Search

During a flood, the geometry of a river channel constrains the flows of water and sediment, however, over many floods, bankfull channel geometry evolves to reflect the longer‐term fluxes of water and sediment supplied by the catchment. Physics‐based models predict the average relationship between bankfull geometry and discharge to within an order of magnitude, however, observed variability about the prediction remains unaccounted for. We used high‐resolution topography to extract continuous measurements of bankfull width from 67 sites spanning the continental United States, yielding a reach‐scale probabilistic description of river width for each site. Within an individual reach, bankfull river width is well‐described by a lognormal distribution. Rivers that spend a greater proportion of time above bankfull are wider for the same bankfull discharge, revealing an unrecognized pathway through which climatic or engineered changes in flow frequency could alter river geometry and therefore impact aquatic habitat and flooding risk. Plain Language Summary: Rivers adjust their shape to transport the water and sediment supplied from their watersheds. Predicting flooding throughout river networks depends on understanding how rivers grow downstream as they accumulate water from larger areas. This relationship between river size and the water discharge that just fills the channel (bankfull) underpins flood routing and hazard models and is remarkably consistent for creeks in the headwaters and throughout a watershed as rivers reach their terminus in lakes and oceans. However, for shorter sections of a river we find that the size can be highly variable, and it remains unclear how to incorporate this reach‐scale variability into watershed and basin scale models. Here we demonstrate that reach‐scale variation has a well‐defined average value and predictable variability. By accounting for the reach‐scale variability, we find that for the same amount of discharge rivers which experience more flooding are wider while those where floods are rarer are narrower. These findings may hold particular relevance for understanding how snowmelt systems which produce long duration floods may adjust to diminishing snowpacks and how to manage river systems below dams where flow durations and magnitudes may be closely regulated. Key Points: Reach‐scale bankfull river width is lognormally distributed with a constant geometric standard deviationVariation in bankfull river width across sites depends on the flow duration where rivers that spend more time above flood stage are widerBankfull flow intermittency, runoff, and basin aridity are correlated, representing an imprint of climate on river geometry [ABSTRACT FROM AUTHOR]

The battering of bedrock by bedload collisions is the primary mechanism by which bedrock rivers erode and landscapes evolve. The energy imparted via impacts acts to detach bedrock via the growth and intersection of surface fractures. We present impact experiments designed to test the influence of particle impact energy on bedrock erosion rates. We found that erosional efficiency increased with increasing impact energy. Notably, these increases in efficiency are not captured by a widely‐used mechanistic bedrock erosion model. Observed increases in erosional efficiency were linked with enhanced elastic energy dissipation captured by differences in the coefficient of restitution. We suggest that this increase in energy dissipation is indicative of enhanced crack extension for high velocity impacts. Our experiments indicate a clear energy‐dependence for bedrock detachment processes that is not yet captured by bedrock incision models but may be integrated into long‐term erosion rates and landscape evolution. Plain Language Summary: Bouncing sediment grains batter the bedrock beds of rivers during floods. As these impacts collide with the bed, they slowly chip away at the riverbed bedrock. It's perhaps intuitive that a grain with more energy when it collides with the riverbed is able to chip away more bedrock. But, how much more destructive are these high energy impacts? We designed an experiment where we hit a bedrock sample with a steel ball at a range of impact energies, over and over again, up to 30,000 times, and measured how much material was chipped away. High energy impacts eroded more material, but this increased efficiency of erosion is not fully captured by impact‐based erosion models. We hypothesize that for higher energy impacts, a larger fraction of the incoming energy is used to crack the bedrock, more easily eroding material. We liken the relative influence of these high energy impacts to a battering ram, where only a few strikes can easily break down a door, whereas lower energy impacts, similar to a simple knock, are unlikely to achieve the same result. We suggest that this effect may be important in determining the influence of extreme events on bedrock erosion over long timescales. Key Points: Impact experiments demonstrate that bedrock erosional efficiency is enhanced with increasing impact energyExperimentally‐derived coefficients of restitution and rock resistance coefficients indicate enhanced elastic energy dissipation with increasing impact energyResults suggest that more energetic impacts enhance crack extension, increasing erosional efficiency [ABSTRACT FROM AUTHOR]