Open-Channel Flow and Urban Drainage

Introduction and context

Open-channel flow occurs when liquid moves with a free surface exposed to atmospheric pressure. Rivers, streams, irrigation canals, roadside gutters, swales, storm drains flowing partially full, culverts, and floodways are all examples. Unlike pressurized pipe flow, the driving force is strongly related to the channel slope and the elevation of the free surface. The geometry, roughness, obstructions, and downstream controls determine whether water flows slowly, rapidly, uniformly, or with changing depth.

For construction technologists, open-channel hydraulics is central to site grading, road drainage, subdivision infrastructure, flood conveyance, erosion control, and the safe integration of buildings with natural drainage systems. A design that conveys the calculated discharge but ignores debris, sediment, maintenance, climate uncertainty, or access can fail in service.

Learning objectives

By the end of this chapter, students should be able to:

1. Define hydraulic depth, hydraulic radius, wetted perimeter, and section factor.
2. Analyze specific energy and identify critical flow.
3. Calculate uniform-flow depth using Manning’s equation.
4. Distinguish subcritical, critical, and supercritical flow.
5. Explain hydraulic jumps and estimate their energy loss.
6. Analyze gradually varied flow and hydraulic controls.
7. Apply principles to culverts, storm drains, gutters, and flood conveyance.
8. Select resilient drainage solutions while recognizing assumptions, regulations, and limitations.

1. Geometry and fundamental variables

For any channel cross-section, the important geometric quantities are:

The hydraulic radius measures how effectively a section conveys flow: a larger value generally means less boundary friction relative to area. For a very wide rectangular channel, R ≈ y, because the wetted perimeter is approximately the channel width. Here y is flow depth.

For a rectangular channel of width b and depth y:

For a trapezoidal channel with bottom width b, depth y, and side slope z horizontal to 1 vertical:

These equations assume prismatic geometry and a single, clearly defined flow section. Natural channels may require surveyed cross-sections, composite roughness zones, and computational models.

The continuity equation is:

It expresses conservation of mass for steady incompressible flow. It is applicable whether flow is uniform or nonuniform, provided A and V represent the same section and discharge does not vary along the reach.

2. Specific energy and critical flow

Specific energy is the energy per unit weight measured relative to the channel bottom:

where E is specific energy (m), y is flow depth (m), V is mean velocity (m/s), g is gravitational acceleration (approximately 9.81 m/s²), and α is the kinetic-energy correction coefficient. In many engineering calculations α is taken as 1.0; it may differ from unity where velocity distribution is highly nonuniform.

Using continuity, the general specific-energy relation is:

For a rectangular channel, A = by, so with α = 1:

At a fixed discharge, specific energy has a minimum. The corresponding depth is the critical depth. The general critical-flow condition is:

where Fr is the Froude number and D = A/T is hydraulic depth. Thus, critical flow occurs when gravitational and inertial effects are balanced. For a rectangular channel:

where q is discharge per unit width (m²/s). The critical specific energy for a rectangular channel is E_c = 3y_c/2 when α = 1.

Flow classification is:

• Subcritical: Fr < 1. Flow is relatively deep and slow; downstream conditions can influence upstream depth.
• Critical: Fr = 1. The flow is at minimum specific energy for the discharge.
• Supercritical: Fr > 1. Flow is relatively shallow and fast; upstream conditions dominate.

Critical depth is useful at weirs, channel contractions, transitions, steep slopes, and control sections. It is not automatically the normal depth, because normal depth is determined by resistance and slope rather than minimum specific energy.

Worked example 1: critical depth

A 3.0 m wide rectangular channel carries 6.0 m³/s. Find the critical depth and critical velocity.

The unit discharge is:

Therefore:

At critical flow, A = by = 3.0 × 0.742 = 2.226 m². Hence:

The result satisfies V_c²/(gy_c) ≈ 1, subject to rounding.

3. Uniform flow and Manning resistance

Uniform flow has constant depth and velocity along a prismatic channel. The water-surface slope, energy slope, and channel-bed slope are approximately equal. The most widely used resistance relationship for open channels is Manning’s equation:

where n is Manning’s roughness coefficient, dimensionless in SI practice; R is hydraulic radius (m); and S_f is the friction or energy slope (m/m). For uniform flow, S_f ≈ S_0, the bed slope.

Manning’s equation is empirical. It assumes steady flow, approximately uniform conditions, hydrostatic pressure, and a roughness value representative of the channel surface and vegetation. It should not be applied uncritically at entrances, outlets, sharp transitions, hydraulic jumps, or rapidly varied regions.

Typical indicative n-values include approximately 0.011–0.015 for smooth finished concrete, 0.020–0.030 for earth channels in good condition, and 0.030–0.080 or more for heavily vegetated natural channels. Local standards and calibrated observations should control final selection.

The normal depth y_n is the depth that satisfies Manning’s equation for a specified Q, geometry, n, and slope. Because the equation is implicit in depth, normal depth is commonly found by iteration, a spreadsheet goal-seek function, or numerical software.

Worked example 2: uniform rectangular channel

A smooth rectangular concrete channel is 2.0 m wide, has a bed slope of 0.004, Manning n = 0.015, and carries 4.0 m³/s. Estimate normal depth.

For a trial depth y:

At y = 1.00 m, A = 2.00 m², R = 0.500 m, and Q_calc ≈ 4.22 m³/s. At y = 0.96 m, A = 1.92 m² and R = 0.492 m, giving Q_calc ≈ 3.99 m³/s. Interpolation gives approximately:

The calculation assumes a prismatic, unobstructed channel and uniform flow. In practice, freeboard must be added above the calculated water depth.

Conveyance and compound sections

Channel conveyance is often written:

so that Q = KS_f^(1/2). In compound sections, the main channel and floodplain may have different roughness and hydraulic radii. A common engineering approach divides the cross-section into subsections and sums conveyances, while avoiding inappropriate treatment of internal interfaces as wetted boundaries. Because methods differ, the selected standard and software documentation should be recorded.

4. Gradually varied flow

Gradually varied flow (GVF) changes depth slowly over a substantial distance, allowing the pressure distribution to remain approximately hydrostatic. Its governing equation is:

where x is the longitudinal coordinate, S_0 is bed slope, S_f is friction slope, and Fr is the local Froude number. The equation becomes numerically sensitive near critical flow, where the denominator approaches zero.

The difference between normal depth and critical depth classifies channel slopes:

• Mild slope: y_n > y_c.
• Steep slope: y_n < y_c.
• Critical slope: y_n = y_c.
• Horizontal or adverse slope: S_0 is zero or negative and requires special treatment.

Profiles are commonly designated M, S, C, H, or A according to slope type and by zones above or below y_n and y_c. A subcritical profile is generally controlled downstream, whereas a supercritical profile is generally controlled upstream. A backwater curve upstream of a dam, weir, or culvert outlet is a typical M1 profile.

The direct-step method is a practical hand-calculation technique for prismatic channels. For two sections with depths y_1 and y_2, the distance can be estimated from the specific-energy change and average friction slope:

This form requires a consistent sign convention and sufficiently small depth increments. Numerical methods such as standard-step calculations are preferred for complex geometry, bridges, floodplains, and variable roughness.

5. Hydraulic jumps and rapidly varied flow

A hydraulic jump is an abrupt transition from supercritical to subcritical flow. It dissipates substantial mechanical energy and produces turbulence, air entrainment, and high local forces. Jumps occur downstream of sluice gates, steep channels, spillways, and culvert outlets.

For a rectangular channel, the sequent-depth relationship is:

where y_1 is the initial supercritical depth, y_2 is the downstream conjugate depth, and Fr_1 is the upstream Froude number. The equation assumes a short, horizontal or mildly sloping reach, negligible weight effects over the jump length, and hydrostatic pressure outside the highly turbulent roller.

The specific-energy loss is:

for a rectangular channel with α = 1. A stilling basin uses this controlled energy dissipation, but its dimensions must follow recognized hydraulic design guidance and account for tailwater, blocks, sills, scour, and structural loading.

6. Culverts and storm drains

A culvert is a conduit conveying water beneath a road, railway, embankment, or site access. Culvert performance depends on inlet geometry, barrel roughness and slope, length, outlet condition, tailwater, debris, and possible inlet or outlet control.

In inlet control, discharge is primarily governed by the entrance and upstream head. In outlet control, the barrel friction, entrance loss, exit loss, bends, and downstream tailwater are significant. A preliminary energy equation may be written:

where H is available head, H_L is total head loss, ∆z is elevation change, and V²/(2g) is velocity head. Losses are often represented as:

where K is the sum of local-loss coefficients, f is a Darcy–Weisbach friction factor, L is barrel length, and D_h is hydraulic diameter. For a noncircular full conduit, D_h = 4A/P. Partially full culverts require open-channel analysis and may involve transitions between free-surface and pressurized states.

Culvert design should check hydraulic capacity, upstream afflux, outlet velocity, scour, debris blockage, headwater limits, structural loading, buoyancy, fish passage where relevant, and safe maintenance access. Oversizing a culvert does not automatically solve flooding if downstream conveyance or tailwater controls the system.

Storm drains are networks of inlets, gutters, pipes, manholes, storage units, and outfalls. Hydrologic design supplies a peak inflow; hydraulic analysis determines whether the network conveys it without unacceptable surcharge. The rational method is often used for small catchments:

In SI-compatible practice, if i is in m/s and A_c is in m², Q_p is in m³/s. C is a dimensionless runoff coefficient. If intensity is expressed in mm/h and area in hectares, a unit conversion factor is required; local design manuals provide the appropriate coefficient. The method assumes a relatively small catchment, uniform rainfall intensity over the relevant duration, and a peak occurring when rainfall duration approximates the time of concentration. It should not replace a design-storm model for large, storage-sensitive, or highly complex catchments.

7. Gutters, inlets, and urban surface drainage

Road and site gutters usually behave as shallow open channels. Their design must balance flow spread, pedestrian and vehicle safety, inlet capture, longitudinal grade, crossfall, curb geometry, and allowable ponding. Manning’s equation can estimate gutter conveyance, but the cross-section may be triangular or composite rather than rectangular.

A triangular gutter with cross-slope S_x and flow width T has approximate area:

If the longitudinal slope is S_0, Manning’s equation may be applied using the actual wetted perimeter and hydraulic radius. Inlet capacity is not necessarily equal to gutter capacity: grate spacing, clogging, bypass flow, and sag-point storage must be assessed separately.

Good urban drainage design uses a hierarchy:

1. Prevent excessive runoff through compact grading, permeable surfaces, and soil restoration.
2. Slow and store runoff using swales, rain gardens, detention, and infiltration where soil and groundwater conditions permit.
3. Convey residual flows safely through gutters, inlets, pipes, culverts, and open channels.
4. Provide a defined exceedance route for storms larger than the minor design event.

This approach is more resilient than relying solely on buried pipes.

8. Flood conveyance and resilient design

Flood conveyance analysis evaluates how a channel, floodplain, bridge, road crossing, or development transports extreme flows. The design discharge may be estimated from hydrologic frequency analysis, rainfall-runoff modelling, regional methods, or regulatory flood studies. Hydraulic calculations then determine water levels, velocities, shear stresses, flood extents, and afflux.

A one-dimensional model is often adequate for long, relatively well-defined channels. Two-dimensional models are valuable where flow spreads across floodplains, streets, buildings, or multiple pathways. Results depend on topographic resolution, roughness zoning, boundary conditions, numerical mesh, structures, and calibration data. Model outputs should be interpreted as scenarios rather than exact predictions.

Resilience requires consideration of:

• larger or more intense rainfall under climate-change scenarios;
• blockage by sediment, vegetation, vehicles, and debris;
• erosion and bank instability;
• loss of power or access during storms;
• safe failure and overflow routes;
• protection of vulnerable people and critical facilities;
• inspection, cleaning, and asset-management budgets.

Freeboard is the vertical allowance between the design water surface and the lowest vulnerable elevation. It is not a substitute for sound hydraulics; it accommodates uncertainty, waves, debris, settlement, and modelling limitations.

9. Illustrative case study: redevelopment of a flood-prone construction site

Background

A fictional illustrative case concerns a 2.4 ha mixed-use site draining toward a small concrete-lined channel. Existing flooding occurs at the downstream road crossing during intense summer storms. The proposed development would increase impervious area from 35% to 70%.

Problem

A preliminary rational-method estimate indicates a peak runoff increase. The existing 600 mm culvert has limited headwater capacity, and a steep downstream reach produces supercritical flow and localized scour. The design team must reduce peak discharge, maintain access, and avoid transferring risk to downstream properties.

Analysis

The team delineates subcatchments and estimates time of concentration using the local standard. It evaluates several rainfall durations rather than assuming one duration. The culvert is checked for both inlet and outlet control, and a blockage sensitivity case is included. Surveyed channel sections are used to estimate normal depth with Manning’s equation and to identify a likely hydraulic jump near the outlet. A two-dimensional model is used for the road crossing because floodwater can bypass the channel along the street.

Approach and solution

The selected scheme combines permeable parking bays, bioretention areas, a vegetated swale, underground detention, and a widened overflow path. The culvert inlet is redesigned with debris-tolerant protection, while outlet protection and a stilling arrangement reduce erosion. The site grading directs exceedance flow away from building entrances and toward a public drainage corridor. Maintenance responsibilities for filters, sediment removal, vegetation, and inspection after major storms are written into the asset plan.

Outcome

The final design reduces the frequent-event peak and keeps the road operational for the specified design event, while the exceedance route limits building inundation during larger events. The analysis also shows that increasing culvert diameter alone would have produced little benefit because downstream tailwater and channel capacity controlled the system.

Learning points

Hydraulic capacity is a system property, not an isolated pipe attribute. Distributed runoff reduction, robust conveyance, blockage allowance, erosion control, and maintainability should be designed together. Every model result should be accompanied by assumptions, sensitivity checks, and a clear statement of residual risk.

10. Limitations, governance, and professional practice

Open-channel formulas are simplified representations of real systems. Manning’s n is uncertain; rainfall and flood estimates are probabilistic; sediment changes geometry; vegetation varies seasonally; and construction tolerances can materially alter grades. A calculation should therefore identify the design event, boundary conditions, roughness values, survey datum, units, software version, and quality-control checks.

Drainage works are commonly governed by municipal standards, environmental permits, floodplain regulations, road authority requirements, accessibility provisions, and occupational-safety rules. Designs may affect wetlands, fish passage, downstream erosion, water quality, and neighboring landowners. Engineers and construction technologists have a duty to avoid foreseeable harm, disclose uncertainty, and not present a model output as a guarantee.

Digital terrain models, GIS, remote sensing, automated sensors, and AI-assisted calibration can improve data processing. However, AI does not replace hydraulic judgement. Training data may omit rare floods, automated parameter fitting may conceal nonphysical values, and black-box predictions can be difficult to defend in regulatory decisions. Human review, independent checks, field inspection, and transparent documentation remain essential.

Best-practice checklist

• Confirm the catchment boundary, datum, units, and design rainfall.
• Survey actual channel geometry and identify obstructions.
• Select roughness values by surface condition, not material name alone.
• Check normal depth, critical depth, Froude number, and freeboard.
• Distinguish inlet control from outlet control in culverts.
• Check transitions, hydraulic jumps, scour, and downstream tailwater.
• Include blockage, maintenance, and exceedance-flow scenarios.
• Use appropriate one- or two-dimensional modelling for floodplain conditions.
• Coordinate hydraulic design with structural, geotechnical, environmental, and construction requirements.
• Record assumptions and independently verify calculations.

Chapter Summary

Open-channel flow is governed by channel geometry, gravity, resistance, and boundary controls. Continuity relates discharge, area, and velocity, while specific energy provides the framework for understanding critical depth and flow transitions. The Froude number classifies flow as subcritical, critical, or supercritical. Manning’s equation estimates uniform-flow capacity and normal depth under suitable steady and approximately uniform conditions.

Gradually varied flow is described by the water-surface-profile equation and is controlled downstream in subcritical regimes and upstream in supercritical regimes. Hydraulic jumps convert supercritical flow to subcritical flow and dissipate energy. Culverts and storm drains require assessment of entrance conditions, friction, tailwater, blockage, surcharge, and outlet scour. Gutters and inlets must be designed for both hydraulic capacity and safe surface operation.

Resilient urban drainage combines runoff reduction, storage, conveyance, overflow routing, maintenance, and consideration of future uncertainty. Sound professional practice requires transparent assumptions, regulatory compliance, field verification, and recognition that drainage is a connected system rather than a collection of independent components.

Key Terms

Open-channel flow: Flow with a free surface exposed to atmospheric pressure.

Discharge: Volume of water passing a section per unit time.

Hydraulic radius: Flow area divided by wetted perimeter.

Hydraulic depth: Flow area divided by top width.

Wetted perimeter: Boundary length in contact with flowing water.

Specific energy: Energy per unit weight measured relative to the channel bottom.

Critical depth: Depth at which specific energy is minimum for a given discharge.

Froude number: Dimensionless ratio indicating the relative importance of inertial and gravitational effects.

Subcritical flow: Flow with Froude number less than one.

Supercritical flow: Flow with Froude number greater than one.

Normal depth: Uniform-flow depth for specified discharge, slope, geometry, and roughness.

Manning roughness coefficient: Empirical parameter representing boundary resistance.

Gradually varied flow: Flow whose depth changes slowly along the channel.

Hydraulic jump: Rapid transition from supercritical to subcritical flow.

Sequent depths: Upstream and downstream depths associated with a hydraulic jump.

Culvert: Conduit carrying water beneath an embankment or transportation facility.

Headwater: Water level upstream of a hydraulic structure.

Afflux: Rise in upstream water level caused by a structure or constriction.

Freeboard: Vertical safety allowance above the calculated design water level.

Exceedance route: Planned pathway for flows greater than the primary design capacity.

Review Questions

Short Answer Questions

1. Define hydraulic radius and explain why it affects channel conveyance.
2. Distinguish normal depth from critical depth.
3. What physical meaning does the Froude number have?
4. State two conditions under which Manning’s equation may be inappropriate.
5. Why should culverts be checked for blockage and outlet scour?

Long Answer Questions

1. Explain specific energy in a rectangular channel and derive the rectangular critical-depth relationship.
2. Describe how channel slope, normal depth, and critical depth are used to classify gradually varied flow profiles.
3. Explain the formation, purpose, and energy loss of a hydraulic jump.
4. Compare inlet control and outlet control in culvert hydraulics.
5. Develop a resilient urban drainage strategy for a rapidly urbanizing catchment.

Analytical / Application Questions

1. A 4.0 m wide rectangular channel carries 8.0 m³/s. Calculate the critical depth and critical velocity, assuming α = 1.0.
2. A proposed development increases impervious cover substantially but has limited downstream pipe capacity. Recommend a technically justified combination of source-control and conveyance measures.
3. A culvert has adequate theoretical barrel capacity but floods upstream during storms. Identify at least four possible causes and propose a field and modelling investigation.

Multiple Choice Questions with Answers

Question 1

For a rectangular channel, which expression defines hydraulic radius?

A. A/T

B. A/P

C. P/A

D. Q/A

Correct Answer: B. A/P

Question 2

A flow with Froude number less than 1 is classified as:

A. Supercritical

B. Critical

C. Subcritical

D. Pressurized

Correct Answer: C. Subcritical

Question 3

A 2 m wide rectangular channel carries 2 m³/s. Using y_c = (q²/g)^(1/3), the critical depth is approximately:

A. 0.22 m

B. 0.47 m

C. 0.78 m

D. 1.02 m

Correct Answer: B. 0.47 m

Question 4

Manning’s equation is most directly used to estimate which condition?

A. Uniform-flow discharge or depth

B. Atmospheric pressure

C. Rainfall frequency only

D. Structural concrete strength

Correct Answer: A. Uniform-flow discharge or depth

Question 5

In subcritical flow, which boundary condition generally has the stronger influence?

A. Upstream condition

B. Downstream condition

C. Wind direction only

D. Bed material color

Correct Answer: B. Downstream condition

Question 6

A hydraulic jump changes flow from:

A. Subcritical to supercritical

B. Supercritical to subcritical

C. Pressurized to vacuum flow

D. Uniform to stagnant flow only

Correct Answer: B. Supercritical to subcritical

Question 7

For a rectangular channel, the critical-flow condition is represented by:

A. Fr = 0

B. Fr < 0

C. Fr = 1

D. Fr → infinity in every case

Correct Answer: C. Fr = 1

Question 8

Which factor is especially important in distinguishing culvert inlet control from outlet control?

A. Paint color of the culvert

B. Entrance geometry and downstream tailwater

C. Air temperature alone

D. Building occupancy only

Correct Answer: B. Entrance geometry and downstream tailwater

Question 9

The rational method peak-discharge equation is commonly written as:

A. Q = AV

B. Q = (1/n)AR^(2/3)S^(1/2)

C. Q_p = CiA

D. E = y + V²/(2g)

Correct Answer: C. Q_p = CiA

Question 10

Which measure best represents resilient urban drainage practice?

A. Relying only on a larger underground pipe

B. Ignoring exceedance flows because they are rare

C. Combining runoff reduction, storage, conveyance, and safe overflow routing

D. Using one roughness value for every surface

Correct Answer: C. Combining runoff reduction, storage, conveyance, and safe overflow routing