Every structure, from a modest residential home to a high-rise tower or a massive bridge abutment, ultimately rests on the ground. The soil beneath a foundation must carry the weight of the structure without failing or settling excessively, and the concept that governs this requirement is soil bearing capacity. It is one of the most fundamental inputs in geotechnical engineering, yet it is often oversimplified on site, where “good soil” and “bad soil” are judged by eye rather than by test. This article explains what bearing capacity is, distinguishes ultimate from allowable values, explains factors of safety, reviews the common field tests, presents typical values by soil type, and shows how this knowledge drives foundation design.
What Is Soil Bearing Capacity?
Soil bearing capacity is the maximum average pressure that a soil can support at the base of a foundation without experiencing shear failure or unacceptable settlement. It is expressed in units of pressure, most commonly kilopascals (kPa) or pounds per square foot (psf). When a foundation presses down on the ground, the soil responds by developing resisting stresses; bearing capacity is the practical limit of that resistance.
An important point often lost in casual conversation is that bearing capacity is not a fixed property of a soil alone, like its density or water content. It is a property of the soil-foundation system: the same clay stratum can support quite different pressures depending on the width and shape of the footing, the depth of embedment, the position of groundwater, and whether the load is vertical and centered or inclined and eccentric. Wider and deeper foundations mobilise soil strength over a larger zone, which is why capacity generally increases with foundation size. A value read from a table is therefore only an approximate starting point, never a final answer.
Bearing capacity must also be distinguished from bearing pressure, the actual stress a foundation imposes on the soil, calculated by dividing the applied load by the footing area. The design objective is to keep the bearing pressure below the allowable bearing capacity throughout the life of the structure.
Ultimate vs Allowable Bearing Capacity
Geotechnical engineers work with two related but very different quantities: ultimate and allowable bearing capacity. Understanding the difference is central to safe foundation design.
Ultimate Bearing Capacity
The ultimate bearing capacity (qult) is the pressure at which the soil beneath a foundation fails in shear: the ground can no longer resist the load, and the foundation plunges into the soil or the soil bulges out from beneath it. This is a true collapse condition. Classical theories, most famously those of Terzaghi and later Meyerhof, Hansen and Vesic, model this failure using the soil’s shear strength parameters, cohesion and the angle of internal friction. The general bearing capacity equation combines terms for soil cohesion, overburden pressure at foundation level, and the soil’s self-weight, each modified by bearing capacity factors and by shape, depth and inclination corrections. On dense sand, failure tends to be a sudden general shear failure at a well-defined load, while soft clays more often exhibit local or punching shear with large deformations before collapse.
Allowable Bearing Capacity
The allowable bearing capacity (qall) is the maximum pressure that may safely be imposed on the soil, obtained by dividing the ultimate bearing capacity by an appropriate factor of safety. It is the value the structural designer actually uses, always well below the ultimate capacity, because collapse is never acceptable.
There is a second, often more restrictive, consideration: the allowable bearing pressure may also be limited by settlement. A foundation on a loose, compressible soil might have a perfectly adequate shear capacity, yet still settle so much that the structure cracks, doors jam, and the building becomes unserviceable. Most codes therefore require the allowable bearing pressure to be checked against both shear failure and allowable settlement criteria, and the governing, lower value is adopted for design. On many real projects, particularly on clay soils, settlement rather than shear is the true controlling factor.
The Factor of Safety in Bearing Capacity Design
The factor of safety (FoS) is the ratio of ultimate to allowable bearing capacity, and it is the designer’s buffer against everything that is uncertain in geotechnical engineering. Typical values for shallow foundations range from 2.5 to 3.0 in most codes, with higher values of 4 to 5 where loads are transient or seismic, or where soil parameters came from limited investigation.
The factor of safety exists because soil properties are inherently variable. An investigation samples a tiny fraction of the ground, laboratory tests introduce their own errors, and analytical models are idealisations of real behaviour. The chosen factor therefore reflects the quality and quantity of investigation data, the reliability of load estimates, the consequence of failure, and local practice. A factor of 3 on a heavily loaded footing founded on variable alluvial deposits, with capacity estimated only from correlations, is not the same margin as a factor of 3 on a uniform, thoroughly tested sandstone.
It is also worth noting that the factor of safety is applied to shear capacity, while settlement is controlled separately against allowable limits. A high factor of safety does not automatically protect a structure from excessive settlement, which is why responsible design always verifies both criteria.
Common Bearing Capacity Tests
Bearing capacity is rarely measured directly for every project. Engineers rely on a combination of field tests, laboratory tests, and empirical correlations. Three field tests dominate practice worldwide: the plate load test, the standard penetration test, and the cone penetration test.
Plate Load Test
The plate load test is the only common test that measures bearing capacity directly. A steel plate, typically 300 mm to 750 mm square or circular, is placed at foundation level in a prepared pit, and load is applied incrementally through a hydraulic jack reacting against kentledge or anchorages. Settlement is recorded at each increment until failure or a target load is reached. The resulting pressure-settlement curve yields the ultimate bearing capacity and, with an allowable settlement criterion and a factor of safety, the allowable bearing pressure.
The test has real limitations. Its influence depth is only about one to two times the plate width, so it tests a shallow zone that may not represent the full stress bulb of a much wider real footing. It is expensive and slow, making it impractical at many locations, and results must be corrected for size effects, footing shape, and groundwater. Despite these drawbacks, it remains the benchmark test for important structures on uniform soils.
Standard Penetration Test (SPT)
The standard penetration test is the most widely used in situ test worldwide. A 63.5 kg hammer is dropped freely through 760 mm to drive a split-spoon sampler 450 mm into the soil at the bottom of a borehole, and the blow count for the final 300 mm is recorded as the N-value. The N-value is a direct measure of resistance to penetration, and decades of experience have produced robust correlations between N-values and the bearing capacity of sands and, less reliably, clays.
The SPT’s advantages are simplicity, low cost, and the fact that it recovers a disturbed soil sample at every test depth, allowing the soil to be identified and logged. Its disadvantages are its empirical nature, significant variability with equipment and operator energy, and poor performance in gravels, cobbles, and very soft clays. Corrected N-values, adjusted for overburden pressure and hammer energy, feed standard charts relating allowable bearing pressure to N-value and footing width.
Cone Penetration Test (CPT)
The cone penetration test pushes a slender instrumented cone, typically with a 10 cm² base area, continuously into the ground at a constant rate of 20 mm per second. It records tip resistance, sleeve friction, and, in the piezocone version, pore water pressure, producing a virtually continuous profile of soil behaviour with depth. Because there is no hammering, the CPT is fast and delivers far more data than borehole tests.
CPT results correlate well with bearing capacity through empirical relationships and, in modern practice, with directly measured soil behaviour type, giving engineers a detailed picture of stratigraphy, strength, and compressibility. The principal drawback is that the CPT recovers no soil samples, so it is normally paired with a few boreholes for identification. In soft ground and for investigating deep layers, the CPT is often the preferred tool.
Typical Bearing Capacity Values by Soil Type
The following ranges of allowable bearing capacity are indicative values drawn from common practice and published correlations, useful for preliminary assessment and for checking the plausibility of test results, but never a replacement for a proper site investigation for design purposes.
- Sound, unweathered rock: typically above 1000 kPa, rising to several thousand kPa for fresh granite, basalt or gneiss, with capacity controlled more by weathering, fractures and bedding than by rock type.
- Dense gravel or dense sandy gravel: 300 to 600 kPa, excellent founding material when compacted.
- Medium dense sand or gravelly sand: 150 to 300 kPa, good founding material for most structures.
- Loose sand or loose sandy gravel: 75 to 150 kPa, marginal for spread footings and better treated or compacted.
- Stiff clay or hard clay: 150 to 300 kPa, acceptable for moderate loads but settlement-sensitive.
- Medium or firm clay: 75 to 150 kPa, commonly requiring raft foundations or ground improvement for heavier structures.
- Soft clay or silty clay: 50 to 75 kPa, generally requiring deep foundations or extensive improvement.
- Very soft clay, organic silt, or peat: below 30 kPa and often unsuitable for any shallow foundation without treatment.
Several cautions apply. Values for granular soils depend heavily on relative density, groundwater depth, and confining pressure; lowering the water table or compacting the soil can increase capacity dramatically. Values for clays depend on undrained shear strength and are strongly governed by consolidation settlement, so the allowable value is frequently dictated by settlement rather than shear. Local building codes often publish tabulated safe bearing capacities for common soil types; these tables are convenient, but should be treated as screening tools rather than design data.
How Bearing Capacity Drives Foundation Design
Bearing capacity is the starting point from which the entire foundation solution flows. The first decision a designer makes, whether to use shallow or deep foundations, is essentially a bearing capacity question: can the near-surface soils carry the applied loads safely, or must the load be carried down to deeper, stronger strata?
Where shallow foundations are viable, bearing capacity directly sets the size of the footing. For a given column load and allowable bearing pressure, the required footing area is simply the load divided by the allowable pressure. Doubling the allowable pressure halves the footing area, changing excavation volumes, concrete quantities, reinforcement, and cost. This is why geotechnical engineers are often asked to squeeze every possible kPa out of a site: the savings in foundation cost are direct and substantial.
Bearing capacity also shapes a range of design decisions. On soft or variable soils, designers may switch from isolated spread footings to a raft foundation, which spreads the load over the whole footprint and reduces the average bearing pressure. Where even a raft cannot achieve the required capacity or settlement limits, driven or bored piles transfer loads to deeper bearing strata, assessed using the same soil parameters. Eccentrically loaded footings, common under retaining walls and moment frames, must be checked so that the maximum edge pressure stays below the allowable value. Groundwater enters the equation too: buoyancy reduces the effective overburden that contributes to capacity, and excavation below the water table can loosen granular soils, requiring dewatering or compaction. On weak sites, ground improvement techniques such as dynamic compaction, vibro-compaction, stone columns, or soil replacement are selected precisely to raise the bearing capacity of the founding stratum.
Finally, the quality of the geotechnical investigation determines the confidence the designer can place in the numbers. A well-instrumented investigation with SPT or CPT soundings across the site allows lower factors of safety, optimised footing sizes, and realistic settlement estimates. A thin investigation forces conservative assumptions that translate directly into larger, deeper, and more expensive foundations, or worse, into unsafe ones. Modern design codes, such as Eurocode 7, formalise this thinking by requiring characteristic soil parameters and partial factors that reflect the variability of the ground and the quality of the investigation.
Conclusion
Soil bearing capacity is the bridge between a structure and the ground that supports it. Understanding the difference between ultimate and allowable bearing capacity, applying appropriate factors of safety, and selecting the right field tests are all essential to answering the central geotechnical question: how much load can this ground safely carry? The plate load test, SPT and CPT each have strengths and limitations, and skilled engineers combine them with laboratory testing and local experience to build a reliable picture of the ground. Typical values by soil type are a useful frame of reference, but no substitute for site-specific investigation, because bearing capacity depends on the unique combination of soil type, strength, groundwater, and foundation geometry at each location. In the end, bearing capacity decides whether a project rests on economical shallow footings, a load-spreading raft, or a deep pile system, protecting both the safety of the structure and the budget of the project.