Australia is a quiet but not silent continent when it comes to earthquakes. Following the recent earthquake in Melbourne, Swinburne University’s Professor John Wilson and Associate Professor Hing-Ho Tsang discuss how good design and adherence to the codes of practice can help mitigate the risk of building collapse during an extreme earthquake event.
Australia experiences one to two magnitude 5 earthquakes on average each year and a magnitude 6 earthquake every five to 10 years.
Significant earthquakes that have struck Australia in recent memory include a magnitude 5.6 event in Adelaide in 1954 and a magnitude 6.5 earthquake in Meckering, Western Australia in 1968.
Three earthquakes with magnitude 6.2-6.5 also occurred within a 12-hour period at Tennant Creek in 1988 and in 1989 a magnitude 5.3 earthquake in Newcastle, NSW tragically claimed 13 lives and caused more than $2 billion in damage.
Many of the major intraplate earthquakes, both locally and globally, come as a surprise in locations that are considered low risk.
This includes the recent magnitude 5.9 earthquake in Victoria, which did not occur near any potentially active faults that have already been identified. There is a consensus among the earthquake engineering community that nowhere in Australia is immune from earthquake hazard.
The Australian insurance industry is aware of the earthquake risk and annually transfers in the order of $300 million to re-insurance companies overseas to reduce their exposure. Interestingly, the reinsurance companies rate an earthquake in Sydney within their top 20 risk exposures worldwide.
Minimum design threshold
The Australian Building Codes Board is similarly aware of the earthquake risk and attempts to mitigate this by ensuring all new buildings satisfy the requirements of the Australian Earthquake Loading Standard AS 1170.4. This was released in 1993 and revised in 2007 and again in 2018.
The most recent update included the introduction of a minimum threshold seismic design coefficient of Z=0.08 across the country to provide some base level of seismic resilience for all new buildings, in recognition of the fact that probabilistic hazard assessments in Australia are difficult due to the inherent lack of data. The Z=0.08 provides modest protection and can be compared with a Z=0.40 factor used in high seismic regions for the ultimate design earthquake event.
“Many of the major intraplate earthquakes, both locally and globally, come as a surprise in locations that are considered low risk.”
The minimum design threshold intends to ensure that new buildings are capable of resisting ground shaking triggered by a magnitude 6 event at around 30 km. Data recorded from the magnitude 5.9 event on 22 September 2021 show that the shaking experienced at locations 60 km or farther away from the earthquake epicentre were significantly lower than the minimum design threshold as expected (reliable data from closer distances is not available).
Hence, inelastic response or significant damage to modern buildings in the metropolitan Melbourne area, which is over 120 km away from the earthquake epicentre, was not expected in that event or any aftershocks of smaller magnitudes, although the shaking was quite perceptible.
Australian earthquakes by their very nature are considered low probability but high consequence events. As it is a large continent and not very urbanised, many events are remote and cause little damage. But when the earthquake is a ‘bullseye’ within an urban area, significant damage can be expected.
In recent years, structural design standards, such as the AS 3700 for Masonry Structures, AS 3600 for Concrete Structures and AS 5100 for Bridge Design have been updated to improve the detailing of structures to improve the inherent ductility and drift capacity and to reduce the overall vulnerability to extreme earthquake events.
However, the standards only apply to new buildings and do not apply to existing building stock, unless the buildings are undergoing significant renovation. The philosophy implicit in this non-retrospective approach is that over time, the overall building stock will be improved through the adoption of the latest building standards in the design and construction process.
Good seismic design involves thinking conceptually about how the whole structure responds as a system to lateral forces and displacements to ensure that premature brittle failure is prevented.
Most buildings fail in earthquakes because of insufficient drift capacity to withstand the displacements imposed, which then results in catastrophic building collapse through the action of gravity loads.
In general, buildings that are regular in plan and elevation with floor diaphragms that are robustly connected to the vertical and horizontal force resisting systems and elements that are well detailed and not excessively stressed under gravity loads, will behave best under extreme earthquake shaking.
Good design also includes particular attention to the non-structural components such as façades, mechanical and electrical services and architectural fittings to avoid excessive damage and business interruption under earthquake conditions.
The most vulnerable building stock in Australia are considered unreinforced masonry buildings (particularly parapets, chimneys and gables), buildings with large discontinuities to create open spaces at ground floor level (known as soft storey buildings), some precast construction (depending on the detailing of the connections), brittle façade systems (such as glass and masonry) and heavily loaded reinforced concrete structures where the vertical gravity load carrying elements are highly stressed, particularly if high strength concrete is utilised (drift capacity may be less than 1 per cent). Also, buildings sitting on soft soils experience stronger shaking, as seismic wave energy could be trapped within soil sediments that amplify shaking at the ground surface.
Melbournians living in medium-to-high-rise apartment blocks felt violent shakings during the recent earthquake. Tall buildings are indeed more flexible laterally due to their slenderness, and hence, larger deflection is expected in a strong wind or earthquake event. Contrary to common perception, taller buildings are actually not at higher risk in earthquakes, as their lateral load resisting systems are typically designed to possess sufficient lateral strength and deformability to ensure overall structural integrity, even in a very rare event of earthquake.
It should be understood that the intent of the building design standards is to prevent collapse and protect life, whilst allowing significant damage to buildings in an extreme earthquake event. The resulting damage can be quite severe and greatly impact society and the economy when the damage from all the buildings and infrastructure is aggregated.
A good example of this was New Zealand’s Christchurch, which is founded on deep, soft, loose and saturated alluvial deposits. On 22 February 2011, a ‘bullseye’ magnitude 6.2 earthquake struck the city. It tragically killed 185 people and caused catastrophic damage that required a $40 billion rebuild of the whole city, which is still going a decade after the devastating event.
In higher seismic regions, performance-based design is being encouraged, so clients/owners can invest in buildings that will remain operational after a significant earthquake event and avoid the hefty repair bills, the significant costs of business interruption and the social and economic impact to the city and country.
Risk-based design is another emerging approach that aims to limit the collapse and fatality risk to tolerable levels. This aggregated approach to seismic design is aimed at improving the community’s resilience to such extreme events and is considered best practice.
Complacency about earthquake design and construction methods has been ingrained in the construction sector for over 25 years, in that good design is missing in most commercial properties and houses. The resilience factors in the NCC and AS1170 are not understood by more than 90% of engineers, and is not taught at university. This unfortunately will see many failures in the years to come and potential losses of upto a billion dollars, in my view of the industry. But its never too late to start doing good design or apply best practice, just getting paid properly to do that is an even bigger hurdle.