Groundwater Flow, Seepage, and Construction Dewatering

Introduction

Groundwater is water occupying the voids and fractures beneath the ground surface. In construction, it may be a valuable resource, but it can also reduce soil strength, cause instability, flood excavations, produce uplift, and initiate piping beneath dams and retaining structures. Safe construction therefore requires an understanding of how water moves through soil and rock and how that movement can be controlled.

This chapter develops the principles of Darcy’s law, permeability, hydraulic gradients, flow nets, well hydraulics, seepage forces, uplift, piping, filtration, drainage, and dewatering. The emphasis is on practical application to foundations, excavations, earth dams, cofferdams, and retaining structures.

Learning Objectives

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

1. Explain groundwater pressure, hydraulic head, hydraulic gradient, and permeability.
2. Apply Darcy’s law to calculate discharge through soil.
3. Distinguish Darcy velocity from actual seepage velocity.
4. Interpret flow nets and estimate seepage discharge and uplift.
5. Calculate discharge from confined and unconfined wells under idealized conditions.
6. Evaluate seepage forces, critical hydraulic gradients, piping risk, and filters.
7. Select appropriate drainage and construction-dewatering systems.
8. Recognize the assumptions, limitations, environmental effects, and professional responsibilities associated with groundwater control.

1. Groundwater and Hydraulic Head

Groundwater commonly occurs in saturated layers called aquifers. An aquifer may be unconfined, with its upper boundary at the water table, or confined between relatively impermeable strata. The water table is the surface at which pore-water pressure is atmospheric. Below it, pore-water pressure is generally positive, although localized negative pressures may occur in unsaturated soil.

The total hydraulic head at a point is

where h is total hydraulic head (m), z is elevation head relative to a chosen datum (m), p is pore-water pressure (Pa), and γ_w is the unit weight of water (N/m³). For water, γ_w ≈ 9.81 kN/m³.

Water flows from higher total head toward lower total head. A difference in water level alone does not fully describe flow direction when elevation and pressure vary; total head must be considered.

For one-dimensional flow in the s direction, the hydraulic gradient is

The negative sign indicates that flow occurs toward decreasing head. In magnitude, i = ∆h/L, where ∆h is head loss and L is the length of the flow path. Hydraulic gradient is dimensionless.

Important terminology

• Aquifer: A geological formation that stores and transmits appreciable groundwater.
• Aquiclude: A formation that may store water but transmits it very slowly.
• Aquitard: A semi-permeable layer that transmits limited flow.
• Piezometric level: The elevation to which water rises in a piezometer.
• Drawdown: Reduction in hydraulic head caused by pumping, usually written as s = h₀ - h.
• Flow net: A graphical representation of groundwater flow lines and equipotential lines.
• Seepage force: The force exerted by flowing water on the soil skeleton.

2. Permeability and Darcy’s Law

Permeability, or hydraulic conductivity, measures the ease with which a fluid moves through a porous medium. It depends on soil characteristics and fluid properties. Coarse gravel generally has much higher permeability than clay. Hydraulic conductivity is denoted by k and has units of m/s.

For saturated, steady, laminar flow through a homogeneous soil, Darcy’s law is

where q is discharge (m³/s), k is hydraulic conductivity (m/s), i is hydraulic gradient (dimensionless), and A is the cross-sectional area normal to flow (m²). The Darcy flux, or discharge per unit area, is

The Darcy velocity is not the actual average velocity of water within the pores because the pores occupy only part of the bulk area. If n_e is effective porosity, the average seepage velocity is approximately

This distinction is important when estimating contaminant travel time or the time required for water to move through a drainage layer.

Assumptions of Darcy’s law

Darcy’s law is most reliable when:

• flow is saturated and approximately steady;
• the soil is sufficiently homogeneous at the scale considered;
• flow is laminar, as normally occurs in sands, silts, and clays;
• the fluid is Newtonian and its viscosity is known;
• the hydraulic gradient is evaluated along the actual flow direction.

Very high velocities in coarse gravel, fractured rock, or open voids may produce non-Darcian flow. In layered soils, an equivalent hydraulic conductivity may be used only after considering the direction of flow.

Equivalent permeability of layered soil

For layers of thicknesses H₁, H₂, ..., H_n and conductivities k₁, k₂, ..., k_n, flow parallel to the layers has the arithmetic weighted average:

Flow perpendicular to the layers has the harmonic average:

Typically, k_h > k_v. These relations assume continuous horizontal layers and one-dimensional flow.

Worked Example 1: Darcy discharge

A saturated sand layer has k = 2.0 × 10^-5 m/s. The head loss over a 12 m flow length is 0.60 m. The cross-sectional flow area is 18 m². Determine discharge.

The hydraulic gradient is

Therefore,

The discharge is approximately 0.018 L/s. If effective porosity is 0.30, the average seepage velocity is

3. Seepage Forces, Effective Stress, and Uplift

Water flowing through a soil exerts a seepage force on the soil skeleton. In a soil volume V, the seepage-force magnitude is approximately

where F_s is seepage force (N), i is hydraulic gradient, γ_w is unit weight of water (N/m³), and V is the affected soil volume (m³). The force acts in the direction of water flow.

The effective-stress principle is

where σ′ is effective normal stress, σ is total normal stress, and u is pore-water pressure. An increase in pore pressure reduces effective stress and may reduce shear strength. For saturated soil, a rising groundwater level can therefore make a slope, excavation face, or foundation less stable.

Under upward seepage, the seepage force acts upward and reduces effective stress. At the condition of boiling or quick soil, the effective stress approaches zero.

Critical hydraulic gradient

For a saturated cohesionless soil, the critical upward hydraulic gradient is approximately

where G_s is the specific gravity of soil solids and e is the void ratio. This follows from balancing submerged soil weight against upward seepage force. The expression assumes uniform saturated soil and neglects cohesion and capillary effects.

A practical factor of safety against piping or boiling may be expressed as

where i_exit is the actual exit gradient. Design criteria commonly require a factor of safety greater than unity, with the selected value depending on consequences, uncertainty, and applicable codes.

Uplift pressure

Water pressure beneath a slab, dam, or foundation produces an upward force. If pressure varies linearly from p₁ to p₂ over area A, the resultant is

For a strip of width b and length L, use the plan area A = bL. If uplift is represented by hydraulic heads, pressure at a point is

provided the point is below the piezometric level and pressure is gauged relative to atmospheric pressure.

4. Flow Nets

A flow net is composed of flow lines, which indicate the direction of seepage, and equipotential lines, which join points of equal hydraulic head. In an ideal homogeneous and isotropic soil, these lines intersect approximately at right angles and form curvilinear squares.

A flow net is useful for estimating seepage, exit gradients, and uplift beneath dams, sheet piles, weirs, and retaining walls. For a two-dimensional flow net, the discharge per unit width is

where q is discharge per unit width (m²/s), H is total head loss (m), N_f is the number of flow channels, and N_d is the number of potential drops. The equation assumes a properly constructed flow net, homogeneous isotropic soil, steady seepage, and negligible three-dimensional effects.

Each potential drop represents head loss

The approximate seepage force near an exit can be assessed from the local head loss divided by the local flow distance. Closely spaced equipotential lines indicate a large hydraulic gradient and therefore a possible piping risk.

Flow-net construction principles

1. Identify impermeable boundaries; they are flow lines because no flow crosses them.
2. Identify constant-head boundaries; they are equipotential lines.
3. Sketch smooth flow lines and equipotential lines.
4. Adjust the network toward approximate curvilinear squares.
5. Count flow channels and potential drops.
6. Check gradients near exits, drains, and sheet-pile toes.

Flow nets are approximations. Numerical groundwater models are preferable when geometry, anisotropy, layering, transient pumping, or irregular boundaries are significant.

5. Well Hydraulics

Wells are used for water supply, groundwater investigation, and construction dewatering. Pumping lowers the water level around the well, creating a cone of depression. The radius of influence depends on soil properties, pumping duration, aquifer boundaries, recharge, and well design.

Confined aquifer

For steady radial flow to a fully penetrating well in a confined aquifer, the idealized discharge is

where Q is pumping rate (m³/s), k is hydraulic conductivity (m/s), b is aquifer thickness (m), h_o is head at radius r_o (m), h_w is head at the well radius r_w (m), and r_o and r_w are radial distances (m). The assumptions include steady flow, a homogeneous isotropic confined aquifer, a fully penetrating well, radial flow, and no nearby boundaries or recharge.

Unconfined aquifer

For steady flow to a well in an unconfined aquifer, the Dupuit approximation gives

where H_o and H_w are saturated thicknesses at the outer radius and well, respectively. This formula assumes predominantly horizontal flow and a mild water-table slope. Near the well, vertical flow may invalidate the approximation.

Worked Example 2: Confined well

A confined aquifer has k = 1.5 × 10^-4 m/s and thickness b = 10 m. The head falls from 22 m at a radius of 100 m to 17 m at a well radius of 0.20 m. Find the ideal steady pumping rate.

Since ln(500) ≈ 6.215,

Thus the idealized pumping rate is approximately 7.6 L/s. Actual capacity may be lower because of well losses, partial penetration, clogging, well-screen limitations, or boundary effects.

6. Piping, Internal Erosion, and Filters

Piping is progressive removal of soil particles by seepage. It often begins at a downstream exit where gradients are high and confinement is low. It can occur beneath hydraulic structures, through earth dams, around conduits, or behind retaining walls.

Internal erosion is broader than visible pipe formation. It includes concentrated leak erosion, backward erosion, contact erosion, and suffusion, in which fine particles migrate through the voids of a coarser soil. Warning signs include cloudy seepage, increasing discharge, sinkholes, sand boils, wet spots, and unexpected settlement.

Filter and drainage criteria

A filter must allow water to pass while retaining the protected base soil. Exact criteria depend on gradation, uniformity, soil type, and governing standards. A commonly used preliminary retention condition is based on comparing filter size D_15,f with base-soil size d_85,b:

A commonly used permeability condition is

These are screening rules, not universal design laws. They must be checked against project specifications and laboratory testing. A filter should also be continuous, properly compacted where appropriate, protected from segregation, and separated from clogging materials.

7. Drainage and Dewatering Systems

Dewatering is the controlled removal or exclusion of groundwater to permit safe construction. The selected method depends on soil permeability, excavation depth, drawdown required, site geometry, inflow rate, nearby structures, and discharge constraints.

Common systems

Sumps and pumping

Sumps collect water at low points, where pumps remove it. They are acceptable when the soil is not vulnerable to erosion, piping, or slaking. In fine sand, direct pumping from an open sump may carry soil particles and cause settlement outside the excavation. A properly graded filter or wellpoint system is safer.

Wellpoints and deep wells

Wellpoints are small-diameter screened pipes installed around an excavation and connected to a header pipe. They are effective where wells can be closely spaced and the required drawdown is moderate. Deep wells are more suitable for large inflows, deep excavations, and highly permeable aquifers.

Dewatering design should estimate pumping rate, drawdown, radius of influence, well spacing, screen length, pump capacity, standby capacity, and water-quality treatment. Pumping should begin gradually to observe actual response and reduce the risk of sudden gradients.

Cutoff and exclusion methods

Impermeable or low-permeability barriers include slurry walls, secant pile walls, sheet piles, grout curtains, and diaphragm walls. These methods reduce inflow and can limit drawdown outside the site. However, barriers may redirect groundwater, generate pressure differences, or create leakage paths at joints and around utilities.

8. Practical Design Sequence

A disciplined groundwater-control investigation normally follows these steps:

1. Review geology, topography, nearby water bodies, existing drainage, and historical groundwater levels.
2. Install piezometers or monitoring wells at multiple elevations and locations.
3. Conduct permeability testing, such as laboratory tests, pumping tests, or slug tests.
4. Define excavation geometry, construction stages, tolerable inflow, and allowable drawdown.
5. Develop a conceptual groundwater model identifying recharge, boundaries, aquifers, aquitards, and preferential paths.
6. Select a control strategy: pumping, cutoff, drainage, recharge, or a combination.
7. Estimate flow using Darcy calculations, flow nets, well equations, or numerical models.
8. Check slope stability, base heave, piping, uplift, settlement, and effects on adjacent structures.
9. Design discharge treatment, energy dissipation, sediment control, and monitoring.
10. Establish trigger levels and contingency actions before excavation begins.

9. Illustrative Case: Dewatering a Deep Foundation Excavation

Background

This illustrative case concerns a 10 m deep basement excavation beside an existing masonry building. The site consists of fill over medium sand, underlain by a lower-permeability silty layer. The groundwater table is 2 m below ground level. The excavation must remain dry enough for foundation construction, but excessive drawdown outside the site could cause settlement of the neighboring building.

Problem

Open sump pumping would probably create steep gradients in the sand and could remove fines. The project therefore needs a method that controls inflow while limiting off-site drawdown.

Analysis

Site investigations identify the sand as the main water-bearing layer. Piezometers show that the silty layer is not fully continuous, so localized leakage is possible. A preliminary flow-net assessment indicates high gradients near the excavation corners. Settlement risk is associated with lowering pore pressure beneath the neighboring building.

Approach and solution

A secant pile cutoff wall is selected around the excavation. The wall extends into the silty layer, but the design allows for local defects and leakage. Inside the wall, a line of wellpoints handles residual inflow. A low-permeability working slab and internal drainage channels collect seepage. Piezometers are installed inside and outside the wall, and settlement points are placed on the neighboring building.

Pumping begins several days before excavation and is increased in stages. Turbidity and discharge are monitored. If outside piezometers fall below predefined trigger levels, pumping is reduced and an engineered recharge system is prepared. Water is passed through settlement tanks before discharge.

Outcome

The combined cutoff and controlled pumping system maintains a dry working area while producing much less external drawdown than a ring of high-capacity deep wells. Monitoring confirms that neighboring settlement remains within the specified limit. Minor leakage at one wall joint is treated locally rather than increasing the overall pumping rate.

Learning points

• Dewatering is a soil–structure–water interaction problem, not merely a pump-sizing exercise.
• Cutoff systems can be valuable where external settlement is a major concern.
• Staged pumping and monitoring are essential because field conditions are uncertain.
• Discharge quality and legal permissions must be addressed before pumping starts.
• Contingency measures should be designed before an emergency occurs.

10. Limitations, Safety, and Professional Responsibility

Groundwater predictions are uncertain because natural deposits are heterogeneous and subsurface information is incomplete. A single permeability value may not represent fissures, lenses, anisotropy, or preferential flow paths. Well equations can overpredict performance when wells are partially penetrating or poorly developed. Flow nets depend strongly on the quality of the geometric interpretation.

Construction teams should monitor groundwater levels, pumping rates, turbidity, piezometric response, wall movement, settlement, and structural condition. Instrument readings should be interpreted against trigger and alarm levels rather than simply archived.

Dewatering may affect neighboring wells, wetlands, streams, vegetation, contaminated sites, and building foundations. Discharged water may contain suspended solids, hydrocarbons, salts, iron, or other contaminants. Permits may be required for abstraction and discharge. Ethical practice requires competent design, transparent communication of risks, protection of third parties, and immediate action when monitoring indicates danger.

Digital groundwater models and automated sensors can improve interpretation, but they do not replace geological judgment. Artificial intelligence may assist with data screening, anomaly detection, or calibration, but input quality, boundary assumptions, and engineering accountability remain essential.

Chapter Summary

Groundwater flows through soil in response to gradients in total hydraulic head. Darcy’s law, q = kiA, provides the fundamental relation for saturated laminar flow, while seepage velocity accounts for the pore space actually available to water. Hydraulic conductivity depends on soil and fluid properties and is strongly influenced by layering and anisotropy.

Flow nets provide approximate estimates of seepage discharge, potential drops, uplift, and exit gradients. Well-hydraulics equations describe idealized radial flow to confined and unconfined wells, but field performance depends on boundaries, well construction, aquifer heterogeneity, and pumping duration. Seepage forces reduce effective stress under upward flow; when the hydraulic gradient approaches the critical value, boiling and piping may occur.

Filters and drainage systems must retain soil particles while allowing safe water passage. Sumps, wellpoints, deep wells, eductor systems, cutoff walls, relief wells, and drainage blankets serve different geological and construction conditions. Reliable dewatering requires investigation, staged implementation, monitoring, discharge management, and contingency planning. The central professional objective is to maintain construction safety while avoiding harmful drawdown, erosion, uplift, settlement, and environmental damage.

Key Terms

Hydraulic head: Elevation plus pressure head, representing the energy state of groundwater.

Hydraulic gradient: Change in total hydraulic head per unit distance in the flow direction.

Hydraulic conductivity: Measure of the ease with which water flows through soil.

Darcy’s law: Relation stating that saturated laminar discharge is proportional to hydraulic gradient and area.

Darcy velocity: Discharge divided by total cross-sectional area, equal to ki.

Seepage velocity: Average velocity of water through the connected pore space.

Aquifer: Geological formation capable of storing and transmitting groundwater.

Aquitard: Semi-permeable geological layer that transmits groundwater slowly.

Piezometer: Instrument used to measure hydraulic head or pore-water pressure.

Drawdown: Reduction in groundwater head caused by pumping.

Flow line: Line tangent to the direction of groundwater flow.

Equipotential line: Line joining points of equal hydraulic head.

Flow net: Network of flow lines and equipotential lines used to analyze seepage.

Seepage force: Force exerted by moving water on the soil skeleton.

Effective stress: Stress carried by the soil skeleton after subtracting pore pressure from total stress.

Critical hydraulic gradient: Gradient at which upward seepage can reduce effective stress to approximately zero.

Piping: Progressive removal of soil particles by seepage.

Filter: Graded material that permits drainage while retaining base-soil particles.

Dewatering: Controlled removal or exclusion of groundwater during construction.

Cutoff wall: Low-permeability barrier used to reduce groundwater flow.

Review Questions

Short Answer Questions

1. Define hydraulic head and explain why groundwater flows from higher to lower total head.
2. State Darcy’s law and identify its variables and units.
3. Distinguish Darcy velocity from seepage velocity.
4. What is the purpose of a filter beneath or downstream of a hydraulic structure?
5. Name two situations in which open-sump pumping may be unsafe.

Long Answer Questions

1. Explain the construction and interpretation of a flow net for seepage beneath a dam or sheet-pile wall.
2. Discuss how groundwater affects effective stress, foundation stability, uplift, and piping.
3. Compare wellpoints, deep wells, eductor systems, and cutoff walls for construction dewatering.
4. Derive or explain the use of the critical hydraulic gradient for cohesionless soil.
5. Describe a complete investigation and monitoring program for a deep excavation below the water table.

Analytical / Application Questions

1. A sand layer has k = 3.0 × 10^-5 m/s, cross-sectional area 25 m², and head loss 0.75 m over a 15 m flow path. Calculate discharge using Darcy’s law.
2. A soil has G_s = 2.65 and void ratio e = 0.65. Calculate its critical hydraulic gradient and explain the engineering significance of the result.
3. An excavation beside an old building requires 8 m of drawdown. Propose a groundwater-control system, identify the principal risks, and specify the field measurements needed to verify performance.

Multiple Choice Questions with Answers

Question 1

Which quantity is the driving force for groundwater flow through saturated soil?

A. Total hydraulic head gradient

B. Soil dry density only

C. Atmospheric temperature only

D. Grain color

Correct Answer: A. Total hydraulic head gradient

Question 2

A soil has k = 2.0 × 10⁻⁵ m/s, hydraulic gradient i = 0.10, and area A = 5 m². What is the Darcy discharge?

A. 1.0 × 10⁻⁶ m³/s

B. 1.0 × 10⁻⁵ m³/s

C. 1.0 × 10⁻⁴ m³/s

D. 1.0 × 10⁻² m³/s

Correct Answer: B. 1.0 × 10⁻⁵ m³/s

Question 3

Why is seepage velocity greater than Darcy velocity in a saturated soil?

A. Water flows through only the connected pore area

B. Water has zero viscosity

C. The hydraulic gradient is always greater than one

D. The soil skeleton flows with the water

Correct Answer: A. Water flows through only the connected pore area

Question 4

For an isotropic homogeneous soil, flow lines and equipotential lines in an ideal flow net intersect approximately at what angle?

A. 0°

B. 30°

C. 45°

D. 90°

Correct Answer: D. 90°

Question 5

What is the principal effect of upward seepage on effective stress?

A. It increases effective stress indefinitely

B. It reduces effective stress

C. It has no effect on pore pressure

D. It changes soil grains into clay

Correct Answer: B. It reduces effective stress

Question 6

For Gs = 2.65 and e = 0.65, the critical hydraulic gradient is approximately:

A. 0.50

B. 0.75

C. 1.00

D. 1.72

Correct Answer: C. 1.00

Question 7

Which system is generally well suited to shallow-to-moderate drawdown in sands using closely spaced screened points?

A. Wellpoint system

B. Concrete gravity dam

C. Open-air ventilation system

D. Rock anchor system

Correct Answer: A. Wellpoint system

Question 8

In a two-dimensional flow net, increasing the number of potential drops while keeping total head loss constant causes each potential drop to:

A. Increase

B. Remain exactly equal to total head loss

C. Decrease

D. Become dimensioned in seconds

Correct Answer: C. Decrease

Question 9

What is a primary purpose of a filter in a seepage-control system?

A. Prevent all water movement

B. Retain soil particles while permitting drainage

C. Increase piping deliberately

D. Replace structural reinforcement

Correct Answer: B. Retain soil particles while permitting drainage

Question 10

Why can high-capacity pumping beside an existing building be dangerous?

A. It may increase external pore pressure and cause frost

B. It may cause drawdown, consolidation, and settlement outside the excavation

C. It always makes soil impermeable

D. It eliminates the need for monitoring

Correct Answer: B. It may cause drawdown, consolidation, and settlement outside the excavation