The low down on liquefaction

By Lynley Hargreaves 22/10/2014

Dr Brendon Bradley
Dr Brendon Bradley

Since the Canterbury earthquakes, most of us are familiar with the effects of liquefaction – sand volcanoes, sunken buildings, and vast, vast quantities of mud. But scientists still don’t fully understand the interactions between the deep soft soils of the Canterbury basin and the shaking seen on the ground. Canterbury University’s Dr Brendon Bradley explains how his Marsden grant will help us understand the ways other cities might move in future, and why Christchurch shook so much.

Did liquefaction in Christchurch catch scientists by surprise?

Prior research already suggested that Christchurch contained soils which would liquefy in future major earthquake shaking. However, the severity and spatial extent of liquefaction were somewhat surprising, with some structures sinking 40-100cm, and liquefaction impacts felt almost everywhere to the east of Hagley Park.

Have many other places had similar problems?

Liquefaction has been observed many times in previous earthquakes around the world, but very rarely to the same extent as in Christchurch (the 1964 Niigata, Japan, earthquake being the closest example). One key difference was that liquefaction in the Christchurch earthquakes occurred in ‘native’ soils, that is, soils that were deposited through natural processes, as opposed to on ‘reclaimed’ or man-made land. There are other locations around the world, and even in New Zealand, where highly-susceptible soils exist like those in Christchurch, it’s just that many of them haven’t experienced strong earthquakes in historic times. Places in New Zealand that might be prone to liquefaction include Wellington Harbour and Petone, Napier/Gisborne, and low-lying regions of Auckland.

How deep into the ground are you looking?

In terms of soil liquefaction, we are looking up to 30 metres below the ground surface. Once you go deeper than this the pressure increases to the point where soil liquefaction is very unlikely, and the consequences are minor. However, the soils in Christchurch are approximately 1000m deep before basement rock is reached, and this large thickness has a significant effect on the way in which seismic waves travel from the earthquake source – at approximately 5km depth – toward the ground surface. The soils basically act like a big ‘bowl of jelly’, vibrating significantly more and longer than the rock (or ‘bowl’) below them. This increases the potential for liquefaction and other associated damage to the city’s infrastructure on the ground surface.

Do you have to drill into this ‘bowl of jelly’?

We use various different experimental techniques to understand soil properties depending of the depth we are interested in. For understanding soils near the ground surface at depths of less than 30m we can test the soil directly using, for example, a Cone Penetration Test, which is a highly-instrumented rod pushed into the ground. It records the soil resistance, friction and water pressure. We also have methods which can remove soil samples from the ground in an ‘undisturbed’ state so they can be taken to the laboratory and tested. For significantly greater depths, we can’t directly measure soil properties in a very economical way (because drilling to deep depths is very expensive), so we use ‘non-invasive’ methods, which are based on the generation of seismic waves at the ground surface and recording these waves as they propagate though the earth at a set of measurement instruments. By performing inversion analysis based on our understanding of wave propagation theory we can determine the geophysical properties of the deep soil layers.

Why hasn’t anyone looked at this before?

People have examined shallow liquefaction and the seismic response of deep soil deposits independently in the past. But usually they don’t have skill sets in both areas – the shallow liquefaction is the realm of the geotechnical engineer and the deep soils of the geophysicist. In my past research I have been involved in both these fields, and particularly in the coupling of the two effects, which can’t simply be ‘added together’. We are working with experts from California who previously have not given much attention to modelling these effects in ground motion simulations, but who know that such effects are important in places like San Francisco.

These interviews showcase researchers supported by the Marsden Fund which, since 1994, has been supporting fundamental, investigator-led research in New Zealand.