07 June 2016

Owen Poland

Engineering Insight: Modelling the Risk - Tsunami Impact Research

Tsunami bore impinging on the model structure

Tsunami bore impinging on the model structure. Photo: Reza Shafiei.

How do you design a building to withstand the unknown variables of a natural disaster? In the case of earthquakes, we’re becoming old hands; but when it comes to tsunami, it’s a different story. Researchers are modelling tsunami to learn how to start mitigating the risk.

The earthquake and tsunami that struck north-eastern Japan with such devastating consequences in 2011 has sparked new research that could improve the design standards of coastal infrastructure in this part of the world. The fact New Zealand has witnessed very few tsunami in recent memory has probably created “a false sense of complacency” about what can potentially happen in a longer time frame, according to GNS Science Senior Geophysicist Dr William Power. After all, he says, the subduction plate boundary beneath the North Island is an example of the “worst case environment” for earthquake-generated tsunami, and that has raised the profile of tsunami risk in New Zealand.

Under the auspices of the Ministry of Business, Innovation and Employment-funded Natural Hazards Research Platform, GNS Science and the Faculty of Engineering at the University of Auckland are investigating the likely frequency, magnitude and impact of tsunami, which will ultimately help focus investment into reducing casualties and building and operational impacts. For its part, GNS has been feeding realistic tsunami sources and seabed bathymetry into the Cornell Multi-grid Coupled Tsunami Model (COMCOT) and Method of Splitting Tsunami (MOST) modelling packages to more accurately forecast the propagation of waves over long distances. The research includes a model to predict inundation (how far inland tsunami waves would travel and how deep the flooding will be).

One important outcome is a precalculated set of models for different New Zealand ports and harbours, which was produced in a 50:50 partnership with Dr Jose Borerro at eCoast Limited. By improving the ability to forecast what will happen in ports during tsunami, William says the models will help avoid unnecessary and potentially expensive shut downs. “The more confidently we can tell them whether it is safe to continue or not – within our margin of accuracy – the better.”

Still in its infancy

For the past few years GNS tsunami data has also been fed into physical modelling, undertaken in wave tanks at the University of Auckland, to better understand the interaction between tsunami and structures like buildings, wharves and bridges. New Zealand may be internationally recognised for its earthquake design standards, but Civil and Environmental Engineering Professor Bruce Melville DistFIPENZ says the equivalent knowledge of how to design such structures for tsunami “is really still in its infancy anywhere in the world.”

Initial experimentation involved measuring the impact of a tsunami wave or “bore” on simple geometrical shapes using a three-dimensional load cell. The cell measured the forces and the moments as a wave or debris struck an object or structure. The experiment involved using pressure transducers to measure local water pressure applied to various points on the structure.

Under Professor Melville’s guidance, PhD student Reza Shafiei, now a civil engineer with Beca, physically modelled simple wharf structures under a variety of different configurations including the angle of the structure and the slope of the coast. He looked at how a structure would respond to the likely uplift forces of a tsunami bore; specifically, whether the structure would lift off its foundations. The height and velocity of the tsunami were also varied to determine how a wharf would behave when initially hit by a wave, and how it responds afterwards to a long wave.

Another important issue is the impact of floating debris like ships and containers, which occupy most ports. “The first thing the wave picks up will be containers,” says Reza, “and these were the main issue during tsunamis in Japan.” While many large and well-built structures survived the initial impact from waves, which reached heights of up to forty metres and travelled ten kilometres inland, Reza says debris “changed the scenario” because it could damage an entire structure. Researchers tried to model a “package” of tsunami impacts including the orientation of structures and waves and the likely scouring around buildings. But the inclusion of floating debris complicated matters. While large ships could simply knock a building over, Reza says smaller debris can knock holes in walls or cladding that will let water in and cause disastrous secondary damage. “It’s very complicated because you don’t know what kind of objects are coming in, their size, weight or material,” he says. The research is ongoing, but initial results indicate that elevated structures with space or voids near the base could provide significant advantages. Known as “Piloti-type” structures, these allow water to flow underneath buildings and can reduce direction forces by as much as 50 per cent as well as a significant reduction in uplift. “It can be a good mitigation strategy in tsunami-prone areas,” says Reza.

Interestingly, some original research at the University of Auckland has found whether a structure was rigid or flexible made little difference. “Everything depends on the shape of the structure as long as it doesn’t deform” says Reza, “so when you have a long wave like a tsunami bore the effect of flexibility was negligible.” Bruce says there’s still an “open question” in terms of the differing hydrodynamics between the initial short duration impulsive force of a tsunami and the longer-duration constant applied force of the bore. The extent of any damage, it seems, can depend on the natural frequency or resonance of structures after impact.

Another piece of research by the engineering faculty has been a desktop study of bridge designs, which will feed into an initial set of guidelines in the NZ Transport Agency’s (the Transport Agency) bridge design manual. Based on post-disaster assessments of different failure modes, the aim was to quantify the maximum tsunami forces being applied to an average bridge super structure. “That’s really the ultimate aim of what we’re doing,” says Bruce, “to provide the structural designer with much better tools to design coastal structures for the effect of tsunami.”

GNS Science is not directly involved in the Transport Agency’s project, but William says future planning needs to take account of likely exposure to tsunami. “If we were to build a bridge on the foreshore at Napier it might need to be built to a higher standard than a bridge on the foreshore at New Plymouth simply because on the East Coast we anticipate larger tsunami.” And while recently-developed US engineering guidelines for building in tsunami-prone areas may be considered a starting point, William says “it makes perfect sense we should have equivalent guidelines for building tsunami-resistant structures in tsunami-exposed places”.

Investing in research

In terms of future research, the focus will be on wharves and bridges using a new tsunami flume at the new engineering campus at Newmarket. “We’re lucky the University of Auckland is investing in newer and much better laboratories,” says Bruce. The University is also grateful for the client-funded contract research from the likes of the Transport Agency, as it opens up more avenues for the funding of long-term research, which normally relies on Government grants.

To that end, the Engineering Faculty’s Centre for Infrastructure Research (CIR) is taking a leading role in promoting a multidisciplinary and multi-sectoral approach to strategic planning. While the engineering and technical solutions are important, CIR Director Dr Jim Bentley says it’s also important to consider the economics around issues like tsunami impact. “Some of those decisions hinge less on the technical aspects and more on the societal impact or benefit, or the economics or political acceptability.”

This article featured in our April/May issue of Engineering Insight, delivered to IPENZ Members. Not a member? Learn about the benefits of becoming one