BRE Driscoll Update 2007

subsidence damage to domestic buildings
subsidence damage to domestic buildings

This guide provides an authoritative, comprehensive guide to the best practice
in all the technical elements of subsidence damage to a domestic building.
It reviews the best current information and provides a baseline of technical
guidance that can be used by practising investigators of subsidence. It is SUBSIDENCE DAMAGE
TO DOMESTIC BUILDINGS
intended particularly for training new investigators.
Key aspects of the investigation and remediation of subsidence are covered.
• Diagnosing subsidence damage
• Investigating clay shrinkage and swelling
A guide to good technical practice
• Subsidence monitoring
• Dealing with the subsidence and repairing the damage
The guide has drawn on the considerable knowledge of BRE and its researchers
Richard Driscoll and Hilary Skinner
on the technicalities of subsidence, and of members of the Subsidence Forum.
To enhance the value of the guide, all the BRE Digests and other relevant
publications that are referred to in the text are provided in PDF format on a
CD-rom. This will be found in a sleeve attached to the inside of the back cover.
about the authors
Driscoll and Skinner

Richard driscoll worked on subsidence problems for over 25 years at the Building Research
Establishment. He was instrumental in several studies of the behaviour of the ground and foundations
affected by clay shrinkage and swelling induced by desiccation by trees. He wrote and commissioned
many BRE publications on shallow foundations and on the consequences of their poor performance.
He has lectured extensively on subsidence and damage, and latterly has provided expert opinion for
lawyers, insurers, engineers and homeowners.
Before joining BRE in 1995, Hilary Skinner carried out research at Cambridge University Engineering
Department into the performance of soils subjected to earthquake loading. Her work at BRE (which
she left in 2006 to join Whitbybird) has included the behaviour of partially saturated and filled ground,
and predictive tools for design; flow through partially saturated soils and the implications for foundation
design; foundations for domestic housing; and design methods for CFA and augered displacement
piles, and the interpretation of installation parameters. Hilary is chairman of the British Geotechnical
Association for 2007 – 09.
IHS BRE Press
IHS BRE Press, Willoughby Road

Bracknell, Berkshire RG12 8FB
www.ihsbrepress.com
FB13
Subsidence damage
to domestic buildings

Subsidence damage
to domestic buildings

This work has been partly funded by the BRE Trust. Acknowledgements
Any views expressed are not necessarily those of
the BRE Trust. While every effort is made to ensure
the accuracy and quality of information and
Particular thanks go to Tim Freeman and Richard Chown of
guidance when it is first published, the BRE Trust, Geo-Serv Ltd for information and advice.
the authors and their sources, and the publishers
can take no responsibility for the subsequent use The support of the BRE Trust in commissioning this guide is
of this information, nor for any errors or omissions
it may contain. gratefully acknowledged.
The mission of the BRE Trust is ‘Through education Members of the Subsidence Forum are also thanked for their
and research to promote and support excellence advice and assistance.
and innovation in the built environment for the
benefit of all’. Through its research programmes
the Trust aims to achieve:
● a higher quality built environment
● built facilities that offer improved

functionality and value for money
● a more efficient and sustainable
construction sector, with
● a higher level of innovative practice
A further aim of the Trust is to stimulate debate on

challenges and opportunities in the built
environment
BRE Trust
Garston, Watford, Herts WD25 9XX, UK
Tel: 01923 664598
secretary@bretrust.co.uk
www.bretrust.org.uk
www.bre.co.uk
BRE Trust and BRE publications are available from:

Website: www.ihsbrepress.com
or
IHS BRE Press
Willoughby Road
Bracknell RG12 8FB The Subsidence Forum aims to provide professional and technical
Tel: 01344 328038 competence in all aspects of subsidence.
Fax: 01344 328005
Email: brepress@ihs.com The Forum is open to any individual or organisation involved with the insured
subsidence risk. Further details are available at:
Requests to copy any part of this publication
should be made to the publisher: www.subsidenceforum.org.uk.
IHS BRE Press
Garston, Watford WD25 9XX
Tel: 01923 664761
Fax: 01923 662477
Email: brepress@ihs.com
FB 13
© BRE Trust 2007

First published 2007
ISBN-13: 978-1-86081-977-3
v
Contents
Preface vii
1 Introduction 1
Background 1
Subsidence: the persistent problem 2
What are subsidence and heave, and what causes them? 3
Erosion, 4
Shrinkage or swelling in clay soil, 4
Why clay soils shrink and swell, 4
Where are shrinkable clays found? 5
Effect of trees, 6
Effect of climate, 8
Preliminary actions in a new case of alleged
subsidence damage 9
2 Diagnosing subsidence damage 11

What are the symptoms? 11
Preliminary assessment, 12
Have the foundations moved? 16
Why foundations move 18
Movement due to shrinkable clay 19
Location, 19
Investigating subsidence caused by leaking drains 20
Other causes of foundation movement 21
3 Investigating clay shrinkage and swelling 23

Trees and desiccation 24
Testing soil for evidence of tree influence 25
Detecting desiccation (suction), 26
Using moisture content and soil plasticity measurements, 26
Soil strength tests, 27
Soil suction tests, 30
Swelling tests, 30
Heave of desiccated clay 31
Is soil testing sufficient? 32
vi Contents
4 Subsidence monitoring 33
Why monitor? 33
Monitoring to confirm that the cause of damage is not subsidence, 35
Monitoring to establish the cause of subsidence, 35
Monitoring to measure the rate of movement, 35
Monitoring to check the success of remedial action, 35
Crack width change monitoring 36
Brass screws and vernier callipers, 36
Steel rule, 36
Plastic tell-tales, 37
Interpreting crack measurements 38
Monitoring vertical movement 39
Interpreting level measurements, 39
Monitoring lateral movements 40
How often and for how long to monitor? 40
Summary of monitoring 41
5 Dealing with the subsidence and repairing the damage 43

Mitigating tree root causes of clay shrinkage 44
Tree removal, 44
Tree pruning, 45
Root pruning, 45
Root barriers, 46
Wetting the ground, 46
Clay stabilisation, 46
Eradicating erosion subsidence by repairing leaking drains 47
Repairing and strengthening the building 47
Options for structural repair, 48
Is remedial underpinning needed? 50
Is structural stability threatened? 50
Is the movement excessive and is it continuing? 50
Partial or full underpinning? 51
Deciding if underpinning is required, 51
What type of underpinning? 53
References 55
Further reading, 56
vii
Preface
Subsidence damage to domestic buildings: a guide to good technical

practice was commissioned by the BRE Trust as part of its support for
the Subsidence Forum, a voluntary pan-industry body which aims to
reduce the amount of time and expense spent on subsidence claims,
and to improve the workings of the subsidence industry.
The guide follows on from earlier work commissioned by the BRE Trust
and published in 2000 as Subsidence damage to domestic buildings:
lessons learned and questions remaining*. It also draws on Has your
house got cracks? A homeowner’s guide to subsidence and heave
damage † which was partially sponsored by BRE.
This guide deals only with the technical and engineering aspects of
subsidence: the investigation, diagnosis, repair, prevention and
mitigation of building cracking and deformation. Its aim is to provide
guidance which can be used for training or as a simple aide-mémoire to
ensure that decisions are based on practices and information that are
the best that are available to the industry.
While the guide has been compiled from well recognised reference
documents that describe the latest information and procedures, it has
also drawn upon the opinions of key members of the industry who have
published and presented at seminars on solving subsidence cases. The
authors have also relied upon 25 years’ experience at BRE researching
subsidence problems, and latterly upon experiences gained while
providing a consultancy service for difficult, disputatious cases.
* R M C Driscoll and M S Crilly. Subsidence damage to domestic buildings: lessons learned and
questions remaining, Report FB1. Published by the Foundation for the Built Environment. Available "
#
CD
from IHS BRE Press, Bracknell, RG12 8FB.
† T J Freeman, R M C Driscoll and G S Littlejohn. Has your house got cracks? A homeowner’s guide to
subsidence and heave damage, 2nd edition. Published by Thomas Telford, London, 2001.
viii Preface
An investigator should follow a logical sequence of thoughts and

activities when confronting a new subsidence case. The guide has been
structured to present this sequence. The language has been kept as
simple and clear as possible to avoid ambiguity and scope for
misinterpretation. In this way it is hoped that the procedures
recommended here will become accepted as the industry standard,
thereby improving the quality of technical service offered by the
subsidence industry. While investigators may each choose slightly
different options, there are sensible minimum requirements that will be
appropriate for the vast majority.
In order to provide additional information and advice to those

investigating cases of subsidence, all the BRE digests and other BRE
publications referred to in this guide are included on a CD-rom disk,
written in PDF format, to be found in a pocket on the inside back cover.
References in the text to the publications included on the disk are
identified by the symbol "# .
CD

June 2007
1
Chapter 1
Introduction
Background
Before tackling a subsidence problem, it is important to understand
some of the background to subsidence and its main causes. This
introductory chapter discusses how and why subsidence has become a
major issue for some homeowners and the insurance market.
Experience has taught that many protracted disputes concerning

insurance, often leading to the expensive involvement of lawyers and
expert witnesses, could have been avoided if certain issues –
particularly about the precise involvement of trees which do not grow
on the policyholder’s property – had been raised at the start of
technical investigations. For this reason, and before going into more
detail on the ingredients of an investigation into a typical subsidence
case, a simple chart is presented in Figure 6 (page 9) which indicates
the initial actions that should take place so that these issues are
resolved before resorting to legal proceedings.
2 Introduction
Subsidence: the persistent problem

Following the low rainfall of some earlier years, the very dry summer of
2003 once again highlighted the fact that subsidence (and, to a lesser
extent, heave) is a persistent problem, both economically and socially.
The number of private domestic insurance claims averages 35,000 per
year and costs over £350 million, increasing to £550 million after dry
spells. Figure 1 indicates year-by-year experience for quarter 1 data:
the first quarter of each year usually produces the highest levels of
subsidence claims for the preceding year year.
Furthermore, local authorities spend large sums on road maintenance,

tree management and repair of buildings caused by the subsidence
phenomenon. Increasingly, subsidence claims are putting a
tremendous strain on the ‘subsidence industry’, both financially and in
terms of resourcing. The claims tend to be complex and protracted,
and therefore emotive, leading to poor relationships with customers
within the industry and with the public in general.
Number of claims Gross incurred costs

25 200
175
20
150
Number of claims (‘000)
Cost of claims (£m)
125
15
100
10
75
50
5
25
0 0
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
First quarters of each year
Figure 1 The number and cost of claims for subsidence in domestic properties.
Only Q1 data are shown since it is this quarter that largely reflects subsidence
occurrences in the previous summer
Introduction 3
What are subsidence and heave, and what causes them?

Subsidence and heave are signs that a building’s foundations have
moved, and may still be moving: downwards (subsidence) and upwards
(heave). The movement may have resulted from a wide range of ground
related factors which include:
● compression of a soft soil layer in the ground as a result of the

applied foundation loads
● shrinkage or swelling of clay soil
● soil erosion
● soil softening
● variations in the groundwater level
● compression of filled ground
● collapse of mine workings or natural cavities
● nearby construction or excavation
● frost heave
● chemical attack on the foundations
● vibration
Most of these processes will start to affect the property from the time
of construction and are therefore usually discovered during the first 10
or 20 years of the building’s life. Subsidence is not caused by the
weight of the building but by processes occurring in the ground.
For an older property without signs of longstanding foundation

problems, the appearance of subsidence (or heave) requires some
change of circumstances. In certain parts of the UK, this change can be
a collapse of an underground cavity or mine; for example, many
properties in Staffordshire, Yorkshire and Durham have been damaged
by the controlled collapse of deep coal mines. In the remainder of the
UK, there are only two common processes that are likely to cause
subsidence in an older property:
● erosion caused by water passing through the ground (such as with

leakage from a defective drain)
● shrinkage or swelling of the near-surface soil

4 Introduction
Erosion
Water seeping through soil has a tendency to transport the finer
particles; this process loosens the soil and ultimately leads to
compression. Permeable soils, such as sands and gravels, are far more
susceptible to erosion than impermeable clay soils. However, it is worth
remembering that drains are often laid in a gravel filled trench, and
escaping water can erode the gravel and soften the sides of the trench.
In older terraced properties the drainage was sometimes laid before

the house was built and runs out underneath the front wall.
Occasionally, therefore, a defective drain can lead to localised damage
where it passes underneath the foundations. Where the drainage runs
outside the house, it is more unusual for escaping water to affect the
foundations. In addition, significant quantities of water are required to
cause damage and this normally occurs only when the drain is blocked
or severely fractured.
Shrinkage or swelling in clay soil

By far the most common cause of foundation movement in the UK is
shrinkage or swelling of clay soil caused by changes in the moisture
content of the layer of soil near the surface and in the zones around
trees. Soil drying causes shrinkage which, in turn, causes subsidence;
and soil wetting causes swelling which, in turn, leads to heave.
The processes are driven by climate and its effects upon vegetation:
the dryer and hotter a summer, the more the vegetation will transpire
through its leaves moisture that it sucks from the soil through its fine
roots; the wetter a winter, the more rehydration of desiccated soil will
occur resulting in swelling and heave of the ground.
It follows that global warming and climate change – predicted to be

seen in more frequent and severe dry summers, with wetter and
stormier winters – are likely to increase the risk of subsidence and
heave damage to domestic buildings on clay soils.
Why clay soils shrink and swell

Clay soils contain a high proportion of extremely small particles with
diameters of less than 0.002 mm. Many of these particles consist of
one of the three common clay minerals: kaolinite, illite and
montmorillonite. Their molecular structure is such that their crystals,
seen under an electron microscope, are shaped like plates. Unlike
coarser grained soils, where any water in the ground simply fills the
voids between the grains, these small plates can hold the water within
their molecular structure, much as a jelly does. An increase in moisture
content forces the plates apart causing the soil to expand; and,
conversely, a reduction in moisture content allows the plates to adopt a
denser packing causing the soil to shrink.
Introduction 5
The moisture content of a clay soil can be reduced in two ways:
● by increasing the pressure on the soil by, for example, constructing a

foundation or by raising the ground level
● through moisture being sucked out of the soil by the roots of

vegetation as water evaporates through the leaves – a process
known as desiccation
If the pressure on the soil is reduced, or the source of suction is

removed, moisture will be drawn back into the soil. Whether the clay is
swelling or shrinking, because of its limited permeability these volume
changes occur only slowly over a period of months or even years.
Where are shrinkable clays found?

Firm shrinkable clays occur widely in the south east of England, as
shown in Figure 2. The geological names of these clays include London,
Gault, Weald, Kimmeridge, Oxford, Lambeth Group, Lias and Barton;
and the glacial drift clays, such as the chalky boulder clays of East
Anglia, in which clay has been mixed with a range of other soils during
the Ice Age. Their moisture contents are close to the Plastic Limit
(page 23); for example, a typical moisture content for weathered
London Clay with a Plastic Limit of say
Glacial clays
26% would be in the range of 25 to 30%.
Eocene
However, close to the ground surface,
L – London the moisture contents are influenced by
Cretaceous evaporation and rainfall, and fluctuations
G – Gault
W – Weald from as little as 15% in dry summer
Jurassic weather to 40% in wet winters can occur.
O – Oxford
K – Kimmeridge
Li – Lias
Some shrinkable clays occur further
north than the areas indicated in Figure 2;
for example, those derived from the
weathering and glaciation of
Carboniferous shales around Sunderland
and north of Shrewsbury. However, in the
north of England the surface clays are
generally sandy and their potential for
shrinkage is, therefore, smaller. In
addition to the firm clays, there are soft,
alluvial clays found in and around
estuaries, lakes and river courses, such
as the Fens, the Somerset levels, the
W
Kent and Essex marshes, and the Firths
of Forth and Clyde. All these clays have a
firm, shrunken crust which is drier than
the body of clay beneath.
Figure 2 The distribution of shrinkable clays in England
6 Introduction
Clay shrinkage is not the only foundation problem in these areas;

excessive settlement due to loading the underlying softer clay and peat
can also occur. More detailed information on the location and
identification of clay soils can be obtained from British Geological
Survey maps.
Effect of trees
Work at BRE has shown that, in grass covered areas, the effect of
evaporation in firm shrinkable clays is largely confined to the
uppermost 1 to 1.5 m of the ground. However, where there are trees,
and to a lesser extent hedges and large shrubs, moisture can be
extracted from depths down to 6 m. (In some extreme cases,
extraction has been measured below 6 m depth.) For high plasticity
clays that tend to have very low permeabilities, rainfall during the winter
cannot fully replenish the moisture removed by large trees during the
summer. Consequently, a zone of permanently desiccated soil
develops under the tree, as shown in Figure 3.
Summer without tree Summer without tree
Winter without tree Winter without tree
Summer with tree Summer with tree
Winter with tree
Root spread Zone of

Area desiccated in summer permanent Permanent moisture deficit
desiccation
Summer and winter without tree
10 20 30 40 10 20 30 40
Moisture content (%) Moisture content (%)
Figure 3 Seasonal variation in moisture content with and without trees

Introduction 7
As the tree grows, the desiccated zone increases in depth and width,
producing more subsidence which is likely to affect any nearby
structures. Figure 4 shows a property that has cracked as the result of
the growth of a poplar tree in the rear garden.
The extent of the desiccated soil depends on the moisture demand of

the tree. In general, broad leaf trees have a greater moisture demand
than evergreens and, because of their size, oak, elm, willow and poplar
are notorious for causing damage. However, these are not the ones
that most commonly cause damage to housing because other trees
with lower moisture demands – notably plane, lime and ash – are more
frequently encountered close to buildings.
Figure 4 A dramatic example of the potential effects of tree growth on a

house with shallow foundations. Subsidence has caused the back wall of
the house to rotate outwards, which is a classic symptom of tree related
damage
8 Introduction
Effect of climate
The degree of desiccation in clay soil is greatest towards the end of
summer and least in late winter or early spring, and this is reflected in
ground movement. To illustrate this point, Figure 5 shows ground
movements measured at various depths at a London Clay site over a
three year period; results are shown both for a grass covered area and
for an area containing some large poplar trees. The movements were
substantially greater in the dry summers of 1989 and 1990 than they
were in 1988, confirming that desiccation increases in hot dry weather.
The site measurements also confirmed that ground movements in the

grass covered area are generally confined to the top metre of soil,
although the unusually dry
weather of 1989 and 1990 did 10
produce movements of 6 mm 0
and 13 mm respectively at a –10
depth of 1 m.
–20
Vertical movement (mm)
Understandably, the –30
movements in the vicinity of the –40

poplar trees were larger and, for –50
example, at a depth of 1 m,
–60 Surface
exceeded 35 mm in an average
–70 1 m deep rod
year such as 1988. During
1989 and 1990, measurable –80 4 m deep rod
ground movements were –90

Grass covered areas away from trees
recorded, even at depths of
–100
4 m, indicating that prolonged
–110
periods of dry weather can Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan
produce deep seated 1988 1989 1990
desiccation. 10
–10
–20
Vertical movement (mm)
–30
–40
–50
–60
Surface
–70 1 m deep rod
–80 4 m deep rod
–90
Areas near trees
–100
–110
Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan
1988 1989 1990
Figure 5 Measurements of ground movement at various depths in London Clay

Introduction 9
Preliminary actions in a new case of alleged

subsidence damage
The decision chart in Figure 6 below suggests those simple actions,
taken initially, that should help to ensure that later disputes about the
cause of and solution to a subsidence problem do not develop.
Investigating possible subsidence need not be a complex business.
However, errors and misconceptions at the start of investigations often
complicate it. Some simple first steps, and constant reviews, can help
to solve even the most difficult of cases. The chart points to later parts
of this guide where detailed information and discussion may be found to
aid decision making.
Only so much can be learned by looking at the damage alone, so

keeping an open mind on whether the damage may be caused by
subsidence is a sensible approach.
● Drift geology
Gather preliminary information
●Age, structural form and materials of

building; drainage
● History of damage; nature, location
Visit the site and scale of cracking and deformation
● Local environs: topography and
vegetation
● Neighbouring properties
● Make plans of building layout, and
Gather all available relevant information damage patterns and scale
(Chapter 1, ‘Preliminary assessment’ – page 12)
Subsidence claim not accepted

Make initial decision on likely cause
(Chapter 2, ‘Why foundations move’ – page 18)
● Desk study to confirm site information

etc
Subsidence? No ● Is drain test required?
● Define ground and foundation
investigation, and, if tree(s) implicated,
Yes
identify roots
● Will crack monitoring be sufficient?
Is further Depending on damage and access to
tree(s), level monitoring may be
investigation needed to confirm initial decision? required to confirm need for
(Chapter 3) underpinning and its design
Yes
Commission further processes ● Define need for recovery actions?

(Chapter 4) Third party tree?
● Confirm or rule out causes of damage?
● Determine scale of problem
● Further intrusive investigations or
monitoring may be needed
Mitigate and repair damage ● Further processes must be correctly
No (Chapter 5) specified since ‘one size does not fit all’
Figure 6 The sequence of actions to be taken in the investigation of potential subsidence

10
11
Chapter 2
Diagnosing subsidence
damage
What are the symptoms?

Identifying the cause of subsidence begins with examining its
consequences. These are almost always damage showing as cracks in
the fabric of the building, or distortions such as door and window
openings deforming, or a combination of the two. Cracking is usually
obvious but distortions may be more difficult to see and may need to be
measured. Many houses, particularly after a dry spell, develop very
minor cosmetic cracks. Except during very dry summers, and in the
absence of nearby trees or vegetation, cracking is due generally to a
cause other than subsidence (an insurable peril), and may be dealt with
in the normal course of property maintenance.
Correctly diagnosing the cause of the damage can often depend on a

correct initial assessment of the nature of the damage itself. It follows
that the initial assessment of the damage is very important in setting
the agenda for the investigations which are described in this chapter.
12 Diagnosing subsidence damage
Preliminary assessment
General information and specific details about cracking and movement
in a building, to be ascertained during the preliminary site assessment,
include the following.
● Estimation of the age of cracks by determining the degree of dirt

ingress
● The approximate location and severity of the damage, and the

magnitude of cracking, including widths and densities. Damage
assessment can be very subjective – words like ‘slight’, ‘bad’,
‘severe’ or ‘unacceptable’ are often used, but are best avoided; what
may seem severe to the owner may in fact be slight in terms of its
effect on the functionality or stability of the building. To help reduce
subjectivity and misunderstandings, a classification of damage
(Table 1 opposite) has been published in BRE Digest 251[1] " # –a
CD
similar table can be found in the Institution of Structural Engineers
guide, Subsidence of low-rise buildings[2]) – which ranges from
Category 0 for slight cosmetic damage to 5 for damage that is likely
to require partial or total rebuilding. This classification is based on
the ease of repairing the damage; it therefore gives no indication of
how and when action is needed to improve the stability of the
foundations. For example, the table indicates that
damage in Category 2, which may include cracks up to
5 mm wide, can be easily filled and covered by
redecoration. The ISE guide[2] states that cracks of up to
2 mm wide, with a seasonal width change up to 1 mm, are
inconsequential
● Physical or anecdotal evidence of crack initiation and

change
● Physical features of the site. BRE Digest 348[3] "

#
CD
describes a walk-over site survey and lists valuable
sources of local site information
● Sticking doors and windows
● Sloping floors and tilting walls. BRE Digest 475[4] "

#
CD
discusses the tilting of buildings
Figure 7 shows cracking in a house. Like this example, not

all signs of movement can be immediately traced to Figure 7 Category 2 cracking (Table 1) in a
subsidence. Many cases require methodical examination by house – there were further, more severe
an expert. cracks elsewhere in the building. The centre of
the movement, in this case, was some
distance away, at a rear corner; distortion of
the structure, though, had transmitted the
movement to cause this remote cracking
Diagnosing subsidence damage 13
Table 1 BRE classification of damage (based on ease of repair of damage)

Category Description of typical damage
of damage (Nature of repair in italic type)
0 Hairline cracking which is normally indistinguishable from other causes such as shrinkage and thermal
movement
Typical crack widths 0.1 mm
No action required
1 Fine cracks. Damage generally restricted to internal wall finishes. Cracks rarely visible in external
brickwork
Typical crack widths up to 1 mm
Easily treated using normal decoration
2 Cracks not necessarily visible externally. Doors and windows may stick slightly
Typical crack widths up to 5 mm
Cracks easily filled. Recurrent cracking can be masked by suitable linings and some
external repointing may be required to ensure weathertightness. Doors and windows may
require easing and adjusting
3 Doors and windows sticking. Service pipes may fracture. Weathertightness often impaired
Typical crack widths 5 – 15 mm; or several of, say, 3 mm
Cracks which require some opening up can be patched by a mason. Repointing of
external brickwork or possibly a small amount of brickwork to be replaced
4 Extensive damage, especially above doors and windows. Windows and door frames distorted.
Floor sloping noticeably (BRE Digest 475[4]). Walls leaning[4] or bulging noticeably; some loss of
bearing in beams. Service pipes disrupted
Typical crack widths 15 – 25 mm, but also depends on number of cracks
Damage requires breaking out and replacing sections of walls
5 Beams lose bearing. Walls lean badly and require shoring. Windows broken with distortion. Danger of
instability
Typical crack widths > 25 mm, but depends on number of cracks
Structural damage which requires a major repair job involving partial or complete rebuilding
Notes
Crack width is one factor in assessing category of damage and should not be used on its own as a direct measure of it.
Local deviation of slope, from the horizontal or vertical, of more than 1/100 will normally be clearly visible. Overall deviations in excess of 1/150
are undesirable.
There are many processes that can cause distortion and cracking in
buildings (BRE Digest 361[5] " # describes why buildings crack). Aside
CD
from foundation movement, common causes of cracking include frost
attack, thermal expansion and contraction in materials, drying
shrinkage, over-stressing of walls or floors (eg as a result of injudicious
structural alterations), and chemical attack. Distinguishing subsidence
damage from that caused by these other processes can sometimes be
difficult, particularly where the damage is relatively minor.
Methods of assessing types and levels of damage are shown in Table 2

on the next two pages.
Table 2 Establishing initial damage, movement and distortion

Damage and Examples Comment
damage assessment
Initial cracking Simple use of a steel rule is

adequate for determining the
Minimum examination size and nature of cracking
Simple assessment by
eye or using, say, the
thickness of a £1 coin
a
(~3 mm)
Advisable examination
Measurement using a
rule (Figures a and b at
right) or optical device
with graticule (Figure c)
(BRE Digest 343 [6] " # )
CD
b c
Initial building There are some commercial

distortion alternatives to the techniques
● Levelness shown here that are equally
(Figure d at right) valid
Minimum examination Distortion of the building may

External visual estimation be shown by changes of level
by sighting and use of a relative to, say, the lowest point
spirit level along brick in the building; but this is not
course likely to establish beyond doubt
that parts of the structure are
Advisable examination rising or falling. Buildings are
Relative levels: not constructed perfectly level
● simple water levelling and up to ~25 mm of level
(Figure e below) d change may have been built in
● accurate optical
levelling (Figure f, BRE Digest 386 describes the
BRE Digest 386 [7] " # )
CD measurement of movement
over time by precise optical
surveying relative to a deep,
stable datum. This is the only
way to establish accurately the
amount and direction of building
movement. For a less precise
measure of movement, a
shallower temporary datum
may be used (eg a nearby
manhole cover). When
interpreting level data, light
buildings with shallow
foundations in clay soil will
move locally and seasonally by
amounts typically of 10 mm;
uniform building movements of
e f about 20 mm are often tolerable
Table 2 Establishing initial damage, movement and distortion (cont)

Damage and Examples Comment
damage assessment
Initial building Distortion of the building may

distortion (cont) show as deviations of walls from
● Verticality the vertical. Walls are not
(Figure g at necessarily constructed perfectly
right) vertical; however, any systematic
departure from the vertical in one
Minimum examination part of the building may indicate
External visual estimation foundation movement
by sighting up wall
returns. Spirit level to
measure out-of-plumb
Advisable examination
Simple plumb-offset
(Figure h at right; also
optical survey, Figure f on g
opposite page)
h h
Summary A good record of the initial damage and

distortion patterns is essential
● The precise manner in
which a structure is
deforming may be more
accurately deduced if
greater care is taken to
observe and record the
nature of the cracking
and its location in the
structure
● Figures i and j illustrate
how careful crack
description can greatly
assist in movement
assessment
A detailed explanation is
given in BRE Digests 343[6]
and 344[8] "#
CD i
j
Have the foundations moved?

Generally, the house foundations must have moved to result in insured
subsidence damage.
Except very locally, causes of cracking other than foundation

movement tend to produce only relatively small cracks less than 3 mm
or so wide. Therefore, although a crack width of more than 3 mm – the
thickness of a £1 coin – is not a necessary condition for the damage to
have been caused by foundation movement, it is a reasonable indicator.
To be significant, the crack should appear in a brick or block wall, rather
than solely in plasterwork or at the edge of a stud partition, for
example.
Some general indicators of foundation

movement are:
● a few cracks at weak points such as wall

openings
● cracks taper, usually downwards

(Figure 8)
● cracks exceed 3 mm in width and are seen

inside and outside buildings
● major cracks extend through damp-proof

courses
Figure 8 Tapering subsidence cracking
● doors and windows stick
● wallpaper rucks at corners and between

walls and ceilings
● gaps appear below skirting boards or

between floor boards and walls
● roof tiles displaced, or other signs of

distortion in roofs
● drains and services disrupted
● floors and walls noticeably out-of-level or

walls out-of-plumb
● cracks consistent with observed (or

recorded) movements
Foundation movement tends to distort openings in walls and often

causes doors and windows to stick. In some cases the distortion may
also affect partitions, ceilings, floors and roofs, resulting, for example,
in rucking of wallpaper in corners and at the junction of walls and
ceilings, gaps below skirting boards or slippage between roof tiles as
shown in Figure 9.
The best way of confirming that the foundations

have moved is to measure how much external
walls are out-of-plumb or how much brick
courses are out-of-level. Brick walls are unlikely
to have cracked unless there have been several
centimetres of differential settlement across a
typical domestic building; this should produce
distortions that can be distinguished from any
variations in level or plumb due to construction
inaccuracies. (The techniques available to
measure the amounts of movement that have
occurred are described in Chapter 4.)
Figure 9 An example of roof distortion caused by foundation

movement. The cause, in this exceptional case, was the
removal, prior to building, of a large tree in the vicinity of the
gable wall. In turn, this caused the gable wall to heave by
160 mm over a number of years leading to the discontinuity in
the roof tiles and the tapering crack beneath the lower
righthand corner of the window
Why foundations move

There are many reasons why foundations may have moved; identifying
some of them may be helped by considering the age of the building.
● In older houses – generally more than 10 years old – the cause is

likely to be either clay shrinkage or swelling involving trees or large
vegetation; or erosion in the ground, usually involving leaking drains.
Other, much less common, causes are:
● down-slope movement in clay soils
● underground cavities
● variations in groundwater levels
● chemical attack on foundations
For identifying down-slope

movement in clay soils and
underground cavities, walk-
over surveys (BRE
Digest 348[3] "# ) and desk
CD
studies (BRE Digest 318[9] "# )
CD
will prove invaluable.
● In newer houses – usually less

than 10 years old – additional
possible causes include:
● compression or collapse
of fill
● compression of soft soil
● heave of swelling soil
Approximately 70% of all

subsidence claims involve
shrinkable clay soil and trees.
Figure 10 A classic example of cracking in asphalt associated with a large
tree. Where a large tree is adjacent to this type of cracking, a similarly close
building – somewhat closer than the house in the background – may be at risk of
damage to its foundations and above-ground structure
Movement due to shrinkable clay
Where, in relation to a damaged property, it has been established that a

clay soil is present and large vegetation – normally a tree – is nearby,
the signs of clay shrinkage are:
● cracks that first appear after a prolonged period of dry weather
● cracks that open in summer and close in winter
● that the largest cracks will be found in the part or parts of the house
closest to the positions of large trees
● obvious damage to garden walls and other structures on shallow

foundations
● cracking and movement of paving and asphalt around trees
In cases of newer buildings or where tree removal has taken place,

cracking may be caused by heaving of the foundations. Swelling of clay
previously desiccated by a tree can occur after either tree removal or
severing of the root system during excavation for foundations. In
severe cases, cracking may result (Figure 9).
The presence of shrinkable clay will often be indicated by the effect of

excessive surface movements on garden walls and other structures on
shallow foundations. Paving may dip noticeably towards large trees;
and asphalt driveways, pavements and roadways surrounding or
adjacent to trees may contain crescent shaped cracks, as shown in
Figure 10 on the opposite page.
Location
The location of cracks is as important as their physical appearance.
Foundation movement often results in cracks at weak points, such as at
window and door openings, or at points where there is a change in
foundation depth, such as the junction of a bay or an extension to the
main structure. Moreover, foundation movement can produce cracks
that are continuous through the damp-proof course. The cracks are
often visible from both sides of the wall and foundation movement is
one of the few processes that can cause cracking in both leaves of a
cavity wall at approximately the same location.
Investigating subsidence caused by leaking drains

Around 30% of valid subsidence claims are not associated with clay and
trees; of these most involve leaking drains. Many drains that are more
than 30 to 40 years old are unlikely to be watertight but, while water
leaking into the soil is generally undesirable, defective drains rarely
cause substantial foundation movement. In most cases, a major
blockage or partial collapse is needed for significant amounts of water
to escape into the soil. Despite this, it is not uncommon for drains to be
identified as the likely cause of damage, even where there is no
evidence to confirm that there has been subsidence. In practice,
leaking drains rarely cause erosion in clay soils, and so an
understanding of the soil type may help in diagnosis.
It follows that a site inspection should give a clear understanding of the

location of all drains and should indicate that the apparent seat of
movement leading to damage coincides with the location of drainage.
Confirmation of leakage using a competent drain testing company is a
necessary condition to prove that leakage has caused the subsidence.
Other causes of foundation movement

The presence of soft or loose soil or poorly compacted fill beneath the
foundations is an occasional cause of foundation movement, generally
in newer properties; the movement invariably starts from the time of
construction and slows progressively as time goes by. A ground
investigation should reveal poor ground under the foundations.
Lateral movements can sometimes

occur on level sites when clay
swelling inside the building footprint
has affected a trench-fill foundation
(Figure 11); otherwise they are
unlikely to be encountered on clay
sites where the ground slopes at less
than 5°.
If clay shrinkage, leaking drains and

poor ground are ruled out by lack of
evidence, and a ground problem is
still suspected, it is essential to seek
the advice of a geotechnical
engineer with experience of
subsidence problems.
As the vast majority of valid

subsidence claims are for damage
Figure 11 Undersailing of brickwork below a damp-proof course caused caused by foundation movements
by swelling clay resulting from desiccation of clay
soils by the actions of tree roots, the
next chapter of this guide deals
specifically with investigating
subsidence and heave involving the
volume change of clay soil.
22
23
Chapter 3
Investigating clay shrinkage

and swelling
Subsidence and heave of near-surface clay soil is a seasonal

phenomenon and may cause some movement to houses; but without
nearby vegetation it is unlikely to cause damage, even in an unusually
dry year, unless the foundations are very shallow, typically less than
0.5 m deep. One possible exception to this general guideline is on a
steeply sloping site where many seasonal cycles may cause a slow,
progressive down-slope movement of the ground.
The tendency of a clay to shrink and swell is linked, among other things,
to the type and amount of clay particles in the soil. Most simply, these
are quantified in index tests to determine the plasticity, specifically the
Liquid Limit is the moisture Plasticity Index which is the difference between the Liquid and
content at which a soil changes
from a plastic solid to a liquid Plastic Limits. Plasticity Index represents the range of moisture
(ie it has no shear strength and contents over which a soil is plastic (ie it can be deformed without
will flow) volume change). The index has been used to classify the volume
Plastic Limit is the moisture change potential of UK clay soils, as shown in Table 3, adapted from the
content at which a soil changes
from a plastic solid to a brittle
classification in BRE Digest 240[10] "#.
CD
solid
Plasticity Index is the Table 3 Classification of clay volume change potential
difference between the Liquid
and Plastic Limits Modified Plasticity Index Volume change potential
(%)
Modified Plasticity Index, I'P >60 Very high

BS 1377 requires that the 40 – 60 High
percentage of material passing 20 – 40 Medium
the 425 µm test sieve (< 425 µm%) <20 Low
is reported for plasticity tests.
This enables the Plasticity Index Notes
to be modified to take account
of any sand or gravel present in This classification applies only to overconsolidated clays (ie clays that have
the test sample, to determine the previously been highly loaded; for example, by glaciation). A normally
consolidated clay (not previously overloaded) may have a considerably greater
volume change potential (BRE shrinkage (but not swelling) potential than is indicated by this classification.
Digest 240 [10]):
The Plasticity Index is modified according to the amount of granular material
I'P = I P x (< 425 µm% / 100%) in a sample (Modified Plasticity Index).
24 Investigating clay shrinkage and swelling
Trees and desiccation Potential for desiccation

If tree roots are present in
BRE Digest 343[6] CD
#" states: ‘It has long been known that trees can ground, tests for desiccation
cause clay soils to shrink …’ and ‘… can damage the building should be performed on soil
samples taken, ideally, from the
structure’. Where trees are suspected of causing damage, it will be site of the suspected movement
necessary to provide evidence of root activity in the soil. An and from a control borehole well
investigation should show fine (water abstracting) roots present at away from the suspect site and
the influence of the suspect tree
depths at least 1 m below foundation level. The roots should be or trees (‘Using moisture
sampled and identified as belonging to the suspect species. content and soil plasticity
measurements’ – page 26)
Research has shown that some species of large tree can reduce clay
soil moisture at considerable distances from them. Attempts have been Relationship between the
made to classify the potential for desiccation of clay soil by different height of a tree and its
species of tree; this was based largely on the incidence of damage distance from a damaged
building Caution should be
involving trees of particular species, and the relationship between exercised when using tree
the height of a tree and its distance from a damaged building height as a measure as there is
(BRE Digest 298 [11] " # ). Table 4, taken from Subsidence of low-rise
CD little to indicate that height is a
good predictor of the lateral
buildings [2], shows the distances beyond which common species of extent of a root system
tree are unlikely to remove significant quantities of moisture from the
soil. Figure 12 shows how these distances should be applied.
Table 4 Maximum distances
of influence of common tree
species from buildings
Trees beyond these distances
are unlikely to be a threat
Species Distance (m)
Apple or pear 10
Large
mature Ash 21
Distance tree Beech 15
between Birch 10
house and Cypress 20
tree Cherry or damson 11
Elm 30
Hawthorn 12
Holly 6
Horse chestnut 23
Laburnum 9
3%
Laurel 6
5m 2%
1%
Lime 20
Magnolia 5
Maple 20
Contours of Oak 30
5m reduction in soil
water content
Pine 8
Plane 22
Plum 11
Associated Poplar 35
50 mm subsidence Spruce 7
trough Sycamore 17
Walnut 14
White beam or rowan 11
Figure 12 Spatial variation of soil water content and associated Willow 40
subsidence profile Yew 5
Investigating clay shrinkage and swelling 25
Testing soil for evidence of tree influence

Even though diagnosis of a subsidence claim may be possible to the
trained eye by means of inspection, and by surveying and monitoring of
crack widths or foundation levels, ground investigations are normally
carried out:
● to expose the type and depth of foundation
● to confirm the diagnosis, or to diagnose the problem
● to enable considered remedial action to be taken
The discovery in a trial pit or trench of a root beneath a foundation is a

necessary, but not sufficient, condition to establishing the culpability of
a nearby tree (Figure 13 shows an excavation to expose the roots of a
Figure 13 A trench showing
street tree). Furthermore, it is necessary to have the root, or roots,
large tree roots from a London
plane tree exposed at the identified as closely as possible as belonging to the species of
entrance and boundary wall to a suspected tree. However, the principal aim of a ground investigation is
property. Finer roots – the main to establish, by means of laboratory testing of recovered soil samples,
means by which trees extract whether or not the soil beneath the foundation is indeed desiccated,
moisture from soil – were found and to what degree and extent. It is very easy when confronted with
beneath the building’s roots in the ground to jump to the conclusion that desiccation therefore
foundations some 6 to 8 m away
must be present; there can be very few houses, though, that do not
have roots growing under their foundations without any signs of
subsidence. Evidence of tree root activity may need to be gathered for
claims involving third party trees, but is often less important where the
tree belongs to the insured. An arboricultural report can be useful for
contentious cases or where a number of trees of different species grow
on the site.
The cause of the desiccation associated with tree root activity is the
negative pressure exerted on the clay soil pore water by the suction
that roots impart as they draw moisture up into the tree; this suction is
generated by transpiration of moisture through the surface of the
leaves. As the moisture content of the clay soil decreases, so does its
volume. The increase in suction which draws the soil particles closer
together causes an increase in the strength and stiffness of the soil.
BRE Digest 412[12] "# describes desiccation in clay soils.
CD
The suction induced in the clay soil also gives it the potential to swell
(with consequent reduction in strength) if free to rehydrate (eg if the
effect of the tree is removed).
Detecting desiccation (suction)

Various means are used to identify desiccation of clay soils.
● Comparing soil moisture content measurements with normal values
● Relating moisture content values to soil plasticity
● Testing samples for swelling in an oedometer
● Comparing soil strength measurements with normal values
● Attempting to measure the state of soil moisture suction in samples
Soil moisture content and plasticity are relatively easy to measure while
strength may be measured in many ways, not all of them reliable.
Suction is notoriously difficult to measure with confidence.
Inevitably the processes involved in drilling holes, and taking and

extruding samples, cause disturbance to the sampled soil. Physical
disturbance can impact on measurements of suction, strength and
swelling. Unprotected or poorly handled samples can change moisture
content; and further alterations in suction, strength and swelling are
also possible. Samples should be waxed to retain moisture, handled
carefully and tested as soon as possible to minimise the possibility of
further disturbance.
Using moisture content and soil plasticity measurements

Moisture content may be reliably measured in disturbed soil samples
taken from trial pits (BRE Digest 381[13] "
# ) or boreholes; samples
CD
must be carefully sealed prior to transporting and testing in the
laboratory or the results can lead to pessimistic assessments of
desiccation. This potential error and the difficulty of determining in a
single profile what is the normal moisture content are reduced by
comparing moisture content values in samples taken at the suspect
location and at a control location remote from it and the suspect tree;
however, such control measurements are often difficult to make owing
to the restrictions of many urban sites.There is often reluctance to pay
for this more reliable comparison.
Control profile A control

An objection to the use of moisture content profiles to assess
profile, away from the area of desiccation is that, in highly plastic clays such as London Clay, large
investigation, provides changes in soil moisture suction occur with only relatively small
information on ground
conditions for comparative
changes in soil moisture content, making desiccation detection
purposes. Where a control unreliable when moisture content deficits are small, even when the
profile is available, great care is deficit is detectable by reference to a control profile. Some people
still required in interpreting
small, apparent moisture
claim to be able to detect desiccation from a difference between the
deficits since only a small measured moisture content and a ‘natural’ value for saturated clay.
change in the soil These claims should be treated with caution as the ‘natural’ moisture
characteristics (eg constituents
and density) can account for
content, even for a well known soil such as London Clay, can vary quite
small moisture differences, appreciably from place-to-place, depending on a variety of factors
even if the soil is saturated including plasticity and density.
Plasticity The problem of

The difficulties and costs of comparative measurement have led to
variable plasticity with respect attempts to identify desiccation by comparing sample moisture content
to interpreting moisture content with some arbitrary value that is deemed to indicate desiccation. Such
profiles can, at least partly, be
overcome by plotting profiles of
an approach, based on shrinkage curves for a number of UK
Liquidity Index, I L, rather than overconsolidated clays (Driscoll [14]), suggested a relationship between
profiles of moisture content. moisture content, w, and Liquid Limit, w L, at different suction values.
I L = (w – wP) / I P The relationship suggested that the onset of desiccation could be
where w = moisture content defined as a suction of 10 kPa at which pressure many of the clay
wP = Plastic Limit and samples tested exhibited a moisture content of w = 0.5 w L. The
I P = Plasticity Index
relationship further suggested that significant desiccation (defined as a
suction of 100 kPa) occurred where w = 0.4 w L.
Suction values were
determined using pressure Largely because of its simplicity, this method is often given
plate apparatus not available
commercially unwarranted authority when assessing suction and establishing tree
causation; its indiscriminate use can lead to the mistaken belief that
desiccation is present (mistaken desiccation). Particularly, the
Suction of 100 kPa was method takes no account of the stress regime of the soil before it was
chosen as equivalent to the
magnitude of swelling pressure sampled, and can also give erroneous results in other than high
likely to overcome the plasticity soil and where the plasticity varies locally. Consequently, it is
downward pressure exerted by now generally considered that this approach should not be primarily
a typical low-rise building
foundation relied upon to indicate desiccation. However, it can provide additional
confirmation of desiccation when comparing moisture content profiles
at the suspect site and at the control site.
Mistaken desiccation
Moisture content in saturated,
overconsolidated, plastic clay Soil strength tests
will tend to reduce with depth so Undrained shear strength can provide a sensitive indication of
that deeper samples may desiccation if a repeatable measurement of strength can be obtained,
appear, erroneously, to exhibit
desiccation and the soil is reasonably homogeneous between and throughout both
the suspect and the control borehole; as with moisture content and
swelling test results, a control borehole profile is required to identify
Undrained shear strength
changes approximately by a
reliably the desiccated soil.
factor of 100 between Plastic
Limit and Liquid Limit indicating
that it is a more reliable
indicator of desiccation than
moisture content change
The undrained strength can be measured in the laboratory on

undisturbed samples using compression test apparatus or hand-held
shear vanes. Using compression test apparatus is relatively expensive
and the shear vane is not designed to deal with very stiff desiccated
soils. Both methods are influenced by the fissured nature of
desiccated soils.
Measurements may be made more practically with a dial penetrometer,

a hand-held device used for on-site assessments of bearing pressure.
When used in combination with driven-tube window samples,
measurements can be made at 250 mm intervals or less, on site, as the
sample tube is withdrawn from the borehole. Use of shear vane
equipment can be facilitated in shallow boreholes by means of
extension rods.
Two examples of desiccation detection using moisture content and

strength profiles are given in the following sections.
Example 1
This example shows, in Figure 14, profiles for a site where significant
heave was occurring, while close by there was continuing extreme
desiccation. It can be seen that the lower w values in borehole 1 at
between 1.5 m and ~5 m depth correspond quite well with the higher
penetrometer readings for the same borehole, disregarding the
anomalous high strength reading at ~3.2 m.
Moisture content (%) Dial penetrometer reading (kg/cm2)

10 20 30 40 0 1 2 3 4 5 6
0 0
>6
1 1
2 2
Depth below ground level (m)
3 3
Equilibrium
penetrometer
4 4 line for >6
weathered
Postulated London Clay
5 equilibrium 5
moisture Typical range of
6 content 6 values for equilibrium
conditions
Borehole 1
7 Borehole 2
7 Borehole 1
Borehole 2
8 8
9 9
Figure 14 The lefthand graph shows profiles of soil moisture content varying with depth for two boreholes, while the
righthand graph shows corresponding values of ground resistance (comparable with strength) measured while inserting
a penetrometer into the ground
Example 2
This example of less extreme desiccation shows, in Figure 15, the
ability of the penetrometer not only to detect desiccation but to monitor
rehydration since, over a number of years, the readings dropped closer
to the typical range of values for undesiccated London Clay.
Moisture content (%) Dial penetrometer reading (kg/cm2)

10 20 30 40 0 1 2 3 4 5 6
0 0

1 1 Equilibrium penetrometer line

for weathered London Clay
Postulated
2 equilibrium 2 Typical range of
moisture values for equilibrium
content conditions
3 3
4 4
5 5
Borehole 1 (1984) Borehole 1 (1984): 3 m from former trees (12 elms)

Borehole 1A (1992) Borehole 1A (1992): 1 m from borehole 1
Borehole 6A (1984) Borehole 6A, control (1984): 25 m from former trees
Figure 15 The lefthand graph shows profiles of soil moisture content varying with depth for three boreholes, while the
righthand graph shows corresponding values of ground resistance (comparable with strength) measured while inserting
a penetrometer into the ground
Used in combination with moisture content profiles, penetrometer

profiles provide a rapid, reliable and low cost method of assessing
desiccation for routine investigations.
The penetrometer does not permit heave potential to be assessed, but

this can be evaluated from moisture content profiles provided results
from a control borehole are available.
Inclusions of claystone or mudstone in desiccated clays may prevent

penetration of the dial penetrometer. Other problems may arise when
there are significant variations in soil types within the profile. In the case
of highly variable soils, a method which is not influenced by variations in
plasticity would be better suited.
Soil suction tests Filter paper test The mass of

The most direct measurement of desiccation is a suction water extracted from a soil
measurement, but moisture suction is one of the most difficult soil sample by a filter paper is a
measure of the suction in the
parameters to determine. The filter paper test, which has gained soil sample. BRE IP 4/93 [15] "
#
CD
quite wide application in subsidence investigations, is a practical, describes how to carry out the
commercially-available test that was developed from correlation with test
known suction values in undisturbed laboratory samples. While best
estimates of suction are obtained from undisturbed samples, this Undisturbed laboratory
makes the method expensive. Therefore, tests are being performed on samples All sampling disturbs
disturbed soil that has been re-compacted into discs in an oedometer the soil to some extent. For a
typical, plastic overconsolidated
ring; these have been found to give reasonably reliable predictions of clay, it has been estimated that
desiccation and its relative severity, but only when used to compare 50 kPA must be added to the
suspect soil with that from a control borehole. A number of other suction value to account for
disturbance
techniques have been developed but they exist largely in the research
world and are not feasible for routine subsidence investigation.
Where a control borehole is available, it is debatable whether suction

tests from a typical soil testing laboratory will give a substantially better
indication of the existence and degree of desiccation than do results
from well conducted tests for moisture content deficit or for soil
strength. However, where decisions have to be made on results from
only the borehole at the suspect site, suction tests are likely to give a
more reliable result provided that they are consistent with measured
moisture contents. Very often data are seen where a relatively high
value of suction corresponds to a relatively high value of water content
for soil of similar constituents and at similar depths, although this is the
opposite of what should be expected.
Swelling tests
Swelling tests may be carried out in an oedometer consolidation test
apparatus; the BS 1377-5 [16] test determines pressure–volume change
relationships for soils. The undisturbed soil sample is laterally confined
within a steel ring and, in the ‘free swell’ test, the swelling of the
vertically unconfined sample is measured after the sample has been
flooded with water. This free swelling is then used as an index of the
desiccation.
A more recent development has been to conduct index swelling tests

on disturbed samples remoulded in the oedometer ring. A relatively
quick swelling test is then performed and the results compared with
those for samples from a control borehole*.
* While these tests appear to give repeatable results, further calibrated testing is being carried out
and results should be available in due course from www.theclayresearchgroup.com.
Heave of desiccated clay

The time taken for the soil to recover from desiccation when the tree
has been removed, allowing the ground to rehydrate, depends largely
on the permeability of the soil. In extreme cases involving highly
shrinkable clays and large deciduous trees, it may take tens of years
for the ground to reach equilibrium even though most of the heave
occurs during the first few years. In one well documented case
(Cheney [17]), where large elm trees were removed from a London Clay
site prior to the construction of some cottages, heave was still
measurable 25 years later and had reached 160 mm. In more
permeable soils, full recovery may achieved in one or two years.
It is possible to estimate heave potential from suction and moisture

content profiles, but only if an undesiccated moisture content profile is
available. Swelling tests give a direct measure of sample heave but
great care is required in extrapolating this to a site heave figure.
Is soil testing sufficient?

A good quality ground investigation – of an extent necessary to
determine clearly the cause of the subsidence and indicate the best
remedial actions – will often be expensive and is therefore uncommon.
Frequently, investigators baulk at this expense and opt for a
combination of a limited ground investigation and some monitoring of
movement of the building. It therefore follows that monitoring plays an
important role in investigating the effects of clay shrinkage and
swelling. Monitoring is discussed in the next chapter.
As a guide, suggested minimum investigations under different

circumstances are shown in Table 5.
Table 5 Suggested minimum investigations

Scenario Purpose of investigation Investigation and Comments
monitoring content
Minor damage, Confirm subsidence Damage and distortion If local geology is mixed, trial pitting may
clear tree or clay survey (Table 2) be required to determine foundation
cause based on adequacy, soil plasticity and moisture
site and geology content
Minor damage Define remedial works Monitoring levels and cracks Can define tree management and
timescale, and confirm adequacy with
level monitoring
Major damage, Confirm subsidence Soil investigation and Should define strengths and
unclear cause or monitoring desiccation. Monitoring confirms
third party trees current movements, location and rates
Major damage Define remedial works Soil and foundation Confirm timescale and adequacy with
investigation and level monitoring. Acquire data on soil
monitoring strength and desiccation for
underpinning design. Determine
foundation adequacy via trial pits
33
Chapter 4
Subsidence monitoring
Why monitor?
Measurements in the ground

In most cases, the only reliable way of assessing if and how the
Very occasionally it is necessary foundations of a damaged or deformed building have moved, or are
to measure movements in the moving, is to carry out measurements on the building (measurements
ground. For example, on sloping in the ground). A monitoring programme may concentrate on one of
sites where there appears to be
some horizontal movement of two aspects, but can include both.
the building, measurements
made in the ground are likely to ● Crack width monitoring which records whether the damage is
be more helpful than
measurements made on the getting worse, or better, and which may indicate how the structure
structure is deforming
● Level monitoring which records the vertical movement of the

foundations
Because inadequate monitoring delays settling an insurance claim, it is

important that the monitoring is properly specified at the outset of the
claim. In this respect, level monitoring has distinct advantages over the
more traditional crack monitoring because it normally provides a clear
picture of what is happening to the property, provided a sufficiently long
period of monitoring, say six months, is conducted. Furthermore, levels
can continue to be monitored, after the repairs have been carried out,
to confirm that the foundations remain stable.
Most often, crack width monitoring is started early in the investigation

as it gives a fairly quick indication of any worsening of, or improvement
in, the cracking. Interpreting crack measurements is subjective and
subtle loading changes in the building structure, or changes in
temperature, can bring about crack changes. Changes in crack widths
are not always associated with foundation movement and changes in
crack width of the order of 1 mm or less should be viewed with caution.
In contentious cases of relatively low severity – often where the cause
of the damage is clay subsidence with a tree not within the
34 Subsidence monitoring
policyholder’s control – crack monitoring might not be sufficiently

conclusive and might be terminated prematurely. In these
circumstances, level monitoring has distinct advantages over crack
monitoring because it will provide a better understanding of what is
happening to the building and of the involvement of the suspect tree.
In practice, level monitoring is the only reliable way of assessing how
the foundations of a damaged building are moving.
Level monitoring has distinct advantages over crack monitoring

because it provides a better understanding of the scale of the problem:
particularly how much foundation movement is occurring, the focal
point of movement, and when the movements occur. It also serves as a
powerful tool in determining when it is best to carry out repairs.
Whichever type of monitoring is adopted, it is important that the

measurement techniques are accurate enough to detect changes
before they have any noticeable and further detrimental effect on the
building. Where the initial observations indicate that the damage is
worsening rapidly, the need for immediate remedial work must be
considered without waiting for the end of the monitoring period.
It also follows that monitoring should be restricted to investigations

where the information will influence the remedial measures. For
example, there would be little merit in monitoring a damaged garage or
extension if the investigator has already decided that it is beyond
economic repair and will require demolition and rebuilding. However,
since most subsidence claims involve trees, and as insurers are
increasingly requiring mitigation of the tree influence in preference to
deepening building foundations, monitoring should play an increasingly
important role in demonstrating the efficacy of reducing or removing
the trees’ influence.
The purposes of monitoring fall into the following four broad

categories.
Subsidence monitoring 35
Monitoring to confirm that the cause of damage is not

subsidence
Where the damage is relatively minor, it is generally difficult to assess
whether it has been caused by foundation movement or by an unrelated
process (eg making minor structural alterations).
Inspections, as described in Chapter 2, will provide a number of clues

that will help identify whether or not the property is suffering from
subsidence, but are rarely conclusive. Crack monitoring will frequently
show that small structural movements are occurring but it will be very
difficult, if not impossible, conclusively to ascribe these to foundation
movement. Level monitoring provides the only definitive evidence for
ongoing movement of the foundations, and whether this movement is
large enough to account for the damage. In some cases policyholders
can be reassured to see results demonstrating that no significant
movement is occurring. This is often enough evidence to convince
them to repair the damage themselves as part of ongoing maintenance.
Monitoring to establish the cause of subsidence

Where an initial investigation has proved inconclusive, monitoring can
be a very cost-effective diagnostic tool. For example, monitoring can
distinguish movement due to seasonal clay shrinkage (which tends to
be cyclic) from that due to processes such as settlement, landslip and
erosion (which tend to develop in one direction only). Again, level
monitoring will usually be far more instructive than crack monitoring.
Monitoring to measure the rate of movement

Where the cause of the damage is self-evident, crack or level
monitoring can be used to establish whether the damage is continuing
to worsen and, if so, whether the rate of movement is slowing down.
This can be a very useful technique where the damage has been caused
by a process that has a limited duration, such as heave following
removal of a tree. Similarly, where the damage has occurred during a
period of abnormal weather, level monitoring is likely to be far more
instructive that crack monitoring.
Monitoring to check the success of remedial action

The most powerful application of monitoring is to gauge the
effectiveness of action that has been taken to remove the cause of the
damage, such as cutting down or pruning nearby trees or installing a
root barrier. The investigator can then make any decisions on the need
for further remedial work, such as underpinning, objectively and fairly.
Crack width change monitoring

The most common and the simplest way of monitoring
subsidence damage is to measure changes in the width of
existing cracks. This can be done in several ways.
Brass screws and vernier callipers

The technique involves fixing small brass screws into the
wall either side of the crack and measuring the distance
between them using accurate, digital, vernier callipers
(Figure 16). This system has the advantages of being
simple, robust, relatively unobtrusive and, by using the
callipers in different modes, capable of measuring cracks
in corners and other awkward positions, as shown.
If three screws are arranged as a right-angled triangle,

both horizontal and vertical movements can be measured
(Table 6, Figure b). With vernier callipers, an accuracy of
better than ±0.1 mm overall should be easily achievable.
Steel rule
Provided sufficient care is taken, crack widths can be
measured to the nearest 0.5 mm using a steel rule.
However, because the readings tend to be subjective, and
it is difficult to ensure that the crack is measured at the
same point each time, this method is normally used only for
recording the state of damage during the initial inspection.
Figure 16 Crack monitoring using vernier

callipers. The shape of the jaws can vary
between different makes of vernier callipers
which can produce slight differences in
measurements. Also measurements can vary
when they are made by different investigators
Plastic tell-tales
The most popular system is shown in Figure 17. It consists of two
overlapping plates screwed to the wall: one marked with a cursor, the
other with a scale graduated in millimetres. The two plates are mounted
on opposite sides of the crack so that the cursor is initially in line with
the centre of the scale; any subsequent movement of the crack can
then be measured to the nearest millimetre on the scale. The advantage
of this system is that a reading can be taken at any time by anyone,
including the occupiers, without any additional measuring equipment.
Also, with a camera and lens of the necessary quality, readings can be
easily photographed.
Figure 17 Crack monitoring using a plastic tell-tale

(Photograph courtesy of Avongard)
The disadvantages are that the tell-tales may be relatively obtrusive,

vulnerable to vandalism or accidental damage depending on their
location, and have only a limited resolution in their simplest form.
Interpreting crack measurements

However accurately crack widths are measured, these are the
symptoms and not the cause of movement. Hence the results can be
ambiguous; a crack may form for one reason and progress for another.
Once a crack has formed, normal thermal expansion and contraction in
the walls can cause small changes in crack width, even though no
further foundation movement has taken place.
As most homeowners appreciate, existing cracks tend to get bigger

year by year. This is mainly due to the walls being stronger in
compression than in tension. The compression caused by thermal
expansion is strong enough to cause the cracks to open, while the
tension produced by contraction may not fully reverse the movement.
There is also a tendency for dirt and debris to fall into the cracks while
they are open which then prevents them closing. Moreover, water filling
a crack can freeze (and therefore expand), opening the crack wider. This
might become a seasonal phenomenon which produces a ratcheting
effect, further widening the gap year-by-year.
A further complication is that all properties founded normally on

shrinkable clay soils will move seasonally over the course of the year
because the surface soil shrinks in summer months and expands in
winter months. In most cases the movements will be no more than a few
millimetres and the effects on the house will be indistinguishable from
thermal expansion and contraction and other changes. However, while
these movements are incapable of causing damage, they can cause an
existing crack to open and close slightly.
For these reasons, it is important that small changes in crack width are
not automatically interpreted as positive evidence of subsidence. The
Institution of Structural Engineers[2] has recommended that cracks of
no more than 2 mm which open and close seasonally by less than 1 mm
are regarded as inconsequential. Many experienced investigators may
disagree with the figure of 1 mm and will have their own limits, which
serves only to illustrate the potential ambiguity associated with crack
monitoring. Investigators should make allowance for masonry built with
cement mortar joints which has a propensity to crack as compared to
masonry with lime mortar joints which accommodates a degree of
movement before cracking.
A further disadvantage of crack monitoring is the time period over

which measurements are needed. For claims involving clay soils, it is
often suggested that measurements are needed over a full year to
assess the effects on the damage of both the shrinkage that occurs
during summer months and the swelling during the winter.
Monitoring vertical movement

Because of the potential disadvantages of crack monitoring, it is
usually preferable to measure the vertical movements of foundations
directly. This should be done using a precision optical level to record
the movements of small screws or other contact points fixed to the
building (BRE Digest 386 [7] "
# ). While other techniques exist, none can
CD
match the overall accuracy of ±0.5 mm which is achievable with a
precision level. For most applications, small screws or masonry nails
can be used as monitoring points and these are normally sufficiently
unobtrusive to avoid acts of vandalism.
Wherever possible, levels should be measured relative to a fixed

reference point or deep datum. For most domestic applications, a
storm-water drain or similar deep feature is sufficiently stable for this
purpose; or sometimes a point on the property remote from the focal
point of damage can be used. However, where there are no deep drains
or where absolute accuracy is imperative, a deep datum can be
specially installed at a suitable depth.
Interpreting level measurements

Level monitoring provides a clear and unambiguous indication of which
parts of the property are moving and by how much. Where the
movement is continually downwards due, for example, to erosion or
compression of a layer of soft soil, even small movements (1 or 2 mm)
are significant as they will have a cumulative effect on the structure. In
this respect, level monitoring can identify movements that may be
causing no discernible changes in crack width.
More commonly, the foundation movement will be seasonal:

downwards in summer months when the surface soil is shrinking and
upwards in winter when the soil is swelling. In these circumstances,
some small seasonal movements are inevitable, and movements of a
few millimetres will have no more effect on the property than normal
thermal expansion and contraction. Larger movements – up to, say,
10 mm over the course of a year – are undesirable but may not
necessarily be causing damage. In some cases, where the movement
is evenly distributed, the property may be capable of withstanding
seasonal movements of 20 mm or more.
In most cases, though, some mitigation (Chapter 5, ‘Mitigating tree root

causes of clay shrinkage’ – page 44) will have taken place early in the
insurance claim to reduce the influence of nearby trees and shrubs, and
the primary purpose of the monitoring will be to assess whether this
action has been effective. The application of this technique is illustrated
by the results shown in Figure 18 on the next page.
10.0
Figure 18 The results of a level
Tree cut back Tree removed Repairs carried out monitoring exercise showing the
benefits of tree management
5.0 (Chart courtesy of Geo-Serv Ltd).
In this particular case, which
involved an exceptionally large
Level change (mm)
0.0 poplar, the tree was first reduced

and, when this proved ineffective,
18 m then removed
6 black poplar
Monitoring points
7 5
–5.0 1 4
2 32
3 Shed
4
5 Garage
6
–10.0 7
–15.0
Aug Nov Feb May Aug Nov Feb May Aug
1996 1997 1998
Monitoring lateral movements

Lateral movements can be critical in cases involving potential landslip.
They are rarely measured in claims involving sloping, shrinkable clay
sites because of the costs involved. The usual technique is to install a
specially slotted plastic tube in a borehole about 15 m deep. Movement
from the vertical can then be measured by lowering an instrument
known as an inclinometer down the tube. The technique is highly
specialised and should only be performed by an investigator with the
necessary expertise.
How often and for how long to monitor?

As subsidence usually involves clay volume change, a crack monitoring
period should cover, ideally, a full seasonal cycle of summer–winter–
summer with readings every 4 to 6 weeks. However, for level and
verticality monitoring, six months should normally be adequate to
establish a clear picture of which parts of the property are moving and
by how much with readings every 6 to 8 weeks.
When confirming the success of remedial treatment such as tree

management, a further period of monitoring will usually be required, as
illustrated in Figure 18 where readings have been taken at quarterly
intervals over the two year period.
Summary of monitoring
Monitoring will be needed on the vast majority of subsidence claims
and the decision on which type of monitoring is needed should be made
as early as possible. While level monitoring has a number of distinct
advantages over crack monitoring, it has to be appreciated that it is
more expensive and requires the use of specialist investigators. It is,
nevertheless, strongly recommended that level monitoring be
specified wherever underpinning is being considered as a remedial
option because of the obvious cost implications of misdiagnosis in
these situations.
Other types of claim where level monitoring has potential benefits

include the following.
• Claims involving third party trees
• Claims involving a number of trees
• Claims involving large trees (especially oak, willow and poplar),

even where these trees are remote from the area of damage
• Claims where it is not clear where the movement is taking place

(eg big cracks near the centre of the property)
• Claims where the cause of the movement is unclear or likely to be

contentious
• Claims involving suspected heave
The various aspects of crack and building monitoring are summarised,

with recommendations for the minimum and advisable amounts of
monitoring, in Tables 1 and 2 (pages 13 to 15), while Table 6 (page 42)
summarises best practice in crack width change monitoring.
Table 6 Accurate measurement of crack width changes
• These are the most commonly used means to establish

subsidence and the effectiveness of remedial work
• Therefore it is essential to ‘get it right’!
• ‘Crack width change monitoring’ (page 36) describes the

merits of the alternative methods
• It is important to use a repeatable method that has sufficient

accuracy, such as the vernier callipers shown in Figure a)
• It is important to ensure that, where appropriate, both

vertical and horizontal movement may be resolved by
measuring between three points (Figure b)
a
• But – beware of reading too much into very
small movements that may be nothing to do
with subsidence!
• Cheaper, but less accurate methods are available for

establishing trends (eg the Avongard gauge, Figure c –
photograph courtesy of Avongard)
• A version with studs can be used to take more accurate

measurements c
• However, these gauges can be vandalised (Figure d)
d
43
Chapter 5
Dealing with the subsidence

and repairing the damage
The optimum solution to a subsidence problem depends on many

factors.
● The cause of the subsidence
● The nature and scale of the damage
● The likelihood of recurrence
● The type and age of the building
The number of possible permutations of these factors is large and a

systematic decision-making approach is desirable in providing a
solution that satisfies all parties. Furthermore, it is important to be
clear if proposed action aims to rectify damage, remove the cause, or
combine the two.
The starting position should be the cause of subsidence. In this chapter

only the predominant clay–tree problem and, to a far less extent,
erosion subsidence from leaking drains are dealt with.
Often a mitigation option will be considered first or as part of the

solution to a subsidence case.
44 Dealing with the subsidence and repairing the damage
Mitigating tree root causes of clay shrinkage

The main options are:
● for many cases of low-level seasonal cracking, do nothing other than

reduce the vegetation perceived to be responsible and repair the
cracks. Future predictions of climate changes, which might lead to
increased seasonal movement, will possibly warrant crack stitching,
even in minor damage cases
● for medium levels of damage, remove offending vegetation and

strengthen or replace the fractured brickwork, underpinning only
when strengthening or replacement is not feasible
● for higher damage levels, and where action against an offending tree
is not possible, consider an underpinning scheme with crack repair
Deciding which of these options is to be followed can be quite difficult

and, for low levels of damage, much will depend on the attitudes of the
parties to the insurance claim. In all cases, the risks have to be carefully
considered before a judgement is made.
Tree removal
Removing the tree altogether will have the greatest and most
immediate effect on the levels of desiccation in the soil. However, there
is frequent opposition to the removal or reduction of an offending tree;
for example, it may belong to a neighbour or the local authority, or be
subject to a Tree Preservation Order.
If removal is feasible and all parties agree, the only risk is that long term
heave – as clay moisture content returns to its natural level – will lift the
foundations to an unacceptable degree. As a rule-of-thumb, any heave
will be acceptable provided that the tree is no older than any affected
part of the house since the heave can, at worst, only return the
foundations to their position before the tree affected them. If there is
any suspicion that tree removal may cause significant heave damage,
an appropriate soil testing programme will be required (Chapter 3,
‘Swelling tests’ – page 30). The length of time that recovery is likely to
take may be a factor in deciding whether or not removing the tree is an
acceptable solution.
Dealing with the subsidence and repairing the damage 45
Tree pruning
Where it is impractical or unsafe to remove the tree altogether and the
cracking is relatively minor, some form of pruning can be considered.
It should be done only by a reputable tree surgeon or qualified
contractor working under the instructions of an arboriculturist. The
pruning options are summarised in Table 7.
Table 7 Tree pruning options

Options Implications Advantages Disadvantages
a Crown reduction Both height and density of More leaf removed than Less visually appealing
crown reduced with b than with b
b Crown thinning Density of crown reduced Less visually damaging than Less leaf removed than
with a with a
c Pollarding Form of a involving > ~50% Larger reduction in moisture Disfigurement of tree
reduction of branch length demand
In the short term, any form of pruning of a healthy tree is likely to

stimulate growth, although the moisture uptake of the tree will still be
reduced because there are fewer leaves transpiring the moisture.
Nevertheless, there is a risk that, if pruning is not regularly repeated,
the tree may end up having a higher moisture demand than before. It is
essential, therefore, that the tree management is repeated periodically
to maintain the tree at a reduced size. Typically, this will involve pruning
every other year; some London local authorities operate a three year
regime. The results of research completed in 2004 (BRE IP 7/06 [18] " # )
CD
implied that severe pruning would be required to mitigate damage and
that therefore pruning should only be considered when regular
treatment can be guaranteed.
Root pruning
Where access to the tree is not available, it is tempting to look for
another way to remove the tree’s influence. One option is root pruning
which is usually performed by excavating a trench between the tree
and the building deep and extensive enough to cut through the roots
close to and beneath the affected foundations. The trench should not
be so close to the tree that severing its roots jeopardises its stability
(BS 5837 [19]). In the short term there might be some heave of the
foundations as the degree of desiccation in the soil under the
foundations reduces. Where the damage has only appeared in a period
of exceptionally dry weather, a return to a normal weather pattern may
prevent further damage occurring.
Permission from the local authority is required before pruning the roots
of a tree which is the subject of a Tree Preservation Order.
Root barriers
Root barriers are an extension of root pruning. Instead of simply filling
the trench with excavated soil after cutting the roots, the trench is
either filled with concrete or lined with a material that, ideally, allows
water but not fine hair roots to pass through. Unless the barrier is
sufficiently extensive laterally and deep enough, the roots may be able
to grow round or under it. It has been known for roots to grow over the
top of a barrier.
There is some limited documented information about the installation

and performance of root barriers (eg from the Arboricultural Advisory
and Information Service [20]); at least one known commercial system is
available. A barrier might be a feasible solution when one or more of the
following circumstances apply.
● The damage is of low severity
● The damage occurred after exceptionally dry weather
● The tree cannot be removed or reduced
● There is resistance to the cost and disruption of underpinning
Wetting the ground

A technique that is occasionally used in an attempt to reduce seasonal
shrinkage and swelling is to supply water to the affected ground.
However, the extremely low permeability of most natural, shrinkable
clay makes the introduction of sufficiently large quantities of water
almost completely impractical. Gravel filled trenches might help to
reduce desiccation levels provided they are maintained full of water and
provided the point of introduction is close to the zone of desiccation.
However, it is unwise to excavate too close to foundations as there
would be a danger of softening the soil and causing further settlement
of the building.
Clay stabilisation
In theory, the characteristics of a clay and its tendency to change
volume can be significantly altered by adding certain chemicals.
Shrinkage potential in particular can be reduced by using lime which
replaces sodium ions in the clay minerals with calcium ions. This
technique is effective in the laboratory and can be used to treat clay fill,
but the introduction of waterborne chemicals is subject to the same
problem of low permeability as in the wetting option.
Eradicating erosion subsidence by repairing leaking

drains
Though not the main focus of this guide, where damage has been
attributed to water escaping from defective drainage or water supply
pipes, it is important to carry out appropriate repairs as soon as
possible. In most cases, the foundations will stabilise rapidly once the
source of the escaping water has been removed.
Where drains have collapsed or are seriously damaged, they will have
to be excavated and replaced. More commonly, though, the defects
can be rectified in-situ by installing a plastic liner inside the existing clay
pipework. There are various proprietary systems on the market, some
of which come with a 10 year guarantee. Before installing the liner, the
drain is mechanically cleaned and visually checked with a video
camera. The resin-impregnated liner is then pushed (or pulled) into
place from a manhole or other access point and fixed in position by
inflating an internal bag which forces the liner against the inside wall of
the pipe until the resin has set.
Although generally cheaper than excavation and replacement, plastic

lining is not cheap and should be reserved for cases where it is evident
that leakage is eroding the surrounding soil and causing damage, or
where there is evidence of significant root growth inside the pipe.
Repairing and strengthening the building

Where it can be shown that the cause of the damage is a process which
is now largely over (such as heave following removal of a tree, or one
that is likely to occur only rarely, such as clay shrinkage during
exceptionally dry weather), it is often possible to prevent a recurrence
of damage by repairing or strengthening the superstructure.
Techniques include using tie bars and straps, resin bonding of
brickwork, brick stitching and mortar bed reinforcement. These
different options are illustrated in Figures 19 to 21 on the next two pages.
A relatively new technique, known as corseting, consists of casting a

reinforced concrete beam around the perimeter of the building, usually
at or below ground level. The beam is connected to the brickwork by
means of vertical steel reinforcing bars or ‘dowels’ and then tensioned
by a torque wrench or hydraulic jack. The corset stiffens the building at
foundation level, and helps it bridge local areas of subsidence; the
system, called ‘Hoopsafe’, is illustrated in Figure 22 (page 49).
Options for structural repair

The options, shown in Figures 19 to 21, for
repairs to the superstructure of a building are:
● various forms of structural strengthening of

brick and blockwork walls
Bed joint
● bonding cracked brickwork reinforcement
● stitching cracked brickwork
Strap twice screwed Wall plate with rear face buttered with
to each joist resin (to provide additional bonding
to brickwork) and secured by
resin-bonded anchor bolts
Ties between floor

joists and brickwork
Existing wall
Strap
notched into joists Bed-joint
Timber noggins between joists reinforcement
at tie bar positions
Extension
wall
External wall separating

from, or not bonding to,
internal wall
Existing wall
Wall starter
Galvanised steel corner straps,

resin-bonded, and anchored to external and Extension wall
internal brick or blockwork walls
Figure 19 Various forms of strengthening brick and blockwork walls

Figure 20 Bonding
cracked brickwork:
before repair (left) and
after repair (right)
Figure 21 Stitching cracked brickwork: before repair (above

left) and after repair (above right). The drawings below show
the arrangement of reinforcing bars installed on a regular grid
Section Elevation
The reinforcing bars are

installed in holes pre-drilled
from both sides of the
cracked wall, at angles
both to the horizontal and
vertical, as shown in the
Plan
section and plan views. In
the elevation view, the Figure 22 Installation of a Hoopsafe corset for
projection of these holes
forms an orthogonal a repair at foundation level (Photograph
pattern in the vertical plane courtesy of Van Elle Ltd)
Is remedial underpinning needed?

Over the past few years there has been a substantial reduction in the
often indiscriminate use of costly and disruptive underpinning to solve
subsidence problems. The emphasis has now changed to considering
alternative, less disruptive means of stabilising affected properties.
There are circumstances, though, where there will be no alternative but
to extend the foundations down to more stable ground to prevent
further, intolerable movement with the risk of recurrent cracking.
Unfortunately, underpinning is not a cure-all for subsidence. In cases of

low levels of damage it may well be ineffective and can sometimes
make a property more susceptible to damage in the future; this can
arise where partial underpinning has stabilised the foundations of some
of the building but the remainder continues to move seasonally.
One problem facing the investigator is the absence of an accepted

method for deciding when underpinning is justified and when it is not.
Consequently there is a wide range of individual opinions which
generally will be based on the following questions and considerations.
Is structural stability threatened?

Where the subsidence damage is so severe that there is doubt about
the ability of the building to continue to carry the loads applied to it – in
other words, there is risk of walls, floors, or the roof becoming unstable
– urgent action will be needed to prevent it collapsing. Although external
shoring or internal propping can be used as a temporary solution, a
permanent solution will almost certainly require underpinning or partial
rebuilding on deeper foundations. In extreme cases, the most cost-
effective solution may be demolition and rebuilding of the whole
structure. This is unlikely to apply unless the damage is Category 5
(Table 1); it is extremely rare for damage due to movement in shrinkable
clay to be as severe as this.
Is the movement excessive and is it continuing?

Where the structural stability of the building is not threatened, the
prognosis for further damage becomes of prime importance.
Engineers will want to know whether the movement is ‘progressive’; in
other words, is there evidence, such as cracks widening, of increasing
damage as a result of further foundation movement? Since small
seasonal movements are common for properties founded on clay soils,
they are rarely, if at all, justification for implementing underpinning
schemes. Small changes in crack widths can give a false impression of
what is happening to the foundations. It is therefore preferable to base
any decisions about the extent and degree of foundation movement on
direct measurements of it using level monitoring. While level monitoring
is often regarded as being too expensive to conduct, its use can
effectively eliminate unnecessary underpinning.
Even evidence of continuing foundation movement does not always

imply that underpinning is essential; movements associated with a large
tree can be effectively reduced by either pruning or removing the tree.
On the other hand, where heave has been caused by removing a tree,
there is nothing which can be done to prevent the ground from swelling
and, if further substantial movement is expected, underpinning is
probably going to be the only, though very expensive, option.
Partial or full underpinning?

The underpinning of the entire foundation of a house is rarely if ever
justifiable when the cause is clay shrinkage. Provided that level
monitoring has been appropriately applied, it should be easy to
determine what part of the foundation is moving and by how much. Only
then can a suitable laterally extensive and deep underpinning scheme
be designed. In the absence of level monitoring, partial underpinning
carries an inherent risk that, by effectively ‘fixing’ the underpinned part
into the ground, the remaining foundations may move at some future
time, rendering the structure around the junction between existing and
underpinned foundation vulnerable to distortion and cracking. It is
customary for underpinning schemes to include some extension of the
underpinning on either side of the unstable part of the foundation in
order to avoid a sudden depth transition from the newly deepened
foundation to the old, shallower foundation that may still be subject to
small seasonal movements.
Any problems associated with partial underpinning (and structural

repair) may become particularly acute where the party wall between a
pair of semi-detached houses is cracked or the party wall requires
underpinning to rectify movement elsewhere. In these circumstances it
will be necessary to comply with the requirements of the Party Wall Act [21].
Partial underpinning may be justified where the underpinning is applied
to an extension that has been built, for example, on ‘shallow’ foundations
within the zone of influence of a tree and has consequently subsided.
Excessive movement
Whether seasonal movements Deciding if underpinning is required
are ‘excessive’ has to be judged Since there must clearly be a connection between the severity of the
by what effect the movements damage and what needs to be done to rectify it, the framework shown
are having on the property.
However, as a general rule, in Table 8 (page 52) aims to help in deciding on a course of action,
movements of more than depending on the scale of damage and other factors. Of course, each
15 mm are likely to cause case has to be judged on its merits.
cracking and will be considered
unacceptable. Where smaller
movements are found to cause Table 8 uses the term ‘movement is progressive or excessive’ when
cracking, it is likely that only a determining the type of repair necessary. Defining ‘excessive’ is difficult
small part of the structure (eg a
porch, front bay or extension) is because there are some properties moving by 25 mm or more over the
being affected and, in such course of the year with little or no evidence of any distress, while there
circumstances, there may be are others where movement of less than 5 mm causes Category 3
alternatives to underpinning
such as movement joints or damage. A reasonable definition of excessive movement is given in
local reinforcement. the box at left.
Table 8 Remedial solutions for subsidence caused by clay volume change

Damage Category Appropriate action
Class* (Relevance of underpinning in italics)
(Monitoring is generally needed to confirm that damage is caused by foundation movement)
0 to 1 Low† Remedial measures are generally unnecessary since cracks can be repaired as part of
routine maintenance. Where cracks recur during periods of dry weather, consider pruning
nearby trees and shrubs.
Underpinning is unlikely to be justified except in very rare circumstances; for
example, where there is recurrent damage to expensive wall finishes
2 Low† Cracks which appear at the end of summer and close during the subsequent winter can be
repaired in the spring and steps taken to reduce the risk of damage recurring (eg pruning
nearby trees and shrubs). Where cracks are growing and are not seasonal, having taken
steps to minimise the movement, monitoring‡ should be used to establish the extent,
magnitude and rate of foundation movement.
Underpinning is unlikely to be cost effective unless foundation movement is
progressive or excessive and there is either a likelihood of recurrent damage which
will be expensive to repair, or the potential for further movement (eg as a result of
heave) will create excessive damage (say Category 4)*
3 Medium† Having taken steps to mitigate the cause of the movement, monitoring‡ should be used to
establish the extent, magnitude and rate of movement; brick arches and other susceptible
features may need propping to prevent deterioration.
Underpinning is likely to be cost-effective where movement is progressive or
excessive and alternatives such as tree removal are impracticable
4 High† Unless there is a risk of instability, monitoring‡ should be used to establish the extent,
magnitude and rate of movement. Wherever practicable, steps to remove the cause of the
movement should be taken prior to monitoring.
Underpinning is needed to prevent instability where movement is progressive or
excessive, unless the cause of the damage is obvious and can be easily removed;
for example, if caused by a large tree and there are no impediments to its removal,
this may be preferable to underpinning
5 High† Temporary support (eg external shoring or internal propping) is probably needed to prevent
collapse. Monitoring‡ may be needed to give warning of instability, but is unlikely to aid the
selection of an appropriate remedy.
Underpinning or rebuilding on deeper foundations is needed to reinstate affected
areas§. Work should be implemented rapidly to prevent unnecessary deterioration
of the structure
As described in Table 1.
*
† The boundaries between these categories are not necessarily fixed in relation to the damage classification, depending on such factors as the
location and structural significance of damage.
‡ Level monitoring should be specified wherever underpinning is being considered.
§ In some circumstances lifting or jacking the structure back to level may provide an economic alternative to rebuilding.
What type of underpinning?

There are various forms of underpinning (Figures 23 to 26, with
comments); the choice is governed largely by ground conditions,
required foundation depth and cost. Foundation strengthening and
other forms of treatment (eg grouting) may be required where voids in
the ground are suspected.
Comments apply for clay volume change and local erosion subsidence,
and may not apply for other causes of subsidence.
Comments
A ● Simple and appropriate for
partial foundation deepening;
depth limited to ~2 m and
1 4 2 1
5 3 2 5 3 therefore not well suited to
2 4
5 A 1 deep seated tree problems
Bays are numbered to indicate a typical sequence of ● Easy to provide transition
excavation, concreting and ‘pinning up’
Existing footing Transition
Pinning up Mass concrete

underpinning
Underpin
depth
Original
Edge footing
of external
excavation in bays
Section A – A
Figure 23 Mass-concrete, or ‘traditional’ underpinning
Beam
Pinning up
A
Comments
Pier
● Suited to localised clay and tree problems requiring
deeper, partial underpinning
● Beam strengthens wall ahead of underpinning
● Not economical for depths exceeding ~4 m Section A – A Section A – A
Alternative 1: beam formed Alternative 2: beam formed in

above existing footing place of existing footing
Figure 24 Pier-and-beam underpinning
Cantilevered Layout of piles

pile cap along ground
beam A Vertical mini-piling Needle piling Cantilever piling
Ground
beam A
Comments
● Suited to clay and tree problems requiring deep
Needle underpinning
capping
beam ● Pile diameter typically varies between ~150 mm and
~400 mm
● In extreme circumstances (eg expected heave of a
floor slab), a piled raft replaces the old slab
● All these methods may need protection from heaving
ground
New piles
1.0 –
1.2 m
New needle
beams
Plan of foundation
Figure 25 Pile-and-beam and piled raft or slab underpinning
Floor finish and screed Comments

Damp-proof membrane ● Slender piles (eg 65 – 150 mm Ø) are not able to resist
Damp-proof course
large lateral pressures; therefore they may not be
Floor slab suitable for clay volume change problems
Fill ● The piles rely on the strength of the existing foundation
Existing footing
Raked mini-piles
Figure 26 Mini-piling underpinning

55
References
[1] BRE. Assessment of damage in low-rise buildings. BRE Digest 251. Garston, IHS BRE
Press, 1995. "
#
CD
[2] Institution of Structural Engineers. Subsidence of low-rise buildings, 2nd edition.

London, ISE, 2000.
[3] BRE. Site investigation for low-rise building: the walk-over survey. BRE Digest 348.
Garston, IHS BRE Press, 1989. " #
CD
[4] BRE. Tilt of low-rise buildings with particular reference to progressive foundation
movement. BRE Digest 475. Garston, IHS BRE Press, 2003. " #
CD
[5] BRE. Why do buildings crack? BRE Digest 361. Garston, IHS BRE Press, 1991. "
#
CD
[7] BRE. Monitoring building and ground movement by precise levelling. BRE Digest 386.
Garston, IHS BRE Press, 1993. " #
CD
[6] BRE. Simple measuring and monitoring of movement in low-rise buildings. Part 1:
cracks. BRE Digest 343. Garston, IHS BRE Press, 1989. "
#
CD
[8] BRE. Simple measuring and monitoring of movement in low-rise buildings. Part 2:
settlement, heave and out-of-plumb. BRE Digest 344. Garston, IHS BRE Press, 1995. " #
CD
[9] BRE. Site investigation for low-rise building: desk studies. BRE Digest 318. Garston,
IHS BRE Press, 1987. " #
CD
[10] BRE. Low-rise buildings on shrinkable clay soils: Part 1. BRE Digest 240. Garston,
IHS BRE Press, 1993. "#
CD
[11] BRE. Low-rise building foundations: the influence of trees in clay soils.
BRE Digest 298. Garston, IHS BRE Press, 1999. " #
CD
[12] BRE. Desiccation in clay soils. BRE Digest 412. Garston, IHS BRE Press, 1996. "
#
CD
56 References
[13] BRE. Site investigation for low-rise building: trial pits. BRE Digest 381. Garston, IHS
BRE Press, 1993. " #
CD
[14] Driscoll R M C. The influence of vegetation on the shrinking and swelling of clay
soils in Britain. Géotechnique (1983) 33 93–105.
[15] Crilly M S and Chandler R J. A method of determining the state of desiccation in

clay soils. BRE Information Paper IP 4/93. Garston, IHS BRE Press, 1993. "
#
CD
[16] British Standards Institution. Methods of test for soils for civil engineering
purposes. Compressibility, permeability and durability tests. British Standard
BS 1377-5:1990. London, BSI, 1990.
[17] Cheney J E. 25 years’ heave of a building constructed on clay after tree removal.
Ground Engineering (1988) 21 (5) 13–27.
[18] Hipps N A, Atkinson C J and Griffiths H. Pruning trees to reduce water use.
BRE Information Paper IP 7/06. Garston, IHS BRE Press, 2006. "#
CD
[19] British Standards Institution. Trees in relation to construction.

Recommendations. British Standard BS 5837:2005. BSI, London, 2005.
[20] Marshall D, Patch D and Dobson M. Root barriers and building subsidence.
Arboricultural Practice Note 4. Farnham, Arboricultural Advisory and Information Service,
1997.
[21] Communities and Local Government. The Party Wall etc Act 1996: explanatory
booklet. Product code 02 BR 00862. Wetherby, Communities and Local Government
Publications, 2002.
Further reading
Freeman T J, Driscoll R M C and Littlejohn G S. Has your house got cracks?

A homeowner’s guide to subsidence and heave damage. London, Thomas Telford, 2001.
Driscoll R M C and Crilly M S. Subsidence damage to domestic buildings: lessons

learned and questions remaining. Foundation for the Built Environment Report FB1.
Garston, IHS BRE Press, 2000. "#
CD
Royal & Sun Alliance. The subsidence handbook. A practical guide to subsidence in
domestic property. London, Royal & Sun Alliance, 2005. Available from
www.subsidenceforum.org.uk
57
Reports from the BRE Trust

(Formerly the Foundation for the Built Environment)
FB1 Subsidence damage to domestic buildings: lessons learned and

questions remaining. R M C Driscoll and M S Crilly. September 2000
FB2 Potential implications of climate change in the built environment.

Hilary M Graves and Mark C Phillipson. December 2000
FB3 Behaviour of concrete repair patches under propped and unpropped

conditions: critical review of current knowledge and practices.
T D G Canisius and N Waleed. March 2000
FB4 Construction site security and safety: the forgotten costs. Bob Knights,
Tim Pascoe and Alice Henchley. December 2002
FB5 New fire design method for steel frames with composite floor slabs.
Colin Bailey. January 2003
FB6 Lessons from UK PFI and real estate partnerships: drivers, barriers and
critical success factors. Tim Dixon, Alan Jordan, Andrew Marston,
James Pinder and Gaye Pottinger. November 2003
FB7 An audit of UK social housing innovation. Keith Ross, James Honour and
Fran Novak. February 2004
FB8 Effective use of fibre reinforced polymer materials in construction.

S M Halliwell and T Reynolds. March 2004
FB9 Summertime solar performance of windows with shading devices.

Paul Littlefair. February 2005
FB10 Putting a price on sustainability. BRE Centre for Sustainable

Construction and Cyril Sweett. May 2005
FB11 Modern methods of house construction: a surveyor’s guide. Keith Ross.

June 2005
FB12 Crime opportunity profiling of streets (COPS): a quick crime analysis –

rapid implementation approach. J Oxley, P Reijnhoudt, P van Soomeren,
C Beckford, A Jongejan and J Jager. November 2005
FB13 Subsidence damage to domestic buildings. A guide to good technical

practice. Richard Driscoll and Hilary Skinner. June 2007
FB14 Sustainable refurbishment of Victorian housing: guidance, assessment

method and case studies. Tim Yates. September 2006
BRE Trust reports are published by IHS BRE Press

www.ihsbrepress.com
CD-rom of
Subsidence damage to domestic buildings

The CD-rom in the pocket opposite contains PDFs of:
Subsidence damage to domestic buildings: lessons learned and questions remaining (FB1)
and the following BRE publications:
Information Papers IP 4/93 and IP 7/06
Digests 240, 251, 298, 318, 343, 344, 348, 361, 381, 386, 412 and 475

BRE Driscoll Update 2007

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BRE Driscoll Update 2007

Uploaded by

Copyright:

Available Formats

subsidence damage to domestic buildings

subsidence damage to domestic buildings

about the authors

Driscoll and Skinner

IHS BRE Press, Willoughby Road

A guide to good technical practice

A guide to good technical practice

Richard Driscoll and Hilary Skinner

● built facilities that offer improved

A further aim of the Trust is to stimulate debate on

BRE Trust and BRE publications are available from:

© BRE Trust 2007

2 Diagnosing subsidence damage 11

3 Investigating clay shrinkage and swelling 23

5 Dealing with the subsidence and repairing the damage 43

Subsidence damage to domestic buildings: a guide to good technical

An investigator should follow a logical sequence of thoughts and

In order to provide additional information and advice to those

Richard Driscoll and Hilary Skinner

Experience has taught that many protracted disputes concerning

Subsidence: the persistent problem

Furthermore, local authorities spend large sums on road maintenance,

Number of claims Gross incurred costs

Cost of claims (£m)

What are subsidence and heave, and what causes them?

● compression of a soft soil layer in the ground as a result of the

● shrinkage or swelling of clay soil

● variations in the groundwater level

● compression of filled ground

● collapse of mine workings or natural cavities

● nearby construction or excavation

● chemical attack on the foundations

For an older property without signs of longstanding foundation

● erosion caused by water passing through the ground (such as with

● shrinkage or swelling of the near-surface soil

In older terraced properties the drainage was sometimes laid before

Shrinkage or swelling in clay soil

It follows that global warming and climate change – predicted to be

Why clay soils shrink and swell

The moisture content of a clay soil can be reduced in two ways:

● by increasing the pressure on the soil by, for example, constructing a

● through moisture being sucked out of the soil by the roots of

If the pressure on the soil is reduced, or the source of suction is

Where are shrinkable clays found?

Clay shrinkage is not the only foundation problem in these areas;

Summer without tree Summer without tree

Winter without tree Winter without tree

Summer with tree Summer with tree

Winter with tree

Root spread Zone of

Summer and winter without tree

Figure 3 Seasonal variation in moisture content with and without trees

The extent of the desiccated soil depends on the moisture demand of

Figure 4 A dramatic example of the potential effects of tree growth on a

The site measurements also confirmed that ground movements in the

Understandably, the –30

movements in the vicinity of the –40

ground movements were –90

Figure 5 Measurements of ground movement at various depths in London Clay

Preliminary actions in a new case of alleged

Only so much can be learned by looking at the damage alone, so

●Age, structural form and materials of

Subsidence claim not accepted