AGRD05B-13 Guide To Road Design Part 5B DrainageOpen Channels Culverts and Floodways

Guide to Road Design Part 5B:
Drainage – Open Channels, Culverts and Floodways
Sydney 2013
Guide to Road Design Part 5B: Drainage – Open Channels, Culverts and Floodways
Publisher
First edition project manager: Richard Fanning
Austroads Ltd.
Level 9, 287 Elizabeth Street
First edition prepared by: Road Design Drainage Technical Sub-committee Sydney NSW 2000 Australia
Phone: +61 2 8265 3300
Abstract
austroads@austroads.com.au
Guide to Road Design Part 5B: Drainage – Open Channels, Culverts and www.austroads.com.au
Floodways contains guidance on the design of open channels, culverts and
floodways to support the operation and management of the road network. About Austroads
This Guide needs to be used in conjunction with the other two Parts of the
Guide to Road Design that relate to drainage design: Austroads is the peak organisation of Australasian
road transport and traffic agencies.
• Part 5: Drainage – General and Hydrology Considerations
Austroads’ purpose is to support our member
• Part 5A: Drainage – Road Surface, Network, Basins and Subsurface. organisations to deliver an improved Australasian
This Guide provides guidance on the fundamentals of open channel, culvert road transport network. To succeed in this task, we
and floodway flows, and includes methods to undertake the design of these undertake leading-edge road and transport
drainage facilities. research which underpins our input to policy
development and published guidance on the
design, construction and management of the road
network and its associated infrastructure.
Keywords Austroads provides a collective approach that
open channels, hydraulic jump, table drains, batter drains, catch drains, delivers value for money, encourages shared
culverts, hydraulic design, headwater, tailwater, inlet control, outlet control, knowledge and drives consistency for road users.
blockage, floodways, submergence, time of closure Austroads is governed by a Board consisting of
senior executive representatives from each of its
Edition 1.1 published August 2018 eleven member organisations:
• Format updated • Roads and Maritime Services New South Wales

Edition 1.0 published May 2013 • Roads Corporation Victoria
• Queensland Department of Transport and Main
Roads
• Main Roads Western Australia
• Department of Planning, Transport and
ISBN 978-1-925037-03-6 Pages 176 Infrastructure South Australia
Austroads Project No. TP1667 • Department of State Growth Tasmania
Austroads Publication No. AGRD05B-13 • Department of Infrastructure, Planning and
Logistics Northern Territory
• Transport Canberra and City Services
© Austroads Ltd 2013 Directorate, Australian Capital Territory
This work is copyright. Apart from any use as permitted under the • Australian Government Department of
Copyright Act 1968, no part may be reproduced by any process without Infrastructure and Regional Development
the prior written permission of Austroads. • Australian Local Government Association
• New Zealand Transport Agency.
This Guide is produced by Austroads as a general guide. Its application is discretionary. Road authorities may vary their practice
according to local circumstances and policies. Austroads believes this publication to be correct at the time of printing and does not
accept responsibility for any consequences arising from the use of information herein. Readers should rely on their own skill and
judgement to apply information to particular issues.
Acknowledgements
The authors acknowledge the role and contribution of the Austroads Road Design Task Force
in providing guidance and information during the preparation of this guide. The Task Force comprised the following members:
• Mr Pat Kenny – Roads and Maritime Services, New South Wales

• Mr Richard Fanning – Roads Corporation, Victoria
• Dr Owen Arndt – Department of Transport and Main Roads, Queensland
• Mr Rob Grove – Main Roads Western Australia
• Mr Noel O’Callaghan – Department for Planning, Transport, and Infrastructure, South Australia
• Mr Shane Gregory – Department of Infrastructure, Energy and Resources, Tasmania
• Mr Bryan Matyorauta – Northern Territory Department of Transport
• Mr Ken Marshall – ACT Department of Territory and Municipal Services
• Ms Sylvia Soon – Australian Local Government Association
• Mr James Hughes – NZ Transport Agency
• Mr Tom Brock – Consult Australia
• Mr Peter Watts – Australian Bicycle Council
• Mr Michael Tziotis – ARRB Group Ltd.
Road Design Drainage Technical Sub-committee
The Austroads Road Design Task Force and Project Manager acknowledge the role and contribution of the AGRD Part 5: Drainage,
Technical Sub-committee in the development and preparation of this guide. The Sub-committee comprised the following members:
• Ms Marianne Robertson – Roads Corporation Victoria (VicRoads) – Project Manager

• Mr Mike Whitehead – Department of Transport and Main Roads Queensland
• Mr Michael Vujcich – BG&E & Main Roads Western Australia (MRWA)
• Mr Peter McCarten – Opus International for NZ Transport Agency
• Mr David House – Roads and Maritime Services, New South Wales
• Mr Jan Mikitczuk – Roads Corporation, Victoria (VicRoads)
• Ms Sylvia Soon – Australian Local Government Association
• Mr Peter Aumann – ARRB Group Ltd – Project Leader.
Catchments and creeks have ownership of figure 3.15 and figure 3.16 and have been reproduced with permission. These images
cannot be reproduced within another publication without the written permission from the director of catchments and creeks pty ltd.
Summary
The Guide to Road Design Part 5B: Drainage – Open Channels, Culverts and Floodways, provides road
designers and other practitioners with guidance on the design of open channels, culverts and floodways.
This guide does not provide all the information necessary to complete a design and needs to be used in
conjunction with the other two parts of the Guide to Road Design Part 5: Drainage, that relate to drainage
design, namely:
• Part 5: Drainage – General and Hydrology Considerations
• Part 5A: Drainage – Road Surface, Networks, Basins and Subsurface.
The guide includes design processes and formula necessary to design effective drainage systems and
infrastructure. It is supported by appendices containing design charts and worked examples that provide
further information.
This guide provides guidelines on good practice in relation to drainage design that will apply in most
situations, rather than specifying mandatory practice. The reason for not specifying mandatory practice is
that there are many factors that influence the design of a road and drainage for a particular situation or site,
and practitioners therefore need to exercise sound judgement in applying the information contained in this
guide.
Austroads2013 | page i
Contents
1. Introduction............................................................................................................................................. 1
1.1 Purpose .................................................................................................................................................... 1
1.2 Scope of this Part ..................................................................................................................................... 2
1.3 Road Safety .............................................................................................................................................. 3
2. Open Drains and Channels ................................................................................................................... 4

2.1 Introduction ............................................................................................................................................... 4
2.1.1 Definition and Use ...................................................................................................................... 4
2.1.2 Factors in Choice of Channel Type ............................................................................................ 5
2.1.3 Environmental Consultation........................................................................................................ 5
2.1.4 Types of Open Drains ................................................................................................................. 6
2.2 General Considerations ............................................................................................................................ 9
2.3 Fundamentals of Open Channel Flow .................................................................................................... 10
2.3.1 Stream Dynamics ..................................................................................................................... 10
2.3.2 Assumptions for Analysis ......................................................................................................... 12
2.3.3 Fundamental Equations ............................................................................................................ 12
2.3.4 Application of Fundamental Equations ..................................................................................... 17
2.3.5 Energy Principles ...................................................................................................................... 20
2.3.6 Hydraulic Jump ......................................................................................................................... 24
2.3.7 Hydraulic Drop .......................................................................................................................... 30
2.4 Erosive Velocities in Natural Streams .................................................................................................... 30
2.5 Backwater ............................................................................................................................................... 31
2.5.1 Tidal Waters.............................................................................................................................. 31
2.5.2 Downstream Tributary .............................................................................................................. 31
2.6 Tailwater Levels ...................................................................................................................................... 31
2.6.1 Tailwater Effects ....................................................................................................................... 31
2.6.2 Design Tailwater Levels ........................................................................................................... 32
2.7 Open Channel Design ............................................................................................................................ 32
2.7.1 Design Methodology ................................................................................................................. 33
2.7.2 Channel Transitions .................................................................................................................. 38
2.7.3 Energy Losses in Channel Bends ............................................................................................ 39
2.7.4 Superelevation in Channel Bends ............................................................................................ 40
2.8 Grassed Channels .................................................................................................................................. 40
2.8.1 Normal Grassed Channels ....................................................................................................... 40
2.8.2 Reinforced Grassed Channels ................................................................................................. 41
2.9 Channels Lined with Hard Facings ......................................................................................................... 42
2.9.1 General ..................................................................................................................................... 42
2.9.2 Riprap and Rock Filled Wire Mattresses/Gabions.................................................................... 42
2.9.3 Concrete Lined Channels ......................................................................................................... 45
2.10 Channel Drops ........................................................................................................................................ 46
2.10.1 Check Dams ............................................................................................................................. 46
2.11 Baffle Chutes .......................................................................................................................................... 46
2.12 Edge Drains and Median Drains ............................................................................................................ 47
2.12.1 Design of Edge and Median Drainage ..................................................................................... 47
2.13 Table Drains and Table Drain Blocks ..................................................................................................... 50
2.13.1 Table Drains ............................................................................................................................. 50
2.13.2 Table Drain Blocks ................................................................................................................... 52
2.13.3 Diversion Drains and Diversion Blocks .................................................................................... 52
2.14 Batter Drains and Chutes ....................................................................................................................... 54
2.14.1 Design Procedures – Batter Drains .......................................................................................... 54
2.15 Catch Drains and Catch Banks .............................................................................................................. 55
2.15.1 Catch Drains ............................................................................................................................. 55
2.15.2 Catch Banks ............................................................................................................................. 56
2.15.3 Design Procedures – Catch Drains and Catch Banks ............................................................. 57
Austroads2013 | page ii
2.16 Worked Examples .................................................................................................................................. 57

2.16.1 Example 1: Manning’s Equation – Determine the Flow Rate, Velocity and State of Flow
for a Stream .............................................................................................................................. 57
2.16.2 Example 2: Manning’s Equation – Determine the Depth and Velocity of Flow Based on a
Known Discharge/Flow Rate in a Stream ................................................................................. 59
2.16.3 Example 3: Manning’s Equation: Determine the Flow Rate and Average Velocity in a
Compound Stream ................................................................................................................... 62
2.16.4 Example 4: Hard-lined/Rigid Boundary Channel Design ......................................................... 64
2.16.5 Example 5: Vegetated/Erodible Boundary Channel Design ..................................................... 67
3. Culverts ................................................................................................................................................. 71
3.1 Introduction ............................................................................................................................................. 71
3.2 Information Required .............................................................................................................................. 71
3.3 Culvert Location...................................................................................................................................... 72
3.3.1 Horizontal Alignment ................................................................................................................ 72
3.3.2 Vertical Profile........................................................................................................................... 73
3.3.3 Non-hydraulic Considerations .................................................................................................. 73
3.4 Culvert Type ........................................................................................................................................... 75
3.4.1 Culvert Shape ........................................................................................................................... 75
3.4.2 Materials ................................................................................................................................... 75
3.5 Culvert Size ............................................................................................................................................ 76
3.5.1 Minimum Culvert Size Allowable .............................................................................................. 76
3.5.2 Other Sizing Considerations ..................................................................................................... 76
3.6 Structural Requirements ......................................................................................................................... 77
3.6.1 Design Loads ............................................................................................................................ 77
3.6.2 Existing Culverts ....................................................................................................................... 77
3.6.3 Space between Pipes in Multiple–Barrel Culverts.................................................................... 78
3.6.4 Cover ........................................................................................................................................ 78
3.7 Hydraulic Design Considerations ........................................................................................................... 78
3.7.1 Design Discharge ..................................................................................................................... 78
3.7.2 Outlet Velocity........................................................................................................................... 79
3.7.3 Culverts in Flat Terrain ............................................................................................................. 79
3.7.4 Siltation and Blockage .............................................................................................................. 80
3.7.5 Allowable Headwater (AHW) .................................................................................................... 80
3.7.6 Tailwater (TW) .......................................................................................................................... 81
3.8 Typical Culvert Operating Conditions ..................................................................................................... 81
3.8.1 General ..................................................................................................................................... 81
3.8.2 Inlet Control Conditions ............................................................................................................ 81
3.8.3 Outlet Control Conditions ......................................................................................................... 84
3.9 Hydraulic Calculations ............................................................................................................................ 84
3.9.1 Control at Outlet ....................................................................................................................... 84
3.9.2 Determination of Tailwater ........................................................................................................ 88
3.10 Design Procedures ................................................................................................................................. 88
3.10.1 Hydraulic Design ...................................................................................................................... 88
3.10.2 Collect Design Data .................................................................................................................. 88
3.10.3 Select a Trial Culvert ................................................................................................................ 91
3.10.4 Design Discharge for Trials ...................................................................................................... 91
3.10.5 Determine Inlet Control Headwater Depth ............................................................................... 92
3.10.6 Determine Outlet Control Headwater Depth ............................................................................. 92
3.10.7 Determine the Controlling Headwater ...................................................................................... 93
3.10.8 Outlet Velocity – Outlet Control ................................................................................................ 93
3.10.9 Outlet Velocity – Inlet Control ................................................................................................... 95
3.10.10 Outlet Flow Energy ................................................................................................................... 96
3.10.11 Design Check ........................................................................................................................... 97
3.10.12 Practical Design ........................................................................................................................ 97
3.11 Blockage of Culverts............................................................................................................................... 97
3.12 Consideration of Large or Extreme Events ............................................................................................ 98
Austroads2013 | page iii

3.13 Culvert Outlet Protection ........................................................................................................................ 99

3.14 Culvert End Treatments ....................................................................................................................... 101
3.14.1 General ................................................................................................................................... 101
3.14.2 Slope Faced............................................................................................................................ 102
3.14.3 Traversable Endwalls ............................................................................................................. 102
3.15 Worked Examples ................................................................................................................................ 103
3.15.1 Example 1: Application of Culvert Design Procedure ............................................................ 103
3.15.2 Example 2: Culvert Outlet Protection ..................................................................................... 109
4. Floodways ........................................................................................................................................... 112

4.1 Introduction ........................................................................................................................................... 112
4.2 Design Considerations ......................................................................................................................... 112
4.2.1 General ................................................................................................................................... 112
4.2.2 Geometric ............................................................................................................................... 114
4.2.3 Environmental ......................................................................................................................... 114
4.2.4 Hydraulic ................................................................................................................................. 115
4.3 Hydraulic Design .................................................................................................................................. 117
4.3.1 Flow Over the Road ................................................................................................................ 117
4.3.2 Full Floodway Calculations ..................................................................................................... 119
4.4 Time of Submergence/Closure............................................................................................................. 121
4.4.1 Introduction ............................................................................................................................. 121
4.4.2 Time of Submergence ............................................................................................................ 121
4.4.3 Time of Closure ...................................................................................................................... 122
4.4.4 Issues Related to Times ......................................................................................................... 122
4.4.5 Calculation of Time of Submergence or Closure.................................................................... 123
4.4.6 Procedure for Estimating Average Annual Time of Closure (AAToC).................................... 124
4.5 Floodway Protection ............................................................................................................................. 125
4.5.1 Introduction ............................................................................................................................. 125
4.5.2 Floodways with Grassed Batters ............................................................................................ 126
4.5.3 Other Floodways .................................................................................................................... 126
4.6 Worked Examples ................................................................................................................................ 131
4.6.1 Example 1: Floodway Design ................................................................................................. 131
4.6.2 Example 2: Floodway Design ................................................................................................. 134
References .................................................................................................................................................... 140

Vegetal Retardance Curves .................................................................................................. 142
Inlet and Outlet Control Nomographs ................................................................................. 147
Velocity and Critical Depth ................................................................................................... 158
Culvert Capacity .................................................................................................................... 164
Solving Manning’s Equation ................................................................................................ 168
Tables
Table 2.1: Manning’s n values for artificial channels/conduits ................................................................. 15
Table 2.2: Manning’s n values for natural channels ................................................................................ 16
Table 2.3: Guide to the selection of vegetal retardance .......................................................................... 17
Table 2.4: Jump types and strength ......................................................................................................... 25
Table 2.5: Erosive velocities in natural streams ...................................................................................... 31
Table 2.6: Recommended maximum velocities (design) for consolidated, bare channels
and vegetated channels .......................................................................................................... 36
Table 2.7: Channel transition energy loss coefficients ............................................................................ 39
Table 3.1: Suggested culvert outlet velocities.......................................................................................... 79
Table 3.2: Culvert entry loss coefficient ................................................................................................... 87
Table 4.1: Design of rock slope protection............................................................................................. 130
Austroads2013 | page iv
Table 4.2: Standard classes of rock slope protection ............................................................................ 131
Figures
Figure 1.1: Flow chart of Guide to Road Design ......................................................................................... 2
Figure 2.1: Bare earth open channel ........................................................................................................... 4
Figure 2.2: Types of open drains ................................................................................................................ 6
Figure 2.3: Batter drain................................................................................................................................ 8
Figure 2.4: Bench drain ............................................................................................................................... 8
Figure 2.5: Contour drain ............................................................................................................................ 9
Figure 2.6: Swale ........................................................................................................................................ 9
Figure 2.7: Velocity distribution across the flow ........................................................................................ 11
Figure 2.8: Wetted perimeter..................................................................................................................... 13
Figure 2.9: Stage-Discharge curve ........................................................................................................... 18
Figure 2.10: Modified Stage-Discharge curve ............................................................................................. 19
Figure 2.11: Compound channel ................................................................................................................. 19
Figure 2.12: Comparing energy between points ......................................................................................... 21
Figure 2.13: Specific energy curve .............................................................................................................. 22
Figure 2.14: Analysis curves and the hydraulic jump .................................................................................. 25
Figure 2.15: Relation between Froude Number and depths of flow in a hydraulic jump ............................27
Figure 2.16: Sequent depth for trapezoidal channels (So ≤ 10%) ............................................................... 28
Figure 2.17: Location of a hydraulic jump ................................................................................................... 29
Figure 2.18: Length of hydraulic jump in terms of sequent depth (y2) in horizontal channels ....................30
Figure 2.19: Trapezoidal channel section ................................................................................................... 34
Figure 2.20: Maximum rates of contraction and expansion ........................................................................ 38
Figure 2.21: Channel transition types ......................................................................................................... 39
Figure 2.22: Relationship between bed velocity and rock diameter............................................................ 44
Figure 2.23: Channel lined with rock filled wire mattresses ........................................................................ 45
Figure 2.24: Diagram of a baffle chute ........................................................................................................ 47
Figure 2.25: Median inlets with levees ........................................................................................................ 49
Figure 2.26: Typical table drains ................................................................................................................. 50
Figure 2.27: Table drain location and shape ............................................................................................... 51
Figure 2.28: Typical catch drain with flat bottom ......................................................................................... 56
Figure 2.29: Catch drain with V shape diagram .......................................................................................... 56
Figure 2.30: Catch bank diagram ................................................................................................................ 56
Figure 2.31: Concrete-lined catch drain ...................................................................................................... 57
Figure 2.32: Stage–Discharge curve ........................................................................................................... 62
Figure 3.1: Stream alignment options ....................................................................................................... 73
Figure 3.2: Stream alignment options (A and B) (preferred option A to avoid bank erosion) ...............................73
Figure 3.3: Culvert grading options ........................................................................................................... 74
Figure 3.4: Development of headwater ..................................................................................................... 80
Figure 3.5: Typical conditions under which standard culverts operate .....................................................83
Figure 3.6: Hydraulics of culvert flowing full under outlet control .............................................................. 85
Figure 3.7: Determination of adopted outlet depth (ho) ............................................................................. 86
Figure 3.8: Form for culvert design calculations ....................................................................................... 89
Figure 3.9: Culvert design flow chart in steps 1 to 10 ............................................................................... 90
Figure 3.10: Flow area for box culverts ....................................................................................................... 94
Figure 3.11: Flow area for pipe culverts ...................................................................................................... 94
Figure 3.12: Full flow/part flow table ........................................................................................................... 96
Figure 3.14: Flood water flowing into box culverts ...................................................................................... 98
Figure 3.15: Single pipe outlet minimum rock size and length of apron ...................................................100
Figure 3.16: Rock apron detail for single pipe outlet ................................................................................. 100
Figure 3.17: Minimum rock size and length of apron for a multi-pipe outlet .............................................101
Figure 3.18: Example of a slope faced and slope faced with grate culvert endwall .................................103
Figure 3.19: Channel cross-section .......................................................................................................... 103
Figure 3.20: Road cross-section for worked example ............................................................................... 104
Figure 3.21: Stage-Discharge curve for culvert design example .............................................................. 105
Austroads2013 | page v
Figure 3.22: Geometric detail of key parameters for worked example .....................................................105
Figure 3.23: Worksheet for example culvert design .................................................................................. 108
Figure 3.24: Application of Figure 3.15 to worked example ...................................................................... 110
Figure 3.25: Solution to example for culvert outlet protection ................................................................... 111
Figure 4.1: Example of a longitudinal shape and cross–section of a floodway ......................................113
Figure 4.2: Indicative velocities of flow over a typical floodway .............................................................. 116
Figure 4.3: Discharge coefficients for flow over floodways ..................................................................... 118
Figure 4.4: Example floodway calculations ............................................................................................. 120
Figure 4.5: Downstream floodway protection (Types 1, 2 & 3) ............................................................... 129
Figure 4.6: Downstream floodway protection (Types 4 & 5) ................................................................... 130
Figure 4.7: Floodway cross-section ........................................................................................................ 132
Figure 4.8: Details of example floodway design...................................................................................... 136
Figure 4.9: Head on roadway .................................................................................................................. 138
Austroads2013 | page vi
1. Introduction
1.1 Purpose
Austroads Guide to Road Design seeks to capture the contemporary road design practice of member
organisations (Austroads Guide to Road Design Part 1: Introduction to Road Design (Austroads 2010a)). In
doing so, it provides valuable guidance to designers in the production of safe, economical and efficient road
designs.
The purpose of the Guide to Road Design Part 5: Drainage is to provide designers with guidance to
appropriately manage stormwater run-off through the design of an efficient and effective road drainage
system. This Guide comprises three parts:
• Part 5: Drainage – General and Hydrology Considerations (AGRD Part 5) (Austroads 2013a).
• Part 5A: Drainage – Road Surface, Network, Basins and Subsurface (AGRD Part 5A) (Austroads 2013b).
• Part 5B: Drainage – Open Channels, Culverts, and Floodways (AGRD Part 5B).
AGRD Part 5 (Austroads 2013a) of the Guide provides an introduction to the three parts, an overview of the
drainage considerations in undertaking a drainage design, safety considerations, environmental issues and
detailed guidance on the hydrologic assessment of a catchment. AGRD Part 5A provides guidance on road
surface flows, design of the pit and pipe systems and basins and introduction to the major/minor flow
concept.
The purpose of this Part is to provide road designers with guidance on the requirements for the design of
open channels, culverts and floodways, to support the operation and management of the road network.
The AGRD Part 5: Drainage is one of eight parts of the Guide to Road Design (see Figure 1.1) that provide
guidance on geometric design, intersection design, roadside design and geotechnical design, all of which
may influence the location and design of the drainage system.
Austroads 2013 | page 1

Figure 1.1: Flow chart of Guide to Road Design
Note:
Part 4 of the Guide to Road Design comprises four parts, namely:
• Part 4: Intersections and Crossings – General.
• Part 4A: Unsignalised and Signalised Intersections.
• Part 4B: Roundabouts.
• Part 4C: Interchanges.
Part 5 of the Guide to Road Design comprises three parts, namely:
• Part 5: Drainage – General and Hydrology Considerations.
• Part 5A: Drainage – Road Surface, Network, Basins and Subsurface.
• Part 5B: Drainage – Open Channels, Culverts and Floodways.
Part 6 of the Guide to Road Design comprises three parts, namely:
• Part 6: Roadside Design, Safety and barriers.
• Part 6A: Pedestrian and Cyclist Paths.
• Part 6B: Roadside Environment.
1.2 Scope of this Part
This Part describes the types of open channels, culverts and floodways and provides design procedures and
guidance on best practice for the detailed design of these components of the drainage system.
When used in conjunction with other relevant parts of the Guide provides the information and guidance
necessary for a road designer to prepare detailed design drawings that are adequate to facilitate the
construction of the drainage system.
Figure 1.1 outlines the structure of the Guide to Road Design. Designers should be aware that there are
nine other subject areas spanning the range of Austroads publications that may also be relevant to the
design of roads (www.austroads.com.au).

1.3 Road Safety
Adopting a Safe System approach to road safety recognises that humans as road users are fallible and will
continue to make mistakes, and that the community should not penalise people with death or serious injury
when they do make mistakes. In a Safe System, roads (and vehicles) should be designed to reduce the
incidence and severity of crashes when they inevitably occur.
The Safe System approach requires, in part (Australian Transport Council 2011):
• roads and roadsides designed and maintained to reduce the risk of crashes occurring and to lessen the
severity of injury if a crash does occur. Safe roads prevent unintended use through design and
encourage safe behaviour by users
• forgiving environments that prevent serious injury or death when crashes occur
• speed limits that complement the road environment to manage crash impact forces to within human
tolerance; and all road users complying with the speed limits.
In New Zealand, practical steps have been taken to give effect to similar guiding principles through a Safety
Management Systems (SMS) approach.
Road designers should be aware of and through the design process, actively support the philosophy and
road safety objectives covered in the Guide to Road Safety.

2. Open Drains and Channels
2.1 Introduction
2.1.1 Definition and Use
An open channel is defined as a conduit or conveyance (artificial or natural) in which water flows with a free
surface. A free surface means that the surface is open to the atmosphere and/or there is no additional
pressure on the flow other than atmospheric pressure. Flow is caused by gravity and streams tend to follow
the path of least resistance.
Open channels can be economical where large flows are to be carried and space is not restricted. An
example of a newly constructed, open channel is shown in Figure 2.1. This channel would then be treated to
avoid erosion, e.g. grassed.
Figure 2.1: Bare earth open channel
Source: DTMR (2010).
Open channels may be constructed to achieve one of three main functions:

• to capture and convey run-off originating from the road formation – e.g. table drains forming part of the
road drainage system where space within the road reserve is sufficient
• to capture and convey overland flows before they reach the road formation, including catch drains at the
top of cut batters to prevent scour of the embankments or side drains to intercept sheet flows and protect
the road drainage system from inundation
• to capture and convey flows beneath the road formation, including flows from the outlets of culverts or
other drainage infrastructure.
This section focuses on the analysis and design of smaller natural streams and creeks and constructed
drains and channels. Assessment of larger streams, creeks, rivers and floodplains is complex and should be
referred to specialists in hydraulics and river engineering.

Where open channels are located adjacent to a road, they should be designed to cater for errant vehicles,
see the Guide to Road Design Part 3: Geometric Design (AGRD Part 3) (Austroads 2010b) and the Guide to
Road Design Part 6: Roadside Design, Safety and Barriers (AGRD Part 6) (Austroads 2010c).
2.1.2 Factors in Choice of Channel Type
In choosing the type of channel to be adopted, consideration should be given to:

• capacity
• erosion
• maintenance.
Capacity
The channel should have adequate capacity for the design flow.
Erosion
Erosion control is a necessary part of good drainage design. Scour may occur unless the channel is
protected where velocities exceed those likely to cause erosion to the material forming the channel. Erosion
control involves the selection of a suitable and economical channel lining (including vegetative cover) which
will give the desired protection.
For further information on erosion control see AGRD Part 5 – Section 3.6. The type of lining should be
consistent with the degree of protection required, overall cost including maintenance, safety requirements
and aesthetic considerations. Erosion control in the form of grass growth may be used in combination with
other types of lining. A channel may be grass lined on the flatter slopes and lined with a more scour resistant
material on the steeper slopes.
Maintenance
Without proper maintenance, a well-designed channel becomes unsightly and will perform unsatisfactorily at
the design flow. Maintenance methods should be considered in the design of drainage channels so that the
type of channel section adopted will be suitable for the methods and equipment that will be used for
maintenance.
2.1.3 Environmental Consultation
Creeks and waterways disturbed by construction work should be left in a way that the channel will continue
to behave in a hydraulic and ecological manner similar to that of the undisturbed natural stream. New open
channels must be appropriately integrated into the surrounding ecological, visual, social and physical
environments.
Environment and drainage authorities should be consulted early in the road planning process as they may
have special requirements for existing channels which may relate to:
• soil conservation programs
• stream rehabilitation or restoration plans
• riparian improvement programs
• natural channel design programs
• other programs relating to ecological sustainability.

2.1.4 Types of Open Drains
Figure 2.2 shows an example of a typical open drain network associated with a road alignment and the
following section provides an outline of each of types of open drains.
Figure 2.2: Types of open drains
Drain type B Drain type B
Drain type C
Notes:
Drain type A – Diversion drain.
Drain type B – Side drain.
Drain type C – Table drain.
Table drains
Table drains are located along the outer edge of the road shoulder in cuts, and beside shallow raised
carriageways in fill. They collect run-off from the pavement, shoulders and cut batters and convey the flows
to a suitable outfall, which could be via a diversion drain or to a culvert.
Diversion drains
The purpose of diversion drains is to convey water collected within the table drain to the designated drainage
outlet. Appropriate outlets include side drains, creeks, or even the open countryside, with run-off spilling
back out to overland flow.
The spacing of diversion drains should be such that the flow in a table drain does not exceed the capacity of
the table drain. Often for simplicity diversion drains may be spaced on set intervals.
Flows which may cause problems in one location can be redirected using diversion drains. Diversion drains
may be employed to ensure that water volumes do not reach velocities sufficient to cause significant
damage.
In order to ensure effective diversion of stormwater away from the table drain, table drain blocks may be
placed immediately downstream of its confluence with the diversion drain.

Catch drains
Catch drains (sometimes known as cut-off drains) intercept the surface water at the top of cut batters in order
to prevent rilling (the forming of small channels across the surface), erosion or scouring of the batter slope.
This type of drain is usually about 0.3 m deep but capacity should also be checked. Alternatively, catch
drains placed at the bottom of fill slopes intercept water from adjacent properties as well as convey road
drainage to an outlet. This type has a flat bottom 2.0 to 2.5 m wide and a depth sufficient to carry the design
flow.
In erodible soils, the catch drain may take the form of a low levee bank along the top of the batter. In such
soils, a drain cut into the surface may rapidly erode and enlarge itself, or cause local slips in the batter by
piping. These issues need to be addressed in the location and design of catch drains. V-shaped drains are
not preferred and should not be used in erodible soils.
Side drains
The main purpose of side drains is to isolate the road from run-off generated from adjoining areas. They are
often constructed as a small levee when it is undesirable to excavate a drain.
Median drains
Median drains collect run-off from the roadway pavement and median and direct the flow to the pavement
drainage system. The main limitation on median drains relates to safe slopes for errant vehicles. A
desirable side slope of one in 10 or flatter severely restricts the capacity of such drains unless the median is
very wide. Road safety requirements may result in the median drain being augmented by grated pits and
underground pipes.
Inlet or outlet drains
These drains direct water towards the culvert inlet on the upstream side and convey water away from the
culvert outlet and surrounding area on the downstream side. They are commonly used where the inlet or
outlet invert levels of the culvert are below the natural surface. In such cases the length of drain required is
determined by ‘daylighting’ the drain, i.e. construction of a drain at a set slope until it breaks out (daylights) at
the natural surface.
Batter drains
A batter drain, also known as a slope drain, is a lined open drain for removing stormwater from the top to the
toe of the batter in order to reduce scour of the batter face, as shown in Figure 2.3. They are spaced at
intervals that meet the maximum flow width criteria.
Further information on these drains identified above is contained in Section 2.12 – Edge Drains and Median
Drains to Section 2.15 – Catch Drains and Catch Banks.

Figure 2.3: Batter drain
Source: D House (used with permission).
Bench drains
A bench drain is provided for the bench (i.e. ledge) that is constructed on a batter, or natural slope. The
purpose of benches is to reduce erosion to the batter faces, reduce the amount of water in cuttings to be
carried by table drains and in some cases to also improve sight distance on horizontal curves. Bench drains
carry water from the bench to suitable drainage outlets. An example of a bench drain is shown in Figure 2.4.
Figure 2.4: Bench drain
Source: Austroads (2010b).
Contour drain
A contour drain, also known as a contour bank, is a surface drain designed to slow the rate of run-off by
diverting water along a gently sloped path, from a site to nearby stable areas, at a discharge velocity that will
not cause erosion. An example is shown in Figure 2.5.

Figure 2.5: Contour drain
Source: RTA (1993).
Swales
Vegetation can be used to provide the filtering surface area to spread and reduce flow velocities that allows
sedimentation as well as providing a substrate for biofilm growth and hence biological uptake of soluble
pollutants. An example of a swale drain is shown in Figure 2.6.
Figure 2.6: Swale
Source: D House (used with permission).
2.2 General Considerations
The design considerations for an open channel include:

• hydrological analysis to determine the discharge for the proposed open drain or channel design
• establishing the ecological requirements for the channel such as
– determining the specific or generic riparian management requirements

– stream bank stability or erosion guidelines

– water quality requirements
– natural channel design requirements
– use of the channel by fauna as a fishway or wildlife corridor
• obtaining permits that may be required for any work in, or over a waterway
• optimising the longitudinal slope of the channel. To reduce the velocity, a flatter slope than that existing
may be adopted by introducing drop structures or meanders. However, shortening the natural channel
alignment will increase velocities
• determining the width available for an open channel
• assessing the geotechnical requirements for the drain or channel side slope stability
• assessing the need for channel lining, particularly for erosion resistance, including vegetation selection
and aesthetics. Ensure future maintainability of lining is considered
• calculating flow characteristics. Design for subcritical flow; see Section 2.3.5 for guidance on Specific
Energy and Froude’s Number. Supercritical flow with its highly erosive velocities should be avoided in a
design flood, except at channel drops or energy dissipators (see Section 2.10 – Channel Drops)
• assessing and ensuring the safety of an open channel
• providing access for maintenance.
2.3 Fundamentals of Open Channel Flow
2.3.1 Stream Dynamics
It is important to have a basic understanding of how streams work or behave to ensure that the planning and
design of drainage infrastructure is appropriate and will work not only immediately after the work is
completed, but also in the future.
Streams are constantly changing and responding to external factors and are therefore dynamic. However, to
simplify analysis and design, it is assumed that the stream cross-section is fixed and uniform flow conditions
occur, that is, the stream is divided into small, isolated parts and/or sections and the flow within the channel
is analysed at a point in time.
Streams and roads
Human interference can easily change or affect streams and it is critical to understand that changes can
affect stream behaviour both upstream and downstream.
Typical encroachments that would affect normal stream flow as related to road infrastructure are:
• roadway embankments
• cross drainage culverts
• floodways
• bridges.
Governing parameters
All streams are governed by the same basic parameters:

• geological factors – soil conditions

• hydrological factors – rainfall

• geometric – channel slope and crossfall
• hydraulic – fluid properties
• vegetation – native vegetation in channels/along banks.
Velocity distribution
Flow velocities within a stream are not uniformly distributed across the cross-section. They are slowest
adjacent to the channel bed and banks, where the friction forces are highest. The highest velocities are
found just below the free surface in the deepest part of the flow.
Figure 2.7 illustrates the velocity distribution across channel flow. To map water particles of equal velocity,
isovels are drawn (isovels trace equal velocity similar to contour lines tracing equal ground height).
Figure 2.7: Velocity distribution across the flow
Types and states of flow
Open channel flow can be described in a number of ways. The following outlines the types of flow:
• Steady flow – is where the depth of flow at the same point does not change over time.
• Unsteady flow – is where the depth of flow at the same point does change over time.
• Uniform flow – is where the depth of flow is the same at every cross-section along the channel. This is
really only possible when the channel flow cross-section, slope and roughness remain constant.
• Non-uniform or varied flow – is where the depth of flow changes along the length of channel. Varied
flow can be further described as either rapid or gradual (i.e. in artificial channels).
The following outlines the states of flow:

• Laminar flow – is where the water particles appear to move in smooth paths or in very thin sheets
(laminas) over the top of each other. Laminar flow is very smooth flow and is rarely seen in the natural
environment.
• Turbulent flow – is where the water particles move in very irregular paths. White water rapids are good
examples of turbulent flow.
Open channel flow can occur as a variety or combination of these types and states.

2.3.2 Assumptions for Analysis
Due to the dynamic nature of open channel flow, it is difficult to model and/or analyse the flow as it occurs in
the natural environment. Therefore a number of assumptions are made to enable and simplify analysis.
These assumptions are considered reasonable and valid in the context of the analysis and design
procedures described in this Guide.
The assumptions made are:

• steady flow
• uniform flow
• velocity is averaged over the whole cross-section
• flow is non-turbulent.
Generally, these assumptions can be applied to most situations encountered in design. However, where it is
obvious that the assumptions cannot be applied, the analysis and/or design should be referred to a suitably
qualified specialist.
2.3.3 Fundamental Equations
The most basic equation in the analysis of open channel flow defines the relationship between flow rate
(discharge), velocity and the cross-sectional area of flow and is represented by Equation 1:
𝑄𝑄 = 𝑉𝑉 x 𝐴𝐴 1
where
Q = Flow rate (m/s)
V = Average velocity (m/s)
A = Cross-sectional area of flow (m2)
This equation also forms the basis of the theory behind the Continuity Equation.
The theory allows simple analysis over changes in the channel irrespective of cross-section, slope or
roughness. The theory assumes no addition to or subtraction from the flow between the two sections being
considered (i.e. Equation 2: Q1 = Q2).
𝑄𝑄1 = 𝑉𝑉1 x 𝐴𝐴1 = 𝑉𝑉2 x 𝐴𝐴2 = 𝑄𝑄2 2
The formula most commonly used for the calculation of steady, uniform flow in open channels is Manning’s
Equation. This equation is used to determine the velocity of flow at a specific point in the channel, and
therefore the variables in the equation must be representative of the point being assessed.

Manning’s Equation (Equation 3) is:
2� 1�
R 3S 2 3
V=
𝑛𝑛
where
V = Average velocity (m/s)
R = Hydraulic radius (m)
S = Slope of energy line (m/m)
𝑛𝑛 = Manning’s roughness coefficient
The hydraulic radius is given by Equation 4:
𝐴𝐴 4
𝑅𝑅 =
𝑃𝑃
where
A = Cross-sectional area (m2)
P = Wetted perimeter (m)
In determining the hydraulic radius, the wetted perimeter is defined as the length of line (normal to the flow)
where the water touches the surface of the ground (channel) (Figure 2.8).
Figure 2.8: Wetted perimeter
A worked example of the application of Manning’s Equation, is shown in Section 2.16.1 – Example 1:
Manning’s Equation. Refer also to Appendix E.
For natural channels the slope of the energy line, S, is almost impossible to determine. Therefore a suitable
estimation of S is required. The slope of the water (flow) surface, Sw, could be used, however this is also
difficult to determine. The only easily available slope that can be used to estimate S is the channel bed, So,
provided that the channel and the bed slope are reasonably uniform near the point being assessed. If the
channel shape and bed slope is irregular, Manning’s Equation will not give acceptable results and therefore
the analysis and design should be referred to a suitably qualified hydraulics practitioner.

In applying Manning’s Equation, particularly to natural channels, the greatest difficulty lies in the
determination of the roughness coefficient n. For artificial channels values of n can be obtained from
Table 2.1.
For natural channels, use the values of n in Table 2.2 based on type and amount of vegetation present. For
artificial channels, e.g. conduits or lined channels, the determination of n must consider the roughness of
both the lining and jointing system.
For grassed channels, and where the hydraulic radius for the flow is less than 1 m, vegetal retardance
curves (see Appendix A) should be used to determine n, when the flow depth is greater than the height of
the grass. Table 2.3 is used to determine the appropriate vegetal retardance curve to be used.

Table 2.1: Manning’s n values for artificial channels/conduits
Type of structure n
1. Closed conduits
1. Concrete pipe 0.011–0.013
2. Corrugated metal pipe or pipe-arch (small corrugation)
(a) plain or unpaved 0.024
(b) paved invert – full flow
25% circumference paved 0.021
50% circumference paved 0.018
Fully paved 0.012
3. Structural plate pipe or pipe-arch 0.030–0.033
4. Monolithic concrete (box culvert) 0.012
5. Vitrified clay pipe 0.012
2. Open channels – lined
1. Concrete – smooth forms or trowelled 0.012
2. Asphalt
(a) Smooth 0.013
(b) Rough 0.016
3. Open channels – excavated (straight alignment and natural lining)
1. Earth – uniform section
(a) Clean – new to weathered 0.016–0.020
(b) With short grass, few weeds 0.022–0.027
(c) In gravelly soil, clean 0.022–0.025
2. Earth – uniform section
(a) No vegetation 0.022–0.025
(b) Grass, some weeds 0.025–0.030
(c) Dense weeds or plants in deep channel 0.030–0.035
(d) Sides clean, gravel bottom 0.025–0.030
(e) Sides clean, cobble bottom 0.030–0.040
3. Dragline excavated or dredged
(a) No vegetation 0.028–0.033
(b) Light bush on banks 0.035–0.050
4. Channels not maintained, weeds and brush uncut
(a) Dense weeds, high as flow depth 0.08–0.12
(b) Clean bottom, brush on sides 0.05–0.08
(c) Same, highest stage of flow 0.07–0.11
(d) Dense brush, high stage 0.10–0.14
Source: Based on Herr and Bossy, 1965. Item 1(b) and 1(c) have been added based on values recommended by
Maccaferri in their publications (cited in DTMR 2010, pp. 8–6).

Table 2.2: Manning’s n values for natural channels
Type of channel n
1. Main channel
1. Fairly regular section
(a) Some grass and weeds, little or no brush 0.030–0.035
(b) Dense growth of weeds, depth of flow materially greater than weed height 0.035–0.05
(c) Some weeds, light brush on banks 0.035–0.05
(d) Some weeds, heavy brush on banks 0.05–0.07
(e) Some weeds, dense willows on banks 0.06–0.08
(f) Trees within channel with branches submerged at high stage Add 0.01–0.02
2. Irregular Section, with pools, slight channel meander Add 0.01–0.02
To (a) to (f) above as applicable
3. Mountain streams, no vegetation in channel, banks usually steep, trees and brush along banks
submerged at high stage
(a) Bottom, gravel, shingle and few boulders 0.04–0.05
(b) Bottom, shingle with large boulders 0.05–0.07
2. Adjacent flood channels
1. Pasture, no brush
(a) Short grass 0.030–0.035
(b) High grass 0.035–0.05
2. Cultivated areas
(a) No crop 0.03–0.04
(b) Mature row crops 0.035–0.045
(c) Mature field crops 0.04–0.05
3. Heavy weeds, scattered brush 0.05–0.07
4. Light brush and trees 0.06–0.08
5. Medium to dense brush 0.10–0.16
6. Dense willows, summer, not bent over by current 0.15–0.20
7. Cleared land with tree stumps (250 to 450 per ha)
(a) No sprouts 0.04–0.05
(b) With heavy growth of sprouts 0.06–0.08
8. Heavy stand of timber, a few fallen trees, little undergrowth
(a) Flood depth below branches 0.10–0.12
(b) Flood depth reaches branches 0.12–0.16
Major Streams (surface width at flood stage > 30 m)
Roughness coefficient is usually less than for minor streams of similar description on account of 0.12–0.16
less effect offered by irregular banks or vegetation on banks. Values of n may be somewhat
reduced. The value of n for large streams of mostly regular section may be in the range.
Source: Based on DTMR (2010).

Table 2.3: Guide to the selection of vegetal retardance
Density of stand Average length of Degree of

Examples
(see Note 1) vegetation (mm) retardance
Thick Longer than 750 A Rhodes grass in ungrazed scrub soil waterway.
Thick 280 to 610 B Wheat 660 mm tall in 180 mm rows Rhodes grass.
B Kikuyu under maximum fertility conditions, long and
green.
B African star grass. Lucerne (see Note 2).
Thick 150 to 250 C Most grasses can be held at this retardance with
mowing or grazing e.g. Rhodes grass, couch grass,
carpet grass, native grasses.
Thick 50 to 150 D African star grass, kikuyu or couch grass, all under
heavy grazing.
Thick Less than 50 E Mowed lawn. Any grass burned short.
Fair Longer than 750 B –
Fair 280 to 610 C Rhodes grass under low fertility conditions .
Fair 150 to 250 D African star grass under low fertility conditions.
Fair 50 to 150 D –
Fair Less than 50 E –
Note: Values of n = 0.15 and greater have been used for fields of standing sugar cane.
1 Thickness of vegetation has an important bearing on retardance, possibly more important than species.
2 Vegetal retardance curves have been based on tests in experimental channels.
Source: Based on Soil Conservation Handbook and Kouwen et al. (cited in DTMR 2010, p. 8–9).
It is important to understand that the depth and velocity of any given flow in a channel operating under
normal, or natural conditions, are called the Normal Depth and Normal Velocity. The depth and velocity of
flow as determined using Manning’s Equation are the Normal Depth and Normal Velocity.
2.3.4 Application of Fundamental Equations
The primary usage of the fundamental equations is to ensure the design discharge can be accommodated
within the designed open drain or channel to the specified freeboard criteria and determine the basic flow
characteristics of a given channel. These characteristics are:
• maximum flow capacity of the channel
• velocity of flow when at maximum capacity
• depth of flow for a specific discharge
• velocity of flow for a specific discharge.
To determine these characteristics, a representative cross-section for the channel needs to be chosen as
well as channel slope (So) and Manning’s n value.
Manning’s Equation gives the velocity of flow in a channel based on a selected depth of flow (which enables
calculation of the cross-sectional area of flow and wetted perimeter). To determine the flow rate (Q), for the
selected depth of flow and using velocity as determined by Manning’s Equation, use the fundamental
equation, Equation 1, which is repeated here:
𝑄𝑄 = 𝑉𝑉 x 𝐴𝐴

If the selected depth of flow was based on the maximum capacity of the channel, the first two characteristics,
as listed above, have now been determined.
To determine the last two characteristics, a Stage-Discharge curve (also known as a rating curve) is
required. A Stage-Discharge curve plots the depth of flow against the flow rate at a particular cross-section
along a stream/channel. To plot a Stage-Discharge curve, a series of iterations of the above
calculations/process is required, with depths selected from zero to maximum capacity.
The curve is plotted using depth of flow for the x-axis and flow rate for the y-axis (Figure 2.9).
Figure 2.9: Stage-Discharge curve
To use the Stage-Discharge curve to determine the depth of flow for a specific discharge, plot the required
flow rate (Qy) onto the left y-axis and extend across the graph to the flow line and then project down to the x-
axis. To determine the velocity of this flow, the cross-sectional area of the flow needs to be calculated using
the determined depth of flow
However, the Stage-Discharge curve can be easily modified to include velocity by plotting the graph using
depth of flow for the x-axis, the flow rate for the left y-axis and velocity for the right y-axis (Figure 2.10).
To use the Modified Stage-Discharge curve to determine the depth and velocity of flow for a specific
discharge, plot the required flow rate (Qy) onto the left y-axis and extend across the graph to the flow line and
then project down to the x-axis, to determine depth, and up/down to the velocity line and then across to the
right y-axis to determine velocity.
The advantage in using a Stage-Discharge curve is that the curve can be used to determine the depth and
velocity of flows for several annual recurrence intervals. It should be noted that the Stage-Discharge curve is
linked to the channel characteristics. Should the analysis shift to a different location where channel
parameters are different (slope, shape n-value and so on), a new Stage-Discharge curve will be required.

Figure 2.10: Modified Stage-Discharge curve
A worked example of the application of Manning’s Equation and the use of a Stage-Discharge curve is
shown in Section 2.16.2 – Example 2: Manning’s Equation.
Compound channels
Simple channels are often represented as a basic trapezoid (see Figure 2.7), assessed using one Manning’s
n-value and/or based on one slope. This methodology is reasonable for simple, small channels. In reality,
the shape of channels is not a basic trapezoid. They are compound in shape with:
• high and low banks with possible small overflow sections
• different n-values for banks and channel bed
• potentially different slopes between the main channel and overflow sections.
Assessment of compound channels uses the same principle and methodology as described earlier in this
section; however the channel cross-section is first divided into smaller sub-sections.
Figure 2.11 shows, a compound channel divided into two sub-sections (labelled A and B).
Figure 2.11: Compound channel

Each section is analysed individually using Manning’s Equation to determine flow velocity based on the
parameters (S or So, n, A and P) specific to the sub-section. It is important to understand that the wetted
perimeter (P) is determined by the length of the ‘Water/Channel Boundary’ for the sub-section. The wetted
perimeter is the component of the equation where effect of friction is applied. The ‘Water/Water boundary’,
as shown by the red dashed line in Figure 2.11, is not to be included as the friction along this boundary is
negligible.
Once all sub-sections have been calculated, the total flow rate can be determined by Equation 5:
𝑄𝑄𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑉𝑉𝐴𝐴 x 𝐴𝐴𝐴𝐴 + 𝑉𝑉𝐵𝐵 x 𝐴𝐴𝐵𝐵 5
Review of the calculations shows that each section has a different average velocity. While an average
velocity can be calculated using QTotal and the cross-sectional area of the whole channel (AA + AB), the
different velocities across the channel represent reality reasonably well.
A worked example for a compound channel is shown in Section 2.16.3 – Example 3: Manning’s Equation.
2.3.5 Energy Principles
The energy in open channel flow can be expressed as the total head in metres (m) of water, which is equal
to the sum of the elevation above a datum (elevation head), the hydrostatic head and the velocity head. The
elevation head and hydrostatic head are components collectively known as the static head.
In conventional terms, energy has two components:

• potential energy
• kinetic energy.
Relating conventional terms to the energy within water, potential energy is equal to the static head and
kinetic energy is equal to the velocity head. As an example, a static body of water such as a lake, has
potential energy while a moving body of water such as a creek, has both potential energy and kinetic energy.
Energy equation
The total energy for an open channel section is given by Equation 6:
𝑉𝑉 2 6
𝐻𝐻 = (𝑧𝑧 + 𝑑𝑑) +
2 𝑔𝑔
where
𝐻𝐻 = Total energy head (m)
𝑧𝑧 = Height of channel bed above some reference datum (m), known as elevation head
𝑑𝑑 = Depth of flow normal to the direction of flow (m), known as hydrostatic head
𝑉𝑉 = Velocity of flow (m/s)
𝑔𝑔 = Acceleration due to gravity (9.81 m/s2)
The term V2/2g within this equation is known as the velocity head.

According to Chow and Henderson (both cited in DTMR 2010) the variable d describes the depth of flow
normal to the direction of flow (typically perpendicular to the channel bed).
It is important to understand that this equation differs slightly from the normal form of the energy equation for
open channels, which is Equation 7:
𝑉𝑉 2 7
𝐻𝐻 = (𝑧𝑧 + 𝑑𝑑 cos 𝜃𝜃) + 𝛼𝛼
2𝑔𝑔
where
𝜃𝜃 = Slope angle of channel bed
𝛼𝛼 = Coefficient accounting for velocity distribution
Channels assessed for the installation of cross drainage culverts are relatively flat and therefore θ can be
assumed to be zero.
Referring to Section 2.3.2 – Assumptions for Analysis, the velocity is averaged across the section of flow,
therefore, α is equal to 1.
The energy equation is useful when comparing two sections within a channel. Typically, as water flows
along a channel, energy is lost. However, the first law of thermodynamics is that energy can neither be
created nor destroyed.
Therefore, when comparing the energy between two sections, the relationship can be expressed with
Equation 8:
𝑉𝑉12 𝑉𝑉22 8
(𝑧𝑧1 + 𝑑𝑑1 ) + = (𝑧𝑧2 + 𝑑𝑑2 ) + + ℎ𝐿𝐿
2𝑔𝑔 2𝑔𝑔
where
ℎ𝐿𝐿 = Head loss (m)
This is best illustrated in Figure 2.12.
Figure 2.12: Comparing energy between points
𝑉𝑉12
2𝑔𝑔
𝑉𝑉22
2𝑔𝑔

Specific energy
Specific energy is defined as the energy per unit mass of water at any section, measured with respect to the
channel bed. Equation 9, the Specific Energy (Es) equation (which is essentially a sub-component of the
energy equation), is:
𝑉𝑉 2
𝐸𝐸𝑠𝑠 = y + 9
2𝑔𝑔
where
y = Flow depth (m)
When the depth of flow is plotted against the specific energy for a given channel section and discharge (only
bed slope changes from flat to steep), a specific energy curve is obtained, as shown in Figure 2.13. The
curve shows that for a given specific energy (Es), there are two possible (conjugate) depths called alternate
depths y1 and y2.
Figure 2.13: Specific energy curve
𝑉𝑉12
2𝑔𝑔 𝑉𝑉12
2𝑔𝑔
𝑉𝑉22
𝑉𝑉𝑐𝑐2 𝑉𝑉𝑐𝑐2
2𝑔𝑔
2 2𝑔𝑔
𝑉𝑉22 2𝑔𝑔𝑉𝑉2
2𝑔𝑔 2𝑔𝑔
𝑉𝑉12
𝐸𝐸𝑠𝑠 = 𝑦𝑦 +
2𝑔𝑔
Source: Franzini and Finnemore (1997) (cited in DTMR 2010).
At point C on the curve, the specific energy is at a minimum and there is only one depth. The depth
corresponding to this point is known as the critical depth, yc. The corresponding velocity is the critical
velocity (Vc) and the slope is the critical slope (Sc).
When the depth of flow is greater than the critical depth, the flow velocity is less than the critical velocity and
is referred to as subcritical flow. When the depth of flow is less than the critical depth, the flow is
supercritical and the velocity will be greater than the critical velocity. Subcritical flow is controlled by
downstream channel conditions, while supercritical flow is controlled from some upstream control condition.

Froude’s Number
Another simpler method to determine subcritical/supercritical flow is by determining Froude’s Number (Fr) for
the flow.
Froude’s Number is a dimensionless number, the ratio of the inertial forces to gravitational forces, and is
defined as:
𝑉𝑉 10
𝐹𝐹𝑟𝑟 =
�𝑔𝑔𝑔𝑔
where
𝑔𝑔 = Gravitational acceleration (9.81 m/s2)
𝐿𝐿 = Characteristic length (m)
The characteristic length (L) in open channel flow is equal to the hydraulic depth, which is defined as the
cross-sectional area of flow divided by the surface width of flow (A/B). Using either the hydraulic depth or
flow rate for the stream, there are two commonly used forms of the above Equation 10, shown here in
Equation 11:
𝑉𝑉 11
𝐹𝐹𝑟𝑟 =
�𝑔𝑔 �𝐴𝐴�𝐵𝐵�
and 𝐹𝐹𝑟𝑟 = 𝑄𝑄 �𝐵𝐵�(𝑔𝑔𝑔𝑔3 )
where
𝐴𝐴 = Cross-sectional area of flow normal to direction of flow (m2)
𝐵𝐵 = Width of flow at surface (m)
𝑉𝑉 Velocity of flow (m/s)
𝑄𝑄 = Flow rate (m3/s)
These equations are valid for irregular channel/stream conditions.
Froude’s Number indicates the state of flow. Critical flow occurs when Fr = 1. When Fr is greater than 1, the
flow is supercritical, and when it is less than 1 the flow is subcritical. When Fr is close to 1 (> 0.9 and < 1.1)
the flow state is unstable and waves appear on the water surface.
It is recommended that design achieves a Froude Number less than 0.9 (subcritical flow). For an example of
application of Froude’s Number see Step 7 in Section 2.3.6 – Hydraulic Jump.
For a description of the subcritical and supercritical flows, see Specific energy earlier in this section.

2.3.6 Hydraulic Jump
A hydraulic jump is an abrupt rise of the water surface which occurs when flow changes from supercritical to
subcritical in response to a downstream control condition causing a higher tailwater level. The rapid (and
turbulent) rise of water dissipates energy. The turbulence persists for some length after the rise before
settling.
The rise and trailing turbulence is called the ‘roller’ and is shown between points 1 and 2 on Figure 2.14.
The depth y1, before the hydraulic jump, is called the initial depth, and the depth y2, after the jump, is called
the sequent depth.
The sequent depth is always lower than the high stage alternate depth, y2, due to the loss of energy, ΔE, in
the turbulence of the hydraulic jump. Due to this loss of energy, the solution to hydraulic jump problems
cannot be found using the Specific Energy curve alone. The principle of momentum must also be used to
develop a Specific Force or Force Momentum curve as shown in Figure 2.14.
The detailed analysis of hydraulic jumps is difficult and complex and full analysis of jumps should therefore
be undertaken by an experienced hydraulic specialist.
However, several models based on accepted theory, experimentation, laboratory testing in rectangular
channels and observation, have been developed to approximate key parameters of hydraulic jumps in
irregular open channels.
These simple methods are considered reasonable for analysing undular and weak hydraulic jumps (see
Jump strength, below) and are suitable for use by designers.
Specific force
As mentioned above, the solution to hydraulic jump problems cannot be found using the Specific Energy
curve alone. The hydraulic jump involves internal energy losses which cannot be simply evaluated using the
energy equation and therefore the use of the momentum principle is required.
In applying the momentum principle to a short horizontal reach of a prismatic channel, the Specific Force (F)
can be defined with Equation 12:
𝑄𝑄2 12
𝐹𝐹 = + 𝑦𝑦�
𝑔𝑔𝑔𝑔
where
𝑄𝑄 = Discharge (m3/s)
𝐴𝐴 = Area of cross-section (m2)
𝑦𝑦� = Depth to the centre of gravity to cross-section (m)

Figure 2.14: Analysis curves and the hydraulic jump
Source: Chow (cited in DTMR 2010).
The first term is the momentum of the flow passing through the channel section per unit time, per unit weight
of water. The second is the force per unit weight of water.
An examination of Figure 2.14 shows that while water flows at a depth (y1) for a given discharge (Q) in any
channel, there will always be another depth where the sum of the force due to velocity plus the hydrostatic
pressure for both depths will be the same.
This means that the specific forces of sections 1 and 2 are equal, provided that the external forces and the
weight effect of water can be ignored. This generally can be applied to horizontal channels with small bed
slopes (So ≤ 10%).
With respect to the rule that there are always two depths, the exception is at critical depth (points C’ & C)
where energy is at a minimum and there is only one depth of flow.
Jump strength
Hydraulic jumps can only form if the upstream flow is supercritical (i.e. has a Froude Number greater than 1).
As Froude’s Number increases, the strength of the jump also increases.
Table 2.4 lists the different types of jump and their defining characteristics. This table has been derived from
research studies undertaken in 1955 by the U.S. Bureau of Reclamation (cited in DTMR 2010).
Table 2.4: Jump types and strength
Jump type Fr Energy dissipated Characteristics

Undular 1.0–1.7 < 5% Very weak jump; standing waves
Weak 1.7–2.5 5–15% Weak jump; small roller; smooth rise
Oscillating 2.5–4.5 15–45% Unstable – avoid
Steady 4.5–9.0 45–70% Stable, well defined jump; best design range
Strong > 9.0 70–85% Choppy; very turbulent; intermittent

Sequent depth in horizontal channels (SO ≤ 10%)
The parameters of the supercritical, upstream flow before the hydraulic jump are known/can be determined.
These parameters are flow velocity, flow depth and Froude Number.
Using these parameters, the alternate or sequent depth (y2) can be determined, considering:
• energy and momentum principles
• using the initial flow velocity as V1
• using the initial flow depth as y1
• experiments using a rectangular channel.
Rectangular channels
Equation 13 was developed to calculate the sequent depth (y2) in a rectangular channel:
𝑦𝑦2 1 13
= ( �1 + 8 𝐹𝐹𝑟𝑟2 − 1)
𝑦𝑦1 2
This can be expressed as:
𝑦𝑦1
𝑦𝑦2 = ��1 + 8𝐹𝐹𝑟𝑟2 − 1�
2
and Froude’s Number (Equation 10)
𝑉𝑉1
𝐹𝐹𝑟𝑟 = � (rectangular section)
�𝑔𝑔 𝑦𝑦1
This relationship is shown graphically in Figure 2.15.

Figure 2.15: Relation between Froude Number and depths of flow in a hydraulic jump
Trapezoidal channels
The determination of sequent depth (y2) in a trapezoidal channel of reasonably small bed slope (So ≤ 10%),
requires an iterative solution. However, when applying the Specific Energy and Specific Force equations, it
is possible to arrange these equations into a dimensionless form, allowing solution by means of a table.
Figure 2.16 assists the designer with the calculation of the sequent depth (depth after the jump). This has
been derived from the Handbook of Hydraulics, 7th Ed, 1996 by Brater et al. (cited in DTMR 2010)

Figure 2.16: Sequent depth for trapezoidal channels (So ≤ 10%)
y1 = Initial depth (m)

y2 = Sequent depth (m)
z = Side slope
1
Esj = Speciﬁc energy before jump (m)
z b= Bottom of channel (m)
J= ratio between sequent depth and Speciﬁc Energy at initial
depth
b y2 = J x Esj
Source: Brater et al. (cited in DTMR 2010).
Sequent depth sloping channels (So > 10%)
While the effect of the weight of water in the jump is negligible for horizontal channels (So ≤ 10%), it is
essential to consider it in the analysis of hydraulic jumps in sloping channels (So > 10%). For channels > 10
% advice from specialist hydraulic engineers should be obtained.

Location of jump
The location of the jump is dependent primarily on the momentum of flow before and after the jump and can
occur either before or after the ‘trigger’ situation that causes the initial supercritical flow to ‘slow’ to subcritical
flow.
Figure 2.17 shows a channel having a ‘break’ in the bed slope (steep to mild slope) and two possible
locations for the jump. If the sequent depth y2 is greater than the alternate depth y′1, (see Section 2.3.5 –
Energy Principles) the jump will occur in the steep region. If the sequent depth y2 is lowered to less than
approximately the alternate depth y′1, the jump will start to move into the mild channel.
Generally, for undular and weak jumps and where channels have a bed slope less than 10%, the jump
occurs close to the trigger point.
Figure 2.17: Location of a hydraulic jump
Length of jump
The length of a hydraulic jump is defined as the distance measured from the front face of the jump to a point
on the surface immediately downstream of the roller and is difficult to model and predict. Two methods to
determine the length of jump are presented here.
Method 1
Observation and measurements have shown that a good approximation of jump length is five times the
sequent depth (y2). In most cases, the jump length is between four and six times y2.
Method 2
The length of the jump can be determined from a relationship based on experimental data prepared by
Peterka (1964) for the US Bureau of Reclamation, as shown in Figure 2.18. This curve was developed
primarily for jumps in rectangular channels but may also be applied to approximate undular and weak jumps
formed in trapezoidal channels.
The length determined by either method should be rounded up to the nearest metre.
As the length of jump indicates the extent of turbulence, it is therefore the minimum length of channel that
should be armoured against scour/erosion. When determining the length of jumps in open channels for
armouring purposes, Method 1 is recommended as it gives a more conservative answer. Method two should
only be used when in constrained situations.

Application of hydraulic jumps
There are many practical applications for the hydraulic jump. The most practical of these applications, for
the purpose of road design, is its use as an energy dissipator to prevent scouring downstream of a hydraulic
structure such as a culvert or to prevent scouring in an open channel.
2.3.7 Hydraulic Drop
The converse of the hydraulic jump, the hydraulic drop, can also occur due to an abrupt change in channel
slope or cross-section. It can frequently occur near the entrance to a culvert flowing under inlet control. An
explanation of culverts flowing under inlet control is given in Section 3.8.3 – Culverts Inlet Control.
Figure 2.18: Length of hydraulic jump in terms of sequent depth (y2) in horizontal channels
2.4 Erosive Velocities in Natural Streams
Natural streams may be part of an open channel design or subject to outlet velocities from a drainage
structure and therefore a key design consideration is ensuring that any design changes to the flow in the
natural stream do not create erosive situations in the stream.
Erosive velocities in natural stream beds are shown in Table 2.5 and have been derived from curves by Neill
(cited in DTMR 2010).

Table 2.5: Erosive velocities in natural streams
Stream bed type Velocity (m/s)

Silt less than 0.3
Sand Fine less than 0.3
Coarse less than 0.3
Gravel 6 mm 0.6 to 0.9
25 mm 1.3 to 1.5
100 mm 2.0 to 3.0
Clay Soft 0.3 to 0.6
Stiff 1.0 to 1.2
Hard 1.5 to 2.0
Rocks 150 mm 2.5 to 3.0
300 mm 3.5 to 4.0
Source: DTMR (2010)
2.5 Backwater
Backwater is the accumulation of water in a stream. For further information, see AGRD Part 5 – Section
3.7.1.
2.5.1 Tidal Waters
Information on the impacts and design considerations of tidal waters and storm surge can be found in AGRD
Part 5 – Section 3.7.2.
2.5.2 Downstream Tributary
Information on the impacts and considerations to the water levels in downstream tributaries can be found in
AGRD Part 5 – Section 3.7.3.
2.6 Tailwater Levels
Tailwater level refers to the normal water level, for a given flow, in a channel immediately downstream of a
drainage structure. For a proposed culvert, it is the depth of flow at the culvert outlet, measured from the
water surface in the downstream channel to the invert of the culvert and the depth can be significant.
Tailwater levels are an important control on the hydraulic performance of road drainage structures, therefore
the estimation of tailwater levels is required for all hydraulic analyses for bridges and culverts.
Typically, tailwater level is defined by the downstream channel properties. If the downstream channel and
flow is uniform, the tailwater level may be determined using Manning’s Equation (see Section 2.3.3 –
Fundamental Equations). If the downstream flow is non-uniform, or if it is being influenced by backwater, a
more sophisticated analysis by a specialist hydraulic engineer is required.
2.6.1 Tailwater Effects
As previously mentioned, tailwater is an important aspect required to determine the hydraulic operation and
performance of drainage structures

A low tailwater typically means the channel can drain the stormwater away fairly quickly and easily. This is
common where drainage channels are less defined with low banks, allowing the stormwater to spread
widely. Drainage structures with low tailwater levels will most likely be controlled by inlet conditions and can
have high outlet velocities; therefore erosion can be an issue. A high tailwater level typically means the
channel does not drain the stormwater away quickly. This is common where drainage channels are
reasonably defined with mid-to-high banks, allowing the channel to contain most stormwater flows. Drainage
structures with high tailwater will most likely be controlled by outlet conditions and can have low outlet
velocities, therefore high headwater levels (afflux) can be an issue.
2.6.2 Design Tailwater Levels
Typically, tailwater levels should be determined based on an assessment of channel flow (and flow depths),
for a given discharge, over a significant length of undisturbed downstream channel. This assessment is
usually commenced at a point well downstream of the proposed structure and calculated back/upstream
towards the proposed structure. The assessment should also include the effects of floods and tailwater
levels. This assessment is complex and hydraulic modelling software is often used.
For the design of most drainage structures within the scope of this Guide, the survey data gathered in
relation to creeks, streams, or channels are usually limited to the road environment (constrained by property
boundaries), unless an extended downstream survey was specifically requested.
Using available survey data and assuming the channel is regular in shape and steady, uniform flow
conditions can be expected, the tailwater level can be determined using Manning’s Equation as follows and:
• adopting a stream/channel cross-section just downstream of the proposed structure outlet, but
representative of the downstream channel in general
• using a series of calculations (Manning’s Equation (Equation 3) and Q = V x A), to develop a Stage-
Discharge curve for the channel
• using the Stage-Discharge curve and design discharge for the proposed structure, to determine flow
depth. This depth is adopted as the tailwater depth for the proposed structure analysis.
Where the design discharge exceeds the maximum capacity of the channel, the channel will overflow and
the design discharge (flow) will spread. Survey data should be reviewed to see if there are any high banks
or if the cross-section can be extended to allow for increased flow depth. Where possible the Stage-
Discharge curve may be extended to incorporate the additional waterway area. If the extended cross-section
cannot contain the design discharge and the spread of flow is significant (i.e. there would be little increase in
flow depth based on design discharge), then the tailwater level could be set at the level of maximum capacity
for the defined channel. Also, if there is not much of an outlet channel (flat terrain/low banks) and the design
discharge easily overtops the banks and/or spread of flow is significant, then a low tailwater level should be
adopted. The survey data should be used to produce a wide flat cross-section and the above process
should be followed to determine this low tailwater level.
Furthermore, anecdotal information with respect to channel flows should be sourced from local residents.
Theoretical predictions (proposed tailwater level) should be checked against the anecdotal evidence to
establish whether they reasonably agree. If so, then the level of confidence in prediction is increased. If not,
then further checks of calculations and gathered information are required. All gathered information should
be recorded.
2.7 Open Channel Design
The primary aim when designing artificial channels is the determination of a channel cross-section that will
convey the design flow whilst maintaining velocities within acceptable limits.

2.7.1 Design Methodology
There are two artificial channel types:

• hard lined channels
• vegetated/bare-earth channels.
Each of these channel types has a specific design methodology to determine the channel cross-section and
these are detailed in the following sections.
General considerations
The following aspects should be considered/established before the design of the channel commences:
• channel velocity – velocities need to be considered for their potential for scouring the channel
• channel slope – generally, flow in channels is intermittent, and the channel must be constructed to allow
all stormwater to drain away (no ponding). Therefore the minimum channel slope (So) should not be less
than 0.25% for a hard lined channel and not be less than 0.5% for vegetated and bare-earth channels
• if a ‘self-cleaning’ channel is required, designers should see specific water-sensitive road design
guidelines for the design requirements
• channel side slopes – slopes should be based on the stability of the material the channel is to be
constructed in (channel is shaped before any lining is applied or grass grows). Typically, this will be in
the range one in one to one in four
• maintenance – safe access for maintenance equipment should be considered in addition to the ease of
maintenance operations based on the cross-section shape
• freeboard – the freeboard to be adopted for open channels is the greater of the following calculations (see
AGRD Part 5 – Section 4.7):
– 300 mmA
– 20% of the flow depth
– velocity headB of the flow.
Note:
– A: Where flooding of adjacent land and buildings does not represent a risk, the 300 mm requirement
can be reduced to 150 mm.
– B: See Section 2.3.5 – Energy Principles.
Hard lined channels
With hard lined or rigid boundary channels, the design procedure is relatively simple as potentially erosive
velocities are less important. Therefore, the cost of excavation and lining are primary factors in determining
the geometry of the channel.
Research and experimentation has found that the most efficient, hydraulic section is when the wetted
perimeter (P) is minimised for a given cross-sectional area (A). Typically, in open channels, this occurs
when the width of flow is twice the depth of flow. Therefore the shape of channel that gives the best
hydraulic efficiency (also known as best hydraulic section) occurs when the hydraulic radius is shown by
Equation 14:
𝑦𝑦 14
𝑅𝑅 = �2

Based on this requirement, the most efficient shape is a semi-circle as it has the least wetted perimeter for a
given area – R is always equal to y/2 irrespective of dimensions. However the cost and difficulty of building
in situ semi-circular channels prohibits their widespread use. Trapezoidal channels are more commonplace.
The most efficient trapezoidal shape is the half-hexagon.
For the design of hard lined channels:
Determine design discharge (Q), bed slope (So), Manning’s n value and channel side slope (X) (Figure 2.19).
Figure 2.19: Trapezoidal channel section
For best hydraulic section, use Equation 15:
𝑏𝑏𝑏𝑏 + 𝑋𝑋𝑦𝑦 2 𝑦𝑦 15
𝑅𝑅 = =
2 �𝑦𝑦 2 + (𝑋𝑋𝑋𝑋)2 + 𝑏𝑏 2
and resolve for b as a function of y.
Combining Equation 1, Manning’s Equation (Equation 3), and Equation 17 gives Equation 16:
1�
𝐴𝐴 𝑅𝑅
2�
3 𝑆𝑆𝑓𝑓2 16
𝑄𝑄 =
𝑛𝑛
where
𝐴𝐴 = 𝑏𝑏𝑏𝑏 + 𝑋𝑋𝑦𝑦 2 17
𝑅𝑅 = y/2 (Equation 14)
𝑆𝑆𝑓𝑓 = Friction slope (𝑆𝑆𝑜𝑜 is used to approximate 𝑆𝑆𝑓𝑓 )
𝑛𝑛 = Manning’s 𝑛𝑛 value
Substitute b with function for b (derived from Equation 15) and solve for y.
Calculate b by substituting y into the function for b.
Channel geometry can now be determined and freeboard added.
Calculate the Froude Number for the channel section/flow and together with velocity and depth of flow; check
that conditions are acceptable where the designed channel discharges into an existing channel.
A worked example of this method is shown in Section 2.16.4 – Example 4: Manning’s Equation.

Vegetated or bare-earth channels
Vegetated/bare-earth channels are erodible boundary channels and the design procedure is relatively
simple, as it is based on the method of maximum permissible velocity.
This method assumes that the channel will remain stable provided that the average velocity of the design
flow is below a determined threshold value.
A key concern for vegetated channels is the consequence when the grass cover cannot be maintained, such
as during drought or after a bushfire. If there is a reasonable risk of this occurring and channel scour is likely
and not desirable, the design should be undertaken assuming bare-earth design values.
When designing vegetated/bare-earth channels the following must be considered:

• the material in which the channel is to be constructed
• a suitable grass species for the channel (where applicable)
• an appropriate Manning’s n value.
A suitable grass species for a stormwater channel should have a relatively short blade
(< 50 mm) and be:
• quick to establish
• self-repairing
• able to survive short durations of inundation
• able to withstand proposed design velocities
• native to the area.
The species should be chosen in consultation with a suitably qualified landscape architect or horticulturalist.
The reason for choosing a short blade length grass species is primarily that longer blade length species can
reduce flow velocity by increased flow resistance. This increases flow depth (surcharge) and ultimately can
cause the channel to break its banks. Swale drains are special vegetated channels, where the flow velocity
is lowered to maximise stormwater treatment. In swale drains, it is common for the flow depth to be less
than the height of grass and care is required when selecting the Manning’s n, see Section 2.8 – Grassed
Channels.
Channels could also have a Water Sensitive Road Design (WSRD) function, so a longer grass blade length
may be preferred.
Table 2.6 gives recommended maximum velocities for use in the design of consolidated, bare-earth channels
and vegetated channels.

Table 2.6: Recommended maximum velocities (design) for consolidated, bare channels and vegetated
channels
Permissible velocities (m/s) when fraction of stable surface cover(1) is:

Channel
gradient (%)
0.02 0.5 0.7 1.0
Erosion resistant soils (e.g. krasnozems and red earths)
0.5 0.8 1.8 2.4 2.8
1 0.7 1.6 2.1 2.8
2 0.6 1.4 1.8 2.5
3 0.5 1.3 1.7 2.4
4 1.3 1.6 2.3
5 1.2 1.6 2.2
6 1.5 2.1
8 1.5 2.0
10 1.4 1.9
15 1.3 1.8
20 1.3 1.7
Easily eroded soils (e.g. black earths and fine surface texture – contrast soils)
0.5 0.6 1.3 1.6 2.3
1 0.5 1.2 1.5 2.1
2 0.5 1.1 1.4 1.9
3 0.4 1.0 1.3 1.8
4 1.0 1.2 1.7
5 0.9 1.2 1.6
6 1.1 1.6
8 1.1 1.5
10 1.1 1.5
15 1.0 1.4
20 0.9 1.3
1 Applies to surface consolidated, but not cultivated.
Notes: Assume the following species under average conditions will provide the fraction of cover indicated:
• Kikuyu, pangola and well maintained couch species – 1.0.
• Rhodes grass, and poorly maintained couch species – 0.7.
• Native species, tussock grasses – 0.5.
Source: Adapted from Department of Environment and Resource Management (2012).
The following method should be used for the design of vegetated/bare-earth channels:
1. Determine design discharge (Q), bed slope (So), Manning’s n value, channel side slope (X) (see
Figure 2.19) and the maximum recommended velocity (Vmax) from Table 2.6. For bare-earth channel
design, adopt 0% cover.
2. Using the basic equation Q=V x A, determine required waterway area.
𝑄𝑄
𝐴𝐴 = �𝑉𝑉
𝑚𝑚𝑚𝑚𝑚𝑚

3. Combining Equation 1 and Manning’s Equation (shown in Equation 16):
2� 1�
2
𝐴𝐴 𝑅𝑅 3 𝑆𝑆𝑜𝑜
𝑄𝑄 =
𝑛𝑛
Re-arrange and solve for wetted perimeter (P):
3�
5 1� 2
𝐴𝐴 �3 𝑆𝑆𝑜𝑜 2
𝑃𝑃 = � �
𝑄𝑄𝑛𝑛
4. Based on channel geometry, two expressions for A and P exist:
𝐴𝐴 = (𝑏𝑏 + 𝑦𝑦𝑦𝑦)𝑦𝑦 (from Equation 17)
and (from part of Equation 15):
𝑃𝑃 = 2𝑦𝑦 �𝑋𝑋 2 + 1 + 𝑏𝑏
Re-arrange expression for A to isolate b:
𝐴𝐴
𝑏𝑏 = − 𝑦𝑦𝑦𝑦
𝑦𝑦
and substitute back into expression for P:
𝐴𝐴
𝑃𝑃 = 2𝑦𝑦�𝑋𝑋 2 + 1 + − 𝑦𝑦𝑦𝑦
𝑦𝑦
5. Calculate depth of flow (y). Solving for y requires rearrangement into the form of a quadratic equation.
𝑎𝑎 𝑦𝑦 2 + 𝑏𝑏𝑏𝑏 + 𝑐𝑐 = 0
The general solution of a quadratic equation is Equation 18:
−𝑏𝑏 ± √𝑏𝑏 2 − 4𝑎𝑎𝑎𝑎 18

𝑦𝑦 =
2𝑎𝑎
solving a quadratic equation gives 2 roots. Substitute roots (y) into:
𝐴𝐴
𝑦𝑦
and solve for b. Dismiss any negative solution and adopt the positive (real) solution for y.
6. Channel geometry can now be determined with freeboard added.
If there are no real roots to the equation, this means that the velocity will never reach the maximum
permissible velocity for the given discharge (Q) and roughness (n), regardless of channel dimensions. In
this case, use the ‘best hydraulic section’ design methodology as detailed in the Section 2.7.1 – Design
Methodology, for the design of the vegetated channel. However, also check that the average velocity of
the resultant channel section does not exceed the maximum recommended velocity (see Table 2.6).
7. Calculate the Froude Number for the channel section/flow and together with velocity and depth of flow;
check that conditions are acceptable where the designed channel discharges into an existing channel.

A worked example of this method is shown in Section 2.16.5 – Example 5: Vegetated/Erodible Boundary
Channel Design.
2.7.2 Channel Transitions
Changes from one channel cross-section to another should be smoothly transitioned as energy losses occur
with the changes.
It is recommended that maximum contraction rates of about one on one, and maximum expansion rates of
about 1 on 4 be adopted as shown in Figure 2.20.
Figure 2.20: Maximum rates of contraction and expansion
Transition head losses can be calculated using Equation 19:
𝑉𝑉12 𝑉𝑉22 19
ℎ𝑡𝑡 = 𝐶𝐶𝑢𝑢 � − �
2𝑔𝑔 2𝑔𝑔
where
ℎ𝑡𝑡 = Transition head loss (m)
𝐶𝐶𝑢𝑢 = Transition energy loss coefficient (dimensionless)
𝑉𝑉1 = Average flow velocity upstream of transition (m/s)
𝑉𝑉2 = Average flow velocity downstream of transition (m/s)
Transition energy loss (contraction and expansion) coefficients (Cu) are shown in Table 2.7 for use in this
formula. Types of transition are shown in Figure 2.21.
It should be noted that the more abrupt the transition, the greater the energy loss.
The difference in velocity head is always to be taken as positive even if mathematically it is negative. In
other words a change in section will always give a head loss.

Table 2.7: Channel transition energy loss coefficients

Transition type Contraction Expansion
(Ref Figure 2.21) coefficient Cu coefficient Cu
Warped 0.10 0.20
Cylindrical quadrant 0.15 0.25
Wedge 0.30 0.50
Straight line 0.30 0.50
Square end 0.30 0.75
Source: Thompson and Kilgore (2006).
Figure 2.21: Channel transition types
Source: Thompson and Kilgore (2006).
2.7.3 Energy Losses in Channel Bends
The Department of Natural Resources and Water, Queensland, DNRW (2007), based on work by Mockmore,
gives an equation to calculate the loss in a bend in an open channel. Equation 20, the equation for channel
bend head loss for bends with changes in direction between 90 to 180 degrees, is:
2𝑏𝑏 𝑉𝑉 2 20
ℎ𝑏𝑏 = � ��
𝑅𝑅𝑐𝑐 2𝑔𝑔
where
ℎ𝑏𝑏 = Channel bend head loss (m)
𝑏𝑏 = Channel width (m)
𝑉𝑉 = Average flow velocity (m/s)
𝑅𝑅𝑐𝑐 = Centreline radius of bend (m)

Results from using this equation should be considered as providing a conservative estimate of head loss.
Linear interpolation is recommended for bends between 0° and 90°.
2.7.4 Superelevation in Channel Bends
Water flowing around a bend in an open channel may show a superelevation with a rise in water level on the
outside of the bend and a corresponding dip in water levels on the inside of the bend. The maximum
difference in these water levels is termed the superelevation (not to be confused with the percentage
superelevation in geometric road design).
The superelevation should be considered when a flood level is reported at a bend (the exact location
needing to be defined), or where it is essential that the design flow does not exit from a man-made channel
at the outside edge of a bend.
Superelevation can be estimated using Equation 21:
𝑅𝑅𝑜𝑜 𝑉𝑉 2 21
ℎ𝑠𝑠𝑠𝑠𝑠𝑠 = 2 log 𝑒𝑒 � ��
𝑅𝑅𝑖𝑖 2𝑔𝑔
where
ℎ𝑠𝑠𝑠𝑠𝑠𝑠 = Superelevation of the water surface across the channel (level difference) (m)
𝑉𝑉 = Average flow velocity (m/s)
𝑅𝑅𝑜𝑜 = Outside radius of bend (m)
𝑅𝑅𝑖𝑖 = Inner radius of bend
2.8 Grassed Channels
This section provides the guidelines for channels that will use grass as a protective lining. Designers should
see specialists for further guidelines when an open channel (such as a swale) is to also be used for
environmental reasons (e.g. Water Sensitive Road Design).
2.8.1 Normal Grassed Channels
The cross-section shapes generally used for grassed channels are trapezoidal (most common), parabolic, or
triangular. Flat side slopes should be used where space permits to assist in minimising maintenance costs.
A desirable maximum steepness of 1:6 and an absolute maximum steepness of 1:4 should be used for the
design of the side slopes.
The maximum permissible velocity of flow in a grass channel is the velocity which, if maintained for a
reasonable length of time, will not cause erosion. Recommended velocities for different vegetal covers,
channel slopes, and soil conditions are given in Table 2.6.
The minimum non-silting velocity is uncertain. However, as a general rule the channel velocity should not fall
below 0.5 m/s.

To function well, grassed channels normally have longitudinal slopes of 0.2% to 0.6%. Where natural
topography is steeper than the desired slope of the channel, channel drops (see Section 2.10 – Channel
Drops), or drop structures should be considered.
Sharp curves which affect the flow in the channel should not be used. In general, curves should be such that
the minimum radius at the centre line is equal to about three times the width of the top of the design flow.
Research by the United States Soil Conservation Service (USSCS) has found that, for a particular channel
lined with a certain species of grass that has a definite length, the value of n varies with the product of the
mean velocity of flow V (m/s) and the hydraulic radius R (m). The relationship is a characteristic of the
vegetation and is the same for most common sections used.
The vegetation characteristics were defined by five degrees of retardance (A to E) depending on the grass
species and length. A channel lined with a grass having a high retardance (A), such as 750 mm long
Rhodes grass, has its flow severely restricted by the grass, while any grass less than 50 mm long, having a
low retardance (E), has little effect on the flow. The relationship between grass species, length and
retardance is given in Table 2.3.
The original USSCS curves for degree of retardance D and E have been replaced in this Guide by curves
based on work by Kouwen et al. (1981) who found that the original curves were not acceptable for short, stiff
vegetation on slopes smaller than 5%.
The use of erosion control cut-off walls at regular intervals in a grassed channel is desirable. Such cut-off
walls will safeguard a channel from serious erosion prior to the grass becoming established or immediately
after a fire. These walls are usually of unreinforced concrete or rock masonry, approximately 200 mm wide
and from 450 mm to 600 mm deep, extending across the entire bed of the channel and conforming to its
shape. Where the banks of the channel may erode, these walls should extend up the banks to the design
water surface.
The designer needs to consider the impact of these walls on the conveyance of the channel and hence flood
levels.
2.8.2 Reinforced Grassed Channels
The use of reinforced grass (or turf) in channel design provides the benefits of significantly increased
allowable flow velocities, however, these products also come with disadvantages, including:
• possible public safety problems resulting from high flow velocities
• maintenance personnel may not be aware of the existence of the grass reinforcing and normal grading
operations can result in costly damage to the reinforcing mats
• damage by grass fires
• degradation of habitat of ground-dwelling wildlife in a bushland environment
• die-off of grass owing to high sediment deposition (reinforced grass or turf can only be installed once the
contributing catchment is stable)
• loss of topsoil in the future when the channel is removed, rebuilt, reconditioned or relocated as it is likely
that the reinforced grass and its enclosed topsoil will need to be removed from the site, thus preventing
the reuse of the topsoil.
The latter point can be a major concern in areas where there is limited topsoil.
There are several proprietary types of reinforced grass which are termed ‘a form of geotextile protection’ in
the U.S.A. Some tests of relatively small flows indicate resistance to erosion for flows up to 5 m/s. Caution
should be exercised before adopting design velocities of this magnitude; the duration and depths of flow on
the actual job site will also need to be considered.

The reinforced grass may be considered an intermediate type of lining between conventional grass and
hard/impermeable linings. Values of Manning’s n for these linings can be found in Table 2.1.
The following provides a comparison on n values for different channel treatments:

• from Table 2.1 – Section 3 – Culverts (b) (ii) n value range 0.025 – 0.030 say use n = 0.028 for
conventional grass
• from Table 2.1 – Section 2 – Open Channels (a) n = 0.012 for hard/impermeable
• average for reinforced grass n = 0.020.
2.9 Channels Lined with Hard Facings
2.9.1 General
These channels are typically lined with concrete, riprap or rock mattresses/gabions and are highly resistant
to erosion. Design velocities may be as high as 5 m/s, much higher than those for grassed channels.
Smaller cross-sections than those for grassed channels are possible to take the same discharges, as side
slopes of 1:1 are practical, although it is common to use 1:1.5. Channels with hard facings should be
considered where the width is limited within the road environment. If the channel is located within the clear
zone of the road, the designer should see AGRD Part 6 (Austroads 2010c), for guidance on appropriate
treatments for errant vehicles.
Other considerations for channels with hard faced linings include the provision of:
• contraction and expansion joints to minimise the risk of cracking and seepage and potential undermining
• step irons (or similar device), to enable safe access, where the channel side slope is steeper than 1:2 and
the channel depth exceeds 0.9 m
• pressure relief weep holes in impermeable linings both within the channel invert and within the channel
side slopes. Weep holes should be 90 mm diameter at 1.8 m maximum centres (depends on detailed
structural design requirements)
• lateral protection against surface flows undermining the side slopes. A hard faced strip having a
minimum width of 0.45 m on both sides at the top of the channel is recommended
• cut-off walls at the ends of the bed and side linings. These should have a minimum depth of 0.5 m
• possible requirements for an energy dissipator at the end of the lined channel section to manage the
expected high flow velocities.
2.9.2 Riprap and Rock Filled Wire Mattresses/Gabions
Riprap is a conventional protection used to prevent channel bed and bank damage upstream and
downstream of hydraulic structures, at bends, at bridges and in other channel areas where there is potential
for erosion.
The thickness of the riprap should be not less than 1.5 times the largest diameter of rock. Figure 2.22 shows
the relationship between bed velocity and rock diameter (Kouwen et al. 1981). When using Figure 2.22, bed
velocity can be taken as approximately 0.7 times the average channel velocity. The designer also needs to
determine the velocity on the bank side slopes, particularly on bends, as this is critical for the stability of the
side slopes.

The flow velocity on side slopes depends on the side slope angle but typically the shear stress is slightly less
on the side slope than that on the channel base. The flow velocity in bends is dependent on the radius of the
bend, the tighter the radius the higher the flow velocity. It is common for the side slope face flow velocity
(parallel to the channel invert) in a tight bend to be 1.3 to 1.5 times the bottom flow velocity in a straight
channel. In channels that can be expected to experience high debris loads, the designer also needs to
consider the mechanical prying force effects that could be generated by say a tree branch on the rock riprap.
Special care is also required for rock riprap design where there is a significant component of flow down the
face of the side slope (perpendicular to the channel invert), see Section 4.5.3 – Floodways with other than
Grassed Batters and the Type 5 floodway (Figure 4.6). For these special cases the designer should seek
the advice of a specialist in the design of rock riprap protection works.
Rock filled wire mattresses or gabions may also be used to line the channel bank or bed, as shown in
Figure 2.23. Smaller sized rocks can be used, because the wire basket surrounding the rock in the mattress,
or gabion, tends to make the mass act as a unit while retaining flexibility. Besides providing protection
against scour, rock filled gabions are useful as drops with either vertical or stepped faces. At drops, the
gabions should be keyed into both banks, and a downstream cut-off wall should be considered. Design and
construction of gabion protection should be in accordance with manufacturer’s
recommendations/specifications.
It is also important to note that the flow velocity on side slopes depends on the side slope angle but typically
the shear stress is slightly less on the side slope than on the channel base. The flow velocity in bends is
dependent on the radius of the bend, the tighter the radius flow velocity. It is common for the side slope face
flow velocity (parallel to the channel invert) in a tight bend to be 1.3 to 1.5 times the bottom flow velocity in a
straight channel. In channels that can be expected to experience high debris loads, the designer also needs
to consider the mechanical prying force effects that could be generated by, say a tree branch, on the rock
riprap. Special care is also required for rock riprap design where there is a significant component of flow
down the face of the side slope (perpendicular to the channel invert), see Section 4.5.3 and the Type 5
Floodway (shown in Figure 4.5). For these special cases, the designer should seek the advice of a specialist
in the design of rock riprap protection works.

Figure 2.22: Relationship between bed velocity and rock diameter

Figure 2.23: Channel lined with rock filled wire mattresses
Open channel lined with riprap
When using riprap or rock filled wire mattresses over erodible underlying materials an appropriate geotextile
fabric should be placed under and behind the riprap, rock filled wire mattresses, or gabions to prevent
erosion and water piping through the lining treatment.
Some of the in-service problems associated with rock filled wire mattresses and gabions are listed below:
• Vegetation problems: In some locations, rock filled wire baskets have been known to quickly vegetate
with both native and non-native weeds and vines. These vines can then migrate into the surrounding
bushland causing problems to native trees.
• Erosion: When used on the outside of channel bends, rock filled mattresses and gabions effectively form
a ‘hydraulically’ smooth surface. This ‘smooth’ surface causes high flow velocities to occur immediately
adjacent to the wire baskets. These high flow velocities can then result in increased bed erosion at the
base of the gabion wall (causing foundation problems), or bank erosion immediately downstream of the
gabion protected bank.
• Vegetation cover: In some installations long-term stability is best achieved by having the gabion and
mattress structures covered with vegetation. When this form of bank stabilisation is placed under a wide-
span bridge, light and rainfall restrictions can limit or prevent vegetation cover, thus resulting in a
reduction of the expected life span of the structure.
• Damage by sediment: In channels that transport large quantities of coarse bed-load sediment, the
expected design life of gabion and rock mattress structures can be reduced due to damage to both the
plastic and galvanised wire coatings.
Site specific designs need to be completed for every gabion or mattress application using design
recommendations from the suppliers of proprietary products.
2.9.3 Concrete Lined Channels
Concrete linings must be protected from hydrostatic uplift forces which can be created by a high water table
after the flow in the channel has dropped. A free draining perforated subsoil drainage pipe may be required
in large channels. In small channels, weepholes and porous backfill may be adequate.

Where undermining along the top of the lining is a risk, a cut-off lip or wall should be provided. In long
lengths of lining, expansion joints are required.
Care should be taken in the design of channels required to carry supercritical flow to ensure that:
• Curves are avoided if possible.
• Channels are free of obstructions which may cause oscillatory waves to extend down the remaining
length of the channels and into culverts.
• Pipes entering a channel are cut off flush with the lining.
• Expansion joints are dowelled to prevent differential movement between segments of the lining, which
could otherwise allow high velocity flow to enter the joints and cause uplift.
• Unintended hydraulic jumps do not form.
Channels designed for supercritical flow require a smooth surface which is provided by concrete lining. The
thickness of lining and type of reinforcement required should be determined in consultation with the road
agency or a specialist structural engineer.
Where channels are large enough, they should generally be accessible to vehicles for maintenance in which
case the lining of the base of the channel should be designed for the appropriate vehicular loading and soil
conditions.
2.10 Channel Drops
The use of channel drops, waterfalls or scour checks may permit flatter channel slopes to be employed. For
safety reasons, it is generally desirable to use several low-head drops rather than fewer higher drops.
However, it is often more difficult to control the hydraulics of low drops.
While vertical drops should generally be avoided to minimise safety concerns and erosion control problems,
they may have to be used in some circumstances. Also, depending on tailwater conditions, supercritical flow
often occurs on the apron of the drop and as a result, either a hydraulic jump or submerged jet will usually
form downstream.
The use and design of channel drops should be referred to a specialist hydraulic engineer.
2.10.1 Check Dams
Another method to reduce the velocity of water flowing in a channel is the use of check dams placed across
the channel. A check dam is constructed across the channel, using a permeable material, such as rocks, to
reduce the velocity of the water. They allow the water to flow through the structure but at a slower rate than
the flow in the channel.
2.11 Baffle Chutes
Baffle chutes (Figure 2.24) provide a satisfactory method of dissipating the energy of flow in a channel where
the slope is steep and to be effective they require no tailwater. The use and design of baffle chutes should
be referred to a specialist hydraulic engineer.

Figure 2.24: Diagram of a baffle chute
2.12 Edge Drains and Median Drains
2.12.1 Design of Edge and Median Drainage
General
It is preferable to allow the pavement run-off to discharge across the shoulder and into the channel but
where the batter slope is steep due to restricted width and erosion of the slope is likely, a kerb should be
provided. However, on high-speed carriageways, kerbs or steep batter slopes should be avoided for safety
reasons.
The main limitation on median drains relates to safe slopes for errant vehicles. Median drains are usually flat
bottomed and 2.0 m to 2.5 m wide at the bottom to accommodate maintenance machinery. For very wide
medians it is desirable from a road safety perspective to have batter slopes of about 1:10 (to cater for trucks)
and grated pits and underground pipes. See AGRD Part 6 (Austroads 2010c) for guidance on driveable
batter slopes and safety barriers.
However, the side slope of 1:10 or flatter severely restricts the capacity of such drains unless the median is
very wide. It may therefore be necessary in many instances to provide a steeper batter slope up to 1:6
(satisfactory for cars) or greater in which case a safety barrier may be required. Reference should be made
to AGRD Part 6 (Austroads 2010c) for guidance on driveable batter slopes and safety barriers.
Pavement run-off
The maximum velocities allowed in grass channels so as to prevent erosion are given in Table 2.6 and the
minimum velocity to prevent silting as a general rule should not fall below 0.5 m/s.

Channel treatment
Where a grassed channel is the preferred treatment, the channel must have sufficient hydraulic capacity to
prevent spread of the water onto the through traffic lanes of the pavement. Also the bottom of the channel
should desirably be at a lower elevation than the pavement sub base in cuts, otherwise subsoil drains may
be necessary to protect the pavement from excessive moisture. Grassed channels are further discussed in
Section 2.8 – Grassed Channels.
In very narrow and steep medians, concrete drains may be necessary to provide sufficient capacity. The
hydraulic design of concrete drains is outlined in Section 2.9 – Channels Lined with Hard Facings.
In very restricted areas, special drainage inlets and conduits may have to be provided.
Inlets
The inlets used to intercept flow in the median should desirably be set flush with the ground so as not to be
hazardous to mowing equipment and vehicles that run off the road. The bars of the grated pits should be set
parallel to the flow to improve hydraulic efficiency. A concrete apron should be used around the inlet to
accelerate the flow into it and to prevent the grate from being overgrown by vegetation.
If a slightly raised grated inlet with the top above the natural surface is adopted to minimise debris collection,
concrete aprons should still be provided to improve conspicuity of the grate for maintenance personnel
undertaking mowing.
Complete interception of all median flow at each inlet gives the most efficient design. For drains on grades,
a levee across the median just downstream of the inlet creates a sag condition and ensures complete
interception (see Figure 2.25).
The levee should be of just sufficient height to maintain the design head over the grate, and have slopes of
1:6 or flatter (1:20 is desirable) for the safety of out-of-control vehicles. The height of the levee is usually that
of the 10 year average recurrence interval (ARI) flood. Furthermore, designers need to also ensure that any
overtopping of the levee will not force stormwater onto the road surface. Where this cannot be achieved, the
road geometrics may need to be reviewed and modified.
The design procedure for inlets is given in Section 2.15 – Catch Drains and Catch Banks. The allowable
depth of ponding is determined as the depth to contain the flow within the median.

Figure 2.25: Median inlets with levees

2.13 Table Drains and Table Drain Blocks
An important design consideration may arise, where new table drains join existing table drains. The impacts
of discharge from the additional flow into the existing drain must be considered in the design of the new
drain. The shape and grade of adjoining table drains should also match. Where the shape and grade do not
match, a suitable transition or other mitigating treatment must be designed to ensure that scour does not
occur and/or that stormwater does not flow out onto the road surface.
2.13.1 Table Drains
Table drains (Figure 2.26) are located along the outer edge of the shoulder in cuts, and beside shallow
raised carriageways on fill. They collect water from the pavement, shoulders and cut batters and convey it to
a suitable turnout, watercourse or culvert. See AGRD Part 3 (Austroads 2010b) for details concerning their
location and use of table drains in road design.
Flat-bottomed table drains are the preferred type or shape, the bottom being wide enough to accommodate
maintenance machinery. Figure 2.27 shows the general location and shape details. The base of the drain is
not flat as the name suggests, but is to be sloped away from the carriageway (at least 3%). These drains:
• spread the flow, reducing depth and velocity (reduced scour potential)
• keep flow and any ponding away from the embankment
• allow easier access for maintenance
• allow safer passage through the drain for errant vehicles
• may provide a source of borrow material for some projects.
Figure 2.26: Typical table drains
The use of ‘V’ drains is to be limited and generally confined to situations where the width available for a drain
is constrained.
Determination of depth and velocity of flow within the table drain can be undertaken by combining Equation 1
and Manning’s Equation (shown in Equation 16).

Figure 2.27: Table drain location and shape
Source: Adapted from DTMR (2010).
Side slopes should not exceed 6:1. Where this slope cannot be achieved designers should see AGRD Part
6 (Austroads 2010c), for guidance on drain profiles that are not hazardous to errant vehicles or guidance on
the need for a safety barrier.
The design water level in the drain should desirably be set below the subgrade level of the road pavement,
particularly in flat country or in high water conditions. It is desirable that the table drain is designed to limit
pavement saturation and scour, and that the water depth limits the encroachment of water onto the shoulder
and carriageway.
Where the material between the table drain and the road pavement is impermeable, the table drain may flow
up to a level which provides 150 mm freeboard against overtopping, or 150 mm below the level of the outer
edge of the shoulder, provided that this does not flood subsurface drain outlets.
The invert of the table drain should be at least 150 mm below the bottom of the pavement (i.e. below
subgrade level), and deeper where subsurface drains discharge into the table drain. The depth of the table
drain will depend upon the design capacity required to safely convey the storm water and the depth of the
subsurface drain outlet.
The slope of grassed table drains (channels) depends on the terrain and the type of soil. The minimum
design slope should be such that ponding and siltation is avoided and the maximum slope should be
selected to avoid soil erosion. However, this balance is often difficult to achieve in practice as most table
drains are graded to follow road grade lines and often do not achieve the most desirable slope.
Nevertheless, the following general guide is provided to assist road designers’ understanding of desirable
slopes:
• To avoid ponding and siltation the desirable minimum slope should be in the range 0.4% to 0.5%.
• To avoid erosion the slope should not exceed 0.6%, depending on the soil.
• Drains steeper than 1% may need scour protection, depending on the erodibility of the soil (see AGRD
Part 5 – Section 3.6) and the vegetative cover.
• A slope of 0.2% may be necessary in very flat terrain.
• A slope of 0.3% may be regarded as the minimum practical slope for construction.
It is important to understand that the slope of the table drain is tied to the grade of the road and therefore,
there are situations where the slope of the table drain is less than the minimum. Examples of this are about
the apex of a crest vertical curve and on flat grades. In these situations, the table drain will have to be
graded independent of the road control line in order to achieve minimum slope.

Where a table drain has been widened as a source of borrow material, it should be graded to a suitable
outlet so that water does not pond against the road formation and weaken the subgrade (Underwood 1995).
Table drains can also function as swales and this requires that they also conform to the principles outlined in
AGRD Part 5 – Section 3.5.
Extreme care is required in the design of drains in very erodible soils or soils with highly dispersive subsoils,
as exposure of such material can have disastrous consequences. Where scour is likely because of the
nature of the material or because of the longitudinal grading, some type of protection of the drain invert is
required. This protection could take the form of rock lining or concrete. In areas known to have dispersive
subsoil, soil chemistry should be analysed to determine whether soil properties can be economically
improved to aid soil stability. The terminal treatment at the bottom of a steep table drain is also important in
order to dissipate the energy of the stormwater.
2.13.2 Table Drain Blocks
On vertical grades (excluding sags), cross drainage culverts can be used to take table drain flow from the
upstream side of the road formation to the downstream side. In order for these culverts (generally of small
diameter) to operate satisfactorily, sufficient head must be generated.
Table drain blocks are small earth embankments (typically 1 m in length) located within the table drain just
downstream of a culvert inlet. These blocks ‘dam’ and divert the flow into the culvert inlet. The design height
of the block should theoretically coincide with the depth of headwater required for efficient operation of the
culvert; however the design height should be no higher than where the subgrade meets the side slope.
Designers also need to ensure that any overtopping of the drain block will not force stormwater onto the road
surface. In determining the location and shape of the table drain block, there is a need to ensure that the
drain block does not create a safety hazard for errant vehicles.
The use of a table drain block for a particular culvert must be specified on the drainage plans along with its
design height.
A marker post should be placed on or adjacent to table drain blocks to alert maintenance personnel of their
existence.
Designers of widening or overlay/rehabilitation projects should check previous project documents for use of
table drain blocks in order to check or preserve the operation of the existing culverts.
2.13.3 Diversion Drains and Diversion Blocks
Diversion drains
On vertical grades, the depth and velocity of flow within a table drain can exceed acceptable limits if no relief
mechanism is included in the design. The steeper the grade, the quicker the limits are reached.
Diversion drains may be required, at intervals based on the grade of the road, to ‘turn’ the stormwater out of
the table drain and away from the road. On the upstream side of the road, the diversion drain could either
discharge flow into a catch drain located roughly parallel to the road and generally near the boundary (drain
discharges into a culvert located further down the slope) or onto the natural surface. On the downstream
side of the road, the diversion drain would normally discharge onto the natural surface, allowing the
stormwater to naturally flow away from the road.

Normally, the shape of the diversion drains should match that of the table drain, however where ‘V’ shaped
table drains are used, flat-bottomed diversion drains should be used unless conflicting requirements exist
that limit the width of the formed road such as:
• minimising disturbance or loss of existing roadside trees
• reducing earthworks costs.
The channel of the diversion drain should have an excavated cross-sectional area at least equal to that of
the table drain discharging water into it. Outlets should discharge onto grassed areas rather than onto
exposed soil.
Diversion blocks
Diversion blocks are earth embankments that are located along the downstream side of diversion drains and
divert stormwater flow from the table drain into the diversion drain. The block extends from the road
formation for several metres along the side of the diversion drain.
The height of the block typically starts (against the road formation) at subgrade level and should taper down
to closely match the natural surface at the end of the block.
Diversion drains and blocks are typically constructed at a 30° to 45° angle to the road however this can be
adjusted in order to reduce the slope within the diversion drain.
The initial grade (‘turnout’ grade) in the diversion drain should approximate the grade of the table drain to
avoid energy loss, and hence siltation and possible bank failure. As the drain increases in length, the grade
in the drain should progressively decrease. Where the diversion drain discharges onto the natural surface,
the last half of the diversion drain’s length should ideally have a surveyed grade of 0.2% to slow the flow
velocity.
Typical spacing of diversion drains along the grade is:

• 120 m for slopes up to 2%
• 60 m for slopes from 2% to 4%
• 30 m for slopes from 4% to 8%
• 15 m for slopes greater than 8%.
If the diversion channel is built through or under a fence (with the landholder’s written agreement/approval), it
is essential that landholders ensure that the outlet is kept clear to allow free draining of the drain and that
maintenance crews are allowed easy access via an adjacent gate. Depending on the fence type and
configuration, the impact of the fence on channel flows will need to be considered.
Discharge from a diversion channel may also be spread over a pasture to assist grass growth.
In areas with known dispersive subsoils, diversion banks (that extend from the diversion block) are preferred
over excavated diversion drains.
Where new and previous projects join, designers should appropriately consider the use and effects of
diversion drains and blocks.

2.14 Batter Drains and Chutes
Batter drains or chutes are structures that are designed to convey run-off from the tops of embankments
down the slope of a cut or fill batter and discharge to the natural ground surface or channel at either non-
erosive velocities or onto a non-erodible surface. Occasionally, in steep country they may be used to convey
run-off from the tops of cuttings down the batter to the roadway table drain or piped drainage system. For an
example of a batter drain see Figure 2.3.
Batter drains can be permanent or temporary. Temporary batter drains are used during the construction
period to control water flow and to protect mulched or newly seeded batters from the erosive forces of
concentrated flow.
On traversable batter slopes, the permanent batter drain or chute should be shaped to a channel which is
traversable by a vehicle.
Care needs to be taken, if batter drains and chutes are proposed in highly dispersive or erodible soils, as
their use is not generally desirable for these conditions.
A major problem with this type of drain is that of the flow overtopping or running parallel to the channel and
eroding the surrounding soil. The problem is prevalent with slope drains constructed from half-round metal
or concrete pipes, pre-cast concrete units or cast-in-situ concrete channels. This situation may be
aggravated by poor installation or by not providing sufficient slope drains, or not satisfactorily locating them,
to cater for the predicted flow. It is preferable to use rock mattresses, as they encourage a wider, less
concentrated flow, are flexible enough to settle and allow water to enter the channel from the side.
The slope drains used on embankments may be constructed progressively as the height increases. Care
should be taken to ensure the efficient collection of water at the top of the batter. It may be desirable to have
the slope drains more closely spaced during construction to minimise the size of flows.
The inlet to the slope drain should ensure that all the run-off is collected and should not allow water to flow
down the slope adjacent to the channel. The bottom of the drain should have some type of energy
dissipating, or erosion control device to prevent scouring at the base of the batter. The opposite bank of the
drain at the bottom of the batter drain or chute should also provide an erosion control device to protect the
bank and turn the flow, if required.
2.14.1 Design Procedures – Batter Drains
A four step process is provided for the design of batter drains:
Step 1 – Dimensions
Batter drains should have a minimum depth of 300 mm.
Hydraulic capacity of a batter drain is normally defined by the allowable head water level upstream of the
drain’s inlet.
Step 2 – Foundations
The lining of the batter drain should be adequately anchored to the foundations to avoid slippage or
separation, with a maximum distance of 3 m between anchorage points.
In cases where prefabricated units need to be bolted together, it is important that all bolt holes are sealed
with a flexible sealant to allow for flexural movement.

Step 3 – Inlet design
The inlet area should be protected against possible scour resulting from accelerating flow velocities (usually
more important on temporary batter drains). This protection is necessary to prevent water from either
undermining the top of the batter drain, or being diverted along the edge of the lining (the most common
cause of failure).
For temporary batter drains (i.e. drainage chutes) and during the early revegetation stage of permanent
batter drains, sand or gravel bags can be used to direct inflow towards the centre of the chute.
Step 4 – Outlet design
Typically, an energy dissipator will be required.
The outlet may consist of a bed of nominal 150 mm rock (minimum) placed with a minimum bed thickness of
250 mm or at least 1.5 times the maximum rock size. Typical dimensions of the rock bed are:
L = 6De metres long (minimum)
T + 0.6 metres wide at the batter drain outlet, expanding to T + 0.5L + 0.3 metres
W =
at the end of the dissipator.
where
De = equivalent pipe diameter (m) of the batter drain flow area
T = top width (m) of flow in the batter drain
Batter drains should be lined each side with a minimum 300 mm wide (turf) grassed filter strip or rock to
control side erosion caused by splash. In areas where the total disturbance is to be minimised, or where
introducing turf is undesirable, other forms of erosion control such as geotextiles or concrete may be
preferred.
2.15 Catch Drains and Catch Banks
2.15.1 Catch Drains
Catch drains (also known as cut-off drains in some jurisdictions) and catch banks are separate devices, but
can be used together. They are generally located on the high side of cuttings clear of the top of batters to
intercept the flow of surface water and upper soil seepage water (Figure 2.28). Their purpose is to prevent
overloading of the table drain and drilling and erosion or scour of the batter face.
Catch drains and/or catch banks can also protect embankments, disturbed areas and stockpile sites from
surface water. Alternatively, catch drains placed at the bottom of fill slopes intercept water from adjacent
properties as well as convey road drainage to an outlet.
These devices are generally located no closer than 2.0 m from the edge of the cuttings in order to minimise
possible undercutting of the top of the batter.
The type of catch drain shown in Figure 2.28 is usually about 0.3 m deep and is a function of the required
capacity. The width of the flat bottom drain will depend on the space available within the road reservation
and the likely run-off. The width and depth should be sufficient to carry the design flow, a width in the range
1.2 m to 2.5 m being typical.

In erodible soils, it is preferable for the catch drain in-cut to take the form of a low mound along the top of the
batter (Figure 2.28) as a drain, cut into the surface, may rapidly erode and enlarge itself, or cause local slips
in the batter by piping. V-shaped drains (Figure 2.29:) are not preferred and should not be used in erodible
soils. Erosion assessment is discussed in AGRD Part 5 – Section 3.6.
Figure 2.28: Typical catch drain with flat bottom
Source: Austroads (2010b).
Figure 2.29: Catch drain with V shape diagram
Depth 0.3 m minimum
Slope 1:6
Drain Slope 1:6
2.15.2 Catch Banks
Catch banks (Figure 2.30) are sometimes used instead of drains to reduce effects of seepage on stability of
the batter slopes and also to minimise disturbed ground and hence scour potential. However use of catch
banks alone may necessitate importation of suitable embankment material as in situ material may not be
suitable.
Figure 2.30: Catch bank diagram

2.15.3 Design Procedures – Catch Drains and Catch Banks
The design of catch drains and banks is based on the same methodology as for open channels. A freeboard
of at least 150 mm should be used. Where site conditions or some other constraint restricts the construction
of a channel with suitable dimensions, supplementary channel treatments such as synthetic channel linings,
riprap or concrete inverts should be utilised to withstand the higher velocities, Figure 2.31 shows an example
of a concrete lined catch drain.
Flow discharging from the catch drain should not be allowed to cause or aggravate erosion. Flow from a
catch drain may be discharged to an existing drain, watercourse, or to a chute.
Figure 2.31: Concrete-lined catch drain
2.16 Worked Examples
2.16.1 Example 1: Manning’s Equation – Determine the Flow Rate, Velocity and State of Flow for a
Stream
This example describes the process, using Equation 1 and Manning’s Equation (Equation 3) to determine
the flow rate, the velocity of a flow and the state of flow in a stream based on the discussion in Section 2.3 –
Fundamentals of Open Channel Flow.
The example commences after the stream data (such as cross-section, terrain, condition of channel and
stream profile to determine site bed slope) has been gathered.
The task
Given the stream data and height of flow (see diagram below); determine the velocity of flow in the channel,
the flow rate and state of flow (i.e. subcritical, critical, or supercritical flow).
Stream data
The stream has the following characteristics:

• bed slope about the site is 0.8%
• the height of the channel bed is 110.60 m
• the channel is regular and considered a little rough with a lot of trees and weeds along the banks.

Height = 111.80 m
sides are
1 on 1
2.5 m
Solution
Determine the velocity of flow using Manning’s Equation first, then the flow rate using the fundamental
equation and finally, determine Froude’s Number to describe the state of flow. (See Section 2.3.3 –
Fundamental Equations, Section 2.3.4 – Application of Fundamental Equations and Section 2.3.5 – Energy
Principles).
Manning’s Equation is Equation 22:

2� 1�
𝑅𝑅 3 x 𝑆𝑆 2 22
𝑉𝑉 =
𝑛𝑛
where
Hydraulic radius – the cross-sectional area of flow (A) divided by the wetted
𝑅𝑅 =
perimeter (P)
𝑆𝑆 = Slope of the energy line
The slope of the energy line (S) is not available; however, as the slope of the bed of the stream (So) is < 1%,
it can be used to approximate S.
Step 1
Calculate the cross-sectional area of the flow given that the depth of flow is 1.2 m.
A = 1.22 + 1.2 x 2.5 = 4.44 m2
Step 2
Calculate the wetted perimeter.
�(1.22 + 1.22 ) x 2 + 2.5 = 5.89 m
Step 3
Calculate the hydraulic radius.
R = A/P = 4.44/5.89 = 0.75 m
Step 4
Now determine an appropriate Manning’s roughness coefficient.

Using Table 2.2 in Section 2.3 – Fundamentals of Open Channel Flow, is channel, being the main channel
and regular in shape, is in the top portion of the table and within Section 1 of Table 2.2. The trees and
weeds would suggest (e), the range 0.06–0.08. Now, with the bed being a little rough, a value n = 0.07 (in
the middle of range) is considered appropriate.
Step 5
All variables have now been determined, therefore calculate velocity.
2� 1�
0.75 3x 0.008 2
𝑉𝑉 =
0.07
V = 1.06 m/s
Step 6
Using fundamental equation Q = V x A, the flow rate in the channel can now be determined.
Q = 1.06 x 4.44
= 4.71 m3/s
Step 7
To determine the ‘state of flow’, Froude’s Number is calculated.
𝐹𝐹𝑟𝑟 = 𝑄𝑄 �𝐵𝐵�𝑔𝑔 𝐴𝐴3
It has been determined that Q = 4.71 m3/s, A = 4.44 m2 and g is acceleration due to gravity (taken as 9.81
m/s2), B, the width of flow across the surface, has to be calculated.
B = 1.2 + 2.5 + 1.2 = 4.9 m
Therefore:
𝐹𝐹𝑟𝑟 = �4.9�(9.81 x 4.443 )
Fr = 0.36
Froude’s Number is below 1.0, therefore the flow is subcritical.
2.16.2 Example 2: Manning’s Equation – Determine the Depth and Velocity of Flow Based on a
Known Discharge/Flow Rate in a Stream
This example describes the process to determine the depth and velocity of flow based on a known
discharge/flow rate in a stream, based on discussion in Section 2.3 – Fundamentals of Open Channel Flow.
The example commences after the stream data (such as cross-section, terrain, condition of channel and
stream profile to determine site bed slope) has been gathered and the flow rate (as determined using the
Rational Method) has been estimated.

The task
Given the stream data and flow rate (see diagram below); determine the depth and velocity of flow in the
channel.
Stream data
The stream has the following characteristics:

• discharge/flow rate (Q) = 17.86 m3/s
• bed slope about the site (S) = 1.2%
• the height of the channel bed is 65.10 m
• the maximum depth of flow (d) is 2.0 m
• Manning’s Number, n = 0.06.
d = ?? m
sides are
1 on 1
4m
Solution
To solve for d, Manning’s Equation is used and a Stage-Discharge curve is developed. Manning’s Equation,
Equation 3, repeated here:
2� 1�
𝑅𝑅 3 x 𝑆𝑆 2
𝑉𝑉 =
𝑛𝑛
A Stage-Discharge curve plots discharge against depth of flow. Therefore several iterations using Manning’s
Equation are required for several depths of flow.
Step 1
Using the maximum channel depth of 2.0 m, calculate stream velocity and flow rate.
Calculate the cross-sectional area of the flow, wetted perimeter and hydraulic radius:
A = 12.00 m2, P = 9.66 m therefore R = 1.24 m

2� 1�
1.24 3x 0.012 2
𝑉𝑉 = = 2.11 m⁄s
0.06
Using Q = V x A
Therefore:
Q = 2.11 x 12
Q = 25.33 m3/s
This flow is greater than the known discharge therefore we know that the channel can easily carry the flow.

Step 2
Using the same method, re-calculate stream velocity and flow rate for several lesser depths (suggest using
even increments).
Wetted Hydraulic
Depth Area (A) Velocity (V) Discharge (Q)
perimeter (P) radius (R)
(m) (m2) (m/s) (m3/s)
(m) (m)
1.50 8.25 8.24 1.00 1.83 15.07
1.00 5.00 6.83 0.73 1.48 7.42
0.50 2.25 5.41 0.42 1.02 2.29
Step 3
Now draw the Stage-Discharge curve, Figure 2.32, for this site/channel.
Step 4
From the curve, the flow depth for the design flow is 17.86 m3/s.
Q = 17.86 m3/s, therefore d = 1.65 m
Step 5
The depth can be used to calculate flow area, then Q = V x A to determine the average flow velocity.
A = 1.652 + 1.65 x 4 = 9.32 m2
17.86 m3/s = V x 9.32 m2
V = 1.92 m/s

Figure 2.32: Stage–Discharge curve
Stage-Discharge Curve
30
25
20
Flow Rate (m3/s)
15
10
0
0 0.5 1 1.5 2 2.5
Flow Depth (m)
Step 5 (Check)
Wetted perimeter (P) = 2√(2 x 1.652) + 4 = 8.66 m
Hence R = 9.32/8.66 = 1.0762
V = 1.07642/3 0.012½/0.06 = 1.92 m/s
Q = 17.86 m3/s
Agreement with design discharge, therefore OK.
2.16.3 Example 3: Manning’s Equation: Determine the Flow Rate and Average Velocity in a
Compound Stream
This example describes the process to determine the flow rate and the average velocity of a flow in a
compound stream based on discussion in Section 2.7.1 – Design Methodology.
The example commences after the stream data (such as cross-section, terrain, Manning’s n and stream
profile to determine site bed slope) has been gathered.
The task
Given the stream data and height of flow (see diagram below); determine the average velocity of flow in the
channel and the flow rate.

Stream data
Bed slope about the site is 0.8%.
d = 1.2 m d = 2.1 m d = 1.0 m n values

A C A = 0.07
sides are B B = 0.035
3m
1 on 1 4m C = 0.06
2m
Solution
To solve for Q, we need to use Manning’s Equation (Equation 3) is used for each sub-section of stream:
2� 1�
𝑅𝑅 3 x 𝑆𝑆 2
𝑉𝑉 =
𝑛𝑛
After calculating V for each sub-section, use Qtotal = VA x AA + VB x AB + VC x AC to determine total flow rate.
Step 1
For sub-section A, calculate VA using Manning’s Equation.
Calculate the cross-sectional area of the flow wetted perimeter and hydraulic radius for sub-section A, using
Equation 4 repeated here:
𝐴𝐴
𝑅𝑅 =
𝑃𝑃
Therefore:
1.22
� � + 1.2 𝑥𝑥 4.0
2
𝑅𝑅 =
√1.22 + 1.22 + 4.0
𝑅𝑅 = 0.97 𝑚𝑚
It is important to remember that the water – water boundary between sub-sections A & B does not contribute
any length to the wetted perimeter.
RA = A/P = 5.52/5.70 = 0.97 m

2� 1�
0.97 3x 0.008 2
𝑉𝑉𝐴𝐴 = = 1.25 𝑚𝑚⁄𝑠𝑠
0.07
Using Q = V x A,
Q = 1.25 x 5.52
Q = 6.90 m3/s

Step 2
For sub-section B, calculate VB using Manning’s Equation.
AB = 7.39 m2
PB = 4.83 m
RB = 1.53 m
2� 1�
1.531 3x 0.008 2
𝑉𝑉𝐵𝐵 = = 3.40 𝑚𝑚⁄𝑠𝑠
0.035
Using Q = V x A,
𝑄𝑄𝐵𝐵 = 3.40 x 7.39
QB = 25.09 m3/s
Step 3
For sub-section C, calculate VC using Manning’s Equation.
AC = 3.50 m2
PC = 4.41 m
RC = 0.79 m
2� 1�
0.793 3x 0.008 2
𝑉𝑉𝑐𝑐 = = 1.28 𝑚𝑚⁄𝑠𝑠
0.06
Using Q = V x A,
𝑄𝑄𝑐𝑐 = 1.28 𝑥𝑥 3.50
QC = 4.47 m3/s.
Step 4
Now calculate:
Qtotal = 6.90 + 25.09 + 4.47 = 36.46 m3/s
and
Atotal = 5.52 + 7.39 + 3.50 = 16.41 m2
Therefore:
Vave = 36.46/16.41 = 2.22 m/s.
2.16.4 Example 4: Hard-lined/Rigid Boundary Channel Design
This example describes the process to design an open channel based on flow rate, selected channel side
slopes and other key properties as discussed in Section 2.7.1 – Design Methodology.

The task
Given the following channel cross-section and data, and using the channel bed slope to approximate the
Friction Slope (Sf), design a concrete-lined, trapezoidal channel.
Channel cross-section:
Solution
For best hydraulic section (Equation 15):
𝑏𝑏𝑏𝑏 + 𝑋𝑋𝑦𝑦 2 𝑦𝑦
𝑅𝑅 = =
2�𝑦𝑦 2 + (𝑦𝑦𝑦𝑦)2 + 𝑏𝑏 2
Step 1
Substitute X=2 and resolve for b as a function of y.
𝑏𝑏𝑏𝑏 + 2𝑦𝑦 2 𝑦𝑦
𝑅𝑅 = =
2�𝑦𝑦 2 + (2𝑦𝑦)2 + 𝑏𝑏 2
𝑏𝑏 = 𝑦𝑦�2√5 − 4�
Step 2
Combine Manning’s Equation (Equation 3) and Equation 1, then substitute in Q = 75 m3/s,

A = by + Xy2, R=y/2, So = 0.0025 (remember; So estimates Sf) and n = 0.014:
2
𝑦𝑦 �3 1
(𝑏𝑏𝑏𝑏 + 2𝑦𝑦 2 ) � � 0.0025 �2
75 = 2
0.014
𝑦𝑦 2�3
21 = (𝑏𝑏𝑏𝑏 + 2 𝑦𝑦 2 ) � �
2
Now substitute:
𝑏𝑏 = 𝑦𝑦�2√5 − 4�

The expression becomes:
𝑦𝑦 2�3
21 = ��𝑦𝑦�2√5 − 4�� 𝑦𝑦 + 2𝑦𝑦 2 � � �
2
and
y = 2.65 m, depth of flow
Step 3
Calculate b using y:
𝑏𝑏 = 𝑦𝑦�2√5 − 4�
𝑏𝑏 = 2.65 x �2 √5 − 4� = 1.25 m, width of bed
Step 4
Calculate freeboard.
In accordance with Section 2.7.1 – Design Methodology, freeboard is the greater value of:
1. 300 mm
2. 20% of 2.65 m = 530 mm
3. Area of flow = 17.36 m2.
Therefore:
𝑄𝑄
𝑉𝑉 =
𝐴𝐴
75
𝑉𝑉 =
17.36
V = 4.32 m/s
𝑉𝑉 2
𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 ℎ𝑒𝑒𝑒𝑒𝑒𝑒 =
2𝑔𝑔
4.322
𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 ℎ𝑒𝑒𝑒𝑒𝑒𝑒 =
2 x 9.81
Velocity head = 950 mm
Select the greater of (1), (2) and (3).
Therefore freeboard is 950 mm.
Step 5
Check the Froude Number (Equation 11):
11.85
𝐹𝐹𝑟𝑟 = 75 �
𝑔𝑔17.363
=1.14 Supercritical

Step 6
Determine final channel geometry:
Width =11.85m
Depth=3.60m V=4.32m/s
y=2.65m
Supercritical
b=1.25m
2.16.5 Example 5: Vegetated/Erodible Boundary Channel Design
This example describes the process to design an open channel based on flow rate, selected channel side
slopes and other key properties, as discussed in Section 2.7.1 – Design Methodology.
Given the following cross-section and data, and using the channel bed slope to approximate the Friction
slope (Sf) design a vegetated, trapezoidal channel in red soil (assume 70% coverage).
Channel cross-section:
Q = 15 m3/s
Side
Sideslope
slopetotobe
be1
on
2:12
Channel bed slope (So) = 2%
Manning’s n for vegetation cover = 0.07
Solution
Step 1
Use basic equation (Equation 1) to determine waterway area:
𝑄𝑄 = 𝑉𝑉 x 𝐴𝐴
𝑄𝑄
𝐴𝐴 =
𝑉𝑉𝑚𝑚𝑚𝑚𝑚𝑚
Using Table 2.6, find Vmax as 1.8 m/s [upper part of table (red soil), 0.70 coverage and 2% channel gradient].
𝑄𝑄 15
𝐴𝐴 = = = 8.33 𝑚𝑚2
𝑉𝑉𝑚𝑚𝑚𝑚𝑚𝑚 1.8

Step 2
Combine Manning’s Equation (Equation 3) and Equation 1, then rearrange into a function of P:
2� 1�2
𝐴𝐴𝑅𝑅 3 𝑆𝑆
𝑜𝑜
𝑄𝑄 =
𝑛𝑛
3�
5� 1�2 2
𝐴𝐴 3 𝑆𝑆
𝑜𝑜
𝑃𝑃 = � �
𝑄𝑄𝑄𝑄
Now substitute known values in the previous equation and solve for P using Q = 15 m3/s,
A = 8.33 m2, R = A/P, So = 0.02 (remember; So estimates Sf) and n = 0.07.
3�
5 1� 2
8.33 �3 0.02 2
𝑃𝑃 = � � = 9.90 𝑚𝑚
15 x 0.07
Step 3
Based on channel geometry, the two expressions for A and P are:
𝐴𝐴 = (𝑏𝑏 + 𝑦𝑦𝑦𝑦)𝑦𝑦
𝑃𝑃 = 2𝑦𝑦�𝑋𝑋 2 + 1 + 𝑏𝑏
Combine these expressions and solve for y:
𝐴𝐴 = (𝑏𝑏 + 𝑋𝑋𝑋𝑋)𝑦𝑦
𝐴𝐴
𝑦𝑦
Now substitute the expression for b into the equation for P:
𝑃𝑃 = 2𝑦𝑦�𝑋𝑋 2 + 1 + 𝑏𝑏
𝐴𝐴
𝑃𝑃 = 2𝑦𝑦�𝑋𝑋 2 + 1 + − 𝑦𝑦𝑦𝑦
𝑦𝑦
𝐴𝐴
𝑃𝑃 = �2 �𝑋𝑋 2 + 1 − 𝑋𝑋� 𝑦𝑦 +
𝑦𝑦
𝑃𝑃𝑃𝑃 = �2�𝑋𝑋 2 + 1 − 𝑋𝑋� 𝑦𝑦 2 + 𝐴𝐴
0 = �2�𝑋𝑋 2 + 1 − 𝑋𝑋� 𝑦𝑦 2 − 𝑃𝑃𝑃𝑃 + 𝐴𝐴
This is a quadratic equation where:
𝑎𝑎 = 2 �𝑋𝑋 2 + 1 − 𝑋𝑋
𝑏𝑏 = −𝑃𝑃
𝑐𝑐 = 𝐴𝐴

Now substitute in known values for X, P and A and solve the quadratic equation
Solution for a quadratic equation is:
−𝑏𝑏 ± √𝑏𝑏 2 − 4𝑎𝑎𝑎𝑎

𝑦𝑦 =
2𝑎𝑎
Therefore:
9.90 ± 3.96
𝑦𝑦 =
4.94
The solutions (roots) for equation (y) are 2.80 and 1.20.
Now:
𝐴𝐴
𝑦𝑦
Solution 1 (when y = 2.8) is –2.63 and Solution 2 (when y = 1.2) is 4.54.
Dismiss negative solution, hence y = 1.20 m and b = 4.54 m
Step 4
Calculate freeboard.
In accordance with Section 2.7.1 – Design Methodology freeboard is the greater of:
• 300 mm
• 20% of 1.20 m = 240 mm
• Velocity head = V2/2g = 1.82/2 x 9.81 = 92 mm
Select greater of (1), (2) and (3). Therefore freeboard is 300 mm.
Step 5
Check the Froude Number (Equation 11):
B = 4.54 + 2(2 x 1.20) = 9.34 m
9.34
𝐹𝐹𝑟𝑟 = 15�(9.81 = 0.61 Subcritical
x 8.333 )

Step 6
Determine final channel geometry:
Width = 9.34 m
Depth = 1.50 m V = 1.8 m/s

y = 1.20 m
b = 4.54 m

3. Culverts
3.1 Introduction
Culverts must be designed to satisfy the hydraulic requirements associated with conveying water under
roads and to have the structural strength to carry construction or highway traffic loads and loads associated
with the road formation.
In designing culverts, a number of issues must be considered including:

• economy (e.g. cost of the project and available budget)
• road immunity – the extent to which flows are passed through culverts under the road rather than allowed
to overtop the road and force its closure
• stream characteristics such as channel shape, slope and natural stream flow (also consider possible
stream morphology over the life of the culvert)
• the culvert structure including type, material, size and number of pipes or cells
• afflux (i.e. the change in water level caused by the road and culvert/restriction)
• tailwater and presence of possible backwater effects
• the culvert’s outlet velocity and flow energy
• erosion, upstream and downstream
• groundwater effects (including longitudinal seepage, piping and erosion – considering the need for cut-off
collars)
• groundwater conditions (including pipe strength, culvert settlement and culvert anchorage when located
on steep slopes)
• the special needs of culverts which are to be used as fishways, for the passage of terrestrial fauna or as
cattle creeps, or underpasses (larger culverts will often be required for fauna or fish passage than for
hydraulic reasons)
• safety and access (i.e. catering for the needs of pedestrians, cyclists, or maintenance crews)
• durability – design life of the structure
• environmental issues (minimising the potential for unacceptable environmental damage).
3.2 Information Required
It is important to gather all required information before design commences. Some information should be
gathered/verified by site inspection.
The catchment area should be carefully defined from maps, as well as aerial photographs and
photogrammetry if available. It is usual to survey the stream bed and adjacent land upstream and
downstream of the culvert site and features such as other culverts, houses and commercial developments
(and possibly their floor levels, if it appears that they might be more at risk of flooding due to the new culvert).
Land use adjacent to the culvert site and any utility services (public or private) which may influence the
location and level of the culvert should also be noted. Important information in relation to the highest known
past flood levels can be obtained by interviewing local residents and road maintenance supervisors. The
designer should also inspect the site thoroughly as the survey may not show all the details relevant to good
design and ease of construction.

A preliminary road design is also required to determine maximum culvert size (height) for possible trial
culvert configurations, to check any clearance or cover requirements and to determine allowable headwater
criteria. Other data required includes hydraulic performance criteria for road section or link, any
environmental requirements/constraints and most importantly rainfall data.
Further information on the data required is located in Section 3.10 – Design Procedures which provides a
generic list of required information. Also designers should see AGRD Part 5 – Section 6 for hydrologic
assessment.
3.3 Culvert Location
In most cases, with the exception of wide floodways, culvert locations are predetermined by the intersection
of a watercourse within an existing or proposed roadway. In general, a culvert should be located to fit within
the natural channel in line and grade as far as is practical. However, the culvert location and size should be
chosen to:
• achieve good hydraulic performance
• ensure stability of the stream bed and banks
• reduce risks for errant vehicles
• minimise construction risks
• minimise construction and maintenance costs.
Where circumstances allow, culverts should be located away from:

• erodible or meandering channel bends or banks
• critical or isolated aquatic habitat areas
• isolated sections of remnant, valued, or protected riparian vegetation.
3.3.1 Horizontal Alignment
A culvert should be laid on a straight alignment, which may be either perpendicular or skewed to the road
centreline. The natural skew in a channel should be adopted wherever possible, as a culvert on a different
skew could cause progressive bank erosion, which may possibly encroach into private property. Protection
of a bank at risk from erosion could be costly.
Realigning short sections of an existing channel to fit the culvert alignment should be avoided as realignment
of a natural channel requires careful design to avoid erosion, and such channel re-alignments may require
the approval of the statutory or other drainage authority.
Where severe changes in channel alignment are necessary, it is recommended that the culvert not be placed
perpendicular to the road as short radius bends in the channel are likely to lead to erosion on the concave
bank and to a build-up of sediment on the opposite bank (Figure 3.1). The road alignment should as much
as possible avoid crossings in such locations.
Where following the natural channel would result in a very long skewed road crossing, it may be cheaper to
construct a stream deviation (see Options A and B in Figure 3.2).

Figure 3.1: Stream alignment options Figure 3.2: Stream alignment options (A and B)
(preferred option A to avoid bank
Source: VicRoads (2003).
3.3.2 Vertical Profile
Most culvert profiles should approximate the natural streambed. Other profiles may be chosen for either
economic or hydraulic reasons. Slopes other than that of the natural stream, can be used to prevent stream
degradation, minimise sedimentation, improve the hydraulic performance of the culvert, shorten the culvert,
or reduce structural requirements. Culverts are typically smoother than the natural channel and hence an
increase in flow velocity is expected. Careful design and detailing is required to control steam erosion and
deposition. Slope alterations should, therefore, be given special attention to ensure that detrimental effects
do not result from the change.
Culvert grading options are shown in Figure 3.3. The designer should check whether energy dissipation is
required so that damage will not occur upstream or downstream (see Section 3.13 – Culvert Outlet
Protection).
The culvert should be designed to:

• suit the outlet conditions even if inlet conditions have to be modified (e.g. a drop inlet to reduce potential
scouring velocities through the culvert). High outlet velocities can cause erosion for a significant distance
downstream of an outlet. Where high outlet velocities are expected, appropriate dissipation measures
and/or protection measures will be required
• be free from sediment deposits, which tend to occur on the inside of stream bends, or where there is an
abrupt change from the stream slope to a flatter grade in the culvert.
3.3.3 Non-hydraulic Considerations
Provision must also be made for fauna and stock movement as required. Where roads traverse broad flood
plains or otherwise interfere with overland flow patterns, regular culverts may be needed to mitigate against
the adverse environmental effects of drainage shadow. See AGRD Part 5 – Section 3.7.10, for this and
other environmental aspects.

Figure 3.3: Culvert grading options
Source: Adapted from VicRoads (2003).

3.4 Culvert Type
The selection of the most appropriate type of culvert is dependent on a range of factors including economics,
site conditions, hydraulic performance, limitations on allowable water level at inlet, height of fill above the
obvert, channel width, structural requirements, durability, ease of construction, the possibility of blocking by
debris and environmental considerations.
3.4.1 Culvert Shape
The most common culvert type is the pipe because it is hydraulically and structurally efficient in most cases,
and is cheaper to produce than a comparable box shaped culvert (including inverted units).
Box culverts are advantageous where:

• there is minimal embankment depth (boxes require less material cover)
• there are channel width restrictions
– for single barrels a pipe having equivalent waterway area to a box is wider for the same height
– for multiple pipe culverts backfill is required between the barrels whereas boxes may be placed side
by side
• a rectangular waterway achieves the allowable headwater
• fauna or livestock passage is required
• there is a need to minimise or control low-flow water depths.
It is important to select the culvert shape that will best fit the waterway. Multiple pipes or rectangular units
are useful in wide channels where the constriction or concentration of flow is to be kept to a minimum. In the
case of a box culvert, it is usually more economical to use multiple units than a wide single span. With multi-
cell construction, box culverts combined with link slabs is often an economical choice. In particular
circumstances (e.g. to minimise channel and bank disturbances), arches may also be appropriate and
economical.
Reference should also be made to jurisdictional standard drawings and/or Australian Standards for
installation requirements including cover, spacing and details for relevant support conditions.
3.4.2 Materials
The most common type of material used in the construction of culverts is steel reinforced concrete.
However, in many circumstances other materials, such as fibre-reinforced concrete, corrugated metals (steel
and aluminium) and plastics (polyvinyl chlorides, polypropylenes and polyethylenes) have been used. This
Guide defines rigid culverts as made of steel reinforced concrete, while semi-rigid culverts are made from all
other materials.
Steel reinforced concrete culverts typically have lower cover requirements and do not deflect as a result of
poor compaction (see AGRD Part 5 – Appendix B). Alternatively, ‘flexible’ culverts (able to deflect under
load) offer ease of transportation and installation as benefits. Depending on the flexible material used they
may also offer advantages such as lower cost, chemical inertness and hydraulic smoothness. It should be
noted that some jurisdictions have restrictions over the use of some flexible culvert materials.
The use of semi-rigid or flexible culverts requires a higher level of workmanship and higher specified backfill
material, as compared to rigid pipes, to ensure that culvert shape and barrel stability is ensured. As a
consequence, these culverts are more susceptible to ground disturbance after installation and hence are not
preferred when there are buried utilities in close proximity.

Unless there are large financial savings, or other construction restraints, more robust and durable materials
should be used. AGRD Part 5 – Appendix B provides a summary of these materials as well as their sizes,
strengths, stock lengths, joint types, restrictions and typical use.
Aluminium culverts referred to in AGRD Part 5 – Appendix B have been used in small quantities by some
road agencies. Aluminium is suitable for use in soils that have a pH in the range 4 to 9. This pH range is
aggressive to many materials and aluminium culverts represent a durable solution in these environments.
Design criteria for aluminium culverts are covered in AS/NZS 2041.1.
Aluminium is not as strong as steel; consequently a larger wall thickness is required to provide an installation
that has an equivalent strength to steel. Aluminium is also more expensive than steel.
In high groundwater situations sealed pipe joints may result in culvert flotation. An alternative is to utilise a
geotextile joint strap and use the culvert as a subsoil pipe. With this detailing a lower groundwater level can
be expected and enhanced pavement protection can be achieved.
In selecting culvert type and material, the work health and safety requirements for inspecting and maintaining
the culvert must be considered.
Further information on the use and applicability of various materials used in drainage is provided in AGRD
Part 5 – Appendix B.
3.5 Culvert Size
3.5.1 Minimum Culvert Size Allowable
The recommended minimum diameter of any pipe culvert subject to road agency requirements is 375 mm to
450 mm diameter in order to reduce the likelihood of blockage. Smaller pipe diameters may require greater
inspection and maintenance regimes.
Whilst the minimum recommended size of culverts across a road formation is 375 mm diameter, a larger size
may be necessary where debris occurs. Where there is the possibility of blockage and pipe settlement, a
750 mm minimum culvert diameter is recommended to allow access for maintenance inspection.
The recommended minimum height of any box section, subject to road agency requirements, is 375 mm.
In selecting the culvert size, the Work Health and Safety requirements for inspecting and maintaining the
culvert must be considered.
3.5.2 Other Sizing Considerations
Considerations for the selection of culvert size are:

• A culvert flowing full makes the most use of the waterway opening available. The culvert opening height
should desirably match the tailwater depth; however, it is recommended that the height not be greater
than 1.25 times the depth of the tailwater. Typically this results in an outlet control culvert which gives a
near full flow with lower outlet velocities. Culverts will generally not flow full if operating under inlet
control.
• The larger the opening, the less prone to blocking of the waterway by debris. The minimum culvert size
required for debris passage should be assessed including the likely debris (type and size), the risk of
blockage and likely impacts of subsequent failure of the culvert.
Environmental considerations, including the needs of fish, fauna and stock in the area should also be taken
into account. See AGRD Part 5 – Section 3, and Austroads Guide to Road Design Part 6A: Pedestrian and
Cyclist Paths (AGRD Part 6A) (Austroads 2009), for more information.

For culverts used as pedestrian and cyclist facilities the minimum recommended invert level is the one in two
year flood level. For further information, see AGRD Part 6A (Austroads 2009).
Following the selection of the size for culverts that are used for drainage and other purposes, the hydrologic
calculations should then be re-checked to confirm adequate hydraulic operation and performance, including
blockage potential.
3.6 Structural Requirements
The design loading on culverts to ensure that they have adequate structural strength must be in accordance
with the road agency’s requirements.
3.6.1 Design Loads
Loads on culverts include:
Dead loads
• fill over the structure, which is a function of:
– height of fill
– type of fill material
– installation conditions e.g. ‘trench’ or ‘embankment’.
Live loads
• design traffic loads
• construction traffic loads
• other or abnormal load conditions.
The load-bearing capacity of a culvert is a function of:

• unit strength (e.g. pipe class)
• type of bedding and backfill material
• pipe diameter (excluding box culverts)
• pipe material.
At sites that are susceptible to earthquakes, longitudinal connectivity of the culvert, robustness of the culvert
foundation and anchorage provided by the inlet and outlet structures, should be considered to mitigate the
effects associated with:
• foundation liquefaction
• road formation spreading laterally
• batter instability.
3.6.2 Existing Culverts
Culverts that are to be retained in a project need to be checked for both hydraulic capacity and structural
durability. Construction staging and final design cover should also be reviewed. See AGRD Part 5 –
Appendix B for further guidance.

3.6.3 Space between Pipes in Multiple–Barrel Culverts
When two or more pipe barrels are laid parallel, they should be separated by a distance that is sufficient to
enable thorough compaction of the back-fill material which is essential for haunch support and the prevention
of settlement. Typical minimum spacing is governed by the type of compaction equipment, however a
controlled low strength material, flowable or stabilised backfill (see AS/NZS 3725) can be compacted using
vibratory rods rather than plate compactors and therefore pipe spacing may be reduced.
Multiple pipe culverts should always be treated as an embankment installation when determining the class of
pipe required from the allowable height of cover. Even for culverts installed in trench conditions, the height
of cover should be calculated for embankment installation.
Box culverts are typically laid without a gap between the lines of culvert units.
3.6.4 Cover
The desirable minimum cover to be adopted for design purposes for box culverts is 0.1 metres below the
subgrade level, to allow for continuous usage of paving machinery. However if necessary the absolute
minimum cover can be 0.1 metres to finished surface level, that is, the culvert can be laid in the pavement
layers. Advice should be sought from experienced structural and pavement engineers before adopting the
absolute minimum condition to ensure appropriate structural integrity is provided and that cracking induced
at the culvert joint does not reflect to the pavement surface.
For box culverts the maximum economic height of fill or depth of trench is about 1.5 metres. Boxes or crown
units required to carry greater depths may be specially designed by an accredited structural designer.
A special treatment such as the induced trench methods may have to be designed so as to retain existing
box culverts where proposed heights of fill exceed 1.5 metres.
Maximum cover limits apply to box culverts, designers should see the manufacturer’s specifications and the
road agency’s specific requirements for cover depths.
For cover requirements for pipe culverts, see AGRD Part 5A – Section 6.4.1.
3.7 Hydraulic Design Considerations
3.7.1 Design Discharge
The design discharge for a culvert is the discharge the culvert is designed to pass based on an accepted
average recurrence interval (ARI). Typically this discharge equals the catchment run–off as calculated by
the method set out in AGRD Part 5 – Section 6.
The performance of a culvert should be checked for a range of design discharges and these shall include
the:
• level of serviceability provided to road users
• serviceability limit state for the culvert – no damage shall occur to the culvert as a result of the design
average recurrence interval (ARI) event or more frequent events
• ultimate limit state for the culvert (culvert or embankment failure may occur under less frequent events)
• environmental flow conditions for fish passage or water quality design.
For design discharge criteria, see AGRD Part 5 – Table 4.2 and Table 4.3, or the relevant road agency’s
policy documents as immunity/accessibility levels vary depending on class and/or function of the road.

3.7.2 Outlet Velocity
Part of the road drainage design process is the determination of acceptable or maximum allowable velocities
for stormwater flows. It should be noted that these velocities are largely based on research that identified the
velocity when erosion/scour started to occur in different soil/stream types. The maximum allowable velocities
for a project are then used in the design of various drainage structures/devices (e.g. culverts and channels)
to ensure design discharge through those devices is below the set maximum allowable velocity for that
location. Some design solutions that may be adopted are based on velocities that are equal to, or just
below, the set maximum allowable velocity.
High outlet velocities can cause bank erosion downstream of an outlet. Where high outlet velocities are
expected, appropriate protection and/or dissipation measures will be required. Where possible, culverts
should be designed to provide acceptable velocities without the need for additional stream bed protection.
Allowable stream velocities to avoid scour vary according to soil type and topography. Some suggested
outlet velocities for various stream bed materials are given in Table 3.1. Scour and preventative measures
are discussed further in Section 3.13 – Culvert Outlet Protection.
Table 3.1: Suggested culvert outlet velocities
Material downstream of culvert endwall Allowable velocity (m/s)
Stable rock 4.5

Stones 150 mm diameter or larger 3.5
Gravel 100 mm or grass cover 2.5
Firm loam or stiff clay 1.2–2.0
Sandy or silty clay 1.0–1.5
Note: These are target velocities at the culvert outlet.

It is desirable that culvert outlet flows be subcritical as this will eliminate a hydraulic jump occurring at or
downstream of the outlet and minimise scour due to the energy dissipation (when flows change from
supercritical to subcritical flow). The Froude Number can be used to determine the flow regime, see
Section 3.10.10 – Outlet Flow Energy.
See Appendix E.1.1, for an example of using the Manning’s Equation to determine flow characteristics for an
existing channel in order to provide conditions for the design of a new culvert.
3.7.3 Culverts in Flat Terrain
In flat terrain, drainage channels are often ill-defined or non-existent and culverts should be located and
designed for least disruption of the existing flow conditions. In these locations multiple culverts can be
considered to have a common headwater elevation, although this will not be precise.
It may also be necessary to construct levee banks (Figure 3.4), to achieve the design headwater at the
culvert location. Where necessary, approval of the local drainage authority should be obtained prior to
construction of any levee banks.

Figure 3.4: Development of headwater
3.7.4 Siltation and Blockage
The likelihood of blockage should be considered for all culverts. Blockage can occur through siltation or
vegetation, though blockage by siltation is more likely to be temporary in nature, because during flood
events, silt deposits can be removed by high velocity flows. To prevent siltation the desirable minimum
velocity in the culvert should be above 0.7 m/s. A check of velocities should be undertaken as part of design.
Where blockage caused by debris (larger objects) is considered likely, larger culvert sizes may be required.
This will allow the debris to pass through the culvert and minimise the risk of blockage and of any adverse
impacts that could occur to the roadway and/or to surrounding properties. Road agencies and/or drainage
authorities may apply a blockage factor, typically 20 to 50%, and therefore guidance should be sought from
the road agency.
Further information is contained in AGRD Part 5 – Section 3.
3.7.5 Allowable Headwater (AHW)
The allowable headwater for a culvert will usually be determined by one or more of the following conditions:
• The elevation of upstream ponding should not cause unacceptable damage or adverse effects to adjacent
properties. The extent of any upstream flooding of properties should be ascertained from field
investigations or by design calculations.
• Where a road is designed not to be overtopped during a set ARI flood event, it is desirable to provide
freeboard of at least 100 mm between the design upstream floodwater surface and the upstream
shoulder edge. Where this is not economically acceptable, pavement design should make allowance for
higher water levels, and the likely duration of inundation.
• In the event of both of the above conditions permitting a high headwater, the associated outlet velocity
may be intolerably high. In this case, the allowable headwater may have to be reduced to limit the outlet
velocity to an acceptable value, that is, one that does not cause unacceptable scouring.
• Where practicable, it is desirable to keep the headwater/afflux in the road reserve or upstream with the
landowner’s permission.
• The elevation of upstream ponding should not cause unexpected hazards to people or stock.
• Where floodwaters are expected to remain on or over the road embankment for some time, the road
embankment design should allow for infiltration of floodwater depending on the anticipated duration of
inundation.
• When overtopping flow needs to be considered, the overtopping flow conditions shall be determined
using Section 4 – Floodways, and superimposed with the culvert flow conditions and the ponding and
scour effects limited or mitigated as necessary.

3.7.6 Tailwater (TW)
Tailwater (TW) is the normal water depth within the existing channel for the design discharge. The tailwater
depth should be determined adjacent to the culvert outlet. Tailwater level is an important aspect of culvert
design because:
• A high tailwater may cause the culvert to flow full or under pressure, so increasing the headwater
necessary to pass the flow.
• A low tailwater relative to the depth of flow in the culvert can result in erosion of the downstream channel.
If the channel is regular in shape and uniform flow conditions can be reasonably expected, the tailwater
depth can be determined using Manning’s Equation see Section 2.7 – Open Channel Design.
For complex channels, backwater models such as the Hydraulic Engineering Centres River Analysis System
(HEC-RAS) can be used, though this would normally only be necessary for large streams. This work should
be undertaken by a specialist hydraulic engineer.
3.8 Typical Culvert Operating Conditions
3.8.1 General
The most important consideration in culvert hydraulic performance is whether the flow is subject to inlet or
outlet control. Figure 3.5 shows eight typical conditions under which culverts may operate (i.e. the range of
flow types commonly encountered in culverts). For inlet control two distinct regimes exist, depending on
whether the inlet is submerged or not submerged. Outlet control occurs in long culverts, culverts laid on flat
grades and culverts with high tailwater depths.
A study of Figure 3.5 will reveal that the prediction of inlet or outlet control is not precise and the standard
procedure for calculation is to assume each control situation in turn, the control situation with the highest
calculated headwater being adopted for design.
Checks should be made to ensure a submerged inlet type (HW > 1.2D). Submerged inlets usually have
more driving head which helps improve culvert capacity/performance, which is preferable.
The desirable control condition is outlet control as shown in Type 7 in Figure 3.5.
3.8.2 Inlet Control Conditions
For inlet control, the capacity of the culvert barrel is greater than that of the inlet. Hence, culvert capacity is
dictated by conditions at the inlet. Under these situations the important factors are the cross-sectional area
of the culvert barrel, the depth of headwater or ponding at the entrance and the entrance conditions,
including the entrance type, existence and angle of headwalls and wingwalls, and the projection of the
culvert into the headwater.
For one-dimensional flow, the relationship between the discharge and the upstream energy can be
computed by an iterative process or by the use of nomographs.
Inlet control can occur with the inlet submerged and the outlet not submerged. Under these conditions, the
flow contracts to a supercritical jet immediately downstream from the inlet. When the tailwater depth
exceeds critical depth, dc and the culvert is laid on a steep grade, flow remains supercritical in the barrel and
a hydraulic jump will form near the outlet. If the culvert is laid on a slope less than critical, then a hydraulic
jump will form in the barrel.

When the culvert flows under inlet control, the roughness and length of the culvert barrel and the outlet
conditions (including the depth of tailwater) are not factors in determining culvert capacity. An increase in the
slope of the culvert reduces headwater only to a small degree, and can normally be neglected for
conventional culverts flowing under inlet control.
In Figure 3.5, Types 3, 4, 5 and 8 are operating under inlet control.

Figure 3.5: Typical conditions under which standard culverts operate
T 7 i th f d ti
Source: Herr and Bossy, cited in DTMR (2010).

3.8.3 Outlet Control Conditions
With outlet control, the culvert flow is restricted to the discharge which can pass through the conduit for a
given level of water in the outlet channel (tailwater level). The slope, cross-sectional area, roughness and
length of the culvert barrel have to be considered as these losses exceed the inlet losses. However, inlet
size, shape and edge geometry can still affect the capacity.
In general the control will be at the outlet if the culvert slope is less than critical. A tailwater depth equal to
80% or more of the height of the culvert barrel/cell will usually indicate outlet control, except in rolling or
mountainous country with the culvert on natural surface slopes. However, a check of the design assuming
inlet control is such an easy process that it forms part of standard design procedure.
Culverts flowing with outlet control can flow with the culvert barrel full or with the barrel part-full for all of the
culvert length. With outlet control, having both the inlet and the outlet submerged, the culvert flows full and
under pressure. The culvert also can flow full over part of its length with part-full flow at the outlet. The point
at which the water surface breaks away from the barrel obvert depends on the tailwater depth and culvert
grade, and can be determined by using flow profile calculations.
If the culvert is laid at a flat grade, outlet control can occur with both inlet and outlet not submerged, and the
part-full flow throughout the culvert length will be flowing under sub-critical conditions.
Variations of these main types of flow conditions can occur, depending on the relative value of critical slope,
normal depth, culvert height and tailwater depth. While the potential flow conditions shown in Figure 3.5 are
the most common for simple culverts, different flow conditions are possible where complex culvert structures
are required. Advice may be required from an expert in such cases.
In Figure 3.5, Types 1, 2, 6 and 7 have outlet control.
3.9 Hydraulic Calculations
The hydraulic design of culverts should be undertaken either manually as described in this Guide, or by
using software approved by the relevant authority.
The following design methodology (this section) and design procedure (Section 3.10 – Design Procedures)
are based on the use of nomographs (see Appendix B for inlet and outlet control nomographs) and provides
an understanding of the hydraulic design of culverts. Other methods of culvert design, including the use of
formula, are available however use of these methods requires approval by the relevant road agency.
Irrespective of the hydraulic analysis method used, the Hydraulic Grade Line for the partially or fully
pressurised culvert shall be determined. The hydraulic parameters to be determined are shown in
Figure 3.6.
3.9.1 Control at Outlet
Flow under outlet control can be calculated from the formulae below, the parameters for which are illustrated
in Figure 3.6. The total head (H) required to convey water through a culvert flowing under outlet control is
determined by Equation 23:

H = He + Hf + Hv 23
where
𝑉𝑉 2
He = Entrance loss (m) = 𝑘𝑘𝑒𝑒
2𝑔𝑔
19.6 𝑛𝑛2 𝐿𝐿 𝑉𝑉 2
Hf = Friction loss (m) = x
𝑅𝑅 1.33 2𝑔𝑔
𝑉𝑉 2
Hv = Velocity head (m) =
2𝑔𝑔
and
𝑉𝑉 = Mean velocity of flow in the culvert barrel (m/s)
𝑔𝑔 = Acceleration due to gravity (assume 9.81 m/s2)
𝑘𝑘𝑒𝑒 = Entrance loss coefficient (see Table 3.2)
𝑛𝑛 = Manning’s roughness coefficient
𝐿𝐿 = Length of culvert barrel (m)
𝑅𝑅 = Hydraulic radius (m)
Figure 3.6: Hydraulics of culvert flowing full under outlet control

2
V1 2
2g V
2g WS = water surface
S = slope of culvert
WS
W.S He
Energy Lin
e Hf
HW Hydraulic G H
V1 h1 ra de Line
Hv W.S
D .
SoS
V h2
LS
LSo dhc c DATUM

Substituting in the first equation (Equation 23) above and simplifying:
19.6 𝑛𝑛2 𝐿𝐿 𝑉𝑉 2
𝐻𝐻 = �1 + 𝑘𝑘𝑒𝑒 + � x
𝑅𝑅1.33 2𝑔𝑔
This equation can be solved for H by the use of the full flow nomographs in Appendix B, Figure B 5 and
Figure B 6 for concrete culverts and Figure B 7 to Figure B 10 for metal culverts.
From the development of this energy equation and Figure 3.7, H is the difference between the elevation of
the hydraulic grade line at the outlet and the energy line at the inlet. Since the velocity head in the entrance
pool usually is small when ponded conditions occur (V2/2g = 0), the water surface of the headwater pool
elevation can be assumed to equal the elevation of the energy line.
Figure 3.7: Determination of adopted outlet depth (ho)
HW H
D SoS
LSo
LS dhcc TW
ho = Greater of hdcc ++ D
D and TW
2
Headwater depth under outlet control is calculated according to Equation 24:
𝐻𝐻𝐻𝐻 = 𝐻𝐻 + ℎ𝑜𝑜 − 𝐿𝐿𝑆𝑆𝑜𝑜 24
where
𝐻𝐻 = Total head (m) determined from Equation 23
ℎ𝑜𝑜 = Adopted outlet depth (m)
𝐿𝐿 = Length of culvert (m)
𝑆𝑆𝑜𝑜 = Slope of culvert barrel (m/m)
The various components of this equation are illustrated in Figure 3.7.

The adopted outlet level (ho) equals TW if TW > D, otherwise it is the greater of TW or Equation 25:
TW or
𝑑𝑑𝑐𝑐 +𝐷𝐷 25
2
where
𝑑𝑑𝑐𝑐 = Critical depth (m)
= Diameter or height of culvert (m)
Table 3.2: Culvert entry loss coefficient
Type of structure and design of entrance Coefficient ke

Concrete/fibre reinforced/poly pipe
Projecting from fill, socket end 0.2
Projecting from fill, square cut end 0.5
Headwall, with or without wingwalls:
• Socket end 0.2
• Rounded edge (radius = D/12) 0.2
• Square edge 0.5
End section conforming to fill slope (precast end unit) 0.5
Mitred/cut to conform to fill slope (field cut) 0.7
Reinforced concrete box
Headwall parallel to embankment (no wingwalls):
• Rounded on 3 edges (radius of 1/12 cell dimension) 0.2
• Square on 3 edges 0.5
Wingwalls at 30° to 75° to cell: 0.5
• Crown edge rounded (radius of 1/12 cell dimension) 0.2
• Crown edge square 0.4
Wingwalls at 10° to 30° to cell: square edged at crown 0.5
Wingwalls parallel (extension of sides) square edged at crown 0.7
Corrugated metal pipe
Headwall rounded edge 0.2
Headwall or headwall and wingwalls: square edged 0.5
End section conforming to fill slope (manufacturer end unit) 0.5
Mitred/cut to conform to fill slope 0.7
Projecting from fill (no headwall) 0.9
Notes:
The effect of wing walls reduces with multi-cell culverts.
For 3–6 cell culverts, assume entrance loss for wing walls 10° to 25° to barrel.
For culverts with more than 6 cells, assume wing walls parallel (extension of sides), regardless of actual wing walls.
Sources: Herr and Bossy (1965), Concrete Pipe Association of Australasia (2012), Schall et al. (2012).

3.9.2 Determination of Tailwater
The tailwater level is an important input to the culvert design process. Tailwater level (TW) can be calculated
using the method described in Section 2.3 – Fundamentals of Open Channel flow.
3.10 Design Procedures
The following procedure for standard culverts is acceptable for all but a very small percentage of culverts.
The procedure does not address the unsteady flow hydraulic effect of changing storage/ponding upstream
caused by the culvert/road embankment, which can modify the discharge through the culvert when flows are
changing significantly.
‘Special’ culverts are those that do not fit the normal design procedure and these should be referred to
specialist hydraulic engineers.
The designer should be familiar with the theory and equations in the previous sections before using these
procedures. Following the design method without an understanding of culvert hydraulics can result in an
inadequate, unsafe or costly structure.
Because of the difficulty in determining whether a culvert will operate under inlet or outlet control, the
standard procedure for calculating the headwater level is to check both inlet and outlet control cases and
adopt the scenario that results in the higher headwater level.
3.10.1 Hydraulic Design
The standard culvert design procedure is illustrated in the flow chart in Figure 3.9 and the following sub-
sections align with this figure.
For this method, calculations should be recorded on a form similar to the one shown in Figure 3.8. The
following procedure provides direction in completing this form.
3.10.2 Collect Design Data
Further to Section 2.3 – Fundamentals of Open Channel Flow, the data to be collected/determined, and then
recorded in the design workings includes:
• required flood immunity (ARI)
• design discharge (Q), check discharge (if required) and extreme event discharge (if available/applicable)
• tailwater height (TW), stream velocity (V) and Froude Number (Fr) for each discharge flow
• road shoulder height and any freeboard requirements
• maximum/allowable headwater height (AHW)
• proposed culvert slope (So)
• proposed culvert length (L)
• inlet/outlet invert heights
• maximum allowable stream velocity (Vmax) for outlet channel.

Figure 3.8: Form for culvert design calculations
Note: Tables and figures referenced in Table 3.8 have been drawn from the same source below.

Figure 3.9: Culvert design flow chart in steps 1 to 10
Collect Design Data

- Required flood immunity (ARI)
- Design discharge (Q) (also consider
A
extreme events).
- Stream velocity (V) and energy (Fr).
1
- Tailwater height (TW).
- Shoulder height and freeboard
requirements. Determine Controlling Headwater
- Maximum / allowable headwater (AHW).
- Proposed culvert slope (So).
- Proposed culvert length (L).
6 Is HWi < HWo?
If yes - HWo controls - go to step 7.
No
- Inlet / outlet invert heights. If no - HWi controls - go to step 8.

- Maximum allowable velocity (Vmax).
Select a Trial Culvert

Determine an initial trial culvert waterway Outlet Velocity – Outlet Control
area using A=Q/Vmax
2
- If Vmax is unknown, use Vmax = 2.0 - 2.5
m/s.
Use area (A) to determine initial trial
7 Calculate outlet velocity using Vo=Q/A.
Where A is depth of flow based on D, dc or

TW as appropriate.
culvert(s).
- Height of culvert (D) should be approx.
TW.
- Size and fit culvert to channel / road.
Outlet Velocity – Inlet Control
8
Calculate outlet velocity using Vo=Q/A.
Design Discharge for Trials
For culverts, divide design discharge Q by Use appropriate nomograph to determine
3
number of cells. part full area for pipes and for boxes, use
For box culverts, a ratio (Q/B) is also Manning’s to determine normal depth.
required. Divide Q per cell by box width
(B).
Outlet Flow Energy (Froude’s Number)

Determine Inlet Control Headwater Depth
Use appropriate nomograph to determine
HW/D.
9 Determine Fr for flow at outlet.
Compare to Fr for channel – is there a
4
Now determine HWi. hydraulic jump?
YES
Check if HWi, > AHW?
- If yes, return to step 2 and select a larger
culvert.
- If no go to step 5.
Design Check
If Vo < Vmax then OK, if not, check that
outlet protection can be provided.
Determine Outlet Control Headwater
Depth
Select entrance loss coefficient ke.
10 Several hydraulically acceptable options
are possible – review costs to determine
required option.
Use appropriate nomograph to determine If an alternative culvert configuration is to
H. be elevated return to step 2.
Use appropriate nomograph for pipes /
formula for boxes to determine dc. If dc
exceeds D, then dc = D A waterway area of culvert
AHW allowable headwater level at inlet (m)
5
Determine ½ (dc + D).
YES B width of culvert (m)
Is TW > D at the outlet? D diameter (pipe) / height (box) of culvert (m)
- If yes, use h0 = TW.
- If no, h0 is the larger of TW and ½ (dc + dc critical depth of flow in culvert (m)
D). Fr Froude’s Number
HWi generated headwater depth at inlet by Inlet control (m)
Determine length by slope, LSo. HWo generated headwater depth at inlet by Outlet control (m)
Determine HW0 = H + h0 – LS0 h0 adopted outlet depth (m)
ke entrance loss coefficient
Check if HW0 > AHW? L length of culvert (m)
- If yes, return to step 2 and select a larger
culvert. Q design discharge through culvert (m3/s)
- If no go to step 6. S0 slope of culvert
TW tailwater depth at outlet (m)
Vmax maximum average velocity in natural channel (m/s)
A
V0 calculated outlet velocity (m/s)

3.10.3 Select a Trial Culvert
To select an initial trial culvert or culverts, first determine an initial trial culvert waterway area (A) using
(Equation 1):
𝑄𝑄
𝐴𝐴 = �𝑉𝑉
𝑚𝑚𝑚𝑚𝑚𝑚
where
𝑄𝑄 = Design discharge (m3/s)
𝑉𝑉𝑚𝑚𝑚𝑚𝑚𝑚 = Maximum allowable outlet velocity (m/s)
Maximum allowable outlet velocity should be based on site conditions or if no data is available, assume an
outlet velocity of 2.0 to 2.5 m/s depending on channel conditions.
Alternatively, if the allowable head loss (H) through the culvert is known (from the allowable afflux to limit
flooding level impacts on upstream development), then the initial trial culvert area may be estimated using
Equation 26:
𝑄𝑄 26
𝐴𝐴 =
3.6 𝐻𝐻0.5
Choose culvert material, shape (pipe or box) and entrance type, allowing for minimum and maximum
allowable cover heights over the culvert.
Select a culvert trial size/configuration using the waterway area determined above. For example, the initial
trial culvert waterway area may have been determined as 1.25 m2. Review of product guides show that a
single barrel 900 mm reinforced concrete pipe (RCP) has a nominal diameter of 0.9 m and a waterway area
of 0.64 m2. Two barrels of 900 mm RCPs would have a waterway area of 1.27 m2, which is slightly larger
than the required waterway area and therefore suitable as an initial trial.
If possible, select a culvert size so that the obvert of the outlet is at, or just below, the tailwater level. This
ensures the probability of the culvert running full, which is usually desirable. This is not always practicable in
wide shallow flood plains, or in steep country. If the trial size is too large because of limited embankment
height or availability of size, multiple culverts may be used by dividing the discharge equally between the
number of cells used.
A further method is the use of culvert capacity charts (shown in Appendix D, Figure D 1 to Figure D 4), which
can be used to determine a preliminary culvert size and corresponding headwater depth for a known
discharge. To use these charts the value of L/(30So), where L is the culvert length (m) and So is the culvert
slope (m/m). The value of L/(30So) will dictate whether to adopt the inlet control line, outlet control line, or
interpolate between the two control lines drawn on the charts, to determine the headwater depth.
In locations where fauna passage is an issue, compare trial culvert size with the minimum fauna
requirements.
Several initial trial culverts of different size/configuration may be selected to start the design process.
3.10.4 Design Discharge for Trials
Where a single barrel culvert is not sufficient, multiple barrels or cells will be required and these
configurations constitute a parallel system.

Provided that all barrels of a multi-cell culvert are the same type, size and roughness (equal conveyance),
and also have the same invert levels/bed slope, flow will distribute evenly.
Design nomographs as shown in Appendix B and Appendix C, are based on a single barrel installation,
therefore divide the design discharge (Q) by the number of barrels for the trial culvert.
For boxes, a ratio (Q/B) is also required. This ratio is determined by dividing Q per cell by the nominal box
width (B).
Any proposed, multi-cell culvert that does not have all cells the same will require specialist advice as the
methodology presented in this Guide does not allow for these type of configurations. These culverts must be
referred to specialist hydraulic engineers.
3.10.5 Determine Inlet Control Headwater Depth
Using the trial culvert(s) from the previous section, find the HW/D value by use of the appropriate inlet control
nomograph (Appendix B, Figure B 1 to Figure B 4). In this case, tailwater conditions are to be neglected.
Three lines are presented in the nomographs for HW/D and the designer needs to select the appropriate line
based on:
• entrance type for pipe culverts
• wingwall flare angle for box culverts.
Headwater under inlet control conditions (HWi) is found by multiplying HW/D by the height of the culvert, D.
Check HWi against AHW and if HWi is greater (or much less) than AHW, try another size until HWi is
acceptable for inlet control before computing headwater for outlet control.
3.10.6 Determine Outlet Control Headwater Depth
Several steps are required to determine the headwater under outlet control conditions (HWo).
Firstly, determine the entrance loss coefficient, ke from Table 3.2.
Calculate the losses through the culvert, H, using the outlet control nomographs, Appendix B, Figure B 5 to
Figure B 10. In using these nomographs, some interpolation can be used for ke and L.
The next step is to determine the critical depth (dc) for the culvert. If dc exceeds D then take dc as D. For
pipe culverts, use nomographs as shown in Appendix C, Figure C 4 to Figure C 6 and for box culverts, use
Equation 27:
2�
𝑄𝑄 3 27
𝑑𝑑𝑐𝑐 = 0.467 � �
𝐵𝐵
where
𝑄𝑄 = 𝑄𝑄 per cell
Calculate (dc + D)/2.
Tailwater (TW) is required for determination of the next variable.
The next step is to establish the adopted outlet depth (ho) for design. Determine if TW > D:
• if yes, ho = TW
• if no, ho = the larger of TW and (dc + D)/2.

Multiply the proposed culvert length (L) and slope (So) to calculate LSo.
Headwater under outlet control conditions (HWo) is calculated using (Equation 24):
𝐻𝐻𝑊𝑊𝑜𝑜 = 𝐻𝐻 + ℎ𝑜𝑜 − 𝐿𝐿𝐿𝐿
Now check HWo against AHW and if HWo is greater than AHW, try another size until both HWi and HWo are
acceptable (< AHW).
3.10.7 Determine the Controlling Headwater
Compare the values of HWi and HWo. The higher headwater governs and indicates the flow control existing
under the given conditions for the trial size selected.
• if HWi > HWo the culvert is under inlet control and the controlling HW = HWi.
• if HWi < HWo the culvert is under outlet control and the controlling HW = HWo.
The controlling headway for trial should be clearly shown in design workings (i.e. design form).
There are now two checks that should be undertaken before completing trial design:
• If controlling HW is less than 1.2D, then the culvert is most likely not operating with a submerged inlet and
therefore may not be operating efficiently. Design should be revised using a different (slightly smaller)
culvert size/configuration.
• If controlling HW is less than 0.75D and the culvert is under outlet control, then the culvert may be flowing
only part-full and using (dc+D)/2 to calculate ho may not be giving accurate results. Design should be
revised using a different culvert size/configuration, where D ≈ TW.
Sometimes, calculations show HWi and HWo to be equal or nearly equal. In this instance, it is not clear
whether or not the culvert will perform under inlet control or outlet condition. In reality, the culvert could also
operate under both conditions (swap from one to the other) during the same rainfall event. Therefore, outlet
velocity calculations should be done for both inlet control and outlet control conditions with the higher velocity
(and associated control condition) being adopted.
3.10.8 Outlet Velocity – Outlet Control
The average outlet velocity for all culvert types can be calculated using (form of Equation 1):
𝑄𝑄�
𝑉𝑉𝑜𝑜 = 𝐴𝐴
where
𝑄𝑄 = Design discharge per culvert barrel/cell (m3/s)
𝐴𝐴 = Cross-sectional area of flow from culvert barrel/cell (m2)
The cross-sectional area of flow (A) depends on the flow depth at the outlet.
Flow depth will be one of the following:

• critical depth (dc) if the tailwater is below critical depth
• tailwater depth (TW) if the tailwater is between critical depth and the top of the barrel
• the height of the barrel (D) if the tailwater is above the top of the barrel.

See Figure 3.10 for guidance in determining the flow area for box culverts.
Figure 3.10: Flow area for box culverts
Determination of flow area for pipes is a little more difficult. The area can be determined using a CADD
package such as AutoCAD™ or calculated using Equation 28:
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 𝜋𝜋𝑅𝑅 2 28
when
y = D
Or Equation 29:
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 𝜋𝜋 𝑅𝑅2 − 1�2 𝑅𝑅2 (𝜃𝜃 − sin 𝜃𝜃) 29
where
𝑅𝑅 = Internal radius of pipe (m)
𝜃𝜃 = Angle in radians
Figure 3.11 provides guidance in determining the flow area for pipes.
Figure 3.11: Flow area for pipe culverts
See Appendix E.1.2 – Method A, for an example of using Manning’s Equation to determine flow
characteristics for an existing channel to provide conditions for the design of a circular pipe.

3.10.9 Outlet Velocity – Inlet Control
The average outlet velocity, Vo, for all culvert types can be calculated using (form of Equation 1):
𝑄𝑄�
𝑉𝑉𝑜𝑜 = 𝐴𝐴
where
𝑄𝑄 = Design discharge per culvert barrel/cell (m3)
𝐴𝐴 = Cross-sectional area of flow from culvert barrel/cell (m2)
The cross-sectional area of flow (A) depends on the flow depth at the outlet which can be approximated by
the normal depth of open-channel flow in the barrel, computed by Manning’s Equation for the discharge flow,
barrel size, roughness and slope of culvert selected.
Pipe culverts
For a pipe culvert, the culvert will not be flowing full at the outlet when under inlet control, meaning only a
part of the full waterway area will be used. To determine this ‘Part Area’, the designer needs to firstly
establish the relationship between ‘Full Flow’ (Qf) and ‘Part Flow’ (Qp) for the culvert trial where:
• ‘Part Flow’ (Qp) is the design discharge per cell
• ‘Full Flow’ (Qf) is the capacity of the trial culvert and its velocity can be determined using Figure C 1.
Now determine the ratio Qp/Qf.
Using this ratio and Figure C 2 determine the percentage factors for velocity (Vo), depth of flow (y), area of
flow (A) and hydraulic radius (R).
Figure C 2 is used as follows:

• Step 1 – plot the Qp/Qf. ratio on the x-axis.
• Step 2 – draw a line up to the Discharge line.
• Step 3 – draw a line left and right to both edges (the left edge being the y-axis).
• Step 4 – read off ‘% Depth of Flow’ from the y-axis.
• Step 5 – drop lines to the x-axis from each intersect between the horizontal line drawn in Step 3 and the
Velocity, Area curves.
• Step 6 – read off % Values from the x-axis for each of these hydraulic elements.
Now draw a table as shown in Figure 3.12 and enter values as follows:
• A is Qf
• B is Qp
• C is the outlet velocity determined using Figure C 1
• D is the nominal diameter of pipe
• E is the waterway area of pipe
• F is the hydraulic element values determined using Figure C 2.

The remaining spaces of the table, including the determination of Vo, are calculated by multiplying the ‘Full’
values by the relevant ‘Factor’.
Figure 3.12: Full flow/part flow table
Applied
Variable Pipe full Pipe part
factor
Q A B
Vo C
y D F
A E
Box culverts
As for a pipe culvert, a box culvert will not be flowing full at the outlet when under inlet control. To determine
the normal flow conditions, depth (y) and velocity (Vo), use Manning’s Equation to develop a Modified Stage-
Discharge Curve for the culvert cell.
Using the Modified Stage-Discharge Curve, the flow depth (y) and outlet velocity (Vo) can be read directly.
See Appendix E.1.3 – Method B, for an example of using Manning’s Equation to determine flow
characteristics for an existing channel to provide conditions for the design of a new culvert.
3.10.10 Outlet Flow Energy
To complete the hydraulic calculations for a trial culvert, the designer must determine Froude’s Number (Fr)
for the flow at the outlet. This is important as the designer can check this against Froude’s Number for the
channel flow and determine if a hydraulic jump will occur.
Froude’s Number can be determined using Equation 11:
𝐹𝐹𝑟𝑟 = 𝑄𝑄 �𝐵𝐵�(𝑔𝑔𝐴𝐴3 )
where B is derived from Equation 30:
𝐵𝐵 = 2�𝑦𝑦(𝐷𝐷 − 𝑦𝑦) 30
Figure 3.13 illustrates D, y and B.

Figure 3.13: Flow width in pipes
When a pipe or box culvert is flowing full, there is some pressure in the system and there is no water surface
(no B). Strictly, Froude’s Number does not apply, however for simplicity and that pressure is minimal, it is
considered suitable to calculate Fr at the moment just before the water surface touches the obvert of the
culvert and adopt this Fr value for the trial culvert. Therefore:
• for box culverts, use full waterway area and B = nominal width of box section
• for pipe culverts, use full waterway area, but assume B as 10% of D (≈ 99.7% of A).
3.10.11 Design Check
At this stage, the outlet velocity (Vo) should be checked against the maximum allowable stream velocity
(Vmax) for the outlet channel. Preferably, Vo should be less than Vmax. If Vo exceeds Vmax then the designer
must include suitable outlet protection for the culvert (see Section 3.5 – Culvert Size) if the trial is to be kept.
Where outlet protection is not suitable (size/cost/Vo > 5 m/s/Fr > 1.7) then a larger/wider culvert should be
trialled, or specialist advice/input obtained.
A worked example of the design steps, Section 3.10.2 – Collect Design Data to Section 3.10.11 – Design
Check, is shown in Section 3.15.1 – Example 1: Application of Culvert Design Procedure.
3.10.12 Practical Design
If a culvert trial design is considered unacceptable, the designer is required to redesign the culvert by trialling
another culvert size/configuration. The choices the designer has in determining a new trial culvert are:
• add another barrel or cell if channel width permits
• increase barrel or cell height if vertical clearance permits
• alter culvert slope (Note: desirable minimum is 0.25%)
• a combination of the above.
Where a designer decides to alter invert levels, it is preferable to lower inlets and leave outlets as close to
the natural surface as possible. Drop inlets (and structures) are better than hanging or buried outlets.
3.11 Blockage of Culverts
The likelihood of blockage should be considered for all culverts. Blockage can occur through siltation or by
debris (such as vegetation). To assist in preventing and/or minimising blockage, the culvert grades should
match the stream grade.

During the times of flood, the water contains silt, vegetation and other debris from the catchment. The
designer should therefore consider the impacts on the effects of the culvert becoming blocked. Allowance
for blockage is commonly provided in the sizing of the culvert (see Section 3.7.4 – Siltation and Blockage).
Blockage reduces the waterway area of the culvert and therefore adversely affects the capacity/performance
of the culvert. The result of blockage is typically:
• an increase in upstream peak water levels/flooding
• an increased potential for water to overtop the road
• an increased risk of failure to the road embankment/culvert.
Silt deposits and some debris can be detected and removed during normal maintenance processes at times
outside of rainfall events and therefore these deposits/debris would not impede any flows; however the
effectiveness of this measure is highly dependent on the efficiency of maintenance in the area.
Where debris blockage during an event is considered likely (typical in catchments that contain significant
woody riparian vegetation), larger culvert sizes may be required, in accordance with the extent of adverse
impacts that could occur to the roadway or to surrounding properties.
Where large or long branches and/or tree trunks are a possibility, sloped extensions to piers, as shown in
Figure 3.14 can be used to ‘turn’ long objects into the culvert barrel.
Figure 3.14: Flood water flowing into box culverts
Designers must consider the potential for and impacts of blockage for each catchment/culvert installation
and, where impacts are considered unacceptable, include appropriate mitigating treatments in the design.
3.12 Consideration of Large or Extreme Events
Irrespective of the design ARI of a culvert, consideration should be given to the impacts of flood events which
are larger than the design event (see AGRD Part 5 – Section 4.9). In determining the appropriate degree of
consideration, issues such as the frequency and consequence of overtopping of the culvert should be
addressed.

3.13 Culvert Outlet Protection
Outlet protection is required in situations where:

• Outlet velocity exceeds the scour velocity of the bed or bank material.
• An unprotected channel bend exists within a short distance of the culvert outlet.
• The outlet channel and banks are actively eroding.
• If an erodible channel bank exists less than 10 to 13 times the pipe diameter downstream of the outlet
and this bank is in line with the outlet jet (i.e. likely to be eroded by the outlet jet), the bank should be
adequately protected to control any undesirable damage as a result of the outlet jetting.
The most appropriate outlet protection is determined by considering the hydraulic performance of the outlet
in the prevailing stream environment.
At outlet structures, the best hydraulic performance is obtained when the confining sidewalls are parallel and
the distribution of flow across the channel is uniform.
Culverts, however, are generally narrower than the natural waterway and a transition section is required to
return the flow to the natural channel. When culvert outlet velocities are high, additional measures at the
outlets may prove to be necessary for energy dissipation. Section 3.15 – Worked Examples provides
additional information on the more specialised energy dissipators.
To check whether standard inlet and outlet structures with headwalls, wingwalls, aprons and cut-off walls are
adequate, the outlet velocity for the culvert requires examination with respect to:
• natural environment (soil and vegetation cover)
• size of peak flow
• duration of large flows.
If outlet velocities exceed the acceptable limits, it may be necessary to check for potential bed scour
problems.
The key design parameters for sizing the culvert outlet protection are the average rock size (d50) and the
length of rock protection (L).
For a single pipe outlet protection the recommended minimum rock size (d50) and length (L) of rock
protection downstream of culvert outlets may be determined from Figure 3.15. The length of an extended
concrete apron can be determined in the same way as for a rock apron. When the tailwater is less than one-
half of the pipe diameter (D), the minimum rock length should be adequate. When the tailwater is between
D/2 and D, energy dissipation decreases and the length of the protection needs to be extended by twice the
minimum length. The minimum recommended width of the protection can be determined using Figure 3.16.
For multi-pipe outlets, the energy generated is higher than for single pipe outlets and the recommended
minimum rock size (d50) and length (L) of the rock protection can be determined from Figure 3.17. The rock
pad length should be doubled if it is required to minimise the risk to scour downstream of the culvert outlet.
The multi-pipe outlet protection is a rectangular shape with the minimum recommended width of protection
being the width of the pipe outlets plus 0.6 m.
When the width of the channel is less than the recommended width of the rock protection, then the rock
protection shall extend up the banks of the channel to a height equal to the obvert of the outlet.
The depth of the rock apron should be twice the size of rock used for the apron.
For Froude Number (Fr) values greater than 1.7 and outlet velocities equal to or greater than 5.0 m/s, an
alternative energy dissipator structure should be considered.

Figure 3.15: Single pipe outlet minimum rock size and length of apron
Source: Catchments and Creeks (2011).
Figure 3.16: Rock apron detail for single pipe outlet
L
La
La/5
B Width of apron
1 La/5
5

Figure 3.17: Minimum rock size and length of apron for a multi-pipe outlet
Note: The pipe diameter is for a single pipe diameter of the multi-pipe outlet, e.g. for an outlet of 2 x 1200 mm pipes, use
a pipe diameter of 1200 mm.
Source: Catchments and Creeks (2011).
A worked example of to determine the dimensions of a rock apron is shown in Section 3.15.2 – Example 2:
Culvert Outlet Protection Example.
3.14 Culvert End Treatments
3.14.1 General
Culvert end treatments, which include wingwalls, cut-off walls and anchorages, and erosion control
measures, may be required to perform one or more of the following functions:
• prevent fill from encroaching on the culvert opening
• prevent erosion of the fill and adjacent channel
• prevent undermining of culvert ends
• help anchor the structure to the ground
• inhibit seepage and piping through the bedding and backfill
• meet traffic safety requirements
• improve the appearance of large culverts
• resist hydraulic uplift forces on corrugated metal pipe culverts
• strengthen the ends of large flexible culverts, especially those with mitred or skewed ends.

Cut-off walls in the form of a vertical wall constructed below the end, or outside edge, of the apron of a
culvert must always be provided at culvert inlets and outlets to prevent undermining and piping failure. For
corrugated metal pipe culverts, the cut-off walls also act to counteract uplift at the culvert inlet.
The treatments at culvert endwalls are a compromise between hydraulic efficiency, traffic safety, and
economy. A flared approach with rounded edges at the pipe entry is the best hydraulic inlet, but such inlets
are too costly for small pipes.
As the size of the culvert increases, protection at the endwall as described in AGRD Part 5 – Section 3.7.6, is
needed. Culvert ends mitred to conform to the fill slope for safety reasons, have higher entry losses than
square headwalls (see Table 3.2), and therefore the headwater depths should be increased accordingly.
Outlet structures should prevent scour damage to the road embankment, the downstream channel and
adjacent property; as well as retaining the road filling and supporting the end of the culvert.
Treatments of culverts to improve roadside safety are discussed in AGRD Part 6 (Austroads 2010c) which
covers provision of roadside barriers across culverts.
3.14.2 Slope Faced
The ends of small culverts that run parallel to the road (e.g. under driveways) should be treated with either a
sloped end, or sloped end with grate, depending on the road agency requirements, to improve roadside
safety for errant vehicles.
Depending on road agency requirements, culverts that carry water under the road should either be extended,
so that the end wall is further from the road, e.g. beyond the clear zone, or have sloped ends that match the
batter slope. Larger culverts may require a grate to be fitted to protect the public. Designers should see
AGRD Part 3 (Austroads 2010b) and AGRD Part 6 (Austroads 2010c) for further information.
3.14.3 Traversable Endwalls
Figure 3.18 shows an example of a traversable end wall for pipes or box culverts up to 1000 mm in diameter
or height. The treatment has rails fitted to prevent errant vehicles approaching the culvert end-on from
snagging a vertical headwall. It is important to note that these end treatments can affect hydraulic
performance and increase maintenance difficulty.

Figure 3.18: Example of a slope faced and slope faced with grate culvert endwall
3.15 Worked Examples
3.15.1 Example 1: Application of Culvert Design Procedure
This example describes a culvert design procedure for a road crossing of a small channel (see Figure 3.19),
based on the following requirements and data:
Design discharge (Q) is 6.25 m3/s for a 50 year ARI event.
Figure 3.19: Channel cross-section
Channel slope = 0.65%

Maximum stream velocity = 2.5 m/s
Side slopes = 1:1 n = 0.06
Channel bed = 5 m
Tailwater has been determined as 0.96 m

Figure 3.20: Road cross-section for worked example
5m
Ht = 125.65 m
Pavement thickness
Freeboard = 100 mm = 460 mm
3
Q50 = 6.25 m /s
RCP culvert Outlet invert

Ht. = 123.85 m Ht. = 123.75 m
Culvert length = 15.6 m
This example can be solved following the design procedure described in Section 3.10 – Design Procedures.
Reference should also be made to Figure 3.9 – Culvert design flow chart.
Step 1 – Collect design data
Use the hydraulic calculation form (Figure 3.8) and fill out the known information.
The shoulder height can be calculated based on the centreline height, road crossfall and traffic lane width.
Shoulder height = Centreline height – road crossfall
= 125.65 m – 5 m x 3%
= 125.5 m
A 100 mm freeboard below shoulder height is required, therefore the Allowable Headwater depth (AHW) will
be:
AHW = Shoulder height – invert height at outlet – freeboard
AHW = 125.50 – 123.85 – 0.100 = 1.55 m
The tailwater depth can be obtained from the Stage-Discharge curve shown in Figure 3.21 below. For
details on how to determine a Stage-Discharge curve see Section 2.3 – Fundamentals of Open Channel
flow.

Figure 3.21: Stage-Discharge curve for culvert design example
14
12
10
8
6.25 m3/s
6
2
0.695 m
0
0 0.2 0.4 0.6 0.8 1 1.2
Depth of flow (m)
Complete top section of the design form as follows:
Figure 3.22: Geometric detail of key parameters for worked example
Step 2 – Select a trial culvert (see Section 3.10.3 – Select a Trial Culvert)
Firstly, determine an indicative waterway area of culvert(s) based on Q and maximum allowable velocity.
A = Q/V
= 6.25/2.5
= 2.50 m2
Use area to select an initial culvert configuration that meets the area required. Size and fit the culvert to the
channel/road.
At this stage, and based on preliminary considerations given above, a RCP Culvert ≥ 1200 will not fit as the
height of the barrel will be within the pavement. That is:
• Maximum barrel height = 125.5 – 0.460 – 123.85 = 1.19 m
• Initial trial size selection area = 2.50 m2
• Try 3/1050 RCP, Area = 2.60 m2 (Check pipe OD ≤ 1.19 – OK).

Step 3 – Design discharge for trials (see Section 3.10.4 – Design Discharge for Trials)
For multiple cells culvert determine Q per cell:
Q per cell = 6.25/3 = 2.08 m3/s
Step 4 – Determine inlet control headwater depth (see Section 3.10.5 – Determine Inlet Control
Headwater Depth)
• Analyse the culvert assuming inlet control.
• Use the nomograph Figure B 2 to determine HW/D and then calculate HWi.
• Q = 2.08 m3/2, D = 1050 mm, from nomograph, HW/D = 1.33.
• HWi = HW/D x D = 1.33 x 1.05 = HWi = 1.40 m < AHW (1.55 m).
Step 5 – Determine outlet control headwater depth (see Section 3.10.6 – Determine Outlet Control
Headwater Depth)
• Analyse the culvert assuming outlet control.
• Select the entrance loss coefficient in accordance with the project description (see Figure 3.20) above.
The culvert entrance type is a square edge with headwall, therefore
ke = 0.5 (see Table 3.2).
• Determine head (H) for RCP culvert flowing full from Figure B 6.
• Plot L = 15.6 m on ke = 0.5, plot D = 1050 and draw line.
• Locate pivot on turning line.
• Plot Q = 2.08 m3/s and draw line crossing pivot on turning line to Head line.
• Read H = 0.50 m.
• Determine the critical depth in a pipe, dc from the nomograph Figure B 6.
• dc = 0.81 m.
• Calculate (dc + D)/2 = (0.81 + 1.05)/2 = 0.93 m.
• Determine the tailwater. TW was earlier determined as 0.695 m (Figure 3.21).
• Determine ho, which is the greater of TW and (dc + D)/2, therefore ho = 0.93 m.
• Calculate length x slope = 15.6 m x 0.0065 m/m = 0.101 m.
• Determine HWO = H + ho – LSo = 0.5 + 0.93 – 0.101 = HWO = 1.33 m < AHW (1.55 m).
Step 6 – Determine the controlling headwater (HW) (see Section 3.10.7 – Determine the Controlling
Headwater)
• The controlling headwater is determined from the larger of HWi = 1.40 m and HWO = 1.33 m
• Therefore, the three 1050 mm diameter RCP pipe culverts are operating under inlet control.
Step 7 – Calculate outlet velocity – inlet control pipe (see Section 3.10.9 – Outlet Velocity – Outlet
control)
• Determine relationship of part full pipe (Qp) and full flow pipe (Qf) = (Qp/Qf).
• Using the nomograph in Figure C 1, Qf = 2.4 m3/s and Vf = 2.5 m/s.
• Qp/Qf = 2.08/2.4 = 87%.

• Using the nomograph in Figure C 2 obtain the following factors:

– 71.5% depth of flow, therefore y = 1.05 x 71.5% = 0.75 m
– 77% waterway area, therefore A = 0.87 m2 x 77% = 0.67 m2
– 112% outlet velocity, therefore V = 2.75 x 112% = 3.08 m/s.
This work can be tabulated as follows (see Figure 3.12):
Variable Full flow Factor Part flow

Q 2.40 – 2.08
V 2.75 1.12 3.08
Y 1.05 0.715 0.75
A 0.867 0.77 0.67
Step 8 – Calculate outlet flow energy
Calculate the Froude Number (Fr) (Equation 11):
𝑉𝑉𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜
𝐹𝐹𝑟𝑟 =
�𝑔𝑔�𝐴𝐴�𝐵𝐵�
In order to calculate the Froude Number, calculate B (top width of flow) using Equation 30:
𝐵𝐵 = 2 �𝑦𝑦 ( 𝐷𝐷 − 𝑦𝑦)
where
𝑦𝑦 = depth flow = 0.715 x 1.05 = 0.75 m
𝐵𝐵 = 2 x √(0.75 x (1.05 – 0.75))
𝐵𝐵 = 0.95 m
Therefore:
3.08
𝐹𝐹𝑟𝑟 = = 1.17 (supercritical flow)
�9.81 �0.67�0.95�
Since Fr < 1.7 a rock pad will be suitable.
Step 9 – Design check (see Section 3.10.11)
The outlet velocity = 3.08 m/s and is therefore greater than allowable stream velocity = 2.5 m/s. It is
recommended that a suitable type of erosion control measure be determined.

Figure 3.23: Worksheet for example culvert design

3.15.2 Example 2: Culvert Outlet Protection
See Section 3.13 – Culvert Outlet Protection for information on culvert outlet protection.
The following example demonstrates the procedure to determine the dimensions of a rock apron required to
provide outlet protection against outlet velocities that are higher than acceptable.
For velocities in excess of 5 m/s, the use of energy dissipators should be considered and specialist advice
should be acquired.
This example will demonstrate the design of an erosion control measure for a 1200 mm diameter RCPs (inlet
control) that have been designed to convey a total discharge of 2.25 m3/s. The maximum allowable stream
velocity for the channel is 2.0 m/s. The depth and velocity at the outlet are 64% of the depth of full flow and
2.97 m/s (per cell) respectively.
Step 1
First determine a suitable type of control measure by calculating the Froude Number Equation 11:
𝑉𝑉
𝐹𝐹𝑟𝑟 =
�𝑔𝑔�𝐴𝐴�𝐵𝐵�
In order to calculate the Froude Number, it is necessary to calculate area of flow at the outlet (Ao) and the top
width of flow (B).
• Ao = Q/Vo
• Ao = 2.25/2.97
• Ao = 0.76 m2.
Using Equation 30:
𝐵𝐵 = 2 �𝑦𝑦 ( 𝑑𝑑 − 𝑦𝑦)
where
𝑦𝑦 = 64% depth of full flow = 0.4 x 1.2 = 0.77 m
𝐵𝐵 = 2 �0.77 (1.2 − 0.77) = 1.15 m
2.97
= 1.16
𝐹𝐹𝑟𝑟 =
�9.81�0.76�1.15�
Since 𝐹𝐹𝑟𝑟 < 1.7, a rock pad will be suitable

Step 2
Determine stone size and length of protection from Figure 3.24 as shown below:
Figure 3.24: Application of Figure 3.15 to worked example
2.97
1200
Source: Adapted from Catchments and Creeks (2011).
From Figure 3.24 above:

• Length = 4D = 4 x 1.2 = 4.8 m. If the tailwater was between D/2 and D, the length should be doubled.
Therefore, the recommended length = 4.8 m x 2 = 9.6 m. (Adopted for this example).
• Also from Figure 3.24, the stone size required, d50 = 300 mm.
Step 3
Determine rock apron dimensions based on a 5:1 expansion ratio. See Figure 3.25.
The outside edge of the normal concrete apron will extend 1.8 m beyond the outlet with a 30° wingwall flare.
The overall width between the wingwalls is therefore= 1.2 + 2(1.8 x tan 30°) = 3.3 m.
The width of the rock apron at its downstream extremity, at a 5:1 expansion ratio is therefore
= 3.3 + 2 (9.6/5) = 7.1 m. See Figure 3.25.

Figure 3.25: Solution to example for culvert outlet protection
L = 11.4 m
La = 9.6 m
9.6/5
B = 1.2 m 3.3 m 7.1 m
9.6/5
1
5

4. Floodways
4.1 Introduction
A floodway consists of a length of pavement on a typically level grade that is designed to be overtopped by
floodwater during relatively low average recurrence interval (ARI) floods. Sections of road likely to be
overtopped should also be designed as a floodway. The objective of a floodway is to provide for floodwater
to be conveyed across the road under controlled conditions at designated places which are specifically
designed and protected to minimise damage to the road asset.
Floodways must be identified by advance warning signs, and by water depth markers along the floodway.
An example of a longitudinal shape and cross-section of a floodway is shown in Figure 4.1.
Floodways are commonly utilised in rural roadways where:

• traffic volumes are relatively low
• some interruption to access is acceptable
• flooding is infrequent
• it is impractical or uneconomical to construct a bridge or culvert
• in appropriate circumstances, as a means of protecting large culverts or a bridge from being washed
away during above-design flood conditions.
The decision to provide a floodway should be supported by adequate consideration of alternatives.
4.2 Design Considerations
4.2.1 General
Design considerations include:

• hydraulic design to ensure that there is sufficient discharge capacity to safely convey the design flow at
the designed depth and flow velocity, thus ensuring the safety of road users (e.g. against washouts) and
keeping maintenance at an acceptable level
• appropriate pavement design to resist against the adverse effects of submergence and high flow velocity
• providing adequate embankment protection against scouring to ensure the stability of the floodway
• upstream flooding and pondage
• providing road signage and road delineation, see AS 1742.2.

Figure 4.1: Example of a longitudinal shape and cross–section of a floodway
General design considerations are that floodways:

• will generally have a low flood immunity but should have short periods of closure depending on the
importance of the road link
• may require costly batter protection and therefore a higher level road together with a larger culvert or
bridge may be a more cost-effective option
• should be designed so that the road level provides the required level of access and trafficability for the
chosen design ARI
• may be supplemented by anti-ponding culvert/s placed at the low point of the natural surface to take the
perennial or low ARI flows under the road. The invert level of the culvert(s) should be such that the
culvert will run full before water flows over the road. In addition, these relatively smaller relieving culverts
may be prone to blockage by debris.

4.2.2 Geometric
Designers should be aware of the following factors that influence the geometric design of floodways:
• It is important that adequate approach sight distance be provided to allow drivers time to recognise water
over the road and to stop.
• As a general principle, floodways should be designed so that the depth of water over the road is as
uniform as possible over the flooded section. The longitudinal gradient of the central part of the floodway
must therefore be level so that it acts as a weir and the possibility of a driver, unexpectedly encountering
deeper water, and possibly stalling or being swept downstream, is avoided.
• The risk of cars being washed off floodways is an important consideration. Risk increases when total
head of water (static plus velocity head) over the road surface is 300 mm or more. Road closure should
be considered when the total head at any point on the road surface reaches 300 mm. Designers should
see reports on the Australian Rainfall & Runoff website, www.arr.org.au.
• The flows over the floodway should also be assessed for suitability for people who may attempt to
traverse the flow on foot (e.g. vehicle occupants leaving their vehicle) and guidance on criteria can be
obtained in Engineers Australia (2010).
• Floodways should not be placed on horizontal curves because:
– there are problems in defining the edge of the pavement for motorists
– any superelevation may change the normal flow distribution i.e. push more water to the non-
superelevated sections of road
– the water depth will be deeper on one side of the road than on the other due to superelevation and
there is a possibility of the high side being trafficable when the low side is not, thus creating a safety
problem.
• The length of a floodway should be limited at about 300 m, so that drivers do not become disorientated
when confronted with wide open stretches of water.
• Where a proposed floodway would be longer than 300 m, it is recommended that the proposed floodway
be broken into shorter lengths by providing sections of road that are raised above the maximum flood
level.
• Avoid the design of superelevated pavements within floodways if possible, as these can result in lateral
depth variations and surface debris on the floodway surface that can compromise safety. Residual silt
and gravel on floodways after isolated rainfall events can result in a hazardous surface and cause serious
injury to vehicle occupants. It is imperative that relevant agencies design new floodways to minimise the
risk and establish effective maintenance programs to remove debris and gravel from floodways.
4.2.3 Environmental
With respect to environmental considerations, floodways:

• may offer environmental advantages over culverts or bridges as they will tend to spread flows more
widely. This means that the risk of scour to waterways and surrounding land is generally reduced
because flow is less concentrated
• should be designed so that they are not covered by water from ponding or backwater for any significant
period of time after a flood event
• should consider fauna movement and fish migration; however if:
– only floodplain culverts are provided (i.e. culverts located away from a watercourse channel),
designers only need to consider terrestrial movement
– fish migration is expected to occur across the floodplain during times of flood, a check should be done
on allowable flow velocities.

4.2.4 Hydraulic
A floodway consists not only of the roadway embankment but also waterway openings which may be
required for one or more of the following functions:
• to reduce the afflux or rise in water level upstream due to the obstruction (embankment)
• to raise the tailwater level so that less batter protection is required on the downstream side e.g. grass
instead of concrete
• to act as anti-ponding structures for low flow stream conditions.
Flow over roadways may be free flow or submerged flow. In the initial stages of overtopping a low tailwater
usually exists and free flow occurs. Under these circumstances flow passes through critical depth over the
road and the discharge is determined by flood levels upstream.
Free flow may be either:

• plunging flow which flows over the shoulder and down the downstream face of the embankment. The
flow then penetrates the tailwater surface producing a submerged hydraulic jump on the downstream
slope. Velocities are likely to be high and erosive
• surface flow which separates from the surface of the road embankment and rides over the surface of the
tailwater. This flow will have less erosion potential downstream.
Submerged flow occurs when the discharge is controlled by the tailwater level as well as the headwater
levels. This occurs when the depth of flow over the road is everywhere greater than the critical depth.
Where a floodway is constructed above ground level, it may or may not act as a control on flow, depending
upon the headwater and tailwater conditions. In the initial stages of overtopping of the road a low tailwater
condition usually exists and a free outfall condition occurs.
Typical velocities of flow over a floodway are shown in Figure 4.2. This figure illustrates the conditions that
result from rising tailwater (e.g. submerged hydraulic jump).

Figure 4.2: Indicative velocities of flow over a typical floodway

4.3 Hydraulic Design
4.3.1 Flow Over the Road
Equation 31, the broad crested weir formula used for flow over a road, is:
𝐶𝐶𝑠𝑠 31
𝑄𝑄 = 𝐶𝐶𝑓𝑓 𝐿𝐿𝐻𝐻1.5 � �
𝐶𝐶𝑓𝑓
where
𝑄𝑄 = Discharge over floodway (m3/s)
𝐶𝐶𝑓𝑓 = Coefficient of discharge ‘free’ flow
𝐶𝐶𝑠𝑠 = Coefficient of discharge flow with submergence
𝐿𝐿 = Length of floodway (m)
𝐻𝐻 = Specific head or specific energy (m)
With reference to Figure 4.3
ℎ = Level difference between the floodway crown and the upstream water surface
𝑉𝑉 = Approach velocity of the stream
𝑙𝑙 = Top width of road formation
The flow over the floodway may be calculated by means of the following design procedure:
1. Calculate the Stage-Discharge curve (height versus discharge) for the unrestricted section, from open
channel hydraulics (see Section 2 – Open Drains and Channels).
2. Select a tailwater level and a headwater level (given h and V) from the Stage-Discharge curve.
3. Calculate H/l with Equation 32.
𝐻𝐻 32
𝑙𝑙
where
𝑉𝑉 2
𝐻𝐻 = ℎ+ = 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 ℎ𝑒𝑒𝑒𝑒𝑒𝑒
2𝑔𝑔
4. From Figure 4.3

• For H/l < 0.15 use curve A to obtain value of Cf.
• For H/l > 0.15 (usual case) use curve B to obtain the free flow coefficient of discharge Cf.
• Calculate D/H x 100 and use curve C to obtain the submergence factor Cs/Cf.
• Calculate the discharge over the road using the broad crested weir formula, Equation 31.

Figure 4.3: Discharge coefficients for flow over floodways
Percentage Submergence D/H x 100
Note: Use this curve for

submergence factor D/H >
0.76
Submergence Factor Cs/Cf

3/2
Overall Q = Cf L H Cs/Cf
Note: Use this curve for free flow

coefficients with H/l ratios < 0.15
Head on Roadway H (m)

Source DTMR (2010).

4.3.2 Full Floodway Calculations
Floodways incorporating culverts and bridges will require calculations in addition to those above.
The basic principle is that the total flow over the road and through the waterway structures equals the flow
downstream in the unrestricted channel.
Given the many combinations of headwater and tailwater possible, it is necessary to fix at least one of these
parameters for each design calculation.
With flow over the road, the issue of whether or not road batter protection is needed becomes important as
the calculations require the tailwater to be fixed when the flood is at the point of overtopping the road.
Guidelines relating to the need for protection are given in Section 4.5 – Floodway Protection. The tailwater
level for floodways with other than grassed batters is usually more than 300 mm and up to 600–700 mm
below the crown of the road when the road is about to be overtopped.
Full floodway design must satisfy Case A and Case B conditions.
Case A: When the flood is at the point of overtopping the road:

• The tailwater level is not below the level specified for the type of protection to be adopted.
• Velocities through the bridges and/or culverts are acceptable. (Scouring velocities may be acceptable, if
additional outlet protection or the formation of scour holes is acceptable). At the point of overtopping, the
velocity through the culvert or bridge will be at its maximum.
Figure 4.4 illustrates results of calculations performed using the steps below for an actual floodway.
The step-by-step design procedure for this case follows:

1. Calculate the rating curve (height versus discharge) for the unrestricted channel in order to determine the
tailwater level.
2. Fix the level of the road as a first trial. The initial level of the road may be based on trafficability, e.g. for
required trafficability in an ARI 20 year flood, the initial road level may be the level of the ARI 20 year
flood level in the unrestricted channel. This allows culverts to be designed for a maximum head of 300
mm with the tailwater at the crown level of the road. It will be found that a lower road level will give less
fill in the embankment, but require more culverts under the road in the ARI 20 year flood.
3. Fix the headwater level at the crown of the road (or the highest edge if superelevated).
4. Find the velocity Vx through a suitable major culvert/bridge with the headwater in (3) above and the
tailwater corresponding to the type of floodway protection to be adopted. Say this tailwater level is RLx on
Figure 4.4 Case A, corresponding to a total discharge, Qx in the unrestricted channel.
5. With this velocity Vx find the total area of waterway required from Total Area Required = Qx/Vx and select
other culverts to give this total.
6. For the fixed headwater and the tailwater level, RLx, calculate the actual flow through each culvert and
bridge and total the discharges to give Qy which may or may not be equal to Qx because of the different
sized openings.
7. Adjust the culvert and bridge areas so that with the fixed headwater and tailwater level, RLx, the total flow
equals Qx. With progressive adjustments the total discharge may be Qy or Qz and so on.

Figure 4.4: Example floodway calculations
Note that the initial assumption of all the waterways having the same velocity is only approximate, as
different size and shape openings have different hydraulic efficiency. This is why the same head will give
different velocities.
The structures giving Qx are to be considered minimum requirements for the floodway regardless of the
requirements calculated for a peak flood above the roadway, as all overtopping floods pass through this
stage.
It should also be noted that the design velocities through culverts are normally in the range 1.8 m/s to 3.0
m/s with 2.4 m/s commonly adopted, depending on the scour resistance of the surface material downstream.
Higher and lower velocities have been adopted. Section 3 – Culverts discusses the hydraulic design of
culverts and methods for erosion control are contained in AGRD Part 5 – Section 3.6.
Case B: At the peak of the flood above the floodway:

• velocities through the waterway structures and over the floodway are acceptable
• afflux caused by the floodway is acceptable.

The step-by-step design procedure for this case follows:
At the peak of the overtopping flood, allowance for flow over the road must be made and afflux calculated for
the design flood (e.g. ARI 20 year in the example in Case A above).
The flow over the road is added to that through the culvert(s) and bridge(s) and the calculations initially
include the structures from Case A. Generally, a range of floods should be considered. With reference to
Figure 4.4 and Case A and Case B, the procedure is:
1. Find the tailwater level corresponding to the design flood discharge from the rating curve for the
unrestricted channel.
2. Select a headwater level slightly above the tailwater level in (1), say 0.1 m above, and calculate the flow
over the road and through each culvert and/or bridge with this headwater-tailwater combination. Add
these flows together to obtain a total flow.
3. Compare this flow with the design flood discharge. If this flow is smaller, a greater head/afflux is required
to ensure that the check flood is passed over the road and through the waterways.
4. Adopt a higher headwater and obtain the total flow and obtain a range of flood discharges for flow over
the road and through the waterway.
5. Plot these flows against their respective headwater levels. Join these points to make the curve shown on
Figure 4.4 Case B.
6. From the curve, find the headwater which gives the required design flood discharge.
7. If the afflux is acceptable, calculations are complete. If the afflux is too high, additional bridge or culvert
waterways are required. Waterways from Case A are not to be reduced.
8. Calculate the afflux for the check flood (usually of ARI 50, or ARI 100 years).
Because of flow over the road, maximum velocities are obtained from Case B. Both velocity and afflux
requirements are to be met.
4.4 Time of Submergence/Closure
4.4.1 Introduction
Where a crossing is designed for overtopping, it is usually important that an estimate be made of the
frequency and duration of the periods during which the crossing will be submerged as well as the times it is
closed to traffic due to flooding. The time of submergence is of importance with respect to stability of
embankments and pavements and consequent maintenance costs. The time of closure is of importance in
consideration of acceptable delays to traffic, since this is as important for defining the disruption to transport
for the flood immunity.
4.4.2 Time of Submergence
Time of submergence (ToS) is defined as the period of time that the road is inundated by flood water, no
matter the depth.
The time of submergence is expressed in one of two main ways. The first is the time of submergence during
a major flood, e.g. the number of hours of submergence during the flood with an ARI of 50 years or a major
historical flood. The second is as the average annual time of submergence, abbreviated as AAToS. This is
the average time per year that the road is submerged, expressed as hours or days per year.
Both expressions are an indication of the frequency and duration of submergence. For example, a crossing
that is submerged frequently for short periods of time may have a similar AAToS as one that is submerged
less frequently for longer times. To understand the concept and to compare options, three parameters are
needed, namely:

• flood immunity
• ToS for the selected flood event
• AAToS.
The absolute value of the time of submergence is not particularly useful by itself for a crossing, but it is useful
to compare several crossings or upgrading options.
4.4.3 Time of Closure
Time of closure (ToC) is similar to time of submergence. However this parameter takes account of the fact
that some inundation of the road may not necessarily close the road, though there may be some hazard in
travelling on the inundated road.
Different types of vehicles can travel on roads with different amounts of inundation with large and heavy
vehicles capable of travelling in water that is deeper and flowing at a higher velocity than a light car.
There are different definitions of the flow conditions when the road is closed by flood waters. Depending on
the velocity of flow and length of floodway, the road may be trafficable with some passage to vehicles. In
some circumstances it may be desirable to close the road at lower depths and, in exceptional cases, the
road should be closed if there is any water over the road (e.g. long floodways and high risk crossings).
A typical adopted limit is when the total head (static plus velocity) across the road exceeds 300 mm above
the crest of a carriageway with a two-way crossfall or above the highest edge of a carriageway with a one-
way crossfall. That is, using Equation 33:
𝑉𝑉 2 33
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 ℎ𝑒𝑒𝑒𝑒𝑒𝑒 (𝐻𝐻) = 𝑑𝑑 +
2𝑔𝑔
where
𝑑𝑑 = Depth of inundation (m)
𝑉𝑉 = Flow velocity (m/s)
Similar to AAToS, there is an average annual time of closure, abbreviated as AAToC.
4.4.4 Issues Related to Times
The concept of time of closure, in association with the time of submergence, adds some additional
information to the question of flood immunity.
Large and flat catchments will respond more slowly than small and steep catchments so the time of
submergence for these catchments will generally be longer for equivalent flood immunity.
In small steep catchments, where the response time is short, the time of submergence will be low, even for a
crossing with low flood immunity.
In small steep catchments, and also for urban catchments where the response time is short, floods may
occur and then recede very quickly. In this case the disruption to traffic may be minimal. Similarly in small
catchments, the depth of inundation may be low even if the flood immunity is low. In this situation, the time
of submergence may be short so the cost of disruption may be very low and the additional cost of providing
for a higher level of flood immunity may not be justified.

On the other hand, large catchments that have a long response time may inundate the road for extended
periods of time. In this case the cost of disruption may be very high and a higher level of flood immunity may
be more easily justified.
There are occasions where the road crosses a tributary close to the junction with a major stream. If the road
is inundated by both the local catchment run-off and backwater from the major stream, the time of
submergence may be quite different for the two flood mechanisms. The local catchment run-off, from a small
catchment, may have a short time of submergence, while the larger catchment may inundate the road for
longer periods of time. Local residents can often identify these two sources of closure and they may be quite
distinct.
When calculating the time of submergence or closure in these situations, the calculation must consider both
sources of inundation as well as the risk that the inundation is independent, in which case the times must be
added together.
When the flood immunity is very low, the time of closure may not be represented accurately by the flood
event hydrographs. In this case, long term stream flow records may need to be analysed.
These situations can occur where the road is inundated for months at a time during the wet season or even
sometimes the road may be inundated by tides. The calculation in this case is very difficult and should use
either a stream gauge or calculated continuous discharge records.
4.4.5 Calculation of Time of Submergence or Closure
There are two approaches to calculating ToS/ToC of a road by flooding:

• analysis of flood hydrographs, the method that is most commonly adopted
• analysis based on stream records of the crossing at the point of submergence or closure where there is a
stream gauge located at the bridge site, which is not a common occurrence.
A hydrograph of the flood may be obtained from actual measurements at the stream crossing or, for
ungauged streams, by the use of a run-off routing method or rainfall run-off model.
The time of submergence or closure may then be calculated by drawing a horizontal line on the hydrograph
at the flow capacity level and measuring the time for which the flow is above this level.
If there is a stream gauge located on the stream at or very near the bridge, this gauge can be used to
analyse the ToS/ToC. Whilst this does not commonly occur, it is the most accurate method for calculating
the ToS/ToC. Use of a stream gauge is also useful since calculated hydrographs are prepared considering
the maximum discharge, and the length of the hydrograph may often vary, and longer duration floods with a
lower peak discharge may be more critical in consideration of time of submergence or closure than the actual
peak discharge.
If there is no stream gauge located close to the crossing, it is possible to calculate a continuous record of
discharge from a hydrologic model and to use this sequence of flows in the calculation of AAToS/AAToC in
exactly the same way as a stream flow record would be used. It should be noted that this approach would
not be commonly attempted.
The procedure for applying the stream gauge records is as follows:

1. Calculate the discharge that will just inundate the road (for time of submergence) or close the road (for
time of closure).
2. Calculate the total period of time where the road is submerged (or closed) from the complete record of the
stream gauge.
3. Calculate the average duration per year that the road is submerged (or closed). This is the AAToS (or
AAToC).

This procedure does not provide the duration of submergence (or closure) for a particular design flood event.
As with all hydrologic analysis, it is important that the period of the stream flow record is sufficiently long to
provide a representative sample of flow at the bridge. If the period of record is too short, or the period of
record is not representative, the result from the calculation may not be reliable.
It is hard to define the period of record that would be adequate for this analysis since it depends on the
variability at the site and the flood immunity considered as well as other conditions. However it is likely that if
there is less than 20 years of record at the site the result may not be reliable.
If there is only a short record, careful analysis can be used to assess the best means of extracting useful
information from the record. It is very likely that a stream gauge will not be located near the bridge, and the
theoretical method discussed here must be used.
The first step in this process is to calculate design flood hydrographs for the crossing site. This can be done
with a catchment hydrologic model, such as RORB or RAFTS™. It is noted that the Rational Method
calculates only the flood peak discharge so is not suitable for application of time of submergence or time of
closure calculations.
It is also noted that the actual time of inundation of the road depends on the flood levels and not necessarily
the discharge so in some cases, especially on large flat floodplains, the hydrographs of water levels may not
be exactly the same as the hydrographs of flood discharges, so this issue may need to be considered in
some cases.
However the normal procedure is to calculate flood hydrographs of discharge with a catchment model and
use these directly for the calculation.
To calculate the time of submergence or time of closure, the following is required:

• a hydrograph of the flood, which may be obtained from actual measurements at the stream crossing or,
for ungauged streams, by the use of a run-off routing method or synthetic hydrograph method. An
experienced hydraulic engineer is required to undertake these analyses.
• the flow capacity of the crossing at the point of submergence or closure.
The time of submergence or closure may then be calculated by drawing a horizontal line on the hydrograph
at the flow capacity level and measuring the time for which the flow is above this level.
4.4.6 Procedure for Estimating Average Annual Time of Closure (AAToC)
The procedure for calculating the average annual time of closure (AAToC) is as follows:
Step 1
Determine the ARI of the flood for which the stream crossing is trafficable (i.e. floodway with or without a
culvert or bridge).
Step 2
Determine the times of closure ty for a series of floods greater than the trafficable capacity flood, and the
ARIs of each of these floods.
Estimate tmax by extrapolating a graph of time of closure versus ARI or by estimating the probable maximum
flood for the catchment. It should be noted that the probable maximum time of closure cannot be
disregarded in the calculation of the AAToC.

Step 3
Calculate the probability FT(t) of the road being closed for each ARI y year flood (Equation 34):
1 34
𝐹𝐹𝑇𝑇 (𝑡𝑡) = 1 −
𝑦𝑦
Step 4
Using the Times of Closure, ty and the probability, FT(t) of the road being closed for each ARI y year flood,
draw the cumulative probability distribution.
This distribution gives the probability of the road being closed in any year for less than t hours.
Step 5
Determine the probability density function (Equation 35):
∆𝐹𝐹𝑟𝑟 (𝑡𝑡) 35
𝑓𝑓𝑇𝑇 (𝑡𝑡) =
∆𝑡𝑡
This is the slope of the line connecting each point on the cumulative probability distribution.
Step 6
Determine AAToC as (Equation 36):
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = � Δ𝑝𝑝 𝑥𝑥 𝑡𝑡̅ 36
where
Δ𝑝𝑝 = Area of each rectangle in the probability density function
The centroidal distance of each rectangle from the fT(t) axis of the probability
𝑡𝑡̅ =
density function. The area of each rectangle in the probability density function
The same procedure is used for AAToS except that a lower flow is used to derive the critical flood level.
4.5 Floodway Protection
4.5.1 Introduction
Selection of the form or type of protection of floodways against scour is governed by:
• whether flow across the floodway is free or submerged
• under free flow conditions, whether plunging or surface flow occurs downstream from the floodway.
The tailwater level when the flood is at the point of overtopping the road usually controls the degree of
protection required for a particular floodway. Therefore, the cost of providing adequate bridge and/or culvert
waterways to raise the tailwater to a high enough level to require minimum protection becomes a prime
consideration as well as the cost of the protection itself.

Floodway protection is considered in two categories:

• minimum protection such as grassed batters
• other than grassed batters.
The tailwater level for floodways with grassed batters is usually not more than 300 mm below the
downstream edge of the road formation when overtopping first occurs.
Overtopping flows of long duration and at frequent intervals may cause pavement failures and softening of
the embankment, thus aggravating any tendency to scour. Even with a high tailwater when the flood is at
the point of overtopping the road, the Times of Submergence may indicate more elaborate protection than
natural grass.
For low standard floodways in flat country the bed levels surveyed downstream should extend sufficiently to
allow design checks to be made regarding the possibility of ponding at the site due to unforeseen natural
deposits, farm dams and access roads downstream.
4.5.2 Floodways with Grassed Batters
Grass in this type of floodway is defined as turf or seeded grass. This is further discussed in the AGRD Part
5 – Section 3.6.2.
Because the physical properties of grass such as species, stiffness, cover density and rooting pattern vary
with soil type and climate, only general guidelines based on constructed floodways in Queensland are
possible.
Floodways with grassed batters should have the following features:

• Bitumen seal or asphalt pavements with concrete or other rigid margins/shoulders (stone pitching, cement
stabilised gravel, etc.) containing the bitumen in place.
• Alternatively, concrete blocks/nib walls along the top edges of the formation with a bitumen seal or
asphalt pavement between them may be constructed. These containing blocks may be as simple as 10%
by volume cement stabilised gravel strips 600 mm wide at the top by 180 mm deep. Concrete pavements
instead of bituminous types will, of course, cover the full width of the formation.
• Culverts under the floodway section to raise the tailwater to not more than 300 mm below the
downstream edge of the road formation when overtopping first occurs.
• Overtopping occurs for a period of less than 12 hours in a 50 year ARI flood. However, the type of
material in the embankment and its saturated strength may require reduction of this allowable time of
submergence. Conversely there are some low floodways which withstand submergence for much longer.
For this type of protection, it is desirable to have good grass cover when the overtopping flood occurs. This
in turn requires an ability to maintain grass cover during the dry season.
4.5.3 Other Floodways
Experience has shown that most failures of floodways with downstream batter protection commences by
scouring at the downstream aprons and/or the downstream edge of the road formation, and therefore a
reduction in standard in these areas should not be considered in order to reduce costs.
Five types of floodway which have performed satisfactorily are described in this section with their associated
limitations noted. All these types of floodways should have the following general features:

• provision of culverts to raise the tailwater to not more than 600 mm below the downstream edge of the
road formation when overtopping first occurs (actual range 300–700 mm). It is important to note that this
afflux may not be acceptable in some areas and non-standard additional outlet protection at culverts may
also be necessary
• full protection of the top surface of the road formation, as for floodways with grassed batters
• protection of at least the downstream batter. Although not clearly defined, it appears that protection of the
upstream batters may only be required in floodways of low flood immunity in major streams. As a
precaution where only downstream protection is adopted, protection for a distance of about 3.0 m on
each side of major culverts on the upstream side may be placed to offset possible scour due to
turbulence from the mixing of longitudinal and direct flows at the culvert inlets
• provision of adequate downstream aprons. For height of road embankment, H, equal or more than 2.0 m,
the downstream apron should extend at least 1.5H metres away from the toe of the embankment. For H
less than 2.0 m, the downstream apron should extend at least H away from the toe of the embankment
unless otherwise specified
• provision of weepholes in the downstream rigid protection to relieve hydrostatic pressure. These
weepholes are normally about 90 mm diameter at 1.8 m (maximum) centres with 300 x 300 x 150 no-
fines concrete blocks behind the weepholes. The weepholes should be placed about 300 mm above the
apron level or just above long standing water level if higher. The more porous types of protection such as
rock on filter cloths or layers and some cement stabilised gravels, depending on the grading, do not
require weepholes.
A brief description of the five successful types of floodway protection follows and sectional details of these
types are shown in Figure 4.5 and Figure 4.6. These types are not in any order of preference and
comparative cost comparisons should be made where more than one suitable protection is considered.
Type 1 floodway
This type of reinforced concrete floodway has been constructed in many areas of Australia. The
reinforcement selected should not only satisfy strength requirements, but also prevent temperature and
shrinkage cracks.
Where cut-off walls have been used without the downstream apron, failures have occurred.
With a suitable width of downstream apron and weepholes, the Type 1 floodway is recommended as suitable
for all crossings where other than grass protection is required, cost permitting.
Type 2 floodway
This is an example of a reinforced concrete floodway where the tailwater depth is uncertain but probably
quite low (perhaps 700 mm or more below the downstream edge of the formation when the flood begins to
overtop the road). No adverse reports are known. Costs are higher than Type 1 Floodways.
Type 3 floodway
The increased use of stone mattresses and gabions has given confidence in this type of construction. Rock
fill with size 70–100 mm and median diameter of 85 mm is considered adequate for most uses.
One of the important features of this type of treatment is that localised forces can be generated at the sharp
change in slope between the road pavement and batter. A rounded shoulder (3 m radius) with a 1:3 slope
should be adopted. Significant forces can act on upstands near the shoulder, therefore any mattress or
dumped riprap protection should not be allowed to project above the slab. This also applies to Type 5.

Type 4 floodway
This type of protection incorporating a bituminous seal is probably the lowest cost of the types shown, but its
use is limited.
It should only be used where:

• fill height is not higher than 900 mm
• tailwater at overtopping is not more than 300 mm below the crown of the road
• time of submergence is small (hours).
Type 5 floodway
This type of floodway (Figure 4.6) shows a typical floodway with mattress and dumped riprap protection,
which was developed following failure and subsequent research carried out by the Centre for Water
Research (Patterson & Abercromby 1986). The important features that should be taken into account in the
design of these floodways are:
• Localised uplift forces can be generated at the sharp change in slope between the road pavement and
batter. A rounded shoulder (3 m radius) with a 1:3 slope should be adopted, as this configuration does
not display negative pressures and the net force on the bed is at all times downwards.
• Significant forces can act on upstands near the shoulder. Therefore, mattresses and dumped riprap
protection should not be allowed to project above the concrete slab or wall shown in Figure 4.6 – Type 5.
• Ingress of water at the upstream edge of the concrete wall shown on Figure 4.6 may result in significant
forces acting on the wall and material downstream of the wall. Leakage at the edge should be prevented
and adhesion between the concrete and stabilised base course maximised.
• As the riprap consists of rock with grading requirements, it may have limited application in some
jurisdictions where supplies of such rock are scarce where floodways are constructed (see Type 3
floodway). Further details of the required grading and riprap thickness may be obtained from Table 4.1
and Table 4.2.

Figure 4.5: Downstream floodway protection (Types 1, 2 & 3)
TYPE 1 – Concrete protection

Most common type.
Widely used.
TYPE 2 – Concrete protection

Type performs well,
but need to justify
cost. Requires
specialist design.
TYPE 3 – Rock mattress protection

Mattresses must be
pinned/anchored.
Consider a cut-off
wall.
Cut off walls may
not be necessary as
mattresses, usually
achieve their
optimum position
with a little scour by
dropping into a
scour proof position

Figure 4.6: Downstream floodway protection (Types 4 & 5)
TYPE 4 – Bitumen seal protection

Common, low cost type.
Suitable for low velocities
over floodway situations
only.
TYPE 5 – Dumped rock (riprap) protection

Variation to Type 3 where
mattresses are not
readily available.
Consider a cut-off wall.
Table 4.1: Design of rock slope protection
Velocity (m/s) Class of rock protection (tonne) Section thickness (m)

<2 None –
2.0–2.6 Facing 0.5
2.6–2.9 Light 0.75
2.9–3.9 ¼ 1.00
3.9–4.5 ½ 1.25
4.5–5.1 1.0 1.60
5.1–5.7 2.0 2.00
5.7–6.4 4.0 2.50
> 6.4 Special –
Source: MRWA (2006).

Table 4.2: Standard classes of rock slope protection
Minimum
Rock class Rock size (m) Rock mass percentage of
rock larger than
Facing 0.4 100 0
0.3 35 50
0.15 2.5 90
Light 0.55 250 0
0.40 100 50
0.20 10 90
¼ tonne 0.75 500 0
0.55 250 50
0.30 35 90
½ tonne 0.90 1000 0
0.70 450 50
0.40 100 90
1 tonne 1.15 2000 0
0.90 1000 50
0.55 500 90
2 tonne 1.45 4000 0
1.15 2000 50
0.75 500 90
4 tonne 1.80 8000 0
1.45 4000 50
0.90 1000 90
Source: MRWA (2006).
4.6 Worked Examples
4.6.1 Example 1: Floodway Design
The task for this example is to design a floodway with 20 year ARI trafficability over a natural open channel
approximately trapezoidal in shape. The floodway would be approximately 90 to 100 m long and for cost
reasons road batters will be grass covered only for protection against scour.
Preliminary considerations:
• Because the batters will be protected by grass only, culverts will be required to build up the tailwater to
not more than 300 mm below the edge of the downstream shoulder when overtopping of the road first
occurs. Allowing for crossfall, there will be a head of 450 mm and a velocity of about 2.30 to 2.45 m/s
through the culverts if this minimum tailwater is adopted. Is this acceptable on this job? (This will be
answered in the example). There is a need to confirm total head over the road is less than 300 mm for an
ARI 20 year flood.
• For 20 year ARI trafficability, the floodway level should be at the 20 year unrestricted flood level to allow
the maximum amount of water over the road and save on culvert and overall costs.
In general it is only in very long floodways with very little velocity in the open channel and/or where costly
protection is unavoidable, that increasing culvert requirements by raising the road, thus decreasing the flow
over the road (to the extreme of a flood free road) may reduce the overall cost of the job.

Step 1
List all relevant criteria:

• required standard: trafficable in a 20 year ARI flood
• time of closure: maximum of one day in a 50 year ARI flood. (Calculated hydrograph shows this maximum
only a matter of hours – not included here)
• batter protection: grass
• width of floodway: 10 m
• road crossfall: 3%.
Step 2
Calculate the rating curve for the unrestricted channel.
Using open channel hydraulic calculations (as discussed in Section 2 – Open Drains and Channels):
• Q50 = 162 m³/s @ Height 322.76 m
• Q20 = 130.4 m³/s @ Height 322.58 m (V = 0.68 m/s)
• Q10 = 108.1 m³/s @ Height 322.44 m
• Q5 = 70 m³/s @ Height 322.13 m.
Step 3
Adopt a road level and calculate the maximum allowable depth of water over the road.
Adopt the road level at the unrestricted 20 year ARI flood level and show the cross-sectional details in
Figure 4.7.
Figure 4.7: Floodway cross-section
V2
2g
H d Q20 Ht 322.58 m
V
10.0 m
Ht 320 55 m
From Section 4.2 – Design Considerations, closure to traffic occurs when (using Equation 33):
𝑉𝑉 2
𝐻𝐻 = 𝑑𝑑 + > 300 𝑚𝑚𝑚𝑚
2𝑔𝑔
The maximum allowable depth of water over the road, y, occurs when H = 0.300 m.

Therefore:
𝑉𝑉 2 0.682
𝑑𝑑 = 𝐻𝐻 − = 0.3 −
2𝑔𝑔 2 𝑥𝑥 9.81
𝑑𝑑 = 0.28 𝑚𝑚
Step 4
Calculate the discharge over the road and through the culverts in a 20 year ARI flood.
The total discharge over the road and through the culverts must equal the discharge in the unrestricted
channel downstream with flow at height 322.58 m.
i.e. Qtot = QR + QC = 130.4 m³/s
Flow over the road using Equation 31:
𝐶𝐶𝑠𝑠
𝑄𝑄 = 𝐶𝐶𝑓𝑓 𝐿𝐿𝐻𝐻1.5 � �
𝐶𝐶𝑓𝑓
where
𝐶𝐶𝑓𝑓 = Coefficient of discharge
From Figure 4.3, Chart A:
H/l 0.30/10 = 0.03, which is < 0.15
Therefore:
Cf = 1.674
L = Length of floodway = 94 m
H = 0.30 m
Therefore:
Q = 1.674 x 94 x 0.31.5
Q = 25.9 m3/s
Discharge through culverts:
QC = Qtot – Q = 130.4 – 25.9 = 104.5 m³/s
Therefore, the culverts are required to take a discharge of 104.5 m³/s, operating under a head of 0.30 m and
under outlet control for this design condition.

Step 5
Detailed culvert design.
Proceed with the design of culverts to take 104.5 m³/s.
Height of culvert opening, D, from Equation 37:
D = Crown level of road – crossfall – minimum fill above culvert – thickness of deck 37
slab – invert level
D = 322.58 – 5 x 0.03 – 0.100 – 0.180 – 320.55 = 1.60 m
For culvert design:

• outlet control with H = 0.30 m
• tailwater at height 322.58 m, HW at height 322.88 m
• invert at height 320.50 m, say.
From the culvert design procedure in Section 3.10 – Design Procedures:

• 13/2700 x 1500 mm RCBC have a capacity of 104 m³/s with outlet velocity = 1.98 m/s.
• Adopt 13/2700 x 1500 mm RCBC with a floodway height 322.58 m and length 94 m.
Step 6
Check for culvert requirements when the flood is at the point of overtopping the road.
By trial and error, calculate flow through the culvert at the point where the headwater height is 322.58 m.
Confirm that flow velocity through the culvert is less than 2.5 m/s.
Step 7
Confirm that velocity criteria are met.
Step 8
Calculate afflux for ARI 50 year flow and confirm that criteria are met.
4.6.2 Example 2: Floodway Design
A new floodway is to be designed for a country road. A sketch plan of the site is shown in Figure 4.8. The
road agency has decided that the road should be flood-free up to the five year ARI. Hydrological estimates
are Q5 = 7 m3/s and Q100 = 20 m3/s.
Using the method in Section 3.10 – Design Procedures, it has been accepted that four cells of 1500 mm x
600 mm box culverts will be used to carry the low flow. Under a total discharge of 20 m3/s the culverts would
carry 8.4 m3/s, and the remaining 11.6 m3/s is to flow across the floodway.
Task
The downstream flood plain has an average slope of 1 in 333, and a cross-section as shown on Figure 4.8.

Analyse the stream using Manning’s Equation (see Section 2.3.3 – Fundamental Equations, Equation 3,
repeated here):
2� 1�
𝑅𝑅 3 𝑆𝑆 2
𝑉𝑉 =
𝑛𝑛
Estimate Manning’s roughness factors for the channel, use Table 2.1 or Table 2.2 as appropriate.
Main channel
For a natural stream channel in the form of a minor stream with a fairly regular cross-section and some
weeds and light brush on the banks, Manning’s coefficient will be in the range
n = 0.035 – 0.05.
Select a value within the range, say n = 0.04.
Flood plain
For a natural stream channel in a flood plain through pasture, no brush, high grass; say
n = 0.035 – 0.05. Because the flow will be shallow relative to the grass height, select the highest roughness
value, n = 0.05.
The slope S is:

1
𝑆𝑆 = = 0.003
333
∴ √𝑠𝑠 = 0.0548
Analysis of stream cross-section
To analyse the stream cross-section, divide it into three segments related to the hydraulic roughness.
Try total depth = 0.85 m and apply Manning’s Equation to each segment of the longitudinal section of the
flood plain.

Figure 4.8: Details of example floodway design

Segment 1
A= 60.25 x 0.25 = 15.06 m2

P = 60 + 0.25 x 2.236 = 60.56 m
𝑃𝑃 15.06
R = 𝐴𝐴 = 60.56 = 0.249 m
2�
𝑅𝑅 3 = 0.395
Segment 2
A = 11.2 x 0.6 + 12.4 x 0.25 = 9.82 m2

P = 10 + 2 x 0.6 x 2.236 = 12.683 m
𝑃𝑃 9.82
R = 𝐴𝐴 = 12.683 = 0.774 m
2�
𝑅𝑅 3 = 0.843
Segment 3
A = 21.85 x 0.25 = 5.46 m2

P = 21.6 + 0.25 x 2.236 = 22.16 m
𝑃𝑃 5.46
R = 𝐴𝐴 = 22.16 = 0.246 m
2�
𝑅𝑅 3 = 0.393
2� 1�
𝐴𝐴𝑖𝑖 𝑅𝑅 3 𝑆𝑆 2
� 𝑄𝑄 = �
𝑛𝑛𝑖𝑖
15.06 𝑥𝑥 0.395 𝑥𝑥 0.0548 9.82 𝑥𝑥 0.843 𝑥𝑥 0.0548 5.46 𝑥𝑥 0.393 𝑥𝑥 0.0548

= + +
0.05 0.04 0.05
� Q = 20.2 m3 /s
Q100 = 20 m3/s which is less than channel capacity of 20.2 m3/s. Therefore, accept 0.85 metres as
downstream depth.
D = d – y = 0.85 – (0.6 + 0.2) = 0.05 m.
Depth of flow
Generally the flow will pass through a critical depth close to the highest point on the road. In this case, with
crossfall in the direction of flow, the critical depth will occur near the upstream edge. Water depths up to 300
mm are generally regarded as trafficable for cars (see Section 4.2 – Design Considerations). As a first trial,
set the critical depth at 0.2 m. If H is not known, and the approach velocity is small, an approximation can be
made of the total head on the weir by:
H = 1.5h.
Therefore,
H = 1.5 x 0.2 = 0.3 m.

Follow the procedure in Section 4.3 – Hydraulic Design.
Step 1
𝐻𝐻 0.3
= = 0.03
𝑊𝑊 10
As this value is less than 0.15, use the design graph A in Figure 4.3.
Step 2
For H = 0.3, read off Cf=1.677 from Figure 4.9 (part of Figure 4.3).
Figure 4.9: Head on roadway
Step 3
The ratio of tailwater to upstream head is:
𝐻𝐻 0.05
= = 0.166 ≤ 0.7
𝐷𝐷 0.3
Therefore inundation does not occur, and
𝐶𝐶𝑠𝑠
= 1.0
𝐶𝐶𝑓𝑓

Step 4
Calculate minimum length of the level section of floodway:
𝐶𝐶𝑠𝑠
𝑄𝑄 = 𝐶𝐶𝑓𝑓 𝐿𝐿 𝐻𝐻1.5
𝐶𝐶𝑓𝑓
𝑄𝑄
𝐿𝐿𝑚𝑚𝑚𝑚𝑚𝑚 =
𝐶𝐶𝑠𝑠 1.5
𝐶𝐶𝑓𝑓 𝐻𝐻
𝐶𝐶𝑓𝑓
11.6
𝐿𝐿𝑚𝑚𝑚𝑚𝑚𝑚 = = 42.0 𝑚𝑚
1.677 x 1.0 x 0.31.5
Round up to 50 m.
Check results
Repeat design steps using actual average velocity:
Flow area (A) and depth.
A = L x h = 50 x 0.2 = 10 m2
𝑄𝑄 11.6
𝑉𝑉 = = = 1.16 𝑚𝑚/𝑠𝑠
𝐴𝐴 10
𝑉𝑉 2
𝐻𝐻 = 𝑑𝑑 +
2𝑔𝑔
1.162
𝐻𝐻 = 0.2 + = 0.268 𝑚𝑚
2 x 9.81
Original assumption was H = 0.3 m.
Check capacity, Q.
𝐶𝐶𝑠𝑠
𝑄𝑄 = 𝐶𝐶𝑓𝑓 𝐿𝐿 𝐻𝐻1.5
𝐶𝐶𝑓𝑓
𝑄𝑄 = 1.675 x 1 𝑥𝑥 50 x 0.2681.5
𝑄𝑄 = 11.62 𝑚𝑚3 ⁄𝑠𝑠
Acceptance criteria
Hydraulically, the floodway capacity nearly equals the design discharge, and is acceptable.
Comparing the results with the trafficability criteria (see Section 4.2 – Design Considerations) – the flow
depth of 0.268 m is less than 0.3 m and the average velocity is about 1.2 m/s, with the V x D < 0.40, and is
therefore acceptable.

References
Australian Transport Council 2011, National road safety strategy 2011-2020, Australian Transport Council,
Canberra, ACT.
Austroads 2009, Guide to road design: part 6A: pedestrian and cyclist paths, AGRD06A/09, Austroads,
Sydney, NSW.
Austroads 2010a, Guide to road design: part 1: introduction to road design, 3rd edn, AGRD01/10, Austroads,
Sydney, NSW.
Austroads 2010b, Guide to road design: part 3: geometric design, AGRD03/10, Austroads, Sydney, NSW.
Austroads 2010c, Guide to road design: part 6: roadside design, safety and barriers, AGRD06/10, Austroads,
Sydney, NSW.
Austroads 2013a, Guide to road design: part 5: drainage – general and hydrology considerations,
AGRD05/13, Austroads, Sydney, NSW.
Austroads 2013a, Guide to road design: part 5A: drainage – road surface, networks, basins and subsurface,
AGRD05A/13, Austroads, Sydney, NSW.
Catchments and Creeks 2011, Use of rock in engineering, webpage, Catchments and Creeks Pty Ltd, Ferny
Hills, Qld, viewed 17 December 2012, <http://www.catchmentsandcreeks.com.au/esc_rock_sizing.html>.
Concrete Pipe Association of Australasia 2012, Hydraulics of precast concrete conduits, CPAA, St Leonards,
NSW.
Department of Environment and Resource Management 2012, Soil conservation measures: design manual
for Queensland, DERM, Brisbane, Qld, viewed 21 March 2012,
<http://www.derm.qld.gov.au/land/management/erosion/index.html>.
Department of Natural Resources and Water 2007, Queensland urban drainage manual, vol. 1, 2nd edn,
DNRW, Brisbane, Qld.
Department of Transport and Main Roads 2010, Road drainage manual, DTMR, Brisbane, Qld.
Engineers Australia 2010, Australian rainfall and runoff: revision project 10: appropriate safety criteria for
people: stage 1 report, Engineers Australia, Barton, ACT.
Franzini, JB & Finnemore, EJ 1997, Fluid mechanics with engineering applications, 9th edn, McGraw-Hill,
USA.
Herr, LA & Bossy, HG 1965, Hydraulic charts for the selection of highway culverts, hydraulic engineering
circular no. 5, Federal Highway Administration, Washington, DC, USA.
Herr, LA & Bossy, HG 1972, Capacity charts for the hydraulic design of highway culverts, hydraulic
engineering circular no. 10, Federal Highway Administration, Washington, DC, USA.
Kouwen, N, Li, RM & Simons, DB 1981, ‘Flow resistance in vegetated waterways’, Transactions of the
American Society of Agricultural Engineers, vol. 24, no. 3, pp. 684-98.
Main Roads Western Australia 2006, Floodway design guide, MRWA, Perth, WA.
Patterson, JC & Abercromby, AK 1986, Floodway model tests, environmental dynamics report no. W-86-027,
University of Western Australia, Centre for Water Research, Nedlands, WA.
Peterka, AJ 1964, Hydraulic design of stilling basins and energy dissipaters, engineering monograph no.25,
Bureau of Reclamation, Department of Interior, Washington, DC, USA.
Roads and Traffic Authority 1993, Road design guide, RTA, Rosebery, NSW.
Thompson, PL & Kilgore, RT 2006, Hydraulic design of energy dissipators for culverts and channels, 3rd edn,
hydraulic engineering circular no. 14, Federal Highway Administration, Washington, DC, USA.

Schall, JD, Thompson, PL, Zerges, SM, Kilgore, RT & Morris, JL 2012, Hydraulic design of highway culverts,
3rd edn, Federal Highway Administration, Washington, DC, USA.
Underwood, R 1995, Road engineering practice, Macmillan Education Australia, Melbourne, Vic.
VicRoads 1998, VicRoads standard drawings for roadworks: culverts, webpage, VicRoads, Kew, Vic, viewed
23 May 2012,
<http://www.vicroads.vic.gov.au/Home/Moreinfoandservices/RoadManagementAndDesign/DesignStandar
dsManualsNotes/RoadDesign/GuidelinesStandardDrawingsForRoadworks.htm#culverts>.
VicRoads 2003, Road design guidelines: part 7: drainage, VicRoads, Kew, Vic.
Standards Australia
AS 1742.2-2009, Manual of uniform traffic control devices: part 2: traffic control devices for general use.
AS/NZS 2041.1-2011, Buried corrugated metal structures: design methods.
AS/NZS 3725-2007, Design for installation of buried concrete pipes.

Appendix A Vegetal Retardance Curves
The vegetal charts are for determination of Manning’s n value. The charts provide guidance for heavy, thick
levels of grass to short lengths of grass (see Table 2.3 for the degree of retardance). The use of these
charts requires the flow depth to be greater than the height of the vegetation as appropriate for each
respective chart.
Figure A 1: Solution of Manning’s Equation for vegetal retardance A

Figure A 2: Solution of Manning’s Equation for vegetal retardance B

Figure A 3: Solution of Manning’s Equation for vegetal retardance C

Figure A 4: Solution of Manning’s Equation for vegetal retardance D

Figure A 5: Solution of Manning’s Equation for vegetal retardance E

Appendix B Inlet and Outlet Control Nomographs
B.1 General
The headwater discharge charts shown in Figure B 1 and Figure B 2 are for the design of concrete culverts
with inlet control. Corresponding charts for corrugated steel and metal culverts are shown in Figure B 3 and
Figure B 4. Design charts for other types of culverts are given in the hydraulic charts for the selection of
highway culverts (Herr and Bossy 1965) and are usually available from the manufacturers of the culverts.
Figure B 5 and Figure B 6 show the outlet head for concrete box culverts and pipes whilst Figure B 7,
Figure B 8, Figure B 9 and Figure B 10 are outlet control nomographs for various corrugated metal structures
running full, including circular and pipe arches, and small and large corrugation size.

B.2 Inlet Control

Figure B 1: Headwater depth for box culvert with inlet control

Figure B 6: Headwater depth for concrete pipe culverts with inlet control

Figure B 7: Inlet control nomograph – CMP and SPCSP culvert

Figure B 8: Inlet control nomograph – CMPA and SPCSPA culvert

B.3 Outlet Control

Figure B 9: Head for concrete box culverts flowing full with outlet control (n=0.012)

Figure B 10: Head for concrete pipes flowing full with outlet control (n=0.012)

Figure B 11: Outlet control nomograph – corrugated metal pipe (CMP) flowing full (n=0.024)

Figure B 12: Outlet control nomograph – structural plate corrugated steel pipe (SPCSP) flowing full (n=0.0328 to
0.0302)

Figure B 13: Outlet control nomograph – corrugated steel pipe arch (CSPA) flowing full (n=0.024)

Figure B 14: Outlet control nomograph – structural plate corrugated steel pipe arch (SPCSPA) flowing full
(n=0.0327 to 0.0306)

Appendix C Velocity and Critical Depth
C.1 General
The outlet velocity for culverts operating under inlet control may be assumed to equal the normal velocity in
open channel in the barrel as computed by Manning’s Equation for the discharge, barrel geometry,
roughness and slope of the barrel.
The velocity may be readily obtained by the use of Figure C 1 and Figure C 2 for round pipes flowing full and
part-full respectively, and Manning’s Equation for rectangular culverts. Critical depth may be obtained from
Figure C 3, Figure C 4, Figure C 5 and Figure C 6.
Figure C 1: Discharge and velocity in round pipes flowing full

Figure C 2: Velocity and discharge in part-full pipes
Af = Area of pipe
df
dp
Ap

Figure C 3: Critical depth in a rectangular section

Figure C 4: Critical depth in a circular pipe

Figure C 5: Critical depth in a corrugated steel pipe arch (CPSA) (hc)
A = Cross-sectional area per cell

Figure C 6: Critical depth in a structural plate corrugated steel pipe arch (SPCSPA) (hc)
A = Cross-sectional area per cell

Appendix D Culvert Capacity
D.1 CULVERT CAPACITY
The following charts Figure D 1 to Figure D 4 are based on a Manning’s n = 0.012.
D.1.1 Socket End of Spigot and Socket Pipes
Figure D 1: Concrete pipes 450 mm – 1650 mm diameter where TW < dc
Reading ACCURACY LIMITATIONS

(1) For L/30So ≤Inlet control value, determine HW from Inlet control The following limitations apply for accurate determination of
line. HW
(2) For L/30So between Inlet and Outlet control values, interpolate (1) TW < dc
(2) For L/30So ≤ Outlet control value, HW < 2D.
to determine HW.
(3) For Outlet control value < L/30So < 1.5 x Outlet
(3) For L/30So in the range 1.0 to 1.5 x Outlet control value, control value, HW < 1.3D.
extrapolate and determine HW.
Source: Herr and Bossy (1972).

D.1.2 Socket End of Spigot and Socket Pipes
Figure D 2: Concrete pipes 1500 mm – 4500 mm diameter where TW < dc
READING ACCURACY LIMITATIONS

(1) For L/30So ≤ Inlet control value, determine HW from Inlet The following limitations apply for accurate determination of HW:
control line. (1) TW < dC
(2) For L/30So between Inlet and Outlet control values, (2) For L/30So ≤ Outlet control value, HW < 2D.
interpolate to determine HW. (3) For Outlet control value < L/30So < 1.5 x Outlet control value,
(3) For L/30So in the range 1.0 to 1.5 x Outlet control value, HW < 1.3D.

D.1.3 Flush Jointed Pipes
Figure D 2: Concrete pipes 900 mm – 4500 mm diameter where TW < dC
READING ACCUARCY LIMITATIONS

(1) For L/30So ≤ Inlet control value, determine HW The following limitations apply for accurate determination of HW
from Inlet control line. (1) TW < dc
(2) For L/30S0 between Inlet and Outlet control value, (2) For L/30So ≤ Outlet control value, HW < 2D
extrapolate and determine HW. (3) For Outlet control value < L/30So < 1.5 x Outlet control value, HW <
(3) For L/30so in the range 1.0 to 1.5 x Outlet control 1.3D
value, extrapolate and determine HW.

D.1.4 Concrete Box Culverts
Figure D 3: Concrete box culverts
READING ACCURACY LIMITATIONS

(1) For L/30So ≤ Inlet control value, determine HW from Inlet The following limitations apply for accurate determination of HW
control line. (1) TW < dc
(2) For L/30So between Inlet and Outlet control values, (2) For L/30So ≤ Outlet control value, HW < 2D.
interpolate to determine HW. (3) For Outlet control value < L/30So < 1.5 x Outlet control
(3) For L/30So in the range 1.0 to 1.5 x Outlet control value, value HW < 1.3D

Appendix E Solving Manning’s Equation
The following provides examples of the application of Manning’s Equation for determining the hydraulic
characteristics of an existing stream in order to generate data that would be used for the proposed culvert
design. The first application is to determine the existing stream, or channel hydraulics. The second
application is to determine the culvert outlet velocity when outlet control conditions govern the design and the
third application is to determine the culvert outlet velocity when inlet control conditions govern the design.
E.1 Introduction
Manning’s Equation:
V = R2/3S1/2/n
and Equation 1:
Q=VxA
are based on ‘steady state’ and ‘uniform flow’ assumptions but can be used to determine prototype hydraulic
characteristics for many flow states.
Manning’s Equation is commonly used to determine the flow characteristics for:

• existing channels. The hydraulic data generated is used as boundary conditions for proposed new
hydraulic structures such as culverts. This data should be verified from site observations or anecdotal
evidence from adjacent property owners
• within barrel flow conditions for proposed or existing culverts flowing part full or ‘just full’
• culvert outlets. In particular it is necessary to determine the culvert outlet velocity and the flow is
commonly only part full.
For simple conduits Manning’s Equation can be solved from a first principle approach. More complex
conduits can be solved using a ‘trial and error’ method but this can be time consuming. This Appendix
presents a method for solving Manning’s Equation using tables of data for pipes (Table E 1) and trapezoidal
conduits (Table E 2) which includes rectangular conduits when z = 0.0. While careful use of the data would
yield an exact solution in practice it is expected a solution within 1% of the exact solution will be satisfactory.
Designers are encouraged to self-check their calculations or have them independently checked in order to
minimise the chance of errors.
Table E 1 and Table E 2 are for use with metric units as indicated by the following applications.
E.1.1 Application 1
Existing channel hydraulics (see Section 3.7.2 – Outlet Velocity).
In order to determine the flow boundary conditions for a new culvert the existing channel hydraulics need to
be determined. The flow depth (which equals the TW depth for the culvert) and the flow velocity (which
should be correlated to the allowable channel velocity) are the hydraulic characteristics that need to be
determined. Site survey is required to determine the typical channel cross-section and longitudinal bed
slope. It is common for the actual channel cross-section to be represented by a symmetrical trapezoidal
channel with side slopes within the range presented in the attached Table E 2.

It is expected the following parameters will be known:
Design discharge : Q (m3/s)
Channel base width : B (m)
Channel side slope : Z (where the side slope is defined by 1 (V):Z (H)
Channel depth : before berm flow occurs – D (m)
Channel roughness :n
Channel longitudinal slope : S (m/m)
See Figure E 1 which shows the channel parameters.
Step 1
Determine Qn/(S1/2 B8/3).
Step 2
Use linear interpolation of Table E 2 to solve for y/B using the appropriate Z column data.
Step 3
With y/B solve for y, the estimated flow depth. If the flow depth y, is greater than the channel depth D, then
berm flow occurs and conveyance in the various section widths (left bank berm, main channel and right bank
berm) needs to be determined – this case is beyond the scope of this procedure. The following steps
assume y < D.
Step 4
With the flow depth determined the flow area can be determined.
Step 5
The channel flow velocity can be determined using V = Q/A.
Step 6 (Check)
Use the determined y, flow area A and the channel data to determine the following:
• Flow wetted perimeter (P).
• Hydraulic radius (= A/P).
• Flow velocity using Manning’s Equation V = R2/3S1/2/n.
• Verify the determined flow velocity in Step 6 is within 1% of that determined in Step 5.
• Flow capacity Q = V x A (channel capacity for flow depth y).
• Verify the determined flow capacity is within 1% of the design discharge.

Example 1:
Design discharge Q = 2 m3/s
Channel base width B = 0.8 m
Channel side slope Z = 1.0
Channel depth D = 1.5 m
Channel roughness n = 0.04
Channel longitudinal slope S = 0.005
Step 1 Qn/(S1/2 B8/3) = 2.0514
Step 2 y/B = (2.0514 – 2.05)/(3.000 – 2.05) x (1.50 – 1.25) + 1.25 = 1.2504

This is less than the channel depth and all the flow is contained in the channel.
Step 3 y = 1.2504 x B = 1.0003 m, say 1.00 m
Step 4 A = 1.80 m2((1.0 + 0.8) x 1.0)
Step 5 V = Q/A = 2.0/1.80 = 1.111 m/s
Step 6 P = 3.6284 m (√2 x 1.0 x 2 + 0.8)

(Check)
R = 0.4961 m (A/P)
V = 1.1078 m/s (= R2/3S1/2/n)
Excellent agreement with the flow velocity – hence OK
Q (Capacity at y = 1.00 m) = 994 m3/s.
Excellent agreement with the design discharge – hence OK.
E.1.2 Application 2
Outlet velocity – outlet control (see Section 3.10.8 – Outlet Velocity – Outlet control).
As outlined in Section 3.10.8 three water depth conditions need to be considered to determine the controlling
boundary water depth for use in determining the outlet flow velocity. The design discharge is known. Once
the controlling water depth is determined barrel geometric data is required to determine the flow area. Once
the flow area is determined Equation 1 can be used to determine the flow velocity.
The application method is hence dependent on the shape of the barrel and two methods are presented
below, one for pipe (circular) barrels using Table E 1 and the other for box (rectangular or square) barrels
using Table E 2.

Method A: For pipe barrels
Step 1 Determine the flow regime that occurs under the design discharge and hence assess the y to
be used for calculating the outlet flow velocity
Step 2 Calculate y/D
Step 3 Use linear interpolation of Table E 1 on y/D to solve for A/D2
Step 4 With A/D2 solve for A, the estimated flow area
Step 5 Determine the flow velocity using V = Q/A
(Note: There is no check procedure available because the flow depth is fixed by the outlet flow depth
determined for the culvert hydraulics.)
Example 2:
Design discharge 2.0 m3/s
Pipe diameter 0.9 m
Step 1 Say y = 0.83 m governed by outlet flow conditions
Step 2 y/D = 0.9222
Step 3 A/D2 = (0.9222 – 0.90)/(0.9500 – 0.90) x (0.7707 – 0.7445) + 0.7445

(determined by interpolating from Table E 1)
= (0.0222)/(0.05) x (0.0262) + 0.7445
= 0.7561
Step 4 A = 0.7561 (0.9)2 = 0.6125 m2
Step 5 V = Q/A = 2.0/0.6125 = 3.266 m/s say 3.27 m/s
Method B: For box culverts
Step 1 Determine the flow regime that occurs under the design discharge and hence assess the y to
be used for calculating the outlet flow velocity
Step 2 Calculate the flow area = y x B
Example 3:
Box culvert size 1.0 m x 1.0 m (B x D)

Step 1 Say y = 0.83 m governed by outlet flow conditions
Step 2 A = y x B = 0.83 x 1.0 = 0.83 m2
Step 3 V = Q/A = 2.0/0.83 = 2.4096 m/s say 2.41 m/s
E.1.3 Application 3
Outlet velocity – inlet control (see Section 3.10.9).
As outlined in Section 3.10.9 when it is determined that inlet control conditions govern culvert hydraulics the
outlet flow velocity is determined assuming uniform flow conditions develop within the culvert barrel and the
flow velocity occurring within the uniform flow section is assumed to represent the outlet flow velocity. The
design discharge is known. Once the uniform flow depth is determined barrel geometric data is used to
determine the flow area. Once the flow area is determined Equation 1 can be used to determine the flow
velocity.
The application method is hence dependent on the shape of the barrel and two methods are presented
below, one for pipe (circular) barrels and the other for box (rectangular or square) barrels. It is noted that for
pipes there is no need to determine the uniform flow depth as the flow area can be determined directly from
the data in Table E 1 and Table E 2.
Method A: For pipe barrels
Step 1 For the design discharge Q pipe sizing calculations will have been undertaken to determine
pipe diameter D which satisfies both inlet and outlet control conditions for the adopted
configuration (inlet type, barrel shape, length, roughness and inlet/outlet flow conditions)
Step 2 Calculate Qn/(S1/2 D8/3)
Step 3 Check If Qn/(S1/2D8/3) < 0.335

This check needs to be met for open channel flow (OCF) within the barrel. Should this check
not be met then either the pipe roughness or the design discharge Q is too large to pass the
flow through the pipe as OCF. The pipe slope or diameter needs to be increased to pass the
flow but this will generate higher outlet energy. If this arises the designer should review the
overall culvert design with a focus on controlling the outlet energy. The following steps assume
OCF occurs within the barrel.
Step 4 Use linear interpolation of Table E 1 to solve for A/D2 assuming the flow conditions are below
peak – which occurs at approximately y/D = 0.95
Step 5 With A/D2 solve for A, the estimated flow area
Step 7 From Step 2 use linear interpolation to solve for R/D

(Check)
Step 8 With R/D solve for R, the estimated hydraulic radius
Step 9 Determine the flow velocity using Manning’s Equation V = R2/3S1/2/n
Verify that the flow velocity (Step 5) and the flow velocity (Step 8) are within a 1% difference.

Example 4:
Pipe diameter 0.9 m
Pipe roughness 0.012
Pipe slope 0.010 m/m
Step 1 It is determined that inlet control conditions govern (low TW and/or steep culvert slope)
Step 2 Qn/(S1/2 D8/3 ) = 0.3179
Step 3 Is Qn/(S1/2 D8/3) < 0.335 – Yes, hence OCF occurs
Step 4 A/D2 = (0.3179 – 0.305)/(0.3210 – 0.305) x (0.7115 – 0.6736) + 0.6736

(determined by interpolating from Table E 1)
= (0.0129)/(0.016) x (0.0379) + 0.6736
= 0.7042
Step 5 A = 0.7042 x (0.9)2 = 0.5704 m2
Step 6 V = Q/A = 2.0/0.5704
Step 7 R/D = (0.3179 – 0.305)/(0.3210 – 0.305) x (0.3033 – 0.3042) + 0.3042

(Check)
= (0.0129)/(0.016) x (– 0.0009) + 0.3042
= 0.3035
Step 8 R = 0.3035 (0.9) = 0.2731 m
Step 9 V= R2/3S1/2/n = 3.508 m/s
Excellent agreement with the flow velocity determined in Step 5 – OK.
(Note: If desired the flow depth can be determined. In this example it is observed the flow depth y/D
approximates 0.84.)
Method B: For box culverts
Step 1 For the design discharge Q pipe sizing calculations will have been undertaken to determine the
box culvert sizing B x D (width x height) which satisfies both inlet and outlet control conditions
for the adopted configuration (inlet type, barrel shape, length, roughness and inlet/outlet flow
conditions)
Step 2 Calculate Qn/(S1/2B8/3)
Step 3 Determine D/B and associated Qn/(S1/2B8/3) when the barrel is full.
(Check)
Determine if Qn/(S1/2B8/3) < Qn/(S1/2B8/3) when the barrel is full.

This check needs to be met for open channel flow (OCF) to occur within the barrel. Should this
check not be met then either the box culvert roughness or the design discharge Q is too large
to pass the flow through the barrel as OCF. The barrel slope or sizing needs to be increased to
pass the flow but this will generate higher outlet energy. If this arises the designer should
review the overall culvert design with a focus on controlling the outlet energy. The following
steps assume OCF occurs within the barrel.
Step 4 Use linear interpolation of Table E 2 to solve for y/B
Step 5 With y/B solve for y, the estimated flow depth and the flow area = y x B
Step 7 From Step 4, y is known and A is known

(Check)
Determine P=2y+B
Step 8 Determine R = A/P, the estimated hydraulic radius
Step 9 Determine the flow velocity using Manning’s Equation V = R2/3S1/2/n
Verify that the flow velocity (Step 6) and that flow velocity (Step 9) are within a 1% difference.
Example 5:
Box culvert size 1.0 m x 1.0 m (B x D)
Geometry Trapezoid with Z = 0.0
Barrel roughness 0.012
Barrel slope 0.010 m/m
Step 1 It is determined that inlet control conditions govern (low TW and/or steep culvert slope)
Step 2 Qn/(S1/2 B8/3) = 0.240
Step 3 Check – For D/B = 1.0 and Z = 0.0 Qn/(S1/2B8/3) when the barrel is full = 0.481.
(Check)
Is Qn/(S1/2B8/3) < 0.481 – Yes hence OCF occurs.
Step 4 y/B = (0.240 – 0.199)/(0.252 – 0.199) x (0.60 – 0.50) + 0.50 = (0.041)/(0.053) x (0.10) + 0.50
= 0.5774
Step 5 y = 0.5774 (1.0) = 0.5774 m and

A = y x B = 0.5774 m2
Step 6 V = Q/A = 2.0/0.5774

= 3.464 m/s say 3.46 m/s

Step 7 y = 0.5774 m
(Check)
A = 0.5774 m2
P = 2 x 0.5774 m + 1.0 m = 2.1548 m
Step 8 R = A/P = 0.2680 m
Step 9 V = R2/3S1/2/n = 3.464 m/s
Excellent agreement with the flow velocity determined in Step 5 – OK.
Table E 1: Uniform flow in circular sections flowing partly full
y = depth of flow
Q = discharge in m3/s
D = diameter of pipe
n = Manning’s coefficient (0.012 for concrete pipe)
A = area of flow
S = slope of channel bottom and of water surface
R = hydraulic radius
y/D A/D2 R/D Qn/D8/3 S1/2 Qn/y8/3 S1/2

0.01 .0013 .0066 .00005 10.31
.05 .0147 .0325 .00150 4.41
.10 .0409 .0635 .00651 3.02
.15 .0739 .0929 .01515 2.38
.20 .1118 .1206 .0273 1.99
.25 .1535 .1466 .0427 1.72
.30 .1982 .1709 .0611 1.51
.35 .2450 .1935 .0820 1.35
.40 .2934 .2142 .1051 1.210
.45 .3428 .2331 .1299 1.092
.50 .3927 .2500 .1560 .990
.55 .4426 .2649 .1820 .900
.60 .4920 .2776 .2090 .818
.65 .5404 .2882 .2360 .744
.70 .5872 .2962 .2610 .676
.75 .6319 .3017 .2840 .613
.80 .6736 .3042 .3050 .553
.85 .7115 .3033 .3210 .496
.90 .7445 .2980 .3330 .440
.95 .7707 .2865 .3350 .384
1.00 .7854 .2500 .3120 .312
Source: Adapted from Thompson and Kilgore (2006).

Table E 2: Flow in trapezoidal channels by Manning’s Equation
𝑄𝑄𝑄𝑄
Values of
𝐵𝐵8⁄3 𝑆𝑆 1⁄2
y/B Z=0 Z = 1/4 Z = 1/2 Z = 3/4 Z=1 Z = 1 1/4 Z = 11/2 Z = 13/4 Z=2 Z=3
.05 .00638 .00649 .00660 .00667 .00673 .00680 .00687 .00693 .00693 .00714
.10 .0191 .0198 .0206 .0210 .0125 .0129 .0222 .0225 .0229 .0241
.15 .0356 .0377 .0394 .0409 .0423 .0435 .0446 .0459 .0466 .0505
.20 .0548 .0590 .0628 .0660 .0687 .0714 .0741 .0761 .0781 .0869
.25 .0761 .0835 .0896 .0956 .101 .106 .110 .114 .118 .134
.30 .0983 .110 .121 .130 .138 .147 .153 .161 .167 .193
.35 .122 .140 .155 .169 .182 .193 .204 .214 .225 .264
.40 .147 .171 .193 .211 .230 .246 .262 .277 .292 .349
.45 .172 .204 .233 .259 .284 .306 .328 .349 .369 .448
.50 .199 .240 .277 .312 .345 .374 .403 .430 .457 .561
.60 .252 .315 .374 .431 .483 .531 .578 .822 .665 .835
.70 .308 .398 .486 .567 .645 .720 .788 .855 .922 1.18
.80 .365 .488 .610 .727 .835 .943 1.04 1.14 1.23 1.60
.90 .422 .586 .747 .902 1.05 1.19 1.33 1.46 1.59 2.09
1.00 .481 .687 .895 1.10 1.30 1.49 1.66 1.84 2.01 2.67
1.25 .630 .976 1.34 1.70 2.05 2.38 2.71 3.02 3.33 4.53
1.50 .781 1.31 1.87 2.44 3.00 3.54 4.07 4.59 5.08 7.00
1.75 .936 1.68 2.51 3.35 4.18 4.99 5.77 6.54 7.34 10.2
2.00 1.08 2.10 3.25 4.43 5.60 6.73 7.88 8.95 10.0 14.2
2.50 1.39 3.08 6.06 7.14 9.22 11.30 13.30 15.3 17.2 24.9
3.00 1.70 4.26 7.41 10.71 14.10 17.40 20.60 23.8 27.0 39.3
4.00 2.33 7.27 13.60 20.54 27.70 34.70 41.70 48.5 55.3 82.1
Source: Adapted from Thompson and Kilgore (2006).
Figure E 1: Open channel terms

AGRD05B-13 Guide To Road Design Part 5B DrainageOpen Channels Culverts and Floodways

Uploaded by

Document Information

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

AGRD05B-13 Guide To Road Design Part 5B DrainageOpen Channels Culverts and Floodways

Uploaded by

Copyright:

Available Formats

Guide to Road Design Part 5B:

Drainage – Open Channels, Culverts and Floodways

• Format updated • Roads and Maritime Services New South Wales

• Mr Pat Kenny – Roads and Maritime Services, New South Wales

• Ms Marianne Robertson – Roads Corporation Victoria (VicRoads) – Project Manager

2. Open Drains and Channels ................................................................................................................... 4

2.16 Worked Examples .................................................................................................................................. 57

Austroads2013 | page iii

3.13 Culvert Outlet Protection ........................................................................................................................ 99

4. Floodways ........................................................................................................................................... 112

References .................................................................................................................................................... 140

Table 4.2: Standard classes of rock slope protection ............................................................................ 131

Austroads 2013 | page 1

Figure 1.1: Flow chart of Guide to Road Design

1.2 Scope of this Part

Austroads 2013 | page 2

1.3 Road Safety

Austroads 2013 | page 3

2. Open Drains and Channels

2.1.1 Definition and Use

Figure 2.1: Bare earth open channel

Source: DTMR (2010).

Open channels may be constructed to achieve one of three main functions:

Austroads 2013 | page 4

2.1.2 Factors in Choice of Channel Type

In choosing the type of channel to be adopted, consideration should be given to:

2.1.3 Environmental Consultation

Austroads 2013 | page 5

2.1.4 Types of Open Drains

Figure 2.2: Types of open drains

Drain type B Drain type B

Austroads 2013 | page 6

Inlet or outlet drains

Austroads 2013 | page 7

Figure 2.3: Batter drain

Source: D House (used with permission).

Figure 2.4: Bench drain

Source: Austroads (2010b).

Austroads 2013 | page 8

Figure 2.5: Contour drain

Source: RTA (1993).

Figure 2.6: Swale

Source: D House (used with permission).

2.2 General Considerations

The design considerations for an open channel include:

Austroads 2013 | page 9

– stream bank stability or erosion guidelines

2.3 Fundamentals of Open Channel Flow

2.3.1 Stream Dynamics

Streams and roads

All streams are governed by the same basic parameters:

Austroads 2013 | page 10

• hydrological factors – rainfall

Figure 2.7: Velocity distribution across the flow

Source: DTMR (2010).

Types and states of flow

The following outlines the states of flow:

Austroads 2013 | page 11

2.3.2 Assumptions for Analysis

The assumptions made are:

2.3.3 Fundamental Equations

Q = Flow rate (m/s)

V = Average velocity (m/s)