Laws of erosion and sediment transport
What laws govern erosion and sediment transport at Earth’s surface, and how much does sediment’s dual role shape landscapes?
Summary
Why it is open
The most widely used model of bedrock river incision is the stream power law [14], which Dimitri Lague calls "a cornerstone of quantitative geomorphology": the incision rate is the product of drainage area and channel slope, each raised to a power (the exponents m and n) [12], times an erodibility coefficient [14]. It is used to infer tectonic deformation from river profiles and to model landscape evolution, but tests have given "mixed results" [12]. In 1999 Kelin Whipple and Gregory Tucker showed that the model's behaviour depends on how nonlinear the link between incision rate and slope is, and that the slope exponent n "emerges as a fundamental unknown" [4].
The second difficulty is sediment. Leonard Sklar and William Dietrich wrote in 2001 that recent theory made incision depend nonlinearly on sediment supply, "challenging the common assumption that incision rate is simply proportional to stream power" [5]. The sediment load supplies the tools that abrade bedrock (the tools effect) and covers and protects the bed (the cover effect) [8]. The transport laws are uncertain too. In a 2020 review Christophe Ancey wrote that scientists "are unable to predict qs to within better than one order of magnitude", where qs is the mean rate of bedload transport, the sediment moved along the river bed [15].
The problem is therefore several questions: which incision law applies, and over what range; how sediment supply and grain size enter it; and how to predict sediment transport. No body sets a bar for a resolution. The sources describe what a test would need. In 2002 Whipple and Tucker found that steady-state topography may not distinguish the competing incision models and that "other methods must be developed to test models" [6]. Lague concluded that the stream power model has "a narrow range of validity", and that going beyond it needs an explicit upscaling of sediment flux and of threshold and flood-variability effects, combined with a channel width that can adjust [12].
What is known
The stream power law (1983–1999). Alan Howard and Gordon Kerby followed badlands that formed after vegetation and soil were stripped from a 13-hectare site in Virginia. In the bedrock channels, the erosion rate was proportional to the 4/9 power of drainage area and the 2/3 power of gradient, consistent with erosion proportional to bed shear stress during high flows [1]. Whipple and Tucker showed that the model's response time to a change in rock uplift depends on climate, rock strength and the size of the change, with n setting how strongly [4].
Tools and cover (2001–2008). Sklar and Dietrich's measurements in laboratory abrasion mills supported the hypothesis that sediment promotes erosion at low supply by providing tools and inhibits it at high supply by burying the bedrock; erosion peaked at a critical supply of coarse sediment, where the bed was only partly exposed [5]. Fine grains made poor tools because they travel in suspension [5]. Their 2004 saltation–abrasion model makes wear depend on the flux of impact energy and on the uncovered fraction of the bed, set by the ratio of coarse sediment supply to transport capacity; it predicts maximum wear at moderate relative supply and intermediate excess shear stress, and laboratory measurements "agree well" with it [7]. Jens Turowski, Lague and Niels Hovius derived a cover term in which the exposed bed area falls exponentially with sediment flux, which predicts higher erosion rates than earlier models [8]. Michael Lamb, Dietrich and Sklar added grains carried in suspension: in their model erosion keeps rising with flow strength even beyond the onset of suspension, and erosion can occur when the supply of sediment exceeds the bedload transport capacity [10].
Field tests (2008–2015). Patience Cowie and colleagues studied rivers that cross active normal faults in the same rock type, with similar climate and tectonic setting, with relative uplift rates that differ by a factor of four [9]. Peak bed shear stress showed no simple relation to uplift, but the long-term sediment supply could explain the data if bedload modulates bedrock incision "in a strongly nonlinear way"; they describe this as the first evaluation in natural systems of theoretical predictions of sediment's role [9]. A. R. Beer and Turowski installed an artificial bedrock slab in a steep mountain stream and recorded hydraulics, bedload and erosion during one transport event [13]. The tools effect dominated, and they proposed a local model in which erosion rate is proportional to bedload transport rate; derivatives of the stream power model still reproduced the overall trend of that event [13].
Bedload transport (1989–2020). Basil Gomez and Michael Church tested 12 bedload formulae for gravel-bed rivers against four river and three flume datasets covering eight orders of magnitude in transport rate, and found that "no formula performs consistently well" [2]. Ancey's review describes experimental scaling laws that relate the mean transport rate to discharge, shear stress or stream power, and newer approaches that treat bedload as the result of random grain motion [15].
Calibration with cosmogenic nuclides (1996–2016). Darryl Granger, James Kirchner and Robert Finkel showed that concentrations of cosmogenic nuclides (isotopes produced by cosmic rays in rock near the surface) in stream sediment give catchment-averaged erosion rates; at two catchments these agreed within one standard error with rates from alluvial fan deposition over the past 16,000 years [3]. Roman DiBiase and Whipple combined such rates from the San Gabriel Mountains of California with streamflow records: a simple stochastic-threshold model, in which bedrock is eroded only in floods large enough to pass a threshold and large floods follow an inverse power law, explained the relation between channel steepness and erosion rate; details of incision mechanics, sediment effects, width adjustment and debris flows did not significantly influence that relation [11]. M.-A. Harel, S. M. Mudd and M. Attal compiled 1,457 basin-averaged erosion rates from beryllium-10 and estimated the stream power parameters for 59 areas [14]. The slope exponent was "generally > 1", which they take to support threshold-controlled incision, and local trends such as a correlation between erodibility and precipitation did not appear at a global scale [14].
Main approaches
Stream power with thresholds. Keep the stream power form but add an erosion threshold and the variability of floods [12]. Strongest evidence: the stochastic-threshold model explains steepness and erosion rates in the San Gabriel Mountains [11]; the global compilation finds a slope exponent generally above 1 [14]; and Lague finds that all published datasets on incising rivers away from knickpoints (steps in the river profile) are dominated by threshold effects [12]. Main difficulty: the threshold-stochastic model fails to predict how slope scales with incision rate where channel width decreases as incision rate rises, and recently proposed models of dynamic width cannot resolve this [12].
Tools and cover. Make incision depend explicitly on sediment supply, grain size and bed cover [5, 7, 8]. Strongest evidence: the abrasion-mill experiments [5, 7], the fault-crossing rivers [9] and the instrumented slab [13]. Main difficulty: the predicted slope–area relation does not deviate enough from a straight line to discriminate between models using empirical data [8], and the models disagree on how erosion responds to strong flows: the saltation–abrasion model predicts a maximum at intermediate excess shear stress [7], while the total-load model keeps rising beyond the onset of suspension [10].
Transport-limited and hybrid channels. Treat some rivers as limited by their capacity to carry sediment rather than by their ability to detach rock, and allow channels between the two [6]. Whipple and Tucker predicted that such hybrid channels behave as transport-limited when rock uplift slows, especially during post-orogenic topographic decline [6], which ties this question to the time-scale of topographic decay. Strongest evidence: the apparent ubiquity of mixed bedrock–alluvial channels in active mountain belts [6]. Main difficulty: preliminary data suggest that "no such diagnostic differences may exist" in steady-state topography between the models [6].
Stochastic bedload transport. Treat bedload as the collective outcome of random grain motions rather than a one-to-one function of flow [15]. Strongest evidence: deterministic formulae have not predicted transport reliably, in the 1989 test [2] or since [15]. Main difficulty: Gomez and Church wrote that limitations of the test data, the constraints of a realistic test and the physics of transport "all may be adduced" for the poor performance [2], and Ancey presents the stochastic approaches as ideas still being explored [15].
Open sub-questions
- Where n exceeds 1. For the 59 areas of Harel et al., does a stochastic-threshold model driven by local discharge records predict the fitted slope exponent, as it did for the San Gabriel Mountains [11, 14]?
- The peak of the tools effect. On instrumented bedrock beds, at what ratio of sediment supply to transport capacity does erosion peak, and which cover term fits the exposed fraction [7, 8, 13]?
- Suspended-load tools. In steep channels with fine sediment, does bedrock erosion keep rising with flow strength beyond the onset of suspension, or peak at intermediate excess shear stress [7, 10]?
- A transient test. Which measurable response to a change in uplift rate, in rivers like the fault-crossing set of Cowie et al., separates detachment-limited from sediment-flux-dependent incision [6, 9]?
- Sources of bedload scatter. In the river datasets of the 1989 test, how much of the misfit remains once a grain-scale stochastic model is used [2, 15]?
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Howard, A. D. & Kerby, G. (1983). Channel changes in badlands. Geological Society of America Bulletin 94(6), 739–752. 10.1130/0016-7606(1983)94<739:CCIB>2.0.CO;2 · Checked 16 Oct 2026
- Gomez, B. & Church, M. (1989). An assessment of bed load sediment transport formulae for gravel bed rivers. Water Resources Research 25(6), 1161–1186. 10.1029/WR025i006p01161 · Checked 16 Oct 2026
- Granger, D. E., Kirchner, J. W. & Finkel, R. (1996). Spatially averaged long-term erosion rates measured from in situ-produced cosmogenic nuclides in alluvial sediment. The Journal of Geology 104(3), 249–257. 10.1086/629823 · Checked 16 Oct 2026
- Whipple, K. X. & Tucker, G. E. (1999). Dynamics of the stream-power river incision model: implications for height limits of mountain ranges, landscape response timescales, and research needs. Journal of Geophysical Research: Solid Earth 104(B8), 17661–17674. 10.1029/1999JB900120 · Checked 16 Oct 2026
- Sklar, L. S. & Dietrich, W. E. (2001). Sediment and rock strength controls on river incision into bedrock. Geology 29(12), 1087–1090. 10.1130/0091-7613(2001)029<1087:SARSCO>2.0.CO;2 · Checked 16 Oct 2026
- Whipple, K. X. & Tucker, G. E. (2002). Implications of sediment-flux-dependent river incision models for landscape evolution. Journal of Geophysical Research: Solid Earth 107(B2). 10.1029/2000JB000044 · Checked 16 Oct 2026
- Sklar, L. S. & Dietrich, W. E. (2004). A mechanistic model for river incision into bedrock by saltating bed load. Water Resources Research 40(6). 10.1029/2003WR002496 · Checked 16 Oct 2026
- Turowski, J. M., Lague, D. & Hovius, N. (2007). Cover effect in bedrock abrasion: a new derivation and its implications for the modeling of bedrock channel morphology. Journal of Geophysical Research: Earth Surface 112(F4). 10.1029/2006JF000697 · Checked 16 Oct 2026
- Cowie, P. A., Whittaker, A. C., Attal, M., Roberts, G., Tucker, G. E. & Ganas, A. (2008). New constraints on sediment-flux-dependent river incision: implications for extracting tectonic signals from river profiles. Geology 36(7), 535. 10.1130/G24681A.1 · Checked 16 Oct 2026
- Lamb, M. P., Dietrich, W. E. & Sklar, L. S. (2008). A model for fluvial bedrock incision by impacting suspended and bed load sediment. Journal of Geophysical Research: Earth Surface 113(F3). 10.1029/2007JF000915 · Checked 16 Oct 2026
- DiBiase, R. A. & Whipple, K. X. (2011). The influence of erosion thresholds and runoff variability on the relationships among topography, climate, and erosion rate. Journal of Geophysical Research (Earth Surface) 116(F4), F04036. 10.1029/2011JF002095 · Checked 16 Oct 2026
- Lague, D. (2014). The stream power river incision model: evidence, theory and beyond. Earth Surface Processes and Landforms 39(1), 38–61 (published online 2013). 10.1002/esp.3462 · Checked 16 Oct 2026
- Beer, A. R. & Turowski, J. M. (2015). Bedload transport controls bedrock erosion under sediment-starved conditions. Earth Surface Dynamics 3(3), 291–309. 10.5194/esurf-3-291-2015 · Checked 16 Oct 2026
- Harel, M.-A., Mudd, S. M. & Attal, M. (2016). Global analysis of the stream power law parameters based on worldwide 10Be denudation rates. Geomorphology 268, 184–196. 10.1016/j.geomorph.2016.05.035 · Checked 16 Oct 2026
- Ancey, C. (2020). Bedload transport: a walk between randomness and determinism. Part 1. The state of the art. Journal of Hydraulic Research 58(1), 1–17. 10.1080/00221686.2019.1702594 · Checked 16 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in geoscience. Checked 08 Oct 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.