Associate Professor Gregory Hancock

Context. Quantifying soil organic carbon (SOC) depth distribution and its vertical transport is needed for both improved understanding of soil properties and behaviour as well as enhanced organic carbon sequestration. This is a global issue, that if better understood, could result in both more agriculturally productive soils as well as enhanced environmental outcomes. Aims. Quantify whole soil-profile SOC and down-profile movement at a series of sites in south-east Australia. Methods. Soil is sampled at regular intervals using cores and assessed for SOC and environmental tracer (137Cs) concentration. Key results. Soils that have a high clay content (Vertosols) and crack (i.e. self mulching) have the highest SOC content. In high clay content soils, 137Cs is present at depths well below that at which it would be present by diffusive processes. Conclusions. Surface soil, labelled with 137Cs is moving down the soil profile by advective processes to depths well below that possible by diffusive processes alone. Using local erosion rates and carbon input, it is estimated that less than 1% of SOC is delivered to the cracking soils by erosional processes and that the majority of SOC must be produced in situ. Implications. Given that 137Cs is a relatively new environmental tracer (1945 onwards), this suggests that surface labelled soil is reaching depths of up to 80 cm at decadal time scales. The methods and findings here have global applicability and provide insights into potential enhancement of carbon sequestration in both cropping and grazing landscapes.

Animals are recognised biological agents that can turn over large amounts of soil, influence soil structure and composition and may allow soil to more easily erode. Feeding activities of feral pigs (Sus scrofa) are known to produce considerable soil disturbance. Here, pig disturbance, together with soil erosion and deposition patterns, are quantified along two hillslope transects in northern Australia over 10 years. Annual disturbance by pigs was ~1% of surface area. The erosion rate was low with a range of 0.008¿0.13 mm year-1. Both transects had similar hillslope profiles, vegetation patterns and similar appearance. There was a greater number of pig disturbances, area and mass of material disturbed by pigs for one transect compared to the other moving downslope. However, for the second transect, the number of disturbances significantly decreased moving downslope while the mass of material exhumed remained relatively constant along the hillslope. For the first transect, this suggests that the greater number of pig digs moving downslope reduces hillslope connectivity and therefore reduces erosion. That is, the greater the amount of disturbance by pigs, the less erosion. The pig digs produce a pit and a mound with the material diffusing locally and the pit capturing material from upslope reducing hillslope connectivity. These pits have been observed to last many years. A control on pig disturbance was likely to be rock content. The average surface rock content of the two transects was significantly different (9% and 20%, respectively) and with the higher rock content reducing pig disturbance and erosion for the second transect. The bio geomorphic influence that pigs represent in the landscape for biogeochemical cycling requires ongoing investigation.

Qualitative and quantitative understandings of the origin, transport and fate of organic carbon (OC) in forest systems is needed to advance our understanding of biogeochemical cycling as well as catchment management and forest harvesting. Here we present the findings from eight steep slope forested catchments dominated by headwater streams (size range 15¿100 ha) in south eastern Australia where bedload, organic carbon in bedload, hillslope soil carbon together with dissolved organic carbon has been measured over multiple years. OC in the bedload was found to be significantly related to catchment area (p < 0.02). Dissolved organic carbon export in stream water declined with catchment area. Combined bedload organic carbon and dissolved organic carbon export was significantly related to catchment area. There was no significant difference between hillslope soil organic carbon and bedload organic carbon concentration. When corrected for area, OC export by bedload was not significantly different to that of dissolved OC export. Using the environmental tracer 137Cs it was found that there was no significant difference in tracer concentration between bedload and hillslope soil. This suggests a direct link between hillslope and channel and delivery of organic carbon to the channel. Of the eight catchments examined here, six had been harvested for timber in previous decades (with area of forest removed ranging from 25 to 60%) while two catchments had minimal disturbance (Control catchments- no harvest). There was no difference in organic carbon export from the harvested and Control catchments. The results demonstrate that although land disturbance had previously occurred the management practices employed in each catchment were effective in the long term.

Mine tailings are a by-product of the processing of minerals. At most mines they are a waste product that needs to be managed. Tailings composition and properties vary widely and are in most cases highly erodible due to their fine particle size and can contain elevated concentrations of unwanted minerals and process chemicals. Therefore, if released to the environment they can be a significant environmental problem. A common management strategy is to store them in ¿tailings dams¿ where they will remain in perpetuity. Little work has been done to assess the long-term erosional behaviour of tailings dams. Computer based Landscape Evolution Models (LEMs) provide information on erosion rates, type of erosion and where erosion is likely to occur. They can therefore provide guidance on long-term behaviour which allows designs to be tested and improved. Here a LEM, SIBERIA, is used to assess two hypothetical tailings dam designs using different surface covers and climates. The results suggest that a tailings dam that can capture rainfall can erode less than a capped design that must shed any runoff. An embankment with a small and steep catchment has minimal erosion potential and any material eroded from the internal wall of the embankment is deposited internally and provides erosion protection. If the external embankment is maintained then there is potential for long-term encapsulation of tailings. The single biggest issue for the employment of LEMs is that of parameterisation and here assumes (1) a uniform and consistent armour or (2) a consistent and self-sustaining vegetation cover. The modelling and methods here provide a template for tailings dam assessment at other sites globally, and will improve tailings dam design and reduce environmental risk.

Explicit relationships and numerical models that link landscape and topographic characteristics with soil organic carbon (SOC) processes are needed. Here we calibrate a digital elevation model based Landscape Evolution Model (LEM) (SIBERIA) and assess both its ability to predict erosion and deposition and the spatial patterns of SOC. The LEM is capable of predicting both erosion and deposition at the hillslope and catchment scale. The LEM is calibrated for the site using a field data approach as well as using a laboratory flume. The predicted soil erosion rates from the LEM (1.7¿2.1 t ha-1 yr-1) compare well with independently determined erosion rates using 137Cs (2.1 to 3.4 t ha-1 yr-1). We also investigate field measured and modelled soil organic carbon movement using the LEM in relation to predicted erosion and deposition patterns and find that erosion and deposition patterns are related to the spatial patterns of SOC. This is the first time that a DEM based LEM has been shown to provide reliable prediction of not just soil erosion but also SOC. The results demonstrate that the majority of SOC is being transported in the near surface soil layer (top 2 cm) and that turnover at greater depths is slower and does not correspond with any modelled patterns. The modelled erosion and deposition suggests that on average 0.06 t ha-1 yr-1 of SOC is exported by erosion from the hillslope assuming a good vegetation cover. However if the hillslope is subjected to disturbance (i.e. tillage, overgrazing) then the site will export 0.46 t ha-1 yr-1of SOC. Laboratory results using flume suggest that there was no enrichment of SOC in the eroded sediment. The methods outlined here provide a new approach to quantify the dynamic movement of sediment and SOC at both the hillslope and catchment scale.

The factors determining the spatial distribution of soil organic carbon (SOC) at large-scales closely align with bioclimate regions; reflecting climate, ecosystem and soil properties. Recent studies of the Köppen-Geiger climate zones of Australia have highlighted an extension of the hot, arid, steppe environment from central Australia into the southeast (SE) under future climate change scenarios (2071¿2100 under RCP 8.5). As SOC concentrations are highest in Australia's SE, it is important the effect of this shift is quantified. This study assesses this and how changes in the factors that control SOC formation may alter SOC concentrations. Field measured SOC concentrations were compared to current climate, soil, topography, vegetation, and soil erosion variables for 12 grassland sites from SE to NW Australia. SOC concentrations ranged from 0.39% in northwest (NW) and Central Australia to as high as 6.88% in the SE. Using regression analyses; temperature, elevation and Normalised Difference Vegetation Index were found to be the only significant drivers (a = 0.95) of SOC across the sites. Partial correlation analyses then identified temperature, elevation and clay content as imparting a significant effect on the relationships between SOC and water availability variables. This indicates that an extension of the arid environment into SE Australia may lead to a decrease in SOC (up to 1.12%), as mean annual temperature exceeds threshold values that limit SOC concentration. This is significant as the majority of Australia's SOC is stored in this area and these environments exert a strong influence on global carbon cycling.

Erosion of soil by water is facilitated by both diffusive and fluvial processes. Here we examine three different soil redistribution processes operating at very different spatial and temporal scales in the monsoonal tropics of northern Australia. The first process, rainsplash, operates across the entire catchment. This process, while subject to annual and seasonal variations in rainfall amount and intensity, can be considered a constant forcing and redistributes on average 9¿t ha-1¿year-1 (range -0.9 to 19 t ha-1¿year-1). The second process, bioturbation, where in this study soil is disturbed by feral pigs (wild boar), occurs in selected areas throughout each year. Pigs exhume 3 to 36.0¿t ha-1¿year-1 (average ~11 t ha-1¿year-1). The effect of this disturbance may last for many years afterwards. The third process is the disturbance of the soil surface by tree throw and creation of pit¿mound topography (also a form of bioturbation), together with the resultant placement of the tree superstructure (above ground biomass) on the ground, which may form debris dams. Tree throw at the scale examined here is likely to occur only once every 50¿100 years, with the influence of this single event lasting for at least 10 years post event. Tree throw in a single event exhumed ~5¿t ha-1 (1.1¿9.5¿t ha-1) of soil. In contrast to rainsplash, pig disturbance and tree throw events are largely point-based phenomena. Field observation suggests that it takes many years for the disturbance from both pigs and tree throw to be removed. We find here that in terms of relative soil redistribution, rainsplash has the largest influence, with any erosional disturbance by pigs and tree throw being within the variability of rainsplash. However, the disruption of surface flow by the pig digs and tree throw disrupts sedimentological and hydrological connectivity.

Growing agricultural water demand is dramatically affecting the implementation of, and compliance with, water sharing plans in regions such as the Murray-Darling Basin (MDB). Problems can arise from water theft, poor resourcing or questionable actions from stakeholders. Recent actions from MDB governments have resulted in improved regulation, although more is required in a technical, governance and cultural space to create a comprehensive and transparent management framework. This is pivotal in improving overall trust in water regulators. We discuss an integrated water resource management approach for improved water regulation, involving the implementation of remote sensing technologies to complement metering, coupled with a focus on a stronger compliance culture in a range of stakeholder groups and regulatory changes that allow quicker adoption of unbiased best practice science and technology.

Baseline estimations of soil organic carbon (SOC) have been made globally using SOC models, earth system models, and digital soil mapping techniques. Digital elevation models (DEMs) underpin these analyses to incorporate the effect of topography on SOC. Here, we test the effect of DEM resolution on the relationships between topography and SOC across catchment to hillslope scales. Samples were collected using a nested field sampling approach in a grazed catchment on Eastern Australia. Topographic attributes were derived from 5 m, 25 m, 30 m, and 90 m resolution DEMs and then used to predict SOC using a Random Forest model. This was trained on a catchment-wide dataset and tested using the repeat samples of this dataset and finer scale field data. SOC was able to be predicted using topography in the catchment-wide datasets (model R2 = 0.37¿0.51), however not in the finer-scale data (model R2 = 0¿0.21). Variable importance was calculated from the modelling process, with elevation (as a surrogate for climate) being the main driver at the catchment-scale. In the finer scale datasets, topographic variables linked to soil redistribution were more important. As a result, SOC estimation methods using coarse resolution DEM data and large-scale sampling may be limited in capturing the effect of topography, having implications for SOC management and modelling.

Field measurement and modelling of soil erosion provides insights into landscape systems as well as the potential for enhanced landscape management. There are a number of field and numerical methods by which soil erosion and deposition can be quantified. Here we examine the capability of the SIBERIA landscape evolution model to quantify short-term erosion and deposition on a well-managed cattle grazing landscape on the east coast of Australia. The model is calibrated by two methods (1) a geomorphological approach using a site digital elevation model (DEM) and soil data and (2) a laboratory-scale flume. The two calibration processes resulted in similar model input parameters and estimated erosion rates of 3.1 t ha-1 year-1 and 4.4 t ha-1 year-1, respectively. These were found to closely match erosion rates estimated using the environmental tracer 137Cs (2.7¿4.8 t ha-1 year-1). However, erosion and deposition estimated at individual points along the hillslope was not well correlated with 137Cs at the same position due to the temporal averaging of the model and microtopography. Sensitivity analysis showed the model was more sensitive to parameterisation than sub-DEM-scale topography. This places confidence in the model's ability to estimate erosion and deposition across an entire hillslope and catchment on decadal time scales. We also highlight the robustness and flexibility of the calibration methods.

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.

Incision as a result of fluvial erosion is an important process to model when simulating landform evolution. For gullies, it is apparent that coupled with the processes that cause incision there must be a range of processes that stop incision. Once started, rills and gullies will grow infinitely without a reduction in support area and/or being arrested by deposition and armouring. Some of these processes have been well studied under the heading of inter-rill erosion. Other limiting processes are related to the shape of the landform and how downstream deposition areas are linked geomorphically to the upstream gullies. Armouring is also an important process that reduces gully incision and extension, where the gully erodes to bedrock and the resistant base limits further development. Post-mining landscapes are new surfaces with new materials and provide the opportunity to examine gully initiation, extension and stabilization. The work presented here has largely been driven by the mining industry, where there has been a need to assess erosion over hazardous wastes like mine tailings and low-level nuclear waste. We demonstrate the usefulness of computer-based landscape evolution models and the more recent soilscape models (that include both surface and subsurface processes) to understand both fluvial and diffusive processes as well as armouring in a digital elevation model framework (as well as landscape evolution). Landscape evolution models provide insights into complex non-linear systems such as gullies. A key need is that of field data to parameterize and validate the models. It is argued that current models have more capability than field data available for parameterization and importantly the validation of model outputs.

The Monsoonal North West (MNW) region of Australia faces a number of challenges adapting to anthropogenic climate change. These have the potential to impact on a range of industries, including agricultural, pastoral, mining and tourism. However future changes to rainfall regimes remain uncertain due to the inability of Global Climate Models to adequately capture the tropical weather/climate processes that are known to be important for this region. Compounding this is the brevity of the instrumental rainfall record for the MNW, which is unlikely to represent the full range of climatic variability. One avenue for addressing this issue (the focus of this paper) is to identify sources of paleoclimate information that can be used to reconstruct a plausible pre-instrumental rainfall history for the MNW. Adopting this approach we find that, even in the absence of local sources of paleoclimate data at a suitable temporal resolution, remote paleoclimate records can resolve 25% of the annual variability observed in the instrumental rainfall record. Importantly, the 507-year rainfall reconstruction developed using the remote proxies displays longer and more intense wet and dry periods than observed during the most recent ~ 100 years. For example, the maximum number of consecutive years of below (above) average rainfall is 90% (40%) higher in the rainfall reconstruction than during the instrumental period. Further, implications for flood and drought risk are studied via a simple GR1A rainfall runoff model, which again highlights the likelihood of extremes greater than that observed in the limited instrumental record, consistent with previous paleoclimate studies elsewhere in Australia. Importantly, this research can assist in informing climate related risks to infrastructure, agriculture and mining, and the method can readily be applied to other regions in the MNW and beyond.

Temperature has a large impact on soil biogeochemical functioning. While this is well recognised there is a lack of data on the variability of soil temperature at the hillslope and catchment scale. Here we examine soil temperature for a series of nested catchments in the east of New South Wales, Australia over a twelve year period. Temperature characteristics between the three catchments were shown to be quite similar, falling into a 1 °C band. Across each catchment, the larger catchments of Krui (590km2) and Merriwa (808km2) had a larger varibaility in temperature range (3.4oC and 4.9 °C repectively) compared to the much smaller Stanley (175 ha) sub-catchment (1.8 °C), indicating that larger catchments will generally have larger variability in soil temperatures. The larger catchments also had a lapse rate of approximately 1 °C per 100 m elevation. Seasonal trends in soil temperature variability were observed to be quite similar between the two larger catchments, with soil temperature exhibiting greater variability for both catchments within the warmer months of the year. The similarity in soil temperature trends between Krui and Merriwa over the year is attributed to their similar size, climate, soils, and north-south orientation. All three catchments had a narrower range of temperatures during the cooler months, when Krui and Merriwa have a similar range to the soil temperatures of Stanley. The variability in soil temperature across the Stanley catchment is relatively stable throughout the year. This was attributed to its smaller size. In terms of quantifying catchment scale soil temperature from a single point, weather station sites that were located closer to the geographic centre of the catchment were shown to be less likely to over or under-estimate the average soil temperature of the catchment. The use of point-scale air temperature as a surrogate for predicting catchment soil temperature was suitable when corrections were applied. The importance of soil temperature to many ecological processes, in particular vegetation growth and soil biological activity, and therefore the soil carbon cycle and soil carbon sequestration, highlights the significance of this research.

Data from field plots describing how new surfaces evolve in the first few years post-construction are scarce in the literature. Here we examine sediment output from four similar 30¿m by 30¿m plots on a rehabilitated mine site over a six year period. Field measurements from the trial plots found that there is an initial high pulse of sediment over the first three years which rapidly reduces to rates similar to that expected for a natural or undisturbed surface. At 6¿years the sediment output is equivalent to that expected from the surrounding undisturbed landscape. This plot data was compared to predictions from a calibrated landscape evolution model. The landscape evolution model used two sets of parameters, one derived from bare waste rock and one derived from an older vegetated surface. The simulations using bare waste parameters produced sediment output that matched the plot data in the first few years while the vegetated parameters produced sediment output which compared well with the field plot data at times >¿3¿years. The results demonstrate that when correctly calibrated the landscape evolution model is able to reliably predict sediment output from these field plots. These results suggest that there is the potential to employ the bare waste rock dump parameters for the first 3¿4¿years then switch to vegetated parameters for the longer term modelling. Both the field plots and landscape evolution model simulations displayed considerable annual variability in total load. This variability is the result of different surface structure from imposed surface roughness (ripping by a bulldozer) and their unique topographic structure. Both initial DEM and model parameters have a large influence on predicted sediment load. The results here support the reliability of the model at the sub-metre grid scale.

This paper generalises the physical dependence of the relationship between contributing area, local slope, and the surface soil grading using a pedogenesis model and allows an exploration of soilscape self-organisation. A parametric study was carried out using different parent materials, erosion, and weathering mechanisms. These simulations confirmed the generality of the area-slope-d50 relationship. The relationship is also true for other statistics of soil grading (e.g. d10, d90) and robust for different depths within the profile. For small area-slope regimes (i.e. hillslopes with small areas and/or slopes) only the smallest particles can be mobilised by erosion and the area-slope-d50 relationship appears to reflect the erosion model and its Shield's Stress threshold. For higher area-slope regimes, total mobilization of the entire soil grading occurs and self-organisation reflects the relative entrainment of different size fractions. Occasionally the interaction between the in-profile weathering and surface erosion draws the bedrock to the surface and forms a bedrock outcrop. The study also shows the influence on different depth-dependent in-profile weathering functions in the formation of the equilibrium soil profile and the grading characteristics of the soil within the profile. We outline the potential of this new model and its ability to numerically explore soil and landscape properties.

A significant issue for the application of numerical Landscape Evolution Models (LEMs) is their calibration/parameterisation and validation. LEMs are now at the stage of development where if calibrated, they can provide meaningful and useful results. However, before use, each LEM requires a set of data and parameter values for it to run reliably and most importantly produce results with some measure of precision and accuracy. This calibration/validation process is largely carried out using parameter values determined from present day, or recent surface conditions which are themselves product of much longer-term geology-soil-climate-vegetation interactions. Here we examine the reliability of an LEM to predict catchment form over geological time (500,000 years) for a potential rehabilitated mine landform using defensible parameters derived from field plots. The findings demonstrate that there is no equifinality in landscape form with different parameter sets producing geomorphically and hydrologically unique landscapes throughout their entire evolution. This shows that parameterisation does matter over geological time scales. However, for shorter time scales (< 10,000 years) the geomorphic differences in hillslope form are minimal as described by the hypsometric curve, area-slope and cumulative area distribution, yet there are large differences in sediment output. Therefore, obtaining reliable and defensible parameters for input to LEMs is essential.

An important part of planning for the rehabilitation of a mine site is the design of stable final landforms for waste rock dumps or spoil piles. Whilst able to be assessed over the short-term (years to decades), the longer term behaviour (centuries to millennia) of such landscapes is not within any meaningful human time frame of observation. Predictive numerical models, therefore, form an important tool with which current landscape behaviour and longer term trajectory can be assessed. However, an important issue associated with the use of models is the ability to assess the reliability and accuracy of the model. Here the SIBERIA and CAESAR-Lisflood Landscape Evolution Models (LEMs) are used to simulate and assess the geomorphic stability of a conceptual rehabilitated landform of the Ranger uranium mine in the Northern Territory, Australia, for a simulated period of up to 1000. years. Utilising both models in this study enabled an independent assessment of likely landscape processes and evolution as well as each model. Results show that SIBERIA and CAESAR-Lisflood produce erosion rates and patterns that are broadly similar. At millennial time scales, short-term processes such as gullying appear to be the dominant erosion features in the proposed landforms and may produce substantial erosion in terms of size and amount of hillslope material eroded and transported downslope. Vegetation was found to have a major effect on the erosion potential of the landform surface. Overall both models produce very similar results.

Pigs (Sus scrofa) are recognised as having significant ecological impacts in many areas of the world including northern Australia. The full consequences of the introduction of pigs are difficult to quantify as the impacts may only be detected over the long-term and there is a lack of quantitative information on the impacts of feral pigs globally. In this study the effect of feral pigs is quantified in an undisturbed catchment in the monsoonal tropics of northern Australia. Over a three-year period, field data showed that the areal extent of pig disturbance ranged from 0.3-3.3% of the survey area. The mass of material exhumed through these activities ranged from 4.3tha-1yr-1 to 36.0tha-1yr-1. The findings demonstrate that large introduced species such as feral pigs are disturbing large areas as well as exhuming considerable volumes of soil. A numerical landscape evolution and soil erosion model was used to assess the effect of this disturbance on catchment scale erosion rates. The modelling demonstrated that simulated pig disturbance in previously undisturbed areas produced lower erosion rates compared to those areas which had not been impacted by pigs. This is attributed to the pig disturbance increasing surface roughness and trapping sediment. This suggests that in this specific environment, disturbance by pigs does not enhance erosion. However, this conclusion is prefaced by two important caveats. First, the long term impact of soil disturbance is still very uncertain. Secondly, modelling results show a clear differentiation between those from an undisturbed environment and those from a post-mining landscape, in which pig disturbance may enhance erosion.

Paired watershed experiments involving the removal or manipulation of forest cover in one of the watersheds have been conducted for more than a century to quantify the impact of forestry operations on streamflow. Because climate variability is expected to be large, forestry treatment effects would be undetectable without the treatment-control comparison. New understanding of climate variability provides an opportunity to examine whether climate variability interacts with forestry treatments, in a predictable manner. Here, we use data from the H. J. Andrews Experimental Forest, Oregon, USA, to examine the impact of the El Niño-Southern Oscillation on streamflow linked to forest harvesting. Our results show that the contrast between El Niño and La Niña events is so large that, whatever the state of the treated watershed in terms of regrowth of the forest canopy, extreme climatic variability related to El Niño-Southern Oscillation remains the more dominant driver of streamflow response at this location. Improvements in forecasting interannual variation in climate might be used to minimize the impact of forestry treatments on streamflow by avoiding initial operations in La Niña years.

The measurement of hillslope erosion can be a difficult, costly and time-consuming activity. Many techniques are available, ranging from using environmental tracers, to LiDAR. Erosion measurements using erosion pins are assessed and compared with regional scale erosion data, hillslope data obtained using ¹³⁷Cs and erosion modelling results. The pins produced erosion rates which are within the range determined using ¹³⁷Cs and model data but above that of regional denudation rates. Our findings demonstrate that inexpensive erosion pins can provide reliable data on hillslope erosion.

The sediment delivery ratio (SDR) has been a common approach developed to understand change in sediment yield and flux through a catchment. In this study we propose that the underlying concept of the sediment delivery ratio is flawed for a number of reasons: its linear extrapolation is physically meaningless; there is no evidence of the magnitude of storage required by the SDR approach on annual to decadal timescales; and the SDR approach assumes suspended sediment transport is conservative yet it is known to undergo both loss and production in-channel. This study considers the sediment yield from 192 UK catchments from 1974 to 2010 for catchment areas between 4 and 9948km2 and shows that linear extrapolation of the SDR approach overpredicts source terms and underpredicts fluxes for large catchments. The SDR approach hides a range of behaviours of suspended sediment flux within catchments with patterns of net deposition, net increase or no change all apparent in UK catchments. The approach proved to be self-correlated which meant that it can result in spurious correlations when compared to catchment area. The change in yield with catchment area can be just as well understood as a change in sediment supply from channels rather than as a change in delivery from hillslope sources. We propose that suspended sediment flux change with catchment area be modelled as a more physically-meaningful Gompertz function (step function) rather than using the traditional SDR approach.

Gully head and wall retreat has commonly been attributed to fluvial and head collapse as a result of soil saturation, sapping or piping. The empirical evidence to substantiate these conceptual models is sparse, however, and often contradictory. This paper explores the hydrological and mechanical controls on gully head and wall stability by modelling the hydrology, stability and elastic deformation of a marl gully complex in Granada Province, southeast Spain. The hydrological and slope-stability simulations show that saturated conditions can be reached only where preferential fissure flow channels water from tension cracks into the base of the gully head, and that vertical or subvertical heads will be stable unless saturation is achieved. Owing to the high unsaturated strengths of marl measured in this research, failure in unsaturated conditions is possible only where the gully head wall is significantly undercut. Head retreat thus requires the formation of either a tension crack or an undercut hollow. Finite-element stress analysis of eroding slopes reveals a build up of shear stress at the gully head base, and a second stress anomaly just upslope of the head wall. Although tension cracks on gully heads have often been attributed to slope unloading, this research provides strong evidence that the so called 'sapping hollow' commonly found in the gully headwall base is also a function of stress release. Although further research is needed, it seems possible that 'pop out' failures in river channels may be caused by the same process. The hydrological analysis shows that, once a tension crack has developed, throughflow velocity in the gully headwall will increase by an order of magnitude, promoting piping and enlargement of this weakened area. It is, therefore, possible to envisage a cycle of gully expansion in which erosion, channel incision or human action unloads the slope below a gully head, leading to stress patterns that account for the tension crack and a stress-release hollow. The tension crack promotes faster throughflow, encouraging hollow enlargement and piping, which undercut the gully head. The tension crack permits the development of positive porewater pressures behind the gully head, leading either to failure or contributing to toppling. Finally the debris may be eroded by fluvial action, unloading a new section of slope and completing the cycle of gully head retreat. Copyright © 2001 John Wiley and Sons, Ltd.

Hypsometry has historically been used as an indicator of geomorphic form of catchments and landforms. Yet there has been little work aimed at relating hypsometry to landform process and scale. This paper uses the SIBERIA catchment evolution model to explore linkages between catchment process and hypsometry. SIBERIA generates results that are qualitatively and quantitatively similar to observed hypsometric curves for physically realistic parameters. However, we show that not only does the hypsometry reflect landscape runoff and erosion process, but it is strongly dependent on channel network and catchment geometry. We show that the width to length ratio of the catchment has a significant influence on the shape of the hypsometric curve, though little on the hypsometric integral. For landforms dominated by fluvial sediment transport, the classic Strahler 'mature' hypsometric curve is only generated for catchments with roughly equal width and length. Narrow catchments show a hypsometric curve more similar to Strahler's 'monadnock' form. For landscapes dominated by diffusive transport, the simulated hypsometric curve is concave-down everywhere, this being consistent with curves reported for some example catchments in France. Because the transition between diffusive dominant to fluvial is scale-dependent, with larger catchments exhibiting greater fluvial dominance, then the hypsometric curve is a scale-dependent descriptor of landforms. Experimental results for simulated landforms from a small-scale rainfall-erosion simulator are reported. It is shown that SIBERIA yields satisfactory fits to the data, confirming its ability to predict the form of the hypsometric curve from a simple model of geomorphic processes.

Modelling complex spatially distributed processes over large areas is a challenge facing many environmental disciplines. In some cases even when the physics and relationships are known they are too complex for large scale simulations. In the field of pedogenesis researchers have a good understanding of the physical processes affecting soil evolution but struggle with predicting their spatial and temporal variability. Spatial and temporal relationships in soil properties are known to be a significant factor in many hydrological and geomorphological processes. Modelling detailed distribution of soil properties is therefore important but yet to be fully achieved over large scales. In order to predict the spatial distribution of soil properties a new approach is needed in the form of detailed large spatial and temporal scale simulation of soil evolution. These simulations could not be done using conventional physically-based model due to their computational complexity and limited dimensionality. A more comprehensive and computationally efficient algorithm is needed. Our solution is a new pedogenesis model called mARM. mARM (matrices-ARMOUR) uses transition matrices in conjunction with physically-based equations to achieve a highly modular and computationally efficient modelling platform. This new approach and the modularity of the mARM platform allow us to simulate a variety of soil processes over long periods and large areas (millions of years and 1000's pixels). In its latest version (presented here) mARM has been extended to three spatial dimensions (mARM3D) by adding a soil profile component. This 3D model is a major breakthrough in pedogenesis modeling since this is the first time profile as well as surface processes are simulated over large scales. The modularity of mARM3D allows us to use it as a virtual laboratory to examine a variety of soil processes and relationships such as: ¿ Depth-dependent weathering rate equations (presented here) ¿ Runoff fluctuations (as a result of climate or environmental changes) ¿ Weathering geometry (proportion of particle split) ¿ Weathering-Erosion relationships and rates In addition to the above capabilities, mARM is intended to be integrated as a component in a landform evolution model (i.e. TelluSim) which will allow such a model, for the first time, to account for time and space variation in soil properties. In this paper we describe the mARM3D model framework which can potentially be used in other complex spatial models. We also present selected results from the mARM3D simulations at hillslope and landscape scale which show the effect of profile weathering physics on surface soil distribution.