Many parts of Queensland have been declared disaster zones and thousands of residents evacuated due to a 1-in-100-year flood. Townsville is at the epicentre of the “unprecedented” monsoonal downpour that brought more than a year’s worth of rain in just a few days, and the emergency is far from over with yet more torrential rain expected.
Such monumental disruption calls for emergency work to safeguard crucial infrastructure such as bridges, dams, motorways, railways, power substations, power lines and telecommunications cables. In turn, that requires accurate, timely mapping of flood waters.
For the first time in Australia, our research team has been monitoring the floods closely using a new technique involving European satellites, which allows us to “see” beneath the cloud cover and map developments on the ground.
Given that the flooding currently covers a 700km stretch of coast from Cairns to Mackay, it would take days to piece together the big picture of the flood using airborne mapping. What’s more, conventional optical imaging satellites are easily “blinded” by cloud cover.
But a radar satellite can fly over the entire state in a matter of
seconds, and an accurate and comprehensive flood map can be produced in less than an hour.
Our new method uses an imaging technology called “synthetic aperture radar” (SAR), which can observe the ground day or night, through cloud cover or smoke. By combining and comparing SAR images, we can determine the progress of an unfolding disaster such as a flood.
In simple terms, if an area is not flooded on the first image but is inundated on the second image, the resulting discrepancy between the two images can help to reveal the flood’s extent and identify the advancing flood front.
To automate this process and make it more accurate, we use two pairs of images: a “pre-event pair” taken before the flood, and a “co-event pair” made up of one image before the flood, and another later image during the flooding.
The European satellites have been operated strategically to collect images globally once every 12 days, making it possible for us to test this new technique in Townsville as soon as flooding occurs.
To monitor the current floods in Townsville, we took the pre-event images on January 6 and January 18, 2019. The co-event pair was collected on January 18 and January 30. These sets of images were then used to generate the accurate and detailed flood map shown below.
The image comparisons can all be done algorithmically, without a human having to scrutinise the images themselves. Then we can just look out for image pairs with significant discrepancies, and then concentrate our attention on those.
Our technique potentially avoids the need to monitor floods from airborne reconnaissance planes – a dangerous or even impossible task amid heavy rains, strong wind, thick cloud and lightning.
This timely flood intelligence from satellites can be used to switch off critical infrastructure such as power substations before flood water reaches them.
Sulawesi’s recent tsunami is a striking reminder of the devastating, deadly effects that the sudden arrival of a large volume of water can have.
Published today, our new research shows what might happen if a tsunami hit Sydney Harbour. A large tsunami could cause significant flooding in Manly. Even very small waves might result in dangerous currents in the entrance of the Harbour and in narrow channels such as at the Spit Bridge.
Beyond Sydney, large areas of the east coast of Australia would also be affected.
Making waves: the tsunami risk in Australia
Our study considered a range of tsunamis, with heights ranging from just 5cm to nearly 1.5m when measured outside the Heads of Sydney Harbour. These wave heights sound small, but because the wavelengths of tsunami are so long (tens to hundreds of kilometres), these waves contain a very large mass of water and can be incredibly powerful and destructive. Wave heights also increase as the tsunami encounters shallower water.
Most tsunamis are caused by earthquakes at sea, where a shift in the sea floor creates the sudden movement of a large volume of water.
Our study approach involved modelling the likely effects of different-sized tsunamis generated by earthquakes on the New Hebrides trench to the northeast (in line with the Vanuatu islands) and the Puysegur trench (south of New Zealand).
For each event we assigned Average Recurrence Intervals (ARI), which provide an average indication of how often tsunamis of different sizes are likely to occur.
The tsunamis we studied range from an ARI of 25 years to 4,700 years. The tsunami with an ARI of 4,700 had a wave height of 1.4m outside the Heads and is the largest tsunami we could reasonably expect in Sydney Harbour. An event with an ARI of 4,700 can also be considered as an event with a 1.5% chance of occurring over a 70-year lifetime.
The tsunamis we’d expect to see in Sydney Harbour would be a sequence of waves with about 15-40 minutes on average between each peak. Some waves might break, and others might appear as a rapid rising and falling of the water level.
The highest water levels would depend on the tide and the size of the event – the largest events could raise the water level up to several metres higher than the predicted tide levels.
The visualisation below represents a tsunami in a fictional location, and shows the rise and fall of water levels (with time sped up).
A tsunami is not just one single wave, but generally a sequence of waves, lasting hours to days. Within the Harbour, larger waves are most likely to breach land, and high tide increases the risk.
The narrow part of Manly – where The Corso part-pedestrian mall is located – is one of the most exposed locations. The largest tsunamis we could expect may flood the entire stretch of The Corso between the open ocean and the Harbour.
The low-lying bays on the southern side of the Harbour could also be affected. A tsunami large enough to flood right across Manly is estimated to have a minimum ARI of 550 years, or at most a 12% chance of occurring over an average lifetime.
Examining these worst-case scenarios over time shows how this flooding across Manly may occur from both the ocean side and the harbour side, isolating North Head.
Even though the smaller tsunamis may not flood the land, they could be very destructive within the Harbour itself. Our modelling shows the current speeds caused by smaller tsunamis have the potential to be both damaging and dangerous.
The map below shows the maximum tsunami current speeds that could occur within the Harbour for the largest event we could reasonably expect.
Areas exposed to the open ocean and locations with a narrow, shallow channel – such as those near the Spit Bridge or Anzac Bridge – would experience the fastest current speeds. A closer look at the area around the Spit Bridge, shows how even smaller tsunamis could cause high current speeds.
The animation below shows a comparison between the current speeds experienced during a regular spring high tide and those that may occur if a tsunami generated by a 8.5 magnitude earthquake on the New Hebrides trench coincided with a spring high tide. A tsunami of this size (0.5m when outside the Harbour) has been estimated to occur once, on average, every 110 years (a 47% chance of occurring over a lifetime).
This video below shows similar current speeds (7m/s based on video analysis) when the Japanese tsunami of 2011 arrived in the marina in Santa Cruz, California, and caused US$28 million of damage.
Historical records show us what happened when a tsunami generated by an earthquake off Chile reached Sydney Harbour in 1960. We didn’t have any instruments measuring current speeds then, but we have witness accounts and we know that many ships were ripped from their moorings.
A whirlpool and significant erosion was also reported in the Spit Bridge area. Photographs from the time show just how much sand was washed away at Clontarf Beach.
A large tsunami affecting Australia is unlikely but possible. Remember that tsunamis are a sequence of waves that may occur over hours to days, and the biggest wave in the sequence could occur at any time.
The Joint Australian Tsunami Warning Centre (JATWC), jointly operated by Geoscience Australian and the Bureau of Meteorology, provides a tsunami warning system for all of Australia.
The bathymetry compilations used by this research are publicly available and can be viewed as a publication with links for free download.