DEBRIS FLOW CONTROL BY USING SLIT DAMS

1 INTRODUTIONTaiwan is a hilly-mountainous island lying across the center of the tropic. The slopeland accounts fortwo thirds of the total area of Taiwan. Most mountains are consisted of geologically young rocks incategories of fissile slates and shales which are easily eroded by weather. TOrrents combined with fissileslates and shales form debris flows with a very strong destrUctive power. This kind of debris flow canpotentially occur for almost all torrents whose gradient is steeper th…     

Experimental investigation of the effects of shrub filter strips on debris flow trapping and interception

Ecological engineering plays an increasingly significant role in mountain hazard control, but the effect of species selection and arrangement (e.g., row spacing and stem spacing) on debris flow suppression is still unclear. To further understand the interception efficiency of shrub arrangement parameters on debris flow and explore the difference with slow hydraulic erosion, sixteen sets of small-scale flume experiments with different stem and row spacings were done to study the effects of shrubs on debris flow severity, flow rate, velocity, and particle size. The results suggest that, for a dilute debris flow, sediment interception effectiveness (27.4%–60.9%) decreases gradually as stem spacing increases. Moreover, as row spacing increases, flow velocity reduction (34.4%–44.9%) and flow reduction (18.5%–47.4%) gradually decrease; and the bulk density reduction (0.5%–5.3%) and sediment interception increase initially and then decrease. In contrast, for a viscous debris flow, the flow reduction, flow velocity reduction, and sedimentation interception decrease gradually as the stem spacing increases. As row spacing increases, the flow velocity reduction, flow reduction, and sediment interception all increase initially and then decrease. A formula for the flow velocity of dilute debris flow after the filter strip was derived based on the energy conservation law and Bernoulli's equation, confirming that debris flow movement is closely related to the degree of vegetation cover. This research strengthens the current understanding of the effectiveness of vegetation in debris flow disaster prevention and control and can guide practical applications.     

Experimental study on blocking and self-cleaning behaviors of beam dam in debris flow hazard mitigation

Blocking is one of the important features when a beam dam intercepts debris flow, while self-cleaning is another when managing suspended debris flow. Both features determine the debris flow control benefits of beam dam but the latter often is not considered in practical engineering design. In this paper, a series of specially designed flume experiments were done to simulate blocking and self-cleaning processes. The blocking ratio and deposition features were measured to contrast the blocking and self-cleaning performance before and after artificial self-cleaning. The experimental results reveal that the beam dam net opening, particle diameter of sediment, sediment concentration, and gradient of the channel are the main factors affecting blocking performance. A new criterion of blocking performance of beam dams that considers the interaction of multiple factors and can provide guidance to practical project design is proposed. For all three types of blocking, sediment deposited upstream of a beam dam can be effectively transported downstream by erosion from post-debris-flow floods, Self-cleaning performance is most efficient for temporary blocking, followed by partial-blocking, and total-blocking. The efficiency of self-cleaning largely depends on the change of the sediment deposit due to erosion. Finally, a discussion is given for the optimal design of an open-type check dam and the feasibility of synergistic effects of self-cleaning in combination with artificial cleaning. Some supporting artificial silt-cleaning should be implemented in practice. A beam dam will, thus, have more storage capacity with which to control the next debris flow event.     

Hydraulic model tests for evaluating sediment control function with a grid-type Sabo dam in mountainous torrents

There are two kinds of Sabo dams in order to control sediment transport by debris flow and flash floods in mountainous area, which are closed and open-type＇s dams. In Japan, open-type＇s Sabo dams are constructed taking into account the continuity of sediment routing from upstream to downstream reach in a basin. A plan to construct a 20 m high grid-type Sabo dam which can capture a sediment volume of 400,000 m3 is proposed in the Amahata river basin in Japan. Hydraulic model tests are conducted to decide on the section for a dam （Section A, B） and the grid size such as clearance of vertical/horizontal bars for evaluating the plan. Several runs of flume tests are conducted and the sediment control function of the Sabo darn is discussed using several experimental data such as dimensionless sediment runoff rate from Sabo dam, temporal changes of bed profile and mean diameter and so on. It was found that sediment deposition in sediment storage area of Sabo dam was affected by curved channel, and that next the grid size of steel bars and thirdly the section of a dam was able to capture sediment in storage area of Sabo dam. Sediment was controlled well in the section B and in the grid size of 1.0×d95, and the problems related to sediment runoffafter sediment capturing in Sabo dam are pointed out.     

Estimation of debris flow discharge coefficient considering sediment concentration

A sabo dam has a purpose to block the path of debris flow. However, when overflow occurs, a sabo dam works as a weir, a vertical obstruction, where the fluid must flow over. Many empirical formulas and discharge coefficients for weirs relating flow depth to discharge have been proposed to calculate overflow discharges. However, only a few studies about overflow discharge coefficients are available in the case of debris flow. In this paper, experiments and numerical simulations were done to estimate debris flow discharge coefficients by considering the sediment concentration. In the numerical simulation, a complete overflow equation and a free overfall equation were implemented to calculate debris overflow discharges at a sabo dam. To determine the discharge coefficients for each equation, single factor regression analysis was used. Laboratory experiments were done to calibrate and to compare with the simulation. Study results showed that the discharge coefficients increase as the sediment concentration increases. This finding suggests debris flow discharge coefficients are derived to calculate the debris overflow discharges at a sabo dam.     

GEOCHEMICAL RECOGNITION OF SPILLED SEDIMENTS USED IN NUMERICAL MODEL VALIDATION

A fixed link (tunnel and bridge, in total 16 km) was constructed between Sweden and Denmark during 1995-2000. As part of the work, approximately 16 million tonnes of seabed materials (limestone and clay till) were dredged, and about 0.6 million tonnes of these were spilled in the water. Modelling of the spreading and sedimentation of the spilled sediments took place as part of the environmental monitoring of the construction activities. In order to verify the results of the numerical modelling of sediment spreading and sedimentation, a new method with the purpose of distinguishing between the spilled sediments and the naturally occurring sediments was developed. Because the spilled sediments tend to accumulate at the seabed in areas with natural sediments of the same size, it is difficult to separate these based purely on the physical properties. The new method is based on the geo-chemical differences between the natural sediment in the area and the spill. The basic propertiesused are the higher content of calcium carbonate material in the spill as compared to the natural sediments and the higher Ca/Sr ratio in the spill compared to shell fragments dominating the natural calcium carbonate deposition in the area. The reason for these differences is that carbonate derived from recent shell debris can be discriminated from Danien limestone, which is the material in which the majority of the dredging took place, on the basis of the Ca/Sr ratio being 488 in Danien Limestone and 237 in shell debris. The geochemical recognition of the origin of the sediments proved useful in separating the spilled from the naturally occurring sediments. Without this separation, validation of the modelling of accumulation of spilled sediments would not have been possible. The method has general validity and can be used in many situations where the origin of a given sediment is sought.     

Variability in rainfall threshold for debris flow after the Chi-Chi earthquake in central Taiwan, China

The purpose of this study is to analyze variability in rainfall threshold for debris flow （critical rainfall for debris flow triggering） after the ML 7.3 Chi-Chi earthquake in central Taiwan in 1999. Two study sites with different geological conditions were surveyed in the earthquake area. Streambed surveys were conducted to continuously monitor debris flows between 1999 and 2006. During the 7-year study period, every debris flow event was identified, and the streambed characterized. Results show that the rainfall threshold for debris flow was remarkably lower just after the Chi-Chi Earthquake, but gradually recovered. To date, this rainfall threshold is still lower than the original level prior to the earthquake. This variability in rainfall threshold is closely related to the mount of sediment material in the initiation area of debris flow, which increased rapidly due to landslides resulting from the earthquake. With the increase in sediment material, the rainfall threshold was lowered severely during the first year following the Chi-Chi earthquake. However, heavy rainfalls mobilized the sediment material, causing debris flows and transporting sediment downstream. With the decrease in sediment material, the rainfall threshold recovered gradually over time. Furthermore, debris flows occurred only in the subbasins that had sufficient sediment material to cause significant movement. Hence, these results confirm that the sediment material in the initiation area of debris flow is a crucial component of the rainfall threshold for debris flow.     

INFLUENCE OF PIPE ANGLE ON BEDLOAD TRANSPORT IN AN INCLINED PIPE

1 INTRODUCTION Sedimentation in reservoirs is a significant problem. A variety of methods have been suggested to alleviate the effects of sedimentation, but none of the methods are inexpensive and entirely effective. Perhaps soil conservation is the best option, but soil conservation only leads to a reduction in sediment load. Dredging is expensive and produces a large amount of spoil. Sediment sluicing can be effective, but for large dams it takes a great deal of time to lower and re…     

EXPERIMENTAL STUDY ON INCIPIENT MOTION OF SEDIMENT PARTICLES ON GENERALIZED SLOPING FLUVIAL BEDS

1 INTRODUCTIONWhen water flows over a fluvial bed, hydro-dynandc force induced by the flow is acting on thesediment particles lying on the bed. A further increase in flow velocity results in an increase in themagnitude of this fOrce; and sediment particles begin to move if a situation is eventu8lly reached whenthe hydro-dynandc force exceeds a certain critical value. This initial movement of sediment pallicles istermed inciPient motion. The erosion and sedimentation of nuvial beds can be…     

Impulsive force of debris flow on a curved dam

Although Sabo dams are an efficient method for river and basin management, traditional Sabo dams have a great impact on ecology and landscape. Moreover, such dams are hit and often damaged by great impulsive force when they block the debris flow. Therefore, alternative shapes for Sabo dam deserve thorough investigation. In this investigation, a curved dam was designed by changing the upstream-dam-surface geometric shape to reduce the impulsive force of the debris flow, with enhanced stability and reduced concrete mass being the anticipated outcomes. In this study, the flume and laboratory facilities simulated the impulsive force of the debris flow to the Sabo dams. Three geometric forms, including vertical, slanted and curved Sabo dams, were used to determine the impulsive force. Impulsive force theories of the debris flow were derived from the momentum equation and the Bernoulli equation. In these, the impulsive force was balanced by the friction force of the Sabo dam and the opposite force of the load cell behind the dam as it was hit by the debris flow. Positive correlations were found when comparing the experimental data with the theoretical results. These findings suggest that our impulsive force theory has predictive validity with regard to the experimental data. The results from both theory and experimental data clearly show that curved darns were sustained less force than the other darns under the same debris flow. This comparison demonstrates the importance of curved geometry for a well-designed Sabo dam.     

带弹簧支撑的新型泥石流拦挡坝抗冲击性能研究

为了解决普通重力式泥石流拦挡坝在大规模泥石流发生时容易被大块石损毁的问题,提出带弹簧支撑的新型泥石流拦挡坝,并采用钢球代替泥石流中的大块石,运用有限元软件ANSYS/LS-DYNA对这种新型坝体进行泥石流大块石的冲击模拟,从坝体的变形、支反力、Von Mises应力、加速度和应变率5个方面与普通重力式拦挡坝进行对比分析。结果表明：与普通泥石流拦挡坝相比,带弹簧支撑的新型泥石流拦挡坝的支反力可减小3/4左右,钢球的冲击加速度可降到1/3.87,坝体的响应加速度可降到1/2.61,说明带弹簧支撑的新型泥石流拦挡坝具有非常好的抗冲击性能。     

IMPACT OF DEBRIS FLOW ON RIVER CHANNEL IN THE UPPER REACHES OF THE YANGTZE RIVER

IINTRODUCTIONDebrisflowisaphenomenonhappeninginextremeseveresit'Uationofsoilerosion(Guan,1996).ItoccursfrequentlyanddistributesdenselyinupperreachesoftheYangtzeRiverandtransportsahugeamountofsedimentilltotheriver.Debrisflowdepositinfluencesthecompositionofsedimentandthemorphologyoftheriverbedfromupstreamt6downstreamandthenaffectsthedynamiccharactersandsedimentbudgetoftheYangtzeRiver.Furthermore,itimpactsonthefloodandwaterconservancyengineering.2DEBRISFLOWGULLY--HIGHWAYOFSEDI…     

WATERSHED SEDIMENT YIELD AND RANGELAND HEALTH

lINTRoDUCTIoNDifferencesintheprevailinglanduseandmanagementofaridandsemiaridareasaredeterminedinlargepartbyclimate.AridareasgenerallyreceivetoolittleprecipitationtosupportdrylandagricultureordomesticlivestockgrazingalthoughtheyaregrazedbywildIife,andattimes,bydomesticlivestock.Incontrast,insemiaridareasadequatemoistureisusuallyavaiIableatsometimeduringtheyeartoproduceforageforlivestockandwildlife,andtherearesomeyearswhendrylandcropproductionissuccessful.However,bothclimatesarecharacterize…     

Design of sediment detention basins: Scaled model experiments and application

Sediment detention basins are implemented on mountain rivers to trap solid material that may aggravate the flooding of downstream settlements.However,retention structures built in the past may unnecessarily retain sediment during non-hazardous flood events,resulting in high maintenance costs and sediment deficit downstream.In addition,the so-called spontaneous self-flushing of previously retained sediment during floods has occasionally been observed.Recent research suggests to design sediment detention basins for controlling sediment passage with a guiding channel across the deposition area upstream of a hybrid barrier.Such barriers consist of a check dam with a slot orifice and an upstream bar screen with a bottom clearance in order to benefit from a combined mechanical-hydraulic retention control.The present paper enhances this pioneering research with the help of new experimental data,including a wide range of sediment mixtures and large wood,as well as variable barrier heights.Improved design criteria are provided regarding the bar screen and the basin storage capacity.The functionality of the enhanced concept for sediment detention is illustrated by a case study on a physical model:the protection of the Villard-Bonnot village(France)against torrential hazards.     

Experimental study of the woody debris trapping efficiency of a steel pipe,open sabo dam

In recent years, the damage caused to human settlements in Japan by large woody debris (LWD) has been increasing. For example, the 2013 Izu Oshima typhoon resulted in a large number of fatalities and missing persons, and the Kagoshima Typhoon Disaster and Northern Kyusyu torrential downpour caused vast infrastructure damage due to the associated LWD. Current countermeasures for preventing LWD are insufficient to maintain the safety of residential areas. One type of protective barrier, the open sabo dam, has been constructed in Japan during the past 30 years. The primary function of open sabo dams is to block the flow of boulders, thereby also reducing sediment flow by reducing the gap size. However, because Japanese open sabo dams are designed specifically for boulder-trapping, the ability of these dams to trap LWD remains uncertain. In particular, many problems have been reported with respect to sediment trapping by driftwood with roots in an open sabo dam setting. The objective of this study was to examine the trapping efficiency of open sabo dams for LWD and sediment. The experimental approach clarified the influence of driftwood, without and with roots, on sediment trapping for a straight-channel flume. The flexible roots of the driftwood were shown to have a significant effect on the sediment trapping efficiency of the dam.     

Modelling of meander migration in an incised channel

An updated linear computer model for meandering rivers with incision has been developed. The model simulates the bed topography, flow field, and bank erosion rate in an incised meandering channel. In a scenario where the upstream sediment load decreases （e.g., after dam closure or soil conservation）, alluvial river experiences cross section deepening and slope flattening. The channel migration rate might be affected in two ways： decreased channel slope and steeped bank height. The proposed numerical model combines the traditional one-dimensional （1D） sediment transport model in simulating the channel erosion and the linear model for channel meandering. A non-equilibrium sediment transport model is used to update the channel bed elevation and gradations. A linear meandering model was used to calculate the channel alignment and bank erosion/accretion, which in turn was used by the 1D sediment transport model. In the 1D sediment transport model, the channel bed elevation and gradations are represented in each channel cross section. In the meandering model, the bed elevation and gradations are stored in two dimensional （2D） cells to represent the channel and terrain properties （elevation and gradation）. A new method is proposed to exchange information regarding bed elevations and bed material fractions between 1D river geometry and 2D channel and terrain. The ability of the model is demonstrated using the simulation of the laboratory channel migration of Friedkin in which channel incision occurs at the upstream end.     

The role of landscape setting in minimizing hydrogeomorphic impacts of flow regulation

The Tongariro Power Development Scheme (TPDS) is used to regulate flow in the headwaters of the largest catchment on the North Island of New Zealand (the Waikato). Two small dams, the Rangipo Dam and the Poutu Intake Dam, were constructed in 1973 and 1983. The flow regime of the river is managed to divert freshes into the power scheme, but allows flows larger than 100 m³ s^?1 to be released, to rework and transport sediment through the catchment. Analysis of aerial photos and maps spanning 1928 to 2007, alongside field measurements, show that there have been few hydrogeomorphic adjustments since dam construction. This includes limited changes to channel geometry, channel planform and bed material organization immediately downstream of the dams. In addition, offsite effects are minimal, both 500 m downstream of each dam, and in the more sensitive, less confined reaches in the lower catchment (11 km downstream of the Poutu Intake dam). The limited changes can be attributed to the locations of the dams within reaches characterised by bedrock gorges and confined within terraces. These locations act to flush sediments and impose margins that allow minimal adjustment of the channel. Bed material within this reach is characterised by the presence of a boulder lag. This is sourced from long-term incision into lahar deposits, and acts to limit the rate of incision, creating a steep and stable base upon which active fractions are transported. Just as importantly, significant storage in the low-relief volcanic plateau located in the upper catchment acts to disconnect and store the high sediment yields generated by active volcanic cones in the western sub-catchment upstream of the dams. This limits the rate of sediment supply to regulated reaches. Findings from this study show that analysis of reach-scale controls is essential in framing dam site locations in relation to the distribution of reaches and landscape units across the catchment. In this instance, tributary inputs downstream of the dams do not replenish the sediment and flow removed at the dam locations, as has been observed in other regulated systems. Rather, the river itself is resilient to change and flow variability is well managed allowing geomorphically effective floods to occur. Landscape setting is a key consideration in determining the hydrogeomorphic impact of flow regulation.     

Test of scale modelling of sediment transport in steady unidirectional flow

In two steady uniform flows at different physical scales in a small open channel, with variables characterizing flow, sediment, and fluid adjusted for dynamic similitude by means of four dimensionless modelling parameters (a Reynolds number, a Froude number, a density ratio, and a length ratio), measured frequency distributions of height, spacing, and migration rate of current ripples were almost identical when scaled, thus verifying that exact Reynolds-Froude modelling of loose-sediment transport is valid and workable. Modelling should be valid as well for a wide range of other transport conditions in the same kind of flow, because no additional kinds of forces or effects would be present in transport of loose grains in modes other than as ripples. In scaled-down modelling, a scale ratio of 2.5 is attainable without recourse to exotic fluids by use of water at 85°C to model natural flows at 10°C.     

Numerical modeling of sediment flushing from Lewis and Clark Lake

Lewis and Clark Lake is located on the main stream of the Missouri River. The reservoir is formed behind Gavins Point dam near Yankton, South Dakota, U.S.A. The Lewis and Clark Lake reach extends about 40 km from the Gavins Point dam. The reservoir delta has been growing since the closure of Gavins Point dam in 1955 and has resulted in a 21% reduction of storage within the maximum pool of the reservoir. Among several sediment management methods, drawdown flushing has been recommended as a possible management technique. The engineering viability of removing sediments deposited in the lake should be examined by numerical modeling before implementing a drawdown flushing. GSTARS4 was used for this study and calibrated by using measured data from 1975 to 1995. Channel cross-section changes and amount of flushed sediment were predicted with four hypothetical flow scenarios. The flushing efficiencies of all scenarios were estimated by comparing the ratios between water consumption and flushed sediment during flushing.     

Numerical modeling of local scour of non-uniform graded sediment for two arrangements of pile groups

A three-dimensional(3D) non-hydrostatic numerical model is established to investigate local scour around four aligned circular piles in uniform and non-uniform sediment mixtures and to provide information for improving scour countermeasures design. In the current study, unsteady Reynolds averaged Navier-Stokes(URANS) equations along with a Re-normalization Group(RNG) k-ε model were applied to simulate the flow field. A non-uniform sediment transport model was applied to estimate the bedload tran...