Bankfull Transport Capacity and the Threshold of Motion in Coarse‐Grained Rivers

The threshold stress for bed sediment transport exerts a primary control on the geometry and stability of coarse‐grained rivers (diameter ≥ 5 mm). Understanding how riverbed mobility couples to channel form is a key mechanistic link for predicting river response to external perturbations such as land use practices and changing climate. Unfortunately, determination of a representative threshold stress is notoriously difficult in the field. Empirical studies have observed that the critical dimensionless shear (Shields) stress ( τ *c ) is correlated with channel slope, a property that is substantially easier to estimate. Mechanistic models have been developed to explain the observed correlation; however, limited field data preclude the widespread application of these models. For practical reasons, the empirical regressions between slope and τ *c are utilized as predictive models. Through a large compilation of field data, we demonstrate that there are two significant problems with using the empirical regressions: (1) they are based on a partial sampling of the observed parameter space of coarse‐grained rivers, and (2) they do not capture the covariation between the bankfull Shields stress ( τ *bf ) and τ *c . These regressions provide spurious predictions for the bankfull transport capacity ( τ *bf / τ *c ) of gravel‐bed rivers. When site‐specific empirical measurements of τ *c are made, coarse‐grained rivers exhibit a remarkably constant transport capacity that is in close agreement with equilibrium channel theory ( τ *bf = 1.2τ *c ). From these data we advocate that, in the absence of measurements, τ *c can be reasonably estimated from the τ *bf using equilibrium channel theory.