Effective Stress
Water pressure pushes the grains apart, so only part of the soil's weight presses them together.
Before this Lesson
- Unit Weights of Soil
- Pressure and Depth · Fluid Mechanics
A vacuum-packed brick of ground coffee is as hard as a block of wood. Snip a corner, hear the hiss, and it turns into a soft bag of grounds you could pour. The air outside presses on the bag just as hard as before. What changed is the air between the grains: there was almost none, and now it pushes back from inside as hard as the air outside pushes in.
Ground below the water table is that bag, with water in place of air. The water's pressure decides how much of the soil's weight the grains press on each other with, and that decides how strong and stiff the ground is. This Lesson finds that number at any depth.
The weight of everything above
Pick a point in the ground and picture the column of ground standing on one square foot around it. Everything in that column rests on that square foot: the grains, the water in their voids, and any lake sitting on top. Its weight per unit area is the total vertical stress, .
You've done this walk before. It's the layered tank from Pressure and Depth, with soil layers in place of oil and water. Start at the ground surface and add each layer's unit weight times its thickness:
Each layer gets the unit weight that matches where it sits, as in Unit Weights of Soil. Above the water table that's the moist unit weight . Below it the voids are full, so it's . A sand that the water table cuts through counts as two layers: the same sand, with above the water table and below. Standing water, such as a lake, is one more layer on top, with a unit weight of .
Stresses come out in psf (lb/ft²) or ksf, where 1 ksf is 1,000 psf, or in kPa (kN/m²). In this topic , , and are vertical unless a question says otherwise.
The water's share
Below the water table, the water in the voids connects from pore to pore all the way up to the water table. The grains make a maze of it, but in still water a container's shape never matters, only depth does. So the pore water's pressure follows , with measured down from its free surface. That free surface is the water table: the level where the groundwater's pressure is zero, which is where water would stand in a hole dug down to it. The Handbook's Effective stress equations write the pore water pressure as
and call the pressure head. In still groundwater it's simply the depth below the water table, so grows by 62.4 psf per foot, or 9.81 kPa per meter. Measure it from the water table, not from the ground surface. Above the water table, take . If a lake covers the site, the lake's surface is the free surface, so counts from there.
What the grains feel
Now cut through the soil at the point with an imaginary horizontal slice, and ask what holds up everything above it. Two things do. The water pushes up on the slice with pressure , and since grains touch only at tiny points, it acts over practically the whole area. The grains below push up through those contacts. Whatever the water doesn't carry, the contacts must, and what they carry per unit area is the effective stress:
The Handbook prints that equation under Effective stress, and its Vertical Stress Profiles diagram writes it as . In words, the water pressure pushes the grains apart, holding up part of the weight that would otherwise press them together.
That's the coffee brick. Sealed, the bag had air outside pressing in at 14.7 psi and almost nothing pushing back from inside, so the grains carried all of it: was about 14.7 psi. Snipped open, the air between the grains rose to 14.7 psi too, and fell to zero.
Why give its own name? Everything a soil does happens at those contacts. Grains slide past each other only when the friction between them gives way, and friction grows with how hard they're pressed together, so sets the strength. A layer compresses only when its grains are pushed into a tighter arrangement, so sets the settlement. That's why most of this topic opens the same way: find first.
One consequence surprises nearly everyone. Take two sites with the same clay, one with the water table at the ground surface and one under 10 ft of standing water. At any depth in the clay, the second site has 624 psf more total stress, from the lake's weight. But its free surface is 10 ft higher, so its pore pressure is 624 psf higher too, and is exactly the same. A lake presses on the soil only through water, and the water in the voids pushes back just as hard. That's why mud on the floor of the deep ocean, under miles of water, is as soft as mud at the edge of a pond.
Skipping the subtraction
Below the water table, each foot you go down adds to and to . So each foot adds their difference to :
That's the effective (submerged) unit weight from Unit Weights of Soil, which most questions call buoyant, and it's no accident that it turns up here. That Lesson found as the weight per cubic foot that Archimedes leaves the grains to press on each other with, which is exactly what counts. So you can build directly, with for each layer above the water table and below it, and never compute at all. For point A:
Same answer, no . The trap is mixing the two routes. has already taken the water out, so a built from is finished, and subtracting from it takes the water out twice.
When a question gives and instead of a unit weight, the Handbook's Phase Relationships section supplies , and is one subtraction away.
The shape of a profile
Plot , , and against depth, and every line is made of straight pieces, because each is a running sum of unit weights times thicknesses. Within a piece the rate of growth is steady, and a line bends only where that rate changes:
- grows at each layer's own unit weight, so it bends at every layer boundary and at the water table.
- is zero down to the water table, then grows at in one straight line.
- matches down to the water table, where is still zero. Below it, grows at , roughly half its rate above.
Depth runs down the page, so a line that grows faster runs flatter. Here are the three profiles for the site you just worked:
Layout adapted from the FE Reference Handbook, Vertical Stress Profiles
The Handbook's Vertical Stress Profiles draws the same three plots for a simpler site, with two layers and the water table at the boundary between them. With a lake on top, and both start at the lake's surface and climb together at through the water, so stays at zero down to the lake bed and only starts growing there.
Working backward
Some questions run the other way. They give the a point must have and ask how deep the water table is, or how thick a layer is. The profile is still a sum of unit weights times thicknesses, so write it with a letter for the unknown thickness and solve.
The water table version has one wrinkle: its unknown shows up twice. The deeper the water table sits, the more of the sand lies above it at and the less lies below it at . Each extra foot of depth trades a foot of for a foot of , and raises by .