Joel S. Scheingross

A model for fire‐induced sediment yield by dry ravel in steep landscapes

Sediment flux from hillslopes to channels commonly increases following wildfires, with implications for the carbon cycle, river habitats, and debris-flow hazards. Although much of this material is transported via dry ravel, existing ravel models are not applicable to hillslopes with gradients greater than the angle of repose, which can constitute the majority of mountainous terrain. To fill this knowledge gap, we develop a continuity model for sediment storage by vegetation dams on steep hillslopes to predict sediment yields following wildfire. The maximum volume of sediment stored prior to wildfire is set to be a function of vegetation density, the capacity of plants to impound sediment, and the contributing hillslope area. Time is required after fire to establish vegetation and replenish hillslope sediment storage, which introduces vegetation regrowth rate, soil production rate, and fire recurrence interval as important variables that affect ravel yield. Model results for the San Gabriel Mountains, California, predict that sediment yield can increase by several orders of magnitude following fire. These results are consistent with field data of ravel yield (~30 mm per contributing area of hillslope in 5 months) we collected following the 2009 Station Fire, as well as postfire sediment flux recorded by 93 debris basins. In contrast to previous work, our model shows that heightened postfire sediment yields can be explained by a change in hillslope sediment storage independent of major changes in the soil production rate and landscape form over geomorphic timescales.

Experimental evidence for fluvial bedrock incision by suspended and bedload sediment

Fluvial bedrock incision sets the pace of landscape evolution and can be dominated by abrasion from impacting particles. Existing bedrock incision models diverge on the ability of sediment to erode within the suspension regime, leading to competing predictions of lowland river erosion rates, knickpoint formation and evolution, and the transient response of orogens to external forcing. We present controlled abrasion mill experiments designed to test fluvial incision models in the bedload and suspension regimes by varying sediment size while holding fixed hydraulics, sediment load, and substrate strength. Measurable erosion occurred within the suspension regime, and erosion rates agree with a mechanistic incision theory for erosion by mixed suspended and bedload sediment. Our experimental results indicate that suspension-regime erosion can dominate channel incision during large floods and in steep channels, with significant implications for the pace of landscape evolution.

Knickpoint formation, rapid propagation, and landscape response following coastal cliff retreat at the last interglacial sea-level highstand: Kaua'i, Hawai'i

Upstream knickpoint propagation is an important mechanism for channel incision, and it communicates changes in climate, sea level, and tectonics throughout a landscape. Few studies have directly measured the long-term rate of knickpoint retreat, however, and the mechanisms for knickpoint initiation are debated. Here, we use cosmogenic ^3He exposure dating to document the retreat rate of a waterfall in Ka’ula’ula Valley, Kaua‘i, Hawai‘i, an often-used site for knickpoint-erosion modeling. Cosmogenic exposure ages of abandoned surfaces are oldest near the coast (120 ka) and systematically decrease with upstream distance toward the waterfall (<10 ka), suggesting that the waterfall migrated nearly 4 km over the past 120 k.y. at an average rate of 33 mm/yr. Upstream of the knickpoint, cosmogenic nuclide concentrations in the channel are approximately uniform and indicate steady-state vertical erosion at a rate of ∼0.03 mm/yr. Field observations and topographic analysis suggest that waterfall retreat is dominated by block toppling, with sediment transport below the waterfall actively occurring by debris flows. Knickpoint initiation was previously attributed to a submarine landslide ca. 4 Ma; however, our dating results, bathymetric analysis, and landscape-evolution modeling support knickpoint generation by wave-induced sea-cliff erosion during the last interglacial sea-level highstand ca. 120–130 ka. We illustrate that knickpoint generation during sea-level highstands, as opposed to the typical case of sea-level fall, is an important relief-generating mechanism on stable or subsiding steep coasts, and likely drives transient pulses of significant sediment flux.

Fault-zone controls on the spatial distribution of slow-moving landslides

Slow-moving landslides (earthflows) can dominate hillslope sediment flux and landscape erosion in hilly terrain with mechanically weak, fine-grained rock. Controls on the occurrence of slow-moving landslides are poorly constrained and need to be understood for landscape evolution models, sediment budgets, and infrastructure and hazards planning. Here, we use airborne interferometric synthetic aperture radar (InSAR) and aerial photographs to document 150 previously unidentified active earthflows along the central, creeping portion of the San Andreas fault, California. The earthflows move seasonally in response to winter rainfall, occur on hillslopes at ∼20%–40% gradients (less than typically associated with rapid, catastrophic landslides), and have similar morphological characteristics to earthflows in different climatic and tectonic settings. Although our data extend up to 10 km from the fault trace, ∼75% of detected landslides occur within 2 km of the active fault. Topographic, precipitation, and rock type metrics alone are not enough to explain the observed spatial distribution of earthflows. Instead, we hypothesize that earthflows cluster near the creeping San Andreas fault because of a fault-induced zone of reduced bulk-rock strength that increases hillslope susceptibility to failure. In addition, similar lithology, topography, and climate exist north of the creeping section of the fault, yet earthflows there are rare. This may be due to large-magnitude earthquakes episodically triggering coseismic rapid landslides, which preferentially remove weak rock from the fault damage zone. Our analysis suggests that the necessary conditions for earthflow formation in central California include some combination of reduced rock strength, fine-grained sedimentary rock, threshold precipitation and relief, and possibly the absence of large-magnitude earthquakes. These conditions likely hold for earthflow development in other areas, and our work suggests that local variations in rock strength and seismicity, such as those associated with fault zones, need to be taken into account in order to accurately predict earthflow occurrence.

Sediment transport through self-adjusting, bedrock-walled waterfall plunge pools

Many waterfalls have deep plunge pools that are often partially or fully filled with sediment. Sediment fill may control plunge-pool bedrock erosion rates, partially determine habitat availability for aquatic organisms, and affect sediment routing and debris flow initiation. Currently, there exists no mechanistic model to describe sediment transport through waterfall plunge pools. Here we develop an analytical model to predict steady-state plunge-pool depth and sediment-transport capacity by combining existing jet theory with sediment transport mechanics. Our model predicts plunge-pool sediment-transport capacity increases with increasing river discharge, flow velocity, and waterfall drop height and decreases with increasing plunge-pool depth, radius, and grain size. We tested the model using flume experiments under varying waterfall and plunge-pool geometries, flow hydraulics, and sediment size. The model and experiments show that through morphodynamic feedbacks, plunge pools aggrade to reach shallower equilibrium pool depths in response to increases in imposed sediment supply. Our theory for steady-state pool depth matches the experiments with an R^2 value of 0.8, with discrepancies likely due to model simplifications of the hydraulics and sediment transport. Analysis of 75 waterfalls suggests that the water depths in natural plunge pools are strongly influenced by upstream sediment supply, and our model provides a mass-conserving framework to predict sediment and water storage in waterfall plunge pools for sediment routing, habitat assessment, and bedrock erosion modeling.

Self-formed waterfall plunge pools in homogeneous rock

Waterfalls are ubiquitous, and their upstream propagation can set the pace of landscape evolution, yet no experimental studies have examined waterfall plunge pool erosion in homogeneous rock. We performed laboratory experiments, using synthetic foam as a bedrock simulant, to produce self-formed waterfall plunge pools via particle impact abrasion. Plunge pool vertical incision exceeded lateral erosion by approximately tenfold until pools deepened to the point that the supplied sediment could not be evacuated and deposition armored the pool bedrock floor. Lateral erosion of plunge pool sidewalls continued after sediment deposition, but primarily at the downstream pool wall, which might lead to undermining of the plunge pool lip, sediment evacuation, and continued vertical pool floor incision in natural streams. Undercutting of the upstream pool wall was absent, and our results suggest that vertical drilling of successive plunge pools is a more efficient waterfall retreat mechanism than the classic model of headwall undercutting and collapse in homogeneous rock.

A Mechanistic Model of Waterfall Plunge Pool Erosion into Bedrock

Landscapes often respond to changes in climate and tectonics through the formation and upstream propagation of knickzones composed of waterfalls. Little work has been done on the mechanics of waterfall erosion, and instead most landscape-scale models neglect waterfalls or use rules for river erosion, such as stream power, that may not be applicable to waterfalls. Here we develop a physically based model to predict waterfall plunge pool erosion into rock by abrasion from particle impacts and test the model against flume experiments. Both the model and experiments show that evolving plunge pools have initially high vertical erosion rates due to energetic particle impacts, and erosion slows and eventually ceases as pools deepen and deposition protects the pool floor from further erosion. Lateral erosion can continue after deposition on the pool floor, but it occurs at slow rates that become negligible as pools widen. Our work points to the importance of vertical drilling of successive plunge pools to drive upstream knickzone propagation in homogenous rock, rather than the classic mechanism of headwall undercutting. For a series of vertically drilling waterfalls, we find that upstream knickzone propagation is faster under higher combined water and sediment fluxes and for knickzones composed of many waterfalls that are closely spaced. Our model differs significantly from stream-power-based erosion rules in that steeper knickzones can retreat faster or more slowly depending on the number and spacing of waterfalls within a knickzone, which has implications for interpreting climatic and tectonic history through analysis of river longitudinal profiles.