Channel Bank Erosion Research Articles

The dominant erosion processes supplying fine sediment to three major rivers in tropical Australia, the Daly (NT), Mitchell (Qld) and Flinders (Qld) Rivers

The tropics of northern Australia have received relatively little attention with regard to the impact of soil erosion on the many large river systems that are an important part of Australia's water resource, especially given the high potential for erosion when long dry seasons are followed by intense wet season rain. Here we use 137Cs concentrations to determine the erosion processes supplying sediment to two major northern Australian Rivers; the Daly River (Northern Territory), and the Mitchell River (Queensland). We also present data from five sediment samples collected from a 100km reach of the Cloncurry River, a major tributary of the Flinders River (Queensland). Concentrations of 137Cs in the surface soil and subsurface (channel banks and gully) samples were used to derive ‘best fit’ probability density functions describing their distributions. These modelled distributions are then used to estimate the relative contribution of these two components to the river sediments. Our results are consistent with channel and gully erosion being the dominant source of sediment, with more than 90% of sediment transported along the main stem of these rivers originating from subsoil. We summarize the findings of similar studies on tropical Australian rivers and conclude that the primary source of sediment delivered to these systems is gully and channel bank erosion. Previously, as a result of catchment scale modelling, sheet-wash and rill erosion was considered to be the major sediment source in these rivers. Identifying the relative importance of sediment sources, as shown in this paper, will provide valuable information for land management planning in the region. This study also reinforces the importance of testing model predictions before they are used to target investment in remedial action.

Channel Adjustment Following Culvert Removal from Forest Roads in Northern California, USA

Forest roads can be a major contributor of sediment entering streams and estuaries. Road restoration and removal has been shown to be effective in reducing sediment delivery over decadal time spans. However, few studies have been conducted to quantify and characterize short-term (1-3 yr) channel adjustment after removal of stream crossing fill material and reconstruction of former stream channels. This evaluation is important for managers needing to assess impacts, cost, and benefits of stream crossing removal. This study examines 20 road stream crossings by surveying channel adjustments and determining rilling and gullying delivery volumes, channel substrate changes, and the role of rock placed in reconstructed channels at Redwood National and State Parks, California, USA. Erosion and channel incision were significantly higher at crossings in the 1st year following removal than at older crossings. First year crossings yielded an average of 7.71 m3 of erosion per crossing, while the 2nd- and 3rd-year crossings showed a net gain of 0.08 m3 of sediment (indistinguishable from 0). Eighty-six percent of the erosion could be attributed to channel incision and bank erosion. The remaining 14% originated from rilling, gullying, and side-slope failures. Coarsening of channel substrate was most pronounced in the 1st year after removal, and the 1st storms of the year were most important in mobilizing sediment. This study underlines the importance of excavating to bedrock or coarse material when recreating a streambed, and the identification of bulges (short convex segments) in the channel profile that are likely to erode.

A multiple watershed field test of hydrogeomorphic functional assessment of headwater streams—Variability in field measurements between independent teams

Ephemeral and intermittent headwater streams are under increasing pressure from disturbance and development. Rapid, repeatable assessment techniques are needed in order to gauge the condition of these stream systems. Several attributes of these headwater streams constrain the use of the most widely used macroinvertebrate or water quality stream assessment techniques. The hydrogeomorphic (HGM) functional assessment is a reference-based alternative technique. To evaluate this alternative, repeated assessments were conducted in eight high-gradient headwaters in West Virginia by four independent teams. Across-site and measurement variance among teams was assessed using a coefficient of variation (CV, expressed as percent). A variability of >50% CV, which suggests less repeatable results, occurred in only 13.8% of measurements, primarily associated with 2 of the 9 variables examined (snag density and substrate size). Between site measurement variance was the greatest at more highly disturbed sample locations, particularly with regard to the large woody debris, tree species richness, and channel bank erosion variables. Variables with the lowest CV were tree diameter, detrital cover, canopy cover, and channel embeddedness. Based on these results, measurements included when applying HGM approach to these streams should focus on direct measurements or directed estimates that yield a large response range across a spectrum of sites while maintaining consistent repeatability among different teams, with special attention paid to their use in highly disturbed sites.

A preliminary investigation of the efficacy of riparian fencing schemes for reducing contributions from eroding channel banks to the siltation of salmonid spawning gravels across the south west UK

The low productivity of salmonids in many river systems across the UK is testament to their intolerance of water quality perturbations including those associated with excessive sedimentation. Catchment and fishery managers concerned with combating such issues require reliable information on the key sources of the sediment problem to target management and on the efficacy of the mitigation options being implemented. In recognition of the latter requirement, a pre- and post-remediation sediment sourcing survey was used to examine the potential for using sediment tracing to assemble preliminary information on the efficacy of riparian fencing schemes for reducing sediment contributions from eroding channel banks to salmonid spawning gravels in the Rivers Camel, Fal, Lynher, Plym, Tamar and Tavy in the south west of the UK. Respective estimates of the overall mean proportion (±95% confidence limits) of the interstitial sediment input to salmonid spawning gravels originating from eroding channel banks during the pre- (1999–2000) and post-remediation (2008–2009) study periods were computed at 97 ± 1% vs. 69 ± 1%, 94 ± 1% vs. 91 ± 1%, 12 ± 1% vs. 10 ± 1%, 92 ± 1% vs. 34 ± 1%, 31 ± 1% vs. 16 ± 1% and 90 ± 1% vs. 66 ± 1%. Whilst the study demonstrates the potential utility of the fingerprinting approach for helping to assemble important information on the efficacy of bank fencing as a sediment source control measure at catchment scale, a number of limitations of the work and some suggestions for improving experimental design are discussed. Equivalent data are urgently required for many sediment mitigation options to help inform the development of water quality policy packages designed to protect aquatic ecosystems.

Variable Responses of Fish Assemblages, Habitat, and Stability to Natural‐Channel‐Design Restoration in Catskill Mountain Streams

AbstractNatural‐channel‐design (NCD) restorations were recently implemented within large segments of five first‐ and second‐order streams in the Catskill Mountains of New York in an attempt to increase channel stability, reduce bed and bank erosion, and sustain water quality. In conjunction with these efforts, 54 fish and habitat surveys were done from 1999 to 2007 at six restored reaches and five stable control reaches to evaluate the effects of NCD restoration on fish assemblages, habitat, and bank stability. A before–after–control–impact study design and two‐factor analysis of variance were used to quantify the net changes in habitat and fish population and community indices at treatment reaches relative to those at unaltered control reaches. The density and biomass of fish communities were often dominated by one or two small prey species and no or few predator species before restoration and by one or more trout (Salmonidae) species after restoration. Significant increases in community richness (30%), diversity (40%), species or biomass equitability (32%), and total biomass (up to 52%) in at least four of the six restored reaches demonstrate that NCD restorations can improve the health and sustainability of fish communities in geomorphically unstable Catskill Mountain streams over the short to marginally long term. Bank stability, stream habitat, and trout habitat suitability indices (HSIs) generally improved significantly at the restored reaches, but key habitat features and trout HSIs did not change or decreased at two of them. Fish communities and trout populations at these two reaches were not positively affected by NCD restorations. Though NCD restorations often had a positive effect on habitat and fish communities, our results show that the initial habitat conditions limit the relative improvements than can be achieved, habitat quality and stability do not necessarily respond in unison, and biotic and abiotic responses cannot always be generalized.

A comparison of average rates of alluvial erosion between the south‐western and south‐eastern United States

AbstractAlluvial channel bed incision and bank widening have been reported in both the south‐western and south‐eastern US throughout the past century. Distinct regional differences in climate and landscape properties likely influence the rate of erosion. This study discusses regional differences in hydraulic driving forces and substrate resistance and tests the hypothesis that regional differences exist in average rates of channel incision, bank erosion, and knickpoint retreat. Specifically, we hypothesize that erosion rates are higher in south‐western US streams and reason that this is because of greater flood magnitudes and limited substrate resistance. A review of the literature documenting incision, bank erosion, and knickpoint retreat, however, indicates that intra‐regional differences are larger than inter‐regional differences and that average rates in the south‐western US are either statistically similar or less than the rates in parts of the south‐eastern US. This could either be a result of strong intra‐regional hydroclimatic and substrate variability or because average erosion rate may not be an appropriate metric for inter‐regional comparisons because of the variability between case studies associated with the field methods to measure erosion, duration of study period, and time since disturbance to the channel. Nevertheless, these findings provide a basis for future evaluations of the relative importance of different controls on driving and resisting forces in these and other landscapes characterized by rapid channel incision and arroyo formation. Copyright © 2010 John Wiley & Sons, Ltd.

Debris Flows and Floods Following the 2003 Wildfires in Southern British Columbia

A number of post-wildfire debris flow and debris flood events occurred in the two years following the extreme fire season of 2003 in the southern interior of British Columbia. Such events had not been previously documented in Canada. Rainstorms following five fires caused significant events in the Okanagan and Kootenay regions of the province, including the Okanagan Mountain Park, Cedar Hills, Kuskonook, Lamb Creek, and Ingersoll fires. Damage to residential property and infrastructure downstream of the fires occurred in three cases. Most of the damaging events were debris flows, although several large debris floods occurred on lower gradient streams. At four of the fires, the events were triggered by short-duration, high-intensity rainstorms. At the fifth fire, a long-duration, low-intensity rainfall triggered the events. In all cases, severe burns with water repellent soils were identified or suspected as contributing to the events. The debris flow and debris flood events described here illustrate three initiation mechanisms: runoff-triggered debris flows caused by erosion of channel bed and banks by critically high discharge; debris flows and floods caused by progressive sediment bulking of runoff with material eroded from headwater slopes; and landslide-triggered debris flows caused by a landslide that enters a steep channel. In the burned areas studied, the first mechanism is the most common, whereas in unburned forested landscapes in the region, the third mechanism is the most common.

Do we know enough about controlling sediment to mitigate damage to stream ecosystems?

Stream and river ecosystems have suffered extensive degradation, and billions are expended annually on restoration efforts. However, few of these projects are monitored, and restoration effectiveness is often unknown. Consequently, there is a poor scientific foundation for restoration designs. Since many stream restoration efforts are at least partially targeted at controlling erosion of channel banks and beds, the effects of a large-scale, long-term stream erosion control effort in six Mississippi watersheds was assessed using 10–16 years of suspended sediment and water discharge records. Flow-adjusted suspended sediment concentrations showed no trends in five of the watersheds and a slight downward trend in one watershed, which was treated with small reservoirs as well as bed and bank erosion protection. Results indicate the inability of orthodox stream management structures (weirs and bank protection) to reduce watershed sediment yield and the need for a stronger scientific basis for stream restoration.

The potential impact of projected change in farming by 2015 on the importance of the agricultural sector as a sediment source in England and Wales

Abstract In association with a major initiative aimed at identifying policy packages for inclusion in the Programmes of Measures (POM's) comprising EU Water Framework Directive (WFD) River Basin Management Plans (RBMP's), recent work has evaluated the gap between current and compliant suspended sediment losses due to farming across England and Wales. The work required national scale sediment source apportionment to assess the current contributions of diffuse agricultural and urban sector losses, channel bank erosion and point source discharges to the total suspended sediment loads delivered to all rivers. Results suggested that the agricultural sector dominates present day (year 2000) sediment inputs to rivers (1929 kt = 76%) compared to eroding channel banks (394 kt = 15%), diffuse urban sources (147 kt = 6%) and point source discharges (76 kt = 3%). Projected change in farming by 2015, represented by the Business as Usual forecast of structural developments and predicted uptake of sediment mitigation methods, suggested an overall 9% reduction in sediment loss from the agricultural sector across England and Wales. The projected reduction is unlikely to deliver sediment compliance in all catchments. Key limitations of the integrated modelling approach are discussed.

Sedimentation, geochemistry and palaeomagnetism of the West Runton Freshwater Bed, Norfolk, England

The sedimentary succession of the West Runton Freshwater Bed (WRFB) is described and interpreted. The sediments represent a fluvial valley floor accumulation that trends west- to east, subparallel to the modern cliffline. Stabilisation of the stream channels was initially followed by deposition of fine-grained silt in standing water, punctuated by intermittent flood events. Decreased flood intensity resulted in vertical accretion of fine-grained, organic, fossiliferous sediments in carbonate-rich spring water.Increased organic deposition produced a transition to black detrital sediment. Periodic flooding continued, but flood frequency decreased in response to reduction in accommodation. Local erosion of channel banks and floors also occurred. The sediments were disturbed by water-release structures and bioturbation, the latter resulting from large mammals locally entering the channel and pools, churning the deposits during walking, bathing and so on. River-flow velocity diminished later as forest vegetation stabilised catchment and valley-floor ground surfaces. Stabilisation of the channel led to deposition from low-velocity flowing and standing water. The uppermost part of the succession indicates vertical accretion, sheet-like deposition. The organic silts were deposited in an anoxic environment, and organic sulphur from rotting organic matter allowed formation of early diagenetic sulphides.Deposition began during the late-glacial (Beestonian Stage) and continued into the first half of a temperate-climate, interglacial event (the Cromerian Stage s.s.). The infill comprises an overall fining-upward sequence, the stream adopted a stable meandering pattern. The WRFB stream showed typical chalk-stream behaviour.The magnetic polarity of the WRFB sediments is normal. The palaeomagnetic data indicates that the sediments are of Brunhes Chron age, and imply that they post-date mid-MIS 19.

Predicting potential change in agricultural sediment inputs to rivers across England and Wales by 2015

Recent work has evaluated the gap between current and compliant suspended sediment losses due to farming across England and Wales and the potential for change by 2015. The study adopted the guideline annual average sediment threshold of 25 mg L–1 cited by the European Union Freshwater Fish Directive. Compliance testing required national scale sediment source apportionment to assess the current contributions of diffuse agricultural and urban sector losses, channel bank erosion and point source discharges to the total suspended sediment loads delivered to all rivers. Results suggested that the agricultural sector dominates present day (year 2000) sediment inputs to rivers (1929 kt = 76%) compared with eroding channel banks (394 kt = 15%), diffuse urban sources (147 kt = 6%) and point source discharges (76 kt = 3%). Projected change in farming by 2015, represented by the Business As Usual forecast of structural developments and predicted uptake of sediment mitigation methods, suggested a 9% reduction in sediment losses from the agricultural sector across England and Wales. Further mitigation of diffuse agricultural sediment transfers to watercourses will therefore be necessary to ensure ‘good ecological status’ in some catchments. A similar modelling framework could be applied in other countries.

Predicting sediment inputs to aquatic ecosystems across England and Wales under current environmental conditions

The awareness of water quality issues has never been higher. As part of its continuing strategic diffuse pollution policy support, ADAS recently undertook to identify catchments across England and Wales that could potentially fail recently proposed suspended sediment yield targets under current environmental conditions. The total suspended sediment loads (SSL) delivered to all rivers were assumed to comprise contributions from diffuse sources in the agricultural and urban sectors, as well as from eroding channel banks and point sources represented by sewage treatment works (STWs). Diffuse agricultural sediment loss to rivers was predicted using the PSYCHIC model. Corresponding inputs from diffuse urban sources were estimated on the basis of an Event Mean Concentration (EMC) methodology. Channel bank sediment inputs were calculated using a prototype national scale model, while point source sediment contributions were based on a register of consented effluent discharges. Modelled SSL were validated ( r 2=68%) against PARCOM data (1999–2003) for the delivery of sediment to different regions of the UK maritime area. The results of the validation were considered to be realistic for a national scale predictor. The modelling exercise suggested that those catchments currently at risk of exceeding proposed suspended sediment yield critical thresholds are largely confined to upland areas across Wales and northwest England and the chalklands of southern and eastern England.

PSYCHIC – A process-based model of phosphorus and sediment transfers within agricultural catchments. Part 2. A preliminary evaluation

Summary This paper describes the preliminary evaluation of the PSYCHIC catchment scale (Tier 1) model for predicting the mobilisation and delivery of phosphorus (P) and suspended sediment (SS) in the Hampshire Avon (1715 km2) and Herefordshire Wye (4017 km2) drainage basins, in the UK, using empirical data. Phosphorus and SS transfers to watercourses in the Wye were predicted to be greater than corresponding delivery in the Avon; SS, 249 vs 33 kg ha−1 yr−1; DP, 2.57 vs 1.26 kg ha−1 yr−1; PP, 2.20 vs 0.56 kg ha−1 yr−1. The spatial pattern of the predicted transfers was relatively uniform across the Wye drainage basin, whilst in the Avon, delivery to watercourses was largely confined to the river corridors and small areas of drained land. Statistical performance in relation to predicted exports of P and SS, using criteria for relative error (RE) and root mean square error (RMSE), reflected the potential shortcomings associated with using longer-term climate data for predicting shorter-term (2002–2004) catchment response and the need to refine calculations of point source contributions and to incorporate additional river basin processes such as channel bank erosion and in-stream geochemical processing. PSYCHIC is therefore best suited to characterising longer-term catchment response.

Assessing the likelihood of catchments across England and Wales meeting ‘good ecological status’ due to sediment contributions from agricultural sources

Abstract This paper is concerned with estimating the gap between current and compliant losses of suspended sediment from the agricultural sector in England and Wales in relation to achieving ‘good ecological status’ (GES) in freshwaters by 2015. Given the emphasis on strategic information for policy support, the assessment necessitated a novel modelling methodology for predicting mean annual total suspended sediment loads (SSL) and time-weighted suspended sediment concentrations (SSC). GES was defined as the guideline annual average SSC of 25 mg l−1 cited by the EC Freshwater Fish Directive. Total suspended sediment inputs to all rivers across England and Wales were estimated using a national sediment source apportionment exercise detailing the contributions from diffuse agricultural and urban sources, eroding channel banks and point sources. The total SSL estimated for each Water Framework Directive (WFD) sub-catchment (n = 7816) across England and Wales was used in conjunction with predicted flow exceedance to derive corresponding SSC time-exceedance plots. Spatial variations in modelled time-averaged SSC compared well with available monitoring data. Given the focus upon national scale, the predictive power of the SSC model (r2 = 33%) was considered realistic. The modelling approach provided a means of mapping the probability of annual average SSC being less than the 25 mg l−1 standard for GES due to sediment losses from all potential, as well as from agricultural sources only. In order to meet GES in non-compliant catchments, suspended sediment losses from diffuse agricultural sources will typically need to be reduced by up to 20%, but by as much as 80% in isolated cases.

Geomorphic response of an Appalachian Valley and Ridge stream to urbanization

AbstractHuman‐induced changes to the channel and 18·6 km2 catchment of Second Creek, in Knox County, Tennessee (USA), have included deliberate channel realignment, channelization of some reaches in culverts or cement‐lined channels, the addition of coarse particles, and intentional and unintentional changes in catchment hydrology. Field observations and measurements made between 1997 and 2001 showed active adjustment of the stream channel. Channel bank erosion is the dominant adjustment, but aggradation also occurs. One change following urbanization is an increase in bed particle size due to the addition of particles of anthropogenic origin. Such particles constitute 2–21 per cent of particles sampled at eight sites along the stream, and their D50 exceeds the D50 of natural particles at five of the sites. The downstream portion of the catchment has been urbanized for more than 150 years, but urbanizing activity has continued throughout the catchment, occurring not as a discrete perturbation, but as a set of disturbances with varying spatial and temporal scales. Spatial patterns of erosion and deposition in the channel are complex and do not show an upstream–downstream trend. Effective, although unintended, decoupling of the most manipulated reaches has hindered the propagation of changes in channel morphology and channel materials in this urbanized stream system. Copyright © 2006 John Wiley & Sons, Ltd.

Model study of the relationship between sediment yield and river basin area

AbstractThe SHETRAN physically based, spatially distributed model is used to investigate the scaling relationship linking specific sediment yield to river basin area, for two contrasting topographies of upland and more homogeneous terrain and as a function of sediment source, land use and rainfall distribution. Modelling enables the effects of the controls to be examined on a systematic basis, while avoiding the difficulties associated with the use of field data (which include limited data, lack of measurements for nested basins and inability to isolate the effects of individual controls). Conventionally sediment yield is held to decrease as basin area increases, as the river network becomes more remote from the headwater sediment sources (an inverse relationship). However, recent studies have reported the opposite variation, depending on the river basin characteristics. The simulation results are consistent with these studies. If the sediment is supplied solely from hillslope erosion (no channel bank erosion) then, with uniform land use, sediment yield either decreases or is constant as area increases. The downstream decrease is accentuated if rainfall (and thence erosion) is higher in the headwaters than at lower elevations. Introducing a non‐uniform land use (e.g. forest at higher elevations, wheat at lower elevations) can reverse the trend, so that sediment yield increases downstream. If the sediment is supplied solely from bank erosion (no hillslope erosion), the sediment yield increases downstream for all conditions. The sediment yield/basin area relationship can thus be inverse or direct, depending on basin characteristics. There still remains, therefore, considerable scope for defining a universal scaling law for sediment yield. Copyright © 2006 John Wiley & Sons, Ltd.

Dam-to-delta sediment inputs and storage in the lower trinity river, Texas

Livingston Dam on the Trinity River in SE Texas, USA disrupts the transport of sediment to the lower Trinity River and the Trinity Bay/Galveston Bay estuary. However, a sediment budget of the lower basin shows that the effects of this disruption are undetectable in the lower river. Sediment trapped in Lake Livingston is partly offset by channel erosion downstream of the dam and by inputs from the lower basin. Most importantly, however, the lower coastal plain reaches of the Trinity are characterized by extensive alluvial storage and are a bottleneck that buffers the bay from effects of upstream changes in sediment flux. Storage is so extensive that the upper Trinity basin and the lowermost river reaches were essentially decoupled (in the sense that very little upper-basin sediment reached the lower river) long before the dam was constructed. Whereas sediment storage in Lake Livingston is extensive, alluvial storage on the Trinity flood plain is even more extensive. Dam-related sediment starvation effects are noted for about 52 km downstream, and the sediment budget suggests that a majority of the sediment in this reach is likely derived from channel scour and bank erosion. The capacious alluvial storage in the lower Trinity not only limits flux to the bay, but the large amount of remobilizable alluvium also allows the system to adjust to localized sediment shortages, as illustrated in the dam-to-Romayor reach. Internal adjustments within the lower Trinity River valley thus buffer the bay from changes in sediment supply upstream.

Forgotten, But Not Gone

ENVIRONMENTAL SCIENCE Many of the most contaminated and polluted sites around the world are related to abandoned mines. Water pollution due to acidification, toxic wastes, or both is common. Heavy metals from ore processing are distributed as particles throughout a watershed in such a way that areas downstream can be more heavily contaminated than sites close to the mine. Identifying these areas and understanding how land use or future climate changes might affect watershed sediments is critical. Coulthard and Macklin present a simulation of how heavy metals have been and may continue to be distributed, focusing on the River Swale in England, an area that has been mined since Roman times. The model includes many aspects of river dynamics, such as channel migration and bank erosion, and accounts for sediment transfer down river networks. The simulations show that metals may accumulate in the downstream floodplain at levels exceeding that of deposits near the mine and that pollutants may persist locally for hundreds of years after mine closure.— BH Geology 31 , 451 (2003).

Geomorphic impacts of flash flooding in a forested headwater basin

Geomorphic impacts of a July 2001 flood in the Hungry Mother basin near Marion, Virginia, were examined to shed light on the relationships between channel characteristics and the frequency of channel-modifying discharges. Creeks in the study area have been observed for many years, with no significant channel changes since at least 1985. The 2001 flood had a recurrence interval of >200 years, and caused the only channel change, bank erosion, and transport of coarse channel material observed in recent decades. The paucity of fines in channels before or after the flood, and the absence of sub-sand sized material in the flood deposits, indicates that normal, frequent, well-below-bankfull flows are sufficient to transport the (apparently limited) supply of fines. The large particles transported during the 2001 flood after years of inactivity indicate that relatively rare floods are necessary to mobilize this material. This suggests the notion of a bimodal ‘dominant’ discharge. On the one hand frequent flows considerably below bankfull levels are sufficient to maintain the channel and prevent significant accumulation of fine sediments and organic matter. On the other hand, rare floods are necessary to transport the coarser bed material and erode channel banks. In the Hungry Mother area, bed material has no relationship to normal flows, or to flows with recurrence intervals on the order of 1–3 years. Bankfull discharge is apparently not related to either the maintenance or channel-changing dominant discharges. These results suggest that the use of channel dimensions and/or bed material as surrogate indicators of hydrologic regimes can be quite complicated, and that in some streams bankfull flow has no particular significance in terms of sediment transport and channel modifications.

Sediment sources in an urbanizing, mixed land-use watershed

The Issaquah Creek watershed is a rapidly urbanizing watershed of 144 km 2 in western Washington, where sediment aggradation of the main channel and delivery of fine sediment into a large downstream lake have raised increasingly frequent concerns over flooding, loss of fish habitat, and degraded water quality. A watershed-scale sediment budget was evaluated to determine the relative effects of land-use practices, including urbanization, on sediment supply and delivery, and to guide management responses towards the most effective source-reduction strategies. Human activity in the watershed, particularly urban development, has caused an increase of nearly 50% in the annual sediment yield, now estimated to be 44 tonnes km −2 yr −1. The main sources of sediment in the watershed are landslides (50%), channel-bank erosion (20%), and road-surface erosion (15%). This assessment characterizes the role of human activity in mixed-use watersheds such as this, and it demonstrates some of the key processes, particularly enhanced stream-channel erosion, by which urban development alters sediment loads.