases in the cohesiveness (silt-and-clay content) of bank materials, which narrows the range of widths a channel can adopt and alters local hydraulics. Riparian vegetation complicates attribution because tree roots reinforce banks: cohesive banks also tend to carry more mature floodplain forest, so bank strength and vegetation effects are covarying rather than independent factors.

Labbe, Hadley, Schipper, Leuven and Gardiner (2011) documented the GST of the Upper Tualatin River, a lowland gravel-to-sand transition reach in Oregon, USA, using site-scale surveys of bed sediment, bank materials, and riparian vegetation across 16 sites. They found that the transition coincides with a valley-slope break, an abrupt decline in bed D50D_{50}, an increase in channel width (BB), and a downstream increase in cohesive (silt/clay-rich) bank material, and they attributed the transition and the associated change in channel form jointly to the valley-slope break and an increase in bank resistance from cohesive bank materials, concluding that riparian vegetation at current densities and compositions is not a significant control on contemporary channel cross section form and may be responding to the bank and channel stability provided by cohesive bank materials in the laterally dynamic gravel-bed channel. However, because these factors change together along the same short reach, the observational design could not separate their relative causal weights or identify which of them is the primary trigger of the transition or its abruptness. Several further questions remain unresolved for this reach: episodic sediment-transport discontinuities (bed-load pulses, substrate abrasion, tributary inputs such as Williams Creek) that may destabilize the bed at and upstream of the transition are not quantified; the covariation of cohesive banks and mature trees leaves open whether restoring denser native floodplain forest or reintroducing large woody debris (fallen trees, logs, and root wads in the channel, which trap sediment and deflect flow) would actively change channel form or merely colonize already-stable banks; a few sites deviate from the general patterns (a sand-bed site with mostly noncohesive banks; gravel-bed sites with cohesive bank toes, i.e. cohesive sediment at the lowest part of the bank where it meets the bed) and small statistical samples constrain the generality of the subreach correlations; and no mechanistic, process-based model coupling hydraulics, sediment transport, and bank-failure mechanics was constructed to test whether the empirical relationships predict the transition under changed conditions.

Later work on GSTs elsewhere has addressed pieces of the general question without closing it. Venditti and Church (2014) analyzed the large Fraser River, British Columbia, where the GST manifests as a terminating gravel wedge with a diffuse gravel extension downstream; they measured a dramatic drop in bed shear stress at the end of the wedge that halts general gravel mobility, and proposed that the slope break there is a consequential feature of the transition rather than its cause — one mechanistic answer for one supply-forced large river. A 2014 Eos summary (Wendel, 2014) of that work stated that scientists remained puzzled by why exactly GSTs occur and why they are abrupt in some rivers and gradual in others. A dedicated review of GST phenomena and the grain size gap (Dingle, Kusack and Venditti, 2021), which cites the Tualatin study, consolidated the phenomenon as an active synthesis topic, and site-specific studies of GST position and migration continue to appear (e.g. He et al., 2023, on the dam-regulated Yangtze River; Yoshida and Koarai, 2026, on identifying GST zones in Japanese alluvial plains with the normalized steepness index, a measure of channel steepness normalized for upstream drainage area and computed from digital elevation data). No published follow-up is known to have returned to the Tualatin reach or to have settled the causal competition between slope break, bank cohesion, and vegetation for transitions of this kind.

Problem Statement

For the gravel-to-sand transition (GST) reach of the Upper Tualatin River, Oregon, determine what causally controls the position and abruptness of the transition, and whether that understanding is predictive. Concretely: (1) disentangle the relative causal roles of the valley-slope break, the downstream increase in bank-material cohesiveness, and episodic sediment-transport discontinuities (bed-load pulses, substrate abrasion, tributary inputs such as Williams Creek) in triggering and maintaining the abrupt co-located changes in bed D50D_{50} (the median bed-sediment grain size) and channel width — including whether the slope break is cause, consequence, or co-symptom; (2) resolve whether riparian vegetation is an independent driver, by establishing whether restoring higher-density native floodplain forest or reintroducing large woody debris would actively alter channel form at and near the transition, or whether mature trees merely colonize already-stable cohesive banks; (3) assess how far the site-level empirical patterns generalize across the documented exceptions (the sand-bed site with mostly noncohesive banks, gravel-bed sites with cohesive bank toes, i.e. cohesive sediment at the lowest part of the bank where it meets the bed), using the existing 16-site data and the cited literature; and (4) determine whether the observed empirical relationships suffice for a mechanistic, process-based model that explicitly couples channel hydraulics, sediment transport, and bank-failure/bank-strength mechanics to reproduce the observed transition and predict its location and abruptness under altered flow regimes, sediment supply, or management interventions. These four components share one objective — a causal, predictive explanation of what forces and maintains this transition — and all four are required to answer it: a submission resolves the task only when every component has been addressed, and resolving one does not substitute for another. Specifically, each component is resolved by an explicit verdict that states the criterion on which it rests: (1) an explicit causal ranking of the three candidate controls, with the evidential basis for the ranking, and a verdict on whether the valley-slope break is cause, consequence, or co-symptom; (2) a yes/no verdict on whether riparian vegetation is an independent driver, i.e. whether restoring higher-density native floodplain forest or reintroducing large woody debris would alter channel form or merely colonize already-stable cohesive banks; (3) a yes/no verdict on whether the site-level patterns generalize across the documented exceptions, argued from the existing 16-site data and the cited literature and stating the criterion used; and (4) a yes/no determination of whether the observed empirical relationships suffice for a coupled mechanistic model, stating the criterion used for sufficiency and the predictive capability that follows.

References

  1. Labbe, J.M., Hadley, K.S., Schipper, A.M., Leuven, R.S.E.W., Gardiner, C.P. (2011). Influence of bank materials, bed sediment, and riparian vegetation on channel form along a gravel-to-sand transition reach of the Upper Tualatin River, Oregon, USA. Geomorphology 125(3), 374–382. doi:10.1016/j.geomorph.2010.10.013. https://doi.org/10.1016/j.geomorph.2010.10.013
  2. Venditti, J.G., Church, M. (2014). Morphology and controls on the position of a gravel-sand transition: Fraser River, British Columbia. Journal of Geophysical Research: Earth Surface 119(9), 1959–1976. doi:10.1002/2014JF003147. https://doi.org/10.1002/2014JF003147
  3. Wendel, J. (2014). Where does the bed of a river change from gravel to sand? Eos, Transactions American Geophysical Union 95(49), 472. doi:10.1002/2014eo490017. https://doi.org/10.1002/2014eo490017
  4. Dingle, E.H., Kusack, K.M., Venditti, J.G. (2021). The gravel-sand transition and grain size gap in river bed sediments. Earth-Science Reviews 222, 103838. doi:10.1016/j.earscirev.2021.103838. https://doi.org/10.1016/j.earscirev.2021.103838
  5. He, Z., Sun, Z., Zhou, W., Li, Y., Hu, Y. (2023). Gravel-Sand Transition of the Yangtze River: Human Disturbances, Migration Processes, and Controlling Factors. Journal of Geophysical Research: Earth Surface 128(6), e2022JF006984. doi:10.1029/2022JF006984. https://doi.org/10.1029/2022JF006984
  6. Yoshida, K., Koarai, M. (2026). Identifying gravel-sand transition zones in alluvial plains using normalized steepness index. Geomorphology 495, 110151. doi:10.1016/j.geomorph.2025.110151. https://doi.org/10.1016/j.geomorph.2025.110151