Effects of particle size and bulk density on erosion of quartz particles. Journal of Hydraulic Engineering, 12 , — Soulsby, R. The bottom boundary layer of shelf seas. Elsevier oceanography series, 35, — Dynamics of marine sands: a manual for practical applications. Thomas Telford. Sumer, B. The mechanics of scour in the marine environment. World Scientific. Tavouktsoglou, N. Whitehouse, R. Scour at marine structures: A manual for practical applications.
Dynamics of estuarine muds. Evaluating scour at marine gravity foundations. Winterwerp, J. Introduction to the physics of cohesive sediment dynamics in the marine environment Vol. Yao, W. Personal communication based on unpublished experimental results. University of Western Australia. Zhao, M. Experimental and numerical investigation of local scour around a submerged vertical circular cylinder in steady currents.
Coastal Engineering, 57 8 , — Experimental study of local scour around subsea caissons in steady currents. Coastal Engineering, 60, 30— Log in to your subscription Username. Peer reviewed only Published between: Published from year: and Published to year: Advanced search Show search help.
Scour caused by turbulence downstream of a bridge.
Bed control using a downstream sill Indicative layouts of sacrificial piles Sacrificial piles upstream of a bridge Indicative layout of vanes Debrishce deflector upstream of a bridge pier Recommended footing locations Principles of pile and pile cap location Concrete repairs to a bridge abutment Typical gradings for different stone sizes Examples of gabion scour protection Interlocking concrete block protection to bridge Examples of grout-filled mattress protection Scour reduction due to protective collars Toe and falling apron details Correct and incorrect ways of arresting pier scour Scour protection at bridge abutments and piers Typical pipeline protection Erosion protection downstream of weir structure Catchment area Cross-sectional area below bank level for regime channel Coefficientin Equation 4.
If the depth of scour becomes significant, the stability of the foundations may be endangered, with a consequent risk of the structure suffering damage or failure. There have been several cases of bridge failures - some causing loss of life and most resulting in significant transport disruption and economic loss - as a result of scour. Many of the well-known cases are mentioned in Section 2.
Further information on some selected failures is included among the case studies in Appendix 2. The factors influencing the development of scour are complex and differ according to the type of structure. Protection works for preventing scour need to be designed to withstand the flow forces imposed on them, and have to be practicable to build and install while minimising environmental effects.
Although it may be greatly affected by the presence of structures encroaching on the channel, scour is a natural phenomenon caused by the flow of water over an erodible boundary.
Scour at marine structures : a manual for practical applications
In a river, scour is normally most pronounced when the bed and river banks consist of granular alluvial materials. It also occurs in cohesive materials, such as clay, and even deeply weathered rock can be vulnerable in some circumstances. Further information on these types of scour is given in Chapter 2; additional definitions appear in the Glossary. Previous page is blank ClRlA C 25 1. The difficulty of making field measurements at structures during high flows tends to hide the potential seriousness of the problem, because scour holes often fill in again after the peak of a flood has passed.
The consequent lack of reliable field data also makes it difficult to verify predictions of potential scour depths obtained from small-scale laboratory tests. To achieve this objective, it has been necessary to combine up-todate technical information with the practical experience that has been gained by engineers in constructing and operating schemes in the field. The manual covers: e different types of scour local, contraction and natural e the scour characteristics of different types of structures ranging from bridge piers to linear revetments e the effects of different types of flow conditions such as uni-directional fluvial flows, bi-directional tidal flows, jet flows from sluices e the design of new structures, and the assessment and repair of existing structures e permanent structures and temporary works e similarities and differences between UK conditions and larger-scale problems overseas e alternative protection systems and methods of installation e assessment of scour risks e costs and benefits of protection works e environmental issues.
Although estuarine conditions are included, the manual is mainly concerned with the fluvial environment and, in particular, scour induced by currents. It does not cover marine scour nor scour predominantly by wave action, for which reference should be made to Scour ut niurine structures Whitehouse, The manual contains limited coverage of risk assessment and the costs and benefits of scour protection works, with the objective of identifying the key issues that need to be considered and suggesting general ways in which these factors should be assessed.
However, it is not possible to give detailed methodologies and quantitative data on risks and costs. Risk assessment is a complex subject and many of the factors that affect scour cannot be determined accurately; risk assessments of different options are therefore likely to be relative. Similarly, estimates of the costs resulting from scour failures or the construction costs of protection schemes vary considerably from case to case, depending on the type, size and location of the structure.
These enable the structures to be screened in terms of whether scour is a significant factor and whether further study or additional protection works may be needed. The assessment procedures are not superseded by the present manual and they should continue to be used for Railtrack and Highways Agency schemes. The manual complements the procedures by providing information on a wider range of structures and topics, such as the design, construction and maintenance of protection schemes.
The flow conditions are very different in those cases and the scale of the problems usually requires them to be investigated on a case-by-case basis. Three aspects of scour assessment and protection are also outside the scope of this manual: 0 0 0 the estimation of flood discharges details of methods for determining or predicting flow conditions at structures for example, backwater calculations and flood-routeing analysis the design of river training works.
Although these issues are described in general terms, readers are referred to other documents for detailed guidance on them. It should be noted that current knowledge about scour is varies greatly. Scour at bridge piers has been the subject of considerable study, for example, whereas little published information is available about scour at pipelines and revetments.
Although the authors have endeavoured to make this manual as comprehensive as possible, some parts of the book inevitably contain more detailed and definitive information than others. Table 1. Appendices Contents 1 Monitoring equipment 2 Case studies Finally, it is important to stress that various factors interact to create scour problems at hydraulic structures. Solutions to these problems therefore require a multi-disciplinary approach. For major projects, inputs from hydrologists, geomorphologists, geotechnical engineers, foundation engineers, as well as from hydraulic engineers, need to be carefully integrated at all stages of the work from preliminary planning through to detailed design, construction and operation.
ClRlA C 27 1. The document is also relevant to designers of non-wet structures on river floodplains, such as access tunnels through road embankments, that are normally dry but that may be at risk from scour from flows occurring during flood conditions. The manual contains six chapters and two appendices see Table 1. There are some variations in terminology and definitions of which readers should be aware.
Nevertheless, it remains widely used and respected. There are several other relevant publications in the same series, as listed in the Bibliography. ClRlA C 29 2 Scour processes The types of scour dealt with in this mariual result from erosion of the channel boundaries by flowing water.
The amount of scour that can occur at a structure placed in the flow, and the speed at which the scour develops, depend on: 0 the position and type of structure 0 the flow conditions affecting it 0 the characteristics of the channel boundary materials in the vicinity of the structure and in the upstream reach. These factors, together with further discussions on the classification of scour and the types of scour failure to which various structures are vulnerable, are the subject of this chapter. Each of the factors that contribute to scour is subject to a significant degree of uncertainty or difficulty in making long-term predictions.
Information available on major floods at the design stage may be limited and, during the life of a structure, the flow conditions may be altered by changes in catchment usage or climate. The responses of natural channels to erosion in short-term floods and over longer periods are hard to predict accurately, partly because of an incomplete understanding of the physical processes involved and partly because they interact in a complex way and are affected by random factors.
Therefore, the risk of a particular depth of scour occurring and of it causing damage to a structure cannot be determined in the same way and to the same accuracy as structural design parameters, where loading, material properties and responses are known more precisely. The question of risk assessment is discussed more fully in Section 6.
Definitions in the literature vary, so it is important to be clear in the definitions used, avoiding where possible those whose meaning may be ambiguous. It is more convenient to consider them in the reverse order, however, starting with the most widespread and moving to the most local. Marine scour, boat scour and scour resulting from high-head jets and energy dissipation are also briefly considered. Degradation, aggradation and regime conditions Degradation and aggradation are the processes of long-term erosion and deposition of bed material in a river, perhaps over a decade or a century, that affect its longitudinal profile.
They normally occur as a series of progressive steps, predominantly during floods, but exclude the more localised effects of scour during a particular flood event. Degradation usually appears as a general lowering of bed levels along a reach of river, and is caused by the reach seeking to adjust its longitudinal gradient to match the requirements of the flows and sediment loads that it carries. If the sediment load entering the reach is lower than the actual transport capacity within the reach, degradation starts at the upstream end and works its way downstream, so as to reduce the overall longitudinal gradient.
However, if the channel downstream of the reach in question has a greater sediment transport capacity, degradation starts at the downstream end of the reach and works its way upstream, leading to an overall increase in the longitudinal gradient. In the case of aggradation, the above causes and effects are reversed. Clearly, channel degradation is the more critical condition when considering scour at structures. From this description of the processes involved, it can be seen that, to obtain early warning of channel degradation problems at a structure, it is necessary to monitor long-term changes in the river over significant distances upstream and downstream.
However, the stable regime conditions to which the river has become adjusted may be disturbed by changes resulting from natural processes andor human interference. ClRlA C The assessment of degradation and aggradation can be a specialist matter, but guidance for routine cases, in particular for evaluating the regime conditions for the design flood, is given in Chapter 3. Further information is available in textbooks such as Pemberton and Lara , Breusers and Raudkivi Channel migration Channel migration may occur naturally or as a result of human activity, and may be associated with any of the causes that give rise to degradation and aggradation.
Migration of the entire river channel as part of the process of meander progression, or movement of the deep-water channel within the same overall channel banks, can affect the scour exposure of a bridge or other structure whose foundations may have been fixed in relation to an earlier channel position. In some cases, migration may occur rapidly in response to a particular flood event, but in other cases it may be gradual.
In a braided channel, the channel positions are continuously changing. The deepest natural scour is likely to be associated with the confluence of two major channels, downstream of a bar or island in the channel. Taking account of the potential for channel migration is an important part of the design or assessment of fluvial and estuarine structures.
As a general rule, if there is potential for channel migration, the foundations should be designed or assessed on the basis of any credible shifts of the deep-water channel or channels. Alternatively, training works may be carried out to limit the possible movement of the deep-water channel. A particular feature of the flow curvature is the generation of secondary spiral currents, which tend to increase the scour towards the outside of the bend. Bend scour, which is illustrated in Figure 2. Melville and Coleman describe the process as follows: Typically, the two streams of flow from converging channels meet at the centreline of the confluence, plunge to the channel bed, and then return to the water surface along the sides of the confluence.
They go on to point out the resemblance to the case of two bends placed back to back. A major part of the contraction is often due to the approach embankments to a bridge, which cause the flows on the floodplain to join the main channel and pass through the bridge opening Figure 2.
Numerical simulation of scour around pipelines using an Euler–Euler coupled two-phase model
It can be estimated from considerations of the stability of the bed material in relation to the flow conditions. It is normal practice to assume that the amount of contraction scour varies across the channel width and distribution factors are available for a variety of situations. Contraction scour may occur in both clear-water or live-bed conditions, as discussed in Section 2.
The structures increase the local flow velocities and turbulence levels and, depending on their shape, can give rise to vortices that exert increased erosive forces on the adjacent bed. As a result, the rates of sediment movement and erosion are locally 34 ClRlA C enhanced around the structures, leading to local lowering of the bed relative to the general level of the channel. Local scour has received wide attention fiom researchers, mainly because it is amenable to physical model testing, resulting in the availability of suitable estimation formulae for a variety of situations.
Particular or unusual cases can also be modeltested as needed. Formulae are available for predicting the equilibrium scour depth in clear-water and live-bed conditions and also for predicting the scour depth during the development phase, before equilibrium is reached. Information on local scour is available for the following features: 0 bridge piers of various shapes 0 bridge abutments and other similar structures 0 river training works and linear revetments 0 spur dikes groynes and other banks at various angles to the flow direction 0 weirs and sills 0 closures of cofferdams and diversion works.
In some cases, the extent, shape and surface gradients within the scour hole can also be predicted. Vertical "wake" vortices caused by flow separation from the sides of the pier can also erode the bed, but, in fluvial conditions, the deepest scour tends to occur at the upstream face of the structure, as a result of the action of the horseshoe vortex. Material eroded fiom this hole is usually deposited towards the downstream end of the structure, to a level above that of the surrounding bed.
The wake vortices are transported downstream by the flow and can create twin longitudinal scour holes; this type of scour may need to be considered if there is another structure farther downstream that is located within the wake created by the fvst structure. ClRlA C 35 As the scour develops, the increase in local flow depth decreases the strength of the erosive action at the bed; as a result, the rate of scour decreases and eventually reaches an equilibrium.
In the case of clear-water scour see Section 2. In this manual, each of these components of the total scour is evaluated separately in the sequence given above, with the local bed elevation resulting from each component being taken as the starting condition for the estimation of the next component Figure 2.
There are, however, additional considerations arising from such issues as tidal flows, littoral drift, the interaction of tidal and fluvial currents in estuaries, and normally a greater exposure to wave action. There are alternative up-to-date guidance documents on marine scour, in particular Scour at marine structiires Whitehouse, and Chapter 7 of Scow manual HofEnans and Verheij, The present manual does not cover marine scour, and is mainly concerned with the fluvial environment, in particular with scour induced predominantly by currents.
In some cases the effects are likely to be minimal, due to the short durations and small velocities in relation to the natural velocities. In canals, the passage of boats is likely to be the principal cause of any scour. Methods of calculating the currents and wave heights produced by boats moving in restricted waterways are given in PIANC Supplenient to Bulletin No 57 High-head scour Erosion of bedrock occurs in high-head situations, affecting massive hard igneous rocks as well as weaker sedimentary rocks.
The primary mechanism for rock detachment in such situations is generally attributed to pressurisation of the joints. This is beyond the scope of the present manual, but further information can be found in various specialist technical papers for example, Mason, , and Whittaker and Schleiss, The bed materials may be non-cohesive typically gravels, sands and silts or cohesive typically, silts and clays. Where the bed material is a competent unweathered bedrock, or is only slightly weathered, then it is essentially resistant to the forms of scour excluding the action of high-head jets that come within the scope of this manual.
Such scour as does occur in these cases would be expected to be governed by the rate of weathering of the bedrock, rather than by the hydraulic conditions, so would not normally be significant within the design life of any bridge or other hydraulic structure. In the cases covered by this manual, scour is therefore essentially a phenomenon concerned with the interaction between the hydraulic conditions and the mobility of readily erodible bed materials.
In some cases, the quantification of the scour depth is virtually independent of the bed material: in other situations, the nature of the bed materials is an important factor and can be affected by issues such as armouring of the scour hole. Non-cohesive sediments Non-cohesive sediments are characterised by a granular structure, with individual particles being susceptible to erosion when the applied fluid forces drag and lift are sufficient to overcome the stabilising forces due to gravity and contact with adjacent bed particles.
I ClRlA C 37 A great deal of research has been carried out into the threshold of movement of uniform sediments in unidirectional flow, the best known being by Shields in Germany in the s. See Box 4.
- Manual on scour at bridges and other hydraulic structures - PDF Free Download.
- The 2-Hour Job Search: Using Technology to Get the Right Job Faster!
- Evaluating scour at marine gravity foundations;
- Fatal Error.?
- Travesties and Transgressions in Tudor and Stuart England: Tales of Discord and Dissension!
The threshold of movement depends on the particle size, density, shape, packing and orientation in the bed. The first two are normally the most important factors, as larger and denser particles are able to resist higher flow velocities. In practical situations, however, the particles are not uniform, the bed has undulations that affect the local hydraulic resistance, and the local velocities and stresses vary spatially and temporally, so that the initiation of movement occurs intermittently and randomly.
Threshold of movement calculations are a key feature of various components of scour estimation, in particular for regime conditions and for contraction scour. Cohesive sediments typically require relatively large forces to detach the particles and initiate movement, but relatively small forces to transport the particles away. HofEnans and Verheij describe the process as follows: Experiments In the initial stage loosened particles and aggregates separate and those with weakened bonds are washed away.
This process leads to the development of a rougher surface. Higher pulsating drag and lift forces increase the vibration and dynamic action on the protruding aggregates. As a result the bonds between aggregates are gradually destroyed until the aggregate is instantaneously tom out of the surface and carried away by the flow. There has been some recent research relating competent velocities to the soil cohesion and other parameters, but the erosion properties of cohesive bed materials are not yet fully understood, particularly in relation to local scour around structures.
As an interim measure, an approximate method for estimating local scour in cohesive materials is proposed in Section 4. In cases where a more resistant layer overlies a less resistant layer, then a conservative approach needs to be taken regarding the risks of the scour breaking through the more resistant layer into the less resistant layer. Reorientation of lensshaped or elongated stones can also be a feature of river bed armouring.
Armouring can be relevant in the consideration of all forms of scour in gravel and cobble bed rivers, but is most likely to be an issue when evaluating regime conditions, degradation and contraction scour. The general approach is to consider the scour in successive depth bands, estimating the particle sizes and corresponding volumes of material that are liable to be washed away in each layer, then revising the grading of the next layer to reflect the addition of the uneroded material from above.
This differential scouring process continues until there is a complete layer of material large enough to resist finther erosion. If a major flood produces more severe flow conditions than those under which the armouring developed, there is a risk that the armour layer could be washed away. This would leave the underlying, non-armoured material exposed and could result in a rapid increase in the depth of scour. Similar problems can occur if sections of channel bed are dredged, or reinstated with excavated material. For these reasons, a cautious approach should be adopted when estimating the possible effects of bed armouring on scour depths, especially in cases of clear-water scour, when there is no supply of potential armour material from upstream.
As the flow increases, so this clear-water scour increases and may, in many granular bed materials, reach a significant depth before there is general sediment movement in the river. Once the flow has increased to the point at which general bed movement occurs, there is then a supply of sediment from upstream to offset the local removal of material. When the flow reduces, clear-water scour may occur again. Both contraction scour and local scour may occur under clear-water and live-bed conditions.
The water in clear-water scour may not literally be clear, as it may contain finer sediments that remain wholly in suspension and do not affect the scouring processes at the bed. At the transition from clear-water scour to live-bed scour, the supply of material from upstream may exceed the rate of local erosion, resulting in a reduction in the depth of the contraction or local scour.
In many cases, there can thus be an initial maximum scour depth at about the point when general movement of the river bed begins, followed by a reduction in scour depth as the flow increases and more material is supplied from upstream. With uniform granular materials, this initial clear-water depth of scour may be reached at moderate flow rates and may not be exceeded even under severe flood conditions.
In the case of graded sediments, however, the scour depth at the limit of clear-water flow is normally less than it would be with a uniform material; increases in velocity that produce live-bed scour are likely to result in greater depths of ClRlA C 39 erosion, but the maximum value is normally no greater than the maximum occurring with a uniformly graded sediment. In some circumstances, for example in channels with coarse or armoured bed material or in vegetated floodplains, or in moderate floods, only clear-water scour occurs.
Clearwater scour approaches the maximum depth asymptotically, generally over a longer period than live-bed scour, and it may take several floods before the maximum is reached. Live-bed scour can reach its maximum much more rapidly and, in cases where the upstream channel contains dunes, the depth tends to fluctuate about a mean equilibrium value.
Live-bed scour Average amount of live-bed scour Clear-water scour Figure 2. Richardson and Davis state that the maximum local pier scour under clear-water conditions is about 10 per cent greater than the equilibrium mean scour depth under live-bed conditions. They also quote a typical fluctuation of live-bed scour in sand rivers of about 10 per cent either side of the equilibrium value, indicating that the maximum values of clear-water and live-bed scour are about equal.
Maximum scour depths are likely to occur near the peak of a flood, and during the recession stages the holes may be partially refilled. This phenomenon, which has been observed from the strata exposed in trial pits in scour holes, is principally a feature of local, contraction, bend and confluence scour in live-bed conditions, where bed material is transported from upstream. However, it can also occur over time following clear-water scour, due to the deposition of finer material or slumping of the scour slopes.
It is important to ensure that, where field measurements are used as part of the scour assessment, the basis of the field measurements is properly documented and understood. Equipment is available that allows either continuous monitoring of scour at structures, or that records the maximum scour reached during an event, but there are practical problems that can affect its successful deployment. Further information is given in Appendix 1.
During the initiation phase erosion is rapid, and material eroded from the upstream slope of the scour hole goes into suspension. In the development phase, the scour hole deepens substantially. The upper part of the upstream slope is in approximate equilibrium and development occurs beyond it, enlarging the hole while its shape remains approximately constant. The development of the hole results in a progressive reduction in near-bed velocities and the rate of erosion.
During the stabilisation phase, the rate of erosion at the base of the hole is very small, but erosion continues at the more exposed position near the top of the downstream slope, resulting in the scour hole lengthening and equilibrium being nearly achieved. In the equilibrium phase, the dimensions of the scour hole are virtually fixed.
In the case of structures such as piers and piles, scour usually first develops along either side of the structure and is deepest at the points of maximum width where the flow velocities are highest. As the holes deepen, they increase in extent and unless the structure is extremely wide join to form a single scour hole around the upstream end of the structure. This hole then continues to deepen towards the equilibrium shape, with the maximum depth normally occurring on the centreline of the upstream face unless the flow is not well aligned with the axis of the structure.
All forms of scour may be subject to tidal effects, with the potential for scour to evolve in steps in response to the tidal flows. In some cases, such as contraction scour, the scour is likely to progress in a similar manner in either flow direction; for local scour around a bridge pier, the equilibrium shape of the scour hole differs for each flow direction; in other cases, such as scour downstream of a sill or apron, scour is likely to occur in different locations on the flood and ebb tides. Consideration needs to be given to the following factors in evaluating scour in tidal locations: 0 the range of tidal amplitude between neaps and springs 0 the peak tidal velocities and the relationship between flood and ebb velocities 0 ClRlA C the peak velocities in fluvial floods and storm surges and their superimposition on the tidal velocities 0 the different locations andor shapes of scour likely under different flow directions 0 the cumulative effect of a series of tides 0 the potential for scour development in a single spring tide andor fluvial flood.
Scour is a potential contributory factor in all cases, and is believed to be the most common cause of failure or damage to bridges during floods, although the immediate cause may be another factor, such as water pressures on a bridge deck, or debris impact forces. Examples and statistics of failures are presented next, with information on failure mechanisms associated with scour and other hydraulic factors being given in the following sections.
The discussion is based mainly on bridges, but the same considerations generally apply to other structures. Richardson and Davis cite several studies of bridge failures in the USA, some of which are summarised below. A more extensive national study in indicated that local pier scour and abutment scour were about equal in occurrence. The flood in the upper Mississippi basin caused 23 bridge failures, of which 14 were from abutment scour, two from pier scour, three from a combination of abutment and pier scour, two from lateral migration, one from debris load and one unattributed.
In a report for the Transport and Road Research Laboratory, HR Wallingford listed the following notable bridge failures due to scour: 0 0 Schoharie Creek bridge, New York State, USA : 10 killed see Appendix 2 Glanrhyd railway bridge, Wales : train ran off partially collapsed bridge, four killed see Appendix 2 0 Wraysbury railway bridge, England : train derailed see Appendix 2 0 River Ness railway viaduct, Scotland : failed during flood 0 Hatchie River bridge, Tennessee, USA : three spans collapsed, eight fatalities see Appendix 2 0 Okuragawa road bridge, Japan : collapsed, 10 killed 0 Kufstein motorway bridge, Austria : pier sank and rotated several metres due to scour Ireland : failure of or damage to 18 small bridges in Wicklow, Leitrim and Waterford, mostly due to scour.
ClRlA C Waitangitaona River road bridge failure , in which one pier and the two adjacent deck spans were lost due to scour, debris accumulation and oblique flow against the pier. Waipaoa River rail bridge failure Cyclone Bola, , in which the abutment and three adjacent piers were affected by scour, the abutment being outflanked and then undermined by bend migration. Oreti River road bridge, affected by degradation aggravated by gravel extraction upstream since construction in and requiring a downstream rock weir to stabilise the bed see Appendix 2.
Bullock Creek road bridge subject to aggradation caused by upstream landslides, reaching 1 m above the deck during a flood in , after which the bridge was removed and replaced at a new site. A recent UK study J Benn, personal commimicufion ,using library searches and questionnaire surveys of bridge owners, identified bridge failures or partial failures due to scour and floods since , of which 54 could reasonably be assigned to a scour failure and 1 1 to water pressure. Of the 54, it is notable that 3 1 were due to undermining or slippage of the approach embankment, rather than scour failure of the actual bridge piers or abutment.
One single event in August accounted for the loss of 11 railway bridges between Berwick and Dunbar. A notable finding of the study was that failure of normally dry access bridges and cattle creeps was more common than failure of bridges over watercourses. Scour in the vicinity of structures may be due to the combined effects of natural, contraction and local scour, and may also be increased by the effects of navigation. A bridge constructed on spread footings is at risk of scour failure when the scour reaches the level of the base of the footing. However, it could be at risk with less scour if the substructure is subject to lateral ground pressures and water forces.
Scour adjacent to piled foundations may result in a loss of skin friction and loadbearing capacity. The piles may also be subject to unplanned bending stresses, from lateral loads and hydrodynamic forces. Local scour at a bridge pier is normally greatest near the upstream nose of the pier, which may lead to the pier being undermined first at the upstream end and thus tilting.
Similar effects may occur at abutments or with groups of piles. Variations in bed and flow conditions may also lead to tilting in other directions. Climate change, if it produces larger peak discharges or changes other relevant factors such as the typical annual hydrograph, could result in increased risks of scour failures in the future.
A bridge or other structure that is sited to suit one location of the main channel may become progressively vulnerable to scour failure as the river attempts to migrate. Abutments or piers located on the original floodplain, if not designed to accommodate channel migration, may be undermined or otherwise destabilised if this occurs.
ClRlA C 43 Changes in channel alignment can also be caused by poorly designed training works or by the uncoordinated construction of other structures upstream; examples of the latter include jetties, and rock weirs or groynes built to provide improved conditions for fish and other fauna. Maintenance repairs involving the placing of rock protection around bridge piers can reduce the flow area of the main span and lead to flow being diverted towards other channels or openings.
In an ideal situation, the forces are in line with the axis of the channel and piers, but in some cases channel movements and other factors may affect the flow direction and the resulting forces to a degree not anticipated in the design. The ability of the structure to withstand the hydraulic forces depends on the foundation design and may be compromised by the scour which occurs. Debris accumulation may also contribute. Such forces can be dangerous because of the overturning moment applied to the foundation, the risk of lifting a simply supported bridge deck off its supports, or the risk of reducing the compression forces around an arch.
The risks of failure would be greater if aggravated by scour, particularly if the scour happens to be greater on the downstream side of the bridge. For small single-span bridges, blockage by up to 90 per cent of the bridge opening has been known; this can lead to large rises in upstream water levels, flooding and overtopping. Accumulation of debris against a bridge structure can increase the amount of scour due to: 0 0 the increased effective width of the pier which is a significant factor in the amount of scour the increased velocities resulting from the flow constriction and the rise in upstream head.
Debris can also be a contributory factor in bridge failure due to: 2. Ice Ice can impose forces against structures due to its expansion during freezing, but this appears unlikely in the fluvial environment. Probably the greater risk comes from the impact of sheets of ice carried in the flow after the ice starts to thaw and break up. Table 2.
Scour at Marine Structures: A Manual for Practical Applications - Richard Whitehouse - Google книги
It is extremely important when calculating the three components to adopt a consistent system for expressing bed levels and scour depths. As an example, in the case of contraction scour, it would be the bed level after allowance has been made for the effects of natural scour. In a major flood this is likely to be below the long-term or regime value. Similarly, the level of the water surface varies according to the particular flow conditions; therefore, if scour depths are being considered for two possible flood events, the datums for determining the two values of scour depth would differ.
For these reasons, in this manual, the final results of scour calculations are expressed in terms of absolute levels relative to an appropriate datum and not in terms of water depths or depths below initial bed level. The return period, N, of a flood is most conveniently defined as the average time interval between years in which a given value of discharge is reached or exceeded. The reciprocal of the return period therefore represents the probability of that discharge being equalled or exceeded in any year.
Detailed methods of determining values of flood discharge are outside the scope of this manual, but the following techniques can be used. It is unlikely that the data would have been obtained in flow conditions as severe as the specified design flood so the results usually need to be extrapolated.
Guidance on how to take stream flow measurements and process the results is given in Parts 3 and 3F of British Standard BS The period for which the records are available is often considerably less than the return period of the design flood, so statistical methods must be used to estimate flow rates for rarer events. Information on these techniques is given in the Flood estimation handbook Institute of Hydrology, and textbooks on hydrology eg Wilson, If the length of the reach of river between the ClRlA C Previous page is blank 47 gauging station and the site is significant, or it contains a tributary or distributary , the values of discharge need to be suitably adjusted.
An approximate estimate can be obtained by multiplying the discharge at the gauging station by the ratio between the catchment area at the site and that at the gauging station. However, for greater accuracy, it is necessary to carry out flood-routeing calculations using onedimensional or two-dimensional numerical models to allow for inflows from the catchment and for the attenuation effects that reduce the magnitudes of flood peaks as they travel downstream.
Statistical methods based on correlations between flood discharge and catchment characteristics such as area, slope, soil type etc. These methods usually only give values of discharge and not information about the duration and shape of the flood hydrographs. In the UK, estimates can be obtained using the information given in the Flood estimation handbook.
RainfalVrunoff methods using statistical data on rainfall and information on the catchment characteristics. This information is used to predict the unit hydrograph for the catchment ie its response to a unit amount of rainfall and is then combined with the rainfall data to determine the shape of the flood hydrograph and the value of peak discharge. Comparison with available data on flood discharges for adjacent catchments or those with similar characteristics.
Calculation of flow rates in the river from survey data defining the cross-sectional shape and longitudinal gradient of the channel and adjacent floodplains where appropriate - see Boxes 3. For major schemes where suitable records from flow gauging stations are not available, it is recommended to use more than one of the alternative methods to check that the estimates obtained are consistent and realistic.
In the case of structures subject to tidal flows in estuaries, the design discharge can be made up of three possible components: 0 a fluvial flow corresponding to the design flood conditions in the upstream catchment a a tidal component produced by astronomical tides occurring approximately twice a day, with the tidal range and the maximum tidal velocity varying in complex spring and neap cycles over periods of months and years 0 a surge component caused by low atmospheric pressure or onshore winds raising sea levels and pushing additional water inland.
The relative importance of these components depends on the size of the river, the magnitude of the tidal range and the relative distance of the site from the upstream tidal limit. In the UK, the tidal component is likely to be dominant if the width of the river or estuary is greater than about m at the site position. It can be noted that the fluvial flow increases the velocities occurring during the ebb portion of a tide and correspondingly decreases the velocities during the incoming flood tide.
On a large lowland river, the duration of the main flood peak is likely to cover both the ebb and flood portions of a tide. The hydraulic resistance of the channel is defined in terms of the Manning coefficient n. However, controls such as weirs and channel constrictions downstream of the site can cause backwater effects that prevent Se being even approximately equal to S. In these cases, backwater calculations may be carried out using a suitable o n e dimensional numerical model to determine the flow conditions at the site more accurately.
Photographs of different types of channel with their corresponding values of n are given by Chow and, for UK rivers, by Hollinrake and Samuels ClRlA C 49 Box 3. Also, in certain cases of contraction scour at bridges, the amount of flow collected by embankments and channelled through the main opening may need to be calculated. The combination of a deep channel with one or more shallower channels having different values of Manning resistance coefficient, n see Box 3.
The following is an approximate method for calculating flow rates in the type of compound channel shown above. The channel cross-section is viewed from upstream and for illustrative purposes shows a bridge with abutments set back from the main channel on either floodplain. Embankments on parts of the floodplain cause the flows in channels 2 and 5 to be diverted through the main opening of the bridge. The Manning resistance coefficient of the incised channel is n, and that of the floodplains is n,.
It is assumed that the level of the water surface, Z m above datum , is the same across the full width of the river and that, upstream of the bridge, the floodplain flows are parallel with that in the main channel, that is, each channel has the same value of energy gradient, Se. As explained in Box 3. The perimeter P, is the the inclusion of the lengths J-N and P-K as part of the distance corresponding to the line J-NP-K; boundaries of the channel is to allow for the drag resistance exerted on the flow in the main channel by the slower moving flows on the floodplains.
The flow in each of the floodplain channels is given by the general formula: A,,, is the flow area within a channel up to the water level, Z. The wetted perimeter, P,,,, is the length of the floodplain boundary within a channel: thus, for channel 2 , P2 is equal to the length of the line L-M, and for channel 3, P3 is equal to the length M-N note that the line J-N is not included for the floodplain channel.
Using equations 3. The problem is more complex in a meandering river, because the path length followed by the flow in the main channel is greater than that on the floodplains so the values of energy gradient, Se,are not necessarily equal. However, the meanders also tend to increase the energy losses on the floodplains.
For this reason, it is normally recommended to use the same value of Se for all the component channels and to calculate it using the path length measured along the main deep-water channel. As an example, unnecessarily severe design conditions might be specified if they were to be based on a year fluvial flood occurring at the same time as the highest astronomical tide and a maximum storm surge, since the return period of this combined event would be very much greater than years. If scour is a significant factor in the design of a tidal scheme, the cost of the required study should be small in relation to the benefits resulting from a more realistic design specification.
Possible sources of information include: 0 a new topographic and bathymetric survey specially commissioned for the project 0 data from previous surveys, if available 0 large-scale survey maps 0 aerial photographs 0 satellite images. For most major projects it is usually necessary to commission a new survey to obtain the specific information needed. However, comparison with older data can be important in identifying longer-term changes and trends in the position and shape of the main channels, particularly for meandering or braided rivers.
The amount of survey information needed depends to a certain extent on the nature of the engineering works but, as a rough guide, cross-sections of the main channel should normally be obtained at intervals of about 50 m in the immediate vicinity of the works. For the floodplains, a larger spacing of about m might be appropriate if their geometry is fairly regular. Contour surveys of the banks will also be necessary in the vicinity of the structure if it is necessary to design protection works.
For some projects, it may be necessary to carry out numerical backwater calculations to establish design water levels at the site see Section 3. The spacings needed between cross-sections farther away from the site should be selected depending on the slope of the river. Some recommended values are given in Table 3. Table 3. The distance over which the survey data is needed depends on the identification of a suitable point for the start of the backwater calculations eg a gauging station, weir or similar control point.
If the river is steep enough to produce supercritical flow in the channel see Box 3. ClRlA C 51 3. In all these cases, it is very important to check and compare the datums used for any information obtained on water levels. Data from different sources are often based on different topographical surveys, some of which may be old and poorly documented.
Mistakes and confusion about the relationships between level datums tend to be common and can have serious consequences for the design and construction of a project. Simple common-sense checks between data sets can often highlight any serious inconsistencies. The datums used in tide tables and navigation charts for particular ports or estuaries are usually chosen so as to indicate minimum water depths for navigation purposes, with positive values on a chart representing the depth of the bed below the chart datum; it is therefore important to establish carefully the relationship between the chart datum and the national datum used for levels on land.
In the case of fluvial rivers, Method 3 above is normally necessary to predict water levels for design flood conditions if the return period of the event is significantly greater than the period for which records are available. For a relatively straight and uniform reach of river, reasonable estimates of flood level may be obtained using the Manning resistance equation and data on the longitudinal gradient of the channel see Boxes 3. However, if the channel geometry varies significantly with distance or there is a downstream control, such as a weir or a junction with a major tributary, it may be necessary to carry out backwater calculations, either by hand or using a suitable computer program.
Computer programs may be classified according to the number of dimensions in which properties of the flow such as the velocity are calculated. Thus, one-dimensional 1-D numerical models predict the variation of mean velocity and water level with distance along the channel. Full 3-D solutions also provide information on the variation of flow velocity with depth within the flow.
Generally, the cost and complexity of a study increases significantly in moving from a 1-D to a 2-D simulation. A 1-D model often gives satisfactory results for fluvial rivers whose cross-sectional properties do not vary too rapidly with distance along the channel. However, in wide estuaries and in fluvial rivers with complex channel and floodplain geometries, it may be necessary to use a 2-D numerical model or a physical model to obtain reliable estimates of flood level.
Use of 3-D numerical models is not normally justified, unless there is also a need to obtain detailed information about local flow conditions at particular locations or structures within the flow. The following definitions are used in this manual. It is assumed that cross-sectional information is available at several points along the river or channel in the vicinity of the site, as shown above. Provided that the variations between the cross-sections are not too large, the following procedure is recommended for calculating representative figures for the bank-full values.
Plot the average best-fit straight line through these points and use this line to determine a separate value of mean bank-full level, z b f m , at each cross-section. A similar averaging technique can be used to determine representative values of channel geometry for other specified water levels and also for the floodplains. However, since for scour calculations it is usually necessary to consider the lowest likely bed levels, it is recommended to calculate Y using the largest value of YmaXobserved at any of the cross-sections included in the averaging process.
If the planning, design and construction of a major project are likely to last for more than about two years, it is highly recommended that water level gauges be installed at the site as early as possible in the programme and a regular system of monitoring arranged. If the site cannot be attended on a regular basis, simple gauges that record the maximum water level occurring between readings can provide very useful information about flood events.
ClRlA C 53 For structures in tidal flows, the heights of astronomical tides can usually be determined from navigation tables or from water levels measured by a local tide gauge. Care is needed to establish what is the correct relationship between the chart datum used for navigation purposes and the national datum.
When determining the maximum design water level for protection works, it is normally necessary to add allowances for the effects of possible storm surges and fluvial floods see Section 3. When estimating possible scour depths, it is worth noting that it is more appropriate to use the mean tide level plus allowances for surges and flood flows as the representative design value, because this is the level at which the maximum tidal velocities normally occur. Flow depths can be calculated in a variety of ways, but particular definitions used in this manual are included in Box 3.
For fluvial rivers and channels, it is unlikely that flow measurements have previously been made under the flood conditions adopted for design. The mean velocity, V d s , over the full cross-section of the flow is given by: In the case of a compound channel consisting of an incised channel with adjacent floodplains, the methods given in Box 3. For design purposes it is normally necessary to use the value of local velocity adjacent to the structure, which in general is not equal to the section-mean velocity V.
Information on typical velocity distributions in channels and around structures can be used to determine multiplying factors that can be applied to the value of V; details of these factors are given in Box 3. Field measurements of flow velocity at the site can be useful even if the conditions under which they were measured are not as severe as specified by the design case. Velocities in the main channel of a river with floodplains tend not to increase substantially when the river is out-of-bank.
Therefore, data obtained at or near bank-full conditions can provide a good indication of velocities that would occur in larger floods. If velocity measurements are made at points across the width of a river, the information can be used to determine multiplying factors for relating local values of velocity to the section-mean velocity, V. This avoids the need to rely on assumed factors, such as those given in Box 3.