NZ Geomechanics News

Shallow replacement of Expansive or Swelling Soils

In the June 2020 issue of NZ Geomechanics News ‘The Shrink Swell Test: A Critical Analysis’, Rogers et al., 2020, was published. This article provided an overview around the use of the Shrink Swell Test and current design practice for foundation design in NZ’s expansive soils in accordance with AS 2870:2011 site soil class.

This paper provides an additional summary of the currently existing methods/ways for identification of expansive soils (defined herein as soils susceptible to both swelling and shrinkage) and presents a discussion on recently observed construction practice comprising shallow 500mm replacement of expansive soils with compacted granular fill as a ‘measure’ to reduce the deformation of the expansive soils and mitigate the risk of structural damage to a slab foundation.


‘Expansive or swelling soils’ are soils that are likely to experience volume change due to changes in moisture content (defined herein as soils susceptible to both swelling and shrinkage). Characteristic expansive or swelling soils include highly plastic clays and clay shales that often contain colloidal clay minerals such as montmorillonites. 

Expansive soils absorb large quantities of water after rainfall or due to local site changes (such as leakage from water supply pipes, sewage or drains), becoming sticky and heavy (Adem and Vanapalli, 2015). Equally, they can become stiffer when dry, resulting in shrinking (reduction of volume) and cracking of the ground (i.e. desiccation cracks). 

Types of structures most often damaged from this ‘hardening-softening’ behaviour (Jones and Jefferson, 2012) of shrink and swell cycles include foundations and walls of residential and light (one or two-story) buildings and retaining walls. Lightweight dwellings, warehouses, garages/sheds, services and pavements are especially vulnerable to damage from swelling soils as these structures are less able to suppress the differential heave of the swelling soil than heavy, multistory structures.

Identification of swelling (expansive) soils

It is important to recognize the existence of swelling soils and to understand the problems that can occur with these soils as early as possible.

Field study is used to assess the presence, extent, and nature of swelling soil and groundwater conditions. The two major phases of field exploration are surface examination and subsurface exploration.

It is not hard to recognize expansive soils in the field. The presence of surface fissures is quite common in expansive soil deposits, and an indication of their swelling potential. 

The swelling potential of a soil basically depends upon its mineral composition along with its in situ moisture content, voids ratio and density. Soil permeability also affects the rate of swelling on site.

In general, clays with plasticity indices (PI) > 25, liquid limits (LL) > 40, and natural water content (w) near the plastic limit or less are more likely to swell.

A number of testing methods and proposed classification systems exist to detect whether a soil is expansive and to predict the magnitude of shrink/swell.

Moreover, although there is no internationally recognised test for soil shrink swell potential or design methodology, a lot of work has taken place for possible correlations between other well-established measures of soil characteristics/behaviour.

AASHTO (T 258-81: Standard Method of Test for Determining Expansive Soils) prescribe a method to detect whether a soil is expansive and to predict the amount of swell. This is done by relating the Atterberg Limits of the soil to the natural soil suction at the time of construction, as shown in the Table 1 below.

In the United Kingdom, the National House Building Council (NHBC), amongst others, provides guidance on the identification of ‘shrinkable’ soils and has adopted a procedure whereby a ‘Modified Plasticity Index (I’p)’ is assessed that gives consideration to the fines content as follows:

The ‘modified plasticity index’ is then related to ‘Volume Change Potential’ as shown in Table 2.

In New Zealand, expansive soils are excluded from the NZ3604:2011 ‘good ground’ and defined as those having:

  • A liquid limit ≥ 50% when tested in accordance with NZS4402 Test 2.2, and
  • A linear shrinkage of more than 15% when tested from the liquid limit in accordance with NZS4402 Test 2.6

If soils are identified as ‘expansive’, the building designer is referred to section 17 (expansive soils), which in turn refers to various site soil classes described in AS 2870:2011 (Rogers et al, 2020). Table 3 presents the likely anticipated surface movements and typical foundation depth to protect the structure against damage by swelling or shrinking soils, as proposed by AS 2870 and MBIE November 2019 amendment.

The site soil class as per AS 2870 is commonly determined based on the shrink-swell index (ISS), as assessed through the shrink-swell test in accordance with AS 1289:2003. Rodgers et al (2020) have provided a comprehensive assessment of the reliability of the shrink-swell test which is not discussed further herein.

Tables 4 and 5 (modified from Asuri and Keshavamurthy, 2016), provide an overview of a number of currently available relationships between index soil properties and swell potential, which may be an indication for the reasons why there is not a unified or internationally accepted method or criteria.

Overlap of categories shown in Table 4 reflects the fact that the values were obtained from multiple sources.

NZ Foundation design in expansive soils / AS 2870:2011 framework

The Australian Standard AS 2870:2011 is the ‘codified’ design approach for slab foundations in areas with expansive soils. The approach to the calculation of the potential shrink-swell movements presented in AS 2870 is based on the moisture flow model developed by Richards (1967) and is generally accepted as being the most appropriate method available for the design of residential slabs and footings.

As mentioned, AS 2870, and NZS 3604, classifies the expansiveness of soil by the characteristic surface movement (ys), which corresponds to ground settlement or heave within the design life of the structure due to change in soil moisture content.

Mitchell (1980) assumed a “trumpet” shaped suction profile (Figure 1) to describe the concept of swelling of expansive soil. 

Figure 1: Theoretical suction profiles with depth for June, September, March and June (reworked after Mitchell, 1980)

The suction profile is approximated to a triangular shape in AS2870 (1996 and 2011) with values of change in suction at the soil surface (i.e. Δu in pF), as well as the recommended depth of design soil suction change are given for various locations in Australia. Soil suction is referred to as the potential energy state of water in the soil (Jury et al., 1991), and therefore, the soil suction depends on the soil moisture condition (and its variation).

It is important to note that in general, within the proposed framework, foundations with adequate stiffness and strength (e.g. shallow waffle rafts designed in accordance with Appendix F of AS 2870:2011) may be designed for any class of expansivity.


Differential heave, and settlement, is the major concern relating to ‘expansive soils’ and can be caused by nonuniform changes in soil moisture and variations in thickness and composition of the expansive soil.

Non-uniform moisture changes below a foundation system/dwelling can occur from local concentrations of water from a number of sources including:

  • surface ponding
  • broken water and sewer lines
  • leaky faucets, defective rain gutters and downspouts
  • transpiration of moisture from nearby trees
  • diffusion of moisture away from heat sources

Managing the water below the foundation/backfill and subgrade is considered as the most critical factor to control the heave potential of an expansive soil.

Essentially treatment of expansive soils can be grouped under two categories:

A. Soil Stabilisation including:

  • removal/replacement
  • remould and compact
  • pre-wetting
  • chemical/cement stabilisation

B. Water content control methods including:

  • horizontal and/or vertical barriers
  • electrochemical soil treatment
  • heat treatment

Partial replacement of expansive soils with granular fill is a widely implemented solution when moderately expansive soils of low thickness are present at the surface of the subsoil foundation. It has been recently observed that a replacement depth limited to 500mm is being specified. 

In particular, the expansive soils can be removed and replaced by:

  • non-expansive, impervious, properly compacted backfill (which also extents past the foundation laterally);
  • non-expansive pervious (granular) backfill provided:
    i. either equipped with drains to carry off infiltrated water, with the addition of an impervious moisture barrier(s) depending on the site-specific details;
    ii. or, the depth of replacement is sufficiently thick for the resulting bearing pressure at the top of the underlying expansive soil to be greater than the swelling pressure.

Chen (1975) states from experience that if the subsoil consists of more than 1.5m (i.e. 5 feet) of granular soils between the base of the foundation and the top of a highly expansive subgrade layer, there is a lower likelihood of foundation movement. Moreover, he recommended a minimum thickness range between 900 to 1500mm below the foundation depth and the top of expansive soils for the backfill.

Providing a granular backfill below and around the foundation is considered an effective technique to mitigate detrimental effects of swelling on the foundations, but not without making sure the water content of the foundation subgrade (especially below the backfill) is controlled, and the replacement is sufficiently thick and extends a suitable distance beyond the foundation footprint.

Shallow replacement of expansive soils with compacted granular fill may be assumed as a measure to balance the net pressure at the base of the foundation (or footings), however, this is probably not the case for a 500mm deep replacement for the majority of sites, due to:

  • The excavation and the granular nature of the backfill will potentially introduce more water to the subgrade;
  • If the excavation is allowed to dry prior to placing the granular material it will propagate the ‘suction profile’ as shown in Figure 1 deeper creating an increased expansion potential;
  • The corners of granular backfill will dry more than the centre, which in turn may cause differential heave / shrinkage.


It is the authors opinion that, although it may be ‘reasonable’ to consider partial replacement of the expansive soils a suitable means to mitigate structural damage, this is not necessarily the case, as (dependent on site-specific conditions), it may have the opposite effect by worsening the condition of the soil (behaviour) due to deeper infiltration (saturation / loss of suction) of the subgrade resulting in heave.

Site-specific geotechnical investigation with targeted laboratory testing to identify the soil characteristics (moisture content, Atterberg limits etc.) and swelling potential (e.g. ASTM D4546 One-dimensional Swell or Collapse of Soils), is considering a ‘probable’ best methodology for a detailed foundation design in expansive soils. 


AASHTO T 258-81 (2018) Standard Method of Test for Determining Expansive Soils, standard by American Association of State and Highway Transportation Officials

Adem and Vanapalli, Review of Methods for Predicting In-Situ Volume Change Movement of Expansive Soil Over Time, Journal of Rock Mechanics and Geotechnical Engineering, 2015, 73-86

ASTM D4546-14e1, Standard Test Methods for One-Dimensional Swell or Collapse of Soils, ASTM International, West Conshohocken, PA, 2014,

Asuri, S., Keshavamurthy, P. Expansive Soil Characterisation: an Appraisal. INAEL 1, 29–33 (2016).

Australian Standard AS 2870:2011, Residential slabs and footings

Chen, F.H. Foundations on Expansive Soils, Elsevier Scientific, 1975

Ice Manuals, Chapter 5 – Expansive Soils

Jones LD, Jefferson I. Expansive soils. In: Burland J, editor. ICE manualof geotech-nical engineering, Vol.1: geotechnical engineering principles, problematic soilsand site investigation. London, UK: ICE Publishing; 2012.

Jury W.A., Gardner W.R. & Gardner W.H. (1991) Soil Physics. John Wiley & Sons, Inc, New York

Mitchell, P. W. The Concepts Defining the Rate of Swell of Expansive Soils. 4th International Conference on Expansive Soils, 1980 Denver. 106-116

NHBC Standards online, Foundations (shrinkable soils), Section 4.5-D5,

Nelson, J.D., Miller, D.J. (1992). Expansive soils: problems and practice in foundation and pavement engineering. John Wiley & Sons, inc. New York, 1992

New Zealand Standard NZS 3604:2011, Timber-framed buildings

Richards BG. Measurement of the free energy of soil moisture by the psychrometric technique using thermistors. In: Aitchison GD, editor. Proceedings of the symposium on moisture equilibria and moisture changes in soils beneath covered areas. Sydney, Australia: Butterworths; 1965. p. 39e46.

Rogers et al. The Shrink Swell Test: A Critical Analysis, NZ Geomechanics News, June 2020, Issue 99

Tags : #expansive soils#plasticity#shrink#soils#swell#swelling soils

NZ Geomechanics News
Andreas Giannakogiorgos, Charles McDermott, David Sullivan, Fabio Parodi
NZ Geomechanics News>Issue 100 - December 2020
New Zealand

Leave a Reply