Soil is rarely uniform across an entire site. Strength, density, moisture, permeability, and layer thickness can change over short distances, so one test result may describe only a small part of the ground. For professionals pursuing civil engineering continuing education, understanding that variability is essential because foundation, pavement, slope, and earthwork decisions depend on how representative the available data really is.
The Site Is Bigger Than the Sample
A soil test gives an engineer useful information about a particular sample or location. It does not automatically describe the entire site. Soil properties can change horizontally, vertically, and even within what appears to be the same soil layer. FHWA guidance specifically notes that properties such as cohesion, relative density, and permeability can vary within a stratum, making engineering judgment important when selecting design parameters.
That sounds obvious, but it has a major effect on civil engineering design. A foundation does not sit on a laboratory specimen. A pavement does not experience the average result from several samples. The structure interacts with the actual ground beneath and around it.
One Boring Shows One Location
Consider a site investigation containing several borings. One boring may encounter stiff clay, another may reach loose sand, and a third may pass through fill before reaching natural soil. None of those results is necessarily wrong. They are observations from different parts of the site.
The problem starts when one result is treated as if it represents the entire project area. FHWA guidance recommends considering the spatial variability of subsurface conditions when determining the number and spacing of borings.
A single test can answer, “What did we find here?” It cannot always answer, “What exists everywhere else?”
Soil Can Change Over a Short Distance
Natural deposits rarely form as perfectly uniform layers. Rivers move sediment. Floods deposit material in different patterns. Glacial activity can leave mixed deposits. Weathering changes rock and soil at different rates. Human activity can make the picture even less predictable through grading and uncontrolled fill. That creates both vertical and horizontal variability.
Two samples taken only a short distance apart can show different density, moisture content, grain size, or strength. FHWA pavement guidance specifically recommends evaluating both vertical and horizontal variability when characterizing subgrade conditions.
This is why a neat soil profile drawn between boring locations should be treated as an engineering interpretation, not a photograph of the subsurface.
The Average Value Can Hide the Weak Spot
Suppose three tests produce strength values of 2,000, 2,200, and 800 psf. An average can be calculated easily. But a foundation does not experience an average value in the mathematical sense. A footing located over the weaker zone responds to that weaker material. This is one reason statistical averages should not replace engineering judgment. FHWA notes that site variability can be assessed using measures such as the coefficient of variation, while final parameter selection still requires engineering interpretation.
The lowest result is not automatically the correct design value either. It may represent a local anomaly, testing issue, or genuinely weak zone. The engineer has to determine what the data actually means.
Testing Methods See Different Parts of the Problem
Soil investigations rarely depend on one testing method. Engineers may use borings, laboratory tests, cone penetration testing, geophysical methods, test pits, and other field techniques. Each method provides a different type of information.
FHWA notes that in-situ testing can supplement conventional sampling and can help identify variations in subgrade support, locate areas that need further sampling, and develop a more continuous picture of subsurface conditions. That combination can be more informative than simply collecting more samples of the same type.
Laboratory Results Still Have Limits
A laboratory test is controlled, which is useful. It also examines a small specimen. The sample may not perfectly represent the structure, density, moisture condition, fabric, or stress history of the soil in the ground. Sampling itself can disturb some soils, while handling and storage can affect moisture-sensitive materials.
FHWA has noted that even apparently homogeneous soil deposits can produce significantly different strength and stress-strain results among specimens. So the laboratory number is valuable evidence, but it still needs context.
Groundwater Can Change the Picture
Water is one of the variables that can alter soil behavior significantly. Groundwater levels can vary seasonally or after major rainfall. Excavation, nearby pumping, drainage changes, and construction activity can also affect groundwater conditions.
A soil sample tested under one moisture condition may not behave the same way under another. For pavement and foundation work, groundwater information can influence drainage, bearing behavior, settlement, excavation conditions, and long-term performance. FHWA specifically identifies groundwater location as an important part of subsurface exploration.
Ignoring water can turn an apparently straightforward soil problem into a much harder one.
Fill Soil Deserves Extra Attention
Man-made fill is often more variable than naturally deposited material because its properties depend on what was placed, how it was placed, and how well it was compacted.
An old site may contain layers of different materials from several construction phases. One area could contain compacted structural fill, while another contains loose debris or undocumented material.
FHWA case material from the Central Artery/Tunnel project describes fill varying in thickness, composition, consistency, and density, partly because of differences in borrow material and placement methods. That is a useful reminder that “fill” is a description of origin, not a guarantee of engineering behavior.
Foundation Design Depends on the Actual Ground
Foundation calculations may use values for allowable bearing pressure, settlement, shear strength, and other soil parameters. Those values are only as useful as the subsurface model behind them.
A footing placed over relatively uniform soil may behave as expected. Another footing on a transition between two materials may experience a different response.
Differential settlement is particularly sensitive to changes in soil stiffness and thickness. Two parts of the same structure can therefore move differently even when the applied loads appear similar.
That is where civil engineering PE continuing education courses can help engineers revisit the connection between field investigation, soil mechanics, and foundation behavior.
Pavements Have the Same Problem
Soil variability is not limited to buildings. A roadway may cross several soil types along its alignment. One section of subgrade may provide strong support, while another contains weak or moisture-sensitive material.
FHWA states that the purpose of pavement subsurface investigation includes identifying subgrade conditions that affect construction and pavement performance and characterizing their variability along the project. A pavement designed around a single representative value may therefore perform differently from one designed using a more realistic understanding of changing subgrade conditions.
How Much Investigation Is Enough?
More testing sounds like the obvious answer, but investigation has a cost. The real engineering question is how much information is needed for the decision being made.
A small, simple project on well-understood ground may need less investigation than a major structure on variable or difficult soil. FHWA notes that boring spacing and depth depend on factors such as soil variability, project type, and available investigation information.
There is no useful universal number that fits every site. The investigation should reduce important uncertainty, not simply produce a larger stack of test results.
When Test Results Disagree
Conflicting results are not necessarily a sign that the investigation has failed. They may reveal genuine site variability. FHWA guidance specifically discusses situations where laboratory and in-situ results conflict and points to engineering judgment, historical information, and further evaluation as ways to reassess the subsurface model.
An engineer should ask what could explain the difference. Is the soil changing? Was the sample disturbed? Are the tests measuring different properties? Is groundwater affecting the result?
Those questions can be more valuable than simply choosing the number that looks most convenient.
Construction Can Reveal What Investigation Missed
The investigation does not end the engineer’s interaction with the ground. Excavation may expose a soil layer that was not encountered in the borings. Groundwater may appear at an unexpected elevation. A contractor may encounter soft material beneath an area that looked consistent during design.
FHWA recommends flexibility in subsurface investigation programs when field conditions differ from expectations. That means construction observations can become important engineering data rather than being treated as an inconvenience.
The Engineer Has to Interpret the Pattern
Soil testing produces data. Engineering turns that data into a ground model. That model should account for the locations of tests, the types of soil encountered, changes in properties, groundwater, site history, construction conditions, and the consequences of being wrong.
A useful investigation does not attempt to pretend the subsurface is perfectly known. It gives the engineer enough information to identify important patterns and make a defensible decision.
Keeping Engineering Judgment Sharp
Good civil engineering depends on knowing when a simple answer is enough and when a problem deserves deeper thought. That kind of judgment grows through experience, project exposure, and continued learning.
DiscountPDH introduces civil engineering continuing education courses around practical engineering issues, helping professionals stay engaged with the technical side of their work while earning the education needed to maintain their professional credentials. The goal is simple: keep learning useful, relevant, and connected to the decisions engineers make every day.