Soil, Bearing Capacity & the Site

The first mental shift is to stop thinking of the ground as a fixed backdrop and start thinking of it as a structural material with properties as real as those of concrete or steel - only far more variable. Soil is broadly sorted by the size of its particles, and that particle size dictates almost everything about how it behaves under a building.

At one end are the coarse, cohesionless soils - gravels and sands. Their grains are large and do not stick to one another, so their strength comes entirely from friction and interlock as the grains press together. Well-compacted, dense sand and gravel make excellent, stable foundation soil that drains freely and settles quickly, mostly during construction. Loose sand, however, is treacherous: it can densify suddenly under vibration, and saturated loose sand can liquefy in an earthquake, briefly behaving like a heavy liquid and letting buildings sink or tilt.

At the other end are the fine, cohesive soils - silts and clays. Their particles are microscopically small and electro-chemically sticky, so they hold together by cohesion even without friction. Clay is the great troublemaker of foundation engineering. It is nearly watertight, so it drains extremely slowly, which means it keeps settling for years after a building is finished. Many clays are also expansive: they swell when wetted and shrink when dried, heaving in the monsoon and cracking away in the dry season. India's notorious black cotton soil is exactly this kind of clay, and it wrecks lightly loaded buildings whose foundations were not taken below its active, moisture-changing zone. Real ground is almost never one pure type but a layered, mixed profile - a metre of made-up fill over soft clay over dense sand over rock - and reading that layering is the heart of the job.

When you place a footing on soil and load it, two things can go wrong, and both must be checked. The first is an outright shear failure: push hard enough and the soil beneath the footing gives way, the ground shears along a curved surface and the footing plunges down while the earth beside it heaves up. The maximum pressure the soil can carry just before this collapse is its ultimate bearing capacity. We never design to that value - we divide it by a generous factor of safety (typically around three) to get the safe bearing capacity, the pressure at which failure is comfortably far away.

But there is a second, quieter limit that usually governs, especially on clay: settlement. Long before the soil is anywhere near collapse, it can compress enough under load to crack finishes, jam doors and tilt the building. So the pressure we are actually allowed to use - the allowable or safe bearing pressure - is the lower of two things: the pressure that keeps us safe from shear failure, and the pressure that keeps settlement within tolerable limits. On strong sand or rock the shear limit rarely bites and settlement is small; on soft clay the settlement limit dominates and the allowable pressure can be shockingly low.

Crucially, bearing capacity is not a fixed property of a soil that you can read off a chart with confidence - it depends on the soil, on the size and depth of the footing, and on the water table. A wider footing spreads load over more soil but also pushes its stress deeper, which changes settlement; a footing taken deeper often finds better soil and gains confinement. This is why the crude allowable-bearing-pressure tables in old textbooks and building bylaws are only a starting guess for preliminary sizing. A real value comes from testing the actual site and is written into the geotechnical report as a recommended safe bearing capacity at a stated founding depth - the single number that sizes almost every shallow foundation in the building.

A footing does not load only the soil directly beneath it. The pressure spreads downward and outward into the ground in a shape engineers call the pressure bulb or stress bulb - concentrated just under the footing and fading with depth, roughly vanishing to insignificance at a depth of about one and a half to two times the footing width. This simple picture explains a great deal.

It explains why settlement happens where it does: the soil that compresses is the soil inside the bulb, so a wide footing, whose bulb reaches deep, will settle more than a narrow one carrying the same pressure, because it squeezes a much larger and deeper volume of ground. It explains why a weak layer buried well below the footing can still matter, if the bulb reaches down into it. And it explains why two footings placed close together can interact - their bulbs overlap, and the soil between them is worked harder than either footing alone would suggest.

The most dangerous form of settlement is not uniform sinking but differential settlement - different parts of the building settling by different amounts. A structure that settles evenly, even by a large amount, may suffer little damage (Mexico City's cathedral and many old buildings have sunk remarkably while staying whole). But differential movement bends and racks the structure, cracking walls, shearing beams from columns and jamming openings, because it forces the frame to accommodate a shape it was never designed for. Differential settlement is what foundation design is really fighting, and it is worst where the soil varies across the plan, where column loads differ greatly, or where part of a building sits on fill and part on natural ground. Managing it - by sizing footings so they settle equally, by tying them together, or by reaching a uniform stiff stratum - is a recurring theme of the next three lessons.

Nothing controls foundation behaviour more quietly and more completely than water in the ground. The water table is the level below which the soil is saturated, its pores full of water, and where it sits - and how it moves with the seasons - reshapes every foundation decision.

Water weakens soil. A rising water table reduces the effective stress between soil grains (the buoyancy of the water carries part of the load), which can substantially lower the bearing capacity of sand and increase settlement. It makes excavation harder and more dangerous, because open pits below the water table flood and their sides slump; building a basement below the water table means fighting uplift - the same buoyancy that reduces soil stress tries to float the whole basement like a boat, a force that must be resisted by the building's weight or by anchors. Water is the agent that makes expansive clays swell and loose saturated sands liquefy. And water is the lifelong enemy of the substructure itself, driving the need for waterproofing and drainage that Lesson 4.4 takes up in detail.

Because the water table moves - rising in the monsoon, falling in the dry months, and drifting over decades with urban development and groundwater extraction - a responsible investigation records not just today's level but the likely highest level the foundation will ever see. Designing a basement or a footing for a dry trial pit dug in summer, only to have it sit in water every monsoon, is a classic and expensive mistake. The honest rule is that you design foundations for the wettest the ground will get, not the driest you happened to see it.

Since the ground is invisible, variable and decisive, the only responsible way to design a foundation is to investigate it first. A geotechnical site investigation is the process of finding out what is actually beneath a plot, and its central tool is the borehole: a hole drilled into the ground, from which the driller logs the layers, lifts out soil samples, and measures the water table. From the samples, a laboratory measures the properties that matter - strength, compressibility, plasticity, moisture content - and from the field, simple in-situ tests read the soil where it lies.

The most widespread field test is the Standard Penetration Test (SPT). A standard sampler is driven into the bottom of the borehole by a standard hammer dropped a standard distance, and the number of blows needed to drive it 300 mm is recorded as the N-value. A low N-value means soft, loose, weak soil; a high N-value means dense, stiff, strong soil. Crude as it sounds, the N-value is enormously useful: it correlates with bearing capacity, with the density of sands and the stiffness of clays, and it is cheap and available everywhere, which is why the SPT profile - N-values plotted against depth down each borehole - is the backbone of most foundation reports in India and much of the world. Other tools (the cone penetration test, plate load tests, laboratory triaxial tests) add precision where needed.

All of this feeds the geotechnical report, which recommends a founding stratum, a safe bearing capacity, an expected settlement, the design water table, and - critically - a foundation type. This is why the sequence of design is often the reverse of what students expect: the geotechnical engineer, reading the ground, effectively tells the architect and the structural engineer what is possible. Good soil near the surface permits cheap shallow footings; weak soil over deep firm strata forces expensive piles; expansive black cotton soil demands special measures. In India this whole domain is framed by IS 1904, the general code for the design and construction of foundations in soils, alongside companion codes for investigation and testing. Skipping the investigation to save time or money is the single most common cause of foundation trouble - because the ground always gets the last vote, and it is far cheaper to hear it in a report than in a crack.