Lesson 1: When does unsaturated soil matter?
Unsaturated Soil Mechanics · Course 1 · about 75 minutesYou will learn
Most of the ground we build on is not saturated
Dig a trial pit on a dry day and the soil in the first few metres is usually neither dry nor full of water. Its pores hold both air and water. That soil is unsaturated, and it behaves differently from the saturated soil that most textbooks start with.
- Explain why designing as if all soil were saturated can be over-conservative in some cases and unsafe in others.
- Describe an unsaturated soil as solids, water, air and the air-water interface, and name the site situations where its behaviour matters.
No calculations in this lesson. Lesson 2 adds the numbers.
Why it matters
Saturated design: safe, but not always right
Classical soil mechanics is built around soil whose pores are completely filled with water. Its central idea, effective stress, describes how the load is shared between the soil grains and the water in the pores (Terzaghi, 1943). For a saturated soil it works extremely well, and much of everyday design assumes the ground is either dry or saturated.
Above the water table, though, the pore water is usually held under tension. It pulls the grains together. Engineers call this tension suction, and you will learn to calculate it in Lesson 2. Suction gives the soil extra strength and stiffness, but only while it lasts. Wetting reduces it, and drying increases it.
Often over-conservative
Ignoring suction usually underestimates the strength of soil above the water table. That is on the safe side, but it can mean more support, more material, more cost and more carbon than the ground needs. Over-design is not free, and it is not sustainable.
Sometimes unsafe
Some soils collapse or swell when they get wetter. For them, the problem is the loss of suction. A design that only considers a fixed saturated state may never ask what happens when the ground gets wet, which is exactly when the damage happens.
The big idea
Unsaturated soil mechanics does not replace classical soil mechanics. It extends it to the ground above the water table, where suction changes with weather, vegetation and drainage. The skill is knowing when suction is helping you, how long you can rely on it, and when losing it is the hazard.
Where you meet it
How common is it?
Unsaturated soil is not a special case. You meet it in:
- Compacted fills such as embankments, earth dams, platforms and road subgrades. Compaction almost never fills every pore with water.
- The upper few metres of most sites, which wet and dry with the seasons, even where the water table is shallow.
- Ground with a deep water table, where the whole depth that matters for a foundation or slope can be unsaturated.
- Tropical, arid and semi-arid regions, which cover a large part of the world’s land, and where residual soils can stay unsaturated to great depth.
Wherever you are
Climate decides how deep and how variable the unsaturated zone is. In a wet temperate climate it may be a seasonal layer a few metres thick. In hot dry or tropical climates it can extend tens of metres, with large swings in suction between the dry and rainy seasons. Find out the climate pattern and the depth to groundwater for your area before deciding which assumptions are reasonable.
The idea, in plain words
Four phases, not three
A saturated soil has two phases: solid grains and water. A dry soil has grains and air. An unsaturated soil has grains, water and air, and wherever water meets air inside a pore, there is a curved surface between them. Fredlund and Morgenstern (1977) suggested treating that surface as a fourth phase, the contractile skin, because it behaves differently from the water and the air on either side of it.
Why the interface matters
A curved surface that pulls grains together
Water molecules attract each other strongly. At a free surface this attraction shows up as surface tension, the same effect that lets a needle float on water or makes water climb up a thin glass tube. Inside a soil, the air-water surfaces curve between the grains, and surface tension holds the water in the small spaces at the grain contacts.
The result is that the pore water is at a lower pressure than the pore air, and the little rings of water pull the grains towards each other. That is why damp sand can be shaped into a sandcastle while dry sand and sand under water cannot. In Lesson 2 you will put a number on this pressure difference.
As the soil dries or wets
As soil dries, air enters more pores, the remaining water retreats into smaller spaces and the pull on the grains changes. As it wets, water reconnects through the pores and the pull reduces. The amount of water, and how tightly it is held, depends on the pore sizes, so two soils with the same name can behave quite differently.
What the model does and does not show. The four-phase picture is a model. It explains why pore-air and pore-water pressures differ, which is the basis for suction. It does not show the real three-dimensional shape of the pores, and it does not mean air and water are spread evenly. The contractile skin is not a material you could collect in a jar; it is a way of describing the surface where air meets water.
Quick recap
Describing how much water is present
You can still describe an unsaturated soil with the usual phase relationships. The most useful here is the degree of saturation, S, the fraction of the void volume filled with water:
S tells you how much water is in the pores. It does not tell you how tightly that water is held. Two soils at the same S can have very different suctions. That second piece of information is what Lesson 2 adds, and it is what makes unsaturated soil behave the way it does.
In practice
Same principle, different problems
Across very different structures, the logic is the same: the weather changes the water in the soil, the water changes the suction, and the suction changes how the soil behaves.
| Situation | What changes | Possible engineering effect |
|---|---|---|
| Excavations and cut slopes | Suction holds the face up; rain or exposure over time reduces it | A face that stood for weeks may slump or collapse after wetting |
| Compacted embankments | Prolonged rain wets the outer zone and reduces suction | Cracking, softening and shallow slips, even with a deep water table |
| Foundations on expansive (active) clay | Seasonal wetting and drying, often driven by trees | Swelling and shrinkage, uneven movement, cracked walls |
| Collapsible soils such as loess and some compacted fills | Wetting under load breaks down an open structure held together by suction | Sudden settlement on first wetting, with no extra load added |
| Road subgrades | Seasonal changes in water content | Changes in stiffness and rutting through the year |
| Landfill and mine covers | Rain, evaporation and plant roots change water storage | Cracking and leakage through the cover |
Two soils, two opposite responses
Collapse and swelling
Collapsible soils, such as wind-blown loess and some loosely compacted or residual soils, have an open structure. While they are dry, suction and weak bonds hold the grains in place and the soil can carry load with little settlement. When they are wetted under load, that support is lost and the structure falls in on itself, sometimes suddenly (Jennings and Burland, 1962).
Expansive soils contain active clay minerals. They take in water and swell when wetted, and shrink and crack when they dry. The movement is often uneven, because one side of a building gets wetter or drier than the other.
Water changes the microstructure
In both cases, wetting changes the soil’s internal structure: how the grains and clumps of grains are arranged. That structure then controls how the soil carries load and how water flows through it. This is why the history of wetting and drying matters, not just the water content today.
Getting it right
Suction is useful, but temporary
Saturated design can cost more than it should
Take an excavation. The usual thinking is that the sides must always be supported as if suction did not exist. Support is often essential. But understanding unsaturated soil lets you ask a better question: how long can suction keep this face stable, and what would remove it? That knowledge can make temporary works safer and more economical, instead of designing every case for a condition the ground may never reach.
Never treat suction as a guarantee
Suction can disappear with one heavy storm, a burst pipe or a blocked drain. Understanding it informs the risk and the timing. It never replaces temporary works design, a site risk assessment or the legal duty to keep people safe. In the UK, for example, the Construction (Design and Management) Regulations 2015 require excavations to be prevented from collapsing where people could be harmed, and nobody should enter an unsupported excavation unless it has been assessed as safe.
Wherever you are
Most countries have their own regulations for excavation safety and temporary works. Find the rules that apply where you work. They always take priority over any argument based on suction.
Your turn
Recognise the problem
Task (15 minutes)
For each situation, decide whether unsaturated soil behaviour could matter, and say what change in water you would worry about.
- A 2.5 m deep trench in stiff clayey sand, well above the water table, to be left open for three weeks in the rainy season.
- A house on shrinkable clay, with a large tree 5 m from one corner.
- A piled jetty in a river.
- A warehouse platform built on compacted fill placed on the dry side of optimum.
Show the answers
1. Yes. The trench may stand at first because of suction. Rain can reduce it and the sides may slump. Plan support and drainage for the wet condition, and keep people out unless the excavation is assessed as safe.
2. Yes. The tree draws water from the clay near that corner, so it shrinks in dry seasons and swells when wet. Uneven movement and cracking are possible.
3. Probably not, for the piles. The soil in a river bed is normally saturated, so saturated soil mechanics applies. Banks and approach embankments above water level are a different matter.
4. Yes. Fill compacted dry of optimum can have an open structure that collapses when wetted under load. Check how the fill will get wet: rain, drains or a rising water table.
Quick check
Five quick questions
Choose an answer to see the explanation.
Key points
What to take away
- Much of the ground we build on is unsaturated: compacted fills, the upper metres of most sites, and large areas of tropical and dry regions.
- Ignoring suction is usually conservative for strength, which can mean over-design. For collapsible and expansive soils, ignoring the change in water can be unsafe.
- An unsaturated soil has four phases: solids, water, air and the air-water interface (the contractile skin).
- Surface tension at the curved air-water surfaces holds water at a lower pressure than the air and pulls the grains together.
- Degree of saturation says how much water is present, not how tightly it is held.
- Suction helps, but it is temporary. It never replaces temporary works design or safety rules.
Up next: Lesson 2
Reading the water state. Where the water is in a ground profile, how to calculate matric suction, and why the strength it gives is only apparent cohesion.
References
- Fredlund, D.G. and Morgenstern, N.R. (1977) ‘Stress state variables for unsaturated soils’, Journal of the Geotechnical Engineering Division, ASCE, 103(GT5), pp. 447–466.
- Fredlund, D.G., Rahardjo, H. and Fredlund, M.D. (2012) Unsaturated Soil Mechanics in Engineering Practice. Hoboken, NJ: John Wiley & Sons.
- Jennings, J.E. and Burland, J.B. (1962) ‘Limitations to the use of effective stresses in partly saturated soils’, Géotechnique, 12(2), pp. 125–144.
- Lu, N. and Likos, W.J. (2004) Unsaturated Soil Mechanics. Hoboken, NJ: John Wiley & Sons.
- Ng, C.W.W. and Menzies, B. (2007) Advanced Unsaturated Soil Mechanics and Engineering. Abingdon: Taylor & Francis.
- Terzaghi, K. (1943) Theoretical Soil Mechanics. New York: John Wiley & Sons.
- The Construction (Design and Management) Regulations 2015 (SI 2015/51). London: The Stationery Office.
Further learning
- Read: Chapter 1 of Fredlund, Rahardjo and Fredlund (2012) gives a clear overview of where unsaturated soils occur and why they matter in practice.
- Try it: make two small piles of sand, one dry and one slightly damp, and try to cut a vertical face in each. Then pour water on the damp one and watch the face. That is suction being gained and lost.
- Look around: next time you pass a cutting or an excavation, look for steep faces in soil standing unsupported, and ask what would happen after a week of rain.