Two-dimensional numerical model for debris flows in the Jiangjia Gully, Yunnan Province

Debris flows are recurrent natural hazards in many mountainous regions.This paper presents a numerical study on the propagation of debris flows in natural erodible open channels,in which the bed erosion and sedimentation processes are important.Based on the Bingham fluid theory,a mathematical model of the two-dimensional non-constant debris flow is developed.The governing equations include the continuity and momentum conservation equations of debris flow,the sediment convection-diffusion equation,the bed erosion-deposition equation and the bed-sediment size gradation adjustment equation.The yield stress and shear stress components are included to describe the dynamic rheological properties.The upwind control-volume Finite Volume Method (FVM) is applied to discretize the convection terms.The improved SIMPLE algorithm with velocity-free-surface coupled correction is developed to solve the equations on non-orthogonal,quadrilateral grids.The model is applied to simulate a debris flow event in Jiangjia Gully,Yunnan Province and to predict the flow pattern and bed erosion-deposition processes.The results show the effectiveness of the proposed numercial model in debris flow simulation and potential hazard analysis.     

Erosion and transport mechanisms of mine waste along gullies

Mine waste debris flows continue to occur in China, and the disaster prevention and mitigation of these flows faces severe challenges since the mechanisms determining erosion and transport of mine waste along gullies are not yet fully understood. The erosion and delivery process of mine waste heaps was reproduced through flume experiments with the method based on field survey data of the Daxicha mine waste debris flow gully in the Xiaoqinling gold mining area. The results showed that the erosion and movement of mine wastes could be divided into three modes: minimal sediment movement, sediment sorting and delivery, and a large amount of sediment transfer. Moreover, there was an obvious amplification effect on peak discharge along with the formation and failure of temporary landslide dams during the erosion process. The correlation between the coefficient of peak discharge amplification and three dimensionless influencing factors, flume gradient, dimensionless volume, and dimensionless particle size, were comprehensively analyzed. An empirical formula for the coefficient of peak discharge amplification was proposed and verified based on 16 sets of experimental data. These preliminary results can provide a scientific reference for future research on disaster prevention and mitigation of mine waste debris flows.     

Influence of inflow discharge and bed erodibility on outburst flood of landslide dam

Accurate prediction of the hydrographs of outburst floods induced by landslide dam overtopping failure is necessary for hazard prevention and mitigation. In this study, flume model tests on the breaching of landslide dams were conducted. Unconsolidated soil materials with wide grain size distributions were used to construct the dam. The effects of different upstream inflow discharges and downstream bed soil erosion on the outburst peak discharge were investigated. Experimental results reveal that the whole hydrodynamic process of landslide dam breaching can be divided into three stages as defined by clear inflection points and peak discharges. The larger the inflow discharge, the shorter the time it takes to reach the peak discharge, and the larger the outburst flood peak discharge. The scale of the outburst floods was found to be amplified by the presence of an erodible bed located downstream of the landslide dam. This amplification decreases with the increase of upstream inflow. In addition, the results show that the existence of an erodible bed increases the density of the outburst flow, increasing its probability of transforming from a sediment flow to a debris flow.     

The Effects of Protruding Rock Boulders in Regulating Sediment Intrusion within the Hyporheic Zone of Mountain Streams

Excessive sedimentation in mountain stream ecosystems is a critical environmental problem due to the clogging of streambeds by sediment particles within the hyporheic zone,with detrimental effects on fish spawning habitat.In this research,the effects of an array of boulders in regulating the intrusion of incoming sand within a gravel substrate were evaluated by performing detailed experiments in a laboratory flume.A unique experimental setup and two different sampling techniques were utilized for measuring the infiltrated sand within the gravel bed under two bed shear stress conditions（moderate vs.high）.For comparison purposes,experiments were performed without and with the presence of partially submerged to the flow（protruding） boulders,which is typical for the average flow conditions found in mountain streams.Results indicated that sand infiltrated primarily in the upper part of the gravel bed creating a surface seal which hindered the penetration of sand particles deeper into the bed.An exponential decrease of the amount of the infiltrated sand within the hyporheic zone was observed in all experiments regardless of the presence of boulders.However,the presence of boulders promoted sediment intrusion of sand particles especially for the moderate applied bed shear stress condition,since the total amount of the infiltrated sand was found to be on average 44% greater whenboulders were present.The findings from this study can provide additional insight regarding the role of boulders on promoting downwelling of flow and sediment within the gravel substrate with potential effects on fish habitat.     

Sedimentary Changes of a Sand Layer in Liquefied Silts

A flume experiment was conducted to investigate the restratification of liquefied sediment strata under a wave load with the focus on the interbedded strata of coarse and fine sediments formed in estuarine and coastal areas. The aim of this research was to study the characteristics and processes of liquefied sediment strata in terms of wave-induced liquefaction. In the experiment, the bottom bed liquefied under the wave action and the liquefied soil moved in the same period with the overlying waves, and the track of the soil particles in the liquefied soil was an ellipse. The sand layer consisting of coarse particles in the upper part, settled into the lower silt layer. The sinking of coarse particles and upward migration of the fine particles of the lower layer induced by liquefied sediment fluctuations are the likely reasons for sedimentation of the sand layer in liquefied silt.     

Experimental study on characteristics of trapping and regulating sediment with an open-type check dam in debris flow hazard mitigation

Beam dams are a highly effective and commonly used open-type check dam in debris-flow hazard mitigation. In this study, dimensional analysis was used to obtain empirical equations for quantitatively determining the sediment-trapping and flow-regulating characteristics of a beam dam. To determine the coefficients of the empirical equations, flume experiments were conducted to simulate the trapping and regulating processes. The flow pattern, trapping, and regulating characteristics were investigated when debris flows passed through a beam dam. Debris-flow bulk density and peak discharge, and sediment-trapping ratios, were measured directly and indirectly. The results showed that three blocking actions occurred, and that blockage-breaking considerably influenced the trapping and regulating performance of the beam dam. The relative opening size and the sediment concentration were the two main factors affecting the performance of the beam dam. The ratio of trapping sediment decreased with relative opening, and increased with sediment concentration as well as reducing ratio of bulk density and reducing ratio of peak discharge. The sediment concentration and relative opening were the leading factors influencing the trapping and regulating sediment of a beam dam, followed by flume gradient. The results showed that the calculated values obtained using empirical equations were in good agreement with the values derived from the experiments, and that the deviation was acceptable. Finally, taking Zechawa Gully as an example, using the empirical equations we designed the opening size of a beam dam aimed to trap sediment and regulate peak discharge of debris flow in the main gully.     

Development and installation of bedload monitoring systems with submerged load cells

Bedload governs riverbed channel variations and morphology, it is necessary to determine bedload discharge through an arbitrary cross section in a mountain river. A new system with submerged load cells has been developed to directly measure bedload discharge. The system consists of: (1) an iron box which is 1 m long, 0.5 m wide and 0.1 m in depth, (2) two submerged load cells 0.7 m apart, (3) a pressure sensor and, (4) an electromagnetic velocity meter. This system has been designed to exclude the effect of the hydraulic pressure of water on direct measurements of bedload particle weight. Initial tests in a laboratory were conducted to examine the accuracy of measurements with the system under aerial conditions. The system has been installed in the supercritical flume in Ashi-arai-dani River of the Hodaka Sedimentation Observatory of the Disaster Prevention Research Institute (DPRI) of Kyoto University to obtain bedload discharge under natural conditions. Flume tests were conducted in this channel by artificial supply of uniform sediment particles of several grain sizes. The average velocity of the sediment particles near the bed was estimated using cross-correlation functions for weight waves obtained by the two load cells. Bedload discharge calculations were based on time integration of the product of sediment velocity and sediment weight obtained by the two load cells. This study clarifies the reasons why bedload measurements are difficult, and provides some solutions using the monitoring systems with submerged load cells through the field measurements. Additionally, the applicability of bedload measurement with the submerged load cells is explained based on experimental artificial sediment supply data.     

Mechanism of downcutting erosion of debris flow over a movable bed

The phenomenon of debris flow is intermediate between mass movement and solid transport. Flows can be sudden, severe and destructive. Understanding debris flow erosion processes is the key to providing geomorphic explanations, but progress has been limited because the physical-mechanical properties, movement laws and erosion characteristics are different from those of sediment-laden flow. Using infinite slope theory, this research examines the process and mechanism of downcutting erosion over a moveable bed in a viscous debris flow gully. It focuses specifically on the scour depth and the critical slope for viscous debris flow,and formulas for both calculations are presented.Both scour depth and the critical conditions of downcutting erosion are related to debris flow properties(sand volume concentration and flow depth) and gully properties(longitudinal slope,viscous and internal friction angle of gully materials,and coefficient of kinetic friction). In addition, a series of flume experiments was carried out to characterize the scouring process of debris flows with different properties. The calculated values agreed well with the experimental data. These theoretical formulas are reasonable, and using infinite slope theory to analyze down cutting erosion from viscous debris flow is feasible.     

Flume simulation of wave-induced release of internal dissolved nitrogen in Taihu Lake, China

A wave flume simulator was used to study internal nitrogen release from the surface sediment collected from Taihu Lake,China.Particulate nitrogen concentrations were positively correlated with the concentrations of suspended solids,primarily from surface erosion related to the shear stress and duration of wave action.In response to 4 cm-and 10 cm-high wave production representing waves generated in Taihu Lake by gentle and gusty winds,respectively,the mean dynamic release rate of ammonium(NH+4) from the sediment to the overlying water was 1×10-3 mg/(m2.s) and the NH4+ concentration in the overlying water increased by 0.016 mg/L,indicating that waves resulting from strong wind can induce the rapid release of dissolved nitrogen from Taihu Lake sediments.The decrease in interstitial NH+4 concentrations at all sediment depths was associated with an increase in NH4+ concentrations in the overlying water by 0.01 mg/L,showing that sediment below the eroded layer was the main source of internal nitrogen release.Changes in the interstitial dissolved oxygen and NH+4 concentrations showed that wave-induced pore water movement can greatly increase the diffusion rate,and that these effects can influence the sediment to a depth of at least 15 cm.Diffusion induced by pore water movement may be very important for the formation of an active sediment layer in Taihu Lake.     

Numerical Analysis of Adaptation-to-capacity Length for Fluvial Sediment Transport

Over the last several decades,various sediment transport capacity formulations have been used by geomorphologists and engineers to calculate fluvial morphological changes.However,it remains poorly understood if the adaptation to capacity could be fulfilled instantly in response to differing inflow discharges and sediment supplies,and thus if the calculation of morphological changes in rivers based on the assumed capacity status is fully justified.Here we present a numerical investigation on this issue.The distance required for sediment transport to adapt to capacity（i.e.,adaptation-to-capacity length） of both bed load and suspended sediment transport is computationally studied using a coupled shallow water hydrodynamic model,in line with varied inlet sediment concentrations.It is found that the adaptation-to-capacity length generally decreases as the Rouse number increases,irrespective of whether the inlet sediment concentration increases or reduces.For cases with vanishing inlet sediment concentration a unified relationship is found between the adaptation-to-capacity length and the Rouse number.Quantitatively,the adaptation-to-capacity length of bed load sediment is limited to tens of times of the flow depth,whilst that of suspended sediment increases substantially with decreasing Rouse number and can be up to hundreds of times of the flow depth.The present finding concurs that bed load sediment transport can adapt to capacity much more rapidly than suspended sediment transport,and it facilitates a quantitative criterion on which the applicability of bed load or suspended sediment transport capacity for natural rivers can be readily assessed.     

Influence of fine sediment on the fluidity of debris flows

Debris flows include a great diversity of grain sizes with inherent features such as inverse grading, particle size segregation, and liquefaction of fine sediment. The liquefaction of fine sediment affects the fluidity of debris flows, although the behavior and influence of fine sediment in debris flows have not been examined sufficiently. This study used flume tests to detect the effect of fine sediment on the fluidity of laboratory debris flows consisting of particles with various diameters. From the experiments, the greatest sediment concentration and flow depth were observed in the debris flows mixed with fine sediment indicating increased flow resistance. The experimental friction coefficient was then compared with the theoretical friction coefficient derived by substituting the experimental values into the constitutive equations for debris flow. The theoretical friction coefficient was obtained from two models with different fine-sediment treatments: assuming that all of the fine sediments were solid particles or that the particles consisted of a fluid phase involving pore water liquefaction. From the comparison of the friction coefficients, a fully liquefaction state was detected for the fine particle mixture. When the mixing ratio and particle size of the fine sediment were different, some other cases were considered to be in a partially liquefied transition state. These results imply that the liquefaction of fine sediment in debris flows was induced not only by the geometric conditions such as particle sizes, but also by the flow conditions.     

An empirical formula for suspended sediment delivery ratio of main river after confluence of debris flow

In order to calculate the suspended sediment discharge of the flood debris flows into the main river,a small scale flume test was designed to simulate the process of confluence of Jiangjia Ravine and Xiangjiang River in Yunnan province,China.By test observation and data analysis,suspended sediment discharge of Debris flow after its entry into the main river was found to have a close relation with the bulk density,the confluence angle of the Debris flow and the main river,the ratio between per unit width discharge of Debris flow and main river.Based on the measured and simulated results,and statistical analysis,an empirical formula was proposed for the suspended SDR(Sediment Delivery Ratio) of the main river after the confluence of Debris flow.Compared with the observed results of Debris flow in 2009,the error between the data calculated by the empirical formula and the monitored data is only about 10%.     

Bed load transport rates during scouring and armoring processes

Dambreak-induced bed scouring may undermine the foundation of bridge piers and other structures, and that destruction can pose a serious threat. Consequently, this paper aims at exploring the mechanisms of scouring and armoring. Firstly, the incipient velocity for nonuniform sediment particles was studied, and a formula was derived based on the angle of repose of nonuniform sediment. The results showed that the mechanism of incipient motion for sand and fine gravel differed from that for coarse gravel and cobbles. Also, comparison between experimental and field data shows that the results from the proposed formula agree well with those observed for all conditions. Secondly, a birth-death, immigration-emigration Markov process was developed to describe the bed load transport rate associated with scouring and armoring. The comparison between experimental data and computed results shows that our model can predict the bed load transport rate, although there may be some limitations, the chief of which is that there are many variables in the model to be determined through experiment. This makes its application in river engineering inconvenient.     

Using two contrasting methods with the same tracers to trace the main sediment source in a mountainous catchment

Assessing the contribution of sediments from different sources is essential to understand erosive processes and formulate further soil conservation strategies to address environmental problems.In this study,we combined the source fingerprinting approach with geochemistry,including comparison of potential sources,and distribution pattern of major elements,to estimate the different land-use relative contributions to streambed sediment in the Luowugou catchment(13.5 km~2)located on the Heishui River,the first branch of the Jinsha River,China.A total of 57 samples from stream bank,cropland,grassland,forest land,and sediments were collected,and then,15 elements were analyzed foreach sample.The fingerprinting results demonstrated that the stream bank(62.4%)had the greatest relative contribution to the bed sediment yield,while cropland,grassland,and forest land contributed 18.6%,14.9%,and 4.1%to the bed sediments,respectively.In comparison to the fingerprinting method,even though the results upon geochemistry only provide a qualitative assessment,the ranking of sediment contributions based on geochemistry was consistent with the sediment fingerprinting ranking,that is,stream bankcroplandgrasslandforest land.Our findings suggest that the focal point for sediment control practices should be the stream bank rather than cropland in the region.Geochemistry can result in an important means in validating the fingerprinting results.     

Non-capacity transport of non-uniform bed load sediment in alluvial rivers

Rivers often witness non-uniform bed load sediment transport. For a long time, non-uniform bed load transport has been assumed to be at capacity regime determined exclusively by local flow. Yet whether the capacity assumption for non-uniform bed load transport is justified remains poorly understood. Here, the relative time scale of non-uniform bed load transport is evaluated and non-capacity and capacity models are compared for both aggradation and degradation cases with observed data. As characterized by its relative time scale, the adaptation of non-uniform bed load to capacity regime should be fulfilled quickly. However, changes in the flow and sediment inputs from upstream or tributaries hinder the adaptation. Also, the adaptation to capacity regime is size dependent, the finer the sediment size the slower the adaptation is, and vice versa. It is shown that the capacity model may entail considerable errors compared to the non-capacity model. For modelling of non-uniform bed load, noncapacity modelling is recommended, in which the temporal and spatial scales required for adaptation are explicitly appreciated.     

Dynamics of loose granular flow and its subsequent deposition in a narrow mountainous river

A large amount of loose debris materials were deposited on the slope of mountainous areas after the 2008 Ms 8.0 Wenchuan earthquake. During and after the earthquake, these loose debris deposits collapsed and slide into valleys or rivers, changing river sediment supply condition and channel morphology. To investigate the mechanisms of granular flow and deposition, the dynamics of slope failure and sediment transportation in typical mountainous rivers of different intersection angles were analyzed with a coupling model of Computational Fluid Dynamics and Discrete Element Method(CFD-DEM). The numerical results show that the change of intersection angle between the granular flow flume and the river channel can affect the deposit geometry and the fluid flow field significantly. As the intersection angle increases, the granular velocity perpendicular to the river channel increases, while the granular velocity parallel to the river channel decreases gradually. Compared to the test of dry granular flow, the CFD-DEM coupling tests show much higher granular velocity and larger volume of sediments entrained in the river. Due to the river flow, particles located at the edge of the deposition will move downstream gradually and the main section of sediments deposition moves from the center to the edge of the river channel. As a result, sediment supply in the downstream river will distribute unevenly. Under the erosion of fluid flow, the proportion of fine particles increases, while the proportion of coarse particles decreases gradually in the sediment deposition. The build-up of accumulated sediment mass will cause a significant increase in water level in the river channel, thus creating serious flooding hazard in mountainous rivers.     

THE USE OF GRAIN SIZE TRENDS IN MARINE SEDIMENT DYNAMICS： A REVIEW

Spatial changes in grain size parameters (i. e. grain size trends) contain information on sediment transport patterns. An analytical procedure has been proposed to transform the grain size trends into an image of trend vectors, which may represent net sediment lransport pathways. A fundamental assumption for such an approach is that the frequency of occurrence of the trend adopted is much higher in the transport direction, than in any of other directions. Preliminary studies show agreement between this assumption and observations. However, further investigations into the physical processes and mechanisms for the formation of grain size trends are required to improve the technique, including flume experiments and numerical modeling. Moreover, attention should be paid to the trends associated with finegrained sediment, for the method of grain size trend analysis is so far designed for coarse-grained material only. The processes of flocculation during settling and the wash-load property must be considered. Appropriate interpretation of grain size data will improve our understanding of the physics of granular materials.