Joel P. L. Johnson

Contrasting bedrock incision rates from snowmelt and flash floods in the Henry Mountains, Utah

Hydrograph variability and channel morphology influence rates of fluvial bedrock incision, but little data exist on these controls in natural channels. Through field monitoring we demonstrate that (1) short-term bedrock channel incision can be rapid, (2) sustained floods with smaller peak discharges can be more erosive than flash floods with higher peak discharges, due to changes in bed alluviation, and (3) bedrock channel morphology varies with local bed slope and controls the spatial distribution of erosion. We present a three-year record of flow depths and bedrock erosion for a human-perturbed channel reach that drains the Henry Mountains of Utah, USA. Starting from a small and steep (∼30% slope), engineered knickpoint in Navajo sandstone, erosion has cut a narrow, deep, and tortuous inner channel in ∼35–40 years. Along the inner channel, we measured up to 1/2 m of vertical incision into Navajo sandstone over ∼23 days, caused by the 2005 season of exceptional snowmelt flow. In contrast, flash floods caused little bedrock incision even when peak discharges were much higher than the peak snowmelt flow. Flash floods were net depositors of coarse sediment while snowmelt flow cleared alluvial cover. We document the formation of a pothole and interpret that it was abraded by bedload rather than fine suspended sediment. Finally, several slot canyons (Peek-a-boo, Spooky, and Coyote Gulch narrows) in the nearby Escalante River drainage basin have erosional morphologies similar to the monitored channel reach. Feedbacks between flow, sediment transport, and transient erosion provide a plausible explanation for the evolution of channel slope, width, and bed roughness of these natural bedrock channels.

A surface roughness model for predicting alluvial cover and bed load transport rate in bedrock channels

Partial alluvial cover in bedrock channels influences downcutting rates and mountain river morphodynamics. Flume experiments have demonstrated that a wide range of stable cover fractions are possible for a given ratio of sediment supply to transport capacity. Existing cover models impose different unique relationships between cover fraction and reach-scale transport variables (e.g., slope, discharge, sediment supply, and grain size). Individually, these models cannot predict the range of cover behaviors observed experimentally, suggesting that additional variables may be important. I propose a 1-D model in which bedrock surface roughness is a control on partial alluvial cover. In the model, surface roughness affects both shear stresses and thresholds of grain motion and therefore sediment transport capacity. The roughness of the combined bedrock-alluvial bed surface varies with the cover fraction. Model trends are reasonably consistent with previous experiments over ranges of roughness, slope, and shear stress. When bedrock roughness is much greater than grain size, cover increases approximately linearly with the ratio of sediment supply to transport capacity. When bedrock roughness is much less than the sediment diameter, the model tends to predict abrupt shifts between complete bedrock exposure and alluvial cover. This runaway alluviation occurs for model solutions that are unstable to perturbations in supply, which is the case when transport capacity decreases as alluvial cover increases. In addition, a model variant in which sediment supply is controlled by the amount of sediment on the bed suggests that a linear cover relation may be sufficient for modeling sediment flux-dependent erosion at landscape scales.

Chemical weathering as a mechanism for the climatic control of bedrock river incision

Feedbacks between climate, erosion and tectonics influence the rates of chemical weathering reactions, which can consume atmospheric CO2 and modulate global climate. However, quantitative predictions for the coupling of these feedbacks are limited because the specific mechanisms by which climate controls erosion are poorly understood. Here we show that climate-dependent chemical weathering controls the erodibility of bedrock-floored rivers across a rainfall gradient on the Big Island of Hawai‘i. Field data demonstrate that the physical strength of bedrock in streambeds varies with the degree of chemical weathering, which increases systematically with local rainfall rate. We find that incorporating the quantified relationships between local rainfall and erodibility into a commonly used river incision model is necessary to predict the rates and patterns of downcutting of these rivers. In contrast to using only precipitation-dependent river discharge to explain the climatic control of bedrock river incision, the mechanism of chemical weathering can explain strong coupling between local climate and river incision.

Using RFID and accelerometer‐embedded tracers to measure probabilities of bed load transport, step lengths, and rest times in a mountain stream

We present new measurements of bed load tracer transport in a mountain stream over several snowmelt seasons. Cumulative displacements were measured using passive tracers, which consisted of gravel and cobbles embedded with radio frequency identification tags. The timing of bed load motion during 11 transporting events was quantified with active tracers, i.e., accelerometer-embedded cobbles. Probabilities of cobble transport increased with discharge above a threshold, and exhibited slight to moderate hysteresis during snowmelt hydrographs. Dividing cumulative displacements by the number of movements recorded by each active tracer constrained average step lengths. Average step lengths increased with discharge, and distributions of average step lengths and cumulative displacements were thin tailed. Distributions of rest times followed heavy-tailed power law scaling. Rest time scaling varied somewhat with discharge and with the degree to which tracers were incorporated into the streambed. The combination of thin-tailed displacement distributions and heavy-tailed rest time distributions predict superdiffusive dispersion.

The effects of precipitation gradients on river profile evolution on the Big Island of Hawai’i

To better understand how climate affects bedrock river incision and long-term landscape evolution in the absence of tectonic forcing, we quantify differences in the longitudinal profiles of eroding bedrock channels across the Kohala peninsula on the northern tip of the Big Island of Hawai’i. An orographic rainfall gradient causes mean annual precipitation rates to vary by over an order of magnitude, from greater than 4000 mm/yr on the wet side to less than 250 mm/yr on the dry side of the peninsula. Channels on the wet side are relatively deeply incised into the surrounding landscape and have developed profiles in which slope increases downstream in upstream reaches (convex form) and decreases downstream in downstream reaches (concave form). Wet-side river channels have relatively large (10–30 m) vertical steps, or knickpoints, in the convex zone. In contrast, channels on the dry side of the peninsula are more shallowly incised, have developed nearly straight longitudinal profiles in which channel slope does not significantly change from upstream to downstream, and have smaller knickpoints. Channel profile form changes from straight to convex-concave at a watershed-averaged mean annual precipitation rate of between ∼1300 and 1750 mm/yr, suggesting a climatic threshold in this landscape, above which bedrock incision rates are enhanced. Valley depth (used as a proxy for the magnitude of channel incision) increases with increasing mean annual precipitation and is consistent with a similar fluvial-incision threshold. While the lithology is entirely basalt, incision patterns also appear to be affected by spatial differences in bedrock weathering due to both local climate and basalt flow age. The older and more weathered Pololū basalts (260–450 ka) are capped by the younger Hawī basalt series (120–260 ka), and we interpret that heterogeneous weathering of these units influences channel form, relief, sediment production, and knickpoint development.

Coarser and rougher: Effects of fine gravel pulses on experimental step‐pool channel morphodynamics

Understanding how steep mountain rivers respond to natural and anthropogenic sediment supply perturbations is important for predicting effects of extreme events (e.g., floods and landslides) and for restoring rivers to more natural conditions. Using flume experiments, we show that stabilized step-pool-like channel beds can respond to pulses of finer gravel by becoming even coarser and rougher than before. Adding finer gravel initially reduces bed roughness and also increases the mobility of previously stable bed grains. Small- and intermediate-diameter clasts are then preferentially winnowed from the bed surface, leaving behind higher concentrations of even larger clasts. Ultimately, this results in both a coarser and rougher bed. Our experiments demonstrate that steep river beds become stable through the coevolution of bed roughness and surface grain size distribution and that these morphological variables can be sensitive to the history of upstream sediment supply.

Arroyo channel head evolution in a flash-flood–dominated discontinuous ephemeral stream system

We study whether arroyo channel head retreat in dryland discontinuous ephemeral streams is driven by surface runoff, seepage erosion, mass wasting, or some combination of these hydrogeomorphic processes. We monitored precipitation, overland flow, soil moisture, and headcut migration over several seasonal cycles at two adjacent rangeland channel heads in southern Arizona. Erosion occurred by headward retreat of vertical to overhanging faces, driven dominantly by surface runoff. No evidence exists for erosion caused by shallow-groundwater–related processes, even though similar theater-headed morphologies are sometimes attributed to seepage erosion by emerging groundwater. At our field site, vertical variation in soil shear strength influenced the persistence of the characteristic theater-head form. The dominant processes of erosion included removal of grains and soil aggregates during even very shallow (1–3 cm) overland flow events by runoff on vertical to overhanging channel headwalls, plunge-pool erosion during higher-discharge runoff events, immediate postrunoff wet mass wasting, and minor intra-event dry mass wasting on soil tension fractures developing subparallel to the headwall. Multiple stepwise linear regression indicates that the migration rate is most strongly correlated with flow duration and total precipitation and is poorly correlated with peak flow depth or time-integrated flow depth. The studied channel heads migrated upslope with a self-similar morphologic form under a wide range of hydrological conditions, and the most powerful flash floods were not always responsible for the largest changes in landscape form in this environment.

Weathering and abrasion of bedrock streambed topography

Our framework for understanding morphodynamic feedbacks in bedrock rivers is built upon the assumption that rock erodibility is reasonably uniform at the sub-reach scale. Here, we demonstrate that climate-controlled rock weathering combined with bedload abrasion can produce systematic spatial variations in erodibility across bedrock streambed topography. Rock strength data from five channel reaches across the Big Island of Hawaiʻi show that upstream-oriented rock surfaces are stronger than downstream-oriented surfaces on the same bedrock protrusion. Moreover, the overall strength of these protrusions correlates with local mean annual precipitation rate, demonstrating climatic control of streambed erodibility. Comparing inferred field abrasion rates with experimental flume measurements, we demonstrate that abrasion rates scale exponentially with the orientation of local bed topography relative to streamflow, independent of weathering. However, the spatial variability in abrasion rate across bedrock protrusions is significantly reduced in the field, where large spatial variations in erodibility occur due to weathering. The methods presented here provide a straightforward field-based approach for evaluating the potential influence of weathering on abrasion in bedrock rivers.

Particle transport mechanics and induced seismic noise in steep flume experiments with accelerometer-embedded tracers: Experimental testing of seismic noise generated by sediment transport

Recent advances in fluvial seismology have provided solid observational and theoretical evidence that near-river seismic ground motion may be used to monitor and quantify coarse sediment transport. However, inversions of sediment transport rates from seismic observations have not been fully tested against independent measurements, and thus have unknown but potentially large uncertainties. In the present study, we provide the first robust test of existing theory by conducting dedicated sediment transport experiments in a flume laboratory under fully turbulent and rough flow conditions. We monitor grain-scale physics with the use of ‘smart rocks’ that consist of accelerometers embedded into manufactured rocks, and we quantitatively link bedload mechanics and seismic observations under various prescribed flow and sediment transport conditions. From our grain-scale observations, we find that bedload grain hop times are widely distributed, with impacts being on average much more frequent than predicted by existing saltation models. Impact velocities are observed to be a linear function of average downstream cobble velocities, and both velocities show a bed-slope dependency that is not represented in existing saltation models. Incorporating these effects in an improved bedload-induced seismic noise model allows sediment flux to be inverted from seismic noise within a factor of two uncertainty. This result holds over nearly two orders of magnitude of prescribed sediment fluxes with different sediment sizes and channel-bed slopes, and particle–particle collisions observed at the highest investigated rates are found to have negligible effect on the generated seismic power. These results support the applicability of the seismic-inversion framework to mountain rivers, although further experiments remain to be conducted at sediment transport near transport capacity.

Predicting paleohydraulics from storm surge and tsunami deposits: Using experiments to improve inverse model accuracy

How accurately can flow depths and velocities of storm surges and tsunamis be predicted from sedimentary deposits? Inverse models have been proposed to quantify hydrodynamics from suspended sediment deposits, but assumptions about how deposit grain size distributions (GSDs) are influenced by flow characteristics remain largely untested. Using laboratory experiments, we evaluate an existing advection-settling model in which suspended sediment transport is assumed to reflect horizontal advection (constraining flow velocity) and vertical settling from the water surface (constraining depth). While the original model assumed that depth and velocity would be best predicted by the deposit D95 (the diameter for which 95% of the cumulative GSD is finer), we find that the median deposit size (D50) tends to better predict mean flow hydraulics. Two key factors influencing how flow characteristics control deposit GSDs are (a) dispersion caused by turbulence and (b) the transport distance required for suspension and settling to effectively sort grains. Deposits proximal to sediment sources primarily reflect the source GSD, while deposits farther from the source preferentially represent transport-dependent sorting. In our experimental data, transport distances longer than 1–2 advection length scales are required for the deposit GSD to reasonably predict flow depths and velocities. These results suggest ways that event deposits can be used to more accurately assess coastal risks from tsunamis and storm waves.