Skip to contentUnited States Department of Transportation - Federal Highway AdministrationFHWA HomeFeedback

Hydraulics Engineering


Bridge Scour and Stream Instability Countermeasures: Experience, Selection, and Design Guidance-Third Edition

Chapter 3



As previously noted, a countermeasure is defined as a measure incorporated into a highway‑stream crossing system to monitor, control, inhibit, change, delay, or minimize stream and bridge stability problems. A monitoring program at structures during and/or after flood events and river stabilizing works over a reach of the river up and downstream of the crossing can also be considered countermeasures.

Countermeasures may be installed at the time of highway construction or retrofitted to resolve stability problems at existing crossings. Retrofitting is good economics and good engineering practice in many locations because the magnitude, location, and nature of potential stability problems are not always discernible at the design stage, and indeed, may take a period of several years to develop.

A countermeasure does not need to be a separate structure, but may be an integral part of the highway. For example, relief bridges on floodplains are countermeasures which alleviate scour from flow contraction at the bridge over the main stream channel. Some features that are integral to the highway design serve as countermeasures to minimize stream stability problems. Abutments and piers properly aligned with the flow reduce local scour and contraction scour. Also, reducing the number of piers and/or setting back the abutments reduces contraction scour.

Countermeasures which are not integral to the highway may serve one function at one location and a different function at another. For example, bank revetment may be installed to control bank erosion from meander migration, or it may be used to stabilize streambanks in the contracted area at a bridge. Other countermeasures are useful for one function only. This category of countermeasures includes spurs constructed in the stream channel to control meander migration.

In selecting a countermeasure it is necessary to evaluate how the stream might respond to the countermeasure, and also how the stream may respond as the result of other activities upstream and downstream.

A countermeasure for scour critical bridges and unknown foundations could also be monitoring a bridge during and/or after a flood event. If monitoring is selected and if the risk of scour failure is high, protection to reduce the risk such as riprap or instrumentation should be provided. Even if riprap is placed around piers or abutments, the high risk bridge should be monitored during and inspected after floods. If monitoring is selected, a plan of action must be implemented which includes a notification process, flood watch procedures, a highway closure process, documentation of available detours, inspection procedures, assessment procedures, and a repair notification process.

The next section provides some general criteria for the selection of countermeasures for stream instability and scour. Then, the selection of countermeasures for specific stream instability and bridge scour problems is discussed.


The selection of an appropriate countermeasure for a specific erosion or scour problem is dependent on factors such as the erosion or scour mechanism, stream characteristics, construction and maintenance requirements, potential for vandalism, and costs. Perhaps more important, however, is the effectiveness of the measure selected in performing the required function.

For example, protection of an existing bank line may be accomplished with revetments, spurs, retards, longitudinal dikes, or bulkheads (see Chapter 2 and Table 2.1). Spurs, longitudinal dikes, and area retardance structures can be used to establish a new flow path and channel alignment, or to constrict flow in a channel. Because of their high cost, bulkheads may be appropriate for use only where space is at a premium. Channel relocation may be used separately or in conjunction with other countermeasures to change the flow path and flow orientation.

3.2.1 Erosion Mechanism

Bank erosion mechanisms include surface erosion and/or mass wasting. Surface erosion is the removal of soil particles by the velocity and turbulence of the flowing water. Mass wasting is by slides, rotational slip, piping and block failure. In general slides, rotational slip and block failure result from the bank being undercut by the flow. Also, seepage force of the pore water in the bank is another factor that can cause surface erosion or mass wasting. The type of mechanism is determined by the magnitude of the erosive forces of the water, type of bed and bank material, vegetation, and bed elevation stability of the stream. These mechanisms are described in HDS 6 (Richardson et al. 2001) and HEC-20 (Lagasse et al. 2001a).

3.2.2 Stream Characteristics

Stream characteristics that influence the selection of countermeasures include (see also Table 2.1):

  • Channel width
  • Bank height
  • Channel configuration
  • Channel material
  • Vegetative cover
  • Sediment transport condition
  • Bend radius
  • Channel velocities and flow depth
  • Ice and debris
  • Floodplain characteristics

Channel Width. Channel width influences the use of bendway weirs and other spur‑type countermeasures. On smaller streams (250 ft (<75 m) wide), flow constriction resulting from the use of spurs may cause erosion of the opposite bank. However, spurs can be used on small channels where the purpose is to shift the location of the channel. Changes in channel width can influence contraction scour and impact bridge pier and abutment countermeasures.

Bank Height. Low banks (<10 ft (3 m)) may be protected by any of the countermeasures, including bulkheads. Medium height banks (from 10 to 20 ft (3 to 6 m)) may be protected by revetment, retardance structures, spurs, and longitudinal dikes. High banks (>20 ft (6 m)) generally require revetments used alone or in conjunction with other measures.

Channel Configuration. Spurs and jack fields have been successfully used as a countermeasure to control the location of the channel in meandering and braided streams. Also, bulkheads, revetments, and riprap have been used to control bank erosion resulting from stream migration. On anabranching streams, revetments, riprap, and spurs have been used to control bank erosion and channel shifting. Also, secondary or side channels that do not carry large flows can and have been closed off. In one case, HDS 6 reports that a large channel was closed off and revetment and riprap used to control erosion in the other channel (Richardson et al. 2001).

Channel Material. Spurs, revetments, riprap, jack fields, or check dams can be used in any type of channel material if they are designed correctly. However, jack fields should only be placed on streams that carry appreciable debris and sediment in order for the jacks to cause deposition and eventually be buried. For channels in the sand size range the movement of bedforms (ripples, dunes, and antidunes) will influence the design of bridge pier and abutment scour countermeasures (see HEC-20 and HDS 6). Some countermeasures may be subject to abrasion and consequent failure from coarse bed load sediment.

Bank Vegetation. Vegetation such as willows can enhance the performance of structural countermeasures for stream instability and may, in some cases, reduce the level of structural protection needed. Meander migration and other bank erosion mechanisms are accelerated on many streams in reaches where vegetation has been cleared.

Sediment Transport. Sediment transport conditions can be described as regime, threshold, or rigid. Regime channel beds are those which are in motion under most flow conditions, generally in sand or silt‑size noncohesive materials. Threshold channel beds have no bed material transport at normal flows, but become mobile at higher flows. They may be cut through cohesive or noncohesive materials, and an armor layer of coarse‑grained material can develop on the channel bed. Rigid channel beds are cut through rock or boulders and rarely or never become mobile. In general, permeable structures will cause deposition of bed material in transport and are better suited for use in regime and some threshold channels than in rigid channel conditions. Impermeable structures are more effective than permeable structures in channels with little or no bed load, but impermeable structures can also be very effective in mobile bed conditions. Revetments can be effectively used with mobile or immobile channel beds. Live-bed and clear-water conditions must be considered when estimating contraction scour and will influence the design of pier scour countermeasures (Richardson and Davis 2001).

Bend Radius. Bend radius affects the design of stream instability countermeasures, because some countermeasures will only function properly in long or moderate radius bends. Thus, the cost per meter (foot) of bank protection provided by a specific countermeasure may differ considerably between short‑radius and longer radius bends. Impinging flow and increased shear stress in short radius bends also affect the stability of scour countermeasures.

Channel Velocities and Flow Depth. Channel hydraulics affect the selection and design of all countermeasures. For stream instability countermeasures, structural stability and induced scour must be considered. Some of the permeable flow retardance measures may not be structurally stable and countermeasures which utilize piles may be susceptible to scour failure in high velocity environments. Velocity and depth have a direct influence on the design of bridge pier and abutment scour countermeasures (see Chapter 4).

Ice and Debris. Ice and debris can damage or destroy any countermeasure and should always be considered during the selection process. On the other hand, the performance of some permeable spurs and area retardance structures is enhanced by debris where debris accumulation induces additional sediment deposition.

Floodplains. In selecting countermeasures for stream stability and scour, the amount of flow on the floodplain is an important factor. For example, if there is appreciable overbank flow, then the use of guide banks to protect abutments should be considered. Also, spurs perpendicular to the roadway approach embankment may be required to control erosion.

3.2.3 Construction and Maintenance Requirements

Standard requirements regarding construction or maintenance such as the availability of materials, construction equipment requirements, site accessibility, time of construction, contractor familiarity with construction methods, and a program of regular maintenance, inspection, and repair are applicable to the selection of appropriate countermeasures. Additional considerations for countermeasures located in stream channels include: constructing and maintaining a structure that may be partially submerged at all times, the extent of bank or channel bed disturbance which may be necessary, and the desirability of preserving streambank vegetative cover to the extent practicable.

3.2.4 Vandalism

Vandalism is always a maintenance concern since effective countermeasures can be made ineffective by vandals. Documented vandalism includes dismantling of devices, burning, and cutting or chopping with knives, wire cutters, and axes. Countermeasure selection or material selection for construction may be affected by concerns of vandalism. For example, rock‑filled baskets (gabions) may not be appropriate in some urban environments.

3.2.5 Countermeasure Selection Based on Cost

Cost comparisons should be used to study alternative countermeasures with an understanding that the measures were installed under widely varying stream conditions, that the conservatism (or lack thereof) of the designer is not accounted for, that the relative effectiveness of the measures cannot be quantitatively evaluated, and that some measures included in the cost data may not have been fully tested by floods.

Life-cycle cost information is difficult to quantify. Initial construction costs are relatively easy to develop; however, even for a specific countermeasure, these costs can vary widely depending on regional availability of materials, site conditions, and access constraints. Therefore, a countermeasure type can be very cost effective in one locale and prohibitively expensive in another. Extending these issues to life-cycle maintenance requirements requires an even broader set of assumptions. Riprap, for example, is a standard countermeasure type in many states; however, alternatives to riprap may need to be investigated because of cost and availability limitations. The risks and consequences of failure at any given site further complicate the issue. Life-cycle costs were , however, the focus of a countermeasure selection methodology for a range of bridge pier scour countermeasures.

A selection methodology was developed under NCHRP Project 24-07(2) (Lagasse et al. 2007) to provide a quantitative assessment of the suitability of six armoring-type pier scour countermeasures based on selection factors that consider river environment, construction considerations, maintenance, performance, and estimated life-cycle cost. With the exception of life-cycle costs, the methodology analyzes the design factors by stepping the user through a series of decision branches, ultimately resulting in a site-specific numerical rating for each selection factor. Countermeasures evaluated by this methodology are:

  • Standard (loose) riprap
  • Partially grouted riprap
  • Articulating concrete blocks
  • Gabion mattresses
  • Grout filled mattresses
  • Grout filled bags

Five factors are used to compute a Selection Index (SI) for each countermeasure:

  • S1: Bed material size and transport
  • S2: Severity of debris or ice loading
  • S3: Constructability constraints
  • S4: Inspection and maintenance requirements
  • LCC: Life-cycle costs

The Selection Index is calculated as:

SI = (S1 x S2 x S3 x S4)/LCC(3.1)

The countermeasure that has the highest value of SI is considered to be most appropriate for a given site, based not only on its suitability to the specific riverine and project site conditions, but also in consideration of its economy. The approach is sensitive to assumptions regarding initial construction cost, remaining service life, assumed frequency of maintenance events, and extent of maintenance required. Each of these factors requires experience and engineering judgment, as well as site- or region-specific information on the cost of materials and delivery, construction practices, and prevailing labor rates. It should be noted that the methodology can be used simply to rank the countermeasures in terms of suitability alone by assuming that the life cycle costs are the same for all countermeasures.

The following sections describe the five factors that comprise the methodology. Flow charts illustrating selection factors S1 - S4 are included in Appendix C.

Bed Material. Bed material is included as a selection factor for two reasons. Abrasion caused by the transport of coarse bed sediments will cause the wire mesh on a gabion mattress to weaken and break, whereas other countermeasure types are relatively resistant to degradation by abrasion. For this reason, when bed material is greater than 2 mm, gabion mattresses are eliminated from the selection process. Bed material size also assists in distinguishing whether dune-type bedforms are anticipated. Grout filled mats are susceptible to failure in the presence of bedforms because the mats do not articulate as well as other countermeasures. When the bed material is less than 2 mm and bedforms are not anticipated, all countermeasures included in the selection process are deemed equally viable.

Ice and/or Debris Loading. Debris in this context is considered floating material such as logs, other woody materials, man-made materials that are typically transported during floods, or ice. The intent of this selection factor is to recognize that high debris loads can be detrimental to gabion mattresses, as indicated in the countermeasure selection matrix in Chapter 2. When a user indicates that anticipated debris loading is high, gabion mattresses receive a low rating of "1" but are not eliminated from the selection process. When debris loading is not anticipated, all countermeasures included in the selection process are deemed equally viable.

Construction Constraints. Construction constraints take into account the different needs and challenges required for placing a countermeasure in the dry versus installation underwater or, in the extreme case, in flowing water. All ratings that consider construction constraints are divided into two categories: piers that have shallow footings versus piers that are more deeply embedded. This results from the fact that riprap-based countermeasures are typically thicker than alternative countermeasures, and they require pre-excavation that may undermine the footer.

In addition, the requirements for specialized equipment are addressed. For example, the equipment requirements, placement techniques, and construction QA/QC requirements for partially grouted riprap are straightforward for working in the dry; however, placement underwater requires construction equipment and placement technologies that are much more sophisticated. Subgrade preparation requirements and placement tolerances also vary among countermeasure types. For example, a relatively thin veneer of articulating concrete blocks requires finer grading techniques than an equivalent, and much thicker, riprap layer.

Working beneath a bridge deck that affords little headroom will dictate the type of equipment that can be used for countermeasure installation. Lastly, alternative placement techniques, particularly for rock riprap, typically dictate the strength requirements for geotextiles in order to meet criteria for geotextile survivability during installation.

The decision box for flow velocity is intended to reflect the relative difficulty in placing a mattress system, such as ACBs, gabion mattresses, or grout mattresses under fast flowing water V > 4 ft/s (1.2 m/s). When the countermeasure does not need to be placed under water and access for construction equipment of all types is good, all countermeasures included in the selection process are deemed equally viable.

Inspection and Maintenance. Inspection and maintenance guidelines vary greatly among countermeasure types. Underwater or buried installations require different considerations to ensure that the countermeasure can be adequately inspected, compared to surficial treatments in ephemeral or intermittent stream environments. The numerical values assigned to this selection factor reflect the relative difficulty of repairing and/or replacing "manufactured" countermeasures, such as ACBs, gabion mattresses, and grout mattresses, versus the relative ease of adding more riprap stone.

The maintenance required for gabion mattresses as presented may be somewhat higher than for other forms of revetment because the wire mesh used to construct the gabion is susceptible to vandalism. When the countermeasure can be inspected and maintenance performed in the dry, all countermeasures included in the selection process are deemed equally viable.

Life-Cycle Costs. The Selection Index calculation is similar to the "Risk Priority Number" method suggested by Johnson and Niezgoda (2004). Johnson and Niezgoda use the concept of "risk categories" in contrast to this selection methodology concept of "suitability categories" to relate various factors. Both methods represent relatively simple techniques for selecting pier scour countermeasures. However, due to the complexity of determining costs associated with countermeasure design and implementation, Johnson and Niezgoda discussed life cycle costs but did not include those costs in the scope of their procedure.

Without consideration of life-cycle cost, the suitability of a countermeasure is dictated solely by the environment of the river and its interaction with the bridge structure, combined with the strengths and vulnerabilities of the countermeasure. This selection methodology attempts to simplify the life-cycle cost estimation process through a series of spreadsheets that assist the user in evaluating regional availability of materials, installation expenses, and an estimation of maintenance based on experience and engineering judgment (see Appendix C).

Life-cycle cost information can be difficult to quantify. Initial construction costs are relatively easy to develop; however, even for a specific countermeasure, these costs can vary widely depending on regional availability of materials, site conditions, and access or constructability constraints. Therefore, a particular countermeasure might be very cost effective in one locale and prohibitively expensive in another. Extending these issues to life-cycle maintenance requires an even broader set of assumptions. This portion of the assessment attempts to ease this process for the practitioner by providing templates for cost estimation.

Estimating life-cycle costs for pier scour countermeasures requires consideration of three major components:

  1. Initial construction materials and delivery costs
  2. Initial construction installation costs associated with labor and equipment
  3. Periodic maintenance during the life of the installation

Each of the above components is comprised of multiple elements, which differ among the various countermeasure types. For example, quantities and unit costs of alternative materials will vary depending on the specific project conditions, as well as local and regional factors. Experience with these factors, as well as project-specific knowledge of the bridge site, are required in order to be as accurate as practicable when using this selection methodology.

The following issues should be considered when developing life-cycle cost estimates:

  • Availability of materials of the required size and weight
  • Haul distance
  • Site access
  • Equipment requirements for the various countermeasures being considered
  • Construction underwater vs. placement in the dry
  • Environmental and water quality issues and permitting requirements
  • Habitat and/or migration issues for threatened and endangered species
  • Traffic control during construction and/or maintenance activities
  • Local labor rates
  • Construction using in-house resources vs. outside contract
  • Design life of the installation
  • Anticipated frequency and extent of periodic maintenance and repair activities

Quantifying each of these factors requires experience and engineering judgment. For this reason, these variables are user inputs in the life cycle cost worksheets referenced in Appendix C. The default values that are provided in the Excel spreadsheet program can and should be changed by the user to reflect both site-specific and state or regional conditions . For access to the Excel spreadsheet, see Forward to Lagasse et al. 2007.

3.2.6 Countermeasure Selection Based on Risk

As suggested by the various scenarios described in Chapter 2, a risk-based analysis may be necessary to develop the plan of action for multiple bridges with scour critical ratings. The level of response and the actions taken will be different from one scour critical bridge to the next. Given limited resources and multiple options, it is up to the interdisciplinary team to formulate the best alternative for any given plan of action considering all available information.

The need to evaluate countermeasure alternatives considering economics and risk led to the development of a systematic, risk-based method to determine the level of resources appropriate for protection of a bridge that is scour critical but has a limited life before scheduled replacement (Pearson et al. 2000). This method determines the optimum level of protection for the bridge and the maximum expenditures that should be accepted to increase the level of protection. An overview of the concepts involved in this approach is provided in this section. Equations to apply the method and an application example are provided in Pearson et al. (2000).

The objectives of the risk-based approach to countermeasure selection are to:

  • Identify bridges at risk from scour using data from the National Bridge Inventory (NBI)

  • Examine the economic feasibility of scour countermeasure alternatives

For convenience, NBI data is used to estimate the relative risk of bridge failure due to scour, where risk is defined as the costs associated with bridge failure multiplied by the probability of failure. The NBI codes can be used to estimate both the cost of failure and the probability of failure as shown in Figure 3.1.

Flow chart to determine Risk showing how Bridge Situation leads to Failure Probability which then can be adjusted by Bridge Age to get a Revised Failure Probability. This Revised Failure Probability and Economic Factors are used to determine Expected Losses and thus Risk in dollars - with details provided in the text
Figure 3.1. Risk-based approach using NBI data.

The failure probability can be derived from the NBI Item 113 coding or, alternatively, from a combination of route class, substructure condition, channel protection, and waterway adequacy. The failure probability can be adjusted based on bridge age (NBI Item 27). Economic factors that influence costs associated with bridge failure include: bridge length, width, and classification, detour length, and average daily traffic (ADT).

Thus, the risk (expected loss) calculated by this method is the product of the probability of scour failure (or heavy damage) and the economic losses associated with such an event. The year-to-year risk (expected loss) of scour failure associated with a bridge installation over water is determined by Equation 3.2.

Risk ($) = KP [(rebuild cost) + (running cost) + time cost)] (3.2)


Risk ($)


Risk of scour failure (annualized for 1 year given current physical condition)



Risk adjustment factor based on foundation type and type of span (NBI Item 43), and



Probability of failure each year (NBI Item 113 or Items 26, 60, 61, and 71)

Rebuild cost, running cost, and time cost can be developed from equations or graphs provided in Pearson et al. (2000).

With the risk of bridge failure established in terms of cost, the risk can be compared with the costs and benefits associated with a range of countermeasure alternatives. This can be done by balancing costs and risks by using a simple benefit/cost ratio, or by a net benefit analysis for candidate countermeasures. As noted by Pearson et al. (2000), balancing the costs and risks represents a traditional approach to risk management-finding the countermeasure cost that corresponds as closely as possible to the risk cost. This approach, however, hides the economic benefits offered by available countermeasures. Maximizing the net benefit illuminates the economic benefits of countermeasures but may result in countermeasure costs greater than necessary to achieve a particular protection goal. Maximizing the benefit/cost ratio offers the soundest economic approach to countermeasure selection since it results in the greatest countermeasure benefit per dollar spent. For an application of risk-based concepts, see Pearson et al. (2000).


The best countermeasure against meander migration is to locate the bridge crossing on a relatively straight reach of stream between bends. At many such locations, countermeasures may not be required for several years because of the time required for the bend to move to a location where it becomes a threat to the highway facility. However, bend migration rates on other streams may be such that countermeasures will be required after a few years or a few flood events and, therefore, should be installed during initial construction [see HEC-20 (Lagasse et al. 2001a)] for further discussion of lateral channel instability, and NCHRP Report 533 (Lagasse et al. 2004) for a methodology for estimating rates of meander migration).

Stabilizing channel banks at a highway stream crossing can cause a change in the channel cross section and an increase in stream sinuosity upstream of the stabilized banks. Figure 3.2a illustrates a natural channel section in a bend with the deeper section at the outside of the bend and a gentle slope toward the inside bank resulting from point bar growth. Figure 3.2b illustrates the scour which results from stabilizing the outside bank of the channel and the resulting steeper slope of the point bar on the inside of the bend. This effect must be considered in the design of the countermeasure and the bridge (see Section 4.3.5). It should also be recognized that the thalweg location and flow direction can change as sinuosity upstream increases.

Figure 3.3a illustrates meander migration in a natural stream and Figure 3.3b, the effects of bend stabilization on upstream sinuosity. As sinuosity increases, meander amplitude may increase, meander radii will become smaller, deposition may occur because of reduced slopes, and the channel width‑depth ratio may increase as a result of bank erosion and deposition, as at the bridge location shown in Figure 3.3b. Ultimately, cutoffs can occur. These changes can also result in hydraulic problems downstream of the stabilized bend.

Sketch (a) of a natural channel cross section in a bend and sketch (b) of a channel cross section with stabilized near vertical outside channel bank and deep scour at toe of bank
Figure 3.2. Comparison of channel bend cross sections (a) for natural conditions, and (b) for stabilized bend (after Brown 1985).

Countermeasures for meander migration include those that:

  • Protect an existing bank line
  • Establish a new flow line or alignment
  • Control and constrict channel flow

The classes of countermeasures identified for bank stabilization and bend control are bank revetments, spurs, bendway weirs, longitudinal dikes, vane dikes, bulkheads, and channel relocations. Also, a carefully planned cutoff may be an effective way to counter problems created by meander migration. These measures may be used individually or in combination to combat meander migration at a site. Some of these countermeasures are also applicable to bank erosion from causes other than bend migration.

Sketch (a) illustrates meander migration in a natural stream progressing downstream with approximately the same sinuosity and Sketch (b) illustrates stabilization of downstream bend geometry at crossing, increasing upstream sinuosity and consequently potential channel cutoff
Figure 3.3. Meander migration in (a) a natural channel, and (b) a channel with stabilized bend (after Brown 1985).


Channel braiding occurs in streams with an overload of sediment, causing deposition and aggradation. As aggradation occurs, the slope of the channel increases, velocities increase, and multiple, interconnected channels develop. The overall channel system becomes wider and multiple channels are formed as bars of sediment are deposited in the main channel.

Braiding can also occur where banks are easily eroded and there is a large range in discharge. The channel becomes wider at high flows, and low-flow forms multiple interconnected channels. In an anabranched stream, flow is divided by islands rather than bars, and the anabranched channels are more permanent than braided channels and generally convey more flow.

A meandering stream may change to a braided stream if the slope is increased by channel straightening or the dominant discharge is increased. Lane's relation may be used to determine if there can be a shift from a meandering channel to a braided one. If, after a change in discharge or slope the stream still plots in the meandering zone, then it will likely remain a meandering stream. However, if it moves closer to or into the braided zone, then the stream may become braided (see HEC-20).

Braided channels change alignment rapidly, and are very wide and shallow even at flood flow. They present problems at bridge sites because of the high cost of bridging the complete channel system, unpredictable channel locations and flow directions, difficulties with eroding channel banks, and in maintaining bridge openings unobstructed by bars and islands.

Countermeasures used on braided and anabranched streams are usually intended to confine the multiple channels to one channel. This tends to increase the sediment transport capacity in the principal channel and encourage deposition in secondary channels. These measures usually consist of dikes constructed from the margins of the braided zone to the channel over which the bridge is constructed. Guide banks at bridge abutments (Design Guideline 15) in combination with revetment on highway fill slopes (Design Guideline 4), riprap on highway fill slopes only, and spurs (Design Guideline 2) arranged in the stream channels to constrict flow to one channel have also been used successfully.

Since anabranches are permanent channels that may convey substantial flow, diversion and confinement of an anabranched stream is likely to be more difficult than for a braided stream. The designer may be faced with a choice of either building more than one bridge, building a long bridge, or diverting anabranches into a single channel.


Bed elevation instability problems are common on alluvial streams. Degradation in streams can cause the loss of bridge piers in stream channels and can contribute to the loss of piers and abutments located on caving banks. Aggradation causes the loss of waterway opening in bridges and, where channels become wider because of aggrading streambeds, overbank piers and abutments can be undermined. At its worst, aggradation may cause streams to abandon their original channels and establish new flow paths which could isolate the existing bridge (see HEC-20 for further discussion of vertical channel instability).

3.5.1 Countermeasures to Control Degradation

Countermeasures used to control bed degradation include check dams and channel linings. Check‑dams and structures which perform functions similar to check‑dams include drop structures, cutoff walls, and drop flumes. A check‑dam is a low dam or weir constructed across a channel to prevent upstream degradation (Design Guideline 3).

Channel linings of concrete and riprap have proved unsuccessful at stopping degradation. To protect the lining, a check‑dam may have to be placed at the downstream end to key it to the channel bed. Such a scheme would provide no more protection than would a check dam alone, in which case the channel lining would be redundant.

Bank erosion is a common hydraulic hazard in degrading streams. As the channel bed degrades, bank slopes become steeper and bank caving failures occur. The USACE found that longitudinal stone dikes, or rock toe‑dikes (see Chapter 8), provided the most effective toe protection of all bank stabilization measures studied for very dynamic and/or actively degrading channels.

The following is a condensed list of recommendations and guidelines for the application of countermeasures at bridge crossings experiencing degradation:

  • Check‑dams or drop structures are the most successful technique for halting degradation on small to medium streams.
  • Channel lining alone may not be a successful countermeasure against degradation problems.
  • Riprap on channel banks and spill slopes will fail if unanticipated channel degradation occurs.
  • Successful pier protection involves providing deeper foundations at piers and pile bents.
  • Jacketing piers with steel casings or sheet piles has also been successful where expected degradation extends only to the top of the original foundation.
  • The most economical solution to degradation problems at new crossing sites on small to medium size streams is to provide adequate foundation depths. Adequate setback of abutments from slumping banks is also necessary.
  • Rock‑and‑wire mattresses are recommended for use only on small (100 ft [<30 m]) channels experiencing lateral instability and little or no vertical instability.
  • Longitudinal stone dikes placed at the toe of channel banks are effective countermeasures for bank caving in degrading streams. Precautions to prevent outflanking, such as tiebacks to the banks, may be necessary where installations are limited to the vicinity of the highway stream crossing.
3.5.2 Countermeasures to Control Aggradation

Currently, measures used in attempts to alleviate aggradation problems at highways include channelization, debris basins, bridge modification, and/or continued maintenance, or combinations of these. Channelization may include dredging and clearing channels, constructing small dams to form debris basins, constructing cutoffs to increase the local slope, constructing flow control structures to reduce and control the local channel width, and constructing relief channels to improve flow capacity at the crossing. Except for debris basins and relief channels, these measures are intended to increase the sediment transport capacity of the channel, thus reducing or eliminating problems with aggradation. Cutoffs must be designed with considerable study as they can cause erosion and degradation upstream and deposition downstream. These studies would involve the use of sediment transport relations given in HDS 6 (Richardson et al. 2001) or the use of sediment transport models such as BRI-STARS (Molinas 1990) or HEC-6 (USACE 1993). The most common bridge modifications are increasing the bridge length by adding spans and increasing the effective flow area beneath the structure by raising the bridge deck.

A program of continuing maintenance has been successfully used to control problems at bridges on aggrading streams. In such a program, a monitoring system is set up to survey the affected crossing at regular intervals. When some pre-established deposition depth is reached, the bridge opening is dredged or cleared of the deposited material. In some cases, this requires clearing after every major flood. This solution requires surveillance and dedication to the continued maintenance of an adequate waterway under the bridge. Otherwise, it is only a temporary solution. A debris basin or a deeper channel upstream of the bridge may be easier to maintain. Continuing maintenance is not recommended if analysis shows that other countermeasures are practicable.

Over the short term, maintenance programs prove to be very cost effective when compared with the high cost of channelization, bridge alterations, or relocations. When costs over the entire life of the structure are considered, however, maintenance programs may cost more than some of the initially more expensive measures. Also, the reliability of maintenance programs is generally low because the programs are often abandoned for budgetary or priority reasons. However, a program of regular maintenance could prove to be the most cost efficient solution if analysis of the transport characteristics and sediment supply in a stream system reveals that the aggradation problem is only temporary (perhaps the excess sediment supply is coming from a transient land use activity such as logging) or will have only minor effects over a relatively long period of time.

An alternative similar to a maintenance program which could be used on streams with persistent aggradation problems, such as those on alluvial fans, is the use of controlled sand and gravel mining from a debris basin constructed upstream of the bridge site. Use of this alternative would require careful analysis to ensure that the gravel mining did not upset the balance of sediment and water discharges downstream of the debris basin. Excessive mining could induce degradation downstream, potentially impacting the bridge or other structures.

The following is a list of guidelines regarding aggradation countermeasures:

  • Extensive channelization projects have generally proven unsuccessful in alleviating general aggradation problems, although some successful cases have been documented. A sufficient increase in the sediment carrying capacity of the channel is usually not achieved to significantly reduce or eliminate the problem. Channelization should be considered only if analysis shows that the desired results will be achieved.
  • Alteration or replacement of a bridge is often required to accommodate maximum aggradation depths.
  • Maintenance programs have been unreliable, but they provide the most cost‑effective solution where aggradation is from a temporary source or on small channels where the problem is limited in magnitude.
  • At aggrading sites on wide, shallow streams, spurs or dikes with flexible revetment have been successful in several cases in confining the flow to narrower, deeper sections.
  • A debris basin and controlled sand and gravel mining might be the best solution on alluvial fans (see HEC-20) and at other crossings with severe problems.


The selection of an appropriate countermeasure for scour at a bridge requires an understanding of the erosion mechanism producing the specific scour problem. For example, contraction scour results from a sediment imbalance across most or all of the channel, while local scour at a pier or abutment results from the action of vortices at an obstruction to the flow. Degradation is a component of total scour, but is considered a channel instability problem (see Section 3.5). Since the selection of a countermeasure depends on the type of scour involved, this section provides a brief overview of the principal scour components.

Scour is the result of the erosive action of running water, excavating and carrying away material from the bed and banks of streams. Different materials scour at different rates. Loose granular soils are rapidly eroded under water action while cohesive or cemented soils are more scour-resistant. However, ultimate scour in cohesive or cemented soils can be as deep as scour in sand-bed streams. Scour will reach its maximum depth in sand and gravel bed materials in hours; cohesive bed materials in days; glacial tills, poorly cemented sand stones and shales in months; hard, dense and cemented sandstone or shales in years; and granites in centuries. Massive rock formations with few discontinuities can be highly resistant to scour and erosion during the lifetime of a typical bridge [see HEC-18 (Richardson and Davis 2001) for detailed discussion and equations for calculating all bridge scour components].

Designers and inspectors need to carefully study site-specific subsurface information in determining scour potential at bridges, giving particular attention to foundations on rock.

Total Scour. Total scour at a highway crossing is comprised of three components. These components are:

  • Aggradation and Degradation. These are long‑term streambed elevation changes due to natural or human-induced causes within the reach of the river on which the bridge is located (see Section 3.5).
  • Contraction Scour. This type of scour involves the removal of material from the bed and banks across all or most of the width of a channel. Most commonly, this scour is the result of a contraction of the flow by the approach embankments to the bridge encroaching onto the floodplain and/or into the main channel which causes an increase in transport of the bed material in the bridge cross section.
  • Local Scour. This scour occurs around piers, abutments, spurs, and embankments and is caused by the acceleration of the flow and the development of vortex systems induced by these obstructions to the flow.

In addition to the types of scour mentioned above, lateral migration of the stream may also erode the approach roadway to the bridge or change the total scour by changing the angle of the flow in the waterway at the bridge crossing. Factors that affect lateral migration and the stability of a bridge are the geomorphology of the stream, location of the crossing on the stream, flood characteristics, and the characteristics of the bed and bank materials (see HEC-20).

3.6.1 Countermeasures for Contraction Scour

Severe contraction of flow at highway stream crossings has resulted in numerous bridge failures at abutments, approach fills, and piers from contraction scour. Design alternatives to decrease contraction scour include longer bridges, relief bridges on the floodplain, superstructures at elevations above flood stages of extreme events, and a crest vertical profile on approach roadways to provide for overtopping during floods exceeding the design flood event (see HEC-20). These design alternatives are integral features of the highway facility which reduce the contraction at bridges and, therefore, reduce the magnitude of contraction scour.

The elevation of bridge superstructures is recognized as important to the integrity of the bridge because of hydraulic forces that may damage the superstructure. These include buoyancy and impact forces from ice and other floating debris (see HEC-18). Contraction scour is another consideration in setting the superstructure elevation. When the superstructure of a bridge becomes submerged or when ice or debris lodged on the superstructure causes the flow to contract, flow may be accelerated and more severe scour can occur. For this reason, where contraction scour is of concern, bridge superstructures should be located with clearance for debris, and, if practicable, above the stage of floods larger than the design flood.

Another design feature which should be considered relative to contraction scour is the effective depth of the superstructure. Present day superstructures often include bridge railings which are solid parapets. These increase the effective depth of the superstructure and, thus, the importance of locating the bridge superstructure above high water with clearance for debris passage. It also increases the importance of alternate provisions for the passage of flood waters in the event of debris blockage of the waterway or superstructure submergence. Possible alternate provisions include relief bridges on the floodplain and a highway profile which provides for overtopping before the bridge superstructure begins to become submerged.

Similarly, pier design, span length, and pier location become more important contributors to contraction scour where debris can lodge on the piers and further contract flow in the waterway. In streams which carry heavy loads of debris, longer, higher spans and solid piers will help to reduce the collection of debris. Where practicable, piers should be located out of the main current in the stream, i.e., outside the thalweg at high flow. There are numerous locations where piers occupy a significant area in the stream channel and contribute to contraction scour, especially where devices to protect piers from ship traffic are provided.

The principal countermeasure used for reducing the effects of contraction is revetment on channel banks and fill slopes at bridge abutments (Design Guidelines 4 and 14). Additional countermeasures used to reduce flow contraction include measures which retard flow along highway embankments on floodplains. Flow along highway fills usually intersects with flow within bridge openings at large angles. This causes additional contraction of the flow, vortices, and turbulence which produce local scour. The contraction of flow can be reduced by using spurs on the upstream side of the highway embankment to retard flow parallel with the highway (Bradley 1978).

Similarly, guide banks (also referred to as spur dikes) at bridge abutments can improve the alignment of the flow in the bridge opening. They reduce contraction scour because they increase the efficiency of the bridge opening. The primary purpose of guide banks, however, is to reduce local scour at abutments (Design Guideline 15).

The potential for undesired effects from stabilizing all or any portion of the channel perimeter at a contraction should be considered. Stabilization of the banks may only result in exaggerated scour in the streambed near the banks or, in a relatively narrow channel, across the entire channel. Stabilization of the streambed may also result in exaggerated lateral scour in any size stream. Stabilization of the entire stream perimeter may result in downstream scour or failure of some portion of the countermeasures used on either the streambed or banks.

3.6.2 Countermeasures for Local Scour

Local scour occurs at bridge piers and abutments. In general, design alternatives against structural failure from local scour consist of measures which reduce scour depth, such as pier shape and orientation, and measures which retain their structural integrity after scour reaches its maximum depth, such as placing foundations in sound rock and using deep piling. Countermeasures which can reduce the risk from local scour include placing armor (e.g., riprap) at the structure or installing monitoring devices.

Abutments. Countermeasures for local scour at abutments consist of measures which improve flow orientation at the bridge end and move local scour away from the abutment, as well as revetments and riprap placed on spill slopes to resist erosion.

Guide banks are earth or rock embankments placed at abutments. Flow disturbances, such as eddies and cross‑flow, will be eliminated where a properly designed and constructed guide bank is placed at a bridge abutment. Guide banks also protect the highway embankment, reduce local scour at the abutment and adjacent piers, and move local scour to the upstream end of the guide bank (Design Guideline 15).

Local scour also occurs at abutments as a result of expanding flow downstream of the bridge, especially for bridges on wide, wooded floodplains that have been cleared for construction of the highway. Short guide banks extending downstream of the abutment to the tree line will move this scour away from the abutment, and the trees will retard velocities so that flow redistribution can occur with minimal scour.

The effectiveness of guide banks is a function of stream geometry, the quantity of flow on the floodplain, and the size of bridge opening. A typical guide bank at a bridge opening is shown in Figure 3.4.

Sketch of guide banks at both sides of a bridge opening showing flow around the tip of the guide bank and through the opening. The guide banks extend upstream from each abutment and curve away from the flow. The curve is in the form of an quarter ellipse with the long axis aligned to the flow. The elipse short axis is given as 0.4 of the long axis. Each guide bank has revetment armor of riprap, or equal, on the inner channel side and upstream tip and vegetative cover on the outer side. Location of scour hole formation is indicated around tip of guide bank.
Figure 3.4. Typical guide bank layout (after Bradley 1978).

Other countermeasures have been successfully used to inhibit scour at abutments where the abutment is located at the streambank or within the stream channel. These measures include dikes to constrict the width of braided streams and retards to reduce velocities near the streambank.

Piers. Three basic methods may be used to prevent damage from local scour at piers. The first method is to place the foundation of the structure at such a depth that the structural stability will not be at risk with maximum scour. This must be done on all new or replacement bridges (Richardson and Davis 2001). The second method (for existing bridges) is to provide protection at or below the streambed to inhibit the development of a scour hole. The third measure is to prevent erosive vortices from forming or to reduce their strength and intensity.

Streamlining the pier nose decreases flow separation at the face of the pier, reducing the strength of the horseshoe vortices which form at piers. Practical application of this principle involves the use of rounded or circular shapes at the upstream and downstream faces of piers in order to reduce the flow separation. However, flow direction can and does change with time and with stage on some streams. Piers oriented with flow direction at one stage or at one point in time may be skewed with flow direction at another. Also, flow direction changes with the passage of bed forms. In general, piers should be aligned with the main channel design flow direction and skew angles greater than 5 degrees should be avoided. Where this is not possible, a single cylindrical pier or a row of cylindrical columns will produce a lesser depth of local scour.

The tendency of a row of columns to collect debris should be considered. Debris can greatly increase scour depths. Webwalls have been used between columns to add to structural strength and to reduce the tendency to collect debris. Webwalls should be constructed at the elevation of stream flood stages which carry floating debris and extended to the elevation of the streambed. When installing a webwall as a countermeasure against debris, the potential for significantly increased scour depths should be considered if the approach flow might impinge on the wall at a high angle of attack.

Riprap is commonly used to inhibit local scour at piers at existing bridges. This practice is not recommended as an adequate substitute for foundations or piling located below expected scour depths for new or replacement bridges. It is recommended as a retrofit or a measure to reduce the risk where scour threatens the integrity of a pier (Design Guideline 11). The use of partially grouted riprap offers the opportunity to use smaller rock for pier scour protection (Design Guideline 12), and geotextile sand containers can provide a methodology to install an appropriate filter in a pier scour hole in flowing water (see Section 5.4.2 and Design Guideline 11). A comprehensive selection methodology for pier scour countermeasures is introduced in Section 3.2.6.

The practice of heaping stones around a pier is not recommended because experience has shown that continual replacement is usually required. Success rates have been better with alluvial bed materials where the top of the riprap was placed at or below the elevation of the streambed.

Piles (sheet, H beams or concrete) have been successfully used as a retrofit measure to lower the effective foundation elevation of structures where footings or pile caps have been exposed by scour. The piling is placed around the pile footings and anchored to the pile cap or seal to retain or restore the bearing capacity of the foundation. The increased mass of the retrofit pile will, however, produce a greater depth of scour.

Where sheet pile cofferdams are used during construction, the sheet piling should be removed or cut off below the level of expected contraction scour in order to avoid contributing to local scour. Cofferdams should not be much wider than the pier itself since the effect may be to greatly increase local scour depth. Leaving or removing cofferdams must be carefully evaluated because leaving a cofferdam that is higher than the contraction scour elevation may increase local scour depth. A study by Jones (1989) gives a method to evaluate the expected scour depths for cofferdams.

3.6.3 Monitoring

Monitoring or closing a bridge during high flows and inspection after the flood may be an effective countermeasure to reduce the risk from scour. However, monitoring of bridges during high flow may not reveal that they are about to collapse from scour. It also may not be practical to close the bridge during high flow because of traffic volume, no (or poor) alternate routes, the need for emergency vehicles to use the bridge, etc. Under these circumstances, scour countermeasures such as riprap could be installed. A countermeasure installed at a bridge to reduce the risk from scour along with monitoring during and inspection after high flows could provide for the safety of the public without closing the bridge (see Chapter 9).

Updated: 09/22/2014

United States Department of Transportation - Federal Highway Administration