Who this guide is for
- Owners who have noticed cracking, sloping floors or doors that no longer close
- Buyers reading a survey that mentions movement and wanting to follow the language
- Owners about to remove or plant trees close to a building
- Self-builders trying to understand why a foundation has been designed the way it has
- Anyone preparing for a conversation with a structural engineer
Downward, and for different reasons
Ground compresses under load, and some movement downward is ordinary. What separates one situation from another is what caused the compression, whether it is finished, and whether it is happening evenly. A whole building settling together is a different problem from one corner going down on its own.
The reference entries keep returning to the same distinction: differential movement is the characteristic risk rather than uniform movement. A pad responds to its own patch of ground, so variation across a site shows up as differences between pads rather than being averaged out along a continuous element.
Ground that loses support beneath a footing
Some downward movement is not compression under load at all but support disappearing. A shrinkable soil drying and reducing in volume, a trench dug alongside a footing that removes ground the footing was relying on, a leaking drain washing fines away, or fill that was never finished consolidating all produce the same symptom from different causes.
That is why the remedy discussion cannot start until the cause is named. Deepening a foundation answers one of those causes and leaves the others running, which is the point the underpinning entry makes when it says the work does not by itself resolve why movement happened.
- Shrinkable soil drying out, seasonally or persistently
- Ground removed alongside or beneath an existing footing
- Water moving fines out of the ground beneath
- Fill that has not finished settling
Heave is the opposite mechanism, not the opposite word
Some soils change volume with moisture in both directions. Where a soil that had been dried out recovers, it swells, and the ground pushes upward and sideways against whatever is buried in it. The library names the clearest case: clay recovering after trees have been removed.
That matters because heave acts on parts of a foundation that downward movement never touches. It can act against the sides of buried beams and caps, and as uplift on pile shafts, which is why the entries describe it as a strategy set by the geotechnical designer rather than a detail.
Why a void exists under a ground beam
A ground beam is not a strip foundation cast at a lower level. It spans between pile caps or piers, and the soil beneath it is deliberately kept out of the load path. The entry is explicit that caps and beams are trying not to touch the ground while piles are trying to engage it.
Where the ground can swell, a void former or compressible layer is placed beneath the beam so the ground has somewhere to move without pushing on the soffit. That layer belongs to the system even though its purpose is to collapse or disappear, and it is a designed absence rather than wasted space.
How that provision gets lost
The entry names the failure directly. If the layer is omitted, crushed before the beam is cast, or filled with spoil, the beam is loaded from below by something the design assumed would not reach it. Void formers have to survive everything that follows until the beam is cast over them, in a confined and often wet zone.
It is also only part of a strategy. Provision beneath a beam does not deal with heave by itself, because swelling ground also acts on the sides of buried elements and as uplift on pile shafts.
Seasonal change is not the same as a defect
Shrinkable soils move with seasonal wetting and drying, and a building founded on them experiences that as a rhythm rather than an event. Which soils behave this way, and what a foundation should do about it, is investigation and design work rather than something to be read from a crack in a spring.
What an engineer will want to see
Movement is diagnosed from evidence rather than from a description, and the evidence is usually a record over time: where cracks are, how wide, whether they follow the mortar joints or run through the units, and whether they are changing. Cracking that follows the joints is describing a different condition from cracking that runs straight through.
- Where the movement appears, and its relationship to openings and returns
- Whether it is changing, and over what period
- What is nearby: trees, drains, a recent excavation, raised ground
- What has been altered in the building before, and when
Framing a foundation movement conversation
- 1Record where movement appears, with dates, rather than describing it from memory.
- 2Note whether cracking follows the mortar joints or runs through the units.
- 3Note whether it appears to be changing, and over what period.
- 4List trees and hedges near the building, including any recently removed.
- 5Note any excavation, trench or landscaping carried out nearby, and when.
- 6Note any drainage that has leaked, been altered or been rerouted.
- 7Note whether external ground levels have been raised against the building.
- 8Collect what is known about the ground beneath the building.
- 9Collect any records of earlier alterations to the structure.
- 10Ask a qualified structural engineer to diagnose before any remedy is designed.
- 11Ask what monitoring is proposed and what would change the diagnosis.
Common mistakes to avoid
- Using settlement, subsidence and heave interchangeably, when they describe different mechanisms with different remedies.
- Choosing a remedy before the cause has been named, when several causes produce the same visible symptom.
- Assuming every crack indicates foundation movement, when the reference entries treat crack patterns as a starting point for assessment rather than a conclusion.
- Removing a mature tree from shrinkable soil without asking what the ground will do as it recovers.
- Filling the void beneath a ground beam with spoil, when it is a designed separation rather than an unfinished excavation.
- Treating heave provision under a beam as the whole strategy, when swelling ground also acts on the sides of buried elements.
- Diagnosing movement from photographs and descriptions instead of from an inspection of the building.
When to involve a professional
- Evidence of movement in a building should be assessed before any alteration is designed, and that assessment belongs with a qualified structural engineer.
- Whether a soil on a particular site swells or shrinks is established by ground investigation and interpreted by geotechnical professionals.
- What provision is called for beneath and beside buried elements where ground can swell is a geotechnical assessment rather than a standard detail.
- Nothing here states or implies what is happening beneath any particular building, and no mechanism described is a diagnosis.
Frequently asked questions
Questions readers ask about this topic
What is the difference between subsidence and settlement?
Both describe downward movement, but the useful distinction for an owner is cause. Some movement is ground compressing under a new load; some is support disappearing beneath a footing that had been standing on it. The remedy follows the cause, so naming it comes first.
Why would removing a tree cause a problem?
Where a shrinkable soil has been kept dry by a mature tree, removing it can let the soil recover and swell. The reference library names clay recovering after trees have been removed as a case where provision is made for ground to move without loading buried elements.
Is the gap under my ground beam a defect?
A ground beam spans between supports and the soil beneath it is deliberately kept out of the load path. Where the ground can swell, a void former or compressible layer gives it somewhere to move without pushing on the beam soffit. That separation is designed rather than accidental.
Do cracks always mean the foundation is moving?
Not necessarily. The library treats crack pattern as evidence to be read rather than a conclusion, noting that cracking following the mortar joints describes a different condition from cracking running through the units. Assessment of a specific building belongs with an engineer.
Will underpinning stop the movement?
It takes an existing foundation to a different bearing level, and the entry states plainly that it does not by itself resolve why movement happened. Trees, drains, changed water regimes and made ground continue afterwards unless each is dealt with separately.
Keep reading