Design of Scour Protection for Bridges and Culverts

Section 8
Design of Scour Protection

GUIDELINE 6
AACRA DRAINAGE DESIGN MANUAL
TABLE OF CONTENTS
8 DESIGN OF SCOUR PROTECTION 8-

1
8.1 Introduction 8-1
8.2 Scour Characteristics 8-1
8.2.1 Types of Scour in Watercourses 8-1
8.2.2 Clear Water and Live-Bed Scour 8-4
8.3 DESIGNING FOR SCOUR 8-5
8.3.1 General 8-5
8.3.2 Design Procedure 8-5
8.4 SCOUR COUNTERMEASURES 8-7
8.4.1 Structural Countermeasures 8-7
8.4.2 Hydraulic Countermeasures 8-8
8.4.3 Monitoring 8-13
8.5 Methods of Estimating Scour 8-13
8.5.1 General 8-13
8.5.2 Design Approach 8-13
8.5.3 Constriction Scour 8-14
8.5.4 Local Scour at Piers 8-19
8.5.5 Scour at Footings and Pile Caps 8-21
8.5.6 Clear Water Constriction Scour 8-21
8.5.7 Critical Velocity for Bed Material 8-22
8.6 Worked Example 8-23
8.7 References 8-28
FIGURES
Figure 8.1: Usual Form of Local Scour Holes at Piers 8-3
Figure 8.2: Scour at an Abutment 8-3
Figure 8.3: Scour at an Abutment and Adjacent Pier 8-4
Figure 8.4: Illustrative Pier Scour Depth in a Sand-Bed Stream as a Function of Time 8-4
Figure 8.5: Guide Bank Details 8-9
Figure 8.6: Chart for Determining Length of Guide Banks 8-10
Figure 8.7: Approximate Length of Embankment Protected by Spurs 8-11
Figure 8.8: The Four Main Cases of Constriction Scour 8-15
Figure 8.9: Definition Sketch for Scour Depths for Case 1(a) 8-17
Figure 8.10: Fall Velocity of Sand-Sized Particles 8-18
Figure 8.11: Definition Sketch for Scour Depths for Case 1(c) 8-19
Figure 8.12: Common Pier Shapes 8-20
TABLES
Table 8.1: Standard Classes of Rock Protection 8-12

Table 8.2: Estimation of Exponent k1 8-17
Table 8.3: Correction Factor, K1 for Pier Nose Shape 8-20
Table 8.4: Correction Factor, K2 for Angle of Attack of Flow 8-20
FINAL – November 2004 Section – 8 Design of Scour Protection:

Page i
GUIDELINE 6
SYMBOLS used in SECTION 8
A = Pier width (m)

An2 = water area (m2) underbridge referred to normal stage
b = length (m) of bridge opening
Dm = effective mean diameter of the bed material (m)
d50 = median diameter of the bed material (m)
Fr1 = Froude number = V1 / (g y1)0.5
k1 = exponent determined from Table 8.2.
K1 = correction factor for pier nose shape from Figure 8.12 and Table 8.3.
K2 = correction for angle of attack of flow from Table 8.4.
Ls = top length (m) of guide bank
Q= total stream discharge (m3/s)
Q1 = flow (m3/s) in upstream channel, not including overbank flow
Q2 = total flow (m3/s) through bridge opening
Qf = lateral or floodplain flow (one side) measured at a point sufficiently upstream so as the flow paths are
not affected by the constriction.
Q30 = q/b x 30 = discharge (m3/s) in 30m of stream adjacent to abutment, measured at a point sufficiently
upstream so as the flow paths are not affected by the constriction
Qf / Q30 = guide bank discharge ratio
Sgr = specific gravity of rock
T= thickness of rock protection (m)
Vn2 = Q/An2 = average velocity (m3/s) through bridge opening
Vp = velocity of parallel flow along tangent bank
Vm = mean velocity through bridge opening
Vi = velocity of impinging flow against curved bank.
W= minimum mass of stone for rock protection (kg)
Wc = Class of rock protection expressed in kg
W1 = bottom width (m) of the upstream main channel
W2 = bottom width (m) of the main channel in the constricted section less the width of piers
yo = existing depth in the contracted section before scour (m)
ys = average scour depth (m) = y2 - y1
y1 = average depth (m) of flow in the upstream main channel
y2 = average depth (m) of flow in the constricted section (including scour)
yps = scour depth (m) below constriction scour level
α= face slope (degrees)
ρ= factor for placement of rock protection
Section – 8 Design of Scour Protection FINAL – November 2004

Page ii
GUIDELINE 6
8 DESIGN OF SCOUR PROTECTION

8.1 Introduction
This chapter presents a methodology for the design of scour protection at inlets and outlets to culverts, in
open channels and at constrictions in channels such as bridges.
In natural channels or unlined floodways the shape of the watercourse may change over time due to scour
and material deposition. A range of factors will influence the rate and extent of change such as:
• Flow regimes (flow may be very seasonal with long periods of low flows);
• Soil characteristics and subsurface geology (this will influence the erodability potential of the channel
and also the sediment loadings from areas upstream);
• Channel grade and alignment;
• Restrictions or changes of flow direction due to structures such as bridges;
• Vegetation (will influence the susceptibility of stream banks to scour and also to accumulate material.
The extent of coverage of vegetation is not normally constant throughout the year and seasonal
changes must be considered).
Scour in watercourses can be addressed in a number of ways one of these is termed scour protection. Scour
protection involves the placement of materials such as rock or concrete lining in areas that may be prone to
scour.
Lining areas prone to scouring is not the only method that can be utilized to address scour. Scour is caused
by either high flow velocities or turbulence such as occurs at a hydraulic jump. Measures can be taken to
reduce the high velocities or turbulence including:
• Increasing channel cross-section thereby reducing velocities;

• Shaping watercourses to ensure that any changes in section are gradual and do not occur abruptly;
• Avoiding areas of super-critical flow by reducing grades or modifying cross-sections.
• Providing relief structures such as culverts and bridges to convey overbank flow thereby reducing
obstruction to the total flow path.
8.2 Scour Characteristics
8.2.1 Types of Scour in Watercourses

Scour in watercourses can take a range of forms. Scouring usually occurs due to one or a combination of the
following factors:
(a) Aggradation and degradation
Aggradation and degradation are long term changes to the watercourse profile due to either manmade or
natural processes. These changes in streambed elevation can be due to a range of factors such as dams,
changes in development within a catchment, diversion of flows for irrigation.
Aggradation is the long term process of material being deposited along a channel reach due to erosion from
the upstream watershed. Degradation is the opposite process whereby the stream bed is lowered by erosion
or scouring.

Page 8-1
GUIDELINE 6
(b) Scour due to River Morphology
The river morphology of natural streams or rivers is constantly changing. The rate of change can vary greatly
with some channel reaches changing in a single storm event. This process includes general stream bed
movement and scouring at channel constrictions or bends. The outside bank of a river tends to have higher
flow velocities and subsequently prone to scouring whereas the slower velocities at the inside bank allow
sediments to accumulate.
(c) Constriction Scour
As the name suggests constriction scour is due to a restriction in the normal flow cross section. The reduced
flow area for instance of bridge approach embankments across a floodplain increases velocities in the main
channel.
Constriction scour prevails until enough material has been removed from the channel bed so that flow
velocities and shear stresses are reduced. When an equilibrium condition prevails the amount of material
being deposited in the subject channel reach is equal to that being exported.
Constriction scour at a bridge opening reflects that in a normal channel the process of scour and deposition
undergoes a cyclic nature. During rising flows of a storm event the bed scours and latter as flows decline
material settles filling the void.
A guide bank at an abutment decreases the risk from scour at the abutment by realignment of the stream
lines of the floodplain flow so that they enter the bridge opening clear of the abutments.
(d) Local Scour
Local scour is confined to a small area around an obstruction in a waterway such as a bridge pier. The
obstruction has two many effects; flow velocities increase around the leading edge of the obstruction, and
vortices form at the base of the obstruction.
The vortices remove material from around the base of the obstruction. As with constriction scour once the
material is removed the flow patterns change and erosive forces are reduced till eventually an equilibrium
point is reached.
As well as a horseshoe shaped vortex at the base of an obstruction or pier, there is another vortex
downstream called the wake vortex. Both of these vortices remove material from the stream bed however the
intensity of the erosive forces reduce rapidly with distance downstream. A deposition of material is often
noticed immediately downstream of a long pier.

Page 8-2
GUIDELINE 6
Figure 8.1: Usual Form of Local Scour Holes at Piers

Source: AUSTROADS, 1994
Figure 8.2: Scour at an Abutment


Page 8-3
GUIDELINE 6
Figure 8.3: Scour at an Abutment and Adjacent Pier

8.2.2 Clear Water and Live-Bed Scour
There two conditions, which affect the duration, required for either a constriction or local scour to reach
equilibrium.
Clear water scour occurs when the upstream channel bed is stable and does not contribute any significant
volumes of material to the point of interest. Clear water scour is most often noted in waterways with very flat
grades, coarse bed material, armoured stream beds or vegetated channels. Live-bed scour however occurs
when the upstream bed is changing.
As shown in Figure 8.4, the clear water scour reaches its equilibrium over a longer period of time than live-
bed scour. Clear water scour occurs during major storm events in coarse bed material streams. Noting the
occurrence interval between such storms it may take numerous years until equilibrium is reached.
Maximum clear-water scour is about 10 percent greater than live-bed scour.
Figure 8.4: Illustrative Pier Scour Depth in a Sand-Bed Stream as a Function of Time

Page 8-4
GUIDELINE 6
8.3 DESIGNING FOR SCOUR
8.3.1 General
In considering the effect of scour it should be considered that whilst repairs can generally be made on
approach embankments relatively easily, a structure undermined by scouring can present a significant cost.
In consideration of the cost of a structure becoming unusable the larger social associated with disruption to a
traffic route needs to be evaluated and compared against providing suitable scour protection measures.
The floods that should be used for scour estimation are as follows (AUSTROADS 1994):
• For the evaluation of bridge foundations – the 200 year ARI flood or overtopping flood if this
produces more severe scour;
• For the design of protection works to the fill around the bridge abutments and to bridge abutments –
the total waterway design flood. The total waterway design flood is the event utilised for the overall
design of the total waterway, generally this is either a 50 or 100 year ARI event.
Waterway investigations of bridge sites should address both the sizing of the bridge waterway and the
designing of the foundations to resist scour. The scope and depth of the investigation should be
commensurate with the importance of the road and the consequences of failure.
8.3.2 Design Procedure

The following design procedures are recommended for designing abutment protection works and evaluating
the type, size and location of a bridge substructure.
(a) Design of Abutment Protection
As most methods available for estimating depth of scour lack field verification and may not give credible
estimates, it is recommended that abutments be protected against scour by the use of rock riprap and/or
guide banks. Properly designed, these two protective measures provide adequate protection to abutments
and make the estimation of the depths of scour less critical.
The procedure for the design of abutment protection is as follows:
1. select either the total waterway design flood or a lesser flood, if it is expected to produce the most
severe scour conditions. When a bridge is designed to be overtopped with a flood with an ARI less
than the total waterway flood, it is quite likely that this flood will produce the worst conditions.
2. Develop water surface profiles for the flood flows.
3. Using the procedures detailed in Section 8.5, estimate the constriction scour and local pier scour.
4. Plot the total scour depths on a cross section of the stream at the bridge site. The cross section
should include the soil profiles obtained from geotechnical investigations.
5. Using engineering judgment consider if the calculated scour depths are reasonable or if an
alternative depth should be utilised.
6. Consider the form and extent of protection required at the abutment with respect to the following
criteria:

Page 8-5
GUIDELINE 6
a. The degree of uncertainty in the scour depth prediction;

b. The potential for and consequences of failure;
c. The added construction cost for installing measures which will reduce the risk of scour at the
bridge abutments and approach;
d. The overall flood flow pattern at the bridge site including flow contained in the main channel
and overbank flow, location of bridge at changes in the river reach.
7. Determine the rock riprap protection required at abutments and guide banks for the depth of
constriction scour adopted. If the local pier scour impacts on the abutment, then provide sufficient
rock protection to protect both the abutment and the pier.
(b) Evaluation of Foundation Design for Ultimate Limit State

The recommended procedure for designing bridge foundations with respect to the ultimate limit state is as
follows (AUSTROADS 1994):
1. Determine whether the bridge will or will not be overtopped with a flood with an ARI less than 2000
years. If it will be overtopped, use the overtopping flood to evaluate the foundation design. If not,
then the 2000 year ARI flood should be used to evaluate the foundation design.
2. Develop water surface profiles for the flood flows.
3. Using the procedures detailed in Section 8.5, estimate the total depths of scour.
4. Plot the total scour depths on a cross section of the stream at the bridge site. The cross section
should include the soil profiles obtained from geotechnical investigations.
5. Using engineering judgment consider if the calculated scour depths are reasonable or if an
alternative depth should be utilised.
6. Evaluate the proposed substructure type, size and location and modify them, if necessary. The
overall flood flow pattern at the bridge site should be visualized to assist in identifying those
elements of the bridge at risk from scour.
7. Substructure design should be carried out on the basis that all stream bed material above the total
scour line has been removed and is not available for bearing or lateral support.

Page 8-6
GUIDELINE 6
8.4 SCOUR COUNTERMEASURES
Scour counter measures in relation to bridges can be classified into following groups (HEC 23 1997):
Group I: Structural countermeasures;

Group II: Hydraulic Countermeasures; and
Group III: Monitoring
8.4.1 Structural Countermeasures

Structural countermeasures essentially involve strengthening of foundation structures to prevent failure from
scour. Foundations in the floodplain should generally be placed at the same level as those in the stream
channel, to allow for possible lateral shifting of the stream. Abutment foundations should be placed below
the elevation of the lowest point of the river bed in the bridge opening.
(a) Spread Footings on Soil
For spread footings on soil the top of the footing should be located beneath the estimated level of scour.
Where there is any risk of scour undermining spread footings, deep foundations in the form of piles should
be used.
(b) Spread Footings on Rock Highly Resistant to Scour
The base of the footing should be placed on solid clean rock. Small embedments should be avoided since
blasting to achieve keying frequently damages the rock structure and makes it more susceptible to scour. If
lateral restraint is required, it should be provided with steel dowels drilled and grouted into the rock.
(c) Spread Footings on Erodible Rock
Excavation into erodible rock should be made with care. All loose rock should be removed from the
excavation and any overbreak beneath the footing made up with lean concrete. The footing should be
poured against the sides of the excavation for the full depth of the footing. The excavation above the top of
the footing should be backfilled with rock riprap sized to withstand flood flow velocities.
(d) Pile Foundations with Pile Caps
The top of the pile cap should be placed at a level below the constriction scour depth. This will minimize the
obstruction to flood flows and resulting scour.
(e) Other Factors related to Bridge Structures
A number of other factors should also be considered to minimize the potential for scouring at bridges:
• Piers should be aligned with the direction of flow;

• Use round piers especially where there are complex flow patterns;
• Piers should be streamlined to decrease turbulence and also the potential to accumulate debris;
• Use spillthrough abutments rather than vertical wall abutments, which produce twice as much scour.

Page 8-7
GUIDELINE 6
(f) Clearing & Earthworks
Designers should not only consider the localized effect of constructing a restriction to flow but also the
broader effect of the works on the environment.
Clearing vegetation or excavating borrow pits to source material for approach embankments will change flow
regimes in the area. These changed flow regimes will consequently have an effect on bridge piers,
abutments and approaches.
Strict controls should be placed on clearing and excavation in the immediate vicinity of a structure and the
associated road embankments.
8.4.2 Hydraulic Countermeasures
Hydraulic countermeasures relevant to Addis Ababa city conditions essentially consist of river training
structures. Of the river training structures the longitudinal structures such as guide banks are considered
appropriate to Addis Ababa city conditions. The transverse structures like spurs also have some application
to the city conditions. These two types of countermeasures are treated in the following description.
(a) Guide Banks
Guide banks are an effective method of decreasing the risk of abutment scour at bridges. Guide banks
transfer the location of the constricting streamlines and associated vortices away from the bridge abutment
to the nose of the guide bank. Guide banks also assist in the protection of approach abutments from scour
by reducing the flow along the face of the embankments.
The required form of the guide is described below:

(i) Geometry
The preferred geometry for a guide bank is a quarter ellipse as described by the following equation.
X2 Y2
+ =1 (8.1)
L S2 (0.4 L S )2
Refer Figure 8.5 for locations of dimensions.

(ii) Height
The guide bank should be higher than the expected design flood level with a freeboard allowance to prevent
overtopping and degradation by wave action.
(iii) Length
Length of guide bank is determined from Figure 8.6 below with the following definitions:
Q= total stream discharge (m3/s)

Qf = lateral or floodplain flow (one side) measured at a point sufficiently upstream so as the flow paths are
not affected by the constriction.
Q30 = Q/b x 30 = discharge (m3/s) in 30m of stream adjacent to abutment, measured at a point
sufficiently upstream so as the flow paths are not affected by the constriction
b= length (m) of bridge opening
An2 = water area (m2) underbridge referred to normal stage
Vn2 = Q/An2 = average velocity (m3/s) through bridge opening
Qf / Q30 = guide bank discharge ratio
Ls = top length (m) of guide bank

Page 8-8
GUIDELINE 6
Figure 8.5: Guide Bank Details

Figure 8.6 is read by entering the ordinate with the proper value of Qf / Q30, moving horizontally to the curve
corresponding with the computed value of Vn2 and then downward to obtain from the abscissa the length of
the guide bank required.
As a general rule, if the calculated length of guide bank is less than 10m then a bank is not required. If the
length from the chart is between 10 and 30m then a 30m minimum should be adopted to ensure that flow
paths are perpendicular when passing under the structure.

Page 8-9
GUIDELINE 6
Figure 8.6: Chart for Determining Length of Guide Banks

The selection of Class of rock protection, as described in latter section, required for impinging and parallel
flow is based on the following assumption:
The velocity ratios are Vp : Vm : Vi = 2 : 3 : 4
where
Vp = velocity of parallel flow along tangent bank
Vm = mean velocity through bridge opening
Vi = velocity of impinging flow against curved bank.
(b) Spurs
Spurs, retards or groynes are structure utilised to deflect flows from along a stream bank or embankment. It
is often cheaper and more effective to provide a series of spurs along a river bank than to provide protection
along its full length.
The noses of spurs are normally rock protected. In the event that a spur begins to fail the consequences are
not as severe as compared to erosion at an expensive structure. As the spur protrudes into the flow path
effectively shielding an embankment immediately upstream and downstream remedial works can be
undertaken on a spur whilst a structure is still effectively protected.
Spurs can also be used to protect approach embankments to bridges. In locations where considerable
overbank flow occurs, the construction of a road crossing can result in a concentrated flow path along the toe
of an embankment as runoff travels towards the main channel and bridge opening. A series of spurs at right
angles to the embankment can protect the approach from scouring.

Page 8-10
GUIDELINE 6
The length of bank or road embankment protected by a spur is approximately three times the perpendicular
length of the spur as illustrated in
Figure 8.7.
Figure 8.7: Approximate Length of Embankment Protected by Spurs

(c) Rock Protection
At abutments and bridge piers where scour is anticipated properly designed rock riprap protection can
provide long term resistance to erosion. Riprap consists of appropriately sized stone placed on the face to be
protected. The stone is normally not secured by wire netting or mortar but resists the force of the water by
the weight of the stones and their natural interlock between the faces of the rock. This kind of protection is
generally the most economical.
Other types of rock protection can include rock filled wire mattresses, grouted stone pitching or hydraulically
filled concrete mattresses.
The California Department of Public Works developed the following relationships for minimum mass of stone
and thickness of the protection layer (AUSTROADS 1994):

Page 8-11
GUIDELINE 6
Minimum mass of stone (kg)
W= 11x10-3 V6 Sgr (8.2)

(Sgr – 1)3 sin3 (ρ - α)
where
Sgr = specific gravity of rock
ρ= 70 degrees for randomly placed rock rubble
α= face slope (degrees)
Thickness of the rock protection (m)

T= 0.3 sin α x Wc 1/3 (8.3)
where
Wc = Class of rock protection expressed in kg
(Refer Table 8.1) (ie. Wc = ¼ tonne = 250kg)
Table 8.1: Standard Classes of Rock Protection
Rock Class Rock Size * (m) Rock Mass (kg) Minimum

Percentage of Rock
Larger Than
Facing 0.40 100
0.30 35 0
0.15 2.5 50
90
Light 0.55
0.40 250 0
0.20 100 50
10 90
¼ tonne 0.75 500
0.55 250 0
0.30 35 50
90
½ tonne 0.90 1000
0.70 500 0
0.40 100 50
90
1 tonne 1.15 2000
0.90 1000 0
0.55 250 50
90
2 tonne 1.45 4000
1.15 2000 0
0.75 500 50
90
4 tonne 1.80 8000
1.45 4000 0
0.90 1000 50
90
* Assuming specific gravity of 2.65 and spherical shape
The stones are to be graded uniformly between the specified minima for Class of rock protection with two
thirds heavier than minimum required on face. The grading of the rock is an important factor for scour
resistance.

Page 8-12
GUIDELINE 6
Note that the mass by which the class of rock protection, Wc is designated does not correspond to the mass
W. The class of rock protection, Wc should be graded so that at least 2/3 of all rocks in the class have a
greater mass than W.
Rock used for riprap should be hard dense and durable stone free from laminations, fractures and otherwise
weak stone.
Stone shape is another important criterion, which will affect the interlock between the stones. Angular stones
provide numerous faces which can interact with adjacent stones. Rounded stones can be used however the
slope face will need to be limited to about 1 in 3. Generally a stones width or breadth should be no less than
1/3 of its length. Flat slab like stones should be avoided.
The placement of stones should commence from a footing trench at the base of the embankment. The base
of the trench must be lower than the level of anticipated scour. Rock should be placed to minimize voids.
In order to protect the finer particle material of the embankment a geotextile layer or gravel filter is used
underneath the rock protection.
8.4.3 Monitoring
Monitoring includes activities aimed at early identification of the developing scour problems. It can be
accomplished by instrumentation or visual inspection. Instrumentation involves fixed or portable instruments.
Fixed instruments are attached to the bridge structure to detect scour at a particular location. Typically, such
instruments are located at piers and abutments. Instruments such as sonar monitors are used to provide a
time line of scour whereas magnet sliding collars are used to monitor maximum scour depth. Data from fixed
monitors is either downloaded manually or telemetered to another location.
Portable instruments are carried manually and used along a bridge whenever required. These are
economical but lack the ability of continuous recording
Visual inspections include measurement of the bed elevation at locations of potential scour. Typically, the
bridges are inspected on a biennial schedule and channel bed elevations compared with previous records to
trace adverse developments. Channel bed elevations should also be taken during and after high flow events.
This type of monitoring, if carried out on regular basis, can help detect serious scours for a timely correction.
8.5 Methods of Estimating Scour

8.5.1 General
Many methods are available to estimate scour at bridges. The reliability and accuracy of the methods varies
greatly with most analysis based on limited available data and scarce field verification.
It is therefore recommending that calculated depths of scour be treated with caution and that sound
engineering judgement based on local field experience should always take precedence.
8.5.2 Design Approach
The method for estimating constriction and local pier scour at bridges is as follows:
• Obtain the fixed bed channel hydraulics. Flow profiles and velocities can be estimated using
backwater analysis programs such as HECRAS by the US Army Corps;

Page 8-13
GUIDELINE 6
• Estimate long term profile changes (channel aggradation or degradation);

• Adjust the fixed bed channel hydraulics to reflect the long term changes;
• Compute the bridge hydraulics;
• Estimate the magnitude of the bridge constriction scour;
• Estimate the magnitude of the local pier scour;
• Plot and evaluate the total scour depths given by the sum of constriction scour and local pier scour);
• Evaluate foundation design;
• Determine the protection required at the abutments in the form of rock riprap and/or guide banks.
It should be noted that the normal method presented in this section assumes that the constriction and local
pier scour occur independently. With this approach the potential local scour is added to the constriction scour
without considering the effects of the constriction scour on the channel and bridge hydraulics. However, if the
constriction scour is significant, the channel and/or bridge hydraulics should be adjusted for the effects of the
constriction scour before estimating the local scour.
8.5.3 Constriction Scour
There are four predominant constriction conditions as illustrated in Figure 8.8. The method used to calculate
constriction scour shall depend upon the prevailing condition.
Case 1- This condition involves a channel with overbank flow which is forced back into the main channel by
the bridge and associated embankment. There are three different situations which can occur:

Page 8-14
GUIDELINE 6
Figure 8.8: The Four Main Cases of Constriction Scour


Page 8-15
GUIDELINE 6
(a) The river channel width becomes narrower, either due to the bridge abutments projecting into the
channel or the bridge being located at a narrowing reach of the river;
(b) No constriction of the main channel but the overbank flow area is completely obstructed by the
embankment;
(c) Abutments are set back from the main channel permitting some overbank flow.
Case 2 – Flow is confined to the main channel and there is no overbank flow. The normal river channel
becomes narrower due to the bridge itself or the bridge site being located at a narrower reach of the river.
Case 3 – A relief bridge in the overbank area with little or no bed material transport in the overbank area (ie.
clear water scour)
Case 4 – A relief bridge over a secondary stream in the overbank area.
Two methods are presented for the calculation of constriction scour; Laursen’s Method and the Mean
Velocity Method. Generally an estimate of constriction scour should be made by both methods as a
comparison and guide to the reliability of the calculated scour depth.
(a) Laursen’s Method
The modified version of Laursen’s Equation (AUSTROADS 1994) for live bed scour at a long contraction can
be used to estimate the depth of live bed constriction scour. Clear water constriction scour can also be
calculated using the same equation but with the removal of the last term involving the ratio of W1 to W2.
y2 / y1 = (Q2 / Q1) 0.86 (W1 / W2) k1 (8.4)
Average scour depth (m) = ys = y2 - y1
Where for Cases 1(a), 1(b), 2(a), 2(b) and 2(c)

y1 = average depth (m) of flow in the upstream main channel
y2 = average depth (m) of flow in the constricted section (including scour)
Q1 = flow (m3/s) in upstream channel not including overbank flow
W1 = bottom width (m) of the upstream main channel
W2 = bottom width (m) of the main channel in the constricted section less the width of piers
k1 = exponent determined from Table 8.2.
Figure 8.9 defines y1, y2, W1 and W2 for Case 1(a). From Figure 8.8 it can be seen that these parameters will
be similar for Cases 1(b), 2(a), 2(b) and 2(c).
Note: For Case 1(b) where W1 and W2 are of similar magnitude and the width of the piers is not significant,
the last part of Equation 8.4 can be ignored.

Page 8-16
GUIDELINE 6
Figure 8.9: Definition Sketch for Scour Depths for Case 1(a)
Table 8.2: Estimation of Exponent k1

V* / w k1 Mode of Bed Material Transport
0.59 Mostly contact bed material discharge
< 0.50
0.50 to 2.0 0.64 Some suspended bed material
discharge
> 2.0 0.69 Mostly suspended bed material
discharge
Where
V* = shear velocity (m/s) in the upstream section = (gy1S1) ½
w= fall velocity (m/s) of bed material based on D50 (mm)

(Refer )
g = acceleration due to gravity = 9.80 m/s2

S1 = slope of energy line of main channel (m/m)
D50 = median diameter (mm) of bed material
Source: AUSTROADS 1994

Page 8-17
GUIDELINE 6
Figure 8.10: Fall Velocity of Sand-Sized Particles
Cases 1(c), 3 and 4 are more complicated and can be treated in two ways:
(1) The main channel can be treated as for Case 1(b) and the overbank flow, between the stream bank
and the bridge abutment, can be treated separately. This approach requires an estimate of the flow
in the main channel and the area of overbank flow, both upstream of the bridge and in the bridge
opening. Estimation of the distribution of flow in the bridge opening poses a difficult problem. Given
the problem of estimating the flow distribution and the limitations of the method for estimating
constriction scour, the complexity involved in this approach cannot be justified.
(2) The alternative approach is to consider the total flow in the bridge opening, when the last term of
Equation 8.4 becomes 1 and the other parameters are as follows:
y1 = average depth (m) of flow in the area of stream channel and floodplain occupied by the
bridge (at normal stage without scour)
y2 = average depth (m) of flow in the bridge opening (including scour)
Q1 = flow (m3/s) in area of stream channel and floodplain occupied by the bridge (at normal stage
and without backwater)
Figure 8.11 defines y1 and y2 for Case 1(c). From Figure 8.8 it can be seen that these parameters will be
similar for Cases 3 and 4

Page 8-18
GUIDELINE 6
Figure 8.11: Definition Sketch for Scour Depths for Case 1(c)
(b) Mean Velocity Method

This method utilizes the concept of cross sectional velocity as a criterion of constriction scour. The method
can be used as a check on the constriction scour estimates determined by Lauren’s Method for Cases 1, 2
and 4. The method is as follows:
(iv) Calculate mean velocity of flow in the unrestricted main channel for the discharge at which
constriction scour will just commence;
(v) Determine average constriction scour level that will make the mean velocity through the bridge
opening equal to the estimated mean velocity calculated in (i) above.
It is assumed that the discharge at which constriction scour will just commence is that of the total waterway
design flood. The experience has shown that it gives realistic estimates of constriction scour. The total
waterway design flood can be either the 50 or 100 year ARI design flood depending upon the standard
adopted. It is not anticipated that estimates of constriction scour will vary greatly with the use of either flood,
as velocities in the main channel of a stream should be of a similar magnitude for both floods.
8.5.4 Local Scour at Piers
The Colorado State University (CSU) Equation can be used to determine depth of local pier scour for both
live-bed and clear water scour conditions. Whilst this equation predicts equilibrium scour depths, maximum
scour depths may be 10% greater.
yps / y1 = 2.0 K1 K2 ( a / y1) 0.65 Fr1 0.43 (8.5)
where
yps = scour depth (m) below constriction scour level (Figure 8.9)
y1 = flow depth (m) just upstream of the pier
K1 = correction factor for pier nose shape from Figure 8.12 and Table 8.3
K2 = correction for angle of attack of flow from Table 8.4.
a= pier width (m)
Fr1 = Froude number = V1 / (g y1)0.5

Page 8-19
GUIDELINE 6
V1 = mean velocity (m/s) of approach flow upstream of pier
Figure 8.12: Common Pier Shapes

Table 8.3: Correction Factor, K1 for Pier Nose Shape
Shape of Nose K1
(a) Square Nose 1.1
(b) Round Nose 1.0
(c) Circular Cylinder 1.0
(d) Sharp Nose 0.9
(e) Group of Cylinders 1.0
Table 8.4: Correction Factor, K2 for Angle of Attack of Flow
Angle L/a = 4 L/a = 8 L/a = 12

0 1.0 1.0 1.0
15 1.5 2.0 2.5
30 2.0 2.5 3.5
45 2.3 3.3 4.3
90 2.5 3.9 5.0
Angle = skew angle of flow
L = length (m) of pier
a = width (m) of pier
Note: The correction factor K1 should be determined using Table 8.4 for flow angle of attack up to 5 degrees.
For greater angles, K2 dominates and K1 should be considered as 1.0. If L/a is larger than 12, use the values
of L/a = 12 as a maximum.
The top width of scour at a pier in cohesionless bed material can be estimated as 2.8 ys (2.8 times the
constriction scour depth) from one side of the pier.
Page 8-20
GUIDELINE 6
(a) Multiple Columns
For multiple columns skewed to the direction of flow, the scour depends upon the spacing between the
columns. The correction factor K2 for angle of attack would be smaller than for a solid pier. The pier width, a
in Equation 8.5 would be the total projected width of all the columns in a single bent, normal to the flow angle
of attack. The correction factor K1, for the multiple columns would be 1.0 regardless of column shape. If
debris is likely to pile up against the multiple columns, it would be logical to consider the multiple columns as
a solid elongated pier. In this instance the appropriate L/a value and flow angle of attack would then be used
to determine K2 from Table 8.4.
(b) Pressure Flow
Pressure flow occurs when the water level beneath a structure rises to an extent where the bridge soffit
affects the flow conditions by being partially or completed submerged. Under these circumstances
constriction scour and local scour can be expected to increase considerably. The depth of scour will depend
on the velocity of the approach flow and the clearance between the superstructure and the stream bed. For
the same approach velocity, constriction and local scour can be expected to increase with decreasing
clearance between superstructure and the stream bed.
8.5.5 Scour at Footings and Pile Caps

Where the top of a footing or the pile cap is at or below the stream bed (after taking into account constriction
scour and long term degradation) it is recommended that the pier width be used for the value of “a” in the
pier scour equation in the previous section.
Where the footing extends above the stream bed a second calculation should be made and the larger of the
two adopted. The second computation should use the width of the footing or pile cap for the value of a, and
the depth and average velocity in the flow zone obstructed by the footing for the y1 and V1 values
respectively in the scour equation.
The average velocity of flow at the exposed footing, Vf, should be determined using the following equation
(Jones, 1989):
Vf = ln [10.93 (yf / ks) + 1] (8.6)

V1 ln [10.93 (y1 / ks) + 1]
where
The values of Vf and yf should be used in Eq 8.5 for V1 and y1 respectively.
(a) Exposed Pile Groups
Pile groups protruding above the stream bed, as a result of constriction scour or long term degradation, can
be conservatively analysed by assuming that the area of the pile group represents a single pier of equal
width ignoring the clear spaces between. Engineering judgment is needed to assess the likelihood of floating
debris being trapped by the pile group and increasing the effective waterway obstruction.
If a pile group is exposed to the flow as a result of local scour then it is unnecessary to consider the piles in
calculating pier scour.
The correction factor K1 in Equation 8.4 should be considered to be 1.0 regardless of the layout of the piles.
If the pile group is square then K2 would be 1.0. However, if the pile group is a rectangle then the dimensions
used would be based on the assumption that they are a single pier and the appropriate L/a value used for
determining K2.
8.5.6 Clear Water Constriction Scour

Clear water constriction scour occurs when there is no bed material transport from the upper reach into the
lower reach or it is transported in the upstream reach only in suspension. A useful relation for predicting clear
water scour due to bridge constriction is given by the following equation (HEC 18 1995):
Page 8-21
GUIDELINE 6
3/ 7
⎡ Q2 ⎤
y=⎢ 2/3 2 ⎥
(8.7)
⎣ 40 Dm W ⎦
Average scour depth, ys = y-yo
Where,
Q = discharge through contraction (m3/s)
Dm = effective mean bed material size = 1.25 d50, (m)
W = bottom width in contraction (m)
yo = existing depth in the contracted section before scour (m)
8.5.7 Critical Velocity for Bed Material
Due to its implication on scour, the designers are invariably interested in the impact of the flow velocity on
mobility of a given bed material size. Critical velocity, Vc, for beginning of motion a given sediment size is
given by the following relation (HEC 18 1995):
K s1 / 2 ( S s − 1)1 / 2 D1 / 2 y 1 / 6
Vc = (8.8)
n
Where,
Vc = critical velocity to move bed material size D or smaller (m/s)
Ks = Shields parameter (0.039 – HEC 18 1995, for this equation)
Ss = Specific gravity of the bed material (2.65)
D = size of the bed material (m)
y = depth of flow (m)
n = Manning’s roughness coefficient.

Page 8-22
GUIDELINE 6
8.6 Worked Example

Bridge over Little McPhees Creek on the Great Northern Highway, Australia
General description of problem – A 54 m long bridge with 1.5:1 spillthrough abutments (see Figure 8.13) is
to be constructed over an ephemeral stream in the Kimberley Region of Western Australia. The bridge will
have a deck level of 210.55 m and a 1.0 m deep superstructure, which will be continuous over the piers.
The three piers, which will be aligned with the flow and evenly spaced, will comprise three 0.6 m diameter
columns founded on spread footings.
The bridge will be overtopped with a flood with a ARI less than 500 years. This overtopping ultimate limit
state flood will be used to evaluate the foundations and the 100 year average recurrence interval flood will be
used to design the abutment protection. The overtopping flood has a discharge of 510 m3/s at a stage height
of 208.80 m and the 100 year design flood discharge is 405m3/s at a stage height of 208.60 m.
The stream has a catchment area of 30 km2 and has a well defined channel with floodplains on either side.
The bridge crossing is located on a slight bend, with a moderately uniform channel upstream and
downstream of the bridge site. The main channel contains a very shallow depth of coarse and overlying
weathered granite. On the floodplains there is a shallow depth of loam over weathered granite. The granite
is distinctly weathered at the surface and readily breaks down into a quartz gravel. Weathering decreases
with depth until fresh granite is reached at a depth of about 12m below bed level. D50 is about 3 to 5 mm for
the quartz gravel of the completely weathered granite.
Hydraulic Characteristics – the distribution of flow in the natural channel (without the bridge) and bridge
hydraulics were determined for the overtopping flood and 10 year design flood using the computer program
AFFLUX. The details are as follows:

Page 8-23
GUIDELINE 6
Figure 8.13: Little McPhees Creek Bridge - Plot of Total Scour


Page 8-24
GUIDELINE 6
Table 8.5: Little McPhees Creek - Distribution of Flow in Natural Channe
l
Distribution of Flow Flow Mean Average
(m3/s) Velocity Depth of Flow
(m/s) (m)
Q overtopping
Main channel 325 2.94 2.52
Q100
Main channel 290 2.78 2.31
Left of bridge, Q c 50 1.45 ---

Bridge opening, Q b 280 2.78 2.31
Right of bridge opening, Q a 75 0.67 ---
The bridge slightly constricts the main channel (Case 1a on Figure 8.8). Because there is very little sand in
the bed of the main channel, it is anticipated that scour will be predominately clear-water in the bridge
opening.
Local pier scour will be estimated for one pier only.
Step 1 – Long term bed elevation changes
It is not anticipated that any bed elevation changes will occur.
Step 2 – Estimate magnitude of constriction scour
(a) For the overtopping ultimate limit state flood

y1 = 2.52 m
Q1 = 325 m3/s
Q2 = 510 m3/s
For clear-water scour, the last part of Equation 8.3 can be ignored, and
0.86
y2 ⎡ 510 ⎤
=⎢ = 1.67
2.52 ⎣ 280 ⎥⎦

Page 8-25
GUIDELINE 6
Table 8.6: Little McPhees Creek - Bridge Hydraulics
Flow (m3/s) Scour Depth (m) Mean Velocity Backwater, h1*

(m3/s) Average Depth* (m)
of Flow (m)
Q overtopping = 500 0.0 4.43 2.52 1.74
1.0 3.33 2.52 1.05
2.0 2.65 2.52 0.70
3.0 2.20 2.52 0.51
4.0 1.89 2.52 0.38
Q 100 = 410 0.0 3.88 2.31 1.15
1.0 2.84 2.31 0.67
2.0 2.23 2.31 0.44
3.0 1.84 2.31 0.31
*Without Scour
Y2 = 1.67 x 2.52 = 4.21 m

and depth of constriction scour,
Ys = 4.21 – 2.52 = 1.69 m
Try Mean Velocity Method utilizing the average velocity of 2.78 m/s in the main channel for the 100 year
design flood. From Table 8.6 it can be seen that a scour depth of about 1.8 m is required to reduce the
velocity to 2.78 m/s. Accept 1.8 m as depth of constriction scour.
(b) For the 100 year flood for the design of protection works
y1 = 2.31 m
Q 1 = 280 m3/s
Q 2 = 405 m3/s
Ignoring the last part of Equation 8.3,
0.86
y2 ⎡ 405 ⎤
=⎢ = 1.37
2.31 ⎣ 280 ⎥⎦
y2 = 1.37x 2.31 = 3.16 m
and the depth of constriction scour,
ys = 3.16 – 2.31 = 0.85 m
With the Mean Velocity Method the scour required to reduce velocity to 2.78 m/s is about 1.1 m.

Page 8-26
GUIDELINE 6
Step 3 – Estimate magnitude of local pier scour
(a) For the overtopping ultimate limit state flood

Estimate local scour for pier 1 (on the outside of the bend) with the lowest natural ground level of
206.12 m, with the bridge hydraulics adjusted for a constriction scour depth of 1.80 m:
y1 = 208.80 – 206.12 + 1.80 = 4.48 m
K1 = 1.0 for a round column

K2 = 1.0 for piers aligned with flow and angle of attack of zero degrees
a = width of column = 0.6 m
V1 = 2.94 m/s for a scour depth of 1.80 m.
Fr1 = 2.94 / (9.80 x 4.48)0.5 = 0.44

0.65
y ps ⎡ 0.60 ⎤
= 2 .0 × 1 .0 × 1 .0 ⎢ 0.44 0.43 = 0.38m
4.48 ⎣ 4.48 ⎥⎦
y ps = 0.38 × 4.48 = 1.70m
Step 4 – Plot and evaluate total scour depths, and evaluate foundation design
The depths of constriction scour and local pier scour are plotted on Figure 8.13. From this plot it can be
seen that the top of the footings need to be 3.50 m (1.8 + 1.7 m) below the stream bed level. This appears
to be excessive given that the weathering of the granite decreases with depth and this will tend to limit the
scour that can occur.
Given the uncertainties in estimating the scour depths, the top of the footings will be placed 3.12 m below the
lowest point in the stream bed at a level of 203.00 m. The footings will be cast against the sides of the
excavation in the weathered rock and the excavation above the footing backfilled with rock riprap. The top of
the abutment footings will be placed at a level of 205.00 m.
Step 5 – Determine protection required at abutments
Rock protected guide banks will be provided at both abutments. Sufficient rock will be provided at the toe of
the spillthrough abutment and guide bank to protect against a depth scour of 1.5 m, which is slightly greater
that the depth required to achieve a velocity of 2.78 m/s (see Table 8.6) through the bridge opening.

Page 8-27
GUIDELINE 6
8.7 References
AUSTROADS (1994), “Waterway Design –A Guide to the Hydraulic Design of Bridges, Culverts and
Floodways”, AUSTROADS National Office, Sydney.
HEC 18 (1995) - Federal Highway Administration, ”Evaluation of Scour at Bridges”, US Department of

Transportation.
HEC 23 (1997) - Federal Highway Administration, ”Bridge Scour and Stream Instability Countermeasures”,
US Department of Transportation.
Institute of Engineers, Australia (2001) ‘Australian Rainfall and Runoff – A Guide to Flood Estimation Volume
1’

Page 8-28

Design of Scour Protection for Bridges and Culverts

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Design of Scour Protection for Bridges and Culverts

Uploaded by

Copyright:

Available Formats

Section 8

Design of Scour Protection

8 DESIGN OF SCOUR PROTECTION 8-

Table 8.1: Standard Classes of Rock Protection 8-12

FINAL – November 2004 Section – 8 Design of Scour Protection:

SYMBOLS used in SECTION 8

A = Pier width (m)

Section – 8 Design of Scour Protection FINAL – November 2004

8 DESIGN OF SCOUR PROTECTION

• Increasing channel cross-section thereby reducing velocities;

8.2 Scour Characteristics

8.2.1 Types of Scour in Watercourses

(a) Aggradation and degradation

FINAL – November 2004 Section – 8 Design of Scour Protection:

(b) Scour due to River Morphology

(c) Constriction Scour

(d) Local Scour

Section – 8 Design of Scour Protection FINAL – November 2004

Figure 8.1: Usual Form of Local Scour Holes at Piers

Figure 8.2: Scour at an Abutment

FINAL – November 2004 Section – 8 Design of Scour Protection:

Figure 8.3: Scour at an Abutment and Adjacent Pier

8.2.2 Clear Water and Live-Bed Scour

Maximum clear-water scour is about 10 percent greater than live-bed scour.

Section – 8 Design of Scour Protection FINAL – November 2004

8.3 DESIGNING FOR SCOUR

8.3.2 Design Procedure

(a) Design of Abutment Protection

The procedure for the design of abutment protection is as follows:

2. Develop water surface profiles for the flood flows.

FINAL – November 2004 Section – 8 Design of Scour Protection:

a. The degree of uncertainty in the scour depth prediction;

(b) Evaluation of Foundation Design for Ultimate Limit State

2. Develop water surface profiles for the flood flows.

Section – 8 Design of Scour Protection FINAL – November 2004

8.4 SCOUR COUNTERMEASURES

 Group I: Structural countermeasures;

8.4.1 Structural Countermeasures

(a) Spread Footings on Soil

(b) Spread Footings on Rock Highly Resistant to Scour

(c) Spread Footings on Erodible Rock

(d) Pile Foundations with Pile Caps

(e) Other Factors related to Bridge Structures

• Piers should be aligned with the direction of flow;

FINAL – November 2004 Section – 8 Design of Scour Protection:

(f) Clearing & Earthworks

8.4.2 Hydraulic Countermeasures

The required form of the guide is described below:

Refer Figure 8.5 for locations of dimensions.

Q= total stream discharge (m3/s)

Section – 8 Design of Scour Protection FINAL – November 2004

Figure 8.5: Guide Bank Details

FINAL – November 2004 Section – 8 Design of Scour Protection:

Figure 8.6: Chart for Determining Length of Guide Banks

The velocity ratios are Vp : Vm : Vi = 2 : 3 : 4

Section – 8 Design of Scour Protection FINAL – November 2004

Figure 8.7: Approximate Length of Embankment Protected by Spurs

(c) Rock Protection

FINAL – November 2004 Section – 8 Design of Scour Protection:

Minimum mass of stone (kg)

W= 11x10-3 V6 Sgr (8.2)

Thickness of the rock protection (m)

Group I: Structural countermeasures;