Site classification myths: six things buyers get wrong

The six most common myths about a site classification, the soil test result for a house, are that the class belongs to the land forever, that a neighbour's soil test will do, that a lower class is always a win, that fill only means digging deeper, that Class P means you cannot build, and that a desktop study can give you the class. All six treat the letter (A, S, M, H1, H2, E or P under AS 2870-2011, the standard for residential slabs and footings) as a measured fact about the land. It is a prediction, built from a few holes, a judgement or a lab test on the clay, and a climate setting.

Much of a classification is judgement. Bruce Hargreaves, a Brisbane geotechnical engineer, kept more than 300 classification reports written by eleven South East Queensland consultants between 1999 and 2001. Only one of those consultants had run a shrink-swell test on the clay (Hargreaves 2005). Classifying from experience is allowed, and often sound. It is also a reason to read the report, not just the letter. For what each class means and how it is worked out, start with what an AS 2870 site classification is; this page deals only with the myths.

Myth 1: Is the site classification a permanent property of the block?

No. It describes the site in the state it was assessed. AS 2870 says the class must be reconsidered if site works change the ground and were not allowed for. On a reactive site that means a cut deeper than the lesser of 0.25Hs (a quarter of the depth of seasonal moisture change) or 0.5 m, or fill deep enough to trigger Class P. The reason is mechanical: the calculation lets the cracked top of a natural clay profile take up some of the movement, and a cut removes that allowance (Cameron 2018). In some SEQ profiles the clay also gets more reactive with depth, so a cut can bring the worse clay up to slab level (Hargreaves 2005).

Hargreaves found many lots classified "as tested" even when the earthworks were already known, producing both over- and under-classified sites. Removing trees, demolishing an old house or filling in a pool can also change the ground moisture enough to matter. If the site has changed since the test, ask for a review.

Myth 2: Will a neighbour's soil test do for my block?

No. A neighbour's result tells you something about the area, not your lot. AS 2870 asks for at least one borehole or pit on each building site, and at least three where seasonal moisture change runs 3 m deep or more, or where the profile is known to vary.

Adelaide shows why. Its clay plains are pocked with gilgai, natural mounds and hollows in the clay surface, so the soil profile can change within a few metres. Adelaide practice is to drill three holes on every house site and take the worst of the three results (Cameron 2018). If profiles can differ that much inside one footprint, the lot next door is not a safe stand-in. Adjacent lots often do match, but they part ways at geological boundaries, filled gullies and dams, cut-and-fill edges and trees. Whether your lot has fill, and how much, only an investigation of your lot can establish.

Myth 3: Is a lower site class always good news?

Only if it is right. A class lower than the ground deserves means a footing designed for less movement than it will see.

Under-classification usually comes from the investigation rather than the ground: holes that miss a variable part of the profile, holes that stop short of the required depth (at least 0.75Hs, with a 1.5 m minimum unless rock is hit first), reactivity assigned by eye on a clay that should have been tested, or the wrong depth of seasonal moisture change for the climate. Hargreaves (2005) saw drilling done to a set formula, possibly to suit a flat fee, rather than to give the engineer what the design needed.

The 2011 edition tightened the by-eye route: anyone classifying reactivity by look and feel must check their calls against lab tests at least every six months and once in every 50 sites (Cameron 2018). Read the report for how many holes were drilled, where and how deep, whether the clay was tested, and which Hs was used.

Myth 4: Does fill just mean digging deeper?

Sometimes, but the standard treats fill more strictly than that. Under clause 2.5.3 of AS 2870-2011:

• Controlled fill (placed and tested to AS 3798) no deeper than 0.4 m, or 0.8 m for sand, leaves the site classified as the natural ground was before filling.
• Deeper controlled fill of material other than sand makes the site Class P, although an engineer can reclassify it after assessing how the fill and the ground under it will move.
• Uncontrolled fill makes the site Class P. Where it is no deeper than 0.4 m (0.8 m for sand), the site avoids P only if every footing is taken through the fill to natural soil.

Well-compacted clay fill is not harmless either. The standard gives it no cracked zone for its first five years, so it is expected to move more than the same clay in place. In Cameron's (2018) worked example, 0.8 m of reactive fill moved a site from H1 to E.

Historical imagery can show where earthworks, filled gullies or dams have been (fill on site), which tells you to brief the investigation for fill. How deep any fill runs on your lot is for the boreholes to find.

Myth 5: Does Class P mean the block cannot be built on?

No. Class P means the site falls outside the normal A to E scale (soft ground, fill, landslip, mine subsidence, abnormal moisture such as trees, and similar), so an engineer designs the footing rather than picking one from the standard. Many P sites are built on with piers to a firmer layer, a raft designed for the site, or removal and replacement of the problem material.

A Class P report should say why the site is P and what further investigation is needed; a bare "P" is not enough to price or design from. See Class P triggers.

Myth 6: Can a desktop study give me the site classification?

No. A desktop study reads the mapped geology, soil mapping, nearby bore records, climate zone and site history, and sets out the reactivity range to expect and what the investigation should look for: fill, mapped landslip or mine subsidence, trees and old structures. It is not a site classification and cannot be used as one. Classification relies on logging the soil on the lot itself and calculating the characteristic surface movement (ys, the predicted rise and fall of the ground surface) from it. The desktop tells you what to expect and what to brief; the site investigation decides the class you build to. Our AS 2870 explainer covers how the class is worked out.

To see the mapped ground and site history for a lot before you commission the test, generate a desktop report for the address or see a sample report.

LayeredGeo compiles geology, soils, groundwater and site data into an automated geotechnical desktop report for any address in Queensland, New South Wales, Victoria and South Australia.

Sources

• Cameron, D. A. (2018). Dealing with reactive clay soils through a national standard. Australian Geomechanics, Vol 53 No 1. geomechanics.org.au
• Hargreaves, B. (2005). 20 years of AS 2870 in South East Queensland: a personal view. Australian Geomechanics, Vol 40 No 3. geomechanics.org.au

About this article. Published by LayeredGeo and written from the published research cited in the Sources section above. It is general information about how property and ground conditions are assessed in Australia, not engineering, planning, legal or financial advice, and it is not specific to any property. Check anything that matters against the source dataset or a suitably qualified professional before you rely on it. If you spot something wrong, tell us at hello@layeredgeo.com.au and we will fix it.

← All articles