A one‐dimensional model for simulating armouring and erosion on hillslopes: 1. model development and event‐scale dynamics

This paper presents an erosion model, ARMOUR, which simulates time‐varying runoff, erosion, deposition and surface armour evolution down a hillslope either as a result of a single erosion event or as the cumulative impact of many events over periods up to decades. ARMOUR simulates sediment transport for both cohesive and non‐cohesive soil and dynamically differentiates between ‘transport‐limited’ and ‘source‐limited’ processes. A variety of feasible processes for entrainment of different size classes can be modelled and evaluated against data. The generalized likelihood of uncertainty estimation (GLUE) technique was used to calibrate and validate ARMOUR using data collected during rainfall simulator experiments at two contrasting sites: (1) non‐cohesive stony sediments at Ranger Uranium Mine, Northern Territory, Australia; and (2) cohesive silty sediments at Northparkes Gold Mine, NSW, Australia. The spatial and temporal variations of model predictions within the individual runoff events showed that some entrainment processes could not model the spikes in concentration and subsequent depletion, while the hiding model of Andrews and Parker best simulated the concentration trends for both calibrated and independent runoff events. ARMOUR also successfully captured the coarsening of the surface material, though small, over the duration of the rainfall simulator trials. This was driven by the depletion of the finest size class of the soil. For a constant discharge, ARMOUR simulated higher sediment flux at the start of the storm with the sediment flux and concentration diminishing with time. For natural rainfall a power law relationship between sediment flux and discharge was observed. The calibration exercise showed that sediment concentration and discharge alone are insufficient to calibrate all aspects of the physics, in particular the armour depth. This appears to be because the armouring during the short duration events is driven by depletion of the finest classes of the sediments (diameters less then 62·5 mm), which are not normally measured. Copyright © 2006 John Wiley & Sons, Ltd.     

Influence of bed clusters and size gradation on operational time distribution for non‐uniform bed‐load transport

The operational time distribution (OTD) defines the time for bed‐load sediment spent in motion, which is needed to characterize the random nature of sediment transport. This study explores the influence of bed clusters and size gradation on OTD for non‐uniform bed‐loads. First, both static and mobile bed armouring experiments were conducted in laboratorial flumes to monitor the transport of mixed sand/gravel sediments. Only in the mobile armouring experiment did apparent bed clusters develop, because of stable feeding and a longer transport period. Second, a generalized subordinated advection (GSA) model was applied to quantify the observed dynamics of tracer particles. Results show that for the static armour layer (without sediment feed), the best‐fit OTD assigns more weight to the large displacement of small particles, likely because of the size‐selective entrainment process. The capacity coefficient in the GSA model, which affects the width of the OTD, is space dependent only for small particles whose dynamics can be significantly affected by larger particles and whose distribution is more likely to be space dependent in a mixed sand and gravel system. However, the OTD for the mobile armour layer (with sediment recirculation) exhibited longer tails for larger particles. This is because the trailing edge of larger particles is more resistant to erosion, and their leading front may not be easily trapped by self‐organized bed clusters. The strong interaction between particle–bed may cause the capacity coefficient to be space‐dependent for bed‐load transport along mobile armour layers. Therefore, the combined laboratory experiments and stochastic model analysis show that the OTD may be affected more by particle–bed interactions (such as clusters) than by particle–particle interactions (e.g. hiding and exposing), and that the GSA model can quantify mixed‐size sand/gravel transport along river beds within either static or mobile armour layers. Copyright © 2016 John Wiley & Sons, Ltd.     

The effects of kinetic sorting on sediment mobility on steep slopes

In poorly mobile static armour, sorting is usually considered the result of hiding/exposure effects. We called this effect ‘static sorting’ in opposition to very efficient grain‐to‐grain mechanisms produced by a mobile mixture, called ‘kinetic sorting’. We hypothesized that kinetic sorting can be an important contributor to the morphodynamics of mountain streams and attempted to demonstrate this with new flume experiments. Two long runs were produced with natural poorly sorted sediments, and with transport stages of the coarse fraction (defined by the ratio between the shear stress and the critical shear stress for transport), smaller and higher than 1, respectively. Both runs produced an efficient transfer downstream of the injected material, but with a major difference: the first run (no kinetic sorting) produced permanent armour figuring clusters, akin to what has already been observed in similar experiments; the second run (with kinetic sorting) also produced bed armouring, but this armour was periodically totally destroyed, leading to substantial bed erosion. This phenomenon was explained by kinetic sorting, the effects of which are to produce an efficient downward migration of fine materials and bed surface armouring. The consequence is that fine materials are hidden to the flow during aggradation, allowing the slope to attain values much steeper than would have been expected at equilibrium for the mixture. However, whereas the surface armouring tends to stabilize the bed, construction of a layer of fine sediments at the subsurface also contributes to making it very unstable. These two contradictory effects explain the complex bed behaviours and the existence of very large bedload and slope fluctuations. Copyright © 2014 John Wiley & Sons, Ltd.     

Surface particle sizes on armoured gravel streambeds: effects of supply and hydraulics

Most gravel‐bed streams exhibit a surface armour in which the median grain size of the surface particles is coarser than that of the subsurface particles. This armour has been interpreted to result when the supply of sediment is less than the ability of the stream to move sediment. While there may be certain sizes in the bed for which the supply is less than the ability of the stream to transport these sizes, for other sizes of particles the supply may match or even exceed the ability of the channel to transport these particles. These sizes of particles are called ‘supply‐limited’ and ‘hydraulically limited’ in their transport, respectively, and can be differentiated in dimensionless sediment transport rating curves by size fractions. The supply‐ and hydraulically limited sizes can be distinguished also by comparing the size of particles of the surface and subsurface. Those sizes that are supply‐limited are winnowed from the bed and are under‐represented in the surface layer. Progressive truncation of the surface and subsurface size distributions from the ?ne end and recalculation until the size distributions are similar (collapse), establishes the break between supply‐ and hydraulically limited sizes. At sites along 12 streams in Idaho ranging in drainage area from about 100 to 4900 km², sediment transport rating curves by size class and surface and subsurface size distributions were examined. The break between sizes that were supply‐ and hydraulically limited as determined by examination of the transport rate and surface and subsurface size distributions was similar. The collapse size as described by its percentile in the cumulative size distribution averaged D₃₆ of the surface and D₇₃ of the subsurface. The discharge at which the collapse size began to move averaged 88 per cent of bankfull discharge. The collapse size decreased as bed load yield increased and increased with the degree of selective transport. Copyright © 2003 John Wiley & Sons, Ltd.     

Suspended sediment transport regime in a debris‐flow gully on Vancouver Island,British Columbia

In debris‐flow‐prone channels, normal fluvial sediment transport occurs (nearly exclusively in suspended mode) between episodic debris‐flow events. Observations of suspended sediment transport through a winter season in a steepland gully in logged terrain revealed two event types. When flows exceeded a threshold of 270 l s⁻¹, events yielded significant quantities of sediment and suspended sediment concentration increased with flow. Smaller events were strongly ‘supply limited’; sediment concentration decreased as flow increased. Overall, there is no consistent correlation between runoff and sediment yield. Within the season, three subseasons were identified (demarcated by periods of freezing weather) within which a pattern of fine sediment replenishment and evacuation occurred. Finally, a signature of fine sediment mobilization and exhaustion was observed within individual events. Fine sediment transport occurred in discrete pulses within storm periods, most of the yield occurring within 5 to 15% of storm runoff duration, so that it is unlikely that scheduled sampling programs would identify significant transport. Significant events are, however, generally forecastable on the basis of regional heavy rainfall warnings, providing a basis for targeted observations. Radiative snowmelt events and rain‐on‐snow remain difficult to forecast, since the projection of temperatures from the nearest regular weather station yields variable results. Copyright © 2004 John Wiley & Sons, Ltd.     

Interrill soil erosion processes and their interaction on low slopes

Soil erosion by water is mostly the result of rainfall‐driven and runoff‐driven processes taking place simultaneously during a storm event. However, the effect of interaction between these two erosion processes has received limited attention. Most laboratory experiments indicate that the rate of erosion in a rain‐impacted flow is greater than for un‐impacted flows of similar depth and velocity; however, negative interaction between the two processes has also been reported. There is no provision for any such interaction in any of the current erosion models. This paper reports on the results of a number of exact experiments on three soil types carried out in the flume of Griffith University's large rainfall simulator to study interaction between rain and runoff processes. The results show that interaction is generally positive under approximately steady state condition and there is very limited sign of negative interaction reported by others. Results provide strong evidence that raindrops continuously peel fine sediment from larger stable aggregates. This mechanism could be the reason for positive interaction during simultaneous rainfall and flow driven erosion in well aggregated soils as a result of increased fine particles in the eroded sediment. Strong positive interaction between rain and runoff erosion also occurs for medium to large aggregates. This strongly suggests that mechanisms that are not well understood are operational. It is quite possible that particle movement can be stimulated by rolling or creeping in a size‐selective manner. Indeed, such additional mechanisms may well be largely responsible for the positive interaction observed between rain and surface flow. Copyright © 2006 John Wiley & Sons, Ltd.     

Evaluating a new landform evolution model: A case study using a proposed mine rehabilitation landform

Landform evolution models are powerful tools for determining long-term erosional stability and denudation rates spanning geological timescales. SIBERIA, CAESAR and CHILD are examples of these model. The newly developed State Space Soil Production and Assessment Model (SSSPAM) coupled soilscape-landform evolution model has the ability to assess overall erosion rates of catchment scale landforms either using short-term precipitation events, variable precipitation or time-averaged precipitation (annual average). In addition, SSSPAM has the capability of developing the subsurface soil profile through weathering and armouring. In SSSPAM, physical processes of pedogenesis such as erosion and armouring, diffusion, sediment deposition and weathering are modelled using a state space matrix approach. In this article we simulate the short-term evolution (100 years) of a proposed post-mining landform using both SIBERIA and SSSPAM and compare the erosion and sediment output results. For the short-term simulations SSSPAM's armouring capability was disabled. The models were then used to simulate the evolution of the catchment for 10,000 years. Results demonstrate that the short-term SSSPAM simulation results compare well with the results from the established landform evolution model SIBERIA. The long-term armouring disabled SSSPAM simulations produces simulated erosion rates comparable with SIBERIA simulations both of which are similar to upper limit of field measured denudation rates. The SSSPAM simulation using armouring demonstrated that armouring reduced the erosion rate of the catchment by a factor of 4 which is comparable with the lower limit of field measured denudation rates. This observation emphasizes the importance of armouring in long-term evolution of landforms. Soil profile cross-sections developed from the same results show that SSSPAM can also reproduce subsurface soil evolution and stratification and spatial variability of soil profile characteristics typically observed in the field.     

Sediment yields from a laboratory catchment and their relationship to rilling and surface armouring

The development and testing of sediment simulation models require continuous monitoring of erosion processes and sediment yields from catchment areas at a wide range of scales. A series of experiments are described in which runoff and sediment yields from a small laboratory catchment were monitored through six consecutive storms applied to each of three soil types. Slope microtopography and the surface particle-size distribution were surveyed between storms. Pronounced peaks in sediment concentration at the start of each storm were not observed for these conditions, but significant variation in yield through a series of storms was shown to result from the interaction of rilling and armouring processes as the source of sediment shifted from the rills to interrill areas. In view of the experimental findings the validity of experiments reporting average or ‘stable’ erosion rates is questioned. The need for dynamic models capable of simulating rill development and changes in sediment availability is emphasized.     

Reintroduced large wood modifies fine sediment transport and storage in a lowland river channel

This paper explores changes in suspended sediment transport and fine sediment storage at the reach and patch scale associated with the reintroduction of partial large wood (LW) jams in an artificially over‐widened lowland river. The field site incorporates two adjacent reaches: a downstream section where LW jams were reintroduced in 2010 and a reach immediately upstream where no LW was introduced. LW pieces were organized into ‘partial’ jams incorporating several ‘key pieces’ which were later colonized by substantial stands of aquatic and wetland plants. Reach‐scale suspended sediment transport was investigated using arrays of time‐integrated suspended sediment samplers. Patch‐scale suspended sediment transport was explored experimentally using turbidity sensors to track the magnitude and velocity of artificially generated sediment plumes. Fine sediment storage was quantified at both reach and patch scales by repeat surveys of fine sediment depth. The results show that partial LW jams influence fine sediment dynamics at both the patch and reach scale. At the patch‐scale, introduction of LW led to a reduction in the concentration and increase in the time lag of released sediment plumes within the LW, indicating increased diffusion of plumes. This contrasted with higher concentrations and lower time lags in areas adjacent to the LW; indicating more effective advection processes. This led to increased fine sediment storage within the LW compared with areas adjacent to the LW. At the reach‐scale there was a greater increase in fine sediment storage through time within the restored reach relative to the unrestored reach, although the changes in sediment transport responsible for this were not evident from time‐integrated suspended sediment data. The results of the study have been used to develop a conceptual model which may inform restoration design. Copyright © 2017 John Wiley & Sons, Ltd.     

Geomorphological evolution and sediment stratigraphy of numerically simulated alluvial fans

When a sediment laden river reaches a flat basin area the coarse fraction of their sediment load is deposited in a cone shaped structure called an alluvial fan. In this article we used the State Space Soil Production and Assessment Model (SSSPAM) coupled landform–soilscape evolution model to simulate the development of alluvial fans in two- and three-dimensional landforms. In SSSPAM the physical processes of erosion and armouring, soil weathering and sediment deposition were modelled using state-space matrices, in both two and three dimensions. The results of the two-dimensional fan showed that the fan grew vertically and laterally keeping a concave up long profile. It also showed a downstream fining of the sediments along the fan profile. Both of these observations are in agreement with available literature concerning natural and experimental fan formations. Simulations with the three-dimensional landform produced a fan with a semicircular shape with concave up long profiles and concave down cross profiles which is typical for fans found in nature and ones developed in laboratory conditions. During the simulation the main channel which brings sediment to the fan structure changed its position constantly leading to the semicircular shape of the fan. This behaviour is similar to the autogenic process of ‘fanhead trenching’ which is the major mechanism of sediment redistribution while the fan is developing. The three-dimensional fan simulation also exhibited the downstream fining of sediments from the fan apex to the peripheries. Further, the simulated fan also developed complex internal sediment stratification which is modelled by SSSPAM. Currently such complex sediment stratification is thought to be a result of allogenic processes. However, this simulation shows that, such complex internal sediment structures can develop through autogenic processes as well. © 2020 John Wiley & Sons, Ltd.     

Weathering and erosion of fractured bedrock systems

We explore the contribution of fractures (joints) in controlling the rate of weathering advance for a low‐porosity rock by using methods of homogenization to create averaged weathering equations. The rate of advance of the weathering front can be expressed as the same rate observed in non‐fractured media (or in an individual block) divided by the volume fraction of non‐fractured blocks in the fractured parent material. In the model, the parent has fractures that are filled with a more porous material that contains only inert or completely weathered material. The low‐porosity rock weathers by reaction‐transport processes. As observed in field systems, the model shows that the weathering advance rate is greater for the fractured as compared to the analogous non‐fractured system because the volume fraction of blocks is < 1. The increase in advance rate is attributed both to the increase in weathered material that accompanies higher fracture density, and to the increase in exposure of surface of low‐porosity rock to reaction‐transport. For constant fracture aperture, the weathering advance rate increases when the fracture spacing decreases. Equations describing weathering advance rate are summarized in the ‘List of selected equations’. If erosion is imposed at a constant rate, the weathering systems with fracture‐bounded bedrock blocks attain a steady state. In the erosional transport‐limited regime, bedrock blocks no longer emerge at the air‐regolith boundary because they weather away. In the weathering‐limited (or kinetic) regime, blocks of various size become exhumed at the surface and the average size of these exposed blocks increases with the erosion rate. For convex hillslopes, the block size exposed at the surface increases downslope. This model can explain observations of exhumed rocks weathering in the Luquillo mountains of Puerto Rico. Published 2017. This article is a U.S. Government work and is in the public domain in the USA     

A mathematical model for steady‐state regolith production at constant erosion rate

It has been hypothesized that many soil profiles reach a steady‐state thickness. In this work, such profiles were simulated using a one‐dimensional model of reaction with advective and diffusive solute transport. A model ‘rock’ is considered, consisting of albite that weathers to kaolinite in the presence of chemically inert quartz. The model yields three different steady‐state regimes of weathering. At the lowest erosion rates, a local‐equilibrium regime is established where albite is completely depleted in the weathering zone. This regime is equivalent to the transport‐limited regime described in the literature. With an increase in erosion rate, transition and kinetic regimes are established. In the transition regime, both albite and kaolinite are present in the weathering zone, but albite does not persist to the soil–air interface. In the weathering‐limited regime, here called the kinetic regime, albite persists to the soil–air interface. The steady‐state thickness of regolith decreases with increasing erosion rate in the local equilibrium and transition regimes, but in the kinetic regime, this thickness is independent of erosion rate. Analytical expressions derived from the model are used to show that regolith production rates decrease exponentially with regolith thickness. The steady‐state regolith thickness increases with the Darcy velocity of the pore fluid, and in the local equilibrium regime may vary markedly with small variations in this velocity and erosion rate. In the weathering‐limited regime, the temperature dependences for chemical weathering rates are related to the activation energy for the rate constant for mineral reaction and to the ΔH of dissolution, while for local equilibrium regimes they are related to the ΔH only. The model illustrates how geochemical and geomorphological observations are related for a simple compositional system. The insights provided will be useful in interpreting natural regolith profiles. Copyright © 2010 John Wiley & Sons, Ltd.     

Modelling centennial sediment waves in an eroding landscape – catchment complexity

Sediment flux dynamics in fluvial systems have often been related to changes in external drivers of topography, climate or land cover. It is well known that these dynamics are non‐linear. Recently, model simulations of fluvial activity and landscape evolution have suggested that self‐organization in landscapes can also cause internal complexity in the sedimentary record. In this contribution one particular case of self‐organization is explored in the Sabinal field study area, Spain, where several dynamic zones of sedimentation and incision are observed along the current river bed. Whether these zones can be caused by internal complexity was tested with landscape evolution model (LEM) LAPSUS (Landscape Process Modelling at Multi‐dimensions and Scales). During various 500 year simulations, zones of sedimentation appear to move upstream and downstream in eroding river channels (‘waves’). These waves are visualized and characterized for a range of model settings under constant external forcing, and the self‐organizing process behind their occurrence is analysed. Results indicate that this process is not necessarily related to simplifications in the model and is more generic than the process of bed‐armouring that has recently been recognized as a cause for complexity in LEM simulations. We conclude that autogenic sediment waves are the result of the spatial propagation in time of feedbacks in local transport limited (deposition) and detachment limited (erosion) conditions. Copyright © 2014 John Wiley & Sons, Ltd.     

Analysis of bedload transport in laminar flows

Bedload transport generally depends on the bed shear stress and Reynolds number. Many studies conducted for the condition of turbulent flows have revealed the dependence of the transport rate on the bed shear stress, while knowledge of the Reynolds number effect on the transport rate is very limited. As an extreme case to reflect the viscous effect on sediment transport, sediment transport in laminar flows is considered in this paper. A stochastic approach is adopted to explore how the transport rate can be associated with characteristics of laminar flows. First, the probability of erosion in the absence of turbulence is assumed to depend only on the randomness of bed particles. The probability is then applied to formulate the sediment transport rate, of which the derivation is made largely based on Einstein’s bedload theory. The theoretical result indicates that the dimensionless transport rate for laminar flows is dependent on the dimensionless shear stress and dimensionless particle diameter or the shear Reynolds number. Comparisons are finally made between the derived formula and an empirical correlation available in the literature.     

Network‐scale dynamics of grain‐size sorting: implications for downstream fining,stream‐profile concavity,and drainage basin morphology

We explore the link between channel‐bed texture and river basin concavity in equilibrium catchments using a numerical landscape evolution model. Theory from homogeneous sediment transport predicts that river basin concavity directly increases with bed sediment size. If the effective grain size on a river bed governs its concavity, then natural phenomena such as grain‐size sorting and channel armouring should be linked to concavity. We examine this hypothesis by allowing the bed sediment texture to evolve in a transport‐limited regime using a two grain‐size mixture of sand and gravel. Downstream ?ning through selective particle erosion is produced in equilibrium. As the channel‐bed texture adjusts downstream so does the local slope. Our model predicts that it is not the texture of the original sediment mixture that governs basin concavity. Rather, concavity is linked to the texture of the sorted surface layer. Two different textural regimes are produced in the experiments: a transitional regime where the mobility of sand and gravel changes with channel‐bed texture, and a sand‐dominated region where the mobility of sand and gravel is constant. The concavity of these regions varies depending on the median gravel‐ or sand‐grain size, erosion rate, and precipitation rate. The results highlight the importance of adjustments in both surface texture and slope in natural rivers in response to changes in ?uvial and sediment inputs throughout a drainage network. This adjustment can only be captured numerically using multiple grain sizes or empirical downstream ?ning rules. Copyright © 2004 John Wiley & Sons, Ltd.     

Modelling water erosion in the Sahel: application of a physically based soil erosion model in a gentle sloping environment

Water is a major limiting factor in arid and semi‐arid agriculture. In the Sahelian zone of Africa, it is not always the limited amount of annual rainfall that constrains crop production, but rather the proportion of rainfall that enters the root zone and becomes plant‐available soil moisture. Maximizing the rain‐use efficiency and therefore limiting overland flow is an important issue for farmers. The objectives of this research were to model the processes of infiltration, runoff and subsequent erosion in a Sahelian environment and to study the spatial distribution of overland flow and soil erosion. The wide variety of existing water erosion models are not developed for the Sahel and so do not include the unique Sahelian processes. The topography of the Sahelian agricultural lands in northern Burkina Faso is such that field slopes are generally low (0–5°) and overland flow mostly occurs in the form of sheet flow, which may transport large amounts of fine, nutrient‐rich particles despite its low sediment transport capacity. Furthermore, pool formation in a field limits overland flow and causes resettlement of sediment resulting in the development of a surface crust. The EUROSEM model was rewritten in the dynamic modelling code of PCRaster and extended to account for the pool formation and crust development. The modelling results were calibrated with field data from the 2001 rainy season in the Katacheri catchment in northern Burkina Faso. It is concluded that the modified version of EUROSEM for the Sahel is a fully dynamic erosion model, able to simulate infiltration, runoff routing, pool formation, sediment transport, and erosion and deposition by inter‐rill processes over the land surface in individual storms at the scale of both runoff plots and fields. A good agreement is obtained between simulated and measured amounts of runoff and sediment discharge. Incorporating crust development during the event may enhance model performance, since the process has a large influence on infiltration capacity and sediment detachment in the Sahel. Copyright © 2005 John Wiley & Sons, Ltd.     

Development of longitudinal profiles of alluvial channels in response to base-level lowering

Degradation of alluvial channels in cohesive sediments was studied in 15 m and 20 m long flumes with a slope of 0°01 cm/cm. Degradation was initiated by lowering base level to a fixed position, and the development of the longitudinal profile of the channel is analysed through a model formulated as a heat (diffusion) equation. It is based on the equation of sediment continuity, combined with an assumption regarding sediment transport, namely that sediment discharge is linearly proportional to the channel slope. In accordance with the boundary and initial conditions imposed by the experimental setup and procedure, the basic equation is amenable to an analytical solution, which defines bed elevation at any distance and time, as a function of the amount of base-level lowering and a ‘diffusion’ coefficient. Additional problems arising from bank erosion and channel armouring are also treated successfully within the framework of the same model. The results show that in homogeneous alluvial sediments, not subject to armouring, the ultimate result of base-level lowering by a certain amount is degradation all along the channel by the same amount. The main impact of erosion is felt in the early stages after initiation of the process, and mainly near the mouth. The rate of degradation at any station along the channel reaches a peak and then slowly decreases with time, and the peak rate is attenuated with distance from the outlet. The model permits the prediction of intermediate stages of profile development at any distance from the outlet and at different times.     

A one‐dimensional model for simulating armouring and erosion on hillslopes: 2. Long term erosion and armouring predictions for two contrasting mine spoils

This paper investigates the dynamics of soil armouring as a result of fluvial erosion for a non‐cohesive sandy gravel spoil from the Ranger Mine, Australia, and a cohesive silt loam spoil from the Northparkes Mine, Australia, using a model for hillslope soil armouring. These long term predictions concentrate on the temporal and spatial changes of the spoil grading and erosion over 100–200 years for the flat cap regions (1–2%) and steep batter edges (10–30%) typically encountered on waste rock dumps. The existence of a significant rock fragment fraction in the Ranger spoil means that it armours readily, while Northparkes does not. For Ranger the waste rock showed reductions in (1) cumulative erosion of up to 81% from that obtained by extrapolating the initial erosion rate out 100 years and (2) the erosion/year by more than 10‐fold. For Northparkes reductions were less marked, with the maximum reduction in erosion/year being 37% after 200 years. For Ranger the reductions were greatest and fastest for intermediate gradient hillslopes. For the steepest hillslopes the armouring decreased because the flow shear stresses were large enough to mobilize all material in the armour layer. Model uncertainty was assessed with probabilistic confidence limits demonstrating that these erodibility reductions were statistically significant. A commonly used hillslope erosion model (sediment flux = β₁ discharge m₁ slope n₁) was fitted to these predictions. The erodibility, β₁, and m₁ decreased with time, which was consistent with our physical intuition about armouring. At Ranger the parameter m₁ asymptoted to 1·5–1·6 while at Northparkes it asymptoted to 1·2–1·3. At Ranger transient spatial trends in armouring led to a short term (50–200 years in the future) reduction in n₁, to below zero under certain circumstances, recovering to an asymptote of about 0·5–1. At Northparkes n₁ asymptoted to about 0·6, with no negative transients predicted. The m₁ and n₁ parameters predicted for Ranger were shown to be consistent with field data from a 10‐year‐old armoured hillslope and consistent with published relationships between erodibility and rock content for natural hillslopes. Copyright © 2007 John Wiley & Sons, Ltd.